UNIVERSIDADE ESTADUAL DE CAMPINAS

FACULDADE DE CIÊNCIAS MÉDICAS

OLADELE SIMEON OLATUNYA

ASSOCIATION BETWEEN HEMOLYSIS INTENSITY, GENETIC MARKERS

AND CLINICAL EVOLUTION IN PATIENTS WITH SICKLE CELL DISEASE

ASSOCIAÇÃO ENTRE A INTENSIDADE DA HEMÓLISE, MARCADORES

GENETICOS E EVOLUÇÃO CLÍNICA EM PACIENTES COM DOENÇA

FALCIFORME

CAMPINAS

2018

OLADELE SIMEON OLATUNYA

ASSOCIATION BETWEEN HEMOLYSIS INTENSITY, GENETIC MARKERS

AND CLINICAL EVOLUTION IN PATIENTS WITH SICKLE CELL DISEASE

ASSOCIAÇÃO ENTRE A INTENSIDADE DA HEMÓLISE, MARCADORES

GENETICOS E EVOLUÇÃO CLÍNICA EM PACIENTES COM DOENÇA

FALCIFORME

Thesis presented to the Faculty of Medical sciences of the University of Campinas in part-fulfilment of the requirement for the award of doctor in medical sciences, with area of concentration in clinical pathology.

Tese apresentada à Faculdade de Ciências Médicas da

Universidade Estadual de Campinas como parte dos

requisitos exigidos para a obtenção do título de doutor em

Ciências Médicas, área de concentração em Patologia

Clínica.

ORIENTADOR: PROF. DR. FERNANDO FERREIRA COSTA

ESTE EXEMPLAR CORRESPONDE À VERSÃO FINAL DA TESE DEFENDIDA PELO ALUNO OLADELE SIMEON OLATUNYA, E ORIENTADO PELO PROF. DR. FERNANDO FERREIRA COSTA

CAMPINAS

2018

FICHA CATALOGRÁFRICA:

BANCA EXAMINADORA DA DEFESA DE DOUTORADO

OLADELE SIMEON OLATUNYA

ORIENTADOR: PROF. DR. FERNANDO FERREIRA COSTA

MEMBROS:

1. PROF. DR. FERNANDO FERREIRA COSTA

2. PROFA. DRA. MARIA STELLA FIGUEIREDO

3. PROFA. DRA. SANDRA FATIMA MENOSI GUALANDRO

4. PROF. DR. ANDRE FATTORI

5. PROFA. DRA. MARINA PEREIRA COLELLA

Programa de Pós-Graduação em [PROGRAMA] da Faculdade de Ciências

Médicas da Universidade Estadual de Campinas.

A ata de defesa com as respectivas assinaturas dos membros da banca

examinadora encontra-se no processo de vida acadêmica do aluno.

Data: 16/01/2018

DEDICATION

This work is dedicated to Almighty God, the Alpha and Omega, beginning and the end,

who gave me the opportunity to start this research and to complete it, may His name be

praised forever.

ACKNOWLEDGEMENTS

My profound gratitude goes to the Almighty God for His sustaining grace,

protection, wisdom, good health and provision throughout the research work. My

undiluted gratitude goes to my amiable supervisor, Professor Fernando Ferreira Costa,

for his mentorship, guidance and supervision during the research work. Your simplicity

and hardwork have impacted me greatly. I am indeed very grateful sir.

I appreciate Drs Dulcinéia Martins de Albuquerque, Carolina Lanaro, Carla

Penteado, Ana Leda Longhini, Irene Santos and Flavia Leonardo Costa for their

inestimable supports during the research. God will continue to lift you up. My sincere

appreciations also go to Ana Luisa de Lorenzo for providing me with very huge and

inestimable administrative and secretariat supports. God bless you richly. To Daniela

Pinheiro, and my students` colleagues who assisted me in times of need, i say a big

thank you. My thanks to all the laboratory staffs at the Hemocentro (Unicamp), and a

host of others who has contributed to the success of this work. I also wish to appreciate

Prof (Mrs) Adeyinka Falusi, Mr Abayomi Odetunde, Mr Kayode Tolorunju Segun and

Mrs Benson Tolulope for providing laboratory supports for some analysis in Nigeria. I

also wish to greatly thank Drs Taiwo Adekunle, Faboya Ayodeji and Ajibola Ayo for

their great contributions. God bless you all.

My special appreciation to the sickle cell support society of Nigeria (SCSSN) for

nominating me for the programme that allowed me to undertake this study. I also wish to

specially appreciate Professor Adekunle Adekile for his mentorships, encouragements

and supports. God bless you sir. Also, my deep appreciations go to Professor Marilda of

FLOCRUZ foundation Bahia for her supports. I wish to specially thank my teachers at

the Ekiti State University for their inestimable roles at seeing me through this programme.

I need to make special mention of the vice chancellor of the Ekiti State University for

approving my study leave. Also, i greatly appreciate Professors GJF Esan, MA Araoye,

KS Oluwadiya, and OAO Oyelami for their timely supports, encouragements and strongly

recommending me for this programme. God bless you richly sirs!

To my father, His Highness Ojo James Olatunya, the Obalese of Oye Ekiti, who

has longed so much to see me complete this research and my wonderful ever supporting

mother Chief Mrs Margaret Modupe Olatunya, i say thank you so much i`m eternally

grateful to you both.

This section will not be complete if i fail to mention the supports of my friend,

my companion and God given wife, Dr Mrs Mercy Ayomadewa Olatunya for her

financial, moral, spiritual and academic supports. My love, you are wonderful! I pray

that God will protect us and help us to reap the fruits of our labour (Amen). Also, i want

to say a big thank you to our children Tijesunimi, Jesutimilehin, and Jesudarasimi who

were always praying and longing for me to complete this research work. I pray that

God will spare your lives and make you great in life in Jesus name. Amen.

RESUMO

Diversas hipóteses têm sido propostas para explicar a diversidade clínica da doença

falciforme (DF). Estas incluem a classificação de um sub-fenótipo hemolítico e sugerem

a potencial contribuição de marcadores genéticos. No entanto, estes não são explicam

totalmente as expressões fenotípicas observadas nos pacientes com DF, assim, há a

necessidade de buscar por mais marcadores candidatos para a DF. Este é um estudo de

correlação fenotípica com alguns marcadores hemolíticos conhecidos e desconhecidos

dentre os pacientes com DF. Este estudo avaliou a relação entre micropartículas

eritrocitárias e outros marcadores tradicionais de hemólise com os fenótipos clínicos de

138 brasileiros com DF, após consentimento, sendo 78 HbSS (63 em uso de hidroxiureia,

15 sem o uso da medicação), 12 HbS-Beta0 talassemia, 12 HbS-Beta+ talassemia e 36

HbSC sem uso de hidroxiurea e em estado estacionário da doença. Ainda, 110 crianças

nigerianas com DF, sendo 102 HbSS e 8 HbSC, foram avaliadas. Um total de 107

indivíduos, 39 do Brasil e 68 da Nigéria, fizeram parte do grupo controle. As

micropartículas eritrocitárias foram quantificadas em plasma por citometria de fluxo,

haptoglobina e hemopexina foram avaliadas por ELISA, e a hemoglobina plasmática e o

heme foram mensurados por ensaio colorimétrico. PCR foi utilizada para confirmar o

diagnóstico de DF. O perfil para alfa talassemia foi determinado por GAP-PCR multiplex,

os genótipos para UGT1A1 foram analisados por meio dos fragmentos gerados, e os

haplótipos βS e a deficiência de G6PD por meio de ensaio TaqMan. Os pacientes

apresentaram níveis elevados de micropartículas, hemoglobina plasmática e heme livre

na seguinte ordem: HbSS > HbSC > HbAA. Por outro lado, haptoglobina e hemopexina

estiveram mais elevadas no grupo controle: HbSS<HbSC<HbAA. As micropartículas

eritrocitárias mostraram correlações significantes com os marcadores tradicionais de

hemólise. Não foram observadas diferenças significantes nos níveis de micropartículas,

heme e hemoglobina plasmática entre os indivíduos HbSS tratados e não tratados com

hidroxiureia. Tanto as micropartículas quanto o heme estiveram associados com a

ocorrência de úlcera de perna e o risco para hipertensão pulmonar determinado pela

velocidade de regurgitação da tricúspide (TRV). Além disso, este estudo mostrou que a

coexistência de alfa talassemia foi significantemente associada com melhores índices

hematológicos, aumento de crises de dor óssea e proteção contra úlcera de perna. O

polimorfismo UGT1A1 foi significantemente associado com níveis mais altos de

bilirrubina e ocorrência de cálculo biliar dentre os pacientes nigerianos. Da deficiência de

G6PD nao foi associada com os eventos clinicos. Os SNPs do gene BCL11A

influenciaram significantemente os níveis de hemoglobina fetal desses pacientes e seus

haplótipos βS foram principalmente a homozigose do tipo Benin/Benin. Em conclusão,

este estudo mostrou que a alfa talassemia, e o polimorfismo UGT1A1 afetam os eventos

clínicos das crianças nigerianas com DF. Além disso, estabeleceu que as micropartículas

eritrocitárias são associadas com a hemólise e os eventos clínicos em brasileiros com DF.

Ambos RMP e heme foram associados com úlcera de perna e TRV (risco para hipertensão

pulmonar) elevado. Ainda, o heme esteve associado com microalbuminúria. Essas

observações sugerem que as micropartículas e o heme podem desempenhar importantes

papeis na fisiopatologia e nas manifestações clínicas da DF. Assim, RMP e heme como

alvo de terapias podem ser uma nova estratégia para tratar a DF.

Palavras chaves: doença falciforme, manifestações clínicas, hemólise, micropartículas

eritrocitárias, marcadores genéticos.

ABSTRACT

Several hypotheses have been proposed for the clinical diversity in SCD. These include

the classification of a sub-hemolytic phenotype and hints on the potential contributions

of genetic markers. However, these do not fully explain the phenotypic expressions

observed in SCD patients. Hence, the need to search for more candidate markers of SCD.

This is a study on phenotype correlation with some known genetic and newer candidate

hemolysis markers among SCD patients. This study evaluated the relationship between

red blood cell microparticles and other traditional markers of hemolysis in relation to

clinical phenotypes of 138 consenting adult Brazilians with SCD made up of 78 HbSS

(63-hydroxyurea treated &15-hydroxyurea naïve), 12 S-Beta0 thalassemia, 12 S-Beta+

thalassemia and 36 HbSC hydroxyurea naive patients in steady state. Also, 110 Nigerian

children with SCD made up of 102 SS and 8 HbSC were studied. A total of 107

individuals, 39 from Brazil and 68 from Nigeria, served as controls. The plasma red blood

cell microparticles were quantified by flow cytometer, haptoglobin and hemopexin by

ELISA, plasma hemoglobin and heme were measured by colorimetric assays. PCR was

used to confirm the diagnosis of SCD. Alpha thalassemia status was determined by

multiplex GAP-PCR, UGT1A1 genotypes by Fragment analysis, βS haplotype and G6PD

by TaqMan SNP genotyping assays. The patients had significant higher levels of red

blood microparticles, plasma hemoglobin, heme in the following order HbSS > HbSC >

HbAA. On the contrary, both haptoglobin and hemopexin were higher in controls than

the patients in the following reverse order HbSS<HbSC<HbAA.

Red blood cell microparticles showed significant correlations with traditional markers of

hemolysis. There were no significant differences in the red blood cell microparticles,

heme and plasma hemoglobin levels of hydroxyurea treated SS patients compared to

untreated patients. Both red blood cell microparticles and heme were associated with leg

ulcer and elevated trancuspid regurgitation velocity (TRV)-risk for pulmonary

hypertension. Also, this study established that coexistence of alpha thalassemia was

significantly associated with better hematologic indices, increased bone pain crisis and

protection against leg ulcer. The UGT1A1 polymorphism was significantly associated

with higher bilirubin levels and occurrence of gallstone among the Nigerian patients.

Similarly, G6PD deficiency was not associated with clinical events. The SNPs of the

BCL11A significantly influenced the fetal hemoglobin levels of the Nigerian patients and

their βS haplotypes were mainly of the homozygous Benin/Benin type. In conclusion,

this study established that alpha thalassemia, and UGT1A1 polymorphisms affect the

clinical events among Nigerian SCD children. It also established that the red blood cell

microparticles are associated with hemolysis and clinical events among Brazilian SCD

patients. Both RMP and heme were associated with leg ulceration and elevated TRV (risk

for pulmonary hypertension). Also, heme was associated with microabuminuria. These

observations suggest that red blood cell microparticle and heme may have important roles

in the pathophysiology and clinical manifestations of SCD. Hence, therapies targeting

RMP and heme could be another strategy to combat SCD.

Keywords: Sickle cell disease, Clinical manifestations, Hemolysis, Red blood cells

microparticles, Genetic Markers.

LIST OF ABBREVIATIONS

SCD – Sickle cell disease

SCA – Sickle cell anaemia

RBC – Red blood cells

WBC – White blood cells

RMP – Red blood cell microparticles

TRV – Trancuspid regurgitation velocity

UACR – Urinary albumin creatinine ratio

VOC – Vaso occlusive crisis

ACS – Acute chest syndrome

NO – Nitric oxide

TLR4 – Toll-like receptor 4

DAMPs – Damage associated molecular pattern molecules

LDH – Lactate dehydrogenase

MCV – Mean corpuscular volume

HU – Hydroxyurea

HbF – Fetal haemoglobin

SNP – Single nucleotide polymorphism

Hp - Haptoglobin

HPX – Hemopexin

Plasma Hb- Plasma Hemoglobin

G6PD – Glucose 6 phosphate dehydrogenase

UGT1A1- Uridine diphosphate glucuronosyl transferase 1A

LIST OF TABLES

Table 1: Summary of parameters among hemolysis study cohorts

Table 2: Effects of hydroxyurea treatment on biologic markers of HbSS patients

Table 3: Effects of hydroxyurea treatment on biologic markers of HbS-β thalassemia patients

Table 4: Associations between acute chest syndrome and biologic markers

Table 5: Associations between acute bone pain crisis (VOC) and biologic markers

Table 6: Associations between leg ulcer and biologic markers

Table 7: Associations between elevated tricuspid regurgitation velocity/risk for

pulmonary hypertension and biologic markers

Table 8: Associations between stroke and biologic markers

Table 9: Associations between sickle cell retinopathy and biologic markers

Table 10: Associations between osteonecrosis and biologic markers

Table 11: Associations between priapism and biologic markers

Table 12: Associations between microalbuminuria and biologic markers

Table 13: Distribution of βs-haplotypes and G6PD deficiency among Nigerian cohorts

Table 14 : Associations between G6PD deficiency and biologic markers of Nigerian SS

cohorts

Table 15: Co-inheritance of sickle cell anemia with G6PD deficiency and clinical events

Table 16: Allele and genotype frequencies of UGT1A1 promoter polymorphism among

participants

Table 17: Comparison of laboratory markers between patients and controls UGT1A1

study cohort

Table 18: Influence of UGT1A1 genotype on laboratory parameters of SS cohort

Table 19: Influence of UGT1A1 genotype on clinical events of SS patients

Table 20: Comparison of parameters in patients with and without gallstones

Table 21: Biodata and frequencies of alpha thalassemia alleles

Table 22: Laboratory parameters of patients and controls of the thalassemia study cohort

Table 23: Alpha thalassemia Alleles and laboratory parameters

Table 24: Co-inheritance of sickle cell anemia with Alpha Thalassemia and clinical events

Table 25: Comparison of parameters in patients with or without leg ulcer in the absence

of Alpha thalassemia

Table 26: Associations between Haptoglobin genotypes and biologic markers of SS cohorts

Table 27: Influence of Haptoglobin genotype on clinical events of SS cohorts

Table28: Table of BCL11A SNPs examined

Table 29: Descriptive statistics and comparisons of BCL11A SNPs between the patients and controls

Table 30: Measures of fetal hemoglobin by allele combination and comparison in Patients group

Table 31: Measures of fetal hemoglobin by allele combination and regression analysis.

Table 32: Linkage disequilibrium between the SNPs pairs.

LIST OF FIGURES

Figure 1: Pathophysiology of sickle cell disease

Figure 2: Sickle cell disease and hemolysis products

Figure 3: The βs-haplotypes

Figure 4: Comparison of red blood cell microparticles across groups

Figure 5: Comparison of plasma hemoglobin across groups

Figure 6: Comparison of heme across groups

Figure 7: Comparison of haptoglobin across groups

Figure 8: Comparison of hemopexin across groups

Figure 9: Comparison of LDH across groups

Figure 10: Comparison of total serum bilirubin across groups

Figure 11: Comparison of unconjugated bilirubin across groups

Figure 12: Comparison of reticulocyte count across groups

Figure 13: Comparison of RBC concentration across groups

Figure 14: Comparison of haemoglobin concentration across groups

Figure 15: Comparison of fetal hemoglobin across groups

Figure 16: Red blood cell correlation plots

Figure 17: Glucose 6 phosphate dehydrogenase agarose gel by restriction analysis

Figure 18: UGT1A1 promoter genotypes found among Nigerian groups

Figure 19: Agarose gel analysis showing α 3.7 deletion by PCR.

Figure 20: Illustration of modeled Red blood cells microparticles in the lumen and endothelial regions of blood vessels

Figiure 21: Illustration of the predictive value of tricuspid regurgitant velocity on functional outcomes and mortality risk in sickle cell disease

Figure 22: Mechanisms of hemolytic anemia in reducing NO bioavailability and association with vasculopathic sub-phenotypes of sickle cell disease

TABLE OF CONTENTS

PAGE

Introduction…………………………………………………………………...17 – 28

Justification……………………………………………………………………29

Objectives……………………………………………………………………..30

Patients and methods………………………………………………………….32 – 44

Results…………………………………………………………………………45 – 101

Discussion……………………………………………………………………..102 – 118

Conclusions……………………………………………………………………119 – 120

References…………………………………….……………………………….121 – 133

Appendices………………………………………………………………… 134 – 201

Annexes……………………………………………………………………. 202 - 209

17

1. INTRODUCTION

Sickle cell disease (SCD) is a common genetic disorder of man that is caused by

a mutation in the β-globin gene1,2. SCD is a disorder of public health importance,

especially in sub-Saharan Africa where the largest burden of the disease exists with about

6 million people affected. Nigeria is the country with the highest burden of the disease

where approximately 2 to 3% of all newborns are born with the disorder3. Due to slave

trade and people migration, the disease has spread from Africa to the other parts of the

world. In Brazil, the prevalence of SCD varies between 0.8 and 60 per 100,000 live births

in different regions of the country and most of those affected are Brazilians of African

descent4,5.

Pathophysiologic basis of SCD

The proximate cause of SCD is a single gene mutation resulting in a single base

change from adenine to thymine (GAG to GTG) in the codon for amino acid. This

mutation results in the replacement of the hydrophilic glutamate by hydrophobic valine

at the sixth amino acid residue of the β-globin polypeptide chain and HbS (Hemoglobin

S) β-globin chain substitute for normal HbA β-globin chain. The HbS can undergo

reversible polymerization when deoxygenated. However, following repeated sickling and

un-sickling cycles, the HbS becomes irreversibly polymerized and injures the RBC (Red

blood cell) by causing irreversible damage to the RBC membrane. The damaged RBCs

have a shortened life span and are removed thus forming the basis for chronic anemia in

SCD. The intravascular component of hemolysis depletes nitric oxide thereby causing

disruption in its balance and this leads to some vascular complications. In addition, other

complex interactions with endothelial cells and some molecules lead to cellular injuries,

inflammation, and other cascades of downstream events.and injuries seen in SCD1,2.

Some aspects of the pathophysiology and complicating events are summarized in Figure

1 below.

18

Figure 1: The Glu6Val mutation leads to HbS formation which polymerises at low O2

tension causing damage to RBC membrane, ↓RBC life span, Hemolysis, NO depletion,

and vasoocclusion. Steinberg MH, 2008. The Scientific World Journal2

Studies have shown that SCD is clinically pleiotropic both within an individual

and among groups of patients. The clinical spectrum ranges from asymptomatic or mild

course to persistent, severe or life-threatening situations often requiring frequent hospital

visits and admissions1,2. These inter- and intra-patient variability and unpredictable

phenotypic expression pose significant management challenges to physicians and

caregivers1,4.

Several hypotheses have been proposed for the phenotypic diversity in SCD with

environmental influences as well as sociodemographic characteristics playing major roles

in tropical Africa2,3,6. Others have attributed the clinical variability to the roles of some

genetic modifiers such as the presence of α-thalassemia, and Glucose 6 phosphatase

dehydrogenase deficiency (G6PD). Also, fetal hemoglobin production which is thought

to be influenced by β-globin gene haplotype and some other factors, has been

implicated6,7,8. In addition, other non-globin genetic factors like the uridine diphosphate

glucuronosyl transferase 1A (UGT1A1) promoter polymorphism has also been found to

19

modify the clinical course of SCD7. However, these factors do not fully explain the

phenotypic diversity observed in SCD patients. Hence, researchers continue to search for

more markers of the SCD clinical expression8,9. Studies on markers of hemolysis such as

reticulocyte count, lactate dehydrogenase, aspartate transaminase, serum bilirubin, heme,

haptoglobin, hemopexin, cell-free plasma haemoglobin and red blood cell micro particles,

have shown some promise in elucidating the clinical course of SCD9-11. Hemolysis,

though considered traditionally as the cause of anemia and gallstone formation in SCD,

has now been shown to cause more than these complications in SCD. Some

pathophysiological processes, such as, endothelial dysfunction, chronic inflammation and

vascular injury have been linked to hemolysis via the release of cell free plasma

haemoglobin, heme, and other toxic products during hemolysis. This leads to a cascade

of downstream events and complications in SCD patients12-15. Due to the damages and

injuries caused by these products of hemolysis, circulating cell free haemoglobin and

heme are now being referred to as erythrocytic damage associated molecules

(eDAMPS)12,15. This is further explained and simplified in figure 2 below. As shown, in

the figure, although the human body has innate mechanisms to neutralise products of

intravascular hemolysis, these endogenous mechanisms are often overwhelm in SCD

patients thus leading to circulating eDAMPS12,15.

Figure 2: Hemolysis releases erythroid DAMP molecules to drive vascular injury and

sterile inflammation, which contribute to the pathogenesis of sickle cell disease.

Hemolysis releases cell free hemoglobin (Hb), which is normally scavenged by

haptoglobin and CD163. Free hemoglobin reacts with and scavenges NO via the

dioxygenation reaction and also reacts with hydrogen peroxide to generate hydroxyl

radicals via the Fenton reaction. This process leads to endothelial dysfunction and

20

pathological vascular remodeling. Oxidized hemoglobin releases free heme, which can

trigger a sterile inflammatory reaction involving TLR4 activation, and stimulates

neutrophils to release NETs. These inflammatory events are proposed to cause

vasoocclusion and acute chest syndrome in sickle cell disease. There are several potential

therapies using the indicated agents (shown in red text) that target multiple stages of this

proposed pathophysiological pathway. RBC=red blood cell (Gladwin MT et al12).

Furthermore, although the impacts of genetic modifiers of SCD are relatively

known in the developed world, there is paucity of information on this from Africa.

Moreover, because of genetic variability in different populations, it is pertinent that more

studies are carried out among cohorts from different ethnic backgrounds to fully

understand the impact of genetic modifiers on SCD.

These observations underscore the inherent potentials of hemolysis and genetic

markers contributions to the clinical course of SCD and fuells the need for search for

more markers of SCD expression that could help to fully understand the disease.

Therefore, putting more focus on some hypothesised markers may help in elucidating

more facts on predictors of SCD phenotypes.

1.2. REVIEW OF SOME HYPOTHESISED SCD PHENOTYPE MARKERS OF

INTEREST

HEMOLYSIS MARKERS

1.2.1 Haptoglobin: This is a protein produced mainly by the liver and regarded

as an acute phase protein because it is elevated in inflammatory conditions. It is an α-

sialoglycoprotein found in mammals but in humans it exhibits polymorphism through two

dominant alleles (Hp1 and Hp2) located on the long arm of chromosome 16q22 and 3

phenotypes have been recognised (HP1-1, HP2-2, & HP2-1)16,17. The HP1-1 allele is

found commonly among Africans and South Americans. It is least prevalent in South east

Asia and it is the most biologically active of the three haplotypes with regards to binding

free plasma Hb (plasma hemoglobin) and suppressing inflammation16. The HP2-2 variant

has the least biological activities with respect to binding free plasma HB and suppressing

inflammation16,17. The HP2-1 variant has intermediate biological and anti-inflammatory

abilities compared to the first two variants. Although haptoglobin has antioxidant and

antibacterial properties, their most striking roles are generally in modulating acute phase

21

responses. They are the first line scavenger of free plasma Hb where they combine and

bind free hemoglobin in the blood stream to form an hemoglobin-haptoglobin (Hb-HP)

complex which is quickly recognised by the scavenger receptor CD163 located on the

surfaces of circulating monocytes and liver macrophages16,17. Haptoglobin disappears

faster than being created when large amounts of RBC are destroyed in the intravascular

compartment due to excess amount of free Hb released into the blood stream. This leads

to reduction in the blood level of haptoglobin. As a result of this, haptoglobin has high

sensitivity and specificity in the diagnosis of hemolytic anemia16,17. Hence, in conjunction

with other markers of hemolysis, and in the absence of other factors like liver disease,

infections, drugs and other inflammatory diseases that may cause alterations in

haptoglobin, blood levels of haptoglobin could be used to ascertain the degree of

hemolysis in chronic hemolytic anemia condition like SCD16,17. Although previous

studies showed the predominance of Hp 1-1 and Hp 2-1 haplotypes among Brazilian and

Nigerian SCD patients respectively,16,18 till date, the influence of Hp polymorphism on

the phenotypic expressions of the SCD patients in the two countries has not yet been

studied. A study of Haptoglobin genotypes’ behaviour among SCD patients will help to

further understand how this parameter influences the clinical outcome of SCD bearing in

mind that haptoglobin is now being used for acute severe haemolytic conditions in some

parts of the world19. This will help to answer the question as to whether the proposition

for its use as an adjunct therapy in the treatment of SCD is justifiable20.

1.2.2 Hemopexin (HPX): Hemopexin is a heme-binding plasma glycoprotein

which is the second line of defense against hemolysis mediated oxidative damage by

mopping up the liberated free plasma heme. It is produced primarily in the liver but in

addition, other parenchymal cells produce it. The heme-hemopexin complex that is

formed is subsequently delivered to the liver cells through receptor- mediated endocytosis

after which the heme is taken off and the HPX is recycled. Decreased plasma level of

HPX indicates higher degree of hemolysis from RBC destruction reflective of increased

hemolysis severity in patients with associated heme toxicities and/or SCD19,21,22,23. To this

end, researchers are currently exploring therapeutic roles for hemopexin infusion in

SCD19,21,22. Hence, further studies are needed to throw more light on the influence of

hemopexin on the clinical expression of SCD in order to justify or dispel the need to

incorporate hemopexin therapy into the care of SCD patients.

22

1.2.3 Lactate dehydrogenase (LDH): Lactate dehydrogenase is an intracellular

enzyme that abounds in many cells and tissues of human body where it plays crucial roles

in generating energy for the cells through two key processes (glycolysis and

gluconeogenesis)24. Different isoenzymes of it have been recognised based on their cells

or tissues of origin. LD1 and LD2 are primarily derived from the red blood cells, heart

and the kidney while LD3 is mainly from lymphoid cells and platelet. LD4 and LD5 are

primarily from the liver and skeletal muscles. The concept about the mechanisms of LDH

as marker of SCD severity is still debatable. While some authors have argued that the

elevated levels of LDH in SCD is due to hemolysis24,25, others have linked the elevation

to tissue destruction26. However, more studies are in agreement with the hemolysis theory

as the primary source of the LD1 and LD2 isoenzymes in patients with SCD as they

strongly correlate with known markers of hemolysis in these patients24,25,27. As a result of

this, LDH serum levels especially the LD1 & LD2 isoenzymes, are now been considered

useful markers of intravascular hemolysis and disease severity in patients with

SCD24,25,27. It is now being considered as parts of routine tests for SCD in most developed

parts of the world as markers of SCD severity based on new findings that it also correlates

well with other complications of SCD9,24,25,27. Kato et al24, found high correlation of

elevated LDH with ntiric oxide insufficiency, increased rates of leg ulcers, pulmonary

hypertension, priapisms and deaths among SCD patients in the USA. Similarly, Mikobi

et al28 found increased disease severity among SCD patients with elevated LDH in Congo.

But of more interest is the suggestion by Mecabo et al27, that LDH levels in SCD could

be used to monitor response to hydroxyurea following their study of LDH behaviour

among the Brazilian cohorts of SCD patients. However, due to lack of resources and

tools, testing for LDH is seldomly performed in most developing countries of Africa

including Nigeria28,29. This is a paradox given the huge burden of SCD in these parts of

the world. Drawing from these, the yield of information that could be obtained from such

studies is therefore huge. Hence, further correlational studies on LDH could unravel more

potentials for this marker in the various aspects of care for the SCD patients and help to

further clarify its roles in SCD expression.

1.2.4 Red blood cell derived microparticles (RMP): These are small

biologically active plasma membrane vesicles released by red blood cells into the blood

stream either as a result of their aging and self preservation processes or as a result of

cellular injuries and destruction of the red blood cell30,31. They are generally less than

23

1µm in size and they have been implicated in some pathological processes where they

exert sundry roles which include: immune modulation, transfer of messages between

cells, activation of coagulation cascade and endothelial injury30,31. In SCD, levels of RMP

have been shown to be elevated both in steady state and in vaso-occlusive crisis31,32. In

addition, they concentrate plasma heme and transfer it to vascular endothelium where it

mediate oxidative stress, vascular dysfunction and vaso-occlussions in SCD31,32,33. Also,

the degree of RMP elevation has also been speculated to be closely related to known

markers of SCD complications/severity like nitric oxide (NO) depletion, generation of

thrombin, rise of plasma free plasma Hb, and increased rate of intravascular

hemolysis30,31,32,33. These deleterious synergies between the RMP and known markers of

SCD disease severity raises a possibility of using the RMP levels of SCD patients to

categorise them into clinical sub-phenotype groups and possibly prognosticate the disease

outcome. Most current studies on RMP in SCD were conducted on patients in the

developed parts of the world. A further study of RMP in SCD patients will help to

explore more on the roles of this biomarker in the clinical expressions of SCD patients.

1.2.5 Free plasma haemoglobin: Although majority (≥70%) of total hemolysis

in SCD occurs in the extravascular space (monocytic-phargocytic systems in spleen and

liver), while approximately ≤30% takes place within the intravascular compartment. In

SCD patients, polymerization of HbS leads to the destabilisation of the RBC membrane

and excessive premature destruction of the erythrocytes. The rate of destruction can be

up to 10% of their total erythrocytes in every 24 hours19,23,34. This process can lead to the

release of as much as 30g of free (decompartmentalised or plasma hemoglobin) per day

and this amount is enough to saturate the endogenous scavenging mechanisms comprising

the plasma haptoglobin and CD163 scavenging receptor. Hence, may result in substantial

amount of free circulating plasma Hb19,23. Free plasma Hb has been found to significantly

scavenge the nitric oxide (NO) levels because of its high affinity for NO. An experimental

study has observed rapid depletion of NO levels in SCD patients by up to 1000 fold by

free plasma Hb and noted that the release of plasma free haemoglobin and its role in

depleting the NO levels may be the main reason for the vascular complications seen in

SCD34. In addition, there is a catastrophic synergy in NO depletion between free plasma

Hb and arginase enzyme released by destroyed RBC during hemolysis in SCD

patients35,36. Nitric oxide is required for the maintenance of vascular integrity and

relaxation to prevent vascular events in SCD patients24,34,35,36. These roles have been

24

found to be totally abrogated when the plasma free haemoglobin level rises up to 6µM

leading to serious vascular complications like systemic and pulmonary hypertensions24,34.

In view of this, some authorities have hinted on the possible need to administer

Haptoglobin (the scavenger of free plasma Hb) to SCD patients in order to ameliorate the

disease complications19,20,23. Therefore, there is need for more studies on the free plasma

Hb of SCD patients in order to unravel more of its roles in the pathophysiology of SCD

as this could help to further strengthen or dispel its roles in the sequaelae of SCD.

1.2.6 Heme: Heme is synthesised in all human cells including the RBC. This

process involves eight enzymatic reactions that take place either within the cells’

mitochondria or their cytosols37. Upon formation, the heme is maintained in constant

equilibrium within the RBC through three regulatory mechanisms. These include

diffusion though the cell membrane, binding to the RBC cell membrane or cytoskeleton

and intracellular degradation by glutathione38. The first two mechanisms are

concentration dependent while the third mechanism requires the presence of adequate

glutathione at a concentration not ˂ 2mM within the RBC otherwise, excess heme could

accumulate and or escape to injure the RBC or other cells38. However, as the RBCs

become senescent they become denatured and some of them undergo autoxidation to

methemoglobin leading to the release of free heme into the plasma. In SCD patients, it

has been shown that sickled HbS has an exaggerated rate of autoxidation compared to

normal HbA leading to excess free heme in these patients38. In addition, the high rate of

destruction of sickled RBC also contributes to excessive amount of free heme in them38.

Free heme has been implicated in the pathophysiology and clinical expression of SCD

through two key methods. Firstly, it is implicated as a co-factor in the promotion of HbS

polymerization (the primary event in the pathophysiology of SCD) causing cascades of

pathological processes. Secondly, it has also been implicated in damaging the RBC

membrane thereby contributing to the increased rate of endothelium adhesion, hemolysis,

RBC removal by monocytic-phargocytic system and these lead to shorter life span of

RBC in SCD patients38. Recently, some researchers found that administration of

hemopexin (a known heme scavenger), prevents these unwanted events in animal

models22. This observation has led to others proposing the possibility of administering

hemopexin to SCD patients in order to ameliorate their disease19. It is therefore pertinent

that, more studies be conducted on SCD patients to further highlight the roles of plasma

free heme as a marker of disease severity in SCD. The major complications attributed to

heme and plasma hemoglobin in the SCD hemolysis phenotypes appear to be mediated

25

through complex and mixed processes involving vascular injuries, inflammation

promotion, oxidative injuries, and networking to signal the activation of other mediatory

molecules like the toll like receptors39-42. Given the impacts of these effects on SCD

patients, researchers are now suggesting that using the scavengers of heme and plasma

hemoglobin on patients with SCD could ameliorate their disease20,42. This raise the need

for more studies to further explore the relationships between these markers as well their

contributions to the clinical manifestations of SCD.

GENETIC MARKERS

1.2.7 Thalassemia syndromes: The thalassaemia syndromes are a group of

inherited disorders in which there is absence or reduced rate of synthesis of one of the

globin chains: either the alpha or the beta-globin chains of hemoglobin A (α2β2), which

is the major human adult haemoglobin approximately 97%. The other minor adult

haemoglobins are haemoglobin A2 (α2δ2) comprising 2.5% and fetal haemoglobin (α2γ2)

comprising less than 1% of the total hemoglobins. The involvement of either globin chain

causes imbalanced globin chain production, ineffective erythropoesis, microcytosis,

hemolysis and various degrees of anaemia43,44. The alpha (α) thalassaemia is usually due

to gene deletion, and is caused by loss of one (αα/-α) or two (-α/-α) of the normal

complement of four alpha genes (αα/αα) leading to a decrease in or absence of α chains.

Alpha thalassaemia is common in Africa where it is believed to offer some protection

against malaria in addition to ameliorating some clinical features of SCD patients43. The

beta (β) thalassaemia is usually due to a point mutation, may be beta (βo) or beta (β+). A

βo thalassaemia is one in which there is no β gene expression and, therefore, no β chain

synthesis and consequently no haemoglobin A (HbA) production. A β+ thalassaemia is

one in which the β gene is expressed but at a reduced rate so that there is some β chain

synthesis and some production of HbA. Beta thalassaemia trait has been found among

2.1% of Nigerian SCD patients44. Beta thalassaemia has been classified into three main

clinical subtypes based on the severity of the clinical manifestations and these include:

(a) Beta thalassaemia major otherwise known as Cooley’s anaemia in which case patients

have severe symptoms and blood transfusion dependent. (b) Beta thalassaemia trait this

is direct opposite of Cooley’s anaemia here patient is usually not symptomatic and the

diagnosis is usually made by chance when patients manifest very mild symptoms upon

the exposure of the patients to extreme stress. (c) Beta thalassaemia intermedia represents

26

the group of patients between the first two groups. They have mild to moderate

intermediate symptoms and are not blood transfusion dependent. The combination of

sickle cell mutation and beta-thalassaemia mutation gives rise to varying heterogenous

groups known as Hb S/β thalassaemia (Hb S/β-Thal). Patients with Hb S/β-Thal exhibit

heterogeneity in their clinical manifestations ranging from nearly asymptomatic state to

severe symptoms depending on the types of mutations involved and the amount β globin

synthesis which is reflected by the amount of HbA produced45. Currently, there is no

consensus about the classification of Hb S/β-Thal. However, two types namely: (Hb S/βo-

Thal and Hb S/β+-Thal) have been recognised generally45. The Hb S/βo-Thal has no

production of HbA and its clinically indistinguishable from sickle cell anaemia.

Therefore, it tend to exhibit a more severe course compared to the Hb S/β+-Thal in which

there is production of variable amounts of HbA which dilute the polymerisation of HbS

thus resulting in milder phenotypic expression. It is however important to note that the

influence of other genetic modifiers of SCD also affect this phenotypic expression45. To

this end, while increased hypercoagulability have been associated with coinheritance of

thalassaemia with SCD46, others have found very mild SCD phenotypic expressions with

Hb S/β+-Thal47 and alpha thalassaemia43 respectively. These observations suggest that the

exact pathophysiologic mechanisms and or influence of the thalassaemias on the

phenotypic expression of SCD are not yet fully elucidated hence, the need for more

studies.

1.2.8 Fetal hemoglobin (HbF) AND BCL11A ,HBS1L-MYB polymorphisms:

HbF has emerged as a central modifier of the several phenotypes like anemia, stroke,

infections and others seen in SCD patients48,49,50. The presence of HbF limits the rate of

polymerisation of HbS which is the primary event that leads to the cascades of both

pathological and clinical manifestations in SCD48,49,50. Luckily, the expression of HbF is

amenable to therapeutic manipulation. To this end, clinicians have used hydroxyurea to

increase the proportion of HbF in SCD patients with the aim of making the patients benefit

from its ameliorative effects on their disease phenotypes51,52. Interestingly, genetic

variations at three principal loci (BCL11A, HBS1L-MYB and HBB) each on chromosomes

2, 6 and 11 respectively, have been shown to account for increased expression of 10-20%

HbF variation among SCD patients in the USA, Brazilian and United Kingdom50. Of

these, only the BCL11A locus has been studied among the African SCD patients. Initial

studies in Tanzania 53 and Cameroon 54,55 have shown that single-nucleotide

27

polymorphism (SNPs) in the BCL11A are prevalent among SCD patients in both

countries with significant association of these SNPs with HbF. Researchers have hinted

on the high degree of variations in the micro allele frequency (MAF) of SNPs of other

genetic variants among the African SCD patients thus making extrapolation of findings

from one African country to another difficult50. These findings suggest that studies of

multiple SCD populations are needed especially from Africa, to further improve the

understanding of the impact of human diversity on HbF expression in SCD.

1.2.9 Glucose 6 phosphate dehydrogenase (G6PD): G6PD is a cytosolic

enzyme in the pentose phosphate pathway which supplies reducing energy to cells by

maintaining the level of the co-enzyme nicotinamide adenine dinucleotide phosphate

(NADPH). G6PDH reduces nicotinamide adenine dinucleotide phosphate (NADP) to

NADPH while oxidizing glucose-6-phosphate (G6PD)56. Humans with a genetic

deficiency of G6PD are predisposed to non-immune hemolytic anaemia and this

deficiency is highly prevalent in sub-Saharan Africa where it is believed that both HbS

and G6PD deficiency confer partial protection against malaria hence, its predominance

in Africa where malaria is endemic57,58. The prevalence of G6PD among the Nigerian

population is between 15-25%59 while that among the Brazilian population varies

between 1-13% depending on the region of the country.56 The most prevalent variant in

the two countries is the G6PD A-variant56,59. There have been conflicting reports about

the genetic polymorphisms of G6PD A-variants in Africa 60,61,62. De-Araujo et al60 and

Capelini et al61 found a predominance of G6PD genotypes with G6PD 202A and G6PD 376G

alleles in Senegal and other parts of Africa similar to the genotypes in Brazil56. However,

Clark et al62 found a predominant of G6PD968C and G6PD376G alleles among the Gambian

population also from the West Africa sub-region like Senegal. This suggests

heterogeneity in the patterns of G6PD A-variants among the African population and this

factor could affect the clinical events in patients with G6PD deficiency. Till date, only

few studies investigated the potential effects of G6PD deficiency on SCD and the studies

have shown conflicting reports. While studies from Saudi Arabia and Burkina Faso, found

no effect of G6PD deficiency on the clinical manifestations and laboratory parameters of

SCD patients63,64, studies from the USA and France found association with increased

cerebral blood flow velocity, increased rate of acute anaemic events, blood transfusions,

vaso-occlussive crisis, and decreased steady state haemoglobin levels58,65,66. The mixed

reports from these studies concerning the effects of G6PD deficiency co-inheritance on

28

the clinical events or laboratory parameters of the SCD raise the need for further studies

to dispel or establish if the co-inheritance impact any influence on SCD manifestations.

1.2.10 Uridine diphosphate glucuronyltransferase 1A (UGT1A)

Polymorphism: Patients with Gilbert's syndrome (GS) have a defect in the uridine

diphosphate glucoronosyl transferase 1A gene that encodes for uridine diphosphate

glucuronyltransferase 1A (UGT1A) enzyme which is the main enzyme responsible for

bilirubin conjugation among the uridine diphosphate glucuronosyl transferase (UGT)

family67,68. Polymorphisms in the UGT1A gene results in a 60-70% reduction in the

enzyme and by extension, the liver's ability to conjugate bilirubin leading to unconjugated

hyperbilirubinemia.69 Bilirubin is an endogenous antioxidant and some authors have

postulated that, Gilbert's syndrome may actually reduce the risk of various age-related

diseases because of the antioxidant properties of bilirubin67,69. One recent study found

that mortality rates observed for people with Gilbert's syndrome in the general population

were shown to be almost half those of people without evidence of Gilbert's syndrome70

and the occurrence of GS had also been thought to give protection to Africans against

malaria among other benefits71. Reports on co-inheritance of SCD with GS have shown

increased rates of cholelithiasis, need for cholecystectomy and increased morbidity72,73,74.

In addition, the response to hydroxyurea treatment was found to be blunted by the

presence of GS in some SCD patients75. These further highlight the deleterious roles GS

could play in modulating the SCD phenotypic expression. Studies on GS behaviour

among Africans and Brazilians are scanty and no specific study has been found from

Nigeria. However, a brief report by Fertrin et al76 found increased rate of

hyperbilrubinemia and possible risk for early onset of symptomatic gallstone disease that

may warrant cholecystectomy among Brazilian SCD patients. These observations raise

the need for more studies to explore the roles of GS in the clinical expression of SCD

patients.

29

2. JUSTIFICATION

Although several hypotheses have been proposed for the phenotypic diversity found

in sickle cell disease, these factors do not fully explain the clinical heterogeneity observed

in SCD patients. Therefore, efforts are on-going in search of new markers that can be

used to better characterize sickle cell disease. The present study aims to establish a link

between markers of hemolysis and some genetic modifiers, and clinical phenotypes of

patients with sickle cell disease. These may be useful for predicting severity as well as

influencing therapeutic decisions in SCD.

30

3. OBJECTIVES

3.1. General objective

To determine the influence of some hemolysis and genetic markers on the

phenotypic expression of patients with sickle cell disease.

3.2 Specific objectives are to:

• Quantify red blood cell microparticles, plasma hemoglobin, heme, hemopexin,

and haptoglobin in plasma samples from Brazillian adult patients with sickle cell

disease treated with or without hydroxyurea, and control subjects

• To correlate the clinical, hematologic and biochemical results of patients with

sickle cell disease with the above quantified markers.

• To evaluate the influence of hydroxyurea treatment on the levels of free plasma

hemoglobin, heme, hemopexin, haptoglobin, and red blood cell microparticles of

SCD patients.

• Identify and try to establish any association between the polymorphisms of G6PD,

UGT1A1, BCL11A, haptoglobin genes and alpha thalassemia trait with clinical

events and laboratory markers of Nigerian children with SCD.

31

4. HYPOTHESIS

Some recent studies have shown that hemolysis and genetic markers play important

roles in the progression of sickle cell disease. The aim of this study is to establish whether

some hypothesized biologic and genetic markers have any link with the laboratory and

clinical expressions of patients with sickle cell disease as these may be useful in predicting

severity as well as influencing therapeutic decisions in SCD.

32

5. PATIENTS AND METHODS

5.1. Type of study

This was a cross-sectional descriptive study

5.2. Place of research

The research was carried out at the Hemoglobin and Genome Laboratory of the

Hematology and Hemotherapy Center of UNICAMP and the pediatrics hematology unit

of the Ekiti State University Teaching Hospital (EKSUTH) Ado Ekiti, Ekiti State Nigeria.

Institutional ethical approvals were obtained (EKSUTH/A67/2016/03/003 and

UNICAMP CAAE 54031115.9.0000.5404).

5.3. Selection of patients and control subjects

The study included patients with sickle cell disease treated at the hematology and

hemotherapy centre, UNICAMP and EKSUTH with diagnostic testing by high pressure

liquid chromatography (HPLC) (Bio-Rad, Hercules, CA, USA) and genetic studies.

UNICAMP

For the hematology and hemotherapy centre arm of the study, the study participants

included patients of both sexes, aged 18 – 60 years and are on regular follow up. The

patients were stable state, i.e, absence of any acute event such as painful crisis, and or

infection, within a month and without blood transfusion for at least three months to

recruitment period. 12mL of each participant's blood were collected for plasma separation

and other analyses.The control group consisted of healthy volunteers, and staff members

at the Hemoglobin and Genome Laboratory of the Hematology and Hemotherapy Center

of UNICAMP, aged 18-60 years. All the controls had HbAA genotype and this was

confirmed by high performance liquid chromatography (HPLC). The determination of

clinical sequale and complications of sickle cell disease in all patients, were as previously

described by Ballas et al77

EKSUTH

The EKSUTH participants consisted of children and adolescents diagnosed with

SCD and aged 2 -21 years of both sexes in steady state. Children who accompanied their

siblings to the EKSUTH to the outpatients` well-child clinic served as controls.

33

To qualify for inclusion, the SCD patients must have been on regular follow up at

the pediatric hematology unit. The determination of clinical sequale and complications of

sickle cell disease in all patients, were as previously described by Ballas et al77.

5.4. Exclusion Criteria:

1. Lack of consent of the patient, or parents or accompanying family members /

caregivers.

2. Patients in crisis or pain crisis in the last month before the study.

3. Patients hospitalized for any medical condition in the last three months prior to the

study.

4. Patients on chronic blood transfusion or those who have received blood transfusion in

the last three months.

5. Patients with other chronic medical conditions.

6. Pregnancy

5.5. Sample size:

A total of 358 people participated in the study. They comprised of 138 patients

with SCD and 39 controls from hematology and hemotherapy centre, UNICAMP, Brazil

and 110 children with SCD alongside 68 controls from the EKSUTH in Nigeria.

5.6. Study Procedure and data collection

After obtaining the consent of participants, parents and or assent as applicable,

peripheral venous blood samples approximately 12 mL (4ml into EDTA tube for

hematological and DNA studies, 4ml into EDTA tube for plasma extraction, and 3.5ml

into Sodium citrate tube for red blood cell microparticle studies), were collected by

venipuncture from the patients.

Blood samples were processed to obtain plasma and stored in freezer at -80 ° C until

analysis as applicable.

5.7. Clinical history and events

A detailed history with special attention to the frequency of acute events such as

pain episodes, and acute chest syndrome in the past 12 months were obtained from the

records. In addition, frequency of other chronic events complicating SCD such as,

cerebrovascular disease, avascular necrosis of the head of femur, leg ulcers, priapism,

34

SCD retinopathy, pulmonary complications and kidney disease, were verified through

past hospital records, and were defined according the SCD co-operative study group77.

A brief description of few of the events by Ballas et al are as follows77.

1. Painful crisis: Acute significant painful episode will be described as painful

event(s) requiring a hospital visit or disruption of normal daily activities and

requiring the use of oral or parenteral analgesics.

2. Avascular necrosis: This will be defined as osteonecrosis or aseptic necrosis of

the head of femur or humerus confirmed by radiography as irregularity of the

articular surfaces of the head of femur/humerus

3. Severe bacterial infections: These will be defined by the presence of one or more

of pneumonia, sepsis, meningitis, osteomyelitis, septic arthritis, confirmed by

positive blood culture and or radiograph as appropriate.

4. Acute chest syndrome: An acute illness characterised by fever, and respiratory

symptoms (dyspnea, chest pain) accompanied by low pulse oxymetry and new

pulmonary infilterates on chest radiodragh

5. Stroke (cerebro-vascular disease): Acute neurologic symptoms or signs

secondarytoocclussion of and orhemorrhage from cerebral vessels confirmed on

computerized tomography (CT) scan or magnetic resonance imaging (MRI)

6. Chronic leg ulcer: Ulceration of the skin and underlying tissue of the lower

extremeties, especially the media or lateral surface of the ankle.

7. Priapism: Sustained, unwanted and painful penile erection.

8. Cholelithiasis: Confirmed cholelithiasis on abdominal USS with or without

abdominal pain.

However, due to the lack of right heart catetherisation to diagnose pulmonary

hypertension the tricuspid regurgitation velocity (TRV) obtained from Doppler

echocardiography was used to categorise patients into two groups based on risk

for pulmonary hypertension. Patients with TRV < 2.5m/sec were categorised as

having normal TRV and those with TRV ≥2.5m/sec were categorised as having

elevated TRV and at risk of pulmonary hypertension78. Urinary albumin-

creatinine ratio (UACR) of less than 30mg/g was taken as no albuminuria (i.e.

normal), between 30mg/g and 300mg/g as microalbuminuria and greater than

300mg/g was classified as macroalbuminuria79. Patients were also classified into

35

two groups based on whether they had presence or absence of proliferative SCD

retinopathy.

Routine laboratory investigations: All laboratory tests were done by standard

procedures.

5.8. Haematological tests:

These included haematocrit; haemoglobin concentration; WBC - total and differential

counts; MCV; MCHC, RDW, reticulocyte count and platelet count and fetal

Haemoglobin level measurements using automated hematology analyzer and HPLC

respectively.

5.9. Biochemical tests: These were done by standard laboratory procedures and they

included serum bilirubin – total and unconjugated, lactate dehydrogenase (LDH) and

urinalysis to determine microalbuminuria.

5.10. Colorimetric Assays:

5.10.1. Plasma hemoglobin: This was done on frozen plasma samples with the

QuantiChromTM Hemoglobin Assay Kit (DIHB-250), BioAssay Systems

USA. This assay is based on an improved Triton/NaOH method, in which

the haemoglobin present in a sample is concerted to a uniform coloured end-

product, and the intensity of colour measured at 400nM is directly

proportional to the haemoglobin concentration present in the plasma80.

5.10.2. Heme: This was done on frozen plasma samples with the QuantiChromTM

Heme Assay Kit (DIHM-250), BioAssay Systems USA. The Assay Kit is

based on an improved aqueous alkaline solution method, in which the heme

is converted into a uniform coloured form. The intensity of colour, measured

at 400 nm, is directly proportional to the heme concentration in the sample81.

5.11. ELISA Tests:

5.11.1. Hemopexin: This was done on frozen plasma samples with the Abcam

Human Hemopexin (ab171576) ELISA kit. This assay employs an affinity

tag labelled capture antibody and a reporter conjugated detector antibody

which immunocaptured the samples analyte in solution. This entire complex

36

(capture antibody/analyte/detector antibody) is in turn immobilised via

immunoaffinity of an anti-tag antibody coating the well. Briefly, to perform

the assay, a Hemopexin specific antibody has been precoated onto 96-well

plates. Standards or test samples were added to the wells and subsequently

biotinylated Hemopexin (antibody mix) was added and incubated on a plate

shaker then followed by washing with wash buffer. After washing away any

unbound substances, an enzyme linked polyclonal antibody specific for

hemopoxin was added to the wells and a substrate solution was then added to

the wells and colour change develops in proportion to the amount of

hemopexin bounded by in the initial step. The development of colour was

stopped by adding stop solution. The density of yellow coloration was

proportional to the amount of Hemopexin captured in the plate. The intensity

was measured at 450nM82. The minimal detectable limit of the kit is

1.44ng/ml.

5.11.2. Haptoglobin: This was done on frozen plasma samples with the Quantikine

Human Haptoglobin (DHAPGO) ELISA kit, R & D Systems USA. The assay

employs the quantitative sandwich enzyme immune assay technique in which

a Haptoglobin specific antibody has been precoated onto 96-wellplates and

immobilised. Standards or test samples were added to the wells and any

Haptoglobin present was bounded by the immobilised antibody. After

washing away any unbound substances, an enzyme linked polyclonal

antibody specific for haptoglobin was added to the wells. This was followed

by a repeat wash to remove any unbound antibody-enzyme reagent, a

substrate solution was then added to the wells and colour develops in

proportion to the amount of haptoglobin bounded by in the initial step. The

development was stopped by adding stop solution. The density of coloration

was proportional to the amount of Haptoglobin captured in the plate. The

intensity of the colour was measured at 450nM83. The minimum detectable

limit of the kit ranged from 0.031 – 0.529ng/ml and the mean minimal

detectable limit is 0.192ng/ml.

5.12. Flow Cytometry:

5.12.1. Red blood cell microparticles quantification: Flow cytometry was used

to characterise and quantify the red blood cells microparticles (RMP). The

processes involved differential centrifugation followed by staining and

37

fluorescence flow cytometry analysis as previously described84. The

processes are briefly described below.

5.12.2. Sample collection and plasma separation for RMP analysis: 3.5ml of

citrated venous sample collected by 21gauge syringe were drawn from each

participants after taking the first 8ml of blood into two EDTA sample tubes

(4mL each) for plasma separation and the other for hematological analysis

with very light tourniquets applied. The citrate samples were not mixed or

rocked and were transported securely in upright positions inside the sample

carrier to the laboratory immediately and the plasma separation was done

within 30 minutes of collection. To obtain the plasma, the topmost 1ml of the

citrated whole blood sample was pipetted out and discarded and the remaining

was processed through a two stage centrifuging, the first at 2,500g x 15min

at 22 °C thereafter the topmost 1ml of the plasma from this first centrifugation

was pipetted out and the remaining discarded. A second centrifugation at

13,000g x 5mins at 22 °C was done for the topmost 1ml plasma obtained from

the first centrifugation. The topmost 700µl of the product of the second

centrifugation was pipetted out into a separate polypropylene tube and the

platelet count of the plasma was checked using Beckman Coulter hematologic

counter (Model 8246-EN, SN- AN37824), USA, to ensure it was platelet

depleted. Thereafter, the samples were aliquoted into polypropylene tubes

and frozen at -80 °C until analysis.

5.12.3. Red blood cell microparticle staining and quantification: The staining

reagents (a. Calcein violet AM (Molecular probes-Invitrogen: 3,125µg (5µL),

b. Bovine Lactadherin FITC (Haematologic Technologies, Inc. 0.83µg

(10µL), c. Anti-CD235a R-PE (Life Technologies: 2µL), d. Sterile filtered

PBS 2 x 0.22 µm membrane) and aliquoted frozen platelet poor plasma were

brought to room temperature. The reagents were prepared according to

manufacturers` specifications. The antibodies (Calcein AM, Anti-CD235a

and Lactadherin were also subjected to a high speed centrifugation at 20,000g

x 5minutes at room temperature to remove false positive events at analysis.

After preparing fresh polypropylene tubes, 10µl of each sample was stained

with 5µL, 10µL, 2µL & 5µL of calcein, lactadherin, anti-CD235a and filtered

sterile PBS buffer respectively. The resultant solution was gently vortexed

and incubated in the dark at 37 ° C, -5%CO2 for 60mins initially. Thereafter,

38

500µL of sterile filtered PBS was added and then incubated for the second

time in the dark at 37 ° C, -5%CO2 for 30mins. Immediately after the second

incubation, the resultant solution was further diluted with 3,468µL of sterile

filtered PBS to make a final dilution of 1:400 with a resultant solution totaling

4000µL. This resultant solution was immediately analysed on a calibrated

flow cytometer, the Beckman Coutler`s CytoFLEX Flow Cytometer, (Model

A00 -1-1102) USA, using the staining buffer as negative control. Each sample

was aspirated and read for 10mins by the Cytoflex machine. The RMP events

were expressed per mL.

5.13. Genetic Studies:

The DNA of the Nigerian patients extracted from each participant by Qiagen

QIAamp DNA (Blood Mini Kit Cat No. 51104 Germany), was used to confirm the

diagnosis of SCD by polymerase chain reaction (PCR) at the genetic laboratory at the

Hematology and Hemotherapy centre, Hemocentro-Universidade Estadual de Campinas

(UNICAMP) Brazil. All DNA studies were carried out blinded regarding the clinical and

laboratory parameters of the participants.

5.13.1. Alpha thalassemia determination: Alpha-thalassemia (α3.7Kb deletion)

was investigated by GAP-PCR according to Dodé et al., 1993. The PCR was

performed in 25 μL reaction volume containing 100ng of DNA sample; 1X

α- Buffer (Tris-HCl 2M (pH 8.6), (NH4)2 SO4 1M, MgCl2 1M, Na2EDTA

0.2M, BSA and β-mercaptoetanol 14.3M); 1X DMSO; 0.3mM of dNTP mix;

0.2 µM of each primer (C2:CCATGCTGGCACGTTTCTGAandC10:

GATGCACCCACTGGACTCCT); 1U of GoTaq® Flexi DNA Polymerase

(Promega Corporation, Madison, USA). Thermal cycle conditions were as

follows: preheating at 94°C by 5 minutes, followed by 35 cycles of 94°C for

45 seconds, 56°C for 1 minute, and 72°C for 2 minutes and a final extension

at 72°C for 7 minutes was performed. After electrophoresis in a 1.2% agarose

gel a fragment of 2.1Kb could be observed for normal alleles and 1.9Kb

fragment for deleted alleles (-α3.7Kb).

Alpha-thalassemia (α4.2Kb deletion) was investigated by Multiplex-PCR.

The PCR was performed in 25 μL reaction volume containing 100ng of DNA

sample; 1X α- Buffer (Tris-HCl 2M (pH 8.6), (NH4)2 SO4 1M, MgCl2 1M,

39

Na2EDTA 0.2M, BSA and β-mercaptoetanol 14.3M); 1X DMSO; 0.3mM of

dNTP mix; 0.4 µM of primer (P71: TACCCATGTGGTGCCTCCATG and

0.3 µM of each of primer P72:TGTCTGCCACCCTCTTCTGAC and P52:

CCTCCATTGTTGGCACATTCC; 1U of Taq DNA Polymerase (Invitrogen,

Carlsbad,CA). Thermal cycle conditions were as follows: preheating at 94°C

by 5 minutes, followed by 35 cycles of 94°C for 45 seconds, 60°C for 1

minute, and 72°C for 2 minutes and a final extension at 72°C for 7 minutes

was performed. After electrophoresis in a 1.2% agarose gel a fragment of

1596 bp could be observed for deleted alleles (-α4.2Kb) and 233 bp as an

internal control to verify the quality of DNA sample.

5.13.2. Uridine diphosphate glucuronosyl transferase 1A (UGT1A1)

promoter polymorphism (rs8175347): DNA samples were used for

genotyping of the (TA)nTAA UGT1A1 promoter polymorphism (GenBank

accession NG_002601). The rs8175347 identification was performed by

Polymerase Chain Reaction (PCR) using a forward primer 5'- (6-FAM)

labelled (*) for detection by fragment analysis in capillary electrophoresis

system. The PCR reaction was prepared in 30 µL volume with 100ng of

genomic DNA; 1X Reaction Buffer (BIOTOOLS B&M Labs, Spain);

2.16mM MgCl2; 1.33 mM of dNTP mix; 133 nM of each primer (Integrated

DNA Technologies, Coralville, Iowa) named: UGT1A1_*F:

GTCACGTGACACAGTCAAAC and UGT1A1_R:

CAACAGTATCTTCCCAGCATG; and 1 U Taq DNA Polymerase

(BIOTOOLS B&M Labs, Spain). Thermal cycle conditions were as follows:

preheating at 96°C by 2 minutes, followed by 25 cycles of 96°C for 30

seconds, 58°C for 40 seconds, and 72°C for 40 seconds. An ended step at

72°C for 30 min was performed to promote adenylation of the PCR products.

The PCR product (1 μL) was added to 8.7 μL Hi-Di Formamide (Applied

Biosystems, Carlsbad, CA) and 0.3 μL of a GeneScan™ 500 LIZ™ size

standard (Applied Biosystems, Carlsbad, CA) and the fragments ranged from

197 - 203 bp, corresponding to (TA)5 - (TA)8 repeats, respectively, were

separated by capillary electrophoresis on a ABI3500 Genetic Analyzer and

analysed by Gene Mapper v 4.1 Software (both Applied Biosystems,

Carlsbad, CA). The UGT1A1 genotypes were further classified into three

40

subgroups namely: low, intermediate and high activity group based on their

activities and number of TA repeats.

5.13.3. Haplotypes (β globin chain): The βs-haplotypes were determined by by

TaqMan SNP genotyping assay of three single nucleotide polymorphisms

(SNPs); Xmn I-rs7482144, Hinc II-rs968857, and Hinf I-rs16911905 as

follows.

Haplotypes linked to the βS mutation provided important anthropological

information regarding the multiple origins of the HbS allele. TaqMan®

Genotyping Assays (Applied Biosystems, Carlsbad, CA) were used for

screening of 3 polymorphic sites located at β-globin gene cluster and to

determine the different haplotypes: SENEGAL, BENIN, CAR (Central

African Republic), ARAB-INDIAN, CAMEROON. The assays related to

restriction sites were analysed: Xmn I (customized - rs7482144) at position

- 158 of the γG gene promoter; Hinc II (C_9599121_10 - rs968857 ) located

at 3' region of ψβ gene; and Hinf I (C_32838989_10 - rs16911905) at 5’ to

the β gene. DNA samples were prepared at 50ng/μL of concentration and the

SNP Genotyping assay was diluted to 5X concentrated, then the reaction

mixture contained: 2.5 μL TaqMan Genotyping MasterMix (2X); 1.0 μL of

diluted assay (5X) and 1.5 μL of DNA sample. The reaction was carried out

using allelic discrimination protocol in ABI7500 FAST and StepOne Plus

Real Time PCR systems (Applied Biosystems, Carlsbad, CA), following the

cycling conditions: 30 seconds at 60°C, 10 minutes at 95°C followed by 45

cycles of 15 seconds at 95°C and 1 minute at 60°C, and a final step named

post-PCR at 60°C for 30 seconds with automated allele calling settings for

the SDS 2.1 software (Applied Biosystems, Carlsbad, CA). The figure below

shows how the haplotypes could be identified.

41

Figure 3: The βs-haplotypes.

5.13.4. Glucose 6 phosphate dehydrogenase deficiency (G6PD): The

Identification of mutations: G6PD G202A - African (rs1050828), G6PD

A376G-(rs1050829), G6PD A542T-(rs5030872), G6PD G680T-

(rs137852328) and C563T - Mediterranean (rs5030868) was as follows.

TaqMan® Genotyping Assays (Applied Biosystems, Carlsbad, CA) were

used for screening of African and Mediterranean mutations of G6PD,

respectively identified as rs1050828 and rs5030868 DNA samples were

prepared at 50ng/μL of concentration and the SNP Genotyping assay was

diluted to 5X concentrated, then the reaction mixture contained: 2.5 μL

TaqMan Genotyping MasterMix (2X); 1.0 μL of diluted assay (5X) and 1.5

μL of DNA sample. The reaction was carried out using allelic discrimination

protocol in ABI7500 FAST and StepOne Plus Real Time PCR systems

(Applied Biosystems, Carlsbad, CA), following the cycling conditions: 30

seconds at 60°C, 10 minutes at 95°C followed by 45 cycles of 15 seconds at

95°C and 1 minute at 60°C, and a final step named post-PCR at 60°C for 30

seconds with automated allele calling settings for the SDS 2.1 software

(Applied Biosystems, Carlsbad, CA). PCR-RFLP was performed in aleatory

samples to confirm G6PD G202A genotypes. The PCR reaction was prepared

in 25 µL volume with 100ng of genomic DNA; 1X Colorless GoTaq® Flexi

Reaction Buffer (Promega Corporation, Madison, USA); 2 mM MgCl2; 0.2

mM of dNTP mix; 0.2 µM of each primer (Integrated DNA Technologies,

Haplotype Name Xmn I (1)

rs7482144

Hinc II (2)

rs968857

Hinf I (3)

rs16911905

Sequence

Senegal + (A) + (T) + (G) ATG

Arab-Indian + (A) + (T) - (C) ATC

CAR - (G) - (C) - (C) GCC

Benin - (G) + (T) - (C) GTC

CAM - (G) + (T) + (G) GTG

42

Coralville,Iowa) named G6PD-202_F:CAAGGGTGGAGGATGATGTATG

and G6PD-202_R: AACGCAGCAGAGCACAGCAG; and 1U of GoTaq®

Flexi DNA Polymerase (Promega Corporation, Madison, USA). Thermal

cycle conditions were as follows: preheating at 96°C by 2 minutes, followed

by 35 cycles of 96°C for 30 seconds, 60°C for 30 seconds , 72°C for 1 minute

and a final extension at 72°C for 5 minutes was performed. The PCR

amplicon corresponding to 527 bp was digested with the restriction enzyme

Nla III and it was possible to observe the following fragments: 376 and 151

bp for wild allele and 253, 151 and 123 bp for mutant alleles. Identification

of other G6PD mutations: G6PD A376G (rs1050829), A542T (rs5030872)

and G680T (rs137852328) was done by direct Sanger sequencing using an

ABI 3500 Genetic Analyzer (Applied Biosystems) from genomic DNA to

screen the others mutations in the G6PD gene. The region that include the

SNPs G6PD A376G (rs1050829), A542T (rs5030872) and G680T

(rs137852328) was amplified followed these conditions in 30 µL volume:

150ng of genomic DNA; 1X Colorless GoTaq® Flexi Reaction Buffer

(Promega Corporation, Madison, USA); 2 mM MgCl2; 0.2 mM of dNTP mix;

0.2 µM of each primer (Integrated DNA Technologies, Coralville,Iowa)

named G6PD-1_F: ACCTGGCCAAGAAGAAGATCTACandG6PD-1_R:

TGATAGCTCAGACACTTAGGTTTTG; and 1,5U of GoTaq® Flexi DNA

Polymerase (Promega Corporation, Madison, USA). Thermal cycle

conditions were as follows: preheating at 96°C by 2 minutes, followed by 35

cycles of 96°C for 30 seconds, 60°C for 40 seconds , 72°C for 2 minutes and

a final extension at 72°C for 5 minutes was performed. The PCR product

(2273bp) was submitted to sequencing reaction by following conditions: 40ng

PCR product purified by ammonium acetate and ethanol method, 1,0 µL

BigDye Terminator v3.1 Ready Reaction Mix (AppliedBiosystems, Foster

City, CA, USA), 1X BigDye Reaction Buffer, 2µM of each primer, in

separated reaction (G6PD-2F:

GAGAAGCTCAAGCTGGAGGACTandG6PD_Seq_R:

GCAGGACTCGTGAATGTTCTTG). After thermal cycling (preheating at

96°C by 2 minutes, followed by 25 cycles of 96°C for 15 seconds, 58°C for

5 seconds, 60°C for 4 minutes and a final extension at 72°C for 5 minutes)

43

the sequencing reaction product was purified by ammonium acetate and

ethanol method, dried at 65ºC and ressuspending in 10 µL of Hi-Di

Formamide for electrophoresis. Resulting sequence data were compared with

the reference NM_000402.

5.13.5. Haptoglobin Genotypes determination: The haptoglobin genotypes of

the patients were determined according to Santos et al16 technique using

the selective amplification of the different alleles.

5.13.6. BCL11A Polymorphism and HbF quantification: The HbF quantitation

was done using HPLC (Bio-Rad Variant D10, USA) based on

manufacturer`s specifications. The BCL11A polymorphisms of rs

4671393, rs11886868, rs766432, rs1427407, rs 7606173, rs 6706648, rs

7557939, rs 6738440, rs 6732518 and rs 13019832 were determined by

real time PCR as described below.

DNA samples were prepared at 50ng/μL of concentration and the SNP

Genotyping assay was diluted to 5X concentrated, then the reaction

mixture contained: 2.5 μL TaqMan Genotyping MasterMix (2X); 1.0 μL

of diluted assay (5X) and 1.5 μL of DNA sample. The reaction was carried

out using allelic discrimination protocol in ABI7500 FAST and StepOne

Plus Real Time PCR systems (Applied Biosystems, Carlsbad, CA),

following the cycling conditions: 30 seconds at 60°C, 10 minutes at 95°C

followed by 45 cycles of 15 seconds at 95°C and 1 minute at 60°C, and a

final step named post-PCR at 60°C for 30 seconds with automated allele

calling settings for the SDS 2.1 software (Applied Biosystems, Carlsbad,

CA).

5.14. Data analysis: Data were analyzed with the GraphPad Prism version 5.0 statistical

software for windows (San Diego, California, USA) and STATA Statistical Software

release 12. (College Station, TX: STATACorp LP; 2011, USA), using both descriptive

and comparative statistics. The frequencies of variables were described and the

significance of differences between groups of participants was assessed using the

Kruskal-Wallis analysis of variance (ANOVA), chi-square, Mann-Whitney or Fisher`s

exact tests as appropriate. For the correlation studies, data were normalised by taking

44

logarithms of the various variables and correlating them with other variables of interest

using the Pearson correlation and the scatter diagrams were plotted as applicable. The

correlational studies were done with STATA Statistical Software. The logarithm of the

variables was used for the correlations because data were not normally distributed and

had some outliers. The level of significance was set at P < 0.05 for all statistical tests.

45

6. RESULTS

HEMOLYSIS STUDIES

6.1. Parameters in patients and controls

Table 1 shows the summary of participants and parameters (UNICAMP)

Parameters a. SS (N=78)

Median (Range)

b. SBeta0

(N=12)

Median (Range)

P Value

(a vs b)

c. SBeta+

(N=12)

Median (Range)

P Value

(a vs c)

d. SC

(N=36)

Median

(Range)

P Value

(a vs d)

e. AA

(N=39)

Median (Range)

P Value

(a vs b vs c d

vs e)

Age in years 41 (18 – 60) 38 (18 – 60) 0.754 39 (21 – 58) 0.501†

45 (27 – 60) 0.08†

35 (20 – 60) 0.0237

*1

Red blood cell

microparticles

(Events/ML)

200000.0 (0 –

17840000.0)

120000.0 (0 -

440000) 0.168

† 40000 (0 –

160000) 0.0002

† 80000.0 (0 –

1320000.0) 0.0006

† 0.0 (0 – 12000.0)

<0.0001 *2

Plasma Hemoglobin

(mg/dL)

84.6 (28.2 – 257.5) 77.5 (42.0 – 134.0) 0.479†

58.7 (34.0 –

196.0) 0.015

† 63.5 (33.4 –

158.9) 0.0001

† 42.7 (27.1 – 90.0)

<0.0001 *3

Haptoglobin

(ng/mL)

1937.0 (279.0 –

17520.0)

2342.0 (540.0 –

5290.0) 0.939

† 5417.0 (480.0 –

18180.0) 0.04

† 2832.0 (424.0

– 29670.0) 0.0381

† 986000.0 (656800.0

– 2455000.0) <0.0001

*4

Heme (µM) 55.6 (15.8 – 206.0) 53.2 (31.6 – 94.0) 0.479†

45.0 (22.9 –

119.0) 0.048

† 42.4 (19.5 –

165.0) 0.0025

† 28.5 (12.9 – 86.6)

<0.0001 *5

Hemopexin (µg/mL) 469.0 (76.7 – 1937.0) 448.0 (227.0 –

1221.0) 0.542

† 704.0 (127.0 –

1805.0) 0.803

† 761.4 (114.0 –

1930.0) 0.045

† 988.0 (800.0 –

1409.0) <0.0001

*6

HbF (%) 13.1 (1.7 – 31.0) 13.9 (2.8 – 35.2) 0.433†

11.7 (0.6 – 27.9) 0.554†

0.8 (0.2 –

12.7) <0.0001

† 0.2 (0.1 – 0.9)

<0.0001 *7

Hb Conc (g/dL) 8.5 (4.2 – 12.9) 8.5 (6.8 – 9.7) 0.680†

9.8 (8.4 – 11.5) 0.0009†

11.3 (9.3 –

15.2) <0.0001

† 13.5 (11.7 – 16.3)

<0.0001 *8

46

Platelet ( X 103/uL) 339.0 (99.0 – 682.0) 377.0 (172.0 –

755.0) 0.413

† 375.0 (76.0 –

458.0) 0.546

† 264.0 (61.0

557.0) 0.04

† 257.0 (171.0 –

453.0) 0.002

*9

WBC ( X 103/uL) 6.4 (3.2 – 14.0) 7.8 (4.2 – 16.4) 0.960†

5.4 (4.1 – 11.8) 0.227†

7.8 (3.9 –

14.7) 0.039

† 6.8 (4.5 – 11.3)

0.174 *

Reticulocyte count

(x 109/L)

271 (23.2 – 755.0) 264.0 (161.0 –

455.0) 0.794

† 261.0 (52.0 –

561.0) 0.785

† 233.0 (57.0 –

467.0) 0.191

† NA

0.574 *

Total bilirubin

(mg/dL)

1.7 (0.4 – 6.8) 1.4 (1.0 – 5.1) 0.315†

1.2 (0.6 – 3.2) 0.031†

1.1 (0.4 – 2.4) <0.0001†

NA <0.0001

*10

Unconjugated

bilirubin(mg/dL)

1.4 (0.3 – 6.0) 1.1 (0.8 – 4.2) 0.210†

0.8 (0.4 – 2.6) 0.029†

0.9 (0.3 – 1.8) <0.0001†

NA <0.0001

*10

LDH (IU) 384.0 (164.0 – 938.0) 311.0 (195.0 –

531.0) 0.046

† 200.0 (128.0 –

496.0) 0.0001

† 237.0 (136.0 –

642.0) <0.0001

† NA

<0.0001*11

RBC count (×1012/L) 2.3 (1.0 – 4.0) 2.8 (2.3 – 4.2) 0.005†

4.0 (2.8 – 6.1) <0.0001†

4.0 (2.2 – 6.3) <0.0001†

4.8 (4.0 – 5.5) <0.0001

*12

NB: Significant P values are indicated in bold fonts, †-Mann-Whitney Test, RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb

Hemoglobin concentration, WBC-White blood cell count, LDH-Lactate dehydrogenase, *- Kruskal-Wallis Test with Dunn`s multiple comparison post-hoc

tests with differences in *1= SC vs AA only; *2= (SS vs SBeta+), (SS vs SC), (SS vs AA), (SBeta0 vs AA), (SC vs AA); *3= (SS vs SC), (SS vs AA), (SBeta0 vs AA),

(SC vs AA); *4= (SS vs AA), (SBeta0 vs AA), (SBeta+ vs AA), (SC vs AA); *5= (SS vs AA), (SBeta0 vs AA); *6= (SS vs AA), (SBeta0 vs AA), (SC vs AA); *7= (SS vs

SC), (SS vs AA), (SBeta0 vs SC), (SBeta vs AA), (SBeta+ vs SC), (SBeta+ vs AA); *8=(SS vs SC), (SS vs AA), (SBeta0 vs SC), (SBeta0 vs AA), (SBeta+ vs AA), (SC vs

AA); *9=SS vs AA only; *10=SS vs SC only; *11=(SS vs SBeta+), (SS vs SC); *12=(SS vs SBeta+), (SS vs SC), (SS vs AA), (SBeta0 vs AA).

SS= Sickle cell anemia, SBeta+=Sickle β+- thalassemia, SBeta0=Sickle β0-thalassemia, SC=HbSC disease, AA= Healthy HbAA controls

47

6.2. Comparison of biologic markers across different categories of SCD patients

6.2.1. RBC Microparticles (RMP)

The levels of red blood cell microparticles of the patients were significantly

different across the SCD patients’ groups with HbSS having the highest levels

(p<0.05) Figure 4.

SS SB0 SB+ SC AA0

50000100000150000200000

400000600000800000

1000000

20000003000000

4000000

P1<0.0001

P2=0.0006

P3=0.0002

P4=0.168

RM

P (

even

ts/m

L)

Figure 4: Comparison of red blood cell microparticles across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.2. Plasma hemoglobin (PHb)

The levels of plasma hemoglobin of the patients were significantly different across

the SCD patients’ groups in the following order (HbSS>SC>AA) with HbSS having

the highest levels (p<0.05) Figure 5.

48

SS SB0 SB+ SC AA0

20

40

60

80

100

150

200

250

300

P1<0.0001

P2=0.0001

P3=0.015

P4=0.47

Pla

sm

a H

b (

mg

/dL

)

Figure 5: Comparison of plasma hemoglobin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.3. Plasma heme

The levels of plasma heme of the patients were significantly different across the

groups in the following order, (HbSS>SC>AA) with HbSS having the highest

levels (p<0.05) Figure 6.

49

SS SB0 SB+ SC AA0

20

40

60

100

150

200

250

P1<0.0001

P2=0.0025

P3=0.04

P4=0.479

He

me

(µµ µµ

M)

Figure 6: Comparison of heme across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.4. Haptoglobin

The levels of plasma haptoglobin of the patients were significantly different

across the groups in the following order, (AA>SC>>SS) with HbSS having the

lowest levels (p<0.05) Figure 7.

50

SS SB0 SB+ SC AA0

500

1000

1500

10000

20000

1000000

2000000

3000000

P1<0.0001

P2=0.038

P3=0.04

P4=0.93

Ha

pto

glo

bin

(n

g\m

l)

Figure 7: Comparison of haptoglobin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.5. Hemopexin

The levels of plasma hemopexin of the patients were significantly different across

the groups in the following order, (AA>SC>>SS) with HbSS patients having the

lowest levels (p<0.05) Figure 8.

51

SS SB0 SB+ SC AA0

100

200

300

400

500

1000

1500

2000

P1<0.0001

P2=0.045

P3=0.80

P4=0.54

He

mo

pe

xin

(µµ µµ

g\m

l)

Figure 8: Comparison of hemopexin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

52

6.2.6. Lactate dehydrogenase (LDH)

The levels of LDH of the patients were significantly different across the SCD

patients’ groups with HbSS having the highest levels (p<0.05) Figure 9.

SS SB0 SB+ SC0

200

400

600

800

1000

P1<0.0001

P2=0.0001

P3=0.046

LD

H (

IU/L

)

Figure 9: Comparison of LDH across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC), P2=Mann-Whitney Test (SS vs SB+),

P3= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), and RMP=Red blood cells microparticles.

6.2.7. Total bilirubin

The levels of total serum bilirubin of the patients were significantly different across the

SCD patients` groups with HbSS having the highest levels (p<0.05) Figure 10.

53

SS SB0 SB+ SC0

2

4

6

8

P1<0.0001

P2=0.031

P3=0.315

To

tal b

ilir

ub

in (

mg

/dL

)

Figure 10: Comparison of total serum bilirubin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC), P2=Mann-Whitney Test (SS vs SB+),

P3= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), and RMP=Red blood cells microparticles.

6.2.8. Unconjugated bilirubin

The levels of unconjugated serum bilirubin of the patients were significantly

different across the SCD patients` groups with HbSS having the highest levels

(p<0.05) Figure 11.

54

SS SB0 SB+ SC0

2

4

6

8

P1<0.0001

P2=0.029

P3=0.210

Un

co

nju

bilir

ub

in (

mg

/dL

)

Figure 11: Comparison of unconjugated bilirubin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC), P2=Mann-Whitney Test (SS vs SB+),

P3= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), and RMP=Red blood cells microparticles.

6.2.9. Reticulocyte count

There were no significant differences across the patients groups regarding their

reticulocytes counts (p>0.05) Figure 12.

55

SS SB0 SB+ SC0

200

400

600

800

P1=0.574

P2=0.785

P3=0.794

Reti

cu

locyte

co

un

t (X

10

9/L

)

Figure 12: Comparison of reticulocyte count across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC), P2=Mann-Whitney Test (SS vs SB+),

P3= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), and RMP=Red blood cells microparticles.

6.2.10. Red blood cell (RBC) count

The levels of RBC counts of the patients were significantly different across the

groups with HbSS having the lowest levels (p<0.05) Figure 13.

56

SS SB0 SB+ SC AA0

2

4

6

8

P1<0.0001

P2<0.0001

P3<0.0001

P4=0.005

RB

C c

ou

nt

(x 1

01

2/L

)

Figure 13: Comparison of RBC concentration across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.11. Hemoglobin concentration (Hb conc)

The levels of RBC counts of the patients were significantly different across the

groups with HbSS having the lowest levels (p<0.05) Figure 14.

57

SS SB0 SB+ SC AA0

5

10

15

20

P1<0.0001

P2<0.0001

P3=0.0009

P4=0.68

Hb

co

nc

(g

\dL

)

Figure 14: Comparison of haemoglobin concentration across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.2.12. Fetal hemoglobin (HbF)

There were significant differences in the levels of HbF between the patients and

controls, the SS patients HbF were significantly higher that of the SC patients and

58

AA respectively p<0.05) with HbSS having the highest levels Figure 15.

SS SB0 SB+ SC AA0

10

20

30

40

P1<0.0001

P2<0.0001

P3=0.554

P4=0.43

Hb

F(%

)

Figure 15: Comparison of fetal hemoglobin across groups

KEY: P1=Kruskal-Wallis ANOVA (SS vs SB0 vs SB+ vs SC vs AA), P2=Mann-Whitney Test (SS vs

SC), P3= Mann-Whitney Test (SS vs SBeta+), P4= Mann-Whitney Test (SS vs SBeta0)

SS= Sickle cell anemia (n=78), SB+=Sickle β+- thalassemia (n=12), SB0=Sickle β0-thalassemia

(n=12), SC=HbSC disease (n=36), AA= Healthy HbAA controls (n=39) and RMP=Red

blood cells microparticles.

6.3. Effects of hydroxyurea on patients’ parameters

None of the HbSC patients was on hydroxyurea treatment. However, 63 (80.7%)

out of the 78 HbSS patients and 12 (50%) of the 24 S-Beta patients comprising

eight (67.0%) S-Beta0, and four (33.0%) S-Beta+, were on hydroxyurea.

HbSS patients on hydroxyurea had significantly higher HbF, MCV, haptoglobin

and hemopexin. On the contrary, they had significantly lower bilirubin, red blood

count, reticulocyte counts, HbS levels, and white blood cell counts. Although the

blood levels of red blood cell microparticles, plasma hemoglobin, and heme were

lower in the HbSS patients on hydroxurea treatment, these values did not reach

statistical significance (Table 2).

59

Table 2: Effects of hydroxyurea treatment on biologic markers of HbSS patients

Parameters HbSS treated with hydroxyurea (N=63)

NO hydroxyurea treatment (N=15)

P values

RMP(events/mL) 160,000.0 (0.0 -17840000.0)

440,000 (0 – 3760000)

0.08

Plasma Hb (mg/dL) 83.9 (28.2 – 169.0) 98.4 (32.6 – 257.5) 0.103

Haptoglobin (ng/mL) 2001.0 (280.0 – 17520)

1013 (543.0 – 8400)

0.023

Heme (µM) 53.3 (15.8 – 206.0) 68.9 (23.5 – 155.6) 0.177

Hemopexin (µg/mL) 686.0 (76.7 – 1937.0)

242.0 (117.0 – 1607)

0.0021

HbF (%) 14.2 (1.7 – 29.5) 5.6 (2.1 – 31.0) 0.014

HbS (%) 76.8 (36.4 – 92.2) 84.7 (54.0 – 92.0) 0.0043

RBC ( million cells/uL)

2.3 (1.0 – 3.7) 2.6 (2.3 – 4.0) 0.006

MCV (fL) 123.7 (85.7 – 147.0) 96.8 (78.6 – 120.0) <0.0001

Total bilirubin (mg/dL)

1.7 (0.6 – 6.1) 3.1 (0.4 – 6.8) 0.0072

Unconjugated bilirubin (mg/dL)

1.3 (0.4 – 5.4) 2.4 (0.3 – 6.0) 0.0073

Hb (g/dL) 8.3 (4.2 – 10.8) 8.5 (7.4 – 12.9) 0.599

Reticulocyte (×109/L) 223.9 (23.2 – 755) 369.6 (76.9 – 548) 0.023

LDH (IU) 384.0 (210.0 – 938.0)

487.0 (164.0 – 683.0)

0.222

Platelet ( X 103/uL) 338.0 (99.0 – 682.0) 372.0 (208.0 – 525.0)

0.368

WBC ( X 103/uL) 5.7 (3.3 – 13.1) 10.3 (8.2 – 14.0) <0.0001

NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test, RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, MCV- Mean corpuscular volume, WBC-White blood cell count, LDH-Lactate dehydrogenase.

For the S-Beta thalassemia patients, the only parameters affected by hydroxyurea

treatment were the RBC and the MCV which were significantly reduced and increased

respectively (Table in Annex 1).

60

6.4. Correlational studies of measured parameters in patients

6.4.1. Red blood cell microparticles

The correlations of RMP with other biologic markers are shown in Figure 16. It showed

that, RMP had positive correlations with total serum bilirubin (r=0.45, p<0.000, figure

16.1), unconjugated bilirubin (r=0.46, p<0.0001, figure 16.2), LDH (r=0.57, p<0.0001),

figure 16.3), plasma hemoglobin (r=0.48, p<0.0001, figure 16.4), heme (r=0.60,

p<0.0001, figure 16.6), and absolute reticulocyte counts (r=0.31, p=0.002, figure 16.8),

of the patients. It also showed a modest significant positive correlation with TRV (r=0.42,

p=<0.0001, figure 16.9), and MCV (r=0.23, p=0.024, figure 16.10). On the contrary, a

modest significant negative correlation was observed between RMP and haptoglobin (r=

-0.228, p=0.039, figure 16.5), hemopexin (r= -0.37, p<0.0001, figure 16.7, red blood cell

count (r= -0.28, p=0.003, figure 16.11) and Hb concentration (r= -0.232, p=0.017, figure

16.12). Also, the HbF of the patients showed a negative correlation with red blood cells

microparticles (r= -0.33, p=0.001), figure 16.13).

61

Figure.16.1

Fig 16.2

r = 0.45, p= 0.000H

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

-1 0 1 2logTotal bilirubin (mg/dL)

logRMPeventsml Fitted values

r=0.46, p= 0.000N

1012

1416

18

log

RB

C m

icro

part

icle

s (R

MP

) (e

vent

s/m

L)

-2 -1 0 1 2log Unconjugated bilirubin (mg/dL)

logRMPeventsml Fitted values

62

Fig 16.3

Fig 16.4

1012

1416

18lo

g R

MP

(ev

ents

/mL)

5 5.5 6 6.5 7log LDH (IU)

logRMPeventsml Fitted values

B r = 0.57 p=0.000

r = 0.48, p = 0.000M

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

3.5 4 4.5 5 5.5log Plasma Hemoglobin (mg/dL)

logRMPeventsml Fitted values

63

Fig 16.5

Fig 16.6

K r= -0.228, p=0.039

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

6 7 8 9 10logHaptoglobin (ng/mL)

logRMPeventsml Fitted values

1012

1416

18lo

g R

MP

(ev

ents

/mL)

3 4 5 6log HemeM (µM)

logRMPeventsml Fitted values

A r = 0.60 p=0.000

64

Fig 16.7

Fig 16.8

r = -0.37, p=0.000

1012

1416

18lo

gRM

Pev

ents

ml (

RM

P)

(eve

nts/

mL)

4 5 6 7 8logHemopexingml (µg/ml)

logRMPeventsml Fitted values

r=0.305, p=0.002A

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

3 4 5 6 7log Absolute Reticulocyte count (x10^9/L)

logRMPeventsml Fitted values

65

Fig 16.9

Fig 16.10

r = 0.42, p=0.000

1012

1416

18lo

gRM

Pev

ents

ml (

RM

P)

(eve

nts/

mL)

.6 .8 1 1.2 1.4 1.6logTRVmsec (m/sec)

logRMPeventsml Fitted values

r = 0.23, p= 0.024G

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

4.2 4.4 4.6 4.8 5log MCV (fl)

logRMPeventsml Fitted values

66

Fig 16.11

Fig 16.12

r = -0.28, p = 0.003F

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

0 .5 1 1.5 2log RBCCount (10^12/L)

logRMPeventsml Fitted values

r = -0.232, p = 0.017I

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(RM

P)

(eve

nts/

mL)

1.5 2 2.5 3log Hb Conc (g/dL)

logRMPeventsml Fitted values

67

Fig 16.13

6.4.2 Plasma haemoglobin

Plasma hemoglobin showed significant positive correlations with total bilirubin (r=0.45,

p<0.0001), unconjugated bilirubin (r=0.43, p<0.0001), LDH (r=0.42, p<0.0001), absolute

reticulocyte counts (r=0.45, p<0.0001), HbS (r=0.357, p=0.0001), of the SCD cohorts.

On the contrary plasma haemoglobin showed significant negative correlations with

haptoglobin (r= -0.35, p <0.0001), modestly with red blood cell count (r= -0.25, p=0.011),

Hb concentration (r= -0.196, p=0.02) and HbF (r= -0.23), p=0.017).

6.4.3 Plasma heme

Heme also showed showed significant positive correlations with total bilirubin (r=0.45,

p<0.0001), unconjugated bilirubin (r=0.44, p<0.0001), (LDH r=0.47, p<0.0001), HbS (r=

0.228, p=0.007), and absolute reticulocyte count (r=0.46, p<0.0001).

It showed significant negative correlations with hemopexin (r= -0.42, p<0.0001), red

blood cell count (r= -0.195, p=0.02) Hb concentration (r= -0.187, p=0.03) and HbF (r= -

0.25, p=0.013).

6.4.4 Plasma hemopexin

1012

1416

18lo

g R

BC

mic

ropa

rtic

les

(eve

nts/

mL)

-1 0 1 2 3 4log HBF (%)

logRMPeventsml Fitted values

r = - 0.33 p=0.001

68

Hemopexin showed modest negative correlations with total bilirubin r= -0.38, p=0.001,

unconjugated bilirubin r= -0.390, p=0.001, reticulocyte count (r= -0.397, p<0.0001),

LDH (r= -0.186, p=0.028) and HbS (r= -0.235, p=0.005).

6.4.5 Plasma haptoglobin

Haptoglobin showed modest negative correlations with total bilirubin r= -0.268, p=0.001,

unconjugated bilirubin r= -0.256, p=0.002, and reticulocyte counts (r= -0.28, p=0.001)

and LDH (r= -0.23, p=0.007).

6.4.6 Tricuspid regurgitation velocity (TRV) and markers of hemolysis

The TRV showed significant positive correlations with RMP (r= 0.42, p<0.0001), heme

(r= 0.25, p=0.004), Total bilirubin (r=0.19, p=0.03), Unconjugated bilirubin (r=0.21,

p=0.019), and LDH (r= 0.32, p=0.001). It however showed no significant correlation with

plasma haemoglobin (r= 0.114, p=0.253) and reticulocyte count (r=0.086, p=0.333). On

the contrary, it showed significant negative correlations with the red blood cell count (r =

-0.43, p<0.0001) and haemoglobin concentration (r= 0.33, p<0.0001).

6.5 Clinical events among the patients

Regarding the acute symptoms or complications of SCD within the last 12 months, 14

(10.1%) patients had acute chest syndrome, and 48 (34.8%) had bone pain crisis.

However, the cumulative and longtime complications found among the patients included:

leg ulcer in 17 (12.3%), stroke 12 (8.6%), proliferative SCD retinopathy 33 (23.9%),

Osteonecrosis 32 (23.2%), elevated tricuspid regurgitation velocity in 53 (38.4%),

priapism in 4 males (3.6%), and albuminuria in 37 (26.8%) out of which 34 (24.6%)

patients were cases of microalbuminuria while the remaining three patients had

macroalbuminuria. In general, the cumulative and longtime complications were more

common among the patients with the HbSS and S-Beta0 genotypes. However, the

occurrence of proliferative SCD associated retinopathy was more among the HbSC

patients (Table 3).

69

Table 3: Clinical events among Brazilian patients

Acute clinical

events and

chronic

complications

(N=138), n (%)

Number

with SS

genotype

N=78

n (%)

Number

with SBeta0

genotype

N=12

n (%)

Number

with SBeta+

genotype

N=12

n (%)

Number

with SC

genotype

N=36

n (%)

Number of

patients (n) and

percentage (%) on

hydroxyurea

therapy per clinical

event and chronic

complications

Bone pain crisis

48 (34.8)

30 (38.4) 5 (41.6) 5 (41.6) 8 (22.2) 32 (66.7)

Osteonecrosis

32 (23.2)

14 (17.9) 1 (8.3) 7 (58.3) 10 (27.7) 15 (46.8)

Acute chest

syndrome

14 (10.1)

12 (15.3) 2 (16.6) 0 (0) 0 (0) 9 (64.2)

Elevated

Tricuspid

regurgitant

velocity (Risk for

pulmonary

hypertension)

53 (38.4)

44 (56.4) 5 (41.6) 1 (8.3) 4 (11.1) 40 (75.5)

Sickle leg ulcer

17 (12.3)

14 (17.9) 3 (25.0) 0 (0) 0 (0) 16 (94.1)

Stroke

12 (8.6)

9 (11.5) 1 (8.3) 2 (16.6) 0 (0) 12 (100)

Proliferative

retinopathy

33 (23.9)

10 (12.8) 0 (0) 1 (8.3) 22 (61.1) 10 (30.3)

Microalbuminuria

34 (24.6)

24 (30.7) 2 (16.6) 2 (16.6) 6 (16.7) 26 (76.5)

Priapism (Male

only N=51)

5 (9.8)

5 (9.8) 0 (0) 0 (0) 0 (0) 5 (100)

NB: SS= Sickle cell anemia, SBeta+=Sickle β+- thalassemia, SBeta0=Sickle β0-thalassemia,

SC=HbSC disease

6.6 Relationships between clinical events and biologic parameters of patients

6.6.1 Acute chest syndrome (ACS)

None of the measured parameters was associated with the occurrence of acute chest

syndrome among the patients (Table 4)

70

Table 4: Associations between acute chest syndrome and biologic markers

Parameters ACS present (N=14) ACS absent (N=124)

P values

RMP(events/mL) 140,000.0 (0.0 -17840000.0)

80,000 (0 – 3760000)

0.3568

Plasma Hb (mg/dL) 99.8 (28.5 – 168.4) 76.8 (31.7 – 257.5) 0.0688

Haptoglobin (ng/mL) 2154.0 (280.0 – 29670) 2275 (279.0 – 11536)

0.826

Heme (µM) 69.8 (15.8 – 177.0) 50.5 (16.0 – 206.0) 0.1309

Hemopexin (µg/mL) 361.1 (76.7 – 1843.0) 648.5 (100.0 – 1937)

0.257

HbF (%) 5.5 (1.7 – 24.6) 9.1 (0.2 – 35.2) 0.969

HbS (%) 69.5 (44.8 – 92.1) 70.3 (23.4 – 92.2) 0.636

RBC ( million cells/uL) 2.9 (1.0 – 4.2) 2.8 (1.2 – 6.3) 0.5319

Total bilirubin (mg/dL) 1.6 (0.6 – 5.4) 1.4 (0.4 – 6.8) 0.2927

Unconj. bilirubin (mg/dL) 1.2 (0.5 – 4.3) 1.1 (0.3 – 6.0) 0.3023

Hb (g/dL) 9.2 (4.2 – 10.8) 9.3 (4.8 – 15.2) 0.599

Reticulocyte (×109/L) 328.0 (23.2 – 755) 225.0 (47.2 – 561) 0.2911

LDH (IU) 334.0 (166.0 – 938.0) 331.0 (128.0 – 907.0)

0.997

Platelet ( X 103/uL) 310.0 (152.0 – 497.0) 339.0 (61.0 – 755) 0.7336

WBC ( X 103/uL) 8.5 (3.2 – 13.1) 6.9 (3.4 – 16.4) 0.3610

NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test, RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-Lactate dehydrogenase.

6.6.2 Vaso occlusive crisis (VOC) / Bone pain crisis

None of the measured parameters was associated with the occurrence of bone pain crisis

among the patients (Table 5)

Table 5: Associations between acute bone pain crisis (VOC) and biologic markers

Parameters VOC present (N=48) VOC absent (N=90) P values

RMP(events/mL) 100,000(0.0 -17840000) 100,000 (0 – 3760000) 0.8594

Plasma Hb (mg/dL) 77.3 (28.2 – 169.0) 78.0 (31.7 – 257.5) 0.884

71

Haptoglobin (ng/mL) 2234.0 (280.0 – 14388) 2164 (279.0 – 29670) 0.5302

Heme (µM) 53.3 (15.8 – 177.0) 50.6 (16.0 – 206.0) 0.5046

Hemopexin (µg/mL) 582.7.1 (76.7 – 1882.0) 677.0 (108.0 – 1937) 0.5046

HbF (%) 6.9 (0.4 – 27.9) 9.4 (0.2 – 35.2) 0.4294

HbS (%) 74.6 (36.4 – 92.1) 69.4 (23.4 – 92.2) 0.174

RBC (million cells/uL)

2.8 (1.0 – 5.6) 2.7 (1.2 – 6.3) 0.8796

Total bilirubin (mg/dL)

1.7 (0.6 – 6.1) 1.3 (0.4 – 6.8) 0.239

Unconj. bilirubin (mg/dL)

1.4 (0.4 – 5.4) 1.0 (0.3 – 6.0) 0.204

Hb (g/dL) 9.4 (4.2 – 15.2) 9.2 (4.8 – 15.2) 0.986

Reticulocyte (×109/L) 252.0 (23.2 – 755) 224.5 (52.0 – 561) 0.7025

LDH (IU) 345.0 (128.0 – 938.0) 325.0 (128.0 – 907.0) 0.5032

Platelet ( X 103/uL) 328.0 (89.0 – 682.0) 338.0 (61.0 – 755) 0.8451

WBC ( X 103/uL) 6.8 (3.2 – 16.4) 7.4 (3.4 – 14.0) 0.6480

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test, RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-Lactate dehydrogenase.

6.6.3 Leg ulcer

The red blood cell microparticles, heme, LDH, HbS conc, RBC, and Hb conc were

significantly higher in patients who had leg ulcer (p<0.05). No association was

found with the other biologic markers (Table 6).

Table 6: Associations between leg ulcer and biologic markers

Parameters Leg ulcer present (N=17)

No leg ulcer (N=121)

P values

RMP(events/mL) 280,000.0 (40,000.0 -2600000.0)

80,000 (0 – 17840000)

0.0134

Plasma Hb (mg/dL) 81.8 (42.9 – 232.9) 76.8 (28.2– 257.5) 0.3043

Haptoglobin (ng/mL) 2143.0 (279.0 – 7400)

2339 (280.0 – 29670)

0.8202

Heme (µM) 69.2 (33.9 – 206.0) 50.1 (15.8 – 177.0) 0.0079

Hemopexin (µg/mL) 559.3 (108.0 – 978.0)

621.8 (76.0 – 1937) 0.429

72

HbF (%) 12.6 (0.4 – 26.8) 7.5 (0.2 – 35.2) 0.104

HbS (%) 79.4 (46.0 – 96.0) 68.4 (23.4 – 92.2) 0.0041

RBC ( million cells/uL)

2.2 (1.7 – 4.0) 2.9 (1.0 – 6.3) 0.0020

Total bilirubin (mg/dL)

1.5 (0.9 – 5.1) 1.4 (0.4 – 6.8) 0.5605

Unconjugated bilirubin (mg/dL)

1.3 (0.6 – 3.5) 1.1 (0.3 – 6.0) 0.4129

Hb (g/dL) 8.6 (7.0 – 11.3) 9.4 (4.2 – 15.2) 0.0424

Reticulocyte (×109/L) 232.0 (178.0 – 548.0)

231 (23.2 – 755) 0.6812

LDH (IU) 393.5 (210.0 – 711.0)

324.0 (128.0 – 938.0)

0.0129

Platelet ( X 103/uL) 327.5 (99.0 – 592.0)

340.5 (61.0 – 755)

0.6812

WBC ( X 103/uL) 6.0 (3.4 – 10.1) 7.3 (3.2 – 16.4) 0.123 NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

6.6.4 Elevated Tricuspid regurgitation velocity (TRV) and risk of pulmonary

hypertension

The red blood cell microparticles, heme, HbF, HbS, and LDH were significantly

higher in patients who had elevated tricuspid regurgitation velocity (TRV) while

their haemoglobin concentration and RBC counts were significantly lower

(p<0.05). No association was found with the other biologic markers (Table 7).

Table 7: Associations between risk for pulmonary hypertension and biologic markers

Parameters Elevated TRV (at risk of pulmonary hypertension) (N=53)

Normal TRV (not at risk of pulmonary hypertension) (N=74)

P values

RMP(events/mL) 200,000.0 (0 -3760000.0)

80,000 (0 – 1800000) 0.0046

Plasma Hb (mg/dL) 79.5 (31.7 – 257.5) 73.1 (28.2– 174.5) 0.5104

Haptoglobin (ng/mL) 2021.0 (279.0 – 11720) 2251 (280.0 – 18180) 0.3196

Heme (µM) 53.4 (16.0 – 206.0) 47.0 (15.8 – 106.0) 0.0375

Hemopexin (µg/mL) 546.8 (76.7 – 1844.0) 668.9 (100.0 – 1937) 0.1620

HbF (%) 12.4 (0.4 – 29.5) 5.5 (0.2 – 35.2) 0.0078

HbS (%) 75.7 (36.4 – 92.0) 62.7 (23.4 – 92.2) 0.0022

73

RBC ( million cells/uL)

2.4 (1.2 – 5.5) 3.2 (1.0 – 6.3) <0.0001

Total bilirubin (mg/dL)

1.5 (0.6 – 6.8) 1.3 (0.4 – 6.0) 0.1404

Unconjugated bilirubin (mg/dL)

1.2 (0.5 – 6.0) 0.9 (0.3 – 5.3) 0.1081

Hb (g/dL) 8.7 (5.1 – 12.8) 9.7 (4.2 – 15.2) 0.0002

Reticulocyte (×109/L) 231.0 (47.2 – 733.0) 213.0 (23.2 – 488.0) 0.370

LDH (IU) 372.0 (128.0 – 668.0) 312.5 (136.0 – 907) 0.0063

Platelet ( X 103/uL) 328.0 (66.0 – 682.0) 344.0 (61.0 – 755) 0.996

WBC ( X 103/uL) 6.6 (3.4 – 16.4) 7.6 (3.3 – 14.7) 0.0747

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase, TRV-Tricuspid regurgitation velocity, 17 patients with no TRV

records not included.

6.6.5 Stroke

The white blood cell count was associated with the occurrence of stroke among the

patients (Table 8).

Table 8: Associations between stroke and biologic markers

Parameters Stroke present (N=12) No stroke (N=126) P values

RMP(events/mL) 120,000.0 (0 -440000.0) 80,000 (0 – 17840000) 0.7869

Plasma Hb (mg/dL) 87.9 (34.0 – 123.4) 77.8 (28.5– 257.5) 0.3275

Haptoglobin (ng/mL) 2096.0 (1141.0 – 28180) 2176.0(280.0 – 29670) 0.852

Heme (µM) 65.0 (22.9 – 206.0) 58.0 (15.8 – 177.0) 0.7181

Hemopexin (µg/mL) 689.0 (230.0 – 1805.0) 703.0 (76.0 – 1937) 0.838

HbF (%) 14.2 (2.8 – 26.7) 7.5 (0.2 – 35.2) 0.0587

HbS (%) 76.4 (38.2 – 82.0) 69.9 (23.4 – 92.2) 0.559

RBC ( million cells/uL)

2.6 (1.8 – 3.2) 2.9 (1.0 – 6.3) 0.08

Total bilirubin (mg/dL)

1.3 (0.7 – 3.5) 1.5 (0.4 – 6.8) 0.1902

Unconjugated bilirubin (mg/dL)

1.0 (0.5 – 2.9) 1.1 (0.3 – 6.0) 0.323

Hb (g/dL) 8.6 (7.2 – 10.0) 9.4 (4.2 – 15.2) 0.1307

Reticulocyte (×109/L) 210.0 (52.0 – 332.0) 241 (23.2 – 755) 0.1103

LDH (IU) 305.0 (220.0 – 503.0) 339.0 (128.0 – 938.0) 0.803

Platelet ( X 103/uL) 297.5 (174.0 – 755.0) 338.0 (61.0 – 682) 0.7237

74

WBC ( X 103/uL) 5.1 (3.9 – 10.6) 7.3 (3.2 – 16.4) 0.0225

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

6.6.6 Sickle cell disease (SCD) retinopathy

The patients with proliferative SCD retinopathy had significantly lower HbF and higher

RBC counts, and Hb concentration (p<0.05) (Table 9).

Table 9: Associations between proliferative SCD retinopathy and biologic markers

Parameters Retinopathy present (N=33)

No retinopathy (N=105)

P values

RMP(events/mL) 80,000.0 (0 -960000.0) 120,000 (0 – 17840000.0)

0.009

Plasma Hb (mg/dL) 68.1 (31.7 – 232.9) 80. (28.5– 257.5) 0.083

Haptoglobin (ng/mL) 2300.0 (424.0 – 29670) 2140.0(279.0 – 18180) 0.197

Heme (µM) 45.0 (15.8 – 165.0) 53.0 (16.0 – 206.0) 0.026

Hemopexin (µg/mL) 848.0 (114.0 – 1882.0) 540.0 (76.7 – 1937) 0.0084

HbF (%) 1.0 (0.4 – 22.0) 11.2 (0.2 – 35.2) <0.0001

HbS (%) 47.3 (23.4 – 89.5) 74.0 (40.0 – 92.2) <0.0001

RBC (million cells/uL)

3.8 (1.0 – 6.3) 2.6 (1.2 – 6.1) 0.0008

Total bilirubin (mg/dL)

1.2 (0.6 – 3.4) 1.5 (0.4 – 6.8) 0.053

Unconjugated bilirubin (mg/dL)

0.98 (0.4 – 2.9) 1.2 (0.3 – 6.0) 0.05

Hb (g/dL) 10.2 (4.2 – 15.2) 8.8 (4.8 – 13.6) <0.0001

Reticulocyte (×109/L) 245.0 (23.2 – 538.0) 232.5 (47.0 – 755) 0.521

LDH (IU) 294.0 (162.0 – 702.0) 340.0 (128.0 – 938.0) 0.209

Platelet ( X 103/uL) 279.0 (66.0 – 557.0) 346.0 (61.0 – 755) 0.07

WBC ( X 103/uL) 6.8 (3.2 – 14.7) 7.2 (3.4 – 16.4) 0.25

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

6.6.7 Osteonecrosis

75

The RBC counts and hemoglobin concentration of patients who had osteonecrosis were

significantly higher (p=0.0103). No association was found with the other biologic

markers (Table 10).

Table 10: Associations between osteonecrosis and biologic markers

Parameters Osteonecrosis (N=32)

No osteonecrosis (N=106)

P values

RMP(events/mL) 80,000.0 (0 -3760000.0)

120,000 (0 – 17840000)

0.2266

Plasma Hb (mg/dL) 75.7 (31.7 – 257.5) 78.3 (28.5– 239.2) 0.8779

Haptoglobin (ng/mL) 2579.0 (480.0 – 18180)

2020.0(279.0 – 29670)

0.1257

Heme (µM) 49.5 (22.9 – 206.0) 51.0 (15.8 – 177.0) 0.8625

Hemopexin (µg/mL) 683.0 (127.6 – 1930.0)

590.7 (76.7 – 1937) 0.7906

HbF (%) 6.7 (0.4 – 28.0) 9.1 (0.2 – 35.2) 0.4128

HbS (%) 65.3 (38.2 – 92.0) 71.9 (23.4 – 92.2) 0.1986

RBC ( million cells/uL)

3.6 (1.8 – 6.1) 2.6 (1.0 – 6.3) 0.0027

Total bilirubin (mg/dL)

1.3 (0.6 – 6.8) 1.5 (0.4 – 6.1) 0.0911

Unconjugated bilirubin (mg/dL)

1.1 (0.4 – 6.0) 1.2 (0.3 – 5.4) 0.126

Hb (g/dL) 10.2 (7.5 – 15.2) 8.9 (4.2 – 15.2) 0.0041

Reticulocyte (×109/L) 214.0 (52.0 – 561.0) 235.7 (23.2 – 755) 0.4186

LDH (IU) 309.5 (128.0 – 907.0)

339.5 (136.0 – 938.0)

0.139

Platelet ( X 103/uL) 352.0 (61.0 – 557.0) 329.0 (66.0 – 755) 0.992

WBC ( X 103/uL) 5.9 (4.0 – 14.7) 7.2 (3.3 – 16.4) 0.773

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

76

6.6.8 Priapism

None of the measured parameters was associated with the occurrence of priapism among

the patients (Table 10)

Table 11: Associations between priapism and biologic markers

Parameters Priapism (N=5) No priapism (N=46) P values

RMP(events/mL) 200,000 (40,000 -1720000) 120,000 (0 – 17840000) 0.4435

Plasma Hb (mg/dL) 83.6 (51.2 – 169.0) 83.0 (32.2– 232.9) 0.703

Haptoglobin (ng/mL) 2596.0 (1141.0 – 17180) 2339.0(280.0 – 18180) 0.230

Heme (µM) 50.0 (21.8 – 102.0) 51.9 (15.8 – 206.0) 0.646

Hemopexin (µg/mL) 709.0 (100 – 1930.0) 457.0 (116.0 – 1937) 0.1887

HbF (%) 15.7 (6.3 – 17.2) 5.6 (0.4 – 29.5) 0.078

HbS (%) 76.3 (74.3 – 84.1) 70.9 (36.4 – 92.2) 0.387

RBC ( million cells/uL)

2.3 (1.3 – 3.0) 2.7 (1.0 – 6.3) 0.154

Total bilirubin (mg/dL)

1.4 (1.2 – 3.4) 1.7 (0.7 – 5.4) 0.657

Unconjugated bilirubin (mg/dL)

1.2 (0.8 – 2.9) 1.3 (0.5 – 4.3) 0.658

Hb (g/dL) 8.9 (4.9 – 10.0) 9.4 (4.2 – 15.2) 0.366

Reticulocyte (×109/L) 225.9 (195.9 – 331.0) 250.5 (23.2 – 755) 0.7157

LDH (IU) 469.0 (287.0 – 563.0) 369.0 (128.0 – 938.0) 0.4284

Platelet ( X 103/uL) 266.0 (170.0 – 369.0) 300.0 (61.0 – 682) 0.5368

WBC ( X 103/uL) 5.6 (4.4 – 7.8) 5.9 (3.3 – 13.8) 0.5161

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

77

6.6.9 Microalbuminuria

Higher heme level was associated with the occurrence of microalbuminuria among the

patients (Table 12).

Table 12: Associations between microalbuminuria and biologic markers

Parameters Microalbuminuria (N=34)

Microalbuminuria absent (N=101)

P values

RMP(events/mL) 160,000.0 (0 -1960000.0) 80,000 (0 – 17840000) 0.1097

Plasma Hb (mg/dL) 84.0 (28.2 – 232.5) 71.9 (31.7– 257.5) 0.1676

Haptoglobin (ng/mL) 2021.0 (280.0 – 29670) 2377.0(279.0 – 18180) 0.445

Heme (µM) 60.7 (21.9 – 206.0) 48.7 (15.8 – 177.0) 0.0410

Hemopexin (µg/mL) 607.0 (76.7 – 1843.0) 680.0 (100 – 1937) 0.8206

HbF (%) 11.7 (0.4 – 35.2) 8.2 (0.2 – 31.0) 0.387

HbS (%) 75.7 (23.4 – 92.0) 69.0 (38.2 – 92.2) 0.2383

RBC ( million cells/uL)

2.6 (1.2 – 5.0) 2.8 (1.0 – 6.3) 0.3410

Total bilirubin (mg/dL)

1.5 (0.6 – 6.1) 1.4 (0.4 – 6.8) 0.8205

Unconjugated bilirubin (mg/dL)

1.2 (0.4 – 5.4) 1.1 (0.3 – 6.0) 0.8095

Hb (g/dL) 9.1 (5.1 – 13.6) 9.3 (4.2 – 15.2) 0.528

Reticulocyte (×109/L) 225.0 (99.7 – 668.0) 232.0 (23.2 – 755) 0.7479

LDH (IU) 410.0 (180.0 – 702.0) 359 (128.0 – 938.0) 0.1391

Platelet ( X 103/uL) 356.0 (99.0 – 557.0) 330.0 (61.0 – 755) 0.368

WBC ( X 103/uL) 7.1 (3.7 – 14.7) 7.2 (3.2 – 16.4) 0.9153

NB: NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test,

RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-

Hemoglobin concentration, HbS- Hemoglobin S, WBC-White blood cell count, LDH-

Lactate dehydrogenase.

78

7. GENETIC STUDIES AMONG NIGERIAN COHORTS

The Nigerian cohorts consisted of 110 SCD patients made up of 102 SS, and 8 HbSC. In

addition, there were 68 controls in the study and they were made up of 22 AS and 46 AA.

7.1. Clinical events

Sixty two (60.7%) of the SS and five (62.5%) HbSC had bone pain crisis in the preceding

12 months. The longtime complications encountered among the SS cohorts included:

stroke five (4.9%), osteonecrosis five (4.9%), gallstone 6 (5.9%), priapism five (4.9%),

and leg ulcer six (5.9%). Also, one (12.5%) SC patient had stroke.

7.2. Haplotypes (β globin chain)

The βs-haplotypes was successfully determined in 107 patients (101 SS & 6 SC) and 64

controls. Four βs-haplotypes were found. Homozygous Benin haplotype (Benin/Benin)

was predominant and found in 94 (93%) patients and all the 19 (100%) HbAS controls.

The Benin/Cameroun, Benin/Bantu and Benin/Arab haplotypes were found in 4, 2 and 1

patients respectively (Table 13).

7.3. Glucose 6 phosphatase dehydrogenase (G6PD) deficiency

Mutation for G6PD deficiency was successfully determined in 107 patients and 64

controls. The African variant of G6PD deficiency (202G>A), and (A376G) was found in

24(22.4%) patients, 8 (42.1%) HbAS, and 6 (13.3%) HbAA controls. None of the

participants had the (563C>T), (A542T), and (G680T) mutations (Table 13).

Table 13: Distribution of βs-haplotypes and G6PD deficiency among Nigerian cohorts

Parameter

SCA (N=101)

n (%)

SC (N=6) n (%)

SCD (SCA+SC)

N=107 n (%)

AS (N=19) n (%)

AA (N=45) n (%)

βs-haplotypes

Benin/Benin 94 (93) 5 (83.3) 99 (92.5) 19 (100) NA

Benin/Cameroun 4 (4) 0 (0) 4 (3.7) 0 (0) NA

Benin/Bantu 2 (2) 1 (1) 3 (2.8) 0 (0) NA

Benin/Arab 1 (1) 0 (0) 1 (1) 0 (0) NA

G6PD

deficiency

21 (20.7) 3 (50.0) 24 (22.4) 8 (42.1) 6 (13.3)

NB: NA=Not applicable

79

Figure 17: Ethidium bromide-stained 1.5% agarose gel showing restriction analysis with

Nla III of PCR products related to G6PD African mutation (G202A – rs1050828).

Lanes: M: 100 bp Molecular Marker; 1: Not digested PCR (527 bp); 2 and 3: female

heterozygote samples (123, 151, 153 and 376 bp); 4, 5, 7 8 and 9: Hemi or Homozygotes

normal samples (151 and 376 bp); 6 and 10: Hemi or Homozygotes Mutant samples (123,

151 and 253 bp).

7.3.1. Influence of G6PD deficiency on biologic markers of patients

The G6PD deficiency did not significantly influence any marker in the patients

(Table 14).

Table 14 : Associations between G6PD deficiency and biologic markers of Nigerian SS cohorts

Parameters G6PD deficient (N=21)

No G6PD deficiency (N =80)

P value

Total bilirubin (mg/dL) 1.7 (0.5 – 7.7) 2.0 (0.4 – 8.1) 0.10

Unconj. bilirubin (mg/dL)

0.7 (0.1 – 6.3) 1.1 (0.1 – 6.2) 0.315

AST (IU) 41.0 (9.0 – 89.0) 45.0 (7.0 – 89.0) 0.373

LDH (IU) 904.0 (322.0 – 1489.0) 780.0 (197.0 – 1860.0) 0.223

RBC ( million cells/uL) 3.0 (2.2 – 3.9) 2.8 (1.8 – 4.8) 0.06

Hb (g/dL) 7.9 (6.5 – 9.4) 7.5 (6.2 – 11.0) 0.392

MCV (fL) 79.2 (60.0 – 95.0) 82.0 (55.9 – 115.0) 0.066

WBC ( X 103/uL) 12.5 (6.1 – 27.3) 13.2 (6.2 – 29.3) 0.340

Platelet ( X 103/uL) 334.0 (133.0 -601.0) 374.0 (108.0 – 832.0) 0.09

HbF (%) 8.9 (2.3 – 19.1) 9.6 (0.9 – 32.3) 0.376

HbS (%) 81.7 (68.6 – 87.6) 78.1 (44.0 – 92.0) 0.09 NB: Test statistics= Mann-Whitney test, RMP-Red blood cell microparticle, HbF-Fetal

Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, WBC-

White blood cell count, LDH-Lactate dehydrogenase

80

7.3.2. Association between G6PD deficiency and clinical events

Presence of G6PD deficiency did not influence the clinical events of patients (Table

15)

Table 15: Co-inheritance of sickle cell anemia with G6PD deficiency and SS clinical events

Clinical events Patients with G6PD

deficiency (N=21)

Patients without G6PD

deficiency (N=80)

P value

Bone pain crisis

(per Year)

Range (0 –6 )

Median 1.5

Range (0 –6 )

Median 1

0.2013†

Blood

Transfusion Per

year

Range (0 – 1)

Median 0

Range (0 – 2)

Median 0

0.299†

Admission Per

year

Range (0 – 6)

Median 2

Range (0 – 8)

Median 1

0.08†

Stroke 2 3 0.60*

Osteonecrosis 0 5 0.324*

Leg ulcer 2 5 0.645*

Gallstone 1 5 1.000*

Priapism n=67 (N=19)

2

(N=48)

2

0.606*

NB- Significant p values are in bold fonts, †= Mann-Whitney Test, *=Fisher`s exact test

7.4. UGT1A1 genotypes (polymorphism)

UGT1A1 genotyping was successfully determined in 107 patients (101-SS &

6HbSC), and 64 controls. Four (TA)n alleles: (TA)5, 6, 7, and 8 were found with gene

frequencies of 0.11, 0.43, 0.41 and 0.05 respectively. The alleles were associated with

10 genotypes:TA5/5, 5/6, 5/7, 5/8, 6/6, 6/7, 6/8, 7/7, 7/8, 8/8. (Figure 18).

The low (TA) 7/7, 7/8, 8/8), intermediate (TA) 6/7, 6/8), and High (TA) 5/5, 5/6, 5/7,

5/8, 6/6,) activity genotypes were found in 38 (22.2%), 63 (38.2%), and 70 (40.9%) of

the participants respectively. Among the SS group, the low activity genotypes were

found in 25 (24.7%) compared to 10 (15.6%) of the controls (P=0.1773). Homozygous

(TA)n TA7/7 was found in 22 (21.7%) of SS compared to 5 (7.8%) controls p=0.018,

(Table16)

81

Table 16: Allele and genotype frequencies of UGT1A1 promoter polymorphism among

participants

Variables SS

N=101

SC

N=6

AS

N=19

AA

N=45

Allelotypes Freq n (%) Freq n (%) Freq n (%) Freq n (%) (TA) 5 18 (11.7) 2 (16.6) 2 (6.7) 10 (12.5) (TA)6 67 (43.5) 3 (25.0) 16 (53.3) 28 (35.0) (TA)7 61 (39.6) 7 (58.4) 10 (33.3) 37 (46.2) (TA)8 8 (5.2) 0 (0) 2 (6.7) 5 (6.3) UGT1A1 Genotypes SS (N=101) SC (N=6) AS (N=19) AA (N=45) Genotypes Freq n (%) Freq n (%) Freq n (%) Freq n (%) TA5/5 0 (0) 0 (0) 0 (0) 1 (2.2) TA5/6 9 (8.9) 1 (16.6) 0 (0) 1 (2.2) TA5/7 6 (6.0) 1 (16.6) 1 (5.2) 8 (17.7) TA5/8 3(2.9) 0 (0) 0 (0) 0 (0) TA6/6 25 (24.7) 1 (16.6) 8 (42.1) 5 (11.1) TA6/7 31 (30.7) 0 (0) 7 (36.8) 21 (46.7) TA6/8 2 (2.0) 0 (0) 1 (5.2) 1 (2.2) TA7/7 22 (21.7) 3 (50.0) 1 (5.2) 4 (8.9) TA7/8 2 (2.0) 0 (0) 1 (5.2) 4 (8.9) TA8/8 1 (1.0) 0 (0) 0 (0) 0 (0) SCA (N=101) SC (N=6) AS (N=19) AA (N=45) UGT1A1 Genotypes

by degree of

Activity

Freq n (%) Freq n (%) Freq n (%) Freq n (%)

Low-Activity genotypes TA (7/7, 7/8, 8/8)

25 (24.7) 3 (50.0) 2 (10.5) 8 (17.8)

Intermediate-Activity genotypes (TA6/7, TA6/8)

33 (32.7) 0 (0) 8 (42.1) 22 (48.9)

High-Activity genotypes TA5/5 , TA5/6, TA5/7, TA5/8, TA 6/6

43 (42.6) 3 (50.0) 9 (47.4) 15 (33.3)

82

Figure 18: UGT1A1 promoter genotypes found among Nigerian groups. A: TA5/TA5; B: TA5/TA6; C: TA5/TA7; D: TA5/TA8; E: TA6/TA6; F: TA6/TA7; G: TA6/TA8; H: TA7/TA7; I: TA7/TA8; J: TA8/TA8.

83

7.4.1. Comparison of serum bilirubin and other laboratory parameters between

patients (SS) and controls

There were significant differences between the patients and controls in all laboratory

parameters p<0.0001 respectively. (Table 17)

Table 17: Comparison of laboratory markers between patients and controls

Parameters SCA (N=101) Median (Range)

Controls (N=64) Median (Range)

Test statistics P value

HbF (%) 8.4 (0.9 – 32.3) 0.9 (0 – 5.8)

p˂0.0001

LDH (IU/L) 771 (197 – 1860)

345(150 – 840) p˂0.0001

Hb Conc (g/dL) 7.4 (6.3 – 11.2) 11.6 (7.8 – 14.8)

p˂0.0001

WBC ( X 103/uL) 12.6 (6.1 – 29.3) 6.6 (4.5 – 14.4)

p˂0.0001

Platelet ( X 103/uL) 343 (118 – 832)

278 (108 – 591)

p˂0.0001

Total Bilirubin (mg/dL)

1.8 (0.42– 8.1)

0.4 (0.1– 2.2)

p˂0.0001

Unconjugated bilirubin (mg/dL)

0.8 (0.1 -6.3) 0.2 (0.03 -0.8) p˂0.0001

AST (IU/L) 42 (17 – 89)

27 (6 – 45) p˂0.0001

ALT (IU/L) 20 (4 – 77)

10 (4 – 51) p˂0.0001

NB:Test statistics =Mann-Whitney , Significant P values are indicated in bold fonts

7.4.2. Effects of UGT1A1 genotype on serum bilirubin and other laboratory

parameters of patients

Both the total bilirubin and unconjugated bilirubin levels showed a trimodal pattern

across the UGT1A1 genotype subgroups with the low affinity genotype group

having the highest levels of serum bilirubin (p<0.0001). The LDH also showed this

trimodal pattern (p=0.0386). However, this was not demonstrated by the other

remaining parameters (Table 18).

84

Table 18: Influence of UGT1A1 genotype on laboratory parameters of SS cohort

Parameter a. Low activity UGT1A1 genotypes

N=25

b. Intermediate activity UGT1A1

genotypes N=33

c. High activity UGT1A1 genotypes

N=43

P1 values

a vs (b+c)

Anova P2 values

(a vs b vs c)

Biochemical and haematologic

Median (Range) Median (Range) Median (Range)

Total Bilirubin (mg/dl)

2.8 (1.2 -8.1) 1.7 (0.8 -4.7) 1.4 (0.4 -3.8) <0.0001* <0.0001**

Unconjugated Bilirubin (mg/dl)

1.8 (0.6 – 6.3) 0.8 (0.1 -3.6) 0.6 (0.1-2.8) <0.0001* <0.0001**

LDH (IU/L) 987(296-1860) 705 (233 -1489) 681 (197-1417) 0.0150* 0.0386**

AST(IU/L) 46 (18 -89) 41 (18 - 89) 37 (17 -89) 0.136* 0.3169**

ALT (IU/L) 25 (4-65) 18 (7- 58) 19 (4 -77) 0.1431* 0.279**

Hb conc (g/dl) 7.3 (6.3 -10) 7.5 (6.3 -11.2) 7.4(6.4 -10) 0.59*65* 0.822**

MCV(fl) 80.6 (66.9 -104.1) 82.3 (60.3 -101.5) 80.9 (55.9 -115) 0.9752* 0.8469**

RBC (x 1012/L) 2.7 (1.9 -4.1) 2.6 (2 -4.7) 2.9 (1.8 -4.8) 0.8427* 0.4438**

WBC (x 109/L) 13 (8.5 -26) 10.5 (6.1 -27) 13.3 (6.4 -29) 0.4814* 0.2997**

Platelet (x 109/L)

367 (118 -771) 342 (167 -593) 358 (118 -832) 0.4717* 0.7494**

HbF (%) 9.7 (1.3 -20.6) 7.3 (1.7 -24.4) 10.4 (0.9 -32) 0.521* 0.7466**

*=Mann-Whitney Test, **=Kruskal-Wallis one way analysis of variance, Significant p

values are indicated in bold fonts.

7.4.3. Effects of UG1TA1 genotype on clinical events

None of the HbSC patients had gallstones. Among the SS group, asymptomatic

gallstones were found in 6 (5.9%) patients. Patients who had gallstones significantly

belonged to the subgroup with low activity genotypes 5 (20%) vs 1(1.3%)

p=0.0024, (Table 19). These were 2 females and 4 males. The two females were

aged 10 and 13 years. The males consisted of a 16-year-old boy with three others

aged 10, 13, and 15 years respectively. Four of them had TA 7/7 genotypes, one

TA 7/8, and TA 6/7 each.

85

Table 19: Influence of UGT1A1 genotype on clinical events of SS patients

Clinical events a. Low activity

UGT1A1 genotypes N=25

b. Intermediate activity UGT1A1 genotypes

N=33

c. High activity UGT1A1 genotypes

N=43 P value

VOC per year 2 (0-6) 0 (0-6) 0 (0-6) 0.09*

Overt Stroke 1 2 2 0.9312†

No overt stroke 24 31 41

Osteonecrosis 1 1 3 0.711†

No osteonecrosis

24 32 40

Leg ulcer 0 1 5 0.102†

No Leg ulcer 25 32 38

Gallstones 5 1 0 0.0024†

No Gallstone 20 32 43

Priapism (Male only, N=67)

Priapism 1 2 2 0.927†

No Priapism 16 26 20

*=Mann-Whitney Test, †=Chi-Square Test, Significant p values are indicated in bold

fonts.

7.4.4. Comparison of laboratory parameters between patients with and without

gallstones.

There were significant differences between the serum bilirubin and HbF levels of

patients with gallstones when compared with those without gallstone in general. No

difference was observed in the LDH and age of the two groups. Also, when those with

gallstones were compared with age- and sex-matched peers within the same UGT1A1

genotype subgroup, only serum bilirubin and HbF showed significant differences

between the two groups (Table 20).

86

Table 20: Comparison of parameters in patients with and without gallstones

Parameter Patients with gallstones

(N=6) Median (Range)

Patients without gallstones (N=95) Median (Range)

P value

Total Bilirubin (mg/dl)

6.4 (2.8 -8.1) 1.8 (0.4 -6.7) 0.0001*

Unconjugated Bilirubin (mg/dl)

4.7 (0.9 -6.3) 0.79 (0.1 -5) 0.0007*

LDH (IU/L) 1004 (592 -1860) 794 (197 -1750) 0.1263*

HbF (%) 4.7 (1.3 -6.8) 10.2 (0.9 -32) 0.0107*

Hb (g/dl) 7.1 (6.3 -8.8) 7.5 (6.2 -10) 0.4210*

Age in years 11.5 (8 -16) 9 (2-21) 0.1368*

Sex

Male (n=66) 4 62 1.000†

Female (n=35) 2 33

Parameter Patients with gallstones (N=6)

Median (Range)

Matched peers without gallstones within same

UGT1A1 genotype activity group N=10

Median (Range)

P value

Total Bilirubin (mg/dl)

6.4 (2.8 -8.1) 2.2 (1.9- 3.2) 0.0023*

Unconjugated Bilirubin (mg/dl)

4.7 (0.9 -6.3) 1.2 (1.0 -2.0) 0.0020*

LDH (IU/L) 1004 (592 -1860) 890 (340 -1603) 0.628*

HbF (%) 4.7 (1.3 -6.8) 14.7 (4.2 -17.9) 0.022*

Hb (g/dl) 7.1 (6.3 -8.8) 8.0 (6.5 – 8.9) 0.137* NB Significant P values are indicated in bold fonts, *=Mann-Whitney test, †=Fisher`s

7.4.5. Relationship between serum bilirubin and other parameters of patients

The total serum bilirubin correlated positively with the age of the patients ( r =0.238,

p=0.013), and their LDH levels (r = 0.218, p=0.028). Conversely, it showed a

negative correlation with the HbF levels (r = -0.210, p= 0.035). No significant

correlation was found between the total serum bilirubin and other biomarkers.

87

7.4.5. Relationship between UGT1A1 genotypes and other parameters by

multivariate analysis.

The unconjugated bilirubin was significantly associated with the low activity

UGT1A1 genotypes (Adjusted Odd Ratio (1.08), 95% Confidence interval

(1.034768 – 1.127873), P=0.000). Also, significant association was found with the

total bilirubin when it was used in place of unconjugated bilirubin in the logistic

regression model (Adjusted Odd Ratio (1.05), 95% Confidence interval (1.029172

– 1.089832), P=0.000).

7.5. Alpha Thalassemia trait

Of the 164 samples successfully studied for Alpha thalassemia, Alpha thalassemia

trait was found in 41 SS, one HbSC, and 24 controls. All were due to 3.7 κb α-globin

gene deletion. There was one case of triplication among the SS.

Figure 19: Ethidium bromide-stained 1.2% agarose gel showing α 3.7 deletion by GAP-

PCR.

Lane M: GeneRuler™ 1Kb DNA Ladder; 1: Homozygote sample for 3.7

alpha globin gene deletion (1.9 Kb fragment); 2, 4, 5 and 7: Homozygotes

samples for normal alpha globin gene (2.1 Kb fragment); 3 and 6:

Heterozygotes samples (2.1 Kb and 1.9 Kb fragments)

88

7.5.1. Comparison of biodata and gene frequencies between patients and

controls

The patients consisted of 66 males and 34 females with a median age of 8.5, range

2 – 21 years. The controls were made up of 22 individuals with HbAS and 41 with

HbAA, median age of 8, range 2- 18years. They consisted of 36 males and 27

females. The patients had been on follow up for a median of 4 years, range 1.5 -

14years. There were no differences in the socio-biographic data of patients and

controls (Table 21).

Table 21: Biodata and frequencies of alpha thalassemia alleles

Parameters SCA (SS) N=100 n (%)

AS N=22 n (%)

AA N=41 n (%)

Chi-square, degree of freedom

ᵡ2, df

P value

Age in years Median (Range)

8.5 (2-21) 9 (3 – 17) 8.5 (2 – 18) NA 0.888a

Sex

Male Female

66 (66.0) 34 (34.0)

13 (59.0) 9 (41.0)

23 (56.0) 18 (44.0)

1.350, 2 0.509b

Social Class

Lower Middle Upper

48 (48.0) 44 (44.0) 8 (8.0)

12 (54.5) 8 (36.4) 2 (9.1)

20 (48.8) 18 (43.9) 3 (7.3)

0.4714, 4 0.9762

b

Alpha thalassemia present

41 (41.0) 8 (36.0) 16 (39.0) 0.2866, 2

0.866 b Alpha thalassemia absent

58 (58.0) 13 (62.0) 25 (61.0)

α-globin gene

deletion

SCA (SS) N=100 n (%)

AS N=22 n (%)

AA N=41 n (%)

αα /-α3.7 34 (34.0) 8(36.0) 16 (39.0) 0.327, 2 0.849 b

-α3.7/-α3.7 7 (7.0) 0 (0) 0 (0) NA 0.043*

-α3.7/-α3.7/-α3.7 1 (1.0) 0 (0) 0 (0) NA

Allele frequency Total chromosome

N=200 n(frequency)

Total chromosome

N=44 n(frequency)

Total chromosome

N=82 n(frequency)

Αα 150 (0.75) 36 (0.82) 66 (0.80) 1.590, 2 0.451 b

-α 49 (0.25) 8 (0.18) 16 (0.20) 0.822, 2 0.662 b

89

Ααα 1 (0.005) 0 (0.0) 0 (0.0) NA

NB: a=Kruskal-Wallis test, b= Chi-square test, *=Fisher`s exact test, NA= Not applicable, One

case of Triplicaion not considered as alpha thalassemia, Significant p value is in bold font.

7.5.2. Allele frequency of alpha thalassemia among SS patients and controls

Alpha-thal trait (3.7 κb α-globin gene deletion) was found in 41 (41.0%) patients;

34 with heterozygous deletion (αα/– α3.7), and 7 with homozygous deletion (-α3.7/-

α3.7), while 58 (58.0%) patients had normal genotype (αα/αα). Twenty-four controls

comprising 8 (36.0%) HbAS, and 16 (39.0%) HbAA had α-thal trait and these were

all heterozygous (αα/– α3.7). Also, 38 controls made up of 13 (62.0%) HbAS and 25

(61.0%) HbAA had normal genotype (αα/αα). Of the 200 patients’ chromosomes

analysed, 150 were αα, 49 were –α, and one was ααα, thus giving gene frequencies

of 0.75 (αα), 0.25 (-α), and 0.005 (ααα) respectively. Similarly, the gene frequencies

among the controls were HbAS - 0.82 (αα), 0.18 (-α) and HbAA- 0.80 (αα), 0.20 (-

α) respectively, with no gene triplication (ααα) detected and these were not

significantly different from the frequencies among the patients (Table 21). Taken

together, the prevalence of α-thal was not different across the groups (SCA 42

(42%) vs HbAS 8 (36.0) vs HbAA 16 (39.0), ᵡ2=0.2866, df=2, p=0.866).

7.5.3. Comparison of laboratory parameters between SS patients and controls

There were significant differences between the patients and controls in all

laboratory parameters p<0.05 except mean corpuscular hemoglobin (MCH). (Table

22)

Table 22: Laboratory parameters of patients and controls

Biomarkers HbAA N=41 Median (Range)

HbAS N=22 Median (Range)

HbSS N=100 Median (Range)

P value

Hb (g/dl) 11.5 (7.8 – 13) 11.6 (8 – 14) 7.5 (6.2 – 11.2) ˂0.0001

MCV (fL) 89.4 (66.4 – 102) 79.6 (60.4 – 74.7)

81.1 (60 – 115) 0.0125

RBC(million cells/uL)

4.5 (3.8 – 5.5) 4.6 (2.3 – 6.0) 2.8 (1.8 – 4.8) ˂0.0001

WBC (x 103/uL) 6.3 (4.5 – 14.4) 6.5 (4.5 -12.5) 13.1(6.1 – 29.30) ˂0.0001

Platelet (x 103/uL) 244 (117 – 514) 275 (125 – 578) 361(108 – 832) 0.0004

HbF (%) 0.75 (0 – 2.1) 0.7 (0.2 – 4.7) 9.3 (0.9 – 32.3) ˂0.0001

HbA2 (%) 2.9 (0.8 – 3.6) 3.6 (0.5 – 4.6) 1.5 (0.2 – 4.0) ˂0.0001

90

MCH (pg) 24.4 (17.6 – 27.9)

25.2 ( 18.2 – 29 )

25.3 (16 – 34) 0.260

Total bilirubin (mg/dl)

0.43 (0.1 – 0.45) 0.45 (0.25 – 1) 1.80 (0.5 – 8.1) ˂0.0001

AST (IU/L) 19 (6 – 64) 27 (8 – 75) 43 (12 – 89) ˂0.0001

LDH (IU/L) 360 (150 – 861) 375 (179 – 993) 789.3 (179 – 1860)

˂0.0001

NB- Significant p values are in bold fonts, Test statistics = Kruskal-Wallis ANOVA, Hb-Hemoglobin concentration, RBC-Red blood cells, MCV-Mean corpuscular volume, WBC-White blood cells count, HbF- Fetal hemoglobin, MCH- mean corpuscular hemoglobin, AST-Aspartate transaminase, LDH-Lactate dehydrogenase

7.5.4. Effects of alpha thalassemia on hematological indices and other

laboratory parameters among patients

Co-inheritance of α-thal was significantly associated with higher hemoglobin

concentration (Hb), red blood cell count (RBC) and HbA2 level. On the contrary, it

was associated with lower mean corpuscular volume (MCV) and mean corpuscular

hemoglobin (MCH), while the white blood cell count (WBC) was significantly

lower in patients with homozygous 3.7 κb α-globin gene deletion compared to the

two other groups. No significant differences were observed across α-thal genotypes

in the other laboratory parameters (Table 23).

Table 23: Alpha thalassemia Alleles and laboratory parameters

Biomarkers Alpha thalassemia with Heterozygous deletion N=34 Median (Range)

Alpha thalassemia with Homozygous deletion N=7 Median (Range)

SS with no Alpha Thalassemia Trait N=58 Median (Range)

P Value

Hb (g/dl) 7.6 (6.3-11.2) 8.2 (7.6-10) 7.2 (6.2-11) 0.0199

MCV (fL) 78 (66.9 -94) 74 (60.3 – 101) 84 (56 – 115) 0.0025

RBC (million cells/uL)

2.8 (2.15 -3.9) 3.1 (2.7 – 4.1) 2.7 (1.8 – 4.8) 0.0264

WBC (x 103/uL) 14.7 (6.4 – 26.2) 7.8 (6.1 – 13.2) 13.3 (6.1 – 29.3) 0.0353

Platelet (x 103/uL)

358 (108 -674) 334 (158 -444) 371 (118 -832) 0.2156

HbF (%) 8.8 (0.6 – 28.5) 13.1 (3.7 – 17.5) 10.7 (2.5 – 32.3) 0.1364 HbA2 (%) 1.7 (0.3 – 3.8) 2.8 (1.5 – 4.0) 1.5 (0.2 – 3.1) 0.0002

MCH (pg) 24.5 (18.6 – 30.6)

24.4 (15.7 – 27.1)

26 (15.9 – 34) 0.0021

Total Bilirubin (mg/dl)

2.1 (0.43 – 7.7) 1.8 (1.03 – 8.1) 1.8 (0.8 – 5.2) 0.729

AST (IU/L) 58 (12 – 89) 40 (22 – 89) 43 (7 – 89) 0.3974 LDH (IU/L) 865 (215 –

1860) 771 (705 – 986) 881 (197 – 1681) 0.6469

91

NB- Significant p values are in bold fonts, Test statistics= Kruskal-Wallis ANOVA, Hb-Hemoglobin concentration, RBC-Red blood cells, MCV-Mean corpuscular volume, WBC-White blood cells count, HbF- Fetal hemoglobin, MCH- mean corpuscular hemoglobin, AST-Aspartate transaminase, LDH-Lactate dehydrogenase.

7.5.4. Effects of alpha thalassemia on clinical events among SS patients

Sixty-one (61.0%) patients had bone pain episodes in the preceding one year, And, 6

(6.0%), 5 (5.0%) each had overt stroke and osteonecrosis while 5 (7.5%) males had

priapism as complications of their disease.

The rate of painful crisis was higher in patients with α-thal compared to those without it

(p<0.0001). However, the presence of α-thal protected against leg ulcer as none of the six

patients with leg ulcer had α-thal vs 6 (10.3%), p=0.0384. The leg ulcers were completely

healed in three (50%) patients after a median duration of five months (range 3 – 6

months), and yet to heal in the remaining three patients, two of whom are having

recrudescence of their ulcers. This gives a prevalence of active ulcer as 3%. One of the

two with active ulcers, a 21-year-old also had overt stroke. The ulcers were mostly located

on the ankles in 5 (83.3%) and right big toe in 1 (16.7%). There was a preceding history

of trauma in 4 (66.7%) and this included a case of traditional scarifications for local

therapeutic purposes. No significant association was found in relation to other clinical

events with respect to the presence or absence of α-thal (Table 24).

Table 24: Co-inheritance of sickle cell anemia with Alpha Thalassemia and clinical events

Clinical events Patients with Alpha Thalassemia (N=41)

Patients without Alpha Thalassemia (N=58)

P value

Bone pain crisis per Year

Range (0 –6) Median 3

Range (0 –6) Median 1

˂0.0001†

Stroke 2 3 1.000*

Osteonecrosis 3 2 0.6471*

Leg ulcer 0 6 0.0384 *

Priapism n=66 (Male only)

(N=34) 1

(N=32) 4

0.1974*

92

Gallstone 4 2 0.235* NB- Significant p values are in bold fonts, †= Mann-Whitney Test, *=Fisher`s exact test

7.5.5. Comparison of parameters between patients with and without leg ulcer.

The 6 (6.0%) patients with leg ulcers (both active and healed) were made up of 3 males

and females each: 4.5% vs 8.8%; p=0.393. They were relatively older compared to their

peers: median age, 17.5years (14 – 21) vs 9years (2 – 18), p=0.0002. They were all from

the low socioeconomic class: low class 6 (12.5%) vs others (middle & upper) 0 (0%),

p=0.0103. They also had less bone pain episodes (Table 25).

The LDH, bilirubin, platelet and WBC counts of the patients who had leg ulcer were

relatively higher compared to their peers without leg ulcer, however, these values did not

attain statistical significance (p>0.05). Similarly, although the Hb, RBC, and HbF of the

patients with leg ulcers were lower, these did not attain statistical significance Table 25.

Table 25: Comparison of patients with or without leg ulcer in the absence of Alpha

thalassemia

Parameters Leg ulcer present but absence of α-thalassemia

(N=6) Median (Range)

Absence of both leg ulcer and α-thalassemia

(N=52) Median (Range)

P value

Age in years 17.5 (14 – 21) 9 (2 – 8) 0.0005

Bone pain crisis per year

0.5 (0 – 1) 1 (0 – 6) 0.0301

Total bilirubin (mg/dl)

2.0 (1.0 – 4.1) 1.8 (0.8 – 5.1) 0.75

AST (IU/L) 40 (28 – 56) 38 (7 – 89) 0.89

RBC (million cells/uL)

2.7 ( 2 – 4.1) 2.8 ( 2.1 – 4.8) 0.396

Hb (g/dl) 7.1 ( 6.2 – 8.1) 7.7 (6.3 – 10) 0.214

LDH (IU/L) 972 (681 – 1681) 789 (197 – 1489) 0.130

HbF (%) 8.5 (5.9 – 12.5) 10.3 ( 2.5 – 22) 0.317

Platelet ( X 103/uL) 473 ( 207 – 669) 375 ( 118 – 832) 0.367

WBC ( X 103/uL) 14.5 ( 10 – 21.7 ) 13.5 ( 6 – 27.3) 0.542

NB- Significant p values are in bold fonts, Test statistics = Mann- Whitney Test, Hb-

Hemoglobin concentration, RBC-Red blood cells, MCV-Mean corpuscular volume,

WBC-White blood cells count, HbF- Fetal hemoglobin, MCH- mean corpuscular

hemoglobin, AST-Aspartate transaminase, LDH-Lactate dehydrogenase

93

7.6 Haptoglobin Genotypes

The haptoglobin genotypes determination was successful in 108 patients (SS-101 & SC-

7). The distribution of the haptoglobin genotypes among the patients were as follows: 48

(44.4%) Hp1-1, 41 (38.8%) Hp2-1 and 19 (17.7%) Hp2-2. Five HbSC patients had Hp1-

1 genotype, one Hp2-1, and one Hp2-2. The distribution of the haptoglobin genotypes of

the controls were as follow: 35 (54.6%) Hp1-1, 24 (37.5%) Hp2-1, and 5 (7.9%) Hp2-2.

The frequency of Hp1 alleles and Hp2 alleles were 0.63 an 0.37 in patients and 0.73 and

0.27 in controls respectively and the occurrence of the genotypes were in Hardy Weinberg

equilibrium for both patients ( ᵡ2 =3.36, p=0.186) and controls ( ᵡ2 =0.1, p=0.951).

7.6.1. Influence of haptoglobin genotype on biologic markers of patients

The haptoglobin genotype distribution did not significantly influence the laboratory

parameters of the patients Table 26

Table 26: Associations between Haptoglobin genotypes and biologic markers of Nigerian SS cohorts

Parameters Genotype Hp1-1

N=43

Genotype Hp2-1

N=40

Genotype Hp2-2

N=18

P Value

Total bilirubin (mg/dL)

2.2 (0.9 – 7.7) 1.5 (0.8 – 8.1) 1.7 (0.4 – 5.1) 0.097

Unconj. bilirubin (mg/dL)

1.0 (0.1 -6.2) 0.7 (0.1 – 6.2) 0.8 (0.3 – 3.7) 0.08

AST (IU) 37.0 (12.0 –

89.0)

45.0 (9.0 – 89.0) 37.0 (7.0 – 89.0) 0.066

LDH (IU) 950.0 (197.0 –

1750.0)

726.0 (197.0 –

1860.0)

592.0 (215.0 –

1399.0)

0.056

RBC ( million cells/uL)

2.8 (1.8 – 4.7) 2.8 (1.8 – 4.8) 2.9 (1.9 -3.9) 0.789

Hb (g/dL) 7.5 (6.2 – 11.1) 7.5 (6.3 – 10.1) 7.7 (6.3 – 8.8) 0.988

MCV (fL) 81.0 (60.0 –

115.0)

81.0 (56.0 –

102.0)

80.0 (70.0 –

104.0)

0.8057

WBC ( X 103/uL)

13.0 (6.1 – 26.2) 12.6 (6.1 – 27.3) 13.0 (6.4 – 29.3) 0.533

94

Platelet ( X 103/uL)

328.0 (118.0 –

771.0)

364.0 (108.0 –

832.0)

402.0 (159.0 –

593.0)

0.207

HbF (%) 6.7 (1.7 – 24.4) 10.8 (0.9 – 32.3) 10.2 (3.1 – 21.8) 0.224

HbS (%) 82.0 (44.0 –

91.0)

80.0 (61.0 –

89.0)

80.0 (73.0 –

88.4)

0.319

NB: Test statistics= Kruskal – Wallis Test, RMP-Red blood cell microparticle, HbF-Fetal

Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, WBC-

White blood cell count, LDH-Lactate dehydrogenase

7.6.2. Influence of haptoglobin genotype on clinical events

The haptoglobin genotype did not significantly influence the patients` clinical events Table 27

Table 27: Influence of Haptoglobin genotype on clinical events of SS cohorts

Clinical events a. Genotype

Hp1-1

N=43

b. Genotype

Hp2-1

N=40

c. Genotype

Hp2-2

N=18 P value

VOC per year 1 (0-6) 1 (0-6) 2 (0-6) 0.864*

Overt Stroke 2 1 2 0.375†

No overt stroke 41 39 16

Osteonecrosis 0 5 0 0.108†

No osteonecrosis

43 35 18

Leg ulcer 3 3 0 0.498†

No Leg ulcer 40 37 18

Gallstones 3 2 1 0.923†

No Gallstone 40 38 17

Priapism (Male only, N=67)

Priapism 3 2 0 0.632†

No Priapism 32 25 10

*=Mann-Whitney Test, †=Chi-Square Test, Significant p values are indicated in bold

fonts.

7.7 BCL11A and Fetal Haemoglobin

95

The BCL11A polymorphisms was significantly associated with the fetal haemoglobin

levels of the participants. The BC11A polymorphism was successfully studied in 110

patients and 66 controls.

Table28: Table of BCL11A SNPs examined. Code SNP

SNP1 rs4671393

SNP2 rs11886868

SNP3 rs766432

SNP4 rs1427407

SNP8 rs7606173

SNP9 rs6706648

SNP10 rs7557939

SNP11 rs6738440

SNP14 rs6732518

SNP15 rs13019832

All the SNPs were in Hardy – Weinberg equilibrium for both patients and controls:

rs4671393 Controls p value (0.6473436), Patients p value (0.8799153);

rs11886868 Controls p value (0.6473436), Patients p value (0.8799153);

rs766432 Controls p value (0.6473436), Patients p value (0.8799153);

rs7606173 Controls p value (0.9459216), Patients p value

(0.9963276);

rs6706648 Controls p value (0.9003801), Patients p value

(0.9731736);

rs7557939 Controls p value (0.6473436), Patients p value

(0.933441);

rs6738440 Controls p value (0.970751), Patients p value

(0.4929772);

rs6732518 Controls p value (0.3468636), Patients p value

(0.9896354);

rs13019832 Controls p value (0.4873589), Patients p value

(0.6341299);

Comparison of parameters of the BCL11A participants: There is no difference in the distribution of the genotype frequencies of the BCL11A polymorphisms of the patients and controls Table 29

96

Table 29: Descriptive statistics and comparisons SNPs between the patients and controls . Variable Controls Patients Total p-

value

Age (Mean ± SD (N)) 8.5 ± 3.9 (N=66) 9.1 ± 4.6 (N=110) 8.8 ± 4.3 (N=176)

0.4914¹

Age (Median (min-max)) 8.5 (2.0-18.0) 9.0 (2.0-21.0) 9.0 (2.0-21.0)

HbF (Mean ± SD (N)) 0.8 ± 0.6 (N=66) 9.8 ± 6.9 (N=110) 6.4 ± 7.0 (N=176)

<.0001¹

HbF (Median (min-max)) 0.8 (0.0-2.1) 8.4 (0.2-32.3) 3.7 (0.0-32.3)

Gender

FEMALE 27 (40.9%) 38 (34.5%) 65 (36.9%)

0.3971²

MALE 39 (59.1%) 72 (65.5%) 111 (63.1%)

Total 66 110 176

SNP1

AA 1 (1.5%) 8 (7.3%) 9 (5.1%)

0.2067²

AG 22 (33.3%) 39 (35.5%) 61 (34.7%)

GG 43 (65.2%) 63 (57.3%) 106 (60.2%)

Total 66 110 176

SNP2

CC 1 (1.5%) 8 (7.3%) 9 (5.1%)

0.2067²

CT 22 (33.3%) 39 (35.5%) 61 (34.7%)

TT 43 (65.2%) 63 (57.3%) 106 (60.2%)

Total 66 110 176

SNP3

AA 43 (65.2%) 63 (57.3%) 106 (60.2%)

0.2067²

AC 22 (33.3%) 39 (35.5%) 61 (34.7%)

CC 1 (1.5%) 8 (7.3%) 9 (5.1%)

Total 66 110 176

SNP4

GG 43 (65.2%) 65 (59.6%) 108 (61.7%)

0.4275³

GT 22 (33.3%) 38 (34.9%) 60 (34.3%)

TT 1 (1.5%) 6 (5.5%) 7 (4.0%)

Total 66 109 175

SNP8

CC 19 (28.8%) 21 (19.1%) 40 (22.7%)

0.2339²

GC 34 (51.5%) 58 (52.7%) 92 (52.3%)

GG 13 (19.7%) 31 (28.2%) 44 (25.0%)

Total 66 110 176

SNP9

CC 16 (24.2%) 35 (31.8%) 51 (29.0%)

0.2244²

CT 31 (47.0%) 55 (50.0%) 86 (48.9%)

TT 19 (28.8%) 20 (18.2%) 39 (22.2%)

Total 66 110 176

SNP10

AA 43 (65.2%) 62 (56.4%) 105 (59.7%)

0.1906²

AG 22 (33.3%) 40 (36.4%) 62 (35.2%)

GG 1 (1.5%) 8 (7.3%) 9 (5.1%)

Total 66 110 176

SNP11

97

AA 30 (45.5%) 60 (54.5%) 90 (51.1%)

0.3279²

AG 30 (45.5%) 45 (40.9%) 75 (42.6%)

GG 6 (9.1%) 5 (4.5%) 11 (6.3%)

Total 66 110 176

SNP14

CC 2 (3.0%) 12 (10.9%) 14 (8.0%)

0.1724²

CT 30 (45.5%) 47 (42.7%) 77 (43.8%)

TT 34 (51.5%) 51 (46.4%) 85 (48.3%)

Total 66 110 176

SNP15

AA 17 (25.8%) 17 (15.5%) 34 (19.3%)

0.1703²

AG 28 (42.4%) 46 (41.8%) 74 (42.0%)

GG 21 (31.8%) 47 (42.7%) 68 (38.6%)

Total 66 110 176

NB: 1= Mann-Whitney Test, 2= Chi-square Test, 3=Fisher Exact Test

Relationships between the SNPs and HbF levels of the patients: All the SNPs except

one (rs 6732518) were significantly associated with HbF levels (Table 30). Further

analysis for possible alleles combinations associated with higher HbF indicated that

patients with ACCTGCGAG combinations significantly belonged to this group while

patient with GTAGCTAAA combinations tend to have lower HbF (Table 31). A look at

the linkage of the SNPs showed that SNPs rs 4671393, rs11886868, rs766432, rs1427407,

are more related to one another and to rs 7557939 (Figure ANNEX 2).

98

Table 30: Measures of fetal hemoglobin by allele combination and comparison in Patients group.

SNP1 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

------------------

AA 8 15.88 9.07 3.40 15.80 32.30

0.0022 AA>GG

AG 39 11.47 6.86 0.20 10.70 28.50

GG 63 7.93 5.92 0.40 6.70 21.80

-----------------------------------------------------------------------------

-------------------

SNP2 N Mean SD Minimum Median Maximum p-

value Result ----------------------------------------------------

---------------------------------------

CC 8 15.88 9.07 3.40 15.80 32.30

0.0022 CC>TT

CT 39 11.47 6.86 0.20 10.70 28.50

TT 63 7.93 5.92 0.40 6.70 21.80

-----------------------------------------------------------------------------

-------------------

SNP3 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

------------------

AA 63 7.93 5.92 0.40 6.70 21.80

0.0022 CC>AA

AC 39 11.47 6.86 0.20 10.70 28.50

CC 8 15.88 9.07 3.40 15.80 32.30

-----------------------------------------------------------------------------

--------------------

SNP4 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

------------------

GG 65 7.93 5.87 0.40 6.70 21.8

0.0001 TT>GT,GG

GT 38 11.08 6.91 0.20 9.85 28.5

TT 6 20.30 6.30 14.90 18.75 32.3

-----------------------------------------------------------------------------

-----------------

SNP8 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

-----------------

CC 21 5.59 5.46 0.70 3.70 18.10

<.0001 GG>GC>CC

GC 58 9.31 6.22 0.20 7.90 28.50

GG 31 13.45 7.22 3.40 12.00 32.30

-----------------------------------------------------------------------------

-----------------

99

SNP9 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

-----------------

CC 35 13.09 6.88 3.40 12.00 32.30

0.0002 CC>CT,TT

CT 55 8.78 6.30 0.20 7.30 28.50

TT 20 6.66 6.39 0.70 4.05 20.60

-----------------------------------------------------------------------------

------------------

SNP10 N Mean SD Minimum Median Maximum p-

value Result

-----------------------------------------------------------------------------

-------------------

AA 62 7.79 5.85 0.40 6.70 21.80

0.0011 GG>AA

AG 40 11.61 6.83 0.20 11.20 28.50

GG 8 15.88 9.07 3.40 15.80 32.30

-----------------------------------------------------------------------------

------------------

SNP11 N Mean SD Minimum Median Maximum p-

value Result -----------------------------------------------------

------------------------------------------

AA 60 11.26 6.80 0.20 10.50 32.30

0.0034 AA>GG

AG 45 8.43 6.64 0.40 6.50 28.50

GG 5 3.88 5.01 0.70 1.30 12.60

--------------

SNP14 N Mean SD Minmum Median Maximum p-

value Result

---------------------------------------------------------------------------

---------------------

CC 12 12.23 6.46 0.20 13.85 20.60

0.2293 -

CT 47 10.23 7.69 0.90 7.50 32.30

TT 51 8.76 6.07 0.40 7.40 23.70

---------------------------------------------------------------------------

----------------------

SNP15

Result N Mean SD Minmum Median Maximum p-

valor

---------------------------------------------------------------------------

--------------------

AA 17 5.41 5.04 0.90 4.20 18.10

0.0002 GG>AA

AG 46 8.91 6.78 0.20 7.85 32.30

GG 47 12.18 6.68 2.60 11.60 28.50

---------------------------------------------------------------------------

---------------------

Test statistics: Kruskal Wallis ANOVA, followed by Tukey post hoc test

100

Table 31: Measures of fetal hemoglobin by allele combination and regression analysis. GTAGGCAAG N Mean SD Minimum Median

Maximum p-value¶

----------------------------------------------------------------------------

------------

0 34 8.70 7.88 0.70 4.60 32.30

0.0909

1 75 10.17 6.39 0.20 9.40 28.50

----------------------------------------------------------------------------

-------------

GTAGCTAAA N Mean SD Minimum Median

Maximum p-value ¶

----------------------------------------------------------------------------

-------------

0 55 11.88 7.35 0.70 11.60 32.30

0.0006

1 54 7.49 5.63 0.20 5.55 20.60

----------------------------------------------------------------------------

-------------

GTAGCTAGA N Mean SD Minimum Median

Maximum p-value ¶

----------------------------------------------------------------------------

-------------

0 70 11.22 7.09 0.20 10.25 32.30

0.0007

1 39 6.99 5.63 0.40 4.50 18.30

----------------------------------------------------------------------------

-------------

ACCTGCGAG N Mean SD Minimum Median

Maximum p-value ¶

----------------------------------------------------------------------------

-------------

0 70 7.91 5.92 0.40 6.30 21.80

0.0002

1 39 12.94 7.37 0.20 12.70 32.30

----------------------------------------------------------------------------

-------------

¶ Simple analysis (univariate)

Multiple analysis

GTAGGCAAG 0.3257

GTAGCTAAA 0.1618

GTAGCTAGA 0.2944

ACCTGCGAG 0.0208

Stepwise

ACCTGCGAG 0.0002-High HbF

GTAGCTAAA 0.0277-Low HbF

Patient with higher levels of Hb F presents the sequence ACCTGCGAG, but not

GTAGCTAA.

101

Table 32: Linkage disequilibrium between the SNPs pairs. Pairwise linkage disequilibrium

----------------------------------

SNP2 SNP3 SNP4 SNP8 SNP9 SNP10 SNP11 SNP15

SNP1 D 0.187 0.187 0.163 -0.108 -0.095 0.186 -0.062 -0.059

SNP1 D' 1.000 1.000 0.948 0.946 0.881 1.000 0.999 0.649

SNP1 Corr. 1.000 1.000 0.896 -0.499 -0.444 0.988 -0.333 -0.283

SNP1 X^2 219.836 219.836 175.064 54.702 43.291 214.601 24.389 17.624

SNP1 P-value <2e-16 <2e-16 <2e-16 1.4e-13 4.72e-11 <2e-16 7.87e-07 2.69e-05

SNP1 n 110 110 109 110 110 110 110 110

SNP2 D 0.187 0.163 -0.108 -0.095 0.186 -0.062 -0.059

SNP2 D' 1.000 0.948 0.946 0.881 1.000 0.999 0.649

SNP2 Corr. 1.000 0.896 -0.499 -0.444 0.988 -0.333 -0.283

SNP2 X^2 219.836 175.064 54.702 43.291 214.601 24.389 17.624

SNP2 P-value <2e-16 <2e-16 1.4e-13 4.72e-11 <2e-16 7.87e-07 2.69e-05

SNP2 n 110 109 110 110 110 110 110

SNP3 D 0.163 -0.108 -0.095 0.186 -0.062 -0.059

SNP3 D' 0.948 0.946 0.881 1.000 0.999 0.649

SNP3 Corr. 0.896 -0.499 -0.444 0.988 -0.333 -0.283

SNP3 X^2 175.064 54.702 43.291 214.601 24.389 17.624

SNP3 P-value <2e-16 1.4e-13 4.72e-11 <2e-16 7.87e-07 2.69e-05

SNP3 n 109 110 110 110 110 110

SNP4 D -0.092 -0.086 0.162 -0.057 -0.051

SNP4 D' 0.879 0.867 0.947 0.999 0.616

SNP4 Corr. -0.438 -0.412 0.885 -0.315 -0.254

SNP4 X^2 41.764 37.049 170.561 21.579 14.061

SNP4 P-value 1.03e-10 1.15e-09 <2e-16 3.39e-06 0.000177

SNP4 n 109 109 109 109 109

SNP8 D 0.226 -0.103 0.136 0.168

SNP8 D' 0.960 0.894 0.999 0.845

SNP8 Corr. 0.917 -0.477 0.632 0.700

SNP8 X^2 184.917 50.011 87.910 107.671

SNP8 P-value <2e-16 1.53e-12 <2e-16 <2e-16

SNP8 n 110 110 110 110

SNP9 D -0.091 0.130 0.171

SNP9 D' 0.825 0.919 0.829

SNP9 Corr. -0.420 0.608 0.719

SNP9 X^2 38.827 81.409 113.768

SNP9 P-value 4.63e-10 <2e-16 <2e-16

SNP9 n 110 110 110

SNP10 D -0.064 -0.054

SNP10 D' 0.999 0.587

SNP10 Corr. -0.337 -0.259

SNP10 X^2 24.994 14.767

SNP10 P-value 5.75e-07 0.000122

SNP10 n 110 110

SNP11 D 0.081

SNP11 D' 0.507

SNP11 Corr. 0.387

SNP11 X^2 32.945

SNP11 P-value 9.48e-09

SNP11 n 110

102

8. DISCUSSION

Hemolysis studies

Chronic hemolysis is a fundamental feature of sickle cell disease (SCD). It

contributes not only to its pathophysiology but also, the phenotypic variabilities15,85. In

order, to further strengthen this widely held belief, some hemolytic sub-phenotypes

comprising Leg ulcer, pulmonary hypertension, priapism and stroke have been

described15,85. However, despite these observations, more studies are needed to further

unravel more associates of hemolysis and possibly, elucidate more on the complex

interactions through which hemolysis affects SCD. Due to the impaired feasibility to

measure hemolysis directly in most clinical settings, clinicians use indirect or surrogate

markers to characterize intravascular hemolysis in SCD.15,86

In this study, we were able to establish that the hypothesized markers contribute

to hemolysis in the study cohorts. The relationships between RMP and other surrogate

hemolysis markers studied strongly suggest that RMP may be associated with

intravascular hemolysis in SCD. The correlational studies showed that RMP, plasma

hemoglobin, heme, haptoglobin, and hemopexin are related to one another as well as other

previously established traditional markers of hemolysis like serum bilirubin, LDH and

reticulocyte count15,24,85,86. One very recent study established that the only direct

reflection of the true hemolysis rate in SCA, is the red blood cell survival86. Furthermore,

the authors found that, out of the other traditional surrogate markers of hemolysis being

used, only the reticulocyte count correlated with the red blood cell survival86. Hence, our

observation that the RMP, Plasma hemoglobin, heme, haptoglobin and hemopexin of our

patients correlated with their reticulocyte counts strongly suggests that these parameters

are true markers of hemolysis. Our findings of highest levels of RMP, plasma

hemoglobin, and heme, and lowest levels of haptoglobin and hemopexin in the HbSS

cohorts in comparison to both the HbSC, and the controls is expected and suggest that the

hemolysis intensity in our study participants occured in the order SS>SC>AA. Also, this

study confirms an earlier report that, SCD patients had higher heme and lower

haptoglobin and hemopexin levels compared to healthy controls87. However, unlike in the

study by Muller-Eberhard et al,87 which was limited to only heme, haptoglobin and

hemopexin, in this study, we further extended the scope of their study by examininig the

levels of both the plasma hemoglobin and RMP. Our findings also showed that the levels

103

of these markers were higher in SCD patients compared to the controls and followed the

order SS>SC>AA. This further suggests higher hemolysis rate among the SCD patients.

Although the red blood cell contains abundant antioxidant enzyme systems like super

oxide dismutase, catalase, the peroxiredoxins and a diffusional barrier that limits NO

catabolism,12 during hemolysis, free hemoglobin is released which in turn, inactivates

nitric oxide (NO) in a deoxygenation reaction that also oxidizes the free hemoglobin to

methemoglobin. The methemoglobin so formed is very unstable and readily loses its

heme thus contributing to the various heme-related injuries seen in patients with SCD12.

Furthermore, reactions of NO with oxyhemoglobin have been shown to seriously

scavenge NO and inhibit its signaling34 These interactions lead to the depletion of NO and

subsequently set in motion, injuries to SCD patients in a way similar to damage-

associated molecular pattern molecules (DAMPs). Few examples of DAMPs include:

mitochondrial and cellular DNA, uric acid, adenosine, and other cytoplasmic and nuclear

proteins,12,15 When DAMPs are released outside of the cell, they activate innate immunity

and cause systemic inflammation in the absence of infection12,13,39,41,42. The release of

hemoglobin and its oxidation products from the red blood cells can drive sterile

inflammation via TLR4, causing vascular injuries, and a host of other damages similar to

those caused by DAMPs12,13,39,41,42. As a results of these, the hemoglobin and heme

released during intravascular hemolysis, are now being referred to as, erythrocytic

damage-associated molecular pattern molecules (eDAMPs)12,15. This observation shows

that the quest for markers of hemolysis and their impacts in SCD is ongoing with the

possibilities of more discoveries that could help in better understanding of the disease.

Given that only very few studies have examined the contributions of RMP to

hemolysis88,89, therefore, one important relevance of this study is the finding of the

association of RMP with traditional markers of intravascular hemolysis. Furthermore, no

previous study has jointly examined the relationships between the RMP, plasma

hemoglobin, heme, haptoglobin, and hemopexin together and by extension validate their

joint relationships to other previously established traditional markers of hemolysis as

done in this study. Similar to the findings in this study, Beers et al,88 found that RMP

correlated positively with plasma hemoglobin, and LDH, while Setty and colleagues89

found that RMP positively correlated with reticulocyte count. In addition, Setty et al89

found that RMP of their patients correlated negatively with the HbF of their patients

further confirming that RMP may be associated with hemolysis. Also, similar to our

104

findings, Donadee et al11 found that, RMP positively correlated with both LDH and

plasma hemoglobin in stored blood. They also showed that RMP scavenged NO in their

in-vitro study. The NO scavenging ability of RMP have also been confirmed by both

Camus et al31 and Liu et al.90 These observations highlight the potential roles RMP could

play in the pathophysiology of SCD especially as it relates to hemolysis and NO depletion

with attendant downstream complications. Therefore, our findings on RMP from this

study strongly suggest that RMP should also be considered as a member of the newly

defined eDAMPs family.

That a substantial proportion of the patients in this study were on treatment with

hydroxyurea is not surprising given that, the study participants were adults who could

have been possibly exposed to the SCD and its complications for long time. Hence, the

need for their treatment with hydroxyurea in order to ameliorate their disease

complications.

Hydroxyurea, through its ability to induce the production.of HbF, has emerged to be a

major drug used in managing patients with SCD90,91. The presence of HbF limits the rate

of polymerisation of HbS which is the primary event that leads to the cascades of

pathological and clinical manifestations in SCD48,49,50. To this end, HbF has been found

to be a modifier of anaemia, VOC, and stroke phenotypes in patients with SCD48,49,50

hence its being used regularly to ameliorate these complications.

The observed influences of hydroxurea on the hematologic parameters of the

patients have been described91 Nonetheless, it appears that the influence of hydroxyurea

therapy is more on our SS cohorts compared to their S-Beta counterparts given that more

hematologic indices were affected by the treatment with this drug in the former compared

to the later. Nevertheless, the small number of S-Beta thalassemia patients in this study,

erodes the possibility of any conclusion. Also, it is difficult to draw conclusions on how

hydroxyurea treatment affects some markers of intravacular hemolysis examined in this

study (RMP, plasma haemoglobin, heme, haptoglobin, and hemopexin) because, only few

studies have been conducted on them in-vivo in humans87,92,93,94,95,96,97and, much fewer

commented on their relationships with hydroxyurea treatment94, 96,97.

Although our SS cohorts with hydroxyurea had relatively lowered RMP, plasma

haemoglobin, and heme, however, these did not reach statistical significance.

Nonetheless, significant differences were observed with haptoglobin and hemopexin. The

105

lack of a significant reduction with hydroxyurea treatment vis a vis the other parameters

is not clear to us. However, we suspect that this could be due to the fact that, only a very

small number of patients in this study were not treated with hydroxyurea. Some other

studies have found that hydroxyurea treatment significantly influenced RMP levels of

SCD patients. Nebor et al94 in France, and Piccin et al96 in Italy as well as that by

Gerotziafas et al.97found that hydroxyurea reduces the levels of RMP in SCD patients.

Interestingly, in contrast to these studies,94,96,97 one study by Brunetta et al98 from Brazil,

actually found higher RMP levels in their SCA patients treated with hydroxyurea. They

proposed the possibility of a selection bias as the reason for their findings. They also

argued that SCD patients treated with hydroxyurea often have more severe disease and,

could possibly, suffer more splenic dysfunction and, are therefore, unable to clear RMP

effectively from circulation given that, the spleen is being considered as a clearance site

for RMP98,99,100. Another explanation proposed by them was the risk of direct cytotoxicity

of hydroxyurea leading to megaloblastic red blood cells that could shed more RMP98.

These observations raised the need for more studies to unravel how hydroxyurea

treatment affects RMP, heme, plasma hemoglobin, haptoglobin and hemopexin in SCD

patients.

Two previous studies92,93, demonstrated similar patterns of levels of plasma

hemoglobin across SCD patients` groups as found in this study hinting on the possibility

that in general, the behaviours of the studied hemolysis markers indicated that they concur

with the few available literature. While Adisa et al93 did not include healthy controls in

their study, Brittain et al92 found that, the levels of plasma hemoglobin was lower in

healthy subjects as found in this study.

Regarding the relationships between the studied markers and clinical events, the

observation that patients with osteonecrosis had higher hemoglobin concentration could

be due to the effects of blood rheology in SCD. SCD patients with higher hemoglobin

concentration levels usually have higher blood viscosity attributable to their rheological

effects and this predisposes them to blood viscosity associated phenotypes like

osteonecrosis and VOC101. Also, the finding of lower HbF in the cohorts with SCD

retinopathy might represent another aspect of the numerous ways through which HbF

ameliorate the severity of SCD48.

106

The associations of RMP and heme with leg ulcer in this study is a further confirmation

of their roles as hemolysis markers as previous studies have categorized leg ulcer as

belonging to hemolysis sub-phenotype15,85. In addition, the patients with leg ulcer had

higher LDH and this further confirms the association of LDH with leg ulcer as found in

previous larger studies25,102,103. However, to the best of our knowledge, the association of

both the RMP and heme with leg ulcer in SCD patients, have not been previously reported

thus making this study the first to report such association and therefore unique to it. We

suspect that, the duo perhaps, mediate and promote leg ulceration through the NO

depletion upon their release during intravascular hemolysis. Camus et al31 successively

demonstrated that, upon their release from the red blood cells, RMPs carry with them

trapped hemoglobin and heme. This observation was also confirmed by Liu et al.90 In

addition, Liu et al90 also demonstrated that, the trapped hemoglobin in RMP can be readily

oxidised to heme from their unstable intermediary (methemoglobin). They also

demonstrated that these RMPs are capable of entering significantly, into the free zone of

vasculature next to endothelia cell lining of blood vessels where they mediate NO

depletion thereby causing vascular damages. This is illustrated in Figure 20. Hence, one

could speculate that findings of associations of RMP and heme with the leg ulcer among

the patients is therefore not surprising given that, the SCD leg ulcer phenotype have been

closely linked to vasculopathy in SCD among other mechanisms.102,103.

107

Figure 20

Illustration of regions of modeled Red blood cells and microparticles in the lumen of the blood

vessel, which includes a red blood cell-free zone. NO is produced in the endothelium which is on

the exterior of the blood vessel. The smooth muscle is the outermost layer around the blood vessel.

The rate of NO scavenging depends on the permeability of the cell free zone which in turn, is

influenced by the amount of RMP present within the blood vessel lumen. Liu et al90

Therefore, it is conceivable to postulate that the joint transportation of trapped

plasma hemoglobin and heme by RMP is probably hazardous to SCD patients and may

be a mechanism for vasculopathy-associated phenotypes of SCD. This observations

further comfirm RMP, plasma hemoglobin, and heme as candidate markers of SCD

manifestations. The lower levels of RBC counts and hemoglobin concentration of our

patients with leg ulcer also confirms that leg ulcer is associated with hgher hemolysis and

anemia as previously described85,102,103.

The observed associations between raised TRV and RMP, heme, and worsening

anemia also suggest that TRV is related to hemolysis. Our findings that the TRV of the

108

patients correlated significantly with their RMP, heme, LDH, RBC counts and

hemoglobin concentration further confirm this association. This is in conformity with

previous reports that TRV, is associated with the hemolytic phenotype of SCD104,105. TRV

is a marker of pulmonary vascular disease and possibly, myocardiac stress. Also, elevated

TRVs are consistently associated with hyperhemolysis 15,104,105 . Studies have shown that

SCD patients with raised TRV are at risk of developing pulmonary

hypertension.15,104,105,106 (Illustration in Figure 21).

Given that pulmonary hypertension has been found to be associated with increased

risk of morbidities and deaths among SCD patients,15, 105, 107, 108 patients with SCD can

readily be screened for risk of pulmonary hypertension using TRV. Furthermore,

109

estimation of TRV can be easily done on outpatient basis through the Doppler

echocardiography thus making it useful in estimating pulmonary artery pressure and

predict the risk for pulmonary hypertension as modelled in figure 21 above. Hence, it

represents a simple, none invasive, and outpatients` investigation that can be carried out

to screen SCD patients for the risk of pulmonary hypertension-related problems so that

those patients at risk of complications can be identified for appropriate interventions.

The observation that the RMP, LDH and heme of our study cohorts with elevated TRV

were higher while their RBC count and haemoglobin were lower strongly suggests that

these parameters are associated with elevated TRV and the possible risk for developing

pulmonary hypertension. Although, RMP was not included in the model described by

Gladwin and Sachdev106 for pulmonary hypertension as shown in figure 22, findings from

this study and other emerging evidences are now showing that RMP and heme constitute

important mediators of some SCD complications. Nothing drives home this observation

than the findings in this study that both the RMP and heme were significantly associated

with at least three (leg ulcer, elevated TRV/risk for pulmonary hypertension and,

microalbuminuria/risk for kidney disease), of the hypothesized hemolytic phenotype of

SCD12,15,85, 106.

Given the increased risk of morbidities among SCD patients with elevated TRV and

pulmonary hypertension,15, 105, 107, 108 this study has further exposed the possibility of also

using the RMP and heme as a biomarkers for identifying high risk SCD patients for

further screening /intervention.

Figure 22

110

Mechanisms of hemolytic anemia in reducing NO bioavailability and association with

vasculopathic sub-phenotypes of sickle cell disease

Hemolysis releases cell-free plasma hemoglobin and arginase 1 into plasma, which

catabolize NO and L-arginine. Activation of vascular oxidases, such as xanthine oxidase,

NADPH oxidase and uncoupled eNOS, generate superoxide which scavenges NO.

Hemolytic anemia and reduced NO bioavailability are associated with vasculopathic

clinical complications in SCD patients (Gladwin & Sachdev106).

The proportion of the patients with microalbuminuria in this study confirms

previous observations that glomerulopathy manifesting as albuminuria, is common in

SCD79,109,110. The heme of our study participants with microalbuminuria was significantly

higher than those without. This suggests that the occurrence of microalbuminuria is

probably related to hemolysis intensity of our patients. However, the lack of significant

associations with other markers of hemolysis examined makes this generalization

difficult. Nonetheless, previous studies have established links between albuminuria and

hemolysis in SCD109,110. One study by Eshbach et al111 actually established that free

hemoglobin (a precursor of heme) inhibits albumin uptake by proximal tubule cells111.

This suggests a strong link between some products of hemolysis and renal disease in SCD

as found in this study. One important aspect of our finding is the fact that

microalbuminuria is a sign of early kidney disease and as such, our findings suggest that

SCD patients with high heme levels should be screened for microalbuminuria in order to

allow for early intervention and possible prevention of progression to chronic kidney

disease.

111

DISCUSSION

Genetic studies

Although quite some studies have been carried out in the developed world on the

influence of genetic markers on SCD, there is dearth of information on these from Africa.

The situation is even worse in Nigeria because of inadequate facilities. This is more

worrisome because of the status of Nigeria as the country with the highest number of

patients with SCD in the world3. Given that the effects of some genetic markers could

vary from place to place because of genetic variability and environmental factors in

different populations, it is pertinent that more studies be carried out among cohorts from

different ethnic backgrounds to fully understand the impacts of genetic modifiers on SCD.

The findings in this study are not only diverse but also interesting.

G6PD deficiency

The prevalence of G6PD deficiency among patients in this study (22.4%) is higher than

between 9 and 12% that has been described among children with SCD in Europe58,66 and

USA.66 It is however similar to between 15 and 25% that has been described among other

Africans59,112. The high prevalence of G6PD deficiency in Africa is thought to due to its

protective advantage against malaria57,58. The lack of association betwwen G6PD

deficiency with any laboratory parameters of the patients in this study confirms the lack

of consensus on how this genetic marker modulate SCD. Studies from Saudi Arabia63,

Burkina Faso64, and Senegal112found no effects of G6PD deficiency on the clinical

manifestations and laboratory parameters of SCD patients. The lack of association of

G6PD deficiency with clinical events as found in this study, is in support of some earlier

reports that, G6PD deficiency does not influence SCD significantly.63,64,112. However,

some studies from the USA and France found association of G6PD deficiency with

increased cerebral blood flow velocity, increased rate of acute anaemic events, blood

transfusions, and decreased steady state haemoglobin levels.58,65,66. There is need for more

studies to fully understand how G6PD deficiency influences SCD.

In conclusion, this study shows that G6PD deficiency does not significantly

influence both the clinical events and laboratory parameters of the patients.

UGT1A1 Polymorphism

This study confirms the variability of bilirubin levels based on the activity of the UGT1A1

genotypes as previously reported71,75,113-119. However, we are not aware of any previous

112

study that has described the trimodal pattern of LDH based on UGT1A1 genotypes

activity as found in this study. While the UGT1A1 modulation of serum bilirubin levels

is well understood71,75,113-121 the exact mechanism through which UGT1A1 could be

associated with LDH is not clear. However, it may be possible that there could be a link

through hemolysis because both bilirubin and LDH are derived from RBC and are

markers of hemolysis.122 UGT1A1 plays an important role in haem catabolism upon its

release from RBC and conversion to bilirubin71,75,113,117. Given the association of LDH

with some phenotypes of SCA,24,120 we suggest there is need for future research to unravel

if there is any link between LDH and UGT1A1 activity.

Gilbert syndrome (GS) has been described in individuals with the TA7/7 genotype

and71,75,113-114,119,120 this was more common among our patients 22 (21.7%) vs 5 (7.8%)

controls (p=0.018). This is probably due to the small number of the controls in this study.

Nonetheless, the proportion of patients with TA7/7 in this study is higher than between 6

-11% described among Europeans,123,124 11.7% in Saudi population,120 and between 3-

18% among Brazilians of different descents.76,125,126 Similarly, it is higher than the 6%

found among Kuwait SCA patients,127 and between 5 and 11% described among other

Africans124. However, this is lower than 32% described among the SCA patients in the

USA128. Nonetheless, the TA7/7 genotype prevalence in this study is comparable to the

18.2% and 20.3% earlier described among Nigerians with non SCD-related illnesses129,130

Besides the TA7/7 genotype, other UGT1A1 genotypes found in this study have been

described among other Africans and Nigerians71,124. These observations indicate that the

UGT1A1 genotype is quite variable among Nigerians and confirm the suggestions that

the expression of the UGT1A1 genotype variants is heterogeneous among Africans

compared to the Caucasians71,123,124.

Our finding that the low-activity UGT1A1 genotype was associated with gallstones

confirms previous observations that SCA patients with the low-activity UGT1A1

genotypes especially the TA7/7, are at risk of developing gallstones113,115,116,120. Also,

others,118,119,131 have reported that some other low-activity genotypes like TA7/8 and

TA8/8 predispose SCA patients to gallstones as found in this study.

The proportion of patients with asymptomatic gallstones in this study, 5.9%, is

comparable to between 4 and 6% earlier reported among Nigerian children of similar age

to our cohorts with SCA132-134. This is also similar to the 4% found in Ghana135, a close

neighbour to Nigeria. However, the gallstone prevalence in this study is lower than

between 9 – 58% that have been reported for some African children of similar age group

113

from other countries136-138. Similarly, higher prevalence of 26% and 30% was reported

for children with SCA and of same age range as our study cohorts in Italy139 and the

USA140 respectively. In the same vein, a 45% prevalence was found among a cohort of

Brazilian children with SCD following a median follow up of 7 years141 These

observations highlight the variations in propensity to gallstone development among

children with SCA. Persistent higher serum bilirubin is a risk factor for lithogenesis

among SCA patients 76, 113, 120,127. In addition, diets and environmental factors have also

been implicated138,140. Therefore, the observed similarities between the rates of gallstones

in this study, and, earlier reports from Nigeria132-134, and Ghana135 study, could be

attributed to the fact that, two countries belong to the West Africa region and are probably

exposed to similar diets and environmental factors different from their colleagues in other

distant African parts. Europeans and Americans are exposed to diets different from

African countries. Also, genetic differences among Africans and between Africans and

Europeans and/or Americans could play some roles. The age of onset and the apparent

lack of initial symptoms attributable to gallstones, as found in this study, have been

reported.131,134,138, 140 -142. However, results of follow-up studies have indicated that the

prevalence of gallstones and its complications increase with age of children with

SCA113,127, 131, 138,141 hence, the need to closely follow up these patients.

Despite the observation that the UGT1A1 low -activity genotype is a leading

factor in hyperbilirubinemia and lithogenesis among SCA patients67,113,115,142, none of the

previous studies from Nigeria examined the UGT1A1 of the patients132-134. Hence, to the

best of our knowledge, the contribution of this polymorphism to lithogenesis and

hyperbilirubinemia among Nigerian patients with SCA was not described prior to this

study. This study has shown that SCA children with the low-activity UGT1A1 genotypes

had higher bilirubin levels compared to others. In addition, we found that the low-activity

genotypes were associated with gallstones.

Similarly, although patients with gallstone in this study had higher serum bilirubin

levels and lower HbF levels as previously reported128, following multivariate analysis,

serum bilirubin was the only parameter associated with the UGT1A1 genotype.

Therefore, it thus appears that the pathway to higher serum bilirubin and gallstone

development in our patients is not exclusively driven by hemolysis and the ameliorating

effect of HbF on hemolysis but also by the influence of UGT1A1 genotype activity.

Beyond bilirubin metabolism and gallstone development, it has been suggested

that moderately elevated serum bilirubin can inhibit bacteria and Plasmodium falciparum

114

replication and that perhaps, the heterogeneity of UGT1A1 genotypes among Africans is

a result of genetic evolution to confer selective advantage by protecting from malaria in

a way similar to other genetic traits like G6PD deficiency and/or alpha

thalassemia71,75,124,143. The occurrence of these possibilities could be investigated through

a thorough prospective study.

In conclusion, this study confirms for the first time that UGT1A1 genotypes are

tightly associated with bilirubin levels and development of gallstone among young

Nigerians with SCA. In addition, it also suggests that the pathway to elevated serum

bilirubin and gallstone development among our study cohorts seems not to be exclusively

driven by hemolysis. These observations are in agreement with earlier reports67,128,144, and

highlight the contribution of UGT1A1 polymorphisms, a non-globin genetic factor, to the

clinical manifestations of SCA patients. Children with SCA in developing countries

should be screened for UGT1A1 polymorphisms and gallstones in order to allow for

holistic care.

Alpha thalassemia trait

That none of the participants had alpha thalassemia due to the 4.2kb globin gene

deletion confirms that 3.7kb globin gene deletion is the common form of alpha

thalassemia among Africans as previously described.43

The prevalence of alpha thalassemia among patients in this study (41%) is higher

than between the 13 and 28% described among patients in the Americas145,146,147,148. This

is however lower than the 60% and 77% described among Congolese,149 and Ugandans43

respectively. It is comparable to the 37.3% among Cameroonians, 150 and the 46% among

SCA patients in France151. It is also in agreement with the 40% to 42.5% earlier described

among Nigerians SCA patients152,153. The high prevalence of α-thal in this study, and

others from Africa,43,149,150 may be due to the selective advantage in conferring protection

and survival against malaria154,155,156. The higher prevalence of α-thal in central

Africa,43,149 relative to this study, may derive from its vital impact on survival of SCA

which is more severe in this part of Africa43,149.

The observed hematological profile of patients with α-thal in this study has been

previously described145,146,148,151-154,157. The increased Hb and red blood cell count are

probably due to the decrease in the intracellular concentration of HbS, and number of

dense red blood cells. These, in turn, lead to increased red blood cell deformability and

decreased rates of both HbS-induced RBC polymerization and hemolysis85,158,159.

115

Similarly, the lack of influence of α-thal on the HbF levels of patients in this study, has

been previously reported145,146,148 .

However, contrary to some earlier reports,146,148 the presence of α-thal was associated with

higher rates of bone pain crisis compared to those without α-thal. This finding is in

keeping with that of Renoux et al151 among children with SCA in France and other

previous reports158,159. Vaso-occlusive crisis (VOC) manifesting as bone pain crisis is a

common complication of SCA and, is thought to be associated with increased blood

viscosity151, 158. Studies have demonstrated that SCA patients with α-thal have increased

blood viscosity because of relatively higher levels of Hb and/or hematocrit, among other

mechanisms85,151,158,159. In this study, our SCA patients with α-thal, had higher Hb that

could lead to increased blood viscosity hence, it is conceivable that they could have higher

rates of painful crisis.

Furthermore, as found in this study, authors from different parts of the world have

reported that α-thal protects against leg ulcer in SCA patients149,157,160. However, some

have reported no such association43,150. These observations reflect lack of consensus on

the roles of α-thal in some phenotypes of SCA. Therefore, there’s a need for more studies

from Africa to explain the relationships between α-thal and leg ulcer, and other SCA

phenotypes.

Leg ulcer is a chronic and debilitating complication that is thought to be associated with

very severe phenotype of SCA85,102,103,161,162. Some studies have associated it closely with

some other hemolysis-related phenotypes of SCA such as priapism, stroke and pulmonary

hypertension85,102,161. We could not confirm the diagnosis of pulmonary hypertension

among our patients due to lack of facilities. However, like their leg ulcer counterparts,

most of the few patients with priapism also did not have α-thal, suggesting the possibility

of protection by α-thal against both leg ulcer and priapism.

Furthermore, the bilirubin, LDH, and AST levels of the patients with α-thal were

relatively lower than their counterparts without α-thal and this suggests lower hemolysis

in the former lending credence to the belief that presence of α-thal ameliorates hemolysis

in SCA patients.

When we evaluated the markers of hemolysis of the non-thalassemic cohorts, we

observed that in contrast to the patients with α-thal, the LDH, and bilirubin levels of leg

ulcer patients were relatively higher while their Hb and HbF were lower, compared to

their matched peers. This is further suggestive of increased hemolysis in the leg ulcer

group. However, these differences did not reach statistical significance probably because

116

of our sample size. Other larger studies have reported association of hemolysis markers

with leg ulcer in SCA patients85,102,103. and the presence of α-thal is thought to ameliorate

hemolysis in SCA85,158,159.

The observations that α-thal protect against the occurrence of stroke in children

with SCA163,164 was not sustained in this study. On the contrary, one previous study 153

found that, in combination with BCL11A variants, α-thal was associated with increased

risk for stroke in older SCA patients. The lack of association between α-thal and stroke

in this study, is in keeping with report by Filho et al165. Nonetheless, these findings need

to be interpreted with caution because of differences in age of study cohorts and

possibility of survival bias. In addition, stroke was determined in this study by overt

clinical history and was, in most cases, not confirmed by magnetic resonance imaging or

angiography and this could lead to detection bias and exclusion of cases of silent infarcts

from our analysis. These observations further underscore the need for more studies to

clearly define the prevalence and associations of α-thal with clinical manifestations

among Nigerian SCA patients.

The lack of association between α-thal and osteonecrosis in this study contrasts

with the report by Milner et al166 on children enrolled in the cooperative study of sickle

cell disease in the USA where they found that both α-thal and increasing age were

significantly associated with osteonecrosis. Possible explanations for our findings

included the fact that our study participants were fewer and younger compared to that by

Milner et al.166 The youngest patient in this study was two years old, while theirs` was

five years old. In addition, only five percent of our patients had osteonecrosis compared

to 9.8% in their study.

Osteonecrosis is a disabling and severe complication of SCA associated with

impairments of both functional activities and growth in children. In SCA, it is thought to

be due to bone microcirculation disturbance in the patients and can be observed in

children with SCA as young as five years old with an increasing incidence throughout

childhood and adolescence, peaking in early adulthood167.

Furthermore, this study is important because the modulating effects of alpha

thalassemia on hematologic indices as found, further reinforce the observations by Borges

et al168 that, Alpha thalassemia significantly influences the hematologic parameters

especially the MCV and MCH to the extent that it could easily be misdiagnosed and

confused with iron deficiency anemia. Given the high rates of iron deficiency anemia

117

among children in developing countries like Nigeria169, it is therefore important to

exclude alpha thalassemia in patients suspected to have iron deficiency in the tropics.

In conclusion, this study shows that coexistence of α-thal influences the

hematologic parameters of patients with SCA as previously described145,157,160,168. It also

showed that α-thal was associated with increased rates of bone pain crisis and protects

against the occurrence of leg ulcer.

Haptoglobin Polymorphism: This study confirm that the Hp1 allele and its

associated genotypes (Hp1-1 and Hp2-1) are common among the Nigerian SCD

patients16. This is similar to the pattern observed among SCD cohorts in Northeast

Brazil.18 However, the Hp1-1 genotype was found to be least represented in both

southeast Brazil170 and Kuwait.16 These differences may reflect the genetic backgrounds

of the patients as well as the influence of environmental factors. The similarity between

the haptoglobin genotypes in this study and that from northern Brazil could stem from the

link between SCD in northern Brazil and migration of people from Africa due to the trans-

atlantic slave trade.18 There was lack of any association between the Hp genotypes and

the examined laboratory parameters of the Nigerian cohort. This observation is not

surprising, as a study among a cohort of SCD patients in Brazil170 did not find any

relationship between haptoglobin genotypes and interleukins 6 and 8. Nonetheless, both

the current study and that from Brazil170 did not examine the blood levels of plasma

hemoglobin which has been proven to have links with haptoglobin genotypes.171 The lack

of association between haptoglobin genotypes and clinical events could be due to the fact

that the study cohorts are relatively young and less at risk of some chronic SCD

complications. It also contradicts the speculations that patients or individuals with certain

haptoglobin genotypes (e.g Hp2-2 are more prone to certain health conditions.16-18

BCL11A Polymorphism

While more than 70% of SCD patients worldwide live in Africa, most advances

in the molecular understanding and management of SCD have been based on research

conducted in either the Europe or the USA. It is now incrovertible that fetal hemoglobin

(HbF) has emerged to be an important disease modifying factor in SCD48-55. Previous

analyses of the cooperative studies on sickle cell disease (CSSCD) showed that increased

HbF levels correlate with less severe complications, fewer pain crises172, and improved

survival173. Also, it has been found that common genetic variation at BCL11A associated

118

with HbF levels lies in noncoding sequences associated with an erythroid enhancer

chromatin signature thus suggesting that, the genome wide association-marked BCL11A

enhancer locus may be an attractive target option for therapeutic genome engineering for

the hemoglobinopathies52. However, despite these great potentials of HbF, information

on the influence of genetic influencers of this impotant modifier of SCD severity are

scanty in Africa. This study confirms for the first time that BCL11A polymorphism

modulate HbF levels of the Nigerian SCD cohorts. The observation that nine: (rs

4671393, rs11886868, rs766432, rs1427407, rs 7606173, rs 6706648, rs 7557939, rs

6738440 and rs 13019832) of the ten BCL11A polymorphisms studied were associated

with the HbF levels of the patients further reinforcing the important roles BCL11A

polymorphism play in the regulation of HbF among the Nigerian patients with SCD. This

observation is in agreement with findings from previous studies among Europeans and

Americans50, Tanzania and African British53, African Brazilian and African American174

SCD patients. It also confirms the association of BCL11A polymorphisms with HbF as

found among a cohort of SCA patients in northern Brazil175 and studies from Cameroon 54,55 and Tanzanian53 cohorts both in Africa. In conclusion, this study shows that BCL11A

polymorphism significantly influenced the HbF levels of the Nigerian children with SCD.

119

9. CONCLUSIONS

Hemolysis studies

This study was able to establish that:

a. Serum levels of RMP, heme, and plasma haemoglobin were significantly higher

in SCD patients compared to control while the hemopexin and haptoglobin levels

of SCD patients were significantly lower compared to controls suggesting higher

hemolysis among the SCD cohorts;

b. The intensity of hemolysis appears to be in the following order SS>SC>AA;

c. RMP, plasma hemoglobin, heme, haptoglobin and hemopexin were significantly

associated with traditional markers of hemolysis thus suggesting their roles as

markers of hemolysis;

d. Hydroxyurea therapy did not significantly influence the levels of RMP, heme, and

plasma hemoglobin of the studied HbSS patients;

e. RMP was significantly associated with leg ulcer and elevated TRV among the

patients;

f. Heme was significantly associated with leg ulcer, elevated TRV and the

occurrence of microalbuminuria;

g. Two of the hypothesized hemolytic phenotypes of SCD: leg ulcers and elevated

TRV (risk for pulmonary hypertension) were associated with biological markers

evaluated in this study thus suggesting their importance in the clinical and

pathologic manifestations of SCD;

h. Osteonecrosis was associated with increased levels of hemoglobin concentration;

i. Higher HbF levels were protective against SCD retinopathy;

j. Based on the above observations, it thus appear that there is a catastrophic synergy

between the RMP and heme in the occurrence of adverse clinical events among

the studied patients. Therefore, therapies targeting heme and RMP may be a new

approach to tackling the SCD.

120

Genetic studies

This study was able to establish the following:

a. UGT1A1 genotypes significantly influenced both the serum bilirubin and LDH

levels of the patients;

b. UGT1A1 polymorphism was found to be significantly associated with the

occurrence of gallstone among the Nigerian patients with SCD. This is the first

study to describe the influence of UGT1A1 polymorphism in Nigerian patients

with SCD;

c. Alpha thalassemia significantly influenced the hematologic indices of the

patients;

d. Alpha thalassemia was associated with higher rates of bone pain crisis and could

protect against leg ulcer phenotype.

e. BCL11A polymorphism significantly influenced the HbF of the Nigerian patients

cohort;

f. G6PD deficienccy did not influence both the clinical and laboratory parameters

of the patients;

g. Haptoglobin genotypes did not influence any of the clinical and laboratory

parameters examined among the Nigerian patients.

121

10. REFERENCES

1. Reeds DC, Williams TN, Gladwin MT, Sickle-cell disease. Lancet. 2010;376:2018–31

2. Steinberg MH. Sickle cell anemia, the first molecular disease: overview of molecular

etiology, pathophysiology, and therapeutic approaches. Scientific World Journal.

2008;8:1295-324.

3. Galadanci N, Wudil BJ, Balogun TM, Ogunrinde GO, Akinsulie A , Hasan-Hanga F et

al. Current Sickle cell disease management practices in Nigeria. Int Health

2013;doi:10.1093/inthealth/iht022

4. Piel FB, Patil AP, Nyangiri OA, Mewahyu PWG, Teperely WH et al Global

epidemiology of sickle cell haemoglobin in neonates. Lancet 2013;381:142-51

5. Lervolino LG, Baldin PE, Picado SM, Viel AA, Campos LA. Prevalence of sickle cell

disease in national screening studies. Rev Bras Hematol Hemoter 2011;33(1):49-54

6. Makani J, Ofori-acquah, Nnodu O, Wonkman A, Ohene-frempong K. Sickle cell disease:

New opportunities and challenges in Afirca The Scientiffic World Journal 2013;

doi:10.1155/2013/193252

7. Steinberg MH. Genetic etiologies for phenotypic diversity in sickle cell anemia. The

scientific world journal 2009;9:46-67

8. Milton JN, Rooks H, Draser E, McCabe EL, Baldwin CT et al. Genetic determinants of

hemolysis in sickle cell anemia. Br J Hematol 2013; 161(2):270-78.

9. Nouraie M, Lee JS, Zhang Y, Kanias T, Zhao X et al The relationship between the

severity of hemolysis, clinical manifestations and risk of death in 415 patients with sickle

cell anemia in the US and Europe. Haematologica 2013; 98(3): 464-72

10. Minniti CP, Sable C, Campbell A, Rana S, Ensing C et al. Elevated triscuspid regugitant

jet velocity in children and adolescents with sickle cell disease:association with

hemolysis and haemoglobin oxygen desaturation. Haematologica 2009; 94(3):340-7

11. Donadee C, Raat NJ, Kanias T, Tejero J, Lee JS et al. Nitric oxide scavenging by red

blood cell microparticles and cell free haemoglobin as mechanism for the red cell storage

lesion. Circulation 2011; 124(4):465-76

12. Gladwin MT, Ofori-Acquah SF. Erythroid DAMPs drive inflammation in SCD. Blood

2014;123 (24):3689-90

13. Mendonca R, Silveira AA, Conran N. Red cell DAMPs and inflammation. Inflamm. Res.

2016;65:665-78

14. Gladwin MT, Kanias T, Kim-Shapiro DB. Hemolysis and cell free haemoglobin drive an

intrinsic mechanism for human disease. J Clin Invest. 2012;122(12):1205-8

15. Kato GJ, Steinberg MH, Gladwin MT. Intravascular hemolysis and the pathophysiology

of sickle cell disease. J Clin Invest. 2017;127(3):750-60

122

16. Adekile AD, Haider MZ. Haptoglobin gene polymorphisms in sickle cell disease patients

with different βS-globin gene haplotypes. Med Princ Pract 2010;19:447-50

17. Wobeto VP, Zaccariotto TR, Sonati MF. Polymorphism of human haptoglobin and its

clinical importance. Genetics and Molecular Biol 2008; 31(3):602-20

18. Santos MN, Bezerra MA, Dormigues BL, Zachariotto TR, Oliveira DM. Haptoglobin

genotypes in sickle cell disease. Genetic Test. Mol. Biomarker 2011;15(10):709-13

19. Schaer DJ, Buehler PW, Alayash AI, Belcher JD, Vercellotti GM. Hemolysis and free

haemoglobin revisited: exploring haemoglobin and hemin scavengers as novel class of

therapeutic proteins. Blood 2013;121(8):1276-84

20. Podmore AH. Haptoglobin for the treatment of sickle cell disease. Bioscience Hypothes

2008;1(1):59-63

21. Gosh S, Adisa OA, Chappa P, Tan Fang, Jackson KA, et al. Extracellular hemin crisis

triggers acute chest syndrome in sickle mice. J Clin Invest 2013;123(1):4809-20

22. Vinchi F, Franceschi L, Chigo A, Townes T, Cinimo J et al. Hemopexin theraphy

improves cardiovascular function by preventing heme induced endothelial toxicity in

mouse models of haemolytic diseases. Circulation 2013; 127:1317-29

23. Smith A, Mc Culloh RJ. Hemopexin and haptoglobin: allies against heme toxicity from

haemoglobin not contenders. Front. Physiol 2015;6:187 doi:10.3389/phys.2015.00187

24. Kato GJ, McGowan V, Machado RF, Little JA, Taylor VI J et al. Lactate dehydrogenase

as a biomarker of hemolysis-associated nitric oxide resistance, priapism, leg ulceration,

pulmonary hypertension, and death in patients with sickle cell disease. Blood 2006;

107(6):2279-85.

25. Kato GJ, Nouraie SM, Gladwin MT. Lactate dehydrogenase and hemolysis in sickle cell

disease. Blood 2013; 122(6): 1091-92.

26. Ballas SK. Lactate dehydrogenase and hemolysis in sickle cell disease Blood 2013;

121(1):243-44.

27. Mecabo G, Yamamoto M, Biassi TP, Figueiredo MS. Lactate dehydrogenase isoenzymes

3 and hemolysis in sickle cell anemia a possible correlation. Blood 2015; 125 (24):3821-

22

28. Mikobi TM, Lukusa Tshilobo P, Aloni MN, Mvumbi Lelo G, Akilimali PZ et al.

Correlation between the lactate dehydrogenase levels with laboratory variables in the

clinical severity of sickle cell anemia in Congolese patients. PLoS ONE

2015;10(5):e0123568. Doi:10.1371/journal.pone.0123568.

29. Adegoke SA, Adeodu OO, Adekile AD. Sickle cell disease clinical phenotypes in

children from South-Western Nigeria. Nig J Clin Pract. 2015;18(1):95-101

30. Rubin O, Canellini G, Delobel J, Lion N, Tissot JD. Red blood cell microparticles:

Clinical relevance. Transfus Med Hemother 2012;39:342-47

123

31. Camus SM, De Moraes JA, Bonnin P, Abbyad P, Le Jeune S et al. Circulating cell

membrane microparticles transfer heme to endothelia cells and trigger vaso-occlussions

in sickle cell disease. Blood 2015;125(24):3805-14

32. Van Beers EJ, Schaap MC, Berckmans RJ, Nieuwland R, Sturk A et al. Circulating

erythrocyte derived microparticle are associated with coagulation activation in sickle cell

disease. Haematological 2009;94:1513-19

33. Camus SM, Gausseres B, Bonnin P, Loufrani L, Grimaud L et al. Erythrocyte

microparticles can induce kidney vaso-occlussions in a murine model of sickle cell

disease. Blood 2012;120(25):5050-58

34. Reiter CD, Wang X, Trans-santos JE, Hogg N, Canon RO III et al. Cell free haemoglobin

limits nitric oxide bioavailability in sickle cell disease. Nat Med 2002;8:1383-89

35. Morris CR, Gladwin MT, Kato GJ. Nitric oxide and arginine dysregulation: A novel

pathway to pulmonary hypertension in haemolytic disorders. Curr Mol Med

2008;8(7):620-32

36. Morris CR, Kato GJ, Poljakovic M, Wang X, Blackweider WC et al. Dysregulated

arginine metabolism, hemolysis-associated pulmonary hypertension and mortality in

sickle cell disease. JAMA 2005;294:81-90.

37. Chiabrando D, Mercurio S, Tolosano E. Heme and erythopoesis: more than a structural

role. Hematologica 2014;99(6):973-83.

38. Uzunova VV, Pan W, Galkin O, Vekilov PG. Free heme and the polymerization of sickle

cell haemoglobin. Biophysical Journal 2010;99(6):1976-85.

39. Chen G, Zhang D, Fuchs TA, Mawani D, Wagner DD. Heme induced neutrophil

extracellular traps contribute to the pathogenesis of sickle cell disease. Blood

2014;123(24):3818-27

40. Sakamoto TM, Canalli AA, Traina F, Franco-Penteado C, Gambero S et al. Altered red

cell and platelet adhesion in haemolytic diseases: Hereditory spherocytosis, paroxysmal

nocturnal hemoglobinuria and sickle cell disease. Clinical Biochemistry 2013;46: 1798-

1803

41. Belcher JD, Nat KA, Vercelotti GM. Vasculotoxic and proinflammatory effects of

plasma heme: cell signalling and cytoprotective responses. ISRN Oxidative Medicine

2013;e831596 doi:dx,doi.org/10.1155/2013/831596

42. Guarda CCD, Santiago RP, Fluza LM, Aleluia MM, Ferreira JRD et al. Heme mediated

cell activation: the inflammatory puzzle of sickle cell anemia. Expert Rev Hematol

2017;10(6):533-41

43. Lubega I, Ndugwa CM, Mworozi EA, Tumwine JK. Alpha Thalassemia among sickle

cell anemia patients in Kampala, Uganda. Afri Hlth Sci 2015; 15 (2):682-89

124

44. Omoti CE. Beta thalassaemia traits in Nigerian patients with sickle cell anaemia. Journal

of Biomedical sciences 2005; 4(1):37-43

45. Figueiredo MS. The compound state: Hb S/beta-thalassemia. Rev Bras Hematol Hemoter

2015;37(3):150-152

46. Ataga KI, Cappellini MD, Rachmilewitz EA. Beta-Thalassemia and sickle cell anaemia

as paradigms of hypercoagulability. Br J Haematol 2007; 139:3-13

47. Belisario AR, Sales RR, Viana MB. Very mild forms of Hb S/beta+-thalassemia in

Brazilian children. Rev Bras Hematol Hemoter 2015;37(3): 198-201

48. Akinsheye I, Solovieff N, Ngo D, Malek A, Sebastiani P et al. Fetal haemoglobin in

sickle cell anemia: molecular characterisation of the unusally high fetal haemoglobin

phenotype in African American. Am J Hematology 2012;87(2):217-19

49. Bhagat S, Patra PK, Thakur AS. Association between Xmnl polymorphism and

hemoglobin F level in SCD patients from Chhattisgarh. Int J Biomed Sci. 2012;8(1):36-

9.

50. Wonkam A, Makani J, Ofori-Aquah S, Nnodu OE, Treadwell M et al. Sickle cell disease

and H3Africa: enhancing genomic research on cardivascular diseases in African patients.

Cardiovasc J Afr 2015;26(2): S50-S55

51. Xu J, Peng C, Sakaran VG, et al. Correlation of sickle cell disease in adult mice by

interference with fetal fetal haemoglobin silencing. Science 2011;334(6058):993-6

52. Bauer DE, Kamran SC, Lessard S et al. An erythroid enhancer of BCL11A subject to

genetic variation determines fetal hemoglobin level. Science 2013;342(6155):253-7

53. Makani J, Menzel S, Nkya S et al. Genetics of fetal hemoglobin in Tanzanian andBritish

patients with sickle cell anemia. Blood 2011;117(4):1390-92

54. Wonkam A, Ngo Bitoungui VJ, Vorster AA et al. Association of variants at BCL11A

and HBS1L-MYB with hemoglobin F and hospitalisation rates among sickle cell patients

in Cameroon. PLoS ONE 2014;9(3):e92506

55. Bitoungui VJ, Ngogang J, Wonkman A. Polymorphism at BCL11A compared to HBS1L-

MYB loci explains less of the variance in HBF in patients with sickle cell disease in

cameroon. Blood cells Mol Dis 2014;Nov 25. Pii:S1079-9796(14)00143-0

56. Monteiro WM, Val FF, Siqueira AM, Franca GP, Samparo VS et al. G6PD deficiency in

Latin America: systematic review of prevalence and variants. Mem Inst Oswaldo Cruz

2014;109(5):553-68

57. Orimadegun AE, Sodeinde O, G6PD status and severity of malaria anemia in Nigerian

children. J Infect Dev Ctries 2011;5:792-8

58. Benkerrou M, Alberth C, Couque N, Ba A Missud F et al. Impact of G6PD deficiency

on sickle cell anaemia expression in infancy and early childhood: A prospective study.

Br J Hematol 2013;163(6):646-54

125

59. Williams O, Gbadero D, Edowhorhu G, Brearley A, Slusher et al. G6PD deficiency in

Nigerian children. PLoS ONE 2013;8(9): e68800.doi:10.1371/journal.pone.

60. De-Araujo C, Migot-Nabias F, Guitard J, Pelleau S, Vulliamy T et al. The roles of the

G6PD376G/968C allele in glucose -6-phosphate dehydrogenase deficiency in the seerer

population of Senegal. Hematologica 2006;91:262-63

61. Cappelini MD, Fiorelli G, Glucose 6 phosphate dehydrogenase deficiency. Lancet 2008;

371:64-74

62. Clark TG, Fry AE, Aubums S, Campino S, Diakite M et al. Alleleic heterogeneity of

G6PD deficiency in West Africa and severe malaria susceptibility. Eur J Hum Genet

2009; 17:1080-85

63. Awamy BH. Effect of G6PD deficiency on sickle cell disease in Saudi Arabia. Indian J

Pediatr 1992;59:331-34

64. Simpore J, Ilboudo D, Damintoti D, Sawadogo L, Maria E et al. Glucose 6 phosphate

dehydrogenase deficiency and sickle cell disease in Burkina Faso. Pak J Biol Sci

2007;10(3): 409-14

65. Bernaudin F, Verlhac S, Cheuret S, Torres M, Coic L et al. G6PD deficiency, abscense

of alpha-thalassemia and hemoltic rate at baseline are significant independent risk factors

for abnormally high cerebral velocities In patients with sickle cell anemia. Blood

2008;112:4314-17

66. Nouraie M, Reading NS, Campbell A, Minnitti CP, Rana SR et al. Association of G6PD 202A,376G with lower Hb concentration but not increased hemolysis in patients with sickle

cell anemia. Br J Hematol 2010;150:218-25

67. Milton JN, Sebastiani P, Solovieff N, Hartley SW, Bhatriagar P et al. A genome-wide

association study of total bilirubin and cholelithiasis risk in sickle cell anemia. PLoS ONE

2012; 7(4): e34741.doi:10.1371/journal.pone.

68. Fertrin KY, Costa FF. Genomic polymorphisms in sickle disease: implications for

clinical diversity and treatment. Expert Rev. Hematol 2010;3(4):443-58

69. Sticova E, Jirsa M. New insights in bilirubin metabolism and their clinical implications.

World J Gastroenterol. 2013 Oct 14;19(38):6398-6407.

70. Horsfall LJ, Nazareth I, Pereira SP, et al. Gilbert's syndrome and the risk of death: a

population-based cohort study. J Gastroenterol Hepatol. 2013 Oct;28(10):1643-7. doi:

10.1111/jgh.12279

71. Horsfal LJ, Zeityln D, Tarekegn A, Bekele E, Thomas MG et al. Prevalence of clinically

relevant UGT1A alleles and haplotypes in African populations. Annals of Human

Genetics 2011; 238-248

126

72. Sebastiani P, Solovieff N, Hartley SW,Milton JN, Riva A. Genetic modifiers of the

severity of sickle cell anemia identified through a genomic-wide association study. Am J

Hematol 2010;85:29-35

73. Buch S, Schafmayer C, Volzke H, Seeger M, Miguel JF et al. Loci from a genome-wide

analysis of bilirubin levels are associated with gallstones risk and composition.

Gastroenterology 2010; 139:1942-51

74. Machado RF, Barst RJ, Yovetich NA, Hassel KL, Kato GJ et al. Hospitalisation for pain

inpatient with sickle cell disease with sidenafil for elevated TRV and low exercise

capacity. Blood 2011; 118: 855-64

75. Italia KY, Jijina FF, Jane D, Merchant R, Nadkarni AH et al. The effect of UGT1A

promoter polymorphism on bilrubin response to hydroxyurea in haemogloboinopathies.

Clin Biochem 2010; 43: 1329-32

76. Fertrin KY, Melo MB, Assis AM, Saad ST, Costal FF. UDP-glucuronosyl transgferase1

gene promoter polymorphism is associated with increased serum bilirubin levels and

cholecstectomy in patients with sickle cell anemia. Clini Genet 2003; 64:160-62.

77. Ballas S.K, Lieff S, Benjamin L.J, Dampier C.D, Heeney M.M, Hoppe C, et al.

Definitions of the phenotypic manifestations of sickle cell disease. Am J Hematol

2010;85:6–13.

78. Kato GJ. TRV: A physiological biomarker in sickle cell disease. Pediatr Blood Cancer

2012;58(6):831-2

79. Quin CT, Saraf SL, Gordeuk VR, Fitzhugh CD, Creary SE et al. Losartan for the

nephropathy of sickle cell anemia: A phase-2, multicentre trial. Am J Hematol

2017;92(9):E520-E528. Doi:10.10002/ajh.24810

80. QuantiChromTM Hemoglobin Assay Kit (DIHB-250), BioAssay Systems USA.

Available at www.bioassaysys.com Retrieved 8/9/2017

81. Quantichrom Heme assay kit (DIHM-250) Available at www.bioassaysys.com

Retrieved 8/9/2017

82. ab108859 Hemopexin human ELISA kit Available at www.abcam.com Retrieved

8/9/2017

83. Quantikine Human Haptoglobin (DHAPGO) ELISA kit, R & D Systems USA.

Available at www.RnDSystems.com Retrieved 8/9/2017

84. Wisgrill L, Lamm C, Hartmann J, Preibing F, Dragosits K et al. Cytometry Part A

2016;89A:663-72

85. Kato GJ, Gladwin MT, Steinberg MH. Deconstructing sickle cell disease:reappraisal of

the role of hemolysis in the development of clinical phenotype. Blood Rev.

2007;21(1):37-47

127

86. Quinn CT, Smith EP, Arbabi S, Khera PK, Lindsell CJ et al. Biochemical surrogate

markers of hemolysis do not correlate with directly measured erythrocyte survival in

sickle cell anemia. Am J Hematol 2016;91:1195-201

87. Muller-Eberhard U, Javid J, Liem HH, Hanstein A, Hanna M. Plasma concentration of

hemopexin, haptogloin, and heme in patients with various hemolytic diseases. Blood

1968;35(5):811-5

88. Van Beers EJ, Schaap MC, Berckmans RJ, Nieuland R, Sturk A et al. Circulating

erthrocyte derived microparticles are associated with coagulation activation in sickle cell

disease. Haematological;2009:94:1513-19

89. Setty BN, Kulkarni S, Rao AK, Stuart MJ. Fetal hemoglobin in sickle cell

disease:relationship to erythrocyte phosphatidylserine exposure and coagulation

activation. Blood 2000;96:1119-24

90. Liu C, Zhao W, Christ GJ, Gladwin MT, Kim-Shapiro DB. Nitric oxide scavenging by

red cell microparticles. Free Radic Biol Med 2013;65:1164-73

91. Ware RE. How I use hydroxyurea to treat young patients with SCA. Blood

2010;115(26):5300-11

92. Brittain EL, Janz DR, Austin O, Bastarache JA, Wheeler LA et al. Elevation of plasma

cell free hemoglobin in pulmnary hyperension. CHEST 2014;146(6):1478-85

93. Adisa OA, Hu Y, Ghosh S, Arycee D, Osunkwo I, Ofori-Acquah SF. Association between

plasma free haem and incidence of vaso-occlussive episodes and acute chest syndrome

in children with sickle cell disease. Br J Haematol 2013;162:702-5

94. Nebor D, Romana M, Santiago R, Vachiery N, Picot J et al. Fetal hemoglobin and

hydroxy carbamide modulate both plasma concentration and cellular origin of circulating

microparticles in sickle cell anemia children. Haematological 2013;98(6):862-67

95. Lezcano NE, Odo N, Kutlar A, Brambilla D, Adams RJ. Regular transfusion lowers

plasma free hemoglobin in children with sickle cell disease at risk of stroke.

Stroke.2006;37:1424-26

96. Piccin A, Murphy C, Eakins E, Kunde J, Corvetta D, Di Pierro A et al. Circulating

microparticles , protein C, free protein S, and enndothelial vascular markers in children

with sickle cell anemia. Journal of Extracellular vesicles 2015;4:28414

97. Gerotziafas GT, Van Dreden P, Chaari M, Galea V, Khaterchi A et al. The acceleration

of the propagation phase of thrombin generation in patients with steady state sickle cell

disease is associated with circulating erythrocyte –derived microparticles. Thromb

Haemost 2012;107:1044-52

98. Brunetta DM, De Santis GC, Silva-Pinto AC, Olivera de Olivera LC, Covas DT.

Hydroxyurea increases plasma concentrations of microparticles and reduces coagulation

128

activation and fibrinolysis in patients with sickle cell anemia. Acta Haematol

2015;133:287-94

99. Hebbel RP, Key NS. Microparticles in sickle cell anemia:Promise and pitfalls. Br J

Haematol 2017;174(1):16-29 doi:10.1111/bjh.14112

100. Westerman M, Porter JB. Red blood cell –derived microparticles:An overview.

Blood cells Mol Dis. 2016;59:134-9

101. Steinberg MH. Predicting clinical severity of sickle cell anemia. Br J Haematol.

2005;129:465-81

102. Minniti CP, Taylor JGt, Hildesheim M, O`Neal P, Wilson J et al. Laboratory and

echocardiography markers in sickle cell patients with leg leg ulcers. Am J hematol

2011;86:705-8

103. Minniti CP, Kato GJ. How we treat sickle cell patients with leg ulcers. Am J

Hematol 2016;91(1):22-30

104. Liem RI, Young LT, Thompson AA. Triscupid regurgitant jet velocity is

associated with hemolysis in children and young adults with sickle cell evaluated for

pulmonary hypertension. Haematologica 2007;92(11):1549-52

105. Mehari A, Alam S, Tian X, Cuttical MJ, Barnett CF. Hemodynamic predictors

of mortality in adults with sickle cell disease. Am J Respir Crit Care Med

2013;187(8):840-7

106. Gladwin MT, Sachdev V. Cardiovascular abnormalities in sickle cell disease. J

Am Coll Cardiol 2012;59(3):10.1016/j.jacc.2011.10900

107. Parent F, Bachir D, Inamo J, Lionnel F, Driss F. A hemodynamic study of

pulmonary hypertension in sickle cell disease. N Engl J Med 2011;365;44-53

108. Mehari A, Gladwin MT, Tian X, Machado RF, Kato GJ. Mortality in adults with

sickle cell disease and pulmonary hypertension. JAMA 2012;307(12):1254-56

109. Day TG, Drasar ER, Fulford T, Sharpe CC, Thein SL. Association between

hemolysis and albuminuria in adults with sickle cell anemia. Haematological

2012;97:201-5

110. Ataga KI, Derebail VK, Archer DR. The glomerulopathy of sickle cell disease.

Am J Hematol 2014;89:907-14

111. Eshbach ML, Kaur A, Rbaibi Y, Tejero J, Weisz OA. Hemoglobin inhibits

albumin uptake by proximal tubule cells: implications for sickle cell disease. Am J

Physiol 2017;312:C733-C740

112. Diop S, Sene A, Cisse M, Toure AO, Sow O et al. Prevalence and morbidity of

G6PD deficiency in sickle disease in the homozygote. Dakar Med 2000;50(2):56-60

129

113. Carpenter SL, Lieff S, Howard TA, Eggleston B, Ware RE. UGT1A1 promoter

polymorphisms and the development of hyperbilirubinemia and gallbladder disease in

children with sickle cell anemia. Am J Hematol. 2008;83:800-3

114. Schwertner HA, Vitek L. Gilbert syndrome,UGT1A1*28 allele, and

cardiovascular disease risk: Possible protective effects and therapeutic applications of

bilirubin. Atherosclerosis 2008;198:1-11

115. Ware R, Filston HC, Schultz WH, Kinney TR. Elective cholecystectomy in

young patients with sickle hemoglobinopathies. Ann Surg. 1988;208:17-22

116. Haberkern CM, Neumayr LD, Orringer EP, ett al. Cholecystectomy in sickle cell

anemia patients: Perioperative outcome of 364 cases from the National preoperative

Transfusion Study. Blood 1997;89:1533-42

117. Perera MA, Inocenti F, Ratan MJ. Pharmacology testing for Uridine Diphosphate

Glucuronosyltransferase 1A1 polymorphisms: Are we there yet? Pharmacotherapy.

2008;28(6):755-68

118. Chaouch L, Talbi E, Moumi I, Chaabene AB, Kalai M, Chaouachi D et al. Early

complication in sickle anemia children due to A (TA)n TAA polymorphism at the

promoter of UGT1A1 gene. Disease Markers;35(2):67-72

119. Chaouch L, Miniar K, Chaouchi D, Mallouli F, Hafsia R, Ben Ammar S et al.

Gilbert syndrome acts as a risk factor of developing gallstone among β hemoglobinopathy

Tunisians patients. Le Tunisie Medicale 2015;93(4):237-41

120. Hamad Z, Aljedai A, Halwani R, Alsultan A. UGT1A1promoter polymorphism

associated with serum bilirubin level in Saudi patients with sickle cell disease. Ann. Saudi

Med 2013;33(4):372-76

121. Haverfield EV, McKenzie CA, Forrester et al. UGT1A1 variation and gallstone

formation in sickle cell disease. Blood 2005;105(3):968-72

122. Damanhouri GA, Jarullah J, Marouf S, Hindawi SI, Mushtaq G, Kamal MA.

Clinical biomarkers in sickle cell disease. Saudi J Biological Sciences. 2015;22:24-31

123. Beutler E, Gelbart T, Demina A. Racial variability in the UDP-

glucuronosyltransferase1 (UGT1A1) promoter: a balanced polymorphism for the

regulation of bilirubin metabolism? Proc Natl Acad Sci1998;95:8170-74

124. Premawardhena A, Fisher CA, Liu YT et al. The global distribution of length

polymorphisms of the promoters of the glucuronoslytransferase 1 gene(UGT1A1):

hematologic and evolutionary implications. Blood Cells Mol Dis 2003;31:98-101

125. Fertrin KY, Goncalves MS, Saad STO, Costa FF. Frequencies of UDP-

Glucuronosyltransferase 1 (UGT1A1) gene promoter polymorphisms among distinct

groups from Brazil. Am J Med Genetics. 2002;108:117-19

130

126. Azevedo LA, Santin AP, Wagner SC, Zaleski CF, Bock H, Saiva-Pereira ML et

al. Prevalence of UGT1A1 Gene polymorphism in patients with haemolytic anemia in

Southern Brazil. Genet. Testing and Molecular Biomarkers. 2011;15 (1&2):107-110

127. AlFadhil S, Al-Jafer H, Hadi M, Al-Mutairi M, Nizam R. The effect of UGT1A1

promoter polymorphism in the development of hyperbilrubinemia and cholelithiasis in

hemoglobinopathy patients. PLoS ONE 8(10):e7768.doi:10.1371/journal.pone.0077681

128. Adekile A, Kutlar F, McKie K, Addington A, Elam D, Holley L. et al. The

influence of uridine diphosphate glucuronosyl transferase 1A promoter polymorphisms,

βs-globin gene haplotype and co-inherited α-thalssemia trait and HbF on steady state

serum bilirubin levels in sickle cell anemia. Eur J Haematol 2005;75:150-55

129. Huo D, Kim H, Adebamowo CA, Ogundiran TO, Akang E et al. Genetic

Polymorphisms in uridine diphosphoglucuronosyltransferase 1A1 and cancer risks in

Africans. Breast Cancer Res Treat. 2008;110:367-76

130. Kaplan M, Slusher T, Renbaum P, Essiet DF, Pam S, Levy-Lahad E et al. (TA)n

UDP-Glucuronosyltransferase1a1 Promoter polymorphism in Nigerian neonates. Pediatr

Res 2008;63:109-111

131. Chaar V, Keclard L, Diara JP, Leturdu C, Elion J, Krishnamoorthy R et al.

Association of UGT1A1 polymorphism with prevalence and age of onset of cholelithiasis

in sickle cell anemia. Haematological 2005;90(2):188-93

132. Adekile AD, Makanjuola D, Ultrasonography in children with sickle cell anemia.

Niger J Paediatr 1983;10:35-8

133. Nzeh DA, Adedoyin MA. Sonographic pattern of gallbladder disease in children

with sickle cell anemia. Pediatr Radiol 1989;19(5):290-2

134. Odunvbun ME, Adeyekun AA. Ultrasonic assessment of the prevalence of

gallstones in sickle cell disease children seen at the University of Benin Teaching

Hospital, Benin City, Nigeria. Niger J Paed. 2014;41(4):370-374

135. Darko R, Rodrigues OP, Oliver-Commey JO, Kotei CN. Gallstones in Ghanian

children with sickle cell disease. West Afr J Med 2005;24(4):295-8

136. Longo-Mbenza B, Ngiyulu R, Kizunda P, Kaluila M, Bikangi N. Gallbladder

disease in young Congolese with sickle cell anemia: an ultrasound survey. J Trop Pediatr

2004;5:73-7

137. Attalla BI, Sonographic findings in Sudanese children with sickle cell anemia.

J Diag Med Sonography 2010;26:276-80

138. Attalla BAI, Karrar ZA, Ibnouf G, Mohamed AO, Abdelwahab O, Nasir EM et

al. Outcome of cholelithiasis in Sudanese children with sickle cell anemia after a 13 years

follow up. Afr Hlth Sci 2013;13(1):154-59

131

139. Russo-Mancuso G, Romeo MA, Guardabasso V, Schiliro G. Survey of sickle cell

disease in Italy. Haematologica 1998;83:875-81

140. Passon RG, Howard TA, Zimmerman SA, Schultz WH, Ware RE. Influence of

bilirubin Uridine Diphosphate Glucuronoslytransferase 1A Promoter polymorphisms on

serum bilirubin levels and cholelithiasis in children with sickle cell anemia. J Pediatr

HematolOncol 2001;23:448-51

141. Gumeiro APS, Bellomo-Brandao MA, Costal-Pinto EAL.Gallstones in children

with sickle cell disease followed up at a Brazillian hematology center. Arq Gastroenterol

2008;45(4):313-18

142. Martins R, Morais A, Dias A, Soares I, Rolao C, Ducla-Soares JL et al. Eearly

modification of sickle cell disease clinical course by UDP-glucuronosyltransferase 1A1

gene promoter polymorphism. J Hum Genet 2008;524-28

143. Kumar S, Guha M, Choubey V, Maity P, Srivastava K, Puri S et al. Bilrubin

inhibits Plasmodium falciparum growth through the generation of reactive oxygen

species. Free Radic Biol Med 2008;44:602-13

144. Chaar V, Keclard L, Etienne-Julan M, Diara JP, Elion J, Krishnamoorthy R et al.

UGT1A1 Polymorphism outweighs the modest effect of deletional (-3.7kb) α-

Thalassemia on cholelithogenesis in sickle cell anemia. Am J Hematol 2006;81:377-79.

145. Figueiredo MS, Kerbauy J, Goncalves MS, Arruda VR, Saad STO et al. Effect of

α-thalassemia and β-globin gene cluster haplotypes on the hematological and clinical

features of sickle cell anemia in Brazil. Am J Hematol 1996;53:72-6

146. Adorno EV, Zanette A, Lyra I, Seixas MO, Reis MG et al. Clinical and molecular

characteristics of sickle cell anemia in the northeast of Brazil. Genetics Mol Biol

2008;31(3):621-25

147. Cardoso G, Guerreiro JF. Molecular characterization of sickle cell anemia in the

northern Brazilian state of Para. Am J Hum Biol 2010;22:573-77

148. Camilo-Araujo RF, Amancio OMS, Figueiredo MS, Cabanas-Pedro AC, Braga

JAP. Molecular analysis and associations with clinical and laboratory manifestations in

children with sickle cell anemia. Rev Bras Hematol Hemoter 2014;36(5):334-9

149. Mikobi TM, Lukusa PT, Aloni MN, Lumaka A, Akilimali PZ et al. Association

between sickle cell anemia and alpha thalassemia reveals a high prevalence of the α3.7

triplication in Congolese patients than in worldwide series. J Clin Lab Anal

2017;00:e22186:doi:10.1002/jcla.22186

150. Rumaney MB, Ngo Bitoungui VJ, Vorster AA, Ramesar R, Kengne AP et al. The

co-inheritance of alpha thalassemia and sickle cell anemia is associated with better

hematological indices and lower consultations rate in Cameroonian patients and could

132

improve their survival. PLoS ONE

2014;9(6):e100516.doi:10.1371/journal.pone.0100516

151. Renoux C, Connes P, Nader E, Skinner S, Faes C et al. Apha thalassemia

promotes frequent vaso-occlusive crises in children with sickle cell anemia through

haemorheological changes. Pediatr Blood Cancer

2017;64:e26455.doi:10.1002/pbc.26455

152. Falusi AG, Esan GJF, Ayyub H, Higgs DR. Alpha thalassemia in Nigeria: its

interactions with sickle cell disease. Eur J Haematol. 1987;38:370-75

153. Saraf SL, Akingbola TS, Shah BN, Ezekekwu CA, Sonubi O et al. Associations

of α-thalassemia and BCL11A with stroke in Nigerian, United States, and United

Kingdom sickle cell anemia cohorts. Blood Advances 2017; 1(11):693-8

154. Adekile AD, Liu JC, Sulzer AJ, Huisman THJ. Frequency of the α-thalassemia-

2 gene among Nigerian SS patients and its influence on malaria antibody titers.

Hemoglobin 1993;17(1):73-9

155. Wambua S, Mwacharo J, Uyoga S, Macharia A, Williams TN. Co-inheritance of

α-thalassemia and sickle cell trait results in specific effects on hematological parameters.

Br J Haematol 2006;133(2):206-9

156. Mockenhaupt FP, Ehrhardt S, Gellert S, Otchwemah RN, Dietz E et al. α+-

thalassemia protects African children from severe malaria Blood 2004;104:2003-6

157. Higgs DR, Aldridge BE, Lamb J, Clegg JB, Weatherall DJ et al. The interaction

of alpha thalassemia and homozygous sickle cell disease. N Engl J Med 1982;306:1441-

6

158. Steinberg MH. Predicting clinical severity of sickle cell anemia. Br J Haematol.

2005;129:465-81

159. Steinberg MH, Sebastiani P. Genetic modifiers of sickle cell disease. Am J

Hematol 2012;87(8):795-803

160. Koshy M, Entsuah R, Koranda A, Kraus AP, Johnson R et al. Leg ulcers in

patients with sickle cell disease. Blood 1989;74:1403-8

161. Minniti CP, Eckman J, Sebastiani P, Steinberg MH, Ballas SK. Leg ulcers in

sickle cell disease. Am J Hematol 2010;85:831-3

162. Serjeant GR, Higgs DR, Hambleton IR. Elderly survivors with homozygous

sickle cell disease. N Engl J Med 2007;356(6):642-3

163. Belisaro AR, Rodrigues CV, Martins ML, Silva CM, Viana MB. Coinheritance

of alpha thalassemia decreases the risk of cerebrovascular disease in a cohort of children

with sickle cell anemia. Hemoglobin 2010;34:516-29

164. Flanagan JM, Frohlich TA, Howard WH, Schultz C, Driscoll R et al. Genetic

predictors for stroke in children with sickle cell anemia. Blood 2011;117:6681-84

133

165. Filho IL, Leite AC, Moura PG, Ribeiro GS, Cavalcante FC et al. Genetic

polymorphisms and cerebrovascular disease in children with sickle cell anemia from Rio

de Janeiro, Brazil, Arq. Neuropsiquiatr 2011;69:431-5

166. Milner PF, Kraus AP, Sebes JI, Sleeper LA, Dukes KA et al. Sickle cell disease

as a cause of osteonecrosis of the femoral head. N Engl J Med 1991;325:1476-81

167. Mahadeo KM, Oyeku S, Taragin B, Rajpathak SN, Moody K et al. Increase

prevalence of osteonecrosis of the femoral head in children and adolescent with sickle

cell disease. Am J Hematol 2011;86(9):806-8

168. Borges E, Wening MR, Kimura EM, Gervasio SA, Costa FF, Sonati MF. High

prevalence of alpha thalassemia among individuals with microcytosis and hypochromia

without anemia. Braz J Med Biol Res 2001;34(6):759-62

169. Stevens GA, Finucane MM, De-Regil LM, Paciorek JC, Flaxman SR, Branca F.

Global, regional and national trends in hemoglobin concentration and prevention of total

and severe anemia in children and pregnant and non pregnant women 1999 – 2011. Lancet

Glob Health 2013;e16-e25.

170. Pierrot-Gallo BS, Vicari P, Matsuda SS, Adegoke SA, Mecabo G, Figueiredo

MS. Haptoglobin gene polymorphisms and interleukins 6 and 8 in patients with sickle

cell anaemia. Rev Bras Hematol Hemoter 2015;37(5):329-35

171. Santos MN. Haptoglobin: an emerging candidate for phenotypic modulation of

sickle cell anemia? Rev Bras Hematol Hemoter 2015;37(5)329-35

172. Platt OS, Thorington BD, Brambilla DJ, Milner PF, Rosse WF et al. Pain in sickle

cell disease. Rates and risk factors. N Engl J Med 1991;325:11-16

173. Platt OS, Brambilla DJ, Rosse WF, Milner PF, Castro O et al. Mortality in sickle

cell disease, Life expectancy and risk for early death. N Engl J Med 1994;330:1639-44

174. Lettre G, Sankaran VG, Bezerra MA, Araujo AS, Uda M et al. DNA

polymorphisms at the BCL11A, HBS1L-MYB, and β-globin loci associate with fetal

hemoglobin levels and pain crises in sickle cell disease. Proc. Natl. Acad.Sci. USA

2008;105(33):11869-74

175. Cardoso GL, Diniz IG, Martins da Silva AL, Cunha DA, Da Silva Junior JS et al.

DNA polymorphisms at the BCL11A, HBS1L-MYB, and Xmn1-HBG2 site loci

associate with fetal hemoglobin levels in sickle cell anemia patients from Northern Brazil.

Blood Cells Mol Dis 2014;53:176-9

134

APPENDICES

PULICATIONS IN CONGRESSES

Oladele Simeon Olatunya, Dulcineia Martins de Albuquerque, Daniela Pinheiro

Leonard, Kleber Y. Fertrin, Adekunle Adekile M, Fernando Ferreira Costa. Effects of

UGT1A1 polymorphism, glucose-6-phosphate dehydrogenase deficiency and

deletional (-3.7кb) α-thalassemia among young Nigerians with sickle cell anemia.

Blood 2017;130:2242 (American Society of Hematology 2017)

Oladele Simeon Olatunya, Carolina Lanaro, Ana Leda Longhini, Carla Fernanda Franco

Penteado, Kleber Y. Fertrin, Sara T.O. Saad, Adekunle Adekile, Fernando Ferreira

Costa. Red blood cells microparticles are associated with hemolysis markers and

leg ulceration in patients with sickle cell disease. Blood 2017;130:3510 (American

Society of Hematology 2017)

Oladele Simeon Olatunya, Dulcineia Martins de Albuquerque, Daniela Pinheiro

Leonard, Kleber Y. Fertrin, Adekunle Adekile M, Fernando Ferreira Costa. UGT1A1

promoter polymorphisms and their effects on clinical events and biomarkers of

Nigerian patients with sickle cell disease. FESBE Brasil 2017

135

MANUSCRIPTS SUBMITTED FOR PUBLICATION

MANUSCRIPT 1

Uridine Diphosphate Glucuronosyl Transferase 1A (UGT1A1) Promoter Polymorphism

is associated with hyperbilirubinemia and gallstone among young Nigerians with sickle

cell anaemia.

Authors:

Oladele Simeon Olatunya,a,b, Dulcineia Martins de Albuquerque,a Daniela Pinheiro

Leonard, a Kleber Y. Fertrin, a Adekunle Adekile,c, Fernando Ferreira Costa,a

Affiliations: aHematology and Hemotherapy Center, University of Campinas, São Paulo, Brazil bDepartment of Paediatrics, College of Medicine, Ekiti State University, Ekiti State, Nigeria cDepartment of Pediatrics, Faculty of Medicine, Kuwait University, Kuwait

Corresponding author and address for correspondence:

Oladele Simeon Olatunya Hematology and Hemotherapy Center (Hemocentro),

University of Campinas (UNICAMP),

Rua Carlos Chagas, 480

Barão Geraldo

Campinas 13083-970-SP, Brazil

Tel: +55 19 3521 8382

Email for correspondence: ladeletunya@yahoo.com

136

ABSTRACT

Background: (TA)n repeat sequence (rs8175347) of UGT1A1 gene promoter

polymorphism is associated with serum bilirubin levels and gallstones among different

sickle cell anaemia (SCA) populations. There are no data on UGT1A1 alleles and their

impact on Nigerian SCA patients.

Objectives: To determine the UGT1A1 genotypes, their association with laboratory

markers and clinical events among young Nigerians with SCA.

Patients and methods: One hundred and one SCA patients were studied and compared

with 64 healthy controls. The influence of UGT1A1 polymorphism on laboratory

parameters and clinical events of the patients was determined.

Results: Four (TA)n alleles:(TA)5, 6, 7, and 8 were found and associated with 10

genotypes:TA5/5, 5/6, 5/7, 5/8, 6/6, 6/7, 6/8, 7/7, 7/8, 8/8. The low- (TA) 7/7, 7/8, 8/8),

intermediate- (TA) 6/7, 6/8), and high-activity (TA) 5/5, 5/6, 5/7, 5/8, 6/5, 6/6,)

genotypes were found in 21.2%, 38.2%, and 40.6% participants respectively. There

were significant differences in serum bilirubin and lactate dehydrogenase (LDH) of the

patients when differentiated by the UGT1A1 genotype activity (p˂0.05). Asymptomatic

gallstones were found in 5.9% of patients and were significantly of low-activity

genotypes sub-group 5 (20%) vs 1(1.3%) p=0.0033. Although, bilirubin and HbF of

patients with gallstones were significantly different from those without gallstone, only

the serum bilirubin was associated with UGT1A1 genotypes on multivariate analysis

(p<0.0001).

137

Conclusion: This study confirms that UGT1A1 genotypes influence bilirubin levels and

development of gallstones among young Nigerians with SCA. Children with SCA

should be screened for UGT1A1 polymorphism and gallstones for holistic care.

Keywords: Sickle cell anaemia, Serum bilirubin, Gallstone, UGT1A1 polymorphism,

Nigeria.

INTRODUCTION

Sickle cell disease (SCD) is a common genetic disorder among Africans. Individuals

with the disease have variable clinical expression but homozygosity for the HbS gene,

also known as sickle cell anaemia (SCA), is the most severe form [1]. Children with

SCA have chronic hemolysis, leading to accumulation of serum bilirubin and

consequent gallstones [2]. Bilirubin is a tetrapyrol that results from the breakdown of

heme in red blood cells. At moderate levels, it is thought to protect against oxidative

stress and inflammatory injuries and some infectious diseases [3-5]. However, excessive

bilirubin levels, as seen in patients with chronic hemolysis, have been linked to

increased incidence of gallstones [1,2,5]. In children with SCA, this risk increases with

advancing age with a cumulative incidence of approximately 50% by adulthood and

some of them may need cholecystectomy [2,6,7].

Uridine diphosphate glucoronosyltransferase 1A isoform 1 (UGT1A1) is a member of

the superfamily of phase II conjugating enzymes that aids the elimination of bilirubin,

drugs and a vast variety of endogenous and exogenous substrates by adding a

glucuronide moiety to the substrates [8,9]. Genetic mutations resulting in absence or

severely reduced UGT1A1 activity leads to Criggler Najjar syndrome, which is

138

characterized by severely elevated serum bilirubin and increased risk of kernicterus

[10]. However, variations in (TA)n tandem repeat sequence within the TATA box

promoter region affect UGT1A1 gene expression and the activity of its (TA)n four

alleles, namely; (TA) 5, 6, 7, and 8 genotypes leading to moderate elevation of serum

bilirubin [9]. There is a negative association between the UGT1A1 and repeat length of

the four alleles attributable to the decreasing promoter activity acting via altered affinity

for the TATA–binding protein [9]. Based on this, the UGT1A1 (TA)n genotypes have

been classified into three sub groups namely; the high-activity genotype subgroup made

up of (TA)5/5, (TA)5/6, (TA)5/7, (TA)5/8, and (TA)6/6; intermediate-activity (TA)6/7,

and (TA)6/8, and low-activity group comprising (TA)7/7, (TA)7/8, (TA)8/8 [9].

Although homozygosity for (TA)7 alleles i.e. (TA)7/7 has been described generally as

Gilbert syndrome genetic hallmark, [2,11], there is an inverse relationship between the

serum bilirubin levels across these subgroups based on the degree of the genotype

activity such that, individuals with low-activity genotypes have elevated levels of serum

bilirubin and are therefore subjected to the modulating effects of higher serum bilirubin

levels including susceptibility to gallstone development [9,11].

Despite the huge burden of SCA in Africa [1], with Nigeria being the country with its

highest burden in the world [12], there is little understanding of the contributions of

genetic modifiers of SCA phenotypes in the country. To the best of our knowledge,

there are no data on the effects of UGT1A1 polymorphisms on the clinical expression of

Nigerian SCA patients. The aim of this study was to determine the UGT1A1 genotypes,

their association with laboratory parameters as well as clinical events in young Nigerian

patients with SCA.

PATIENTS AND METHODS

139

Study participants and settings

The study was conducted on 101 hydroxyurea-naïve children and adolescents with SCA

aged between 2 and 21years (median of 9 years) who are regular attendees at the

paediatric hematology unit of the Ekiti State University Teaching Hospital (EKSUTH),

Ado Ekiti, Ekiti State, in Southwest Nigeria. Sixty four healthy children who

accompanied their siblings to the paediatric haematology clinic or attended the

paediatrics outpatients` well-child clinic of the hospital served as the controls. To

qualify for inclusion, the SCA participants must have been on regular follow up at the

clinic for a minimum of one year prior to recruitment with up-to-date hospital records.

Participants with confirmed or suspected liver or other chronic diseases apart from SCA

were excluded. Also excluded were the few SCA patients on regular blood transfusion

and hydroxyurea therapies.

Ethical considerations

The study was approved by the Ethics and Research Committee of EKSUTH no:

A67/2016/03/003. Written informed consent of parents/caregivers as well as patients’

assents and consents were obtained as applicable after explaining the purpose of the

study to them in clear and plain language.

Data collection

Clinical and laboratory data

A tested chart review form was used to extract relevant information from the hospital

records of participants regarding their steady-state laboratory parameters and clinical

events. Average of at least two steady state results of laboratory parameters performed

between 3 to 6 months intervals by standard techniques were recorded for each

140

participants. The steady state parameters included the complete blood performed by

Sysmex KX21N Hematology analyser (Sysmex Corporation, Kobe, Japan). The serum

lactate dehydrogease (LDH), bilirubin, aspartate transaminase (AST), and alanine

transaminase (ALT) were measured with standard techniques. The quantitative

assessment of HbF, HbA, HbA2, HbS, and HbC was done by high performance liquid

chromatography (HPLC, Bio-Rad Variant D10, USA). Steady state was defined as

being free from any acute event(s) for at least one month and transfusion free for at least

four months [13].

Other information retrieved from patients’ charts included the biodata and details of the

clinical evolution of SCA such as number of bone pain crises requiring admission

and/or administration of opioids within the preceding one year, presence of leg ulcer,

priapism, overt osteonecrosis and/or overt stroke as well as presence of gallstone as

determined by serial abdominal ultrasound scans conducted on the patients as clinically

indicated. Gallstone was diagnosed on the basis of echodense images within the gall

bladder with acoustic shadowing or gravitational changes [14]. In addition, the clinical

records of the patients diagnosed with gallstone by ultrasound were examined for the

presence or absence of symptoms, and or treatment(s) for gallstone complications. The

definitions of clinical events were as previously described [15].

Genetic studies

These were carried out at the Centro de Hematologia e Hemoterapia (Hemocentro),

UNICAMP, Campinas, Sao Paulo State, Brazil. SCA was initially diagnosed by

hemoglobin electrophoresis and high performance liquid chromatography (HPLC) and

was confirmed by DNA studies.

141

The DNA of each participant extracted from each participant by Qiagen QIAamp DNA

Blood Mini Kit Cat No. 51104 Germany), was used to confirm the diagnosis of SCD by

polymerase chain reaction (PCR).

The rs8175347 identification was performed by Polymerase Chain Reaction (PCR) using

a forward primer 5'- (6-FAM) labelled (*) for detection by fragment analysis in capillary

electrophoresis system. The PCR reaction was prepared in 30 µL volume with 100ng of

genomic DNA; 1X Reaction Buffer (BIOTOOLS B&M Labs, Spain); 2.16mM MgCl2;

1.33 mM of dNTP mix; 133 nM of each primer (Integrated DNA Technologies,

Coralville, Iowa) named UGT1A1_*F: GTCACGTGACACAGTCAAAC and

UGT1A1_R: CAACAGTATCTTCCCAGCATG; and 1 U Taq DNA Polymerase

(BIOTOOLS B&M Labs, Spain). Thermal cycle conditions were as follows: preheating

at 96°C by 2 minutes, followed by 25 cycles of 96°C for 30 seconds, 58°C for 40 seconds,

and 72°C for 40 seconds. An ended step at 72°C for 30 min was performed to promote

adenylation of the PCR products. The PCR product (1 μL) was added to 8.7 μL Hi-Di

Formamide (Applied Biosystems, Carlsbad, CA) and 0.3 μL of a GeneScan™ 500 LIZ™

size standard (Applied Biosystems, Carlsbad, CA) and the fragments ranged from 197 -

203 bp, corresponding to (TA)5 - (TA)8 repeats, respectively, were separated by capillary

electrophoresis on a ABI3500 Genetic Analyzer and analysed by Gene Mapper v 4.1

Software (both Applied Biosystems, Carlsbad, CA).16,17 All DNA studies were carried

out blinded regarding the clinical and laboratory parameters of the participants. The

UGT1A1 genotypes were further classified into three subgroups namely: low,

intermediate and high activity subgroups as previously described [9].

Data Analysis

142

Statistical analysis was performed with the GraphPad Prism Program, version 5 for

Windows (San Diego, California, USA). The normal distribution of the quantitative

variables was verified by the Kolmogorov-Smirnov and Shapiro-Wilk tests. The

frequencies of variables were described and the significance of differences between

groups of patients was assessed using the Kruskal-Wallis analysis of variance

(ANOVA), chi-square, Mann-Whitney or Fisher`s exact tests as appropriate. Odd ratios

were obtained by applying logistic regression to determine the effects of UGT1A1

promoter polymorphism using the UGT1A1 genotype as the independent variable and

other outcomes of interest as the dependent variables. The test for Hardy-Weinberg

equilibrium was performed using the R-Project for statistical computing web tool

available at https://www.R-project.org/. Level of significance was set at P < 0.05 for

all statistical analyses.

RESULTS

The 101 patients with SCA consisted of 66 males and 35 females with median

age of 9, range 2 - 21years. The controls were made up of 19 sickle cell trait (HbAS)

and 45 haemoglobin HbAA, median age of 8 range 2 - 18years (p=0.4260), and 41

males. The SCA patients have been on follow up for a median of 4years, range 1-

14years.

UGT1A1 genotypes

Four (TA)n alleles: (TA)5, 6, 7, and 8 were found with gene frequencies of 0.11, 0.43,

0.41 and 0.05 respectively. The alleles were associated with 10 genotypes:TA5/5, 5/6,

5/7, 5/8, 6/6, 6/7, 6/8, 7/7, 7/8, 8/8 (Figure 1). The low (TA) 7/7, 7/8, 8/8), intermediate

(TA) 6/7, 6/8), and High (TA) 5/5, 5/6, 5/7, 5/8, 6/6,) activity genotypes were found in

35 (21.2%), 63 (38.2%), and 67 (40.6%) of the participants respectively. The low

143

activity genotypes were found in 25 (24.7%) patients and 10 (15.6%) of the controls

(P=0.1773). Homozygous (TA)n TA7/7 was found in 22 (21.7%) patients and 5 (7.8%)

controls p=0.018, (Table1) The observed genotype distributions of the patients and

control group were not significantly different from the values expected under Hardy-

Weinberg equilibrium (ᵡ2=15.10, df= 9, p=0.09), and (ᵡ2=11.86, df= 9, p=0.22),

respectively.

Comparison of serum bilirubin and other laboratory parameters between patients

and controls

There were significant differences between the patients and controls in all laboratory

parameters p<0.0001 respectively. (Table 2)

Effects of UGT1A1 genotype on serum bilirubin and other laboratory parameters

of patients

Both the total bilirubin and unconjugated bilirubin levels showed a trimodal pattern

across the UGT1A1 genotype subgroups with the low affinity genotype group having

the highest levels of serum bilirubin (p<0.0001). The LDH also showed this trimodal

pattern (p=0.0386). However, this was not demonstrated by the other remaining

laboratory parameters (Table 3).

Effects of UG1TA1 genotype on clinical events

Asymptomatic gallstones were found in 6 (5.9%) patients. Patients who had

gallstones significantly belonged to the subgroup with low activity genotypes 5 (20%)

vs 1(1.3%) p=0.0033, (Table 4). These were 2 females and 4 males. The two females

were aged 10 and 13 years. The males consisted of a 16-year-old boy with three others

144

aged 10, 13, and 15 years respectively. Four of the patients with gallstone had TA 7/7

genotypes, one TA 7/8, and TA 6/7 each.

Comparison of laboratory parameters between patients with and without

gallstones.

There were significant differences between the serum bilirubin and HbF levels

of patients with gallstones when compared with those without gallstone in general. No

difference was observed in the LDH and age of the two groups. Furthermore, when

those with gallstones were compared with age- and sex-matched peers within the same

UGT1A1 genotype subgroup, only serum bilirubin and HbF showed significant

differences between the two groups (Table 5).

Relationship between serum bilirubin and other parameters of patients

The total serum bilirubin correlated positively with the age of the patients ( r =0.238,

p=0.013), and their LDH levels (r = 0.218, p=0.028). Conversely, it showed a negative

correlation with the HbF levels (r = -0.210, p= 0.035). No significant correlation was

found between the total serum bilirubin and other biomarkers.

Relationship between UGT1A1 genotypes and other parameters by multivariate

analysis.

The unconjugated bilirubin was significantly associated with the low activity UGT1A1

genotypes (Adjusted Odd Ratio (1.08), 95% Confidence interval (1.034768 –

1.127873), P<0.0001). Also, significant association was found with the total bilirubin

when it was used in place of unconjugated bilirubin in the logistic regression model

(Adjusted Odd Ratio (1.05), 95% Confidence interval (1.029172 – 1.089832),

P<0.0001). No association was found with the other laboratory parameters.

145

DISCUSSION

There are gaps on the knowledge of impacts of genetic modifiers of SCA among

Africans. This is partly due to the lack of data and adequate infrastructure. Given that

genetic influences described among a specific population are not necessarily found in

others, it is pertinent that more studies are carried out among many cohorts to fully

understand the impact of genetic markers on SCA.

This study confirms the variability of bilirubin levels based on the activity of the

UGT1A1 genotypes as previously reported [2,4-5, 8-9,14]. However, we are not aware

of any previous study that has described the trimodal pattern of LDH based on UGT1A1

genotypes activity as found in this study. While the UGT1A1 modulation of serum

bilirubin levels is well understood [2,4,8,9,11,14,18-19], the exact mechanism through

which UGT1A1 could be associated with LDH is not clear. However, it may be possible

that there could be a link through hemolysis because both bilirubin and LDH are

derived from RBC and are markers of hemolysis [20]. UGT1A1 plays an important role

in haem catabolism upon its release from RBC and conversion to bilirubin [2,5,8,9].

Given the association of LDH with some phenotypes of SCA [20,21], we suggest there

is need for future research to unravel if there is any link between LDH and UGT1A1

activity.

Gilbert syndrome(GBS) has been described in individuals with the TA7/7 genotype and

[2,4,5,9,11,14,18], this was more common among our patients 22 (21.7%) vs 5 (7.8%)

controls (p=0.018). This difference could probably be due to the small number of the

controls. Nonetheless, the proportion of patients with TA7/7 genotype, in this study, is

higher than between 6 -11% described among Europeans [22,23], 11.7% in Saudi

146

population [18], and between 3-18% among Brazilians of different descents [24-26].

Similarly, it is higher than the 6% found among Kuwait SCA patients [27], and

between 5 and 11% described among other Africans [11,23]. However, this is lower

than 32% described among the SCA patients in the USA [28]. Nevertheless, the TA7/7

genotype prevalence in this study is comparable to between 18.2% and 20.3% earlier

described among Nigerians with non SCD-related illnesses [29,30]. Besides the TA7/7

genotype, other UGT1A1 genotypes found in this study have been described among

Africans [9,23]. These observations indicate that the UGT1A1 genotype is quite

variable among Nigerians and confirm the suggestions that the expression of the

UGT1A1 genotype variants is heterogeneous among Africans compared to the

Caucasians [9,22,23].

Our finding that the low-activity UGT1A1 genotype was associated with gallstones

confirms previous observations that SCA patients with the low-activity UGT1A1

genotypes especially the TA7/7, are at risk of developing gallstones [2,6,7,11,14,18].

Also, as found in this study, others [11,14,31], have reported that some other low-

activity genotypes like TA7/8 and TA8/8 predispose SCA patients to gallstones.

The proportion of patients with asymptomatic gallstones in this study, (5.9%), is

comparable to between 4-6% earlier reported among Nigerian children of similar age to

our SCA cohorts [32-34], This is also similar to the 4% found in Ghana [35], a close

neighbour to Nigeria. However, the gallstone prevalence in this study is lower than

between 9 – 58% that have been reported for some other African children with SCA of

similar age group [36-38]. Similarly, higher prevalence of between 26% and 30% was

reported for SCA cohorts of same age range as our patients in Italy [39] and the USA

[40] respectively. In the same vein, a higher prevalence of 45% was found among a

cohort of Brazilian children with SCD following a median follow up of 7 years [41].

147

These observations highlight the variations in propensity to gallstone development

among children with SCA from different backgrounds. Persistent higher serum bilirubin

is a risk factor for lithogenesis among SCA patients [2,18,25,27]. Also, diets and

environmental factors have also been implicated in explaining the variations in

propensity for gallstone formation among SCA [38,40]. Hence, the observed similarity

between the rates of gallstones in this study and earlier reports from Nigeria [32-34]

and Ghana [35], could be attributed to the two countries belonging to the West Africa

region and being perhaps, expose to similar diets and environmental factors compared

to their colleagues in other distant African parts, or in Europe, and Americas. In

addition to these earlier mentioned possibilities, genetic variabilities among Africans

and between Africans and Europeans and/or Americans could play some roles. The age

of onset and the apparent lack of initial symptoms attributable to gallstones, as found in

this study, have been reported [31,34,38, 40-42]. However, results of follow-up studies

have indicated that the prevalence of gallstones and its complications increase with age

of children with SCA [2,27,31,38,41] hence, the need to closely follow up these

patients.

Despite the observation that the UGT1A1 low -activity genotype is a leading factor in

hyperbilirubinemia and lithogenesis among SCA patients [2,6,7,42,43], the impact of

UGT1A1 polymorphism on the gallstone phenotype among the Nigerian SCA patients

was unknown prior to this study as none of the previous studies from Nigeria examined

the UGT1A1 of the patients [32-34]. Hence, to the best of our knowledge, the

contribution of this polymorphism to lithogenesis and hyperbilirubinemia among

Nigerian patients with SCA is being described for the first time in this study. This study

has shown that SCA children with the low-activity UGT1A1 genotypes had higher

148

bilirubin levels compared to others. In addition, we found that the low-activity

genotypes were associated with gallstones.

Although patients with gallstone in this study had higher serum bilirubin levels and

lower HbF levels as previously reported [28], following multivariate analysis, serum

bilirubin was the only laboratory parameter associated with the UGT1A1 genotype.

Therefore, it thus appears that the pathway to higher serum bilirubin and probably

gallstone development in our patients is not exclusively driven by hemolysis and the

ameliorating effect of HbF on hemolysis but also, by the influence of UGT1A1

genotype activity.

Beyond bilirubin metabolism and gallstone development, it has been suggested that

moderately elevated serum bilirubin can inhibit bacteria and Plasmodium falciparum

replication. Also it was hypothesised that, perhaps, the heterogeneity of UGT1A1

genotypes among Africans is a result of genetic evolution to confer selective advantage

by protecting from malaria in a way similar to other genetic traits like G6PD deficiency

and/or alpha thalassemia [5,9,23,44], hence, the occurrence of these possibilities could

be investigated through a thorough prospective study.

In conclusion, this study confirms for the first time that UGT1A1 genotypes are tightly

associated with bilirubin levels and development of gallstone among young Nigerians

with SCA. In addition, it also suggests that the pathway to elevated serum bilirubin and

gallstone development among our study cohorts seems not to be exclusively driven by

hemolysis. These observations are in agreement with earlier reports that UGT1A1

polymorphisms influence bilirubin metabolism [11,14,28,43,45], and highlight the

contribution of UGT1A1 polymorphisms, a non-globin genetic factor, to the clinical

149

manifestations of SCA patients. Children with SCA in developing countries should be

screened for UGT1A1 polymorphisms and gallstones in order to allow for holistic care.

FUNDING AND DISCLOSURES

This study was supported by grants No 2014/00984-3 from FAPESP, and grants No

2015/141693-0 from CNPq, Brazil.

The authors declare no conflict of interest with respect to this study. All authors

contributed to critical aspects of the study. OSO, AA, and FFC, conceived the study.

OSO Wrote the paper, participated in data collection, analysis and patients`

management and follow up. DMA & DPL, performed the genetic studies, OAF, & TSK

data collection & analysis. KYF manuscript review. All authors participated in

reviewing the manuscript for important intellectual contents and agreed to the final

version.

ACKNOWLEDGEMENTS

Authors acknowledge with thanks the supports received from participants and their

caregivers/parents during the study.

REFERENCES

[1]. Rees DC, Williams TN, Gladwin MT. Sickle cell disease. Lancet.

2010;376:2018-2031

[2]. Carpenter SL, Lieff S, Howard TA, Eggleston B, Ware RE. UGT1A1

promoter polymorphisms and the development of hyperbilirubinemia and

gallbladder disease in children with sickle cell anemia. Am J Hematol.

2008;83:800-803

150

[3]. Stocker R, Yamamoto Y, McDonagh A, Glazer A. Bilirubin is an

antioxidant of possible physiological importance. Science. 1987;235:1043-1046

[4]. Schwertner HA, Vitek L. Gilbert syndrome,UGT1A1*28 allele, and

cardiovascular disease risk: Possible protective effects and therapeutic

applications of bilirubin. Atherosclerosis 2008;198:1-11

[5]. Italia KY, Jijina FF, Jain D, Merchant R, Nadkarni AH, Mukherjee M, et al.

The effect of UGT1A1 promoter polymorphism on bilirubin response to

hydroxyurea therapy in haemoglobinopathies. Clinical Biochemistry

2010;43:1329-1332

[6]. Ware R, Filston HC, Schultz WH, Kinney TR. Elective cholecystectomy in

young patients with sickle hemoglobinopathies. Ann Surg. 1988;208:17-22

[7]. Haberkern CM, Neumayr LD, Orringer EP, ett al. Cholecystectomy in sickle

cell anemia patients: Perioperative outcome of 364 cases from the National

preoperative Transfusion Study. Blood 1997;89:1533-1542

[8]. Perera MA, Inocenti F, Ratan MJ. Pharmacology testing for Uridine

Diphosphate Glucuronosyltransferase 1A1 polymorphisms: Are we there yet?

Pharmacotherapy. 2008;28(6):755-768

[9]. Horsefall LJ, Zeitlyn D, Tarekegn A, Bekele E, Thomas MG, Bradman N et

al. Prevalence of clinically relevant UGT1A1 Alleles and haplotypes in African

population. Annals of Human Genetics 2011;75:236-246

[10]. Kadakol A, Ghosh SS, Sappal BS et al. Genetic lesions of bilirubin

uridine-diphosphoglucuronate glucuronosyltransferase (UGT1A1) causing

Criggler-Najar and Gilbert Syndromes: correlation of genotype to phenotype.

Hum Mutat 2000;16:297-306

151

[11]. Chaouch L, Talbi E, Moumi I, Chaabene AB, Kalai M, Chaouachi D et al.

Early complication in sickle anemia children due to A (TA)n TAA

polymorphism at the promoter of UGT1A1 gene. Disease Markers;35(2):67-72

[12]. Diakwu-Akinwumi IN, Abubakar SB, Adegoke SA, Adeleke S, Adewoye

O, Adeyemo T. et al Blood Transfusion Services for patients with Sickle Cell

Disease in Nigeria. Int. Health. 2016;8(5):330-335

[13]. Ballas SK. More definitions in sickle cell disease: Steady state v baseline.

Am J Hematol 2011;86:338

[14]. Chaouch L, Miniar K, Chaouchi D, Mallouli F, Hafsia R, Ben Ammar S et

al. Gilbert syndrome acts as a risk factor of developing gallstone among β

hemoglobinopathy Tunisians patients. Le Tunisie Medicale 2015;93(4):237-241

[15]. Ballas SK, Lieff S, Benjamin LJ, Dampier CD, Heeney MM, Hope C.

Definitions of phenotypic manifestations of sicle cell disease. Am J Hematol.

2010;85:6-13

[16]. Abou Tayoun AN, de Abreu FB, Lefferts JA, Tsongalis GJ. A clinical PCR

frangment analysis assay for TA repeat sizing in the UGT1A1 promoter region.

Clinca Chimica Acta 2013;422:1-4

[17]. Gupta N, Benjamin M, Kar A, Munjal SD, Sarangi AN, Dalal A et al.

Identification of promoter and Exonic variations, and functional characterisation

of a splice site mutation in Indian patients with unconjugated hyperbilrubinemia.

PLoS ONE 10 (12): e0145967, doi:10.1371/journal.pone.0145967

[18]. Hamad Z, Aljedai A, Halwani R, Alsultan A. UGT1A1promoter

polymorphism associated with serum bilirubin level in Saudi patients with sickle

cell disease. Ann. Saudi Med 2013;33(4):372-376

152

[19]. Haverfield EV, McKenzie CA, Forrester et al. UGT1A1 variation and

gallstone formation in sickle cell disease. Blood 2005;105(3):968-972

[20]. Damanhouri GA, Jarullah J, Marouf S, Hindawi SI, Mushtaq G, Kamal

MA. Clinical biomarkers in sickle cell disease. Saudi J Biological Sciences.

2015;22:24-31

[21]. Kato GJ, McGowan V, Machado RF, Little JA, Taylor VI J et al. Lactate

dehydrogenase as a biomarker of hemolysis-associated nitric oxide resistance,

priapism, leg ulceration, pulmonary hypertension, and death in patients with

sickle cell disease. Blood 2006; 107(6):2279-2285.

[22]. Beutler E, Gelbart T, Demina A. Racial variability in the UDP-

glucuronosyltransferase1 (UGT1A1) promoter: a balanced polymorphism for the

regulation of bilirubin metabolism? Proc Natl Acad Sci1998;95:8170-8174

[23]. Premawardhena A, Fisher CA, Liu YT et al. The global distribution of

length polymorphisms of the promoters of the glucuronoslytransferase 1

gene(UGT1A1): hematologic and evolutionary implications. Blood Cells Mol

Dis 2003;31:98-101

[24]. Fertrin KY, Goncalves MS, Saad STO, Costa FF. Frequencies of UDP-

Glucuronosyltransferase 1 (UGT1A1) gene promoter polymorphisms among

distinct groups from Brazil. Am J Med Genetics. 2002;108:117-119

[25]. Fertrin KY, Melo MB, Assis AM, Saad STO, Costa FF. UDP-

glucuronosyltransferase 1 gene promoter is associated with increased serum

bilirubin levels and cholecystectomy in patients with sickle cell anemia. Clin

Genet. 2003;64:160-162

[26]. Azevedo LA, Santin AP, Wagner SC, Zaleski CF, Bock H, Saiva-Pereira

ML et al. Prevalence of UGT1A1 Gene polymorphism in patients with

153

haemolytic anemia in Southern Brazil. Genet. Testing and Molecular

Biomarkers. 2011;15 (1&2):107-110

[27]. AlFadhil S, Al-Jafer H, Hadi M, Al-Mutairi M, Nizam R. The effect of

UGT1A1 promoter polymorphism in the development of hyperbilrubinemia and

cholelithiasis in hemoglobinopathy patients. PLoS ONE

8(10):e7768.doi:10.1371/journal.pone.0077681

[28]. Adekile A, Kutlar F, McKie K, Addington A, Elam D, Holley L. et al. The

influence of uridine diphosphate glucuronosyl transferase 1A promoter

polymorphisms, βs-globin gene haplotype and co-inherited α-thalssemia trait and

HbF on steady state serum bilirubin levels in sickle cell anemia. Eur J Haematol

2005;75:150-155

[29]. Huo D, Kim H, Adebamowo CA, Ogundiran TO, Akang E et al. Genetic

Polymorphisms in uridine diphosphoglucuronosyltransferase 1A1 and cancer

risks in Africans. Breast Cancer Res Treat. 2008;110:367-376

[30]. Kaplan M, Slusher T, Renbaum P, Essiet DF, Pam S, Levy-Lahad E et al.

(TA)n UDP-Glucuronosyltransferase1a1 Promoter polymorphism in Nigerian

neonates. Pediatr Res 2008;63:109-111

[31]. Chaar V, Keclard L, Diara JP, Leturdu C, Elion J, Krishnamoorthy R et al.

Association of UGT1A1 polymorphism with prevalence and age of onset of

cholelithiasis in sickle cell anemia. Haematological 2005;90(2):188-193

[32]. Adekile AD, Makanjuola D, Ultrasonography in children with sickle cell

anemia. Niger J Paediatr 1983;10:35-38

[33]. Nzeh DA, Adedoyin MA. Sonographic pattern of gallbladder disease in

children with sickle cell anemia. Pediatr Radiol 1989;19(5):290-292

154

[34]. Odunvbun ME, Adeyekun AA. Ultrasonic assessment of the prevalence of

gallstones in sickle cell disease children seen at the University of Benin

Teaching Hospital, Benin City, Nigeria. Niger J Paed. 2014;41(4):370-374

[35]. Darko R, Rodrigues OP, Oliver-Commey JO, Kotei CN. Gallstones in

Ghanian children with sickle cell disease. West Afr J Med 2005;24(4):295-298

[36]. Longo-Mbenza B, Ngiyulu R, Kizunda P, Kaluila M, Bikangi N.

Gallbladder disease in young Congolese with sickle cell anemia: an ultrasound

survey. J Trop Pediatr 2004;5:73-77

[37]. Attalla BI, Sonographic findings in Sudanese children with sickle cell

anemia. J Diag Med Sonography 2010;26:276-280

[38]. Attalla BAI, Karrar ZA, Ibnouf G, Mohamed AO, Abdelwahab O, Nasir

EM et al. Outcome of cholelithiasis in Sudanese children with sickle cell anemia

after a 13 years follow up. Afr Hlth Sci 2013;13(1):154-159

[39]. Russo-Mancuso G, Romeo MA, Guardabasso V, Schiliro G. Survey of

sickle cell disease in Italy. Haematologica 1998;83:875-881

[40]. Passon RG, Howard TA, Zimmerman SA, Schultz WH, Ware RE.

Influence of bilirubin Uridine Diphosphate Glucuronoslytransferase 1A

Promoter polymorphisms on serum bilirubin levels and cholelithiasis in children

with sickle cell anemia. J Pediatr HematolOncol 2001;23:448-451

[41]. Gumeiro APS, Bellomo-Brandao MA, Costal-Pinto EAL.Gallstones in

children with sickle cell disease followed up at a Brazillian hematology center.

Arq Gastroenterol 2008;45(4):313-318

[42]. Martins R, Morais A, Dias A, Soares I, Rolao C, Ducla-Soares JL et al.

Eearly modification of sickle cell disease clinical course by UDP-

155

glucuronosyltransferase 1A1 gene promoter polymorphism. J Hum Genet

2008;524-528

[43]. Milton JN, Sebastiani P, Solovieff N, Hartley SW, Bhatnagar P, Arking

DE et al. A Genome-wide association study of total bilirubin and cholelithiasis

risk in sickle cell anemia. PLoS ONE 2012; 7(4):e34741.

Doi10.1371/journal.pone.0034741

[44]. Kumar S, Guha M, Choubey V, Maity P, Srivastava K, Puri S et al.

Bilrubin inhibits Plasmodium falciparum growth through the generation of

reactive oxygen species. Free Radic Biol Med 2008;44:602-613

[45]. Chaar V, Keclard L, Etienne-Julan M, Diara JP, Elion J, Krishnamoorthy R

et al. UGT1A1 Polymorphism outweighs the modest effect of deletional (-3.7kb)

α-Thalassemia on cholelithogenesis in sickle cell anemia. Am J Hematol

2006;81:377-379.

156

Figure 1: UGT1A1 promoter genotypes found among participants.

A: TA5/TA5; B: TA5/TA6; C: TA5/TA7; D: TA5/TA8; E: TA6/TA6; F: TA6/TA7; G: TA6/TA8; H: TA7/TA7; I: TA7/TA8; J: TA8/TA8

Table 1: Allele and genotype frequencies of UGT1A1 promoter polymorphism among

participants

157

Variables SCA

N=101

AS

N=19

AA

N=45

Allelotypes Freq n (%) Freq n (%) Freq n (%)

(TA) 5 18 (11.7) 2 (6.7) 10 (12.5)

(TA)6 67 (43.5) 16 (53.3) 28 (35.0)

(TA)7 61 (39.6) 10 (33.3) 37 (46.2)

(TA)8 8 (5.2) 2 (6.7) 5 (6.3)

UGT1A1

Genotypes

SCA (N=101) AS (N=19) AA (N=45)

Genotypes Freq n (%) Freq n (%) Freq n (%)

TA5/5 0 (0) 0 (0) 1 (2.2)

TA5/6 9 (8.9) 0 (0) 1 (2.2)

TA5/7 6 (6.0) 1 (5.2) 8 (17.7)

TA5/8 3(2.9) 0 (0) 0 (0)

TA6/6 25 (24.7) 8 (42.1) 5 (11.1)

TA6/7 31 (30.7) 7 (36.8) 21 (46.7)

TA6/8 2 (2.0) 1 (5.2) 1 (2.2)

TA7/7 22 (21.7) 1 (5.2) 4 (8.9)

TA7/8 2 (2.0) 1 (5.2) 4 (8.9)

TA8/8 1 (1.0) 0 (0) 0 (0)

SCA (N=101) AS (N=19) AA (N=45)

UGT1A1

Genotypes by

degree of Activity

Freq n (%) Freq n (%) Freq n (%)

Low-Activity

genotypes

TA (7/7, 7/8, 8/8)

25 (24.7) 2 (10.5) 8 (17.8)

158

Intermediate-

Activity genotypes

(TA6/7, TA6/8)

33 (32.7) 8 (42.1) 22 (48.9)

High-Activity

genotypes

TA5/5 , TA5/6,

TA5/7, TA5/8, TA

6/6

43 (42.6) 9 (47.4) 15 (33.3)

159

Table 2: Comparison of laboratory markers between patients and controls

Parameters SCA (N=101)

Median (Range)

Controls (N=64)

Median (Range)

Test statistics P value

HbF (%) 8.4 (0.9 – 32.3) 0.9 (0 – 5.8)

p˂0.0001

LDH (IU/L) 771 (197 – 1860)

345(150 – 840) p˂0.0001

Hb Conc (g/dL) 7.4 (6.3 – 11.2) 11.6 (7.8 – 14.8)

p˂0.0001

WBC ( X 103/uL) 12.6 (6.1 – 29.3) 6.6 (4.5 – 14.4)

p˂0.0001

Platelet ( X 103/uL) 343 (118 – 832)

278 (108 – 591)

p˂0.0001

Total Bilirubin

(mg/dL)

1.8 (0.42– 8.1)

0.4 (0.1– 2.2)

p˂0.0001

Unconjugated

bilirubin (mg/dL)

0.8 (0.1 -6.3) 0.2 (0.03 -0.8) p˂0.0001

AST (IU/L) 42 (17 – 89)

27 (6 – 45) p˂0.0001

ALT (IU/L) 20 (4 – 77)

10 (4 – 51) p˂0.0001

NB:Test statistics =Mann-Whitney , Significant P values are indicated in bold fonts

160

Table 3: Influence of UGT1A1 genotype on laboratory parameters

Parameter a. Low activity UGT1A1 genotypes

N=25

b. Intermediate activity UGT1A1 genotypes

N=33

c. High activity UGT1A1 genotypes

N=43

P1 values

a versus (b+c)

Anova

P2

values

(a vs b vs c)

Biochemical and haematologic

Median (Range) Median (Range) Median (Range)

Total Bilirubin (mg/dl)

2.8 (1.2 -8.1) 1.7 (0.8 -4.7) 1.4 (0.4 -3.8) <0.0001* <0.0001**

Unconjugated Bilirubin (mg/dl)

1.8 (0.6 – 6.3) 0.8 (0.1 -3.6) 0.6 (0.1-2.8) <0.0001* <0.0001**

LDH (IU/L) 987(296-1860) 705 (233 -1489) 681 (197-1417) 0.0150* 0.0386**

AST(IU/L) 46 (18 -89) 41 (18 - 89) 37 (17 -89) 0.136* 0.3169**

ALT (IU/L) 25 (4-65) 18 (7- 58) 19 (4 -77) 0.1431* 0.279**

Hb conc (g/dl)

7.3 (6.3 -10) 7.5 (6.3 -11.2) 7.4(6.4 -10) 0.59*65* 0.822**

MCV(fl) 80.6 (66.9 -104.1)

82.3 (60.3 -101.5) 80.9 (55.9 -115) 0.9752* 0.8469**

RBC (x 1012/L)

2.7 (1.9 -4.1) 2.6 (2 -4.7) 2.9 (1.8 -4.8) 0.8427* 0.4438**

WBC (x 109/L)

13 (8.5 -26) 10.5 (6.1 -27) 13.3 (6.4 -29) 0.4814* 0.2997**

Platelet (x 109/L)

367 (118 -771) 342 (167 -593) 358 (118 -832) 0.4717* 0.7494**

HbF (%) 9.7 (1.3 -20.6) 7.3 (1.7 -24.4) 10.4 (0.9 -32) 0.521* 0.7466**

*=Mann-Whitney Test, **=Kruskal-Wallis one way analysis of variance, Significant p values are indicated in bold fonts.

161

Table 4: Influence of UGT1A1 genotype on clinical events of patients

Clinical events

a. Low activity UGT1A1 genotypes

N=25

b. Intermediate activity UGT1A1 genotypes

N=33

c. High activity UGT1A1 genotypes

N=43

P value

VOC per year 2 (0-6) 0 (0-6) 0 (0-6) 0.09*

Overt Stroke 1 1 2 1.000†

No overt stroke

24 32 41

Osteonecrosis 1 1 3 1.000†

No osteonecrosis

24 32 43

Leg ulcer 0 1 5 0.331†

No Leg ulcer 25 32 38

Gallstones 5 1 0 0.0033†

No Gallstone 20 32 43

Priapism (Male only event, N=67)

Priapism 1 2 2 1.000†

No Priapism 16 26 20

*=Mann-Whitney Test, †=Fisher`s exact test, Significant p values are indicated in bold fonts.

162

Table 5: Comparison of parameters in patients with and without gallstones

Parameter Patients with gallstones (N=6)

Median (Range)

Patients without gallstones (N=95)

Median (Range)

P value

Total Bilirubin (mg/dl)

6.4 (2.8 -8.1) 1.8 (0.4 -6.7) 0.0001*

Unconjugated Bilirubin (mg/dl)

4.7 (0.9 -6.3) 0.79 (0.1 -5) 0.0007*

LDH (IU/L) 1004 (592 -1860) 794 (197 -1750) 0.1263*

HbF (%) 4.7 (1.3 -6.8) 10.2 (0.9 -32) 0.0107*

Hb (g/dl) 7.1 (6.3 -8.8) 7.5 (6.2 -10) 0.4210*

Age in years 11.5 (8 -16) 9 (2-21) 0.1368*

Sex

Male (n=66) 4 62 1.000†

Female (n=35) 2 33

Parameter Patients with gallstones (N=6)

Median (Range)

Matched peers without gallstones within same UGT1A1 genotype activity group N=10

Median (Range)

P value

Total Bilirubin (mg/dl)

6.4 (2.8 -8.1) 2.2 (1.9- 3.2) 0.0023*

Unconjugated Bilirubin (mg/dl)

4.7 (0.9 -6.3) 1.2 (1.0 -2.0) 0.0020*

LDH (IU/L) 1004 (592 -1860) 890 (340 -1603) 0.628*

HbF (%) 4.7 (1.3 -6.8) 14.7 (4.2 -17.9) 0.022*

Hb (g/dl) 7.1 (6.3 -8.8) 8.0 (6.5 – 8.9) 0.137*

163

NB Significant P values are indicated in bold fonts, *=Mann-Whitney test, †=Fisher`s Exact

MANUSCRIPT 2

Title: Influence of Alpha Thalassemia on Clinical and Laboratory Parameters among

Nigerian Children with Sickle Cell Anemia

Running title: Alpha Thalassemia and Sickle Cell Anemia

Authors:

Oladele Simeon Olatunya,a,b, Dulcineia Martins de Albuquerque,a, Adekunle

Adekile,c, Fernando Ferreira Costa,a

Affiliations: aHematology and Hemotherapy Center, University of Campinas, São Paulo, Brazil bDepartment of Paediatrics, College of Medicine, Ekiti State University, Ekiti State, Nigeria cDepartment of Pediatrics, Faculty of Medicine, Kuwait University, Kuwait

Corresponding author and address for correspondence:

Oladele Simeon Olatunya Hematology and Hemotherapy Center (Hemocentro),

University of Campinas (UNICAMP),

Rua Carlos Chagas, 480

Barão Geraldo

Campinas 13083-970-SP, Brazil

Tel: +55 19 3521 8382

Email for correspondence: ladeletunya@yahoo.com

164

ABSTRACT

Alpha thalassemia modulates both the clinical and laboratory parameters of sickle cell

anemia (SCA). There is paucity of data on these among Nigerian SCA patients. This

study aimed to determine the prevalence of alpha thalassemia and its association with

laboratory and clinical manifestations among young Nigerians with SCA.

One hundred patients with SCA and 63 healthy controls were studied. Alpha

thalassemia genotyping was done by multiplex gap-PCR. Laboratory parameters,

including complete blood count, hemoglobin quantitation, serum lactate dehydrogenase

(LDH) and bilirubin were determined with standard techniques.

Alpha thalassemia was found in 42 (42.0%) patients compared to 24 (38.1%) controls

(p=0.7433) and all were due to the 3.7 κb α-globin gene deletion. Alpha thalassemia

associated with more frequent bone pain crisis, higher hemoglobin concentration, red

blood cell count and HbA2 level, and lower mean corpuscular volume, mean

corpuscular hemoglobin, and white blood cell count (WBC) in the SCA patients

(p˂0.05). There were 6 (10.3%) patients with leg ulcers, none of whom had alpha

thalassemia, p=0.0384.

This study confirms that coexistence of alpha thalassemia significantly influences both

the clinical and laboratory manifestations of young Nigerians with SCA. The

coexistence of this genetic modifier is associated with increased bone pain crisis and

protects against leg ulcers in SCA patients.

Keywords: Sickle cell anemia, Alpha thalassemia, Laboratory parameters, Clinical

manifestations, Children, Nigeria.

165

INTRODUCTION

Sickle cell anemia (SCA) is very common among Africans [1]. Globally, about 305,800

neonates are born with SCA annually and almost two-thirds of these occur in Africa1and

Nigeria has the highest burden in the world [2]. SCA is characterized by heterogeneous

clinical phenotypes due to the influence of genetic modifiers [3,4].

Alpha thalassemia (α-thal) trait secondary to one (αα/– α3.7) or two (-α3.7/-α3.7) α-gene

deletion has been found to, partly protect patients with SCA from early hyposplenism

[5], gallstones [6], stroke [7], priapism [8], glomerulopathy [9], and leg ulcers [7,10-12].

However, others have reported higher rates of painful crises [13], osteonecrosis [14] and

retinopathy [14] in SCA patients with coexistent α-thal. Although some studies have

described the possible effects of α-thal among patients with SCA in the developed world

[5-7, 9-14], there is paucity of information about the influence of this genetic factor in

African patients with SCA. The few available studies have established that the 3.7 κb α-

globin gene deletion is the common α-thal allele among Africans [15-17], and Nigerians

[18-20]. However, there is lack of consensus on its impacts on the laboratory and

clinical manifestations of patients with SCA [15-20].

Because of genetic variability in different populations, it is pertinent that more studies

are carried out among cohorts from different ethnic backgrounds to fully understand the

impact of genetic modifiers on SCA. The aim of this study was to assess the prevalence

of α-thal trait and its influence on laboratory and clinical events in a group of young

Nigerians with SCA.

166

PATIENTS AND METHODS

The study was conducted on 100 hydroxyurea-naive SCA children and adolescents who

were regular attendees at the pediatric hematology clinic of the Ekiti State University

Teaching Hospital (EKSUTH), Ado Ekiti, Ekiti State, Southwestern Nigeria. Sixty three

healthy children, who accompanied their siblings to or attended the outpatients` well-

child clinic served as controls. To qualify for inclusion, the SCA patients must have

been on regular follow up at the clinic for a minimum of one year prior to recruitment

with up-to-date hospital records. Participants with confirmed or other suspected chronic

diseases apart from SCA, were excluded. Also excluded were the few SCA patients on

regular blood transfusion and hydroxyurea. SCA was initially diagnosed by hemoglobin

electrophoresis and high performance liquid chromatography (HPLC) and was

confirmed by DNA studies.

Ethical considerations

Written informed consent of parents/caregivers as well as patients’ assents and consents

were obtained as applicable. The study was approved by the hospital`s Institutional

Ethics and Research Committee no: A67/2016/03/003.

Data collection

Clinical and laboratory data

A pre-tested chart review form was used to extract relevant information from the

patients’ hospital records regarding their steady-state laboratory parameters and clinical

167

events. An average of, at least, two steady-state results of laboratory parameters

performed between 3 to 6 months intervals were recorded for each participant. The

steady-state parameters included the complete blood count performed by Sysmex

KX21N Haematology analyser (Sysmex Corporation, Kobe, Japan); serum lactate

dehydrogease (LDH), bilirubin, and aspartate transaminase (AST) were measured with

standard techniques. HbF quantitation was done using HPLC (Bio-Rad Variant D10,

USA). Steady state was defined as being free from any acute event(s) for at least one

month and transfusion free for at least four months [21].

Other information retrieved from patients’ charts included their biodata. Data on the

clinical evolution of SCA among the patients were obtained and this included the

number of bone pain crises requiring hospital visit and administration of analgesics

within the preceding one year. In addition, presence or absence of complications of

SCA like leg ulcers, priapism, and overt osteonecrosis or overt stroke were examined.

The definitions of clinical events were as previously described [22].

Genetic studies

These were carried out at the Centro de Hematologia e Hemoterapia (Hemocentro),

UNICAMP, Campinas, Sao Paulo State, Brazil.

DNA extracted from each participant by Qiagen QIAamp DNA (Blood Mini Kit Cat

No. 51104 Germany), was used to confirm the diagnosis of SCA by polymerase chain

reaction (PCR).

Alpha thalassemia determination

Alpha-thalassemia (α3.7Kb deletion) was investigated by GAP-PCR according to Dode

et al [23]. Briefly, the PCR was performed in 25 μL reaction volume containing 100ng

168

of DNA sample; 1X α- Buffer (Tris-HCl 2M (pH 8.6), (NH4)2 SO4 1M, MgCl2 1M,

Na2EDTA 0.2M, BSA and β-mercaptoetanol 14.3M); 1X DMSO; 0.3mM of dNTP

mix; 0.2 µM of each primer (C2: CCATGCTGGCACGTTTCTGA e C10:

GATGCACCCACTGGACTCCT); 1U of GoTaq® Flexi DNA Polymerase (Promega

Corporation, Madison, USA). Thermal cycle conditions were as follows: preheating at

94°C by 5 minutes, followed by 35 cycles of 94°C for 45 seconds, 56°C for 1 minute,

and 72°C for 2 minutes and a final extension at 72°C for 7 minutes was performed.

After electrophoresis in a 1.2% agarose gel a fragment of 2.1Kb could be observed for

normal alleles and 1.9Kb fragment for deleted alleles (-α3.7Kb).

Alpha-thalassemia (α4.2Kb deletion) was investigated by Multiplex-PCR according to

Oron-Karni et al [24]. Briefly, the PCR was performed in 25 μL reaction volume

containing 100ng of DNA sample; 1X α- Buffer (Tris-HCl 2M (pH 8.6),

(NH4)2 SO4 1M, MgCl2 1M, Na2EDTA 0.2M, BSA and β-mercaptoetanol 14.3M); 1X

DMSO; 0.3mM of dNTP mix; 0.4 µM of primer

(P71: TACCCATGTGGTGCCTCCATG e 0.3 µM of each primer P72:

TGTCTGCCACCCTCTTCTGAC and P52: CCTCCATTGTTGGCACATTCC; 1U

of Taq DNA Polymerase (Invitrogen, Carlsbad,CA). Thermal cycle conditions were as

follows: preheating at 94°C by 5 minutes, followed by 35 cycles of 94°C for 45

seconds, 60°C for 1 minute, and 72°C for 2 minutes and a final extension at 72°C for 7

minutes was performed. After electrophoresis in a 1.2% agarose gel a fragment of 1596

bp could be observed for deleted alleles (-α4.2Kb) and 233 bp as an internal control to

verify the quality of DNA sample. All DNA studies were carried out blinded regarding

the clinical and laboratory parameters of the participants.

169

Data Analysis

Statistical analysis was performed with the GraphPad Prism Program, version 5 for

Windows (San Diego, California, USA). The normal distribution of the quantitative

variables was verified by the Kolmogorov-Smirnov and Shapiro-Wilk tests. The

frequencies of variables were described and the significance of differences between

groups of patients was assessed using the Kruskal-Wallis analysis of variance

(ANOVA), chi-square, Mann-Whitney or Fisher`s exact tests as appropriate. Level of

significance was set at P < 0.05

RESULTS

The patients consisted of 66 males and 34 females with a median age of 8.5, range 2 –

21 years. The controls were made up of 22 individuals with HbAS and 41 with HbAA,

median age of 8, range 2- 18years. They consisted of 36 males and 27 females. The

patients had been on follow up for a median of 4 years, range 1.5 -14years. There were

no differences in the socio-biographic data of patients and controls (Table 1).

Allele frequency of alpha thalassemia among patients and controls

None of the participants had the Alpha-thal trait 4.2Kb α-globin gene deletion.

Alpha-thal trait (3.7 κb α-globin gene deletion) was found in 41 (41.0%) patients; 34

with heterozygous deletion (αα/– α3.7), and 7 with homozygous deletion (-α3.7/-α3.7),

while 58 (58.0%) patients had normal genotype (αα/αα). Twenty-four controls

comprising 8 (36.0%) HbAS, and 16 (39.0%) HbAA had α-thal trait and these were all

heterozygous (αα/– α3.7). Also, 38 controls made up of 13 (62.0%) HbAS and 25

(61.0%) HbAA had normal genotype (αα/αα). Of the 200 patients’ chromosomes

170

analysed, 150 were αα, 49 were –α, and one was ααα, thus giving gene frequencies of

0.75 (αα), 0.25 (-α), and 0.005 (ααα) respectively. Similarly, the gene frequencies

among the controls were HbAS - 0.82 (αα), 0.18 (-α) and HbAA- 0.80 (αα), 0.20 (-α)

respectively, with no gene triplication (ααα) detected and these were not significantly

different from the frequencies among the patients (Table 1). Taken together, the

prevalence of α-thal was not different across the groups (SCA 42 (42%) vs HbAS 8

(36.0) vs HbAA 16 (39.0), ᵡ2=0.2866, df=2, p=0.866).

Comparison of laboratory parameters between patients and controls

There were significant differences between the patients and controls in all laboratory

parameters p<0.05 except mean corpuscular hemoglobin (MCH). (Table 2)

Effects of alpha thalassemia on hematological indices and other laboratory

parameters among patients

Co-inheritance of α-thal was significantly associated with higher hemoglobin

concentration (Hb), red blood cell count (RBC) and HbA2 level. On the contrary, it was

associated with lower mean corpuscular volume (MCV) and mean corpuscular

hemoglobin (MCH), while the white blood cell count (WBC) was significantly lower in

patients with homozygous 3.7 kb α-globin gene deletion compared to the two other

groups. No significant differences were observed across α-thal genotypes in the other

laboratory parameters (Table 3).

Effects of alpha thalassemia on clinical events among patients

Sixty-one (61.0%) patients had bone pain episodes, 6 (6.0%), 5 (5.0%) each had overt

stroke and osteonecrosis while 5 (7.5%) males had priapism.

171

The rate of painful crisis was higher in patients with α-thal compared to those without it

(p<0.0001). However, the presence of α-thal protected against leg ulcer as none of the

six patients with leg ulcer had α-thal 0 (0.0%) vs 6 (10.3%), p=0.0384. No significant

association was found in relation to other clinical events with respect to the presence or

absence of α-thal (Table 4).

DISCUSSION

There is scarcity of information on the roles of genetic modifiers on SCA in Africa

despite its huge burden in the continent and this makes comparison of data difficult.

The two α-globin genes are present within a 4-kb duplicated region leading to the

possibility of rearrangements including deletions and triplications with many attendant

downstream consequences [24]. This study confirmed earlier reports that the 3.7 κb α-

globin gene deletion is the common α-thal allele among Africans [15-20] as none of the

participants had the 4.2Kb α-globin gene deletion.

The prevalence of alpha thalassemia among patients in this study (41%) is higher than

between the 13 and 28% described among patients in the Americas [25-28].This is

however lower than the 60% and 77% described among Congolese [17], and Ugandans

[16] respectively. It is comparable to the 37.3% among Cameroonians [15], and the

46% among SCA patients in France [13]. It is also in agreement with the 40% to 42.5%

earlier described among Nigerians SCA patients [19,20]. The high prevalence of α-thal

in this study, and others from Africa [15-17], may be due to the selective advantage in

conferring protection and survival against malaria [18,29,30]. The higher prevalence of

α-thal in central Africa [16,17], relative to this study, may derive from its vital impact

on survival of SCA, which is more severe in this part of Africa [16,17].

172

The observed hematological profile of patients with α-thal in this study has been

previously described [10,13,18-20,25,26,28]. The increased Hb and red blood cell count

are due to the decrease in the intracellular concentration of HbS, and number of dense

red blood cells. These, in turn, lead to increased red blood cell deformability and

decreased rates of both HbS-induced RBC polymerization and hemolysis.3,7,14 Similarly,

the lack of influence of α-thal on the HbF levels of patients in this study, has been

previously reported [25,26,28].

Beyond these, alpha thalassemia hematologic profile also mimicks that of iron

deficiency anemia and individuals with alpha thalassemia may be misdiagnosed for iron

deficiency and mistreated with iron therapy [31]. The contributions of iron deficiency to

the clinical phenotype of SCA is still not fully explored. Nevertheless, some studies

have found that iron deficiency anemia do occur among children with SCA [32,33].The

similarities in the hematological profile of patients with iron deficiency and alpha

thalassemia make it important that alpha thalassemia status of children with SCA are

determined in order to prevent misdiagnosis and mistreatment for iron deficiency. This

is more important for patients in the developing countries given the high prevalence of

both conditions among children in these parts of the world [1,2,34].

Contrary to some earlier reports that alpha thalassemia did not influence the clinical

events in SCA [26,28], the presence of α-thal in this study, was associated with higher

rates of bone pain crisis compared to those without α-thal. This finding is in keeping

with that of Renoux et al [13], among children with SCA in France and other previous

reports [7,14]. Vaso-occlusive crisis (VOC) manifesting as bone pain crisis is a

common complication of SCA and, is thought to be associated with increased blood

viscosity [7,13]. Studies have demonstrated that SCA patients with α-thal have

increased blood viscosity because of relatively higher levels of Hb and/or hematocrit,

173

among other mechanisms [3,7,13,14]. In this study, our SCA patients with α-thal, had

higher Hb that could lead to increased blood viscosity. It is therefore conceivable that,

they could have higher rates of painful crisis.

Furthermore, as found in this study, authors from different parts of the world have

reported that α-thal protects against leg ulcer in SCA patients [10,11,17]. However,

some have reported that, no such association exist [15,16]. These observations reflect

lack of consensus on the roles of α-thal in some phenotypes of SCA. Leg ulcer is a

chronic and debilitating complication that is thought to be associated with very severe

phenotype of SCA [3,12,35-37]. Therefore there’s a need for more studies from Africa

to explain the relationship between α-thal and leg ulcer, and other SCA phenotypes.

The observations that α-thal is associated with lower frequency of stroke in children

with SCA [38,39] was not sustained in this study. On the contrary, one previous study

[20] found that, in combination with BCL11A variants, α-thal was associated with

increased risk for stroke in older SCA patients. The lack of association between α-thal

and stroke in this study, is in keeping with report by Filho et al [40]. Nonetheless, these

findings need to be interpreted with caution because of differences in age of study

cohorts and possibility of survival bias. In addition, stroke was determined in this study

by overt clinical history and was, in most cases, not confirmed by magnetic resonance

imaging or angiography and this could lead to detection bias and exclusion of cases of

silent infarcts from our analysis. These observations further underscore the need for

more studies to clearly define the prevalence and associations of α-thal with clinical

manifestations among Nigerian SCA patients.

The lack of association between α-thal and osteonecrosis in this study contrasts with the

report by Milner et al [41] on children enrolled in the cooperative study of sickle cell

174

disease in the USA where they found that both α-thal and increasing age were

significantly associated with osteonecrosis. Possible explanations for our findings

included the fact that our study participants were fewer and younger compared to that

by Milner et al [41]. The youngest patient in this study was two years old, while theirs`

was five years old. In addition, only five percent of our patients had osteonecrosis

compared to 9.8% in their study.

Osteonecrosis is a disabling and severe complication of SCA associated with

impairments of both functional activities and growth in children. In SCA, it is thought

to be due to bone microcirculation disturbance in the patients and can be observed in

children with SCA as young as five years old with an increasing incidence throughout

childhood and adolescence, peaking in early adulthood [42].

In conclusion, this study shows that coexistence of α-thal influences the hematologic

parameters of Nigerian children with SCA. It also showed that α-thal was associated

with increased rates of bone pain crisis and seems to protect against the occurrence of

leg ulcer.

FUNDING AND DISCLOSURES

This study was supported by grants No 2014/00984-3 from FAPESP, and grants No

2015/141693-0 from CNPq, Brazil.

The authors declare no conflict of interest with respect to this study.

ACKNOWLEDGEMENTS

Authors acknowledge with thanks the supports received from participants and their

caregivers/parents during the study.

REFERENCES

175

[1]. Piel FB, Patil AP, Howes RE et al. Global epidemiology of sickle

hemoglobin in neonates: a contemporary geostatistical model-based map and

population estimates. Lancet 2013; 381:142-51

[2]. Diaku-Akinwumi IN, Abubakar SB, Adegoke SA et al. Blood transfusion

services for patients with sickle cell disease in Nigeria. Int Health

2016;8(5):330-5

[3]. Kato GJ, Gladwin MT, Steinberg MH. Deconstructing sickle cell disease:

Reappraisal of the role of hemolysis in the development of clinical

subphenotypes. Blood Rev.2007;21:37-47

[4]. Goncalves MS. Comment on molecular analysis and association with

clinical and laboratory manifestations in children with sickle cell anemia. Rev

Bras Hematol Hemoter2014;36(5):315-18

[5]. Adekile AD, Tuli M, Haider K et al. Influece of α thalassemia trait on spleen

function in sickle cell anemia patients with high HbF. Am J Hematol 1996;53:1-

5

[6]. MZ Haider, Ashedu S, Aduh P, Adekile AD. Influence of α-thalassemia on

cholelithiasis in SS patients with elevated HbF. Acta Haematol 1998;100:147-50

[7]. Steinberg MH. Predicting clinical severity of sickle cell anemia. Br J

Haematol. 2005;129:465-81

[8]. Ballas SK, Larner J, Smith ED, Surrey S, Schwartz E, Rappaport EF.

Rheology predictors of the severity of the painful crisis. Blood.1988;72:1216-23

[9]. Lamarre Y, Romana M, Lemonne N et al. Alpha thalassemia protects sickle

cell anemia patients from macro-albuminuria through its effects on red blood

cell rheological properties. Clin Hemorheal Micro-circ. 2014;57:63-72

176

[10]. Higgs DR, Aldridge BE, Lamb J et al. The interaction of alpha thalassemia

and homozygous sickle cell disease. N Engl J Med 1982;306:1441-6

[11]. Koshy M, Entsuah R, Koranda A et al. Leg ulcers in patients with sickle

cell disease. Blood 1989;74:1403-8

[12]. Minniti CP, Eckman J, Sebastiani P, Steinberg MH, Ballas SK. Leg ulcers

in sickle cell disease. Am J Hematol 2010;85:831-3

[13]. Renoux C, Connes P, Nader E et al. Alpha thalassemia promotes frequent

vaso-occlusive crises in children with sickle cell anemia through

haemorheological changes. Pediatr Blood Cancer

2017;64:e26455.doi:10.1002/pbc.26455

[14]. Steinberg MH, Sebastiani P. Genetic modifiers of sickle cell disease. Am J

Hematol 2012;87(8):795-803

[15]. Rumaney MB, Ngo Bitoungui VJ, Vorster AA et al. The co-inheritance of

alpha thalassemia and sickle cell anemia is associated with better hematological

indices and lower consultations rate in Cameroonian patients and could improve

their survival. PLoS ONE 2014;9(6):e100516.doi:10.1371/journal.pone.0100516

[16]. Lubega I, Ndugwa CM, Mworozi EA, Tumwine JK. Alpha thalassemia

among sickle cell anemia patients in Kampala, Uganda. Afr Health Sci

2015;15(2):682-89

[17]. Mikobi TM, Lukusa PT, Aloni MN et al. Association between sickle cell

anemia and alpha thalassemia reveals a high prevalence of the α3.7 triplication in

Congolese patients than in worldwide series. J Clin Lab Anal

2017;00:e22186:doi:10.1002/jcla.22186

177

[18]. Adekile AD, Liu JC, Sulzer AJ, Huisman THJ. Frequency of the α-

thalassemia-2 gene among Nigerian SS patients and its influence on malaria

antibody titers. Hemoglobin 1993;17(1):73-9

[19]. Falusi AG, Esan GJF, Ayyub H, Higgs DR. Alpha thalassemia in Nigeria:

its interactions with sickle cell disease. Eur J Haematol. 1987;38:370-75

[20]. Saraf SL, Akingbola TS, Shah BN et al. Associations of α-thalassemia and

BCL11A with stroke in Nigerian, United States, and United Kingdom sickle cell

anemia cohorts. Blood Advances 2017; 1(11):693-8

[21]. Ballas SK. More definitions in sickle cell disease: Steady state v baseline.

Am J Hematol 2011;86:338

[22]. Ballas SK, Lieff S, Benjamin LJ, Dampier CD, Heeney MM, Hope C.

Definitions of phenotypic manifestations of sicle cell disease. Am J Hematol.

2010;85:6-13

[23]. Dode C, Krishnamoorthy R, Lamb J, Rochette J. Rapid analysis of – α3.7

thalassaemia and αααanti3.7 triplication by enzymatic amplification analysis. Br J

Haematol 1993;83:105-111

[24]. Oron-Karni V, Filon D, Oppenheim A, Deborah R. Rapid detection of the

common Mediterranean α-globin deletions/rearrangements using PCR. Am J

Hematol 1998;58:306-310

[25]. Figueiredo MS, Kerbauy J, Goncalves MS et al. Effect of α-thalassemia

and β-globin gene cluster haplotypes on the hematological and clinical features

of sickle cell anemia in Brazil. Am J Hematol 1996;53:72-6

[26]. Adorno EV, Zanette A, Lyra I et al. Clinical and molecular characteristics

of sickle cell anemia in the northeast of Brazil. Genetics Mol Biol

2008;31(3):621-25

178

[27]. Cardoso G, Guerreiro JF. Molecular characterization of sickle cell anemia

in the northern Brazilian state of Para. Am J Hum Biol 2010;22:573-77

[28]. Camilo-Araujo RF, Amancio OMS, Figueiredo MS, Cabanas-Pedro AC,

Braga JAP. Molecular analysis and associations with clinical and laboratory

manifestations in children with sickle cell anemia. Rev Bras Hematol Hemoter

2014;36(5):334-9

[29]. Wambua S, Mwacharo J, Uyoga S, Macharia A, Williams TN. Co-

inheritance of α-thalassemia and sickle cell trait results in specific effects on

hematological parameters. Br J Haematol 2006;133(2):206-9

[30]. Mockenhaupt FP, Ehrhardt S, Gellert S et al. α+-thalassemia protects

African children from severe malaria Blood 2004;104:2003-6

[31]. Borges E, Wenning MR, Kimura EM et al. High prevalence of alpha

thalassemia among individuals with microcytosis and hypochromia without

anemia. Braz J Med Biol Res 2001;34(6):759-62

[32]. Vichinsky E, Kleman K, Embury S, Lubin B. The diagnosis of iron

deficiency anemia in sickle cell disease. Blood 1981;58:963-8

[33]. Akodu SO, Kehinde OA, Diakwu-Akinwumi IN, Njokanma OF. Iron

deficiency anemia among pre-school children with sickle cell anemia: Still a rare

diagnosis? Mediterr J Hematol Infect Dis 2013;5(1):e2013069

doi:10.4084/MJHID.2013.069

[34]. Stevens GA, Finucane MM, De-Regil LM, Paciorek JC, Flaxman SR,

Branca F. Global, regional and national trends in hemoglobin concentration and

prevention of total and severe anemia in children and pregnant and non pregnant

women 1999 – 2011. Lancet Glob Health 2013;e16-e25.

179

[35]. Serjeant GR, Higgs DR, Hambleton IR. Elderly survivors with

homozygous sickle cell disease. N Engl J Med 2007;356(6):642-3

[36]. Minniti CP, Taylor JGt, Hildesheim M et al. Laboratory and

echocardiography markers in sickle cell patients with leg leg ulcers. Am J

hematol2011;86:705-8

[37]. Minniti CP, Kato GJ. How we treat sickle cell patients with leg ulcers. Am

J Hematol 2016;91(1):22-30

[38]. Belisaro AR, Rodrigues CV, Martins ML, Silva CM, Viana MB.

Coinheritance of alpha thalassemia decreases the risk of cerebrovascular disease

in a cohort of children with sickle cell anemia. Hemoglobin 2010;34:516-29

[39]. Flanagan JM, Frohlich TA, Howard WH et al. Genetic predictors for stroke

in children with sickle cell anemia. Blood 2011;117:6681-84

[40]. Filho IL, Leite AC, Moura PG et al. Genetic polymorphisms and

cerebrovascular disease in children with sickle cell anemia from Rio de Janeiro,

Brazil, Arq. Neuropsiquiatr 2011;69:431-5

[41]. Milner PF, Kraus AP, Sebes JI et al. Sickle cell disease as a cause of

osteonecrosis of the femoral head. New Engl J Med 1991;325:1476-81

[42]. Mahadeo KM, Oyeku S, Taragin B et al. Increase prevalence of

osteonecrosis of the femoral head in children and adolescent with sickle cell

disease. Am J Hematol 2011;86(9):806-8

180

Table 1: Biodata and frequencies of alpha thalassemia alleles

Parameters SCA (SS)

N=100

n (%)

AS

N=22

n (%)

AA

N=41

n (%)

Chi-square,

degree of

freedom

ᵡ2, df

P value

Age in years

Median

(Range)

8.5 (2-21) 9 (3 – 17) 8.5 (2 – 18) NA 0.888a

Sex

Male

Female

66 (66.0)

34 (34.0)

13 (59.0)

9 (41.0)

23 (56.0)

18 (44.0)

1.350, 2 0.509b

Social Class

Lower

Middle

Upper

48 (48.0)

44 (44.0)

8 (8.0)

12 (54.5)

8 (36.4)

2 (9.1)

20 (48.8)

18 (43.9)

3 (7.3)

0.4714, 4 0.9762 b

Alpha thalassemia present

41 (41.0) 8 (36.0) 16 (39.0) 0.2866, 2

0.866 b

Alpha thalassemia absent

58 (58.0) 13 (62.0) 25 (61.0)

α-globin

genotype

SCA (SS)

N=100

n (%)

AS

N=22

n (%)

AA

N=41

n (%)

αα/–α

34 (34.0) 8(36.0) 16 (39.0) 0.327, 2 0.849 b

-α/-α

7 (7.0) 0 (0) 0 (0) NA 0.043*

ααα 1 (1.0) 0 (0) 0 (0) NA

181

Allele

frequency

Total

chromosome

N=200

n(frequency)

Total

chromosome

N=44

n(frequency)

Total

chromosome

N=82

n(frequency)

αα 150 (0.75) 36 (0.82) 66 (0.80) 1.590, 2 0.451 b

-α 49 (0.25) 8 (0.18) 16 (0.20) 0.822, 2 0.662 b

ααα 1 (0.005) 0 (0.0) 0 (0.0) NA

NB: a=Kruskal-Wallis test, b= Chi-square test, *=Fisher`s exact test, NA= Not applicable, One

case of Triplicaion not considered as alpha thalassemia, Significant p value is in bold font.

182

Table 2: Laboratory parameters of patients and controls

Biomarkers HbAA

N=41

Median

(Range)

HbAS

N=22

Median

(Range)

HbSS

N=100

Median

(Range)

P value

Hb (g/dl) 11.5

(7.8 – 13)

11.6

(8 – 14)

7.5

(6.2 – 11.2)

˂0.0001

MCV (fL) 89.4

(66.4 – 102)

79.6

(60.4 – 74.7)

81.1

(60 – 115)

0.0125

RBC ( million

cells/uL)

4.5

(3.8 – 5.5)

4.6

(2.3 – 6.0)

2.8

(1.8 – 4.8)

˂0.0001

WBC (x

103/uL)

6.3

(4.5 – 14.4)

6.5

(4.5 -12.5)

13.1

(6.1 – 29.30)

˂0.0001

Platelet (x

103/uL)

244

(117 – 514)

275

(125 – 578)

361

(108 – 832)

0.0004

HbF (%) 0.75

(0 – 2.1)

0.7

(0.2 – 4.7)

9.3

(0.9 – 32.3)

˂0.0001

HbA2 (%) 2.9

(0.8 – 3.6)

3.6

(0.5 – 4.6)

1.5

(0.2 – 4.0)

˂0.0001

MCH (pg) 24.4

(17.6 – 27.9)

25.2

( 18.2 – 29 )

25.3

(16 – 34)

0.260

Total

bilirubin

(mg/dl)

0.43

(0.1 – 0.45)

0.45

(0.25 – 1)

1.80

(0.5 – 8.1)

˂0.0001

AST (IU/L) 19

(6 – 64)

27

(8 – 75)

43

(12 – 89)

˂0.0001

LDH (IU/L) 360

(150 – 861)

375

(179 – 993)

789.3

(179 – 1860)

˂0.0001

183

NB- Significant p values are in bold fonts, Test statistics = Kruskal-Wallis ANOVA, Hb-

Hemoglobin concentration, RBC-Red blood cells, MCV-Mean corpuscular volume, WBC-White

blood cells count, HbF- Fetal hemoglobin, MCH- mean corpuscular hemoglobin, AST-Aspartate transaminase, LDH-Lactate dehydrogenase

Table 3: Alpha thalassemia Alleles and laboratory parameters

Biomarkers Alpha

thalassemia

with

Heterozygous

deletion

N=34

Median (Range)

Alpha

thalassemia

with

Homozygous

deletion

N=7

Median (Range)

SS with no

Alpha

Thalassemia

Trait

N=58

Median (Range)

P Value

Hb (g/dl) 7.6 (6.3-11.2) 8.2 (7.6-10) 7.2 (6.2-11) 0.0199

MCV (fL) 78 (66.9 -94) 74 (60.3 – 101) 84 (56 – 115) 0.0025

RBC (million

cells/uL)

2.8 (2.15 -3.9) 3.1 (2.7 – 4.1) 2.7 (1.8 – 4.8) 0.0264

WBC (x 103/uL) 14.7 (6.4 – 26.2) 7.8 (6.1 – 13.2) 13.3 (6.1 – 29.3) 0.0353

Platelet (x

103/uL)

358 (108 -674) 334 (158 -444) 371 (118 -832) 0.2156

HbF (%) 8.8 (0.6 – 28.5) 13.1 (3.7 – 17.5) 10.7 (2.5 – 32.3) 0.1364

HbA2 (%) 1.7 (0.3 – 3.8) 2.8 (1.5 – 4.0) 1.5 (0.2 – 3.1) 0.0002

MCH (pg) 24.5 (18.6 –

30.6)

24.4 (15.7 –

27.1)

26 (15.9 – 34) 0.0021

Total Bilirubin

(mg/dl)

2.1 (0.43 – 7.7) 1.8 (1.03 – 8.1) 1.8 (0.8 – 5.2) 0.729

AST (IU/L) 58 (12 – 89) 40 (22 – 89) 43 (7 – 89) 0.3974

LDH (IU/L) 865 (215 –

1860)

771 (705 – 986) 881 (197 –

1681)

0.6469

NB- Significant p values are in bold fonts, Test statistics= Kruskal-Wallis ANOVA, Hb-

Hemoglobin concentration, RBC-Red blood cells, MCV-Mean corpuscular volume, WBC-White

blood cells count, HbF- Fetal hemoglobin, MCH- mean corpuscular hemoglobin, AST-Aspartate transaminase, LDH-Lactate dehydrogenase

184

Table 4: Co-inheritance of sickle cell anemia with Alpha Thalassemia and clinical events

Clinical events Patients with

Alpha

Thalassemia

(N=41)

Patients without

Alpha

Thalassemia

(N=58)

P value

Bone pain crisis

per Year

Range (0 –6)

Median 3

Range (0 –6)

Median 1

˂0.0001†

Stroke 2 3 1.000*

Osteonecrosis 3 2 0.6471*

Leg ulcer 0 6 0.0384 *

Priapism n=66

(Male only)

(N=34)

1

(N=32)

4

0.1974*

NB- Significant p values are in bold fonts, †= Mann-Whitney Test, *=Fisher`s exact test

185

MANUSCRIPT 3

Evaluation of socio-demographic, clinical and laboratory markers of sickle leg ulcers

among young Nigerians.

Authors:

Oladele Simeon Olatunya,a,b, Dulcineia Martins de Albuquerque,a, Adekunle

Adekile,c, Fernando Ferreira Costa,a

Affiliations: aHematology and Hemotherapy Center, University of Campinas, São Paulo, Brazil bDepartment of Paediatrics, College of Medicine, Ekiti State University, Ekiti State, Nigeria cDepartment of Pediatrics, Faculty of Medicine, Kuwait University, Kuwait

Corresponding author and address for correspondence:

Oladele Simeon Olatunya Hematology and Hemotherapy Center (Hemocentro),

University of Campinas (UNICAMP),

Rua Carlos Chagas, 480

Barão Geraldo

Campinas 13083-970-SP, Brazil

Tel: +55 19 3521 8382

Email for correspondence: ladeletunya@yahoo.com

186

ABSTRACT

Sickle leg ulcer (SLU) is a chronic and debilitating complication of sickle cell disease

(SCD) associated with huge physical and psychosocial discomfort. The occurrence of

SLU has remained steady despite successful preventive strategies and advances in SCD

care. Although multifactorial factors have been implicated in SLU, these are not fully

understood and data on how these relate to children with SCD are scanty. This study

aimed to evaluate the socio-demographic, clinical and laboratory markers of SLU in a

young SCD cohort.

One hundred and nine young SCD patients had their clinical, socio-demographic, and

laboratory parameters evaluated for SLU and their parameters compared with 67

healthy peers.

There were no differences in the socio-demographic parameters of patients and controls.

However, their laboratory parameters differ significantly (p<0.0001). Six patients with

HbSS genotype had SLU giving a prevalence of 5.9%. There was a preceding history of

trauma in 4 (66.7%) and this included a case of traditional scarifications for local

therapeutic purposes. The ulcers were mostly located on the ankles in 5 (83.3%) and

right big toe in 1 (16.7%). Two of the three males with SLU also had priapism. Patients

with leg ulcer were older, had less frequent bone pain crises, and significantly belonged

to the low socioeconomic class (p<0.05). Although patients with SLU had relatively

higher LDH, platelet count, AST, bilirubin, WBC, and lower Hb concentration and

HbF, these did not attain statistical significance (p>0.05).

This study confirms that SLU is common among HbSS genotype, older children and

those with low socio-economic background. In addition, this study suggests that SLU

could be more related to hemolysis-associated SCD phenotypes thus supporting the

187

previously held hypothesis that SLU is a distinct phenotype that is probably not related

to blood viscosity/rheology complications of SCD.

Keywords: Sickle cell anemia, Children, leg ulcer, Socio-demographic status,

Laboratory parameters, Clinical manifestations, Nigeria.

INTRODUCTION

Sickle cell disease (SCD), is an inherited hemoglobin defect that is very

common among Africans.1 The disease results from the inheritance of an abnormal

hemoglobin known as HbS.1 Patients with SCD present with variable disease severity

but the most severe form is the sickle cell anemia (SCA), which is the homozygous state

for the abnormal HbS.1 Globally, about 305,800 neonates are born with the SCA

annually and almost two-thirds of these occur in Africa2and Nigeria is the country with

the highest burden of SCA in the world.3

The structural abnormality in SCD leads to sickling of the red blood cells at low

oxygen tension. This results in clogging at microvascular level and subsequent vaso-

occlusion, hypoxia and other downstream clinico-pathologic manifestations of the

disease.1 Different aspects of the disease have been described based on these clinico-

pathologic manifestations.1,4

Leg ulcer complicating SCD, otherwise known as sickle leg ulcer (SLU), is a

chronic, and debilitating condition associated with both physical and emotional

disturbances.5,6,7,8 The prevalence of SLU varies among patients with SCD. It is low

before the age of ten years and rare in SCD genotypes other than SCA.6,8 Despite the

fact that SLU was among the complications found in the first documented SCD patient,9

188

advances in understanding its pathophysiology and management have been slow. 6,8

However, it is generally believed that SLU has multifactorial etiology.6,8 Hemolysis

intensity, and deposition of hemolysis products, depletion of nitric oxide and endothelial

function impairments have been implicated in its pathophysiology. 4,5,6,7,8,10 In addition,

SLU is believed to be rare in some parts of the world where prevailing genetic factors

are thought to ameliorate its occureence thus highlighting possible roles for

environmental and geographical factors. 6,8,11 Although, some studies have been

dedicated to the description of SLU,4,5,6,7,8,,10,12 very few studies have solely described

SLU among the young SCD population.4-8,10,12,13

In Nigeria, there is scanty information on SLU among young SCD patients.

Most of the previous studies were conducted among mixed populations of adults and

children.7,13,14 This leaves information on SLU among young SCD patients largely

unknown. This study was carried out on a cohort of young Nigerians with SCD. We

aimed to corroborate or dispute whether SLU is related to some laboratory markers,

hemolysis sub-phenotype and socio-demographic factors. Findings from this study

could help in improving the knowledge on factors influencing the occurrence of SLU

phenotype among young SCD patients.

PATIENTS AND METHODS

This was a cross-sectional study conducted on 109 hydroxyurea-naïve children

and adolescents with SCD in steady state and, who are regular attendees at the pediatric

hematology unit of the Ekiti State University Teaching Hospital (EKSUTH), Ado Ekiti,

Ekiti State, in Southwest Nigeria. Sixty-seven healthy children who accompanied their

siblings to the pediatric hematology clinic or attended the pediatrics outpatients` well

child clinic of the hospital served as the controls. To qualify for inclusion, the SCD

participants must have been on regular follow up at the clinic for a minimum of one

189

year prior to recruitment with up to date hospital records. Participants with confirmed or

suspected to have other chronic diseases apart from SCD were excluded. Also excluded

were the few SCD patients on regular blood transfusion and/or hydroxyurea therapies as

well as those whose caregivers declined to participate. SCD was initially diagnosed by

hemoglobin electrophoresis and high performance liquid chromatography (HPLC) and

was confirmed by DNA studies at the Centro de Hematologia e Hemoterapia

(Hemocentro), University of Campinas (UNICAMP), Brazil.

Ethical considerations

Written informed consents of parents/caregivers were obtained. Also, assents and

consents of patients were obtained as applicable. The study was approved by the

hospital`s Institutional Ethics and Research Committee no: A67/2016/03/003.

Data collection

Clinical, socio-demographic and laboratory data

A tested chart review form was used to extract relevant information from the hospital

records of participants regarding their steady state laboratory parameters and clinical

events. Average of at least two steady state results of laboratory parameters performed

between 3 to 6 months intervals by standard techniques were recorded for each

participant. The laboratory parameters examined for both patients and controls included

the complete blood count performed by Sysmex KX21N Haematology analyser

(Sysmex Corporation, Kobe, Japan). Serum lactate dehydrogease (LDH), bilirubin,

alanine transaminase (AST) and aspartate transaminase (AST) were measured with

standard techniques. The quantitative assessment of HbF, HbA, HbA2, HbS, and HbC

was done by high performance liquid chromatography (HPLC, Bio-Rad Variant D10,

190

USA). Steady state was defined as being free from any acute event(s) for at least one

month and transfusion free for at least three months.1

Details of the clinical events among the patients like: presence of leg ulcer and

its evolution, overt osteonecrosis, overt stroke, priapism, and the number of bone pain

crises requiring admission and/or administration of opioids within the preceding one

year were obtained. The definitions of clinical events were as previously described.15

Other information retrieved from patients’ charts included the biodata and patients were

classified into socioeconomic groups based on the educational level and occupation of

the parents/caregiver.16

Data Analysis

The GraphPad Prism Program, version 5 for Windows (San Diego, California, USA)

was used for the statistical analysis. The normal distribution of the quantitative variables

was verified by the Kolmogorov-Smirnov and Shapiro-Wilk tests. Continuous variables

with non-normal distribution were expressed in median and analyzed by the Mann-

Whitney tests for comparison of two independent groups. Chi-Square test or Fisher`s

Exact Test was used to compare categorical variables as applicable and level of

statistical significance was set at P < 0.05.

RESULTS

The patients were made up of 101 HbSS (SCA) and 8 HbSC. They consisted of

72 males and 37 females with median age of 9, range 2 - 21years. The controls

consisted of 40 males and 27 females and were made up of 22 sickle cell trait (HbAS) &

45 HbAA, with median age of 9, range 2- 18years. The SCD patients have been on

191

follow up for a median of 4 years, range 1.5 -14years. There were no differences in the

socio-biographic data of patients and controls (Table 1).

Comparison laboratory parameters between patients and controls

There were significant differences between the patients and controls in all

laboratory parameters p<0.0001 respectively. (Table 1)

Prevalence and description of leg ulcer among patients

In general, only six (5.5%) patients had SLU (both active and healed) and they

were all HbSS patients (three males and females each) with median age of 17, range 14-

21years, thus representing 5.9% of this group. The leg ulcers were completely healed in

three (50%) after a median duration of five months (range 3 – 6 months), and yet to heal

in the remaining three patients, two of whom are having recrudescence of their ulcers.

One of the two with recrudescence ulcers, a 21-year-old also had overt stroke three

months after her SLU recrudescence and was counselled for hydroxyurea treatment but

never took the drug. Two of the three males with SLU also had priapism. The ulcers

were located on the ankles in 5 (83.3%) and right big toe in 1 (16.7%). There was a

preceding history of trauma in 4 (66.7%) and this included a case of traditional

scarifications for local therapeutic purposes.

Comparison of laboratory parameters between SCA patients with and without leg

ulcer

Although the LDH, bilirubin, AST, platelet and WBC counts of the patients with

SLU were relatively higher compared to their peers without leg ulcer, these values did

not attain statistical significance (p>0.05). Similarly, the lower values of Hb, RBC, and

192

HbF of the patients with SLU compared to their non-SLU HbSS counterparts, did not

attain statistical significance (Table 2).

Associations between leg ulcer phenotype, clinical events and socio-demographic

data of SCA patients

The patients with SLU were relatively older compared to their peers: median

age, 17.5 years (14 – 21) vs 9 years (2 – 18), p=0.0002. They were all from the low

socioeconomic class: low class 6 (12.5%) vs others (middle & upper) 0 (0%), p=0.0097.

They had less bone pain episodes and the males among them were more associated with

priapism (p=0.0132). However, the occurrence of leg ulcer was not differentiated by

the sex of the patients (p>0.05) Table 3.

DISCUSSION

Although improved treatment strategies like vaccinations and prophylaxis

against infections, transfusion services, use of hydroxyurea and other supportive care,

have favourably influenced the prognostic outlook of SCD, SLU still remains a

worrisome complication of the disease.7.8,10

There is scarcity of information on SLU among young African patients with

SCD despite the huge burden of SCD in the continent and this makes comparison of

data difficult. Nonetheless, although fewer than their HbSS counterparts, the

observation that none of the patients with HbSC genotype had SLU confirms the widely

held belief that, SLU is not common among this group of SCD patients.6,8,10,12 Similarly,

the observation that most cases of SLU in this study were preceded by trauma has been

previously reported in both children and adults.8,17,18 and the locations of the ulcers as

found in this study have been previously reported.8,10,17,18

193

However, the observation that SLU occurred as a result of local and unorthodox

therapeutic scarification in a child in this study, may not be unconnected with the strong

belief of Africans in some traditional practices that may be inimical to the health of

children. In some cases, these may be deleterious to the health of the patients18 as found

in this case or constitute some forms of abuse.19 There is therefore the need for more

education of caregivers to desist from such harmful practices.

Similar to previous local17 and international reports 8,10,12 the occurrence of leg

ulcer among our patients is not common within the first decade of life. Also, the 6%

prevalence of leg ulcer in this study is very similar to the 5% among the children in the

USA, 12 between 5.4 and 7.6% described locally by Akinyanju et al17 and Ideawor et al14

respectively in their studies of mixed populations of both children and adults from

Western Nigeria.

However, this is quite higher than 1.3% found among children from Southeast

Nigeria, 13 2.6% among the pediatric cohorts of a mixed study comprising both adults

and children in Brazil,20 It is nonetheless lower than the prevalences of 27% and 22%

by Madu et al7 and Bazuaye et al18 respectively, in their mixed studies of both children

and adults from Eastern and Midwestern Nigeria respectively. Similarly, it is lower

than the 75% described from groups of children and adults in Jamaica.21 These

observations suggest that the occurrence of leg ulcer increases with age and varies from

place to place.8 The studies by both Junior et al20 and Serjeant21 clearly demonstrated

that the prevalence of SLU increases with age. However, the apparently lower

prevalence of SLU among Brazilian children20, could be because of the use of

hydroxyurea by some of the children in the study. Hydroxyurea is known to ameliorate

the severity of SCD and reduce the occurrence of leg ulcer through its induction of HbF

194

and lessening of hemolysis in SCD patients.20 As noted, none of our participants was on

treatment with this drug.

In addition, the findings of association of leg ulcer with low socioeconomic

condition in this study is in tandem with observations by Minniti et al8 and Cumming et

al10 that poor socioeconomic status is a risk factor for leg ulcer among SCA patients.

The lack of sex predilection for leg ulcer in the current study has been reported among

Nigerians14 and patients in other parts of the world.5 However, one Nigerian study,17

found a male preponderance.

There is still a debate about classifying SLU, while some authors thought it is

associated with very severe phenotype of SCD,.4,5,6,20 others have argued that this is

probably not so.7 However, there is a general consensus that, it is closely associated

with hemolysis intensity and some other hemolysis-related phenotypes of SCA such as

priapism, stroke and pulmonary hypertension.4,5,6,20-22 Due to lack of facilities, we could

not confirm the diagnosis of pulmonary hypertension among our patients and only few

patients with overt stroke were diagnosed. There is therefore the possibility of missing

cases of silent infarcts. Nonetheless, most of the few male patients with priapism also

had SLU, suggesting the possibility of higher hemolysis in those affected with both leg

ulcer and priapism.

Furthermore, although not statistically significant, we observed a trend of higher

LDH, AST, and bilirubin among the SLU group while their Hb and HbF were lower,

compared to those without SLU. This is suggestive of possible higher hemolysis in the

leg ulcer group. That these differences did not attain statistical significance might be

due to the sample size of this study given that some larger studies have reported

association of hemolysis markers with leg ulcer in SCA patients.4,6,7,22

195

In this study we found that the SLU patients had less episodes of painful crises (VOC),

which is in keeping with findings by previous authors,7,23 and suggests that the two

complications (SLU and VOC) are distinct phenotypes. This supports the hypothesis

that while VOC is related to blood rheology/viscosity-associated complications, 24,25 the

SLU is linked to hemolysis.4-8,22

This study, being from a single centre, is limited by size. The apparently fewer

complications attributable to SCD phenotypes, could be due to the fact that, the study

participants were children who were relatively young as such yet to develop some of

SCD complications. Despite these limitations, the study was able to show that SLU is

not uncommon among children with SCA. In addition, some socio-demographic and

clinical associates of SLU were found.

In conclusion, this study showed that SLU occurs among young Nigerian SCA

patients with poor socio-economic background and in older children. In addition, this

study suggests that SLU is more related to hemolysis-associated SCD phenotypes thus

supporting the previous hypothesis that SLU is a distinct phenotype that is probably not

related to blood viscosity/rheology complications of SCD. There is need for more

studies on the SLU phenotype among children in order to fully understand how this

phenotype affects children with SCD.

FUNDING AND DISCLOSURES

This study was supported by grants No 2014/00984-3 from FAPESP, and grants No

2015/141693-0 from CNPq, Brazil.

The authors declare no conflict of interest with respect to this study.

ACKNOWLEDGEMENTS

196

Authors acknowledge with thanks the supports received from participants and their

caregivers/parents during the study.

REFERENCES

1. Rees DC, Williams TN, Gladwin MT. Sickle cell disease. Lancet 2010;376:179-

84

2. Piel FB, Patil AP, Howes RE, Nyangiri OA, Gething PW et al. Global

epidemiology of sickle hemoglobin in neonates: a contemporary geostatistical

model-based map and population estimates. Lancet 2013; 381:142-51

3. Diaku-Akinwumi IN, Abubakar SB, Adegoke SA, Adeleke S, Adewoye O et al.

Blood transfusion services for patients with sickle cell disease in Nigeria. Int

Health 2016;8(5):330-5

4. Kato GJ, Gladwin MT, Steinberg MH. Deconstructing sickle cell disease:

Reappraisal of the role of hemolysis in the development of clinical

subphenotypes. Blood Rev.2007;21:37-47

5. Minniti CP, Eckman J, Sebastiani P, Steinberg MH, Ballas SK. Leg ulcers in

sickle cell disease. Am J Hematol 2010;85:831-3

6. Minniti CP, Taylor JGt, Hildesheim M, O`Neal P, Wilson J et al. Laboratory and

echocardiography markers in sickle cell patients with leg leg ulcers. Am J

hematol 2011;86:705-8

7. Madu AJ, Ubesie A, Madu KA, Okwor B, Anigbo C. Evaluation of clinical and

laboratory correlates of sickle leg ulcers. Wound Rep Reg 2013:21:808-12

8. Minniti CP, Kato GJ. How we treat sickle cell patients with leg ulcers. Am J

Hematol 2016;91(1):22-30

9. Herrick J. Peculiar elongated and sickle shaped red blood cell corpuscles in a

case of severe anemia. Arch Intern Med 1910;6:517-21

197

10. Cumming V, King L, Serjeant G, Reid M. Venous incompetence, poverty and

lactate dehydrogenase in Jamaica are important predictors of leg ulceration in

sickle cell anemia. Br J Haematol 2008;142:119-25

11. Adekile AD. Mild –Phenotype of sickle cell disease: molecular basis, clinical

presentation and management recommendations. Current Paediatrics

2005;15:57-61

12. Koshy M, Entsuah R, Koranda A, Kraus AP, Johnson R et al. Leg ulcers in

patients with sickle cell disease. Blood 1989;74:1403-8

13. Chinawa JM, Chukwu BF, Ikefuna AN, Emordi IJ. Musculoskeletal

complications among children with sickle cell anemia admitted in university of

Nigeria teaching hospital (Ituku-Ozalla) Enugu. Ann. Med Health Sci

2013;3:564-7

14. Idaewor PO, Enosolease ME, Momoh MI. Leg ulcer in a population of Nigerian

patients with sickle cell anemia-twenty years experience. JnI Med &Biomedical

Res 2002;110:18-21

15. Ballas SK, Lieff S, Benjamin LJ, Dampier CD, Heeney MM, Hope C.

Definitions of phenotypic manifestations of sicle cell disease. Am J Hematol.

2010;85:6-13

16. Oyedeji GA, Socioeconomic and cultural background of hospitalized children in

Ilesa. Niger J Paediatr 1985;13:111-17

17. Akinyanju O, Akinsete I. Leg ulceration in sickle cell disease in Nigeria. Trop

Geogr Med 1979;31(1):87-91

18. Bazuaye GN, Nwannadi AI, Olayemi EE. Leg ulcers in adult with sickle disease

patients in Benin City Nigeria. Gomal J Med Sci. 2010;8(2):190-4

198

19. Olatunya OS, Isinkaye AO, Oluwadiya KS. Profile of non accidental childhood

injuries at tertiary hospital in South-west Nigeria. J Trop Pediatr 2015;

61(3):174-81

20. Junior EB, Silva DG, Torres L, Okumura JV, Lobo CL, Bonini-Domingos CR.

Severity of Brazilian sickle cell disease patients: Severity scores and feasibility

of the Bayesian network model use. Blood Cells Mol Dis 2015;54:321-7

21. Serjeant GR, Serjeant BE, Mohan JS, Clare A. Leg ulceration in sickle cell

disease; medieval medicine in a modern world. Hematol Oncol Clin North Am

2005;19:943-9ix

22. Kato GJ, McGowan V, Machado RF, Little JA, Taylor VI J, Morris CR et al.

Lactate dehydrogenase as a biomarker of hemolysis-associated nitric oxide

resistance, priapism, leg ulceration, pulmonary hypertension, and death in

patients with sickle cell disease. Blood 2006;107:2279-85

23. Ballas SK. Sickle cell anemia with few painful crises is characterized by

decreased red cell deformability and increased number of dense cells. Am J

Hematol 1991;36:122-30

24. Steinberg MH. Predicting clinical severity of sickle cell anemia. Br J Haematol

2005;129:465-81

25. Renoux C, Connes P, Nader E, Skinner S, Faes C, Petras M et al. Alpha

thalassemia promotes frequent vaso-occlussive crises in children with sickle cell

anemia through haemrheological changes. Pediatr Blood Cancer

2017;64(8):doi:10.1002/pbc.26455

199

Table 1: Bio-demographic and laboratory parameters in patients and control

Parameters SCD Patients

N = 109

Median (Range)

Controls

N = 67

Median (Range)

P value

Age in years 9.0 (2.0 – 21.0) 9.0 (2.0 – 18.0) 0.8786a

Sex Male 72 40 0.4227 b

Female 37 27

Social Class

Upper 10 5 0.7050 c

Middle 49 27

Lower 50 35

Hb (g/dl) 7.5 (6.2 – 11.2) 11.6 (8.0 – 14.2) <0.0001 a

RBC (million cells/uL) 2.8 (1.8 – 4.8) 4.5 (2.5 – 6.5) <0.0001 a

MCV (fL) 80.9 (55.9 – 115.0) 79.0 (60.4 – 102.6) <0.0001 a

HbF (%) 8.4 (0.2 – 28.5) 0.8 (0 – 5.8) <0.0001 a

WBC (x 103/uL) 12.6 (4.1 – 29.3) 6.6 (4.5 – 14.4) <0.0001 a

Platelet (x 103/uL) 343.0 (108.0 – 832.0) 278.0 (117.0 – 591.0) <0.0001 a

LDH (IU/L) 771.4 (166.0 –

1860.0)

360.0 (150.0 – 993.0) <0.0001 a

AST (IU/L) 42.0 (12.0 – 89.0) 20.0 (6.0 – 75.0) <0.0001 a

ALT (IU/L) 20.0 (8.0 – 77.0) 10.0 (4.0 – 45.0) <0.0001 a

Total bilirubin

(mg/dl)

1.8 (0.4 – 8.1) 0.4 (0.1 – 2.2) <0.0001 a

NB: Statistical significant p values are in bold fonts, a=Mann-Whitney Test, b=Fisher`s Exact

Test, c= Chi-Square Test, SCD=Sickle cell disease.

200

Table 2: Laboratory parameters in SCA patients with and without leg ulcer

Parameters SCA wIth leg ulcer

N = 6

SCA wIthout leg

ulcer

N = 95

P value

Hb (g/dl) 7.1 (6.2 – 8.1) 7.5 (6.3 – 10.1) 0.3356 a

RBC (million cells/uL) 2.5 (2.0 – 2.8) 2.8 (1.8 – 4.8) 0.067 a

HbF (%) 7.1 (3.7 – 10.7) 9.4 (0.9 – 28.5) 0.377 a

WBC (x 103/uL) 14.5 (10.0 – 21.3) 13.2 (6.1 – 29.3) 0.4421 a

Platelet (x 103/uL) 473.5 (207.0 – 669.0) 361.0 (108.0 – 832.0) 0.247 a

LDH (IU/L) 972.0 (681.7 –

1682.0)

800.0 (197.0 –

1860.0) 0.095

a

AST (IU/L) 40.0 (35.0 – 56.0) 36.0 (7.0 – 89.0) 0.1541 a

Total bilirubin

(mg/dl)

2.1 (1.3 – 4.1) 1.8 (0.4 – 8.1) 0.570 a

NB: Statistical significant p values are in bold fonts, a=Mann-Whitney Test, SCA=Sickle cell

anemia.

201

Table 3: Associations between leg ulcer phenotype, clinical events and bio-demographics

of SCA patients

Parameters SCA with leg ulcer

N = 6

SCA without leg ulcer

N = 95

P value

Age in years 17.5 (14.0 – 21.0) 9.0 (2.0 – 18.0) 0.0002 a

Sex Male 3 64 0.4016 b

Female 3 31

Social class

Lower 6 42 0.0097 b

Others

(Upper & Middle)

0 53

Bone pain crisis

(VOC)

0.5 (0 -1.0) 2 (0 – 6.0) 0.0382 a

Osteonecrosis 1 4 0.2686 b

Stroke 1 3 0.2203 b

Priapism (Males only

total number = 67)

(N = 3)

2

(N =64)

3

0.0132 b

NB: Statistical significant p values are in bold fonts, a=Mann-Whitney Test, b=Fisher`s Exact

Test, SCA=Sickle cell anemia.

202

ANNEXES

Annex 1

Table : Effects of hydroxyurea treatment on biologic markers of HbS-β thalassemia patients

Parameters HbSS treated with hydroxyurea (N=12)

NO hydroxyurea treatment (N=12)

P values

RMP(events/mL) 60,000.0 (0.0 -440000.0)

60,000 (0 – 280000)

0.8593

Plasma Hb (mg/dL) 62.6 (34.0 – 134.0) 75.6(40.0 – 196) 0.312

Haptoglobin (ng/mL) 2832.0 (480.0 – 11730)

2569.0 (540.0 – 18180)

0.5834

Heme (µM) 48.0 (22.9 – 94.0) 52.3 (23.0 – 119.0) 0.977

Hemopexin (µg/mL) 556.0 (108.0 – 1805.0)

310.0 (127.0 – 1642)

0.6236

HbF (%) 18.9 (0.6 – 35.2) 9.3 (2.0 – 26.2) 0.1487

HbS (%) 67.6 (56.7 – 82.4) 70.8 (46.0 – 89.7) 0.583

RBC ( million cells/uL)

2.8 (2.4 – 4.4) 4.0 (3.2 – 6.1) 0.0009

MCV (fL) 96.7 (79.4 – 118.4) 73.8 (61.8 – 85.9) 0.0002

Total bilirubin (mg/dL)

1.5 (0.6 – 5.1) 1.3 (0.6 – 2.2) 0.418

Unconjugated bilirubin (mg/dL)

1.2 (0.4 – 4.2) 0.9 (0.4 – 1.8) 0.543

Hb (g/dL) 8.7 (6.8 – 11.5) 9.5 (7.2 – 11.0) 0.193

Reticulocyte (×109/L) 232.0(52 – 455) 270.9 (136.0 – 561.0)

0.544

LDH (IU) 276.0 (192.0 – 531.0)

294.0 (128.0 – 496.0)

0.623

Platelet ( X 103/uL) 391.0 (172.0 – 479.0)

332.0 (76.0 – 755.0)

0.298

WBC ( X 103/uL) 5.5 (4.2 – 11.1) 9.2 (4.1 – 16.4) 0.119

NB: Significant p values are in bold fonts, Test statistics= Mann-Whitney test, RMP-Red blood cell microparticle, HbF-Fetal Hemoglobin, RBC-Red blood cell, Hb-Hemoglobin concentration, HbS- Hemoglobin S, MCV- Mean corpuscular volume, WBC-White blood cell count, LDH-Lactate dehydrogenase.

203

Annex 2

Figure annex 02: Linkage disequilibrium map of the SNPs pairs.

204

205

ANEXO:

206

207

208

209