The Natural History of Sickle Cell Diseaseperspectivesinmedicine.cshlp.org/content/3/10/a011783.full.pdf · The Natural History of Sickle Cell Disease ... enough under normal physiological

The Natural History of Sickle Cell Disease

Graham R. Serjeant

Sickle Cell Trust (Jamaica), Kingston 6, Jamaica

Correspondence: [email protected]

The term sickle cell disease embraces a group of genetic conditions in which pathologyresults from the inheritance of the sickle cell gene either homozygously or as a doubleheterozygote with another interacting gene. The spectrum of resulting conditions is thereforeinfluenced by the geography of individual hemoglobin genes, but in most populations, thecommonest genotype at birth is homozygous sickle cell (SS) disease. Because this genotypegenerally manifests a greater mortality, the relative proportion of sickle cell genotypes isinfluenced by age as well as the geographical distribution of individual genes.

Inheritance of the abnormal sickle cell genefrom one parent and a gene for normal HbA

from the other parent results in the sickle celltrait. Persons with the sickle cell trait have a rel-ative resistance to falciparum malaria, are lesslikely to get the disease, run lower parasitecounts, and are less likely to die. This survivaladvantage is most marked during a window inearly childhood between the loss of passivelyacquired maternal immunity and the develop-ment of active immunity, but the length andtiming of this window may vary between com-munities and be influenced by the pattern ofmalarial transmission. The survival advantagerelative to normal individuals with HbA hascontributed to high frequencies of the sicklecell trait in areas with a history of malaria. Theprevalence of the sickle cell trait varies mark-edly between different regions but reaches levelsas high as 40% in some areas of sub-SaharanAfrica, eastern Saudi Arabia, and central India.Persons with the sickle cell trait have both HbAand HbS, with HbS levels between 20% and

45%, which are largely genetically determined.The levels of HbS in the sickle cell trait are notenough under normal physiological conditionsto cause the problems relating to sickling exceptin the renal medulla where vascular damagecauses an inability to concentrate the urine anda tendency to painless hematuria. Other featurespossibly associated with the trait include pul-monary embolism and an increased risk of sud-den death under unusual physiological condi-tions, but these events are rare and the sicklecell trait, under normal conditions, is consideredgenerally harmless and must remain outside thedefinition of sickle cell disease.

GEOGRAPHY OF SICKLE CELL DISEASE

Populations of African Origin

The sickle cell trait is widespread throughoutAfrica with low frequencies (,1%–2%) in thenorth and south of the continent and highbut variable frequencies throughout much of

Editors: David Weatherall, Alan N. Schechter, and David G. Nathan

Additional Perspectives on Hemoglobin and Its Diseases available at www.perspectivesinmedicine.org

Copyright # 2013 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a011783

Cite this article as Cold Spring Harb Perspect Med 2013;3:a011783

1

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org

on June 15, 2018 - Published by Cold Spring Harbor Laboratory Press http://perspectivesinmedicine.cshlp.org/Downloaded from

http://perspectivesinmedicine.cshlp.org/

equatorial Africa. Characterization of the DNAstructure flanking the b-globin locus of HbSsuggests that the mutation has arisen on at leastthree independent occasions in the African con-tinent, referred to as b-globin haplotypes andnamed after the areas where they were first de-scribed: Benin, Senegal, and Central African Re-public or Bantu (Pagnier et al. 1984; Nagel et al1985; Chebloune et al. 1988). The HbC trait isbelieved to be a relatively recent mutation lim-ited to West Africa where it occurs at high fre-quencies (.20%) in central Ghana and Burki-na Faso, in only 2% in Nigeria, and does notoccur, except in peoples of West African origin,in East and Central Africa. Only limited data areavailable on the type and distribution of a- andb-thalassemia genes in the African continent.

The most comprehensive data on popu-lations of West African origin derive from theAmericas and particularly from the UnitedStates and from the Caribbean. Screening of100,000 consecutive nonoperative deliveries inthe Jamaica Cohort Study found the sickle celltrait in 10.0% and the HbC trait in 3.6% (Ser-jeant et al. 1986) and the relative frequenciesamong Afro-Americans were 8% and 2%. InSouth America, these genes are largely confinedto Brazil where they may occur in up to 7% ofpeoples of African origin.

The b-thalassemia trait derives from manydifferent molecular mutations but their clinicalsignificance depends on the degree of b-chainsynthesis and, hence, HbA. Most commonamong Jamaicans are the mild bþ-thalassemiamutations, which occur in about 1%, and areusually owing to mutations in the promoter re-gion (288 C . T, 229 A . G). The more se-vereb0-thalassemia mutations occur in �0.5%,and are usually owing to mutations in the inter-vening sequence (IVSII-849 A . G). When in-herited with the HbS gene, sickle cell-bþ-thal-assemia generally results in 15%–25% HbA anda mild clinical course, whereas sickle cell-b0-thalassemia has no HbA and is often clinicallyindistinguishable from homozygous sickle cell(SS) disease.

The frequencies of the genotypes of sicklecell disease at birth in Jamaica are one in 300 forSS disease, one in 500 for sickle cell-hemoglobin

C (SC) disease, one in 3000 for sickle cell-bþ-thalassemia, and one in 7000 for sickle cell-b0-thalassemia; slightly lower figures occur in theUnited States. Rare forms of sickle cell diseasemay result from the interaction of HbS withHbD Punjab, HbO Arab, and Hb Lepore.

Populations in India and the Arabian Gulf

In Saudi Arabia, the sickle cell trait is wide-spread but reaches its highest prevalence in theeastern Province. In India, the trait occurs mostcommonly among the tribal peoples in centralIndia (southeastern Gujarat, Maharastra, Madh-ya Pradesh, Chhattisgarh, western Odisha) witha smaller focus in the south of the country(northern Tamil Nadu and Kerala), and traitfrequencies as high as 40% have been describedin some groups. In these areas, the DNA struc-ture flanking the b-globin locus differs fromthat in African peoples, suggesting that this isa fourth independent occurrence of the HbSmutation and is referred to as the Asian haplo-type. This haplotype is often associated withhigh levels of fetal hemoglobin and frequenta-thalassemia, both of which tend to inhibitsickling and ameliorate the clinical course. Sick-le cell disease in the Arabian Gulf and India ispredominantly SS disease.

COMPARISON OF GENOTYPES

SS disease and sickle cell-b0-thalassemia aremost severe clinically and hematologically al-though both genotypes may vary markedly inseverity. SC disease is generally mild but moreprone to proliferative sickle retinopathy (PSR).Sickle cell-bþ-thalassemia presents a very wideclinical spectrum depending on the molecularmutation for the b-thalassemia gene and theamount of HbA produced. In peoples of WestAfrican origin, the sickle cell-bþ-thalassemiagene is associated with a mild reduction of nor-malb-chain synthesis, HbA levels of 15%–25%,and generally a mild clinical course, althoughthis genotype is also more prone to PSR. Theincreased risk of PSR in the generally milderforms of sickle cell disease, SC and Sbþ-thalas-semia, remains largely unexplained. In Indians,

G.R. Serjeant

2 Cite this article as Cold Spring Harb Perspect Med 2013;3:a011783

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



the common bþ-thalassemia gene occurring inassociation with HbS produces only 3%–5%HbA and a generally severe clinical course.

The frequency and severity of clinical com-plications varies between the genotypes butthere is a pattern for symptoms to be moremarked in the generally severe SS and Sb0-thal-assemia and more mild in SC disease and theWest African form of Sbþ-thalassemia with thenotable exception of PSR. Much of the subse-quent presentation will focus on the featuresof SS disease and be presented by patient agebecause this is a major determinant of clinicalfeatures.

CLINICAL FEATURES

The First Year

Symptoms develop as the g-chain synthesis,which results in 60%–80% fetal hemoglobin(HbF) at birth, is replaced by b-chain synthesisand increasing levels of the abnormal HbS. Thistransition occurs at different rates in differentindividuals but in some, pathological levels ofHbS may be reached within 8–10 wk of birthand potentially life-threatening complicationsmay occur from that age. Early complicationsinclude dactylitis and death may result fromthe acute chest syndrome and abnormal splenicfunction rendering the infant prone to over-whelming septicemia and to acute splenic se-questration.

The onset of significant mortality between 6and 12 mo implies that detection of the diseaseat birth is vital to implement family educationand prophylactic measures; this may be a par-ticular challenge because the mother has anapparently normal, asymptomatic child andmay not understand the urgency. Overwhelm-ing blood infections may be largely prevented bypneumococcal prophylaxis, which includes theconjugate pneumococcal vaccine at 2, 4, and6 mo (Halasa et al. 2007), and regular penicillin(Gaston et al. 1986) although there are concernsthat the conjugate vaccine may select nonvac-cine serotypes (McCavit et al. 2011). Acutesplenic sequestration may commence as earlyas 3 mo of age but is most common in the second

6 mo of life; two-thirds of attacks occur before2 yr of age, and events become rare after 6 yr ofage. Splenectomy has been advocated after twoattacks, and instructing mothers in the detec-tion of splenic enlargement and when to seekurgent attention has significantly reduced mor-tality from this complication (Emond et al.1985). The mother also receives an importantpsychological message that although her childhas a currently incurable disease, much can bedone by close observation and simple interven-tions to improve its outcome.

Dactylitis, a complication caused by avascu-lar necrosis of the bone marrow in the smallbones of the hands and feet starts as early as10–12 wk of age and affected 50% of childrenby the age of 2 yr in the Jamaican Cohort Study(Stevens et al. 1981). Attacks usually resolveover 5–7 d with simple analgesia, but common-ly recur, and become rare after the age of 5 yrwhen the active bone marrow withdraws fromthe small bones. Although most attacks resolvecompletely, clinically and radiologically, infec-tion superimposed on the avascular tissue maysometimes cause premature fusion and a per-manent shortening of the affected small bones.Dactylitis must be carefully explained to themother and medical attendants because the oc-casional misdiagnosis of child abuse may causegreat distress to the family. A history of dactylitismay predict a more severe clinical course (Mil-ler et al. 2000).

Early Years

Acute chest syndrome is a term that has becomewidely used in sickle cell disease because of thecomplex pathology which may include compo-nents from infection, infarction, fat embolism,and pulmonary sequestration. Characterized bynew pulmonary infiltrates and a very variableclinical picture including chest pain, cough, anddyspnea, this syndrome should be monitoredclosely clinically and by pulse oximetry. Ex-change transfusion can be lifesaving when rapiddeterioration signals acute pulmonary seques-tration. This syndrome may also follow hypo-ventilation associated with the pleuritic pain ofavascular necrosis of the ribs or sternum and

Natural History of Sickle Cell Disease

Cite this article as Cold Spring Harb Perspect Med 2013;3:a011783 3

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



may be prevented by incentive spirometry insuch cases (Bellet et al. 1995). The acute chestsyndrome is one of the commonest causes ofdeath after the age of 2 yr.

Stroke is also a feature of early childhoodwith a median age of onset of 6 yr, some casesoccurring as early as the second year of life, withan incidence of 8% by the age of 14 yr in theJamaican cohort (Balkaran et al. 1992). Theprincipal pathology at this age is cerebral infarc-tion secondary to stenosis of major cerebral ves-sels, even as large as the internal carotid artery.The mechanism of this pathology is poorly un-derstood and clearly different from the smallvessel occlusion characterizing other manifesta-tions of the disease. Recurrent strokes occurwithin 3 yr of the initial event in 50%–70%and many recurrent episodes may be preventedwith chronic transfusion programs (Sarnaik etal. 1979; Pegelow et al. 1995). Risk factors in-clude narrowing of cerebral vessels, whichmay be detected by transcranial Doppler, andin these, chronic transfusion has prevented theinitial stroke from occurring (Adams et al.1998).

Hypersplenism, characterized by persistentsplenic enlargement, significant red cell seques-tration, marked bone marrow expansion, anda new hematological equilibrium, also emergesaround this age and affected 5% of childrenin the Jamaican cohort. Some events resolvespontaneously; others are treated by splenec-tomy or chronic transfusion, but if unresolved,this complication may lead to significant mor-bidity and mortality. The relationship betweenacute splenic sequestration and hypersplenismis complex and ill understood; some episodesof acute splenic sequestration evolve into hy-persplenism and episodes of acute splenic se-questration may become superimposed on hy-persplenism, but the two pathologies tend tooccur at different ages and have different riskfactors.

The aplastic crisis also emerges as a com-mon problem at this age and is an avoidablecause of the occasional death. Almost all clini-cally defined aplastic crises are owing to humanparvovirus B19, which destroys the red cell pre-cursors in the bone marrow for 8–10 d until

neutralized by the production of parvovirus-specific IgG. In hematologically normal people,such infection reduces the reticulocyte count tozero but with a mean red cell survival of 120 d,the hemoglobin decreases by only 0.5 g/dL.However, with red cell survival of 10–12 d typ-ical of African SS disease, events in the bonemarrow are rapidly reflected in the peripheralblood and the hemoglobin falls by �1 g/d tolife-threatening levels unless treated by transfu-sion. The outcome is entirely predictable andbenign as long as oxygen delivery is maintainedby a single unit transfusion, and immunity toparvovirus seems to be lifelong. Seroconversion,assessed by parvovirus-specific IgG, occurredin 70% by age 20 yr in both Jamaica (Serjeantet al. 2001) and Uganda (Ndugwa et al. 2004),and most cases present clinically between 2 and10 yr of age.

Later Childhood and Early Adolescence

Many problems emerge in adolescence, whichunfortunately coincides with the transitionfrom pediatric to adult care when many patientsare lost to regular follow-up. These problemsinclude nocturnal enuresis, an increasing prev-alence of bone pain crisis, avascular necrosis ofthe femoral head, leg ulceration in some com-munities, priapism, problems associated withdelayed growth and puberty, pregnancy, andthe need for effective contraception.

Nocturnal enuresis has received little atten-tion because it is not life threatening but it mayhave a profound effect on behavior and socialperformance. Bedwetting more than or equal totwice weekly at the age of 8 yr occurred in 45%of children in the Jamaican cohort comparedwith 19% among normal (AA) controls (Read-ett et al. 1990a). Enuretic patients tended tohave higher urinary volumes and lower bladdercapacity (Readett et al. 1990b). Enuresis alwaysresolves eventually but may persist in some pa-tients up to age 18–20 yr. The deleterious ef-fects are more social and psychological ratherthan medical because of the obvious stresscaused to patient and family.

The bone pain crisis, generally resultingfrom avascular necrosis of bone marrow, is the

G.R. Serjeant


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



adult counterpart of dactylitis. Although com-monly believed to result from vaso-occlusionand often referred to as the vaso-occlusive crisis(VOC), the bilateral, symmetrical involvement(Serjeant et al. 1994) and the common precip-itation by skin cooling are difficult to explain onthis basis. Bone pain crises increase in adoles-cence and early adult life, especially in males(Baum et al. 1987; Platt et al. 1991), and declinein both severity and frequency after the age of25–30 yr, in most patients. Risk factors includea high hemoglobin level, low HbF level, and inthe third trimester of pregnancy precipitatingfactors include cold exposure, infections, andstress. Although this complication is the causeof �90% of hospital admissions in the UnitedStates (Yang et al. 1995) and the United King-dom (Brozovic et al. 1987), it is not an intrin-sically severe pathology and death is unusual.The mechanism of the progressive ameliorationof bone pains after 30 yr of age is unclear butmay reassure patients.

Avascular necrosis of the femoral head alsoresults from avascular necrosis of bone marrowbut because of its weight bearing in walking,secondary damage may occur to the articularsurface of the femoral head. The prevalencein unselected series is uncertain although prob-ably �10% and most commonly occurs inlate adolescence. The outcome is determinedby age at occurrence; infarction of the imma-ture capital epiphysis allows remodeling andreasonable function, whereas involvement ofthe mature femoral head typically results insegmental damage with painful limitation ofmovement, which may require prosthetic sur-gery. Early diagnosis, for which magnetic res-onance imaging (MRI) seems most sensitive,and avoidance of weight bearing may allowthe bone to heal without deformity. Core de-compression, presumably by relieving the intra-medullary pressure, may relieve pain and im-prove movement (Mukisi-Mukaza et al. 2009)and possibly delay the use of total hip replace-ment.

Chronic leg ulceration is a major problemin adolescence but its frequency varies widelyin different communities, affecting over 50%of Jamaican patients (Cumming et al. 2008),

but only 5%–10% of those in the United States(Koshy et al. 1989). Although rarely a cause ofdeath, leg ulceration has a profound impact oneducation and subsequent employment poten-tial, on social relationships, and may cause de-pression. Although characterized by a chronichealing/relapsing course, ulcers eventually healin most patients by the age of 30 yr. The cause ofthe geographic variability is not currently un-derstood.

Physical growth and sexual developmentare commonly delayed from a variety of mecha-nisms and it is necessary to reassure the familiesthat this is characteristic of the disease and gen-erally does not require special therapy becausefamilies may use expensive tonics to “fatten” thechild with little or no effect. Body habitus maybe permanently affected with the appearance oflong thin limbs, narrow shoulders and hips, anda relative absence of body fat. Puberty, assessedby menarche, is delayed by an average of 2.5 yrbut there is no difference in the interval be-tween menarche and first pregnancy. Pregnan-cy is associated with an increased risk of bonepain crisis, acute chest syndrome, and maternaldeath, especially in the last trimester and imme-diate postpartum period, and close antenatalmonitoring with hospital delivery is advised.If contraception is requested, there is no evi-dence that any form of contraception is contra-indicated in SS disease.

Priapism is a distressing complication af-fecting �40% of postpubertal males in Jamaica(Emond et al. 1980). Defined as painful erec-tions unassociated with sexual desire, there aretwo clinical forms. Stuttering priapism, whichis nocturnal, lasts 2–4 h, with normal interven-ing sexual function and major attacks lasting.12 h and usually followed by permanent vas-cular damage and impotence. Stuttering attacksinterfere with sleep, leading to daytime somno-lence but more importantly may be a prodromefor major events. Stuttering attacks are usuallyunderreported because of embarrassment andthe lack of realization that it is owing to sicklecell disease, so direct questioning may be nec-essary in its detection. The natural history ispoorly documented but stuttering events gen-erally resolve with time.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Later Adult Life

A second peak of stroke occurs around 25 yr ofage and these are predominantly hemorrhagicincluding subarachnoid hemorrhage often sec-ondary to berry aneurysms or direct intracere-bral or intraventricular hemorrhage, which havea poor prognosis.

The acute chest syndrome continues to be amajor clinical problem and increasing pulmo-nary fibrosis may be associated with pulmonaryhypertension, placing further stress on cardiacfunction. Congestive cardiac failure becomesmore common, possibly secondary to pulmo-nary hypertension, but also to a reduced com-pliance of the left ventricular wall secondary tothe sustained hypertrophy and intraventricularfibrosis. Hemoglobin levels commonly declineover the age of 40 yr compromising cardiacfunction further.

Deteriorating renal function is a major con-tributor to morbidity beyond the age of 40 yr,progressive glomerular fibrosis being associ-ated with declining glomerular filtration rates(GFR), falling erythropoietin levels, and a grad-ual decline in total hemoglobin (Morris et al.1991). Often clinically silent, patients do notpresent with the renal abnormalities but thegradually decreasing hemoglobin level impairscardiac function. Renal failure in SS diseaseshould be defined on the basis of lower serumcreatinine levels than in the general population,and an upper limit of normal of 70–80 mmol/Lwas associated with significantly lowered GFR inthe Jamaican cohort (Thompson et al. 2007).

PERSPECTIVE

From the foregoing summary of major clinicalevents, these are clearly influenced by patientage. The disease is asymptomatic in the first1–2 mo of life but between 3–6 mo, dactylitisand splenic problems emerge. The second 6 moof the first year hold serious dangers and thegreatest mortality rate for the patient from acutesplenic sequestration, overwhelming septice-mias, and acute chest syndrome, most of whichcan be prevented or more effectively treated, butonly if the underlying diagnosis of sickle celldisease is known, which is a major justification

for newborn screening. From 1–5 yr of age,symptoms continue to be common with dacty-litis and acute splenic sequestration (both ofwhich may recur), acute chest syndrome con-tinues as an important problem, and strokeemerges. After age 5 yr, dactylitis becomes rarebut is replaced by the adult counterpart of thebone pain crisis as active bone marrow dis-appears from the small bones of the hands andfeet and becomes focused in the juxta-articularareas of the long bones, the vertebra, ribs, andsternum.

Adolescence is a serious time for many pa-tients faced with enuresis, leg ulceration, in-creasing bone pain crisis, delayed growth andsexual development, priapism, and hip disease.Major clinical and social support is needed atthis time.

In early adult life, many of these problemscontinue and may be amplified by the difficultiesof management of pregnancy. After 25–30 yr ofage, bone pain crises wane in frequency and se-verity and there may be a long period free ofbone pains when the quality of life is better thanpreviously and patients can engage in full-timeemployment and social activities and common-ly default from clinic follow-up. After the ageof 40 yr, renal impairment with the associat-ed decrease in hemoglobin level becomes com-mon and may lead to cardiac and other prob-lems.

Overall, survival in SS disease in developedsocieties now shows a median survival of 45–55 yr in the United States and Jamaica (Plattet al. 1994; Wierenga et al. 2001). However,the absence of survival data from earlier periodsmakes it difficult to assess whether this im-provement is real or apparent because of lesssymptomatic bias in the observed populations.This bias, which was undoubtedly strong in thepast, has been reduced by widespread popula-tion screening and especially neonatal detec-tion, although even in the relatively short timeframe of the Jamaican cohort, significantly im-proved survival was shown when comparing thefirst with subsequent terciles (Lee et al. 1995).Furthermore, there should be continued im-provement because many of the interventionsdeveloped in the past 30 yr address problems

G.R. Serjeant


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



in early childhood and the beneficial effectof these is yet to filter through to greater sur-vival.

The outcome of the disease in developingsocieties where services are less organized andresources more limited remains unknown andin sub-Saharan Africa, median survival may be,5 yr.

DETERMINANTS OF SEVERITY

It is likely that many factors, both genetic andenvironmental, determine expression and finaloutcome of the disease.

Fetal Hemoglobin

High levels of HbF, present at birth in virtuallyall patients, protect against vaso-occlusive andhemolytic features of the disease in the first fewmonths. The level at which HbF persists varieswidely between patients and between geo-graphic areas but patients maintaining higherlevels tend to have less hemolysis and moremild clinical courses. In African SS disease he-molysis was well established with lower hemo-globin and higher reticulocyte counts from1 mo of age in the Jamaican cohort (Serjeantet al. 1981). In the Asian haplotype, higher HbFlevels persist and are associated with a milderclinical course and probably greater survivalbut may render patients more prone to hyper-splenism. The genetics of the persistence of g-chain synthesis allowing HbF to persist is com-plex, but patients with higher HbF levels tendto have one or both parents with the sickle celltrait and a modest elevation of HbF suggestinga genetic mechanism that is not yet clearly de-fined.

a-Thalassemia

Deletion of one of a pair of closely linked a-thalassemia genes (heterozygous aþ-thalasse-mia) in SS disease tends to lower the mean cellhemoglobin concentration (MCHC), which as adeterminant of HbS polymerization (Seakinset al. 1973), is likely to reduce intravascularsickling. Deletion of both of the closely linked

a-thalassemia genes (homozygousaþ-thalasse-mia) has an even greater effect in lowering theMCHC but anyconsequent reduction in sicklingmay be offset by an increase in total hemoglobin.a-Thalassemia therefore may provide a fascinat-ing model to distinguish the effects of inhibitingsickling in small vessels and the potentiallydeleterious effect of a higher hematocrit impair-ing blood flow in larger vessels. In African SSdisease in Jamaica, heterozygous aþ-thalasse-mia occurred in 35% and homozygousaþ-thal-assemia patients in 4% (Higgs et al. 1982),whereas in the Asian haplotype even higher lev-els have been found (Karet al. 1986; Padmos et al.1991).

MCHC

Greece is the only population so far described inwhich SS disease is not associated with highfrequencies of a-thalassemia, yet Greek patientswere found to have significantly lower MCHCand features of milder disease compared withJamaicans (Christakis et al. 1990). The loweredMCHC in this population remains to be con-firmed and the mechanism explored.

Environmental Factors

The importance of environmental factors inthe outcome of sickle cell disease is undoubted,whether it relates to climatic factors or skincooling precipitating bone pain crises or appar-ently random events such as contact with Strep-tococcus pneumonia or parvovirus B19. Malariacontinues to be a major determinant of mor-bidity and mortality in affected areas such assub-Saharan Africa and central India. Suscept-ibility to other potentially serious infections bySalmonella spp. may reflect carriage rates in thegeneral population. Other important factorsare nutrition, access to public health measuressuch as immunization, and socioeconomic sta-tus, which may determine the families’ access tocommunication, transport, and medical care.The importance of environmental influencesand their mechanisms hold great potential forincreasing the understanding and further ame-lioration of sickle cell disease.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Genetic versus Environmental FactorsModifying SS Disease

One of the classic methods for distinguishinggenetic from environmental influences is thestudy of identical twins. Identical or monozy-gotic twins occur in 0.3%–0.4% of births, butto be useful for studies of hematology and clin-ical features, both twins must survive longenough to provide adequate data. These condi-tions are not frequently met and the largestavailable series contained six pairs of identicaltwins in Jamaica (Weatherall et al. 2005). Acomparison of twin pairs showed that geneticfactors influenced growth and hematologicalindices but that clinical events in the twin pairswere frequently discordant, suggesting the im-portance of nongenetic influences.

GEOGRAPHIC VARIABILITY

Reflecting both genetic and environmental fac-tors is the variability of SS disease in differentgeographic areas. The b-globin haplotypes pro-bably represent separate occurrences of the HbSmutation and do show minor differences in dis-ease features depending on the DNA structureson which such mutations occurred. Compari-son of features in the three principal Africanhaplotypes, Benin, Bantu, and Senegal, havebeen difficult to conduct because of the domi-nance of the Benin haplotype in the UnitedStates and the Caribbean but the approximatelyequal prevalence in Brazil should allow compar-ison of Benin and Bantu haplotypes. Most stud-ies currently available have been small andreached conflicting conclusions. In the EasternProvince of Saudi Arabia and central India, theAsian haplotype appears to have less hemolysis,greater persistence of splenomegaly, generallymore mild clinical features, and probably greatersurvival, although the extent to which this isowing to the haplotype or to the commonly co-inherited genes for a-thalassemia and persis-tence of HbF remains unclear.

CAUSES OF DEATH

The causes of death are strongly influenced bythe prevalence of malaria and other infections,

and almost certainly by the availability and so-phistication of medical and other services. Nosystematic data are available in sub-Saharan Afri-ca but survival is likely to be markedly shortenedand median duration may be as short as 5 yr.

African SS disease in the United States,where malaria is absent, has a much better prog-nosis, although even under the most sophisti-cated conditions, median survival is shorten-ed by at least 20 yr relative to the generalpopulation. Mortality patterns are undoubted-ly changing because of diagnosis by newbornscreening and more effective interventions,but an earlier review in Jamaica indicated thatthe highest mortality occurred in the second6 mo of the first year of life due principally toacute splenic sequestration, the acute chest syn-drome, and septicemias (Thomas et al. 1982).At later ages, acute chest syndrome persists asa major cause of death and the pattern of septi-cemias may be changing with the selectionof penicillin-resistant and/or nonvaccine sero-types of Streptococcus pneumoniae and the emer-gence of Salmonella as an important organism(Wright et al. 1997). After the age of 30 yr, cu-mulative organ damage impairs function espe-cially in the lungs with pulmonary fibrosis andpulmonary hypertension and in the kidney witha progressive glomerular fibrosis.

In SS disease with the Asian haplotype, al-most no data are available on survival or on thecauses of death although there are suggestionsthat cumulative end organ damage may be lesscommon than in African SS disease.

Ascertaining the cause of death is vital todefining strategies to prevent the responsiblepathologies and the dearth of available data,lack of blood and viral cultures, and of autopsiesare major impediments in Africa, Arabia, andIndia. However, even in Jamaica, with autopsyrates as high as 90%, detailed histology andblood cultures sometimes failed to detect a plau-sible cause of death. Furthermore, the effects onsurvival of recent long-term interventions suchas hydroxyurea and chronic transfusion pro-grams will need to be assessed and the currentdata on mortality and hydroxyurea (Steinberget al. 2003; Bakanay et al. 2005) illustrate thecomplexity of interpreting results.

G.R. Serjeant


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



PREVENTION OF SICKLE CELL DISEASE

It is now widely recognized that prevention ofdisease is more cost effective than trying to treatestablished pathology. Nowhere is this challengemore relevant than sub-Saharan Africa wherefigures based on population size, birth rates,and sickle cell trait frequency estimate thatmore than 250,000 babies with SS disease areborn each year. The available resources cannotprovide adequate clinical care to a problem ofthis scale and prevention of affected births mustbe one approach to the future control of thedisease. Because the disease results from the in-heritance of abnormal hemoglobin genes fromboth parents, affected children could be avoidedif one parent had a normal hemoglobin (AA)genotype. Genotype identification with educa-tion and counseling on its significance wouldallow persons to make informed decisions ontheir choice of partner and should they wish,avoid the risks of a child with sickle cell disease.In Bahrain, a voluntary premarital screeningprogram halved the births with SS disease andscreening became mandatory in 2004 (Al-Ar-rayed 2005). The Kingdom of Saudi Arabiaimplemented a similar program of premaritalscreening in February 2004, and a review ofthe first 6 yr of the program (2004–2009)indicates that out of 1.5 million marriage pro-posals, there had been a steadily increasing rateof marriage cancellations among at-risk couples(Memish and Saeed 2011). Among at-risk cou-ples who continue to marry, other potential op-tions include prenatal diagnosis and preimplan-tation diagnosis (Xu et al. 1999).

It remains unknown whether, in societieswithout traditions of arranged marriages, ge-notype identification and counseling at youn-ger ages will influence reproductive decisionsand such studies should have a high priorityat this time. A project in Orchomenos, Greecein 1966 examined 2300 families with a sicklecell trait frequency of 23%, but assessment ofgenotypes in 101 marriages occurring over thesubsequent 7 yr showed no significant devia-tion from random (Stammatoyannopoulos1973). Factors militating against success inthis study may have included the small num-

bers and lack of awareness of the disease amongthis relatively uneducated farming community.The Manchester Project, set in the Parish ofManchester in central Jamaica, has advantagesof scale and of a population with greater aware-ness of sickle cell disease. Screening for the de-tection of HbS, HbC, b-thalassemia trait, andother abnormal hemoglobins has been per-formed in approximately 16,000 senior schoolchildren aged 16–19 yr over the last 6 yr. Insteadof looking at the genotypes of selected partners,this study will be based on the offspring ofscreened students and whether the distributionof genotypes differs from that predicted by theHardy-Weinberg hypothesis. Newborn screen-ing of more than 32,000 births in the parishesserving this area identified 770 babies so far bornto mothers screened as school girls but muchgreater numbers will be needed to test the hy-pothesis.

Great advances have been made in the man-agement of sickle cell disease but many of theseare expensive and dependent on medical infra-structure, which may take a long time to de-velop in high-risk societies. In the meantimeprevention, where possible, will be an essentialcomponent to overall management of the prob-lem of sickle cell disease.

REFERENCES

Adams RJ, McKie VC, Hsu L, Files B, Vichinsky E,Pegelow C, Abboud M, Gallagher D, Kutlar A, Nichols FT,et al. 1998. Prevention of a first stroke by transfusions inchildren with sickle cell anemia and abnormal results ontranscranial Doppler ultrasonography. N Engl J Med 339:5–11.

Al Arrayed S. 2005. Campaign to control genetic blood dis-eases in Bahrain. Community Genet 8: 52–55.

Bakanay SM, Dainer E, Clair B, Adekile A, Daitch L, Wells L,Holley L, Smith D, Kutlar A. 2005. Mortality in sicklecell patients on hydroxyurea therapy. Blood 105: 545–547.

Balkaran B, Char G, Morris JS, Serjeant BE, Serjeant GR.1992. Stroke in a cohort study of patients with homozy-gous sickle cell disease. J Pediatr 120: 360–366.

Baum KF, Dunn DT, Maude GH, Serjeant GR. 1987. Thepainful crisis of homozygous sickle cell disease: A studyof risk factors. Arch Int Med 147: 1231–1234.

Bellet PS, Kalinyak KA, Shukla R, Gelfand MJ, Ruck-nagel DL. 1995. Incentive spirometry to prevent acutepulmonary complications in sickle cell diseases. N EnglJ Med 333: 699–703.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Brozovic M, Davies SC, Brownell AI. 1987. Acute admis-sions of patients with sickle cell disease who live in Brit-ain. Br Med J 294: 1206–1208.

Chebloune Y, Pagnier J, Trabuchet G, Faure C, Verdier G,Labie D, Nigon V. 1988. Structural analysis of the 50 flank-ing region of the b-globin gene in African sickle cellanemia patients: Further evidence for three origins ofthe sickle cell mutation in Africa. Proc Natl Acad Sci 85:4431–4435.

Christakis J, Vavatsi N, Hassapopoulou H, Papadopou-lou M, Mandraveli K, Loukopoulos D, Morris J, Ser-jeant BE, Serjeant GR. 1990. Comparison of homozygoussickle cell disease in Northern Greece and Jamaica. Lancet335: 637–640.

Cumming V, King L, Fraser R, Serjeant G, Reid M. 2008.Venous incompetence, poverty and lactate dehydroge-nase in Jamaica are important predictors of leg ulcerationin sickle cell anaemia. Br J Haematol 142: 119–125.

Emond A, Holman R, Hayes RJ, Serjeant GR. 1980. Priapismand impotence in homozygous sickle cell disease. ArchIntern Med 140: 1434–1437.

Emond AM, Collis R, Darvill D, Higgs DR, Maude GH,Serjeant GR. 1985. Acute splenic sequestration in homo-zygous sickle cell disease: Natural history and manage-ment. J Pediatr 107: 201–206.

Gaston MH, Verter JI, Woods G, Pegelow C, Kelleher J,Presbury G, Zarkowsky H, Vichinsky E, Iyer R, Lobel JS,et al. 1986. Prophylaxis with oral penicillin in childrenwith sickle cell anemia. A randomized trial. N Engl J Med314: 1593–1599.

Halasa NB, Shankar SM, Talbot TR, Arbogast PG, Mit-chel EF, Wang WC, Schaffner W, Craig AS, Griffin MR.2007. Incidence of invasive pneumococcal disease amongindividuals with sickle cell disease before and after theintroduction of the pneumococcal conjugate vaccine.Clin Infect Dis 44: 1428–1433.

Higgs DR, Aldridge BE, Lamb J, Clegg JB, Weatherall DJ,Hayes RJ, Grandison Y, Lowrie Y, Mason KP, Serjeant BE,Serjeant GR. 1982. The interaction of a-thalassemia andhomozygous sickle-cell disease. N Engl J Med 306: 1441–1446.

Kar BC, Satapathy RK, Kulozik AE, Kulozik M, Sirr S,Serjeant BE, Serjeant GR. 1986. Sickle cell disease in Oris-sa State, India. Lancet 328: 1198–1201.

Koshy M, Entsuah R, Koranda A, Kraus AP, Johnson R,Bellvue R, Flournoy-Gill Z, Levy P. 1989. Leg ulcers inpatients with sickle cell disease. Blood 74: 1403–1408.

Lee A, Thomas P, Cupidore L, Serjeant B, Serjeant G. 1995.Improved survival in homozygous sickle cell disease: Les-sons from a cohort study. Br Med J 311: 160–162.

McCavit TL, Quinn CT, Techasaensiri C, Rogers ZR. 2011.Increase in invasive Streptococcus pneumonia infectionsin children with sickle cell disease since pneumococcalconjugate vaccine licensure. J Pediatr 158: 505–507.

Memish ZA, Saeed MY. 2011. Six-year outcome of the na-tional premarital screening and genetic counseling pro-gram for sickle cell disease and b-thalassemia in SaudiArabia. Ann Saudi Med 31: 229–235.

Miller ST, Sleeper LA, Pegelow CH, Enos LE, Wang WC,Weiner SJ, Wethers DL, Smith J, Kinney TR. 2000. Pre-diction of adverse outcomes in children with sickle celldisease. N Engl J Med 342: 83–89.

Morris J, Dunn D, Beckford M, Grandison Y, Mason K,Higgs DR, De Ceulaer K, Serjeant BE, Serjeant GR.1991. The haematology of homozygous sickle cell diseaseafter the age of forty years. Br J Haematol 77: 382–385.

Mukisi-Mukaza M, Manicom O, Alexis C, Bashoun K,Donkerwolcke M, Burny F. 2009. Treatment of sicklecell disease’s hip necrosis by core decompression: A pro-spective case-control study. Orthop Traumatol Surg Res95: 498–504.

Nagel RL, Fabry ME, Pagnier J, Zohoun I, Wajcman H,Baudin V, Labie D. 1985. Hematologically and geneticallydistinct forms of sickle cell anemia in Africa. The Senegaltype and the Benin type. N Engl J Med 312: 880–884.

Ndugwa C, Kiggundu M, Cohen B, Serjeant GR. 2004.Aplastic crises and acute anaemic episodes in homozy-gous sickle cell disease. E Afr Med J 81: 493–494.

Padmos MA, Roberts GT, Sackey K, Kulozik A, Bail S,Morris JS, Serjeant BE, Serjeant GR. 1991. Two differentforms of homozygous sickle cell disease occur in SaudiArabia. Br J Haematol 79: 93–98.

Pagnier J, Mears JG, Dunda-Belkhodja O, Schaefer-RegoKE, Beldjord C, Nagel RL, Labie D. 1984. Evidence forthe multicentric origin of the sickle cell hemoglobin genein Africa. Proc Natl Acad Sci 81: 1771–1773.

Pegelow CH, Adams RJ, McKie V, Abboud M, Berman B,Miller ST, Olivieri N, Vichinsky E, Wang W, Brambilla D.1995. Risk of recurrent stroke in patients with sickle celldisease treated with erythrocyte transfusions. J Pediatr126: 896–899.

Platt OS, Thorington BD, Brambilla DJ, Milner PF, RosseWF, Vichinsky E, Kinney TR. 1991. Pasin in sickle celldisease. Rates and risk factors. N Engl J Med 325: 11–16.

Platt OS, Brambilla DJ, Rosse WF, Milner PF, Castro O,Steinberg MH, Klug PP. 1994. Mortality in sickle celldisease. Life expectancy and risk factors for early death.N Engl J Med 330: 1639–1644.

Readett DRJ, Morris JS, Serjeant GR. 1990a. Nocturnal en-uresis in sickle haemoglobinopathies. Arch Dis Child65: 290–293.

Readett DRJ, Morris J, Serjeant GR. 1990b. Determinants ofnocturnal enuresis in homozygous sickle cell disease.Arch Dis Child 65: 615–618.

Sarnaik S, Soorya D, Kim J, Ravindranath Y, Lusher J. 1979.Periodic transfusions for sickle cell anemia and CNS in-farction. Am J Dis Child 133: 1254–1257.

Seakins M, Gibbs WN, Milner PF, Bertles JF. 1973. Erythro-cyte Hb-S concentration. An important factor in the lowoxygen affinity of blood in sickle cell anemia. J Clin Invest52: 422–432.

Serjeant GR, Grandison Y, Lowrie Y, Mason K, Phillips J,Serjeant BE, Vaidya S. 1981. The development of haema-tological changes in homozygous sickle cell disease: Acohort study from birth to six years. Br J Haematol 48:533–543.

Serjeant GR, Serjeant BE, Forbes M, Hayes RJ, Higgs DR,Lehmann H. 1986. Haemoglobin gene frequencies in theJamaican population: A study of 100,000 newborns. Br JHaematol 64: 253–262.

Serjeant GR, De Ceulaer C, Lethbridge R, Morris JS, Sing-hal A, Thomas PW. 1994. The painful crisis of ho-

G.R. Serjeant


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



mozygous sickle cell disease—Clinical features. Br J Hae-matol 87: 586–591.

Serjeant BE, Hambleton IR, Kerr S, Kilty CG, Serjeant GR.2001. Haematological response to parvovirus B19 infec-tion in sickle-cell disease. Lancet 358: 1779–1780.

Stammatoyannopoulos G. 1973. Problems of screening andcounseling in the hemoglobinopathies. In Proceedingsof the IVth International Conference on Birth Defects,Vienna, pp. 268–276.

Steinberg MH, Barton F, Castro O, Pegelow CH, Ballas SK,Kutlar A, Orringer E, Bellevue R, Olivieri N, Eckman J,et al. 2003. Effect of hydroxyurea on mortality and mor-bidity in adult sickle cell anemia: Risks and benefits up to9 years of treatment. JAMA 289: 1645–1651.

Stevens MCG, Padwick M, Serjeant GR. 1981. Observationson the natural history of dactylitis in homozygous sicklecell disease. Clin Pediatr 20: 311–317.

Thomas AN, Pattison C, Serjeant GR. 1982. Causes of deathin sickle-cell disease in Jamaica. Br Med J 285: 633–635.

Thompson J, Hambleton IR, Reid M, Serjeant GR. 2007.Albuminuria and renal function in homozygous sickle

cell disease: Observations from a cohort study. Arch In-tern Med 167: 701–708.

Weatherall MW, Higgs DR, Weiss H, Weatherall DJ, Ser-jeant GR. 2005. Genotype/phenotype relationships insickle cell disease: A pilot twin study. Clin Lab Haematol27: 384–390.

Wierenga KJ, Hambleton IR, Lewis NA. 2001. Survival esti-mates for patients with homozygous sickle-cell disease inJamaica: A clinic-based population study. Lancet 357:680–683.

Wright J, Thomas P, Serjeant GR. 1997. Septicemia causedby salmonella infection; an overlooked complication ofsickle cell disease. J Pediatr 130: 394–399.

Xu K, Shi ZM, Veeck LL, Hughes MR, Rosenwaks Z. 1999.First unaffected pregnancy using preimplantation genet-ic diagnosis for sickle cell anemia. JAMA 281: 1701–1706.

Yang Y-M, Shah AK, Watson M, Mankad V. 1995. Compar-ison of costs to the health sector of comprehensive andepisodic health care for sickle cell disease patients. PublicHealth Rep 110: 80–86.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



June 28, 20132013; doi: 10.1101/cshperspect.a011783 originally published onlineCold Spring Harb Perspect Med

Graham R. Serjeant The Natural History of Sickle Cell Disease

Subject Collection Hemoglobin and Its Diseases

The Natural History of Sickle Cell DiseaseGraham R. Serjeant Hemoglobin Synthesis

Transcriptional Mechanisms Underlying

Nathaniel J. Pope, et al.Koichi R. Katsumura, Andrew W. DeVilbiss,

Current Management of Sickle Cell Anemia

WarePatrick T. McGann, Alecia C. Nero and Russell E. Disease

Iron Deficiency Anemia: A Common and Curable

Jeffery L. Miller

TherapiesTargetedNew Disease Models Leading the Way to

Cell-Free Hemoglobin and Its Scavenger Proteins:

Dominik J. Schaer and Paul W. Buehler

Management of the ThalassemiasNancy F. Olivieri and Gary M. Brittenham

-ThalassemiaαClinical Manifestations of Elliott P. Vichinsky

-ThalassemiaβThe Molecular Basis of Swee Lay Thein

Erythroid Heme Biosynthesis and Its DisordersHarry A. Dailey and Peter N. Meissner

Erythropoiesis: Development and DifferentiationElaine Dzierzak and Sjaak Philipsen

Clinical CorrelatesHemoglobin Variants: Biochemical Properties and

Gell, et al.Christopher S. Thom, Claire F. Dickson, David A.

ErythropoietinH. Franklin Bunn

The Prevention of ThalassemiaAntonio Cao and Yuet Wai Kan

Classification of the Disorders of HemoglobinBernard G. Forget and H. Franklin Bunn

The Switch from Fetal to Adult HemoglobinVijay G. Sankaran and Stuart H. Orkin

-ThalassemiaαThe Molecular Basis of Douglas R. Higgs

http://perspectivesinmedicine.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2013 Cold Spring Harbor Laboratory Press; all rights reserved


http://perspectivesinmedicine.cshlp.org/cgi/collection/

http://perspectivesinmedicine.cshlp.org/cgi/collection/


Documents

The Natural History of Sickle Cell Diseaseperspectivesinmedicine.cshlp.org/content/3/10/a011783.full.pdf · The Natural History of Sickle Cell Disease ... enough under normal physiological