By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Please be advised that we experienced an unexpected issue that occurred on Saturday and Sunday January 20th and 21st that caused the site to be down for an extended period of time and affected the ability of users to access content on Wiley Online Library. This issue has now been fully resolved. We apologize for any inconvenience this may have caused and are working to ensure that we can alert you immediately of any unplanned periods of downtime or disruption in the future.
Professor G. R. Serjeant, Sickle Cell Trust, 14 Milverton Crescent, Kingston 6, Jamaica.
Objective To compare pregnancy outcome in sickle cell-haemoglobin C (SC) disease with that in homozygous sickle cell (SS) disease and age-matched controls with a normal haemoglobin (AA) genotype.
Design A cohort study followed from birth.
Setting Sickle Cell Clinic, University Hospital and other Jamaican hospitals.
Population Ninety-five pregnancies in 43 patients with SC disease, 94 pregnancies in 52 patients with SS disease and 157 pregnancies in 68 controls.
Methods Systematic review of all pregnancies occurring in sample population. Kaplan–Meier analysis for interval to first pregnancy, and the t test, χ2 test or Fisher's exact test as appropriate; correction was made for multiple testing and multiple linear regression was used for analysis of determinants of birthweight.
Main outcome measures Age at menarche, interval to first pregnancy, outcome of pregnancy, maternal complications and possible predictors of low birthweight.
Results Menarche was marginally delayed in SC disease compared with AA controls (median age 13.7 vs 13.0 years, P= 0.02) but age at first pregnancy was similar (median age 22.5 vs 20.1 years, P= 0.32). Pregnancy outcome in SC disease did not differ from AA controls but compared with SS disease there were marginally fewer miscarriages, more live deliveries and greater birthweight. The prevalence of pregnancy-induced hypertension, pre-eclampsia, antepartum or postpartum haemorrhage in SC disease did not differ from AA controls but the prevalence of sickle-related complications was similar to SS disease.
Conclusions Contrary to some claims, pregnancy outcome in SC disease is generally benign compared with SS disease.
There is general agreement that homozygous sickle cell (SS) disease runs a generally severe, but markedly variable, clinical course compared with sickle cell-haemoglobin C (SC) disease. However, pregnancy-related problems have been claimed to be more common in SC disease. Distinction of the genotypes was not always possible until the introduction of haemoglobin electrophoresis in 1954, but an early review of pregnancy1 reported three maternal deaths, two confirmed and one suspected to have SC disease. This study found complications in all five SC patients, whereas pregnancies were uneventful in five SS patients, the authors concluding that ‘the morbidity and mortality during pregnancy of patients with SC disease is in sharp contrast to the relatively benign pregnancies of our patients with sickle cell anemia’. Perinatal mortality rates were also higher in SC compared with SS disease.2 There was an impression of greater severity during pregnancy3–7 and pooling of data from both genotypes in another report8 assumed similar severity. These studies are subject to ascertainment biases likely to emphasise more severely affected patients. The Jamaican Cohort Study in which a representative sample of patients with SC and SS disease, along with matched controls with a normal haemoglobin (AA) genotype, have been followed from birth, affords an opportunity to compare obstetric performance in these genotypes. The data in SS disease and in AA controls have been published9 and we now present pregnancy outcome in SC disease.
SUBJECTS AND METHODS
A Cohort Study from birth10,11 based on screening 100,000 consecutive deliveries at the main Government maternity hospital (Victoria Jubilee Hospital) from June 1973 detected 315 babies with SS disease and 201 with SC disease over an 8 1/2 year period. The first 125 SS babies were each matched with two controls with an AA phenotype of the same sex born closest in time (usually immediately before and after) resulting in 250 controls collected over the first 2 1/2 years. Of the 201 babies with SC disease, only the initial 173 were recruited to the pregnancy study and outcome is compared with normal controls in Fig. 1. Patients with SC disease were matched to the nearest SS subject at a median interval of +2 days (range −22 to +31 days). At the study date (31 October 2004), the age range for mothers with SC disease (23.9–31.3 years) was wider than for AA controls (27.4–29.9).
Subjects and controls were followed at three- to six-month intervals, encouraged to attend at any time when sick and defaulters were actively traced with rates of annual default (missed appointments for one year) consistently below 5% in SC disease. Clinical management was generally supportive and comprehensive but without specific interventions such as chronic transfusion or hydroxyurea. Haematology was monitored at six monthly intervals when well, routine haematological indices were measured in electronic counters (Coulter Electronics, Miami) and fetal haemoglobin by alkali denaturation.12 At each clinic attendance, postpubertal female patients were routinely questioned about suspected or confirmed pregnancies and the majority of women were interviewed in the month prior to the study date. Hospital notes for all deliveries were reviewed except for five SC pregnancies (two delivered at home, three hospital notes were untraceable). Prophylactic transfusion is not used in the management of pregnancy in Jamaica but transfusions were indicated in six pregnancies of five SC mothers (four acute chest syndrome, one haemorrhage [Hb 5.8 g/dL], one anaemia [Hb 4.8 g/dL]), which may be compared with published data for SS disease and AA controls.9 Antenatal clinic attendance, assessed among pregnancy durations of 24 weeks or more in those with adequate data, showed a median of eight visits (range 0–14) for 62 pregnancies in 38 SC mothers, six (range 0–18) for 49 pregnancies in 35 SS mothers and three (range 0–17) for 113 pregnancies in 67 control mothers. No antenatal care occurred in 11/62 (18%) SC pregnancies, 9/49 (18%) SS pregnancies and 47/113 (42%) controls.
Spontaneous pregnancy loss before 24 weeks of gestation was defined as an abortion and after this age, as a stillbirth. In all early losses, pregnancy was confirmed by positive pregnancy tests. The diagnosis of abortion was based on history and only occasionally histologically confirmed. Neonatal deaths applied to live births dying within 28 days. Gestational age was calculated from last menstrual period (LMP) and was considered accurate in 59 SC, 43 SS and 109 controls. Age at first pregnancy was defined by the outcome date. Prematurity or preterm deliveries referred to a gestational age less than 259 days (37 weeks) and low birthweight as below 2500 g. Because ultrasound examinations were not routinely performed, the frequency of intrauterine growth retardation could not be ascertained. Acute chest syndrome was defined as pulmonary symptoms and signs associated with a new pulmonary infiltrate on chest radiograph. Painful crisis referred to typical bone pain of sufficient severity to require narcotic analgesia. Pregnancy-induced hypertension was defined as a systolic pressure ≥140 mmHg or a diastolic pressure ≥90 mmHg after 20 weeks of gestation in the absence of proteinuria, in mothers who were not chronically hypertensive and in whom blood pressure returned to normal by one month after delivery.13 Pre-eclampsia was defined by the same blood pressure criteria but with proteinuria. Proteinuria was defined by Dipstick as trace proteinuria or more in SC and SS mothers and 1+ or more in control mothers; the difference recognising the lower urine concentration in SC and SS disease. Lesser amounts were considered as absence of proteinuria. Other definitions have been described elsewhere.9
Age at first pregnancy and the interval between menarche and first pregnancy were calculated by Kaplan–Meier estimates and differences tested by the log-rank test. Because SC and SS patients were recruited over a longer period and younger subjects had less time to become pregnant, survival calculations also controlled for year of birth using the Cox proportional hazards model. The probability of having a baby (yes/no) was compared in patients and controls using logistic regression, and the number of births compared using linear regression. Pregnancy outcome and pregnancy-related complications were assessed by the t test for mean summaries, the χ2 test for counts of events and Fisher's exact test when counts were <10. Because 26 univariate tests were performed, there was a reasonable chance of a false-positive result, and a conservative correction for multiple testing required a P value of 0.002 before assuming significance at the 5% level. Potential determinants of birthweight were assessed by multiple linear regression, examining gestational age (weeks), mother's age (years), parity, mother's pre-pregnant weight and babies' gender; three regression analyses were performed for the separate genotypes. The regressions in SC and SS disease also considered the potential predictive effects of maternal total haemoglobin and fetal haemoglobin levels (defined as the average of all steady state observations over the age of five years) and the total number of clinical events (painful crisis, acute chest syndrome and urinary tract infections) during pregnancy up to one week post-delivery as a single continuous variable. A robust estimate of standard error14 was used to account for repeated births by the same mother in all univariate tests and regressions. All analyses were conducted using Stata 8 (Stata Corp 2003, Stata Statistical Software: Release 8.0, College Station, Texas, USA).
The Cohort Study was approved by the Ethical Committee of the University Hospital of the West Indies. Mothers whose children were recruited to the Cohort gave informed consent and each subject was re-interviewed at age 18 years and gave informed consent to continue in the study. Specific permission was obtained to present pregnancy data.
The study group included 95 completed singleton pregnancies in 43 SC mothers and 157 pregnancies in 68 AA control mothers (Fig. 1); published data show the results of 94 pregnancies in 52 SS mothers.9 The mean number of pregnancies in those becoming pregnant was 2.2 (range 1–7) in SC, compared with 1.8 (1–5) in SS, and 2.3 (1–6) among controls. The proportion of SC subjects becoming pregnant did not differ from AA controls before (odds ratio 1.4 times, 95% CI 0.7–2.7, P= 0.38) or after (odds ratio 0.7, 95% CI 0.2–2.2, P= 0.56) adjustment for year of birth.
Age at menarche in SC women was 13.7 [1.9] years compared with 15.4 [1.6] years in SS and 13.0 [1.1] years in controls. Menarche was significantly earlier in SC women compared with SS (1.7 years, P < 0.001) and marginally later than AA controls (0.6 years, P= 0.02), similar to previous reports in this group.15 Age at first pregnancy (Fig. 2) in SC disease did not differ from SS disease (median age 22.5 years, 95% CI 19.7–24.2 vs 23.7 years, 95% CI 21.5–26.5; hazard ratio 1.4, 95% CI 0.9–2.1, P= 0.12) or AA controls (median age 20.1 years, 19.2–21.3; hazard ratio 1.2, 95% CI 0.8–1.9, P= 0.40). The interval between menarche and first pregnancy in SC disease (median 8.9 years, 95% CI 6.6–11.1) was similar to SS disease (9.0 years, 95% CI 6.8–10.5; hazard ratio adjusted for year of birth 1.0, 0.7–1.5, P= 0.97) and AA controls (7.1, 6.2–8.9; hazard ratio adjusted for year of birth 1.2, 0.8–1.8, P= 0.46).
SC mothers had marginally fewer miscarriages, more live deliveries and higher mean birthweights when compared with SS mothers and did not differ from AA controls (Table 1). The frequency of pregnancy terminations was similar in all genotypes and was generally for social reasons. No differences occurred in stillbirth frequency. Of the two stillbirths in SC disease, one occurred in a 29-year-old unbooked primipara, presenting with pyrexia and painful crisis at 31 weeks of gestation. An ultrasound scan diagnosed intrauterine death, and labour was induced. The other was a 25-year-old primipara admitted for three days at 35 weeks of gestation with a painful crisis and pregnancy-induced hypertension. A week later she was readmitted with lower abdominal pain and ultrasound confirmed intrauterine death. There was one neonatal death following an uneventful pregnancy up to 33 weeks of gestation, when the 24-year-old mother developed acute chest syndrome. A 2.2-kg female infant was delivered by caesarean section for fetal distress at 34 weeks of gestation, but died within 8 hours from sepsis and respiratory distress.
Table 1. Pregnancy outcome in SC disease, SS disease and AA controls. χ2 test adjusted to allow for multiple pregnancies per subject.
SC versus SS
SC versus AA
Includes assisted deliveries by forceps (2 SS, 1 AA).
Excludes four individuals (3 SS, 1 AA) in whom gestational ages >43 weeks were most likely due to incorrect dates.
Significant after adjusting for multiple testing (0.002 required for significance at 5% level).
Mode of delivery did not differ between genotypes. In SC mothers, 14 deliveries were by caesarean section, indicated for cephalopelvic disproportion (2), intrauterine growth retardation (1), fetal distress (2), malpresentation (1), severe pre-eclampsia (3) and previous sections (5). Pregnancy outcomes were successful in 85% SC mothers, 57% SS and 89% controls.
Compared with a highly significant effect of maternal genotype on birthweight in SS mothers (estimate −0.30, 95% CI −0.49 to −0.11, P= 0.002), maternal SC disease had no discernible effect (SC vs AA controls, estimate 0.07, 95% CI −0.09 to 0.23, P= 0.39). Birthweight was associated with gestational age in all genotypes but the only additional factors in SC mothers were maternal weight (regression model adjusted for all other terms 0.06 (95% CI 0.01–0.12, P < 0.03) and babies' gender [male compared with female, 0.23 (0–0.46), P= 0.05]. Of interest was the lack of an effect of sickle-related clinical events during pregnancy in SC mothers compared with a highly significant contribution previously shown in SS mothers.
Problems during pregnancy are compared in the three genotypes in Table 2. There were no genotype differences in pregnancy-induced hypertension, pre-eclampsia, antepartum or postpartum haemorrhage. No SC pregnancies manifested eclampsia compared with one SS and one AA and none had a retained placenta. There were no genotype differences in sickle-related complications and the number of painful crises in SC disease was influenced by a small subgroup (nine painful crises occurred during four pregnancies in two mothers). Most sickle-related problems occur late in pregnancy and in these SC patients, this applied to all episodes of painful crises (5 second trimester, 14 third, 8 intrapartum or postpartum) and the acute chest syndrome (2 third trimester, 4 postpartum).
Table 2. Pregnancy-associated complications in SC disease, SS disease and AA controls beyond 24 weeks of gestation.
SC versus SS
SC versus AA
Total pregnancies (n)
Uneventful pregnancies (%)
Painful crises (n)
Pregnancies affected (%)
Acute chest syndrome (n)
Pregnancies affected (%)
Urinary tract infections (n)
No maternal deaths occurred among SC patients but an 18-year-old primipara became hypertensive and febrile during labour. After delivery she developed respiratory distress and thrombocytopenia, was transfused two units of packed cells, became hypertensive (202/100 mmHg) and proteinuric and increasingly hypoxic with a Po2 of 62.6 mmHg on 10 L O2. On day 3 postpartum, she became verbally unresponsive and was transferred to the ICU for respiratory monitoring and support with diagnoses of pre-eclampsia and pulmonary embolism, remained in the ICU for four days and was discharged from hospital well on day 11 postpartum. She subsequently had two uneventful pregnancies.
The present study leaves little doubt that pregnancy is more benign in SC disease than in SS disease and differed only minimally from AA controls. Menarche in SC disease was marginally delayed compared with AA controls and the age at first pregnancy and interval between menarche and first pregnancy in SC disease were intermediate between SS disease and AA controls. Birthweight in SC pregnancies was similar to AA controls but there was a tendency towards a lower gestational age. This benign outcome is consistent with the behaviour of the SC genotype, which is often mild and may not be diagnosed until later in adult life. Manifestations cited as more severe in SC disease include eye problems, bone marrow embolism, hip necrosis and pregnancy. Proliferative sickle retinopathy is undoubtedly more common16 in SC disease compared with SS disease and the mechanism for this is still not understood. Severe bone pain with bone marrow embolism has been described in pregnant patients with SC disease,17–20 but the evidence that this is more common in SC than SS disease is not convincing. Hip necrosis and pregnancy-related problems are both significantly more common in SS disease than in SC disease and this false impression appears to have resulted from the symptomatic selection of patients.
This symptomatic selection may have been accentuated by differential survival of the two genotypes especially in communities where there is a high mortality of SS disease. Thus, SS disease is more common at birth in Nigeria, but because of the higher early mortality, pregnancy is seen more frequently among SC disease. In one Nigerian review,21 there were 21 pregnancies in SS disease and 145 in SC disease, whereas in the Cooperative Study in the United States,22 these relative frequencies were reversed (174 SS, 36 SC). Thus, in Africa especially 30–40 years ago, SC patients were more likely to become pregnant and pregnancy-related problems were therefore seen more frequently in SC disease. Furthermore, the SS patients who did survive to adult life in these populations tended to have ameliorating factors determining a benign outcome in pregnancy. So hospital-based observations may have selected mildly affected cases of SS disease and provided larger pregnant SC groups among whom problems could develop. The impression of a severe outcome in SC disease was also derived from reports in the United States; a review of literature23 from 1956 to 1973 found 581 pregnancies in 201 SC subjects with a maternal mortality of 2.2 per 100 pregnancies and significant morbidity especially in some series.2,7,24,25 Selection of mildly affected cases of SS disease may also have exaggerated the relative severity in SC disease in the United States, prompting Curtis7 to comment that although fewer women with SS disease became pregnant, those who did so had milder courses than women with SC disease. A further factor could be that pregnancy in SC disease may precipitate sickle cell problems for the first time, occurring against the background of a generally mild disease and the ensuing distress more apparent than problems occurring against the more symptomatic background of SS disease. Studies with substantial numbers of patients of both genotypes confirm a more severe clinical course during pregnancy in SS disease21,23,26–33 and this impression is strengthened by the mild course of SC patients noted in the Jamaican Cohort Study.
Although most SC patients are mildly affected, some do have serious complications. It is not yet possible to predict those SC patients who will develop severe complications in pregnancy and it would be wise to monitor all pregnancies in SC disease closely with delivery in hospital.
Most of this work was conducted during the tenure of the MRC Laboratories (Jamaica), University of the West Indies, Kingston, Jamaica.