- Email: [email protected]
Sickle cell disease in pregnancy: Twenty years of experience at Grady Memorial Hospital, Atlanta, Georgia
Sickle cell disease in pregnancy: Twenty years of experience at Grady Memorial Hospital, Atlanta, Georgia Phoebe Mou Sun, MD, Wanda Wilburn, MD, B. Denise Raynor, MD, and Denise Jamieson, MD, MPH Atlanta, Georgia OBJECTIVE: We compared pregnancy outcomes among women with sickle cell disease with outcomes for African American women without the disease. STUDY DESIGN: We selected 127 deliveries in women with sickle cell disease (hemoglobin SS or hemoglobin SC) that occurred between 1980 and 1999. A control group of 129 deliveries by African American women with normal hemoglobin (hemoglobin AA) was also selected. Evaluated pregnancy outcomes included low birth weight, prematurity, intrauterine growth restriction, antepartum hospital admission, preterm labor or preterm premature rupture of membranes, postpartum infection, preeclampsia, pyelonephritis, intrauterine fetal death, perinatal mortality, and maternal mortality. RESULTS: Compared with deliveries among women with hemoglobin AA, deliveries among women with hemoglobin SS or hemoglobin SC were at increased risk for intrauterine growth restriction, antepartum hospital admission, and postpartum infection. In addition, deliveries among women with Hb SS were more likely to be complicated by low birth weight, prematurity, and preterm labor or preterm premature rupture of membranes when compared with deliveries among women with hemoglobin AA. There were no significant differences among the groups (hemoglobin SS, hemoglobin SC, and hemoglobin AA) in terms of perinatal deaths; there were no maternal deaths in the study population. CONCLUSION: Those caring for women with sickle cell disease should be aware that they are at increased risk for pregnancy complications, although overall pregnancy outcome is favorable. (Am J Obstet Gynecol 2001;184:1127-30.)
Key words: sickle cell disease, pregnancy, pregnancy outcomes
Sickle cell disease is the most common hemoglobinopathy in the United States, with 8% of African American subjects carrying the sickle hemoglobin gene.1 Sickle cell disease results from the pairing of hemoglobin S (Hb S) with another abnormal hemoglobin; the most frequent and severe phenotypes are hemoglobin SS (Hb SS) and hemoglobin SC (Hb SC). In the past, pregnancy for patients with sickle cell disease was fraught with complications, including increased rates of maternal and perinatal mortality. For example, a study published by Hendrickse et al2 in 1972 reported maternal mortality as high as 11.5%. In response to reports of poor maternal and fetal outcomes among women with sickle cell disease, some clinicians advocated counseling sickle cell patients to avoid pregnancy and recommended primary sterilization, elective abortion, and postpartum sterilization.3 More recently, most authorities suggest careful prenatal monitoring at an institution accustomed to sickle cell complications.4 From the Department of Gynecology and Obstetrics, Emory University School of Medicine. Reprint requests: Phoebe Mou Sun, MD, 69 Butler St SE, Suite 142, Atlanta, GA 30303. Copyright © 2001 by Mosby, Inc. 0002-9378/2001 $35.00 + 0 6/1/115477 doi:10.1067/mob.2001.115477
In a 1986 study reviewing maternal mortality before and after 1972, Powars et al5 reported a maternal mortality rate averaging 4.1% before 1972 and dropping to 1.7% after 1972. Perinatal mortality, although still substantial, had also decreased from 52.7% to 22.7% after 1972 according to the study of Powars et al.5 These significant decreases in maternal and perinatal mortality may be attributed to advancements in general medical care for patients with sickle cell disease, improvements in transfusion medicine, and earlier and more frequent prenatal care, as well as advancements in neonatalogy. In this study we sought to describe the characteristics of pregnancies complicated by sickle cell disease, which were followed at a large, urban, public hospital in the United States. This study reviews maternal and fetal outcomes for pregnancies managed at our institution within the last two decades. Material and methods Using a computerized database of deliveries at our institution, we identified 119 deliveries by 100 women with sickle cell disease (SS) and 118 deliveries by 90 women with sickle hemoglobin C disease (SC), who received obstetric care at our institution. Of these patients, medical records were available for 58% (69/119) of deliveries among women with SS and 49% (58/118) of deliveries 1127
1128 Sun et al
May 2001 Am J Obstet Gynecol
Table I. Characteristics of study population by hemoglobin genotype (percent)
Age <20 y 20-25 y 26-30 y 31-35 y >35 y Parity 0 1-3 ≥4 First prenatal visit 1st trimester 2nd trimester 3rd trimester No prenatal care Route of delivery Cesarean Vaginal
Statistical significance
AA
SS
SC
30 43 16 7 4
13 50 22 12 3
28 52 10 5 5
P = .22*
35 56 9
38 56 6
19 74 7
P = .11*
46 36 12 7
47 33 10 10
38 27 15 20
P = .29*
22 78
36 64
29 71
P = .11*
*Not statistically significant.
among women with SC. Data for the remaining patients were not available because of missing medical records or transfer of care to another hospital before delivery. This study was based on prenatal records, hospital admissions, and a database of deliveries at >20 estimated weeks’ gestation; thus elective abortions and unreported spontaneous abortions were not included. A control group of Hb AA African American women, who were delivered of infants during the same time period (1980-1999), was randomly selected from the computerized database from the same month and year as SS and SC deliveries; medical records were available for 65% of AA deliveries. For each hospitalization and delivery, the following information was abstracted: age, parity, phenotype, gestational age at delivery, mode of delivery, birth weight, birth weight percentile, Apgar score at 5 minutes, number of prenatal visits and gestational age at first visit, antepartum hospital admissions and length of stay, baseline hematocrits, number of transfusions, and antepartum, intrapartum, and postpartum complications including pain crises, pneumonia, heart failure, pyelonephritis, preterm labor, preterm premature rupture of membranes, chorioamnionitis, preeclampsia, postpartum infection, postpartum hemorrhage, perinatal mortality, and maternal mortality. Hemoglobin phenotype was based on hemoglobin electrophoresis performed at the institutional clinical laboratory; in this study phenotype was characterized as SS, SC, or AA (normal). Estimated gestational age was based on last menstrual period confirmed by ultrasonography, when available, at initiation of care. Low birth weight was defined as birth weight <2500 g. Intrauterine growth restriction (IUGR) was defined as birth weight <10th percentile for gestational age. Blood transfusion was performed for symp-
tomatic anemia or anemia with concurrent pain crisis or infection; prophylactic transfusions to maintain a certain hemoglobin or hematocrit level throughout pregnancy were not performed. Preterm delivery was defined as delivery at a gestational age <37 weeks. The sickle cell crises noted in this study were defined as acute pain episodes necessitating hospitalization for analgesia and intravenous hydration. Hypertensive disorders included chronic hypertension, pregnancy-induced hypertension, and preeclampsia. Preeclampsia was not differentiated into mild or severe in several cases and thus was not defined as such in our study. Pulmonary complications included pneumonia, acute chest syndrome, and pulmonary edema; patients had documented acute abnormalities on chest radiographs and increased oxygen requirement. Diagnoses of pyelonephritis were based on positive urine cultures with either fever or costovertebral angle tenderness. Data analysis was performed with the EpiInfo version 6 (Centers for Disease Control and Prevention, Atlanta, Ga) to calculate Mantel-Haenszel χ2 P values. A P value ≤ .05 was considered statistically significant. Unadjusted relative risks with 95% confidence intervals were also calculated. Power calculations were also performed for each variable to determine adequate sample size for detecting a significant difference among patients with SS, SC, and AA. For these calculations the significance level was chosen to be α = .05, and the power of the test was chosen to be β = .80. The study was submitted to and approved by the human investigations committee at our institution. Results This study involved 256 deliveries, categorized by the hemoglobin phenotype of 251 mothers; this included 69 SS deliveries, 58 SC deliveries, and 129 AA deliveries. An overview of the study population showed that most of the women in the study were young, with 72% being ≤25 years old. A total of 30% of the study population was primiparous, whereas only 7% of the women had had ≥4 previous pregnancies. The 3 study groups (SS, SC, AA) were similar in terms of age, parity, trimester of first prenatal visit, and route of delivery (Table I). Antepartum period. SS patients required significantly more admissions than AA patients; the relative risk for antepartum admission was 6.8 (95% confidence interval, 3.8-11.9) for SS patients compared with the referent group, the AA patients (Table II). The average length of stay for SS patients was 5.3 days. SC patients also showed an increased risk for antepartum admission with a relative risk of 2.8 (95% confidence interval, 1.4-5.6) when compared with AA patients (Table III). Antepartum admissions for SS and SC patients were most commonly the result of sickle pain crises, pyelonephritis, and anemia. In 34 SS pregnancies (48%) and 11 SC pregnancies (19%) the complications included one or more pain crises that re-
Sun et al 1129
Volume 184, Number 6 Am J Obstet Gynecol
Table II. Poor pregnancy outcomes for SS pregnancies
Table III. Poor pregnancy outcomes for SC pregnancies Relative risk and % of SC 95% confidence pregnancies interval*
Relative risk and % of SS 95% confidence pregnancies interval* Antepartum hospital admission Preterm labor or preterm premature rupture of membranes Prematurity Low birth weight IUGR Pyelonephritis Preeclampsia Postpartum infection Intrauterine fetal death Perinatal mortality
62 13
6.8 (3.8-11.9) 2.8 (1.1-7.6)
45 46 45 7 10 22 4 11
4.1 (2.4-7.2) 2.7 (1.7-4.3) 4.9 (2.7-8.9) 3.1 (0.8-12.8) 1.9 (0.7-4.2) 9.4 (2.8-31.4) 5.7 (0.6-53.0) 3.0 (1.0-8.9)
*Referent group, AA pregnancies.
quired antepartum admission (Table IV). Other less common diagnoses for antepartum admissions included pneumonia, uncontrolled diabetes, preterm labor, preterm premature rupture of membranes, and congestive heart failure. SS patients also had lower baseline hematocrits (average, 23%) and required more antepartum blood transfusions than SC patients (59% vs 21%). There were 42 SS patients (59%) who required antepartum blood transfusions (Table IV), and SS patients received an average of 4.64 units of packed red blood cells. SC patients were significantly less likely than SS patients to receive blood transfusion; 12 SC patients (21%) required blood transfusions, receiving an average of 2.16 units of packed red blood cells. The incidence of transfusion antibodies was not significantly greater for SS patients than for SC patients (11% vs 5%; P = .2)). No AA patients received transfusions in this study. Intrapartum period. The risk of preterm labor and premature rupture of membranes (Tables II and III) was significantly increased for SS patients compared with AA patients, with a relative risk of 2.8 (95% confidence interval, 1.1-7.6), but not for SC patients (relative risk, 1.5; 95% confidence interval, 0.4-5.1). SS patients were also more likely than AA patients to undergo preterm delivery with an average gestational age at delivery of 34.1 weeks; relative risk for prematurity was 4.1 (95% confidence interval, 2.4-7.2) for SS patients. Infants of SS patients also had a higher risk of low birth weight (relative risk, 2.7; 95% confidence interval, 1.7-4.3) and IUGR (relative risk, 4.9; 95% confidence interval, 2.7-8.9) compared with AA patients. SC patients were also at higher risk for IUGR (relative risk, 2.2; 95% confidence interval, 1.1-4.7) but not for preterm delivery (relative risk, 2.1; 95% confidence interval, 1.0-4.1) or low birth weight (relative risk, 1.0; 95% confidence interval, 0.5-2.0). Five-minute Apgar scores were not significantly different for the 3 phenotypes; approximately 90% of SS, SC, and AA deliveries
Antepartum hospital admission Preterm labor or preterm premature rupture of membranes Prematurity Low birth weight IUGR Pyelonephritis Preeclampsia Postpartum infection Intrauterine fetal death Perinatal mortality
26 7
2.8 (1.4-5.6) 1.5 (0.4-5.1)
22 17 21 5 3 10 2 2
2.1 (1.0-4.1) 1.0 (0.5-2.0) 2.2 (1.1-4.7) 2.2 (0.5-10.8) 0.6 (0.1-2.9) 4.5 (1.2-17.3) 2.2 (0.1-35.2) 0.5 (0.1-3.8)
*Referent group, AA pregnancies.
Table IV. Description of morbidity and mortality for SS and SC pregnancies
Blood transfusions (%) Blood antibodies (%) Pain crises (%) Maternal mortality (No.) Perinatal mortality (No.)
SS
SC
Statistical significance
59 11 48 0 9
21 5 19 0 1
P < .001 P = .2* P < .001 — P = .1*
*Not statistically significant.
had Apgar scores >7 at 5 minutes. Hypertensive disorders and pulmonary complications were not significantly increased in SS or SC patients compared with AA patients. There was no significant increase in risk of preeclampsia for SS patients (relative risk, 1.9; 95% confidence interval, 0.69-5.2) or SC patients (relative risk, 0.6; 95% confidence interval, 0.1-2.9) compared with AA patients. Postpartum period. SS and SC patients were more likely than AA patients to be diagnosed with postpartum infections (Tables II and III), most commonly endomyometritis or pyelonephritis. The relative risk for postpartum infection was 9.4 (95% confidence interval, 2.8-31.4) for SS patients and 4.5 (95% confidence interval, 1.2-17.3) for SC patients. No significant increase in risk of postpartum hemorrhage was noted for SS or SC patients when compared with AA patients. Comparison of route of delivery (Table I) showed no significant difference in the rate of cesarean delivery (36% SS vs 28% SC vs 22% AA deliveries). Cesarean delivery was most often performed because of fetal distress, failure to labor to progress, malpresentation, or elective repeat cesarean; however, the study did not distinguish between elective and nonelective cesarean deliveries. Rates of spontaneous abortions and intrauterine fetal deaths (7% and 4%, respectively) were not significantly different for SS patients in our study. Likewise, there was no significant difference in perinatal mortality rates;
1130 Sun et al
8 perinatal deaths were reported for SS pregnancies (11%) vs 1 (2%) for SC pregnancies and 5 (4%) for AA pregnancies. No maternal deaths were reported for the patients in this study. Comment Within the last several decades, modern technology and scientific advancement have revolutionized sickle cell disease management and obstetric and perinatal care, allowing more aggressive intervention and subsequent improvement of pregnancy outcomes for women with sickle cell disease. Both maternal and perinatal mortality have dropped significantly for patients with sickle cell hemoglobinopathies5; thus the recommendations of many years ago for primary sterilization or pregnancy termination are no longer warranted for sickle cell patients.3 Pregnancy for these patients is, however, still characterized by a higher rate of complications. This study shows that SS patients have an increased risk for maternal complications such as preterm labor and premature rupture of membranes, antepartum admission, and postpartum infection. Furthermore, SS patients are also at higher risk for fetal complications such as low birth weight, IUGR, and preterm delivery. Women with hemoglobin SC have an increased risk for IUGR, antepartum hospital admission, and postpartum infection, although to a lesser extent than SS patients. These findings are consistent with other earlier studies performed in the United States.5, 7, 9, 10 Of note, the risk of postpartum infection may be affected by the increased rate of cesarean delivery for Hb SS; however, the cesarean rate increase was not statistically significant. Several limitations of this study should be noted. Because of the relatively small sample sizes, the study had inadequate power to detect clinically significant differences for the following outcomes: intrauterine fetal death, pyelonephritis, preeclampsia, and perinatal mortality. There may also have been confounding factors not controlled for in our study, which included only univariate analysis. When antepartum complications were considered, SS patients were most commonly hospitalized for pain crises, infections, and anemia, requiring an average of 5 days of hospitalization. More than half of the SS patients required at least one blood transfusion, compared with one out of five SC patients who required at least one blood transfusion. Likewise, almost half of the SS patients had one or more documented pain crises during pregnancy, compared with one out of five SC patients who had one or more pain crises. These findings revisit the issue of improving pregnancy outcome with prophylactic blood
May 2001 Am J Obstet Gynecol
transfusions. Koshy et al6 showed that prophylactic transfusions did not change perinatal outcome but decreased the frequency of pain crises. Previous studies have also shown an increased risk for preeclampsia, pyelonephritis, intrauterine fetal death, and perinatal mortality.5, 7, 9 Our study found no significant change in these outcomes for SS or SC patients; however, our sample size had insufficient power to detect significant differences for these variables. Early pregnancy outcomes such as spontaneous or elective first-trimester abortions were excluded from this study, because the perinatal database does not include deliveries at <20 weeks’ estimated gestational age. Sickle–β-thalassemia patients were excluded from this study because of the small number of these patients available for study. Although maternal and perinatal mortality rates have decreased significantly in the last several decades, sickle cell disease in pregnancy remains associated with a higher risk of multiple maternal and fetal complications. These findings warrant close monitoring of all patients with sickle cell disease throughout pregnancy. Prenatal care should be available at an institution prepared to manage sickle cell disease complications and high-risk pregnancies. Despite the risks, most women with sickle hemoglobinopathies can expect good pregnancy outcomes. REFERENCES
1. Motulsky AG. Frequency of sickling disorders in U.S. blacks. N Engl J Med 1973;288:31-3. 2. Hendrickse JP, Watson-Williams EJ, Luzzatto L, Ajabor LN. Pregnancy in homozygous sickle cell anemia. J Obstet Gynaecol Br Commonw 1972;79:396-409. 3. Fort AT, Morrison JC, Berreras L, Diggs LW, Fish SA. Counseling the patient with sickle cell disease about reproduction. Pregnancy outcome does not justify the maternal risk. Am J Obstet Gynecol 1971;111:324-7. 4. Gabbe SG, Niebyl JR, Simpson JL. Obstetrics: normal and problem pregnancies. 3rd ed. Philadelphia: WB Saunders; 1996. 5. Powars DR, Sandhu M, Niland-Weiss J, Johnson C, Bruce S, Manning PR. Pregnancy in sickle cell disease. Obstet Gynecol 1986; 67:217-28. 6. Koshy M, Burd L, Wallace D, Moawad A, Baron J. Prophylactic red-cell transfusions in pregnant patients with sickle cell disease. N Engl J Med 1988;319:1447-52. 7. Smith JA, Espeland M, Bellevue R, Bonds D, Brown AK, Koshy M. Pregnancy in sickle cell disease: experience of the cooperative study of sickle cell disease. Obstet Gynecol 1996;87:199-204. 8. Koshy M, Ashenhurst J. Management of pregnancy in sickle cell anemia. Tex Rep Biol Med 1981;40:273-8. 9. Muhieddine AFS, Cantwell C, Nobles G, Levy DL. Outcome of pregnancies complicated by sickle cell and sickle-C hemoglobinopathies. Am J Perinatol 1994;11:187-91. 10. Brown AK, Sleeper LA, Pegelow CH, Miller ST, Gill FM, Waclawiw MA. The influence of infant and maternal sickle cell disease on birth outcome and neonatal course. Arch Pediatr Adolesc Med 1994;148:1156-62.
Key words: sickle cell disease, pregnancy, pregnancy outcomes
Sickle cell disease is the most common hemoglobinopathy in the United States, with 8% of African American subjects carrying the sickle hemoglobin gene.1 Sickle cell disease results from the pairing of hemoglobin S (Hb S) with another abnormal hemoglobin; the most frequent and severe phenotypes are hemoglobin SS (Hb SS) and hemoglobin SC (Hb SC). In the past, pregnancy for patients with sickle cell disease was fraught with complications, including increased rates of maternal and perinatal mortality. For example, a study published by Hendrickse et al2 in 1972 reported maternal mortality as high as 11.5%. In response to reports of poor maternal and fetal outcomes among women with sickle cell disease, some clinicians advocated counseling sickle cell patients to avoid pregnancy and recommended primary sterilization, elective abortion, and postpartum sterilization.3 More recently, most authorities suggest careful prenatal monitoring at an institution accustomed to sickle cell complications.4 From the Department of Gynecology and Obstetrics, Emory University School of Medicine. Reprint requests: Phoebe Mou Sun, MD, 69 Butler St SE, Suite 142, Atlanta, GA 30303. Copyright © 2001 by Mosby, Inc. 0002-9378/2001 $35.00 + 0 6/1/115477 doi:10.1067/mob.2001.115477
In a 1986 study reviewing maternal mortality before and after 1972, Powars et al5 reported a maternal mortality rate averaging 4.1% before 1972 and dropping to 1.7% after 1972. Perinatal mortality, although still substantial, had also decreased from 52.7% to 22.7% after 1972 according to the study of Powars et al.5 These significant decreases in maternal and perinatal mortality may be attributed to advancements in general medical care for patients with sickle cell disease, improvements in transfusion medicine, and earlier and more frequent prenatal care, as well as advancements in neonatalogy. In this study we sought to describe the characteristics of pregnancies complicated by sickle cell disease, which were followed at a large, urban, public hospital in the United States. This study reviews maternal and fetal outcomes for pregnancies managed at our institution within the last two decades. Material and methods Using a computerized database of deliveries at our institution, we identified 119 deliveries by 100 women with sickle cell disease (SS) and 118 deliveries by 90 women with sickle hemoglobin C disease (SC), who received obstetric care at our institution. Of these patients, medical records were available for 58% (69/119) of deliveries among women with SS and 49% (58/118) of deliveries 1127
1128 Sun et al
May 2001 Am J Obstet Gynecol
Table I. Characteristics of study population by hemoglobin genotype (percent)
Age <20 y 20-25 y 26-30 y 31-35 y >35 y Parity 0 1-3 ≥4 First prenatal visit 1st trimester 2nd trimester 3rd trimester No prenatal care Route of delivery Cesarean Vaginal
Statistical significance
AA
SS
SC
30 43 16 7 4
13 50 22 12 3
28 52 10 5 5
P = .22*
35 56 9
38 56 6
19 74 7
P = .11*
46 36 12 7
47 33 10 10
38 27 15 20
P = .29*
22 78
36 64
29 71
P = .11*
*Not statistically significant.
among women with SC. Data for the remaining patients were not available because of missing medical records or transfer of care to another hospital before delivery. This study was based on prenatal records, hospital admissions, and a database of deliveries at >20 estimated weeks’ gestation; thus elective abortions and unreported spontaneous abortions were not included. A control group of Hb AA African American women, who were delivered of infants during the same time period (1980-1999), was randomly selected from the computerized database from the same month and year as SS and SC deliveries; medical records were available for 65% of AA deliveries. For each hospitalization and delivery, the following information was abstracted: age, parity, phenotype, gestational age at delivery, mode of delivery, birth weight, birth weight percentile, Apgar score at 5 minutes, number of prenatal visits and gestational age at first visit, antepartum hospital admissions and length of stay, baseline hematocrits, number of transfusions, and antepartum, intrapartum, and postpartum complications including pain crises, pneumonia, heart failure, pyelonephritis, preterm labor, preterm premature rupture of membranes, chorioamnionitis, preeclampsia, postpartum infection, postpartum hemorrhage, perinatal mortality, and maternal mortality. Hemoglobin phenotype was based on hemoglobin electrophoresis performed at the institutional clinical laboratory; in this study phenotype was characterized as SS, SC, or AA (normal). Estimated gestational age was based on last menstrual period confirmed by ultrasonography, when available, at initiation of care. Low birth weight was defined as birth weight <2500 g. Intrauterine growth restriction (IUGR) was defined as birth weight <10th percentile for gestational age. Blood transfusion was performed for symp-
tomatic anemia or anemia with concurrent pain crisis or infection; prophylactic transfusions to maintain a certain hemoglobin or hematocrit level throughout pregnancy were not performed. Preterm delivery was defined as delivery at a gestational age <37 weeks. The sickle cell crises noted in this study were defined as acute pain episodes necessitating hospitalization for analgesia and intravenous hydration. Hypertensive disorders included chronic hypertension, pregnancy-induced hypertension, and preeclampsia. Preeclampsia was not differentiated into mild or severe in several cases and thus was not defined as such in our study. Pulmonary complications included pneumonia, acute chest syndrome, and pulmonary edema; patients had documented acute abnormalities on chest radiographs and increased oxygen requirement. Diagnoses of pyelonephritis were based on positive urine cultures with either fever or costovertebral angle tenderness. Data analysis was performed with the EpiInfo version 6 (Centers for Disease Control and Prevention, Atlanta, Ga) to calculate Mantel-Haenszel χ2 P values. A P value ≤ .05 was considered statistically significant. Unadjusted relative risks with 95% confidence intervals were also calculated. Power calculations were also performed for each variable to determine adequate sample size for detecting a significant difference among patients with SS, SC, and AA. For these calculations the significance level was chosen to be α = .05, and the power of the test was chosen to be β = .80. The study was submitted to and approved by the human investigations committee at our institution. Results This study involved 256 deliveries, categorized by the hemoglobin phenotype of 251 mothers; this included 69 SS deliveries, 58 SC deliveries, and 129 AA deliveries. An overview of the study population showed that most of the women in the study were young, with 72% being ≤25 years old. A total of 30% of the study population was primiparous, whereas only 7% of the women had had ≥4 previous pregnancies. The 3 study groups (SS, SC, AA) were similar in terms of age, parity, trimester of first prenatal visit, and route of delivery (Table I). Antepartum period. SS patients required significantly more admissions than AA patients; the relative risk for antepartum admission was 6.8 (95% confidence interval, 3.8-11.9) for SS patients compared with the referent group, the AA patients (Table II). The average length of stay for SS patients was 5.3 days. SC patients also showed an increased risk for antepartum admission with a relative risk of 2.8 (95% confidence interval, 1.4-5.6) when compared with AA patients (Table III). Antepartum admissions for SS and SC patients were most commonly the result of sickle pain crises, pyelonephritis, and anemia. In 34 SS pregnancies (48%) and 11 SC pregnancies (19%) the complications included one or more pain crises that re-
Sun et al 1129
Volume 184, Number 6 Am J Obstet Gynecol
Table II. Poor pregnancy outcomes for SS pregnancies
Table III. Poor pregnancy outcomes for SC pregnancies Relative risk and % of SC 95% confidence pregnancies interval*
Relative risk and % of SS 95% confidence pregnancies interval* Antepartum hospital admission Preterm labor or preterm premature rupture of membranes Prematurity Low birth weight IUGR Pyelonephritis Preeclampsia Postpartum infection Intrauterine fetal death Perinatal mortality
62 13
6.8 (3.8-11.9) 2.8 (1.1-7.6)
45 46 45 7 10 22 4 11
4.1 (2.4-7.2) 2.7 (1.7-4.3) 4.9 (2.7-8.9) 3.1 (0.8-12.8) 1.9 (0.7-4.2) 9.4 (2.8-31.4) 5.7 (0.6-53.0) 3.0 (1.0-8.9)
*Referent group, AA pregnancies.
quired antepartum admission (Table IV). Other less common diagnoses for antepartum admissions included pneumonia, uncontrolled diabetes, preterm labor, preterm premature rupture of membranes, and congestive heart failure. SS patients also had lower baseline hematocrits (average, 23%) and required more antepartum blood transfusions than SC patients (59% vs 21%). There were 42 SS patients (59%) who required antepartum blood transfusions (Table IV), and SS patients received an average of 4.64 units of packed red blood cells. SC patients were significantly less likely than SS patients to receive blood transfusion; 12 SC patients (21%) required blood transfusions, receiving an average of 2.16 units of packed red blood cells. The incidence of transfusion antibodies was not significantly greater for SS patients than for SC patients (11% vs 5%; P = .2)). No AA patients received transfusions in this study. Intrapartum period. The risk of preterm labor and premature rupture of membranes (Tables II and III) was significantly increased for SS patients compared with AA patients, with a relative risk of 2.8 (95% confidence interval, 1.1-7.6), but not for SC patients (relative risk, 1.5; 95% confidence interval, 0.4-5.1). SS patients were also more likely than AA patients to undergo preterm delivery with an average gestational age at delivery of 34.1 weeks; relative risk for prematurity was 4.1 (95% confidence interval, 2.4-7.2) for SS patients. Infants of SS patients also had a higher risk of low birth weight (relative risk, 2.7; 95% confidence interval, 1.7-4.3) and IUGR (relative risk, 4.9; 95% confidence interval, 2.7-8.9) compared with AA patients. SC patients were also at higher risk for IUGR (relative risk, 2.2; 95% confidence interval, 1.1-4.7) but not for preterm delivery (relative risk, 2.1; 95% confidence interval, 1.0-4.1) or low birth weight (relative risk, 1.0; 95% confidence interval, 0.5-2.0). Five-minute Apgar scores were not significantly different for the 3 phenotypes; approximately 90% of SS, SC, and AA deliveries
Antepartum hospital admission Preterm labor or preterm premature rupture of membranes Prematurity Low birth weight IUGR Pyelonephritis Preeclampsia Postpartum infection Intrauterine fetal death Perinatal mortality
26 7
2.8 (1.4-5.6) 1.5 (0.4-5.1)
22 17 21 5 3 10 2 2
2.1 (1.0-4.1) 1.0 (0.5-2.0) 2.2 (1.1-4.7) 2.2 (0.5-10.8) 0.6 (0.1-2.9) 4.5 (1.2-17.3) 2.2 (0.1-35.2) 0.5 (0.1-3.8)
*Referent group, AA pregnancies.
Table IV. Description of morbidity and mortality for SS and SC pregnancies
Blood transfusions (%) Blood antibodies (%) Pain crises (%) Maternal mortality (No.) Perinatal mortality (No.)
SS
SC
Statistical significance
59 11 48 0 9
21 5 19 0 1
P < .001 P = .2* P < .001 — P = .1*
*Not statistically significant.
had Apgar scores >7 at 5 minutes. Hypertensive disorders and pulmonary complications were not significantly increased in SS or SC patients compared with AA patients. There was no significant increase in risk of preeclampsia for SS patients (relative risk, 1.9; 95% confidence interval, 0.69-5.2) or SC patients (relative risk, 0.6; 95% confidence interval, 0.1-2.9) compared with AA patients. Postpartum period. SS and SC patients were more likely than AA patients to be diagnosed with postpartum infections (Tables II and III), most commonly endomyometritis or pyelonephritis. The relative risk for postpartum infection was 9.4 (95% confidence interval, 2.8-31.4) for SS patients and 4.5 (95% confidence interval, 1.2-17.3) for SC patients. No significant increase in risk of postpartum hemorrhage was noted for SS or SC patients when compared with AA patients. Comparison of route of delivery (Table I) showed no significant difference in the rate of cesarean delivery (36% SS vs 28% SC vs 22% AA deliveries). Cesarean delivery was most often performed because of fetal distress, failure to labor to progress, malpresentation, or elective repeat cesarean; however, the study did not distinguish between elective and nonelective cesarean deliveries. Rates of spontaneous abortions and intrauterine fetal deaths (7% and 4%, respectively) were not significantly different for SS patients in our study. Likewise, there was no significant difference in perinatal mortality rates;
1130 Sun et al
8 perinatal deaths were reported for SS pregnancies (11%) vs 1 (2%) for SC pregnancies and 5 (4%) for AA pregnancies. No maternal deaths were reported for the patients in this study. Comment Within the last several decades, modern technology and scientific advancement have revolutionized sickle cell disease management and obstetric and perinatal care, allowing more aggressive intervention and subsequent improvement of pregnancy outcomes for women with sickle cell disease. Both maternal and perinatal mortality have dropped significantly for patients with sickle cell hemoglobinopathies5; thus the recommendations of many years ago for primary sterilization or pregnancy termination are no longer warranted for sickle cell patients.3 Pregnancy for these patients is, however, still characterized by a higher rate of complications. This study shows that SS patients have an increased risk for maternal complications such as preterm labor and premature rupture of membranes, antepartum admission, and postpartum infection. Furthermore, SS patients are also at higher risk for fetal complications such as low birth weight, IUGR, and preterm delivery. Women with hemoglobin SC have an increased risk for IUGR, antepartum hospital admission, and postpartum infection, although to a lesser extent than SS patients. These findings are consistent with other earlier studies performed in the United States.5, 7, 9, 10 Of note, the risk of postpartum infection may be affected by the increased rate of cesarean delivery for Hb SS; however, the cesarean rate increase was not statistically significant. Several limitations of this study should be noted. Because of the relatively small sample sizes, the study had inadequate power to detect clinically significant differences for the following outcomes: intrauterine fetal death, pyelonephritis, preeclampsia, and perinatal mortality. There may also have been confounding factors not controlled for in our study, which included only univariate analysis. When antepartum complications were considered, SS patients were most commonly hospitalized for pain crises, infections, and anemia, requiring an average of 5 days of hospitalization. More than half of the SS patients required at least one blood transfusion, compared with one out of five SC patients who required at least one blood transfusion. Likewise, almost half of the SS patients had one or more documented pain crises during pregnancy, compared with one out of five SC patients who had one or more pain crises. These findings revisit the issue of improving pregnancy outcome with prophylactic blood
May 2001 Am J Obstet Gynecol
transfusions. Koshy et al6 showed that prophylactic transfusions did not change perinatal outcome but decreased the frequency of pain crises. Previous studies have also shown an increased risk for preeclampsia, pyelonephritis, intrauterine fetal death, and perinatal mortality.5, 7, 9 Our study found no significant change in these outcomes for SS or SC patients; however, our sample size had insufficient power to detect significant differences for these variables. Early pregnancy outcomes such as spontaneous or elective first-trimester abortions were excluded from this study, because the perinatal database does not include deliveries at <20 weeks’ estimated gestational age. Sickle–β-thalassemia patients were excluded from this study because of the small number of these patients available for study. Although maternal and perinatal mortality rates have decreased significantly in the last several decades, sickle cell disease in pregnancy remains associated with a higher risk of multiple maternal and fetal complications. These findings warrant close monitoring of all patients with sickle cell disease throughout pregnancy. Prenatal care should be available at an institution prepared to manage sickle cell disease complications and high-risk pregnancies. Despite the risks, most women with sickle hemoglobinopathies can expect good pregnancy outcomes. REFERENCES
1. Motulsky AG. Frequency of sickling disorders in U.S. blacks. N Engl J Med 1973;288:31-3. 2. Hendrickse JP, Watson-Williams EJ, Luzzatto L, Ajabor LN. Pregnancy in homozygous sickle cell anemia. J Obstet Gynaecol Br Commonw 1972;79:396-409. 3. Fort AT, Morrison JC, Berreras L, Diggs LW, Fish SA. Counseling the patient with sickle cell disease about reproduction. Pregnancy outcome does not justify the maternal risk. Am J Obstet Gynecol 1971;111:324-7. 4. Gabbe SG, Niebyl JR, Simpson JL. Obstetrics: normal and problem pregnancies. 3rd ed. Philadelphia: WB Saunders; 1996. 5. Powars DR, Sandhu M, Niland-Weiss J, Johnson C, Bruce S, Manning PR. Pregnancy in sickle cell disease. Obstet Gynecol 1986; 67:217-28. 6. Koshy M, Burd L, Wallace D, Moawad A, Baron J. Prophylactic red-cell transfusions in pregnant patients with sickle cell disease. N Engl J Med 1988;319:1447-52. 7. Smith JA, Espeland M, Bellevue R, Bonds D, Brown AK, Koshy M. Pregnancy in sickle cell disease: experience of the cooperative study of sickle cell disease. Obstet Gynecol 1996;87:199-204. 8. Koshy M, Ashenhurst J. Management of pregnancy in sickle cell anemia. Tex Rep Biol Med 1981;40:273-8. 9. Muhieddine AFS, Cantwell C, Nobles G, Levy DL. Outcome of pregnancies complicated by sickle cell and sickle-C hemoglobinopathies. Am J Perinatol 1994;11:187-91. 10. Brown AK, Sleeper LA, Pegelow CH, Miller ST, Gill FM, Waclawiw MA. The influence of infant and maternal sickle cell disease on birth outcome and neonatal course. Arch Pediatr Adolesc Med 1994;148:1156-62.