The risk of venous thromboembolism (VTE) is increased in pregnancy and sickle cell disease (SCD), yet the rates of pregnancy-related VTE are not firmly established in SCD, nor are other SCD-related complications, including pneumonia, vasooclusive crisis (VOC), and acute chest syndrome (ACS), which may be clinically indistinguishable from VTE. Moreover, the American College of Chest Physicians has made no recommendations regarding thromboprophylaxis in pregnant women with SCD.
Inpatient hospital discharge data for PE for the most recent 5-year period available, 2007-2011, were obtained from the Pennsylvania Health Care Cost Containment Council (PHC4). We compared VTE, pregnancy complications, medical co-morbidities, and mortality among pregnant women with and without SCD. Among pregnant SCD women with and without VTE, we also compared rates of pneumonia, vasooclusive crisis (VOC), and acute chest syndrome (ACS). All patient identifiers were removed. Diagnostic categories and co-morbidities were obtained using ICD-9 codes. Data from categorical variables were analyzed by chi-square or Fisher's exact test; and from continuous variables by two-sample Student t-test.
The prevalence of VTE was 2.8% among pregnant women with SCD. Among pregnant women with SCD and VTE, the rate of pneumonia, 28.6% vs. 4.4%, p=0.043, was significantly higher than in those without VTE. While somewhat higher, the rates of VOC, 57.1% vs. 24.7%, p=0.073, and ACS, 14.3% vs. 2.4%, p=0.177, were not significantly different between pregnant SCD VTE and non-VTE groups. Comparing VTE and non-VTE pregnant SCD women, the rate of pregnancy complications did not differ, p=0.999; nor did the rates of medical co-morbidity, other than diabetes, 28.6% vs. 3.6%, p=0.031. Among the subset of pregnant SCD with pneumonia, the prevalence of VOC, 80.0% vs. 36.1%, p=0.001; ACS, 35.0% vs. 2.9%, p<0.001; and length of stay, 12.5 vs. 4.6 days, p=0.030, were significantly greater than in those without pneumonia. Among the subset of SCD pregnancies with VOC, the prevalence of preeclampsia, 28.4% vs. 13.5%, p=0.003; pneumonia, 15.7% vs. 2.6%, p=0.001; ACS, 12.8% vs. 0.6%, p<0.001; and length of stay, 7.7 vs. 3.6 days, p<0.001, were significantly more common than in those without VOC. Among the subset of SCD pregnancies with ACS, the rates of intrauterine fetal death, 14.3% vs. 1.6%, p=0.036; postpartum infection, 28.6% vs. 6.6%, p=0.016; pneumonia, 50.0% vs. 5.3%, p<0.001; and VOC, 92.9% vs. 36.5%, p<0.001, were significantly higher than in those without ACS. Overall, the rate of VTE, among SCD women with pneumonia, VOC, and/or ACS, 6.6%, was significantly higher than among those without these conditions, 2.2%, p<0.001.
The prevalence of pregnancy-related VTE in women with SCD, while low, 2.8%, appears to be at least 10-fold greater than the general 0.2% incidence of pregnancy-related VTE in the unaffected population. Further, the higher rates of VTE we observed among pregnant women with SCD-related complications, including pneumonia, VOC, or ACS, and the well-recognized potential clinical overlap with VTE, suggest that VTE may be missed in such women, and that VTE rates in pregnant women with SCD may be higher than herein reported or previously recognized. The presence of pneumonia, VOC, or ACS appears to be associated with VTE and may be a potential marker(s) of its occurrence. Prospective studies, however, are needed to determine the incidence of VTE in pregnant women with SCD with and without complicating pneumonia, VOC, and/or ACS. We conclude that pregnancy-related VTE may be more common than previously recognized in pregnant women with SCD and, if confirmed, such women might be candidates for thromboprophylaxis.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.