Abortions Outcomes of People with Sickle Cell Disease

Introduction

People with sickle cell disease (SCD) are at risk for SCD-associated complications with routine procedures. Whether people with SCD seeking abortions require special consideration is unknown. At our center, 28% of surveyed women with SCD reported having had an abortion, and 32% believed abortion to be unsafe for people with SCD. This cohort also reported low rates of long-acting contraception use. Information about abortion outcomes among people with SCD is important to inform care and minimize unnecessary delays, particularly as people in the U.S. face barriers to abortion access. The purpose of this study is to describe a single-center experience with abortions for people with SCD.

Methods

This single-center JHU IRB-approved study included people with SCD ages 18–55 years who had a medication or procedural abortion between April 1, 2013, and December 31, 2022. Systematic data collected from electronic health records included patient demographics, pregnancy information (e.g., gravidity), SCD characteristics (e.g., genotype), healthcare utilization, and abortion information, including the type of abortion, anesthesia and steroid use, pre- and post-abortion transfusion use, and the number and type of post-abortion complications. These complications included SCD complications (e.g., pain crises and acute chest syndrome) and non-SCD complications (e.g., infections and hemorrhage). SAS was used to generate descriptive statistics.

Results

There were 19 subjects with 27 abortions between 2013 and 2022. The median age at the time of abortion care was 24 years (IQR: 20-29). About half of the subjects with known genotypes (10/18, 56%) had HbSS/HbSβ0 thalassemia and accounted for an equal proportion of abortions (15/27, 56%). The nearest median hemoglobin level to abortion was 8.9 g/dL (IQR: 8.1-10.6). In 15/27 (56%) abortions, the subject was on disease-modifying therapy (hydroxyurea or chronic transfusions) before pregnancy, and two subjects stopped taking hydroxyurea during pregnancy. Six subjects who had 11/27 (41%) abortions in the cohort had ≥ five hospitalizations for pain crises in the year before abortion.

Most abortions (24/27, 96%) were managed with a <30-minute dilation and curettage or dilation and evacuation. The median gestational age at the time of abortion care was 10 weeks (IQR: 8-14). Among procedural abortions (N=24), seven were treated with steroids. The rates of 7-day and 30-day complications for procedural abortions were 6/24 (26%) and 10/24 (42%), respectively, of which a majority were SCD pain crises (Table 1). Baseline pain crisis frequency was six times higher in subjects with post-abortion pain crises compared to those without. There was no difference in pain crisis after abortion between those with steroid use (4/7, 56%) compared to those without steroid use (8/16, 50%).

Conclusions

In this cohort, most abortions were procedural and required brief anesthesia. This does not routinely require prophylactic transfusion. Steroids, a risk factor for pain crises, were commonly used in procedural abortions. The high rate of baseline pain crises in this cohort confounded the association between abortions and post-abortion pain crises. People with HbSS/HbSβ0 thalassemia were underrepresented in our cohort when compared to the general SCD population. This may reflect differences in pregnancy rates or abortion decision-making within this group. Limitations include that this study is single-center and did not capture abortions performed outside our healthcare system. Larger studies are needed to define management and inform abortion care for people with SCD.