Fertility Preservation Outcomes for People with Ovaries and Sickle Cell Anemia: A Multi-Center Study

Background: People with sickle cell anemia (SCA) may undergo fertility preservation (FP) due to risks of diminished ovarian reserve (DOR) and curative therapy regimens. Ovarian tissue, oocytes or embryos may be cryopreserved; pubescent individuals with ovaries are offered controlled ovarian hyperstimulation (COH) with oocyte cryopreservation (OC). Little data addresses FP outcomes in SCA. FP benefits are balanced against procedural risks, potential low oocyte yield with DOR, and access barriers.

The purpose of this study is to describe (1) clinical factors associated with oocyte recovery and (2) COH complications in SCA.

Methods: This five-center retrospective study captured embryo and OC outcomes from January 2016 - May 2024. Individuals included had sickle cell disease (SCD) and an OC attempt. All subjects received expert hematology and reproductive endocrinology care.

We reviewed electronic medical records for OC indications, pre-OC laboratory values, SCD complications, and disease modifying therapy. OC outcomes included baseline anti-Mullerian hormone (AMH) level, transfusion before OC, peak estradiol (E2) level during OC (a measure of ovarian response), total gonadotropin use (medication to stimulate follicle growth) during OC, and oocyte maturity rate (mature oocytes used for future embryos; mature oocyte(s) frozen/total retrieved). OC complications captured included SCD-related complications, pre- or post-OC pain, infection, unplanned transfusion, and fluid overload associated with ovarian hyperstimulation syndrome [OHSS]).

Median and interquartile range (IQR) describe continuous variables. We categorized SCD characteristics and pre-OC descriptors as binary variables to analyze by logistic regression for association with occurrence of >1 complication from beginning of COH until two weeks after OC.

Results: There were 46 subjects, all had SCA (93% HbSS, 7% HbSβ0-thalassemia), median age 23.7 (IQR 18, 28). Most (44/46) underwent FP before curative therapy; other reasons were DOR (1/46) or in vitro fertilization with pre-implantation genetic testing (1/46). Insurance coverage varied: private (n=15), public (n=9), self-pay (n=6), or research protocol (n=15).

In the year before OC, the median number of acute care visits for pain was 4 (IQR 2, 7). All subjects had at least one SCD complication: cerebrovascular disease (16/46), acute chest syndrome (23/46), >2 vasoocclusive episodes (VOE) per year (31/46), thromboembolism (4/46), or pulmonary hypertension (1/46). At COH, most subjects received chronic transfusion therapy (25/46), hydroxyurea (HU) (37/46), or both (19/46).

There were 55 COH cycles. Procedures were performed in two settings: ambulatory (33/55) or operating room (18/55). Most (n=36) subjects underwent one COH cycle. Three cycles were canceled due to poor ovarian response. Six subjects required multiple cycles, either for low oocyte yield (n=5) or cycle cancelation (n=1).

Median AMH (2.1 ng/mL) was lower than the expected AMH for our subjects' age range (norm 2.8 – 3.4 ng/mL). Median gonadotropins were 4088 IU (IQR 2700, 4912) and peak E2 levels were 1108 pg/mL (IQR 644, 2617). Median oocytes retrieved per cycle was 14 (IQR 7, 24) and mature oocytes cryopreserved was 11 (IQR 4.5, 15.5); maturity rate was 72%.

Complications occurred in 25/55 cycles; 11 subjects had >1 complication. These included unplanned transfusion (pre-OC n=6, post-OC n=4), mild-moderate OHSS (n=2), bacteremia needing intensive care (n=1), mid-COH VOE (n=8), and post-retrieval VOE (n=21). Most (19/21) post-retrieval VOE were managed with emergency department care or hospitalization. Any OC complication was associated with > 3 VOE in the year before COH (mean of 3 VOE per subject without complication vs. 6 per subject with complication; p = 0.036). No center differences were detected.

Conclusion: Among 55 COH cycles at 5 large academic centers, 50% of subjects used FP services through public insurance or a research protocol, affirming access enables FP use in SCA. Here, 13% required >1 cycle due to cancellation or low oocyte yield. Complications were common and >3 VOE in the year before COH predicted pain following COH, underscoring the need for individualized pain plans before COH. We did not capture steroid use in anesthesia, which may impact pain outcomes. The study is limited by future embryos and live birth data. These results may inform counseling, care, and advocacy efforts.