Advertisement
Poster session II Clinical obstetrics, diabetes, labor, medical-surgical-disease, physiology/endocrinology, prematurity: Abstracts 237 - 386| Volume 208, ISSUE 1, SUPPLEMENT , S124, January 01, 2013

272: Pregnancy outcomes among women with type 1 diabetes managed with continuous insulin pumps versus multiple injections: a propensity-score based analysis

      Objective

      Compare pregnancy outcomes among women with type 1 diabetes managed with a continuous subcutaneous insulin pump (pump) to women managed with multiple insulin injections (injections).

      Study Design

      This was a retrospective cohort study of pregnancies among women with type 1 diabetes who delivered at UCSF since 2006. Pregnancies prior to 2006 among women who also delivered after 2006 were included. Outcomes in women who were managed with a pump were compared to women managed with injections using a propensity-score based approach. The propensity to be on a pump was based on age, age at diagnosis, parity, distance from UCSF (> 50 mi or < 50 mi) and median income of the patient's county of residence. Primary outcomes were a composite maternal outcome (vaginal delivery, no antepartum admissions or preeclampsia and peripartum hospitalization ≤ 5 days) and neonatal outcome (delivery ≥ 36 weeks, no hypoglycemia or hyperbilirubinemia, neonatal hospitalization ≤ maternal stay and care in the well-baby nursery).

      Results

      There were 48 pregnancies included, 35 among women on a pump, 13 among women using injections. Women on pumps were more likely to be older, live closer to UCSF and have a higher median income in their home county. Optimal maternal and neonatal outcomes occurred among 22.7% and 25.7% of women on a pump versus 7.7% of women on injections (p = 0.2). When subjects were propensity-score matched, however, women on pumps were more likely to have both optimal maternal and neonatal outcomes (p < 0.01).

      Conclusion

      Using propensity-scores for matching, women on insulin pumps appear to have better pregnancy outcomes than women on multiple daily injections. Additional investigation is needed to determine if this association is a function of improved glycemic control or other factors. Although geographic and socioeconomic factors may be important barriers, consideration of initiation of an insulin pump before or even during pregnancy may be warranted.
      Figure thumbnail grr33
      † Propensity matching based on maternal age, age at diagnosis, parity, distance from UCSF (> 50 miles or < 50 miles) and median income of the patient's county of residence.