Women with type 1 diabetes are at greater risk of adverse pregnancy outcomes and complications compared with the general population. Whether the number and outcome of pregnancies predict mortality risk, however, is not well studied. We thus assessed this hypothesis among women with childhood-onset type 1 diabetes. Female participants (n=325, mean age 28 and duration 19 years) of the Pittsburgh Epidemiology of Diabetes Complications (EDC) study, a prospective investigation of childhood onset type 1 diabetes, were selected for study. Pregnancy data were available for 320 women (98.5%). During 25 years of follow-up, 25.9% died (n=83). Excluding women (n=4) with twin pregnancies, no differences were observed in the proportion having been pregnant or in the number of pregnancies by mortality status. However, among those with at least one pregnancy (and adjusting for the total number of pregnancies), deceased were more likely to have had fewer live births and induced abortions but a greater number of spontaneous abortions and stillbirths (all p≤0.05). In multivariable Cox models adjusting for traditional baseline risk factors, having ever been pregnant was associated with a borderline inverse mortality risk (HR=0.65, p=0.07). In addition, among women with at least one pregnancy, adjusting (among other risk factors) for the total number of pregnancies, ≥ 2 spontaneous abortions (HR=3.5, p=0.03); ≥ 2 stillbirths (HR=9.1, p=0.02); and ≥ 2 occurrences of any adverse pregnancy outcome (i.e., tubal/ectopic pregnancies, spontaneous abortions, stillbirths; HR=3.9, p=0.007) were associated with an increased risk of death. Women whose pregnancy outcomes were all spontaneous abortions were at tenfold increased risk for mortality (p=0.004) whereas those with all preterm births were 76% less likely to die (HR=0.24, p=0.06).

In conclusion, adverse pregnancy outcomes among women with type 1 diabetes are associated with a greater mortality risk even after adjustment for traditional risk factors and diabetes control.


T. Costacou: None. D. Rubinstein: None. T.J. Orchard: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.