Prepregnancy fasting insulin as an early indicator of the risk of spontaneous preterm birth: A retrospective cohort study

Background: Identifying modifiable risk factors for spontaneous preterm birth (sPTB) is crucial for developing effective interventions. While maternal overt diabetes is known to increase the risk of sPTB, the impact of impaired insulin sensitivity and insulin response\textemdash early stages of glucose dysmetabolism affecting a larger proportion of reproductive-age individuals\textemdash remains underexplored. This study aimed to identify maternal prepregnancy insulin sensitivity and response indices with prognostic value for the risk of dysmetabolic-associated sPTB. Methods: This retrospective study included 1,898 non-diabetic individuals who underwent an oral glucose tolerance test (OGTT) before pregnancy and had recorded delivery outcomes. Adjusted odds ratios (aORs) for quintiles of prepregnancy OGTT-derived insulin sensitivity and response indices in relation to sPTB before 37 weeks were estimated using multivariable logistic regression models, with the middle quintile as the reference group. The index of interest was selected based on its accessibility and prognostic significance. Models incorporating restricted cubic splines, a statistical method for modeling nonlinear associations, were developed to assess the relationship between the index of interest as a continuous variable and sPTB risk. The relationships between the identified index and individual characteristics were analyzed to explore possible mechanisms. Results: Most participants (85.2%) exhibited normal prepregnancy glucose tolerance. Prepregnancy fasting insulin was selected as the index of interest due to its accessibility and comparable aORs for sPTB risk relative to other OGTT-derived insulin sensitivity and response indices. The highest and lowest quintiles of prepregnancy fasting insulin had sPTB rates of 11.8% and 7.9%, respectively, approximately twice the rate observed in the middle quintile (4.2%). The aORs were 2.81 [95% confidence interval (CI): 1.47\textendash5.65] and 2.36 (95% CI: 1.21\textendash4.78). A significant nonlinear relationship was identified between prepregnancy fasting insulin as a continuous variable and sPTB risk ($P$ for nonlinearity = 0.018). Additionally, prepregnancy fasting insulin correlated significantly with prepregnancy body mass index (BMI) and glucose levels. Conclusions: Prepregnancy fasting insulin is associated with an increased risk of sPTB at both high and low levels in individuals predominantly exhibiting normal glucose tolerance.