Menstrual cycle abnormalities as distinctive sign of type 1 diabetes mellitus: results from a meta-analysis

Background: Type 1 diabetes mellitus (T1DM) management profoundly changed across years, with increasing emphasis on stringent glycaemic control. While it is well demonstrated that the progressive improvement of glycaemic control allows a tighter command of diabetes-related complications, the positive implications thereof on reproductive functions are still unclear. Indeed, it is well known that oligomenorrhea and amenorrhea are more frequently detected in young women with T1DM compared to healthy age-matched controls. However, whether the menstrual abnormalities incidence changed across years is still matter of debate. Aim of the study: To evaluate the menstrual cycle abnormalities rate in T1DM young women, compared to healthy subjects, and to search for potential T1DM-related factors influencing female reproductive system. Secondary aim was the evaluation of the possible effects of the change in T1DM management, occurred in the late 90’s, on menstrual cycle dysfunction. Methods: A meta-analysis was performed considering all clinical trials in which menstrual cycle abnormalities in T1DM young women were reported, compared to healthy age-matched subjects. Primary endpoint was the rate of oligomenorrhea/amenorrhea and secondary objective was age at menarche. Sensitivity analysis was conducted dividing studies into two groups, i.e. before and after 2000, according to the change in T1DM management. Three meta-regression analyses were performed, considering the influence of diabetes duration, body mass index (BMI) and glycated haemoglobin (HbA1c) serum levels on menstrual irregularities. Results: From 623 papers initially identified, 12 studies were finally included. Menstrual cycle dysfunction rate was significantly higher in T1DM women compared to controls, also considering only studies published after 2000 (OR:2.08; 95%CI: 1.43,3.03, P<0.001). Age at menarche was significantly higher in T1DM women compared to controls (P<0.001) also when studies published after 2000 were evaluated separately (mean difference:0.53; 95%CI: 0.32,0.74 years, P<0.001). In meta-regression analyses, the menstrual abnormalities rate in T1DM women were inversely related to diabetes duration (R2 =0.396,P=0.023), but not to BMI (R2 =0.134,P=0.373) and HbA1c serum levels (R2 =0.083,P=0.409). Conclusion: The meta-analytic approach confirmed the high incidence of menstrual cycle dysfunction in T1DM young women. The improvement in T1DM management, introduced after 2000, seems not able to influence this rate, leaving menstrual cycle abnormalities one of the distinctive signs of this chronic condition. Indeed, T1DM-related menstrual dysfunction is associated neither to anthropometrical variables, nor glycemic control. Although actual pathogenetic mechanisms are not fully understood, here we demonstrate a potential association with T1DM duration, suggesting that the process of disease acceptance could underlie these irregularities.