Effects of different insulin sensitisers in the management of polycystic ovary syndrome: A systematic review and meta-analysis.

OBJECTIVE: Characteristic features of polycystic ovary syndrome (PCOS) include insulin resistance and an increased risk for type 2 diabetes. To promote improved insulin sensitivity, insulin sensitisers have been used in PCOS. However, direct comparisons across these agents are limited. This study compared the effects of metformin, rosiglitazone and pioglitazone in the management of PCOS to inform the 2023 International Evidence-based PCOS Guideline. DESIGN: Systematic review and meta-analysis of the literature. PATIENTS: Women with PCOS and treatment with insulin sensitisers. MEASUREMENTS: Hormonal and clinical outcomes, as well as side effects. RESULTS: Of 1660 publications identified, 13 randomised controlled trials were included. Metformin was superior in lowering weight (mean difference [MD]: -4.39, 95% confidence interval [CI]: -7.69 to -1.08 kg), body mass index (MD: -0.95, 95% CI: -1.41 to -0.49 kg/m2 ) and testosterone (MD: -0.10, 95% CI: -0.18 to -0.03 nmol/L) versus rosiglitazone, whereas there was no difference when comparing metformin to pioglitazone. Adding rosiglitazone or pioglitazone to metformin did not improve metabolic outcomes. However, rosiglitazone seemed superior to metformin in lowering lipid concentrations. CONCLUSIONS: Metformin should remain the first-line insulin sensitising treatment in adults with PCOS for the prevention and management of weight and metabolic features. The addition of thiazolidinediones appears to offer little benefit.

Risk of induced abortions in childhood cancer survivors.

BACKGROUND: The relative probability of pregnancy and parenthood in cancer survivors is reduced. Studies have shown that cancer survivors are concerned about the health of their offspring and the recurrence of their own cancer. This could lead to an increased risk of induced abortion. The aim of this study was to examine whether pregnancies of childhood cancer survivors (CCSs) who were 0 to 14 years old at diagnosis in 1971-2012 were more likely to result in induced abortions in comparison with population controls. METHODS: Data from Finnish registries for cancer, births, and induced abortions were merged to identify 420 first pregnancies of CCSs and 2508 first pregnancies of age-matched population controls in 1987-2013. Poisson regression and logistic regression modeling were used to estimate incidence rates and relative risks (RRs) with 95% confidence intervals (CIs) of first pregnancies and induced abortions in CCSs in comparison with population controls. RESULTS: The risk of first pregnancy was reduced in CCSs in comparison with population controls (RR, 0.72; 95% CI, 0.64-0.80), whereas the risk of a first pregnancy resulting in an induced abortion was similar in CCSs and population controls (RR, 1.01; 95% CI, 0.77-1.33). In subanalyses stratifying by decade of diagnosis and cancer treatment, the risk of induced abortion was similar in CCSs and population controls. CONCLUSIONS: Female CCSs do not have an overall increased risk of induced abortions. The reduced probability of pregnancy among CCSs highlights the continued need for interventions to preserve fertility at the time of a cancer diagnosis.