RESUMO
OBJECTIVE: Women with polycystic ovary syndrome (PCOS) are associated with increased levels of insulin resistance (IR). Other than treatment with insulin-sensitizing drugs, specialized diets have also been implemented to reduce the patient's IR. However, the capacity of certain diets, concerning with the severity of the patient's IR, to improve insulin sensitivity has not fully been explored. Therefore, we conducted a meta-analysis to determine in PCOS subjects from low to severe IR, if hypocaloric diets improve insulin sensitivity. STUDY DESIGN: PubMed, SCOPUS, EBSCO, and LILACS databases and retrieved studies' bibliographies were searched for prospective studies that investigated the association between diets and IR in PCOS women until October 2018. Diet was defined as a modification of the patients' nutrition intake according to caloric restriction, change in protein intake, or by using a specialized diet. IR measures (HOMA1-IR), pre- and post-intervention were extracted. Using Comprehensive Meta-Analysis software, depending on the level heterogeneity, determined by the ψ2-based Q-test and the I2-test, fixed-effects or random-effects models were used to calculate the pooled standard paired differences (SPD) and 95 %CI. RESULTS: 20 publications (25 studies) fulfilled the inclusion criteria. Due to the heterogeneity of the diets, the random-effects model was used. In 48 % of the studies, the diets led to a decrease of IR, where 44 % had no effect. In 2 studies, the diets increased IR. Overall, the diets decreased IR (SPD=-0.58; 95 %CI: -0.81 to -0.36). Subjects with severe IR (HOMA1-IR>4.2) had a marked improvement (SPD=-1.22; 95 %CI: -1.61 to -0.84). Moreover, diets low in carbohydrate (<50 %) was also determined to improve IR (SPD=-0.86; 95 %CI: -1.23 to -0.50). CONCLUSIONS: Here, we demonstrate that diets are more likely to improve IR in PCOS women with severe IR. Therefore, it is crucial to determine a subject's IR status before considering any intervention containing a diet.