Intake of polyunsaturated fatty acids and risk of preclinical and clinical type 1 diabetes in children—a systematic review and meta-analysis

The association between the intake of polyunsaturated fatty acids (PUFAs) and the risk of preclinical and clinical type 1 diabetes (T1D) in children has generated conflicting results. Thus, we aimed to evaluate the definite effects of PUFAs on the risk of preclinical and clinical T1D. Three databases were systematically searched up to July 18, 2017 to identify relevant observational studies, without language restriction. Any study included should report the risk of preclinical or clinical T1D in children with PUFAs supplementation compared with the controls, and report relative risks (RRs) or odds ratios (ORs) or provide data for estimation. Pooled RRs (or ORs) with 95% confidence intervals (CI) were calculated using random-effects models irrespective of statistical heterogeneity assessed by I2 statistic. We identified seven studies (three prospective cohort studies and four case-control studies) on PUFAs intake during pregnancy or during early life in children. The pooled RR between the risk of preclinical T1D and n-3 PUFAs supplementation against controls was 0.98 (95%CI, 0.85–1.13), with no heterogeneity. The results were similar after the intake during pregnancy, but not during early life in children (pooled RR, 0.45; 95%CI, 0.21–0.96; P = 0.039). N-3 PUFAs supplementation was not associated with a significant reduction in the risk of clinical T1D in children (pooled RR, 0.87; 95%CI, 0.71–1.08), with substantial heterogeneity(I2 = 64.7%). No association was also found between n-6 PUFAs intake and the risk of preclinical (1.07; 0.97–1.017) or clinical T1D (1.05; 0.92–1.20) in children. The result of the meta-analysis does not support that n-3 or n-6 PUFAs supplementation in children affects the overall risk of preclinical or clinical T1D. However, n-3 PUFAs intake in early life might reduce the risk of preclinical T1D. Therefore, this finding should be verified by more and well-designed prospective research in the future.