Abstract
AIMS/HYPOTHESIS: Previous meta-analyses have suggested that diabetes confers a greater excess risk of coronary heart disease, stroke, vascular dementia, and heart failure in women compared to men. While the underlying mechanism that explains such greater excess risk is unknown, in the current meta-analysis we hypothesized that we would find a similar sex difference in the relationship between diabetes and peripheral arterial disease (PAD). METHODS: PubMed MEDLINE, the Cochrane Database of Systematic Reviews, and Embase were systematically searched for prospective population-based cohort studies, with no restriction on publication date, language, or country. We included studies that reported the relative risk (RR), and its variability, for incident PAD associated with diabetes in both sexes. We excluded studies that did not adjust at least for age, and in which participants had pre-existing PAD. In cases where sex-specific results were not reported, study authors were contacted. Random-effects meta-analyses with inverse variance weighting were used to obtain summary sex-specific RRs and the women: men ratio of RRs for PAD. The Newcastle-Ottawa scale was used to assess study quality. RESULTS: Data from seven cohorts, totalling 2071,260 participants (49.8% women), were included. The relative risk for incident PAD associated with diabetes compared with no diabetes was 1.96 (95% CI 1.29-2.63) in women and 1.84 (95% CI 1.29-2.86) in men, after adjusting for potential confounders. The multiple-adjusted RR ratio was 1.05 (95% CI 0.90-1.22), with virtually no heterogeneity between studies (I(2) = 0%). All studies scored 6-8, on the Newcastle-Ottawa scale of 0-9, indicating good quality. Eleven of the 12 studies that met review inclusion criteria did not report sex-specific relative risk, and these data were collected through direct correspondence with the study authors. CONCLUSION/INTERPRETATION: Consistent with other studies, we found evidence that diabetes is an independent risk factor for PAD. However, in contrast to similar studies of other types of cardiovascular disease, we did not find evidence that diabetes confers a greater excess risk in women compared to men for PAD. More research is needed to explain this sex differential between PAD and other forms of CVD, in the sequelae of diabetes. In addition, we found that very few studies reported the sex-specific relative risk for the association between diabetes and PAD, adding to existing evidence for the need for improved reporting of sex-disaggregated results in cardiovascular disease research.