Abstract
BACKGROUND: The aim of this study was to comprehensively analyze the bidirectional association between cardiovascular disease (CVD) and hip fracture (HF). METHODS: We searched PubMed, EMBASE, Web of Sciences, Cochrane Library, ScienceDirect and China National Knowledge Infrastructure for relevant studies. The Newcastle-Ottawa scale was used to evaluate the risk of bias. We conducted random effects model for meta-analysis and subgroup analysis of different ethnic groups. Sensitivity analysis and publication bias of this study were also evaluated. This study followed the PRISMA and MOOSE guidelines for systematic reviews and meta-analyses. RESULTS: This research included 18 cohort studies and case-control studies with a total sample of 1,854,441 individuals. The results showed ischemic heart disease might increase the risk of HF (OR = 1.41, 95%CI[1.05, 1.89], I(2) = 96%). Stroke might be a risk factor for HF (OR = 2.23, 95%[1.18, 4.19], I(2) = 97%), and HF might likewise be a risk factor for Stroke ( OR = 2.22, 95% CI [1.81, 2.71], I(2) = 78%). Heart failure might increase the risk of HF (OR = 2.89, 95%CI [1.22, 6.85], I(2) = 91%), and HF might increase the risk of heart failure (OR = 2.74, 95%CI [1.27, 5.89], I(2) = 92%). Hypertension might increase the risk of HF (OR = 1.55, 95%CI[1.34, 1.8], I(2) = 87%), and HF might increase the risk of hypertension (OR = 3.75, 95%CI[3.3, 4.26], I(2) = 98%). Cerebrovascular disease (OR = 1.96, 95%CI[1.61, 2.4], I(2) = 79%) and diseases of arteries, arterioles, and capillaries (OR = 1.58, 95%CI[1.49, 1.68], I(2) = 0%) might increase the risk of HF. HF might increase the risk of myocardial infarction (OR = 2, 95%CI[1.17, 3.41], I(2) = 97%) and CVD-related death (OR = 1.78, 95%CI[1.05, 3.02], I(2) = 50%). Subgroup analyses showed that among Asians IHD might not raise the risk of HF (OR = 1.33, 95% CI [1.00, 1.78], I(2) = 95%). In caucasians, IHD might also not raise HF risk (OR = 1.52, 95%CI [0.64, 4.56], I(2) = 95%). CONCLUSIONS: This study supports possible bidirectional associations between CVD and HF, but more mechanistic studies of CVD and HF were warranted. However, high heterogeneity and potential confounding by unmeasured variables warrant cautious interpretation.