[Is diabetes mellitus a coronary heart disease equivalent? Results of a meta-analysis of prospective studies].

INTRODUCTION AND OBJECTIVES Several guidelines on the treatment of cardiovascular risk factors base their recommendations on the assertion that diabetes mellitus (DM) is a coronary heart disease (CHD) or cardiovascular disease (CVD) risk equivalent. To date, no systematic review of studies substantiating this assertion has been carried out. METHODS A systematic search of the PubMed database up to February 2006 was performed to identify prospective studies meeting the following criteria: a) follow-up was >5 years; b) groups of subjects with DM and without CHD (i.e., DM+CHD-), without DM and with CHD (DM-CHD+), and without either DM or CHD (DM-CHD-) were all included; and c) data on CHD or CVD mortality was reported. The characteristics of the studies were assessed, and data were combined separately for men and women using a random effects model and taking the DM-CHD- group as a reference. RESULTS In total, 13 studies met the inclusion criteria. Overall, CHD mortality was non-significantly lower in DM+CHD- men than in DM-CHD+ men, hazard ratio [HR] (95% confidence interval [CI]), 3.06 (2.45-3.83) vs 4.28 (3.24-5.66), respectively (P=.066); as was CVD mortality, HR (95% CI), 2.55 (2.00-3.26) vs 3.61 (2.81-4.62), respectively (P=.051). In women, there was no significant difference between the DM+CHD- and DM-CHD+ groups with regard to either CHD mortality, HR (95% CI), 4.68 (3.40-6.45) vs 3.51 (1.75-7.04), respectively (P=.42), or CVD mortality, HR (95% CI), 4.70 (4.23-5.22) vs 3.39 (1.51-9.02), respectively (P=.59). CONCLUSIONS The findings of this meta-analysis support the view that women in the DM+CHD- group have similar CHD and CVD mortality to those in the DM-CHD+ group, whereas men in the DM+CHD- group demonstrated a non-significant trend towards lower CHD and CVD mortality than those in the DM-CHD+ group.