Predictive value of coronary computed tomography angiography in asymptomatic individuals with diabetes mellitus: Systematic review and meta-analysis

Background: Coronary CT angiography (CTA) is generally not established as a screening tool for asymptomatic individuals. However, it is controversial whether this test may have a role for screening asymptomatic individuals with diabetes mellitus (DM) due to the high prevalence of asymptomatic coronary artery disease (CAD) in this subgroup. Methods: We searched PubMed and EMBASE through May 2017 for studies that reported on the association between findings at coronary CTA and future cardiac events in asymptomatic individuals with DM. Summary hazard ratios for the presence of obstructive CAD (>= 50% stenosis), presence of non-obstructive plaque (< 50% stenosis), segment involvement score, and segment stenosis score were derived using a random effects regression model. I-2 was calculated to quantify between-study heterogeneity and causing factors were identified using meta-regression. Results: A total of 10 studies reporting on 5012 individuals with DM (median age: 62.3 years, median proportion of women: 40.5%) were included in the analysis. The presence of obstructive CAD on coronary CTA (vs. non-obstructive or no CAD) was associated with a significantly elevated risk for adverse events (summary HR: 4.07, 95% CI: 2.30 to 7.21). The estimated summary HR for non-obstructive plaque (vs. no CAD) was 2.17 (95% CI: 1.11 to 4.25). The pooled HRs per unit for segment stenosis score and segment involvement score were 1.44 (95% CI: 0.98 to 2.12), and 1.73 (95% CI: 1.07 to 2.80) respectively. On meta-regression analysis, we observed a trend towards a higher risk estimate in studies with a higher proportion of females (p=0.1063). Conclusion: The presence and extent of CAD on coronary CTA are strong, independent predictors of cardiovascular events in asymptomatic individuals with DM despite heterogeneity between studies in endpoints, study population and length of follow-up.