New diagnosis of diabetes in patients with myocardial infarction or stroke: a systematic review and meta-analysis

Abstract

Aims What proportion of patients with an acute myocardial infarction (MI) or stroke get a new diagnosis of diabetes (NDD) at the time of hospitalization is unclear. Methods and results We systematically searched MEDLINE and Embase from database inception to 30 January 2025, to select English-language observational studies that included adult patients with MI or stroke and reported the number of patients with NDD. The denominator was patients without known diabetes. Study quality was assessed using the Joanna Briggs Institute checklist. Random-effects meta-analyses were used to calculate the pooled proportion of NDD in patients with MI or stroke. Heterogeneity was explored in subgroup analyses, and any change over time was evaluated using meta-regression. 82 studies that included 9440 patients with NDD were identified. Included studies were of good methodological quality: 17 (22%) tested all eligible patients for diabetes, and 19 (23%) used registry-based samples. 16.0% [95% confidence interval (CI), 14.3–17.7, n = 52] patients with MI had NDD, and 15.3% (95% CI 12.5–18.0, n = 30) patients with stroke had NDD, albeit with high heterogeneity. The pooled proportion was higher when oral glucose tolerance test was used to diagnose diabetes. The proportion of NDD did not change over the last 30 years (0.1% per year decline; 95% CI −0.3% to 0.1%). Conclusion Among patients admitted with MI or stroke, one in six gets a new diagnosis of diabetes. Improving diabetes screening could help identify people before having a diabetes-related cardiovascular event.