Fasting but not changes of plasma metabolome during oral glucose tolerance tests improves the diagnosis of severe coronary arterial stenosis

Background Noninvasive stress tests for the diagnosis of significant coronary arterial stenosis requiring intervention are not perfect. We investigated whether plasma metabolome during the oral glucose tolerance test (OGTT) can improve the diagnosis. Methods A total of 117 subjects with positive stress test results who received coronary angiography were recruited. After excluding subjects with a history of myocardial infarction and subjects who did not receive OGTT, the 18 subjects without significant stenosis were selected as controls. Another 18 age- and sex-matched subjects with significant stenosis were selected as cases. Plasma metabolome from samples obtained in fasting, 30 and 120 min after OGTT was measured using liquid chromatography combined with time-of-flight mass spectrometry. Results We found five metabolites which can identify patients with significant stenosis independent to clinical risk factors, including diabetes, hypertension, hypercholesterolaemia, smoking and history of percutaneous coronary intervention (all P < 0·05). The area under the receiver operating characteristic (AUROC) curve of these metabolites was 0·799-0·818 at fasting and 30 min after OGTT. The addition of metabolites to clinical factors increases the AUROC (0·616, 95%CI 0·429-0·803 for model with clinical factors only; 0·824, 95%CI 0·689-0·959 for model with four metabolites and clinical factors). The changes of plasma metabolite levels during OGTT did not significantly improve the diagnostic performance. Conclusions Fasting plasma metabolome, but not change of plasma metabolome during OGTT, can improve the diagnosis of significant stenosis in patients with positive noninvasive stress test results.