The aim of the current study is to determine the impact of elevated lipoprotein(a) [Lp(a)] on cardiovascular events (CVEs) in stable coronary artery disease (CAD) patients with different glucose metabolism status. In this multicenter study, we consecutively enrolled 5,143 patients from March 2011 to February 2015. Patients were categorized according to status of glucose metabolism (diabetes mellitus [DM], pre-diabetes mellitus [pre-DM], and normal glucose regulation [NGR]) levels and further classified into 12 groups by Lp(a) levels. CVE end points included nonfatal acute myocardial infarction (MI), stroke, and cardiovascular mortality. All subjects were followed up for the occurrence of the CVEs. During a median of 6.1 years' follow-up, 435 (8.5%) CVEs occurred. No significant difference in occurrence of CVEs was observed between NGR and pre-DM groups (hazard ratio 1.131 [95% CI 0.822-1.556], P > 0.05). When status of glucose metabolism was incorporated in stratifying factors, 30 ≤ Lp(a) < 50 mg/dL and Lp(a) ≥50 mg/dL were associated with significantly higher risk of subsequent CVEs in pre-DM (2.181 [1.099-4.327] and 2.668 [1.383-5.415], respectively; all P < 0.05) and DM (3.088 [1.535-5.895] and 3.470 [1.801-6.686], all P < 0.05). Moreover, adding Lp(a) to the Cox model increased the C-statistic by 0.022 and 0.029 in pre-DM and DM, respectively, while the C-statistic was not statistically improved when Lp(a) was included for CVEs prediction in NGR. Our findings, for the first time, indicated that elevated Lp(a) levels might affect the prognosis in patients with pre-DM with stable CAD, suggesting that Lp(a) may help further stratify stable CAD patients with mild impaired glucose metabolism. © 2019 by the American Diabetes Association.