Abstract Background Early glucose lowering intervention in subjects with type 2 diabetes mellitus was demonstrated to be beneficial in terms of micro- and macrovascular risk reduction. However, most of currently ongoing cardiovascular outcome trials are performed in subjects with manifest atherosclerosis and long-standing diabetes. Therefore, the aim of this study is to investigate the effects of the dipeptidylpeptidase-4 inhibitor linagliptin in subjects with coronary artery disease (CAD) but early type 2 diabetes mellitus (T2DM) on a set of cardiovascular surrogate measurements. Methods In this randomized, placebo-controlled, double-blind, single-center study, we included subjects with early diabetes (postchallenge diabetes (2 h glucose > 200 mg/dl) or T2DM treated with diet only or on a stable dose of metformin monotherapy and an HbA1c < 75 mmol/mol) and established CAD. Participants were randomized to receive either linagliptin (5 mg) once daily orally or placebo for 12 weeks. The primary outcome was the change in flow mediated dilatation (FMD). The secondary objective was to investigate the effect of linagliptin treatment on arginine bioavailability ratios [Global arginine bioavailability ratio (GABR) and arginine to ornithine ratio (AOR)]. Arginine, ornithine and citrulline were measured in serum samples with a conventional usual amino acid analysis technique, involving separation of amino acids by ion exchange chromatography followed by postcolumn continuous reaction with ninhydrin. GABR was calculated by l-arginine divided by the sum of (l-ornithine plus l-citrulline). The AOR was calculated by dividing l-arginine by l-ornithine levels. Group comparisons were calculated by using a two-sample t-test with Satterthwaite adjustment for unequal variances. Results We investigated 43 patients (21% female) with a mean age of 63.3 ± 8.2 years. FMD at baseline was 3.5 ± 3.1% in the linagliptin group vs. 4.0 ± 2.9% in the placebo group. The change in mean FMD in the linagliptin group was not significantly different compared to the change in the placebo group (0.43 ± 4.84% vs. − 0.45 ± 3.01%; p = 0.486). No significant improvements were seen in the arginine bioavailability ratios (GABR; p = 0.608 and AOR; p = 0.549). Conclusion Linagliptin treatment in subjects with CAD and early T2DM did not improve endothelial function or the arginine bioavailability ratios. Trial registration ClinicalTrials.gov, NCT02350478 (https://clinicaltrials.gov/ct2/show/NCT02350478)