In this observational prospective study, we assessed the role of clinical variables and circulating biomarkers in graft occlusion at 18 months to identify a signature for graft occlusion.Methods
A total of 330 patients undergoing primary elective coronary artery bypass grafting were enrolled. Blood collection for biomarker assessment was performed before surgery and discharge. Patients were then scheduled to undergo coronary computed tomography angiography at 18 months follow-up, and 179 patients underwent coronary computed tomography angiography 18 ± 2 months postoperatively.Results
There were 46 of 503 (9.1%) occluded grafts; of these, 29 (63%) were venous and 17 (37%) were arterial grafts; overall, 43 of 179 patients (24%) had at least 1 occluded graft. Logistic mixed effects model assessing independent factors associated with graft occlusion identified that lower D-dimer levels at baseline (odds ratio [OR], 2.58; 95% confidence interval [CI], 1.36-4.89; P = .00) and total protein content at discharge (OR, 1.09; 95% CI, 1.01-1.19; P = .028) were related to overall graft occlusion at follow-up, along with an arterial graft other than the left internal thoracic artery (OR, 2.92; 95% CI, 1.24-6.9; P = .078); moreover, a venous graft emerged was possibly associated with graft occlusion (OR, 1.51; 95% CI, 0.95-2.39; P = .078). By separately analyzing saphenous vein and arterial grafts, D-dimer levels (OR, 2.67; 95% CI, 1.15-6.2; P = .022 and OR, 2.5; 95% CI, 1.01-7.0; P = .05 for venous and arterial graft, respectively) were still associated with arterial and venous graft occlusion at follow-up.Conclusions
We identified D-dimer as a biomarker associated with arterial and venous grafts occlusion. This may help stratify patients at risk of graft failure and identify new molecular targets to prevent this complication.