Preventive medication efficacy after 1-year follow-up for graft failure in coronary artery bypass surgery patients: Bayesian network meta-analysis

Abstract

To compare preventive medications against graft failures in coronary artery bypass graft surgery (CABG) patients after a 1-year follow-up. Systematic review with Bayesian network meta-analysis and meta-regression analysis. We searched PubMed, Scopus, and Web of Science databases in February 2023 for randomized controlled trials, comparing preventive medications against graft failure in CABG patients. We included studies that reported outcomes at 1 year after surgery. Our primary outcome was graft failure After screening 11,898 studies, a total of 18 randomized trials were included. Acetylsalicylic acid (ASA) [odds ratios (OR) 0.51, 95% credibility interval (CrI) 0.28-0.95, meta-regression OR 0.54, 95% CrI 0.26-1.00], Clopidogrel + ASA (OR 0.27, 95% CrI 0.09-0.76, meta-regression OR 0.28, 95% CrI 0.09-0.85), dipyridamole + ASA (OR 0.50, 95% CrI 0.30-0.83, meta-regression OR 0.49, 95% CrI 0.26-0.90), ticagrelor (OR 0.40, 95% CrI 0.16-1.00, meta-regression OR 0.43, 95% CrI 0.15-1.2), and ticagrelor + ASA (OR 0.26, 95% CrI 0.10-0.62, meta-regression OR 0.28, 95% CrI 0.10-0.68) were superior to placebo in preventing graft failure. Rank probabilities suggested the highest likelihood to be the most efficacious for ticagrelor + ASA [surface under the cumulative ranking (SUCRA) 0.859] and clopidogrel + ASA (SUCRA 0.819). The 95% CrIs of ORs for mortality, bleeding, and major adverse cardio- and cerebrovascular events (MACE) were wide. A trend towards increased bleeding risk and decreased MACE risk was observed when any of the medication regimens were used when compared to placebo. Sensitivity analysis excluding studies with a high risk of bias yielded equivalent results. Of the reviewed medication regimens, dual antiplatelet therapy combining ASA with ticagrelor or clopidogrel was found to result in the lowest rate of graft failures.

Keywords: Antiplatelet; Coronary artery bypass grafting; Dual antiplatelet therapy; Graft failure.

Figure 1

PRISMA flowchart of the study selection process.

Figure 2

Within-study risk of bias of the included studies.

Figure 3

The network plots between the medication regimens. (A) The network of graft failures per patient. (B) The network of graft failures per graft. The edge width corresponds with the number of studies on the given comparison.

Figure 4

Crude and adjusted odds ratios for graft failure per patient and per graft of each medication regimen with placebo set as a control treatment. 95% Crl, 95% credibility interval.

Figure 5

Rank probability distributions for each medication regimen. (A) Graft failures per patient. (B) Graft failures per graft. The leftmost bar signifies the first rank, i.e. the relative probability be the most efficacious medication regimen whereas the rightmost bar signifies the last rank, i.e. the relative probability to be the least efficacious regimen. The height of a bar shows the probability of the given rank. The surface under the cumulative ranking (SUCRA) score indicates the likelihood of a medication regimen to be the most efficacious with higher value indicating higher likelihood.

Figure 6

Sensitivity analysis of low risk of bias studies only for per-patient data.

Figure 7

Sensitivity analysis of low risk of bias studies only for per-graft data.

Figure 8

Odds ratios for complications of each medication regimen with placebo set as a control treatment. 95% Crl, 95% credibility interval.

Figure 9

Sensitivity analysis of complications with high risk of bias studies excluded. ASA is set as a control treatment. 95% Crl, 95% credibility interval.