Exploring the best second conduit in coronary artery bypass grafting: a never-ending debate

We have read with interest the individual patient data meta-analysis by Gaudino et al. [1] exploring the best second conduit in coronary artery bypass grafting (CABG) surgery. It suggested that left internal mammary artery (LITA) plus radial artery (RA) is the CABG strategy associated with the best long-term survival and the lowest major adverse cardiac and cerebrovascular events rate. The conclusions of the authors about the better outcomes in patients receiving an RA graft appeared to us rather strongly worded. In particular, the ‘take home message’ in the visual abstract seems not fully supported by the evidence provided in the study. We believe that the selection criteria for the randomized clinical trials (RCTs) used in this meta-analysis suffer from an excessive arbitrariness. Of the 4 selected RCTs, the only 1 comparing bilateral internal thoracic artery (BITA) with LITA plus RA, the RAPCO trial [2], was excluded from the propensity score matching process because of missing variables. The populations of the 3 other RCTs, each of them comparing different outcomes and different CABG populations receiving different treatments [3–5], were pooled together for the analysis. We appreciate the authors’ efforts to develop a complex propensity score to reduce the selection biases inherent in this type of analysis, but as they recognize, there may be biases and residual imbalance that could influence the findings. Moreover, there are 2 issues that deserve further clarification. First is why, in the supplementary table, there are more BITA and LITA plus RA patients in the matched group than in the unmatched one and, second, whether the authors considered the patient groups from the ART trial, the largest contributor of BITA patients to the current study, according to the ‘intention-to-treat’ or ‘as-treated’ analysis.