Performance of LITA-LAD Anastomosis by a Resident Yields Equivalent Outcomes to an Attending in the REGROUP Trial

A. Gikandi¹, E. Stock², S. Hirji³, F. Bakaeen⁴, J. Quin⁵, M. Haime⁶, B. Hattler⁷, K. Biswas⁸, M. Zenati^9
1Harvard Medical School, Brighton, Massachusetts ²West Roxbury VA Boston healthcare System, West Roxbury, Massachusetts ³Brigham and Women's Hospital, Allston, Massachusetts ⁴The Cleveland Clinic Foundation, Cleveland, Ohio ⁵VA Boston Healthcare System, West Roxbury, Massachusetts ⁶West Roxbury V A Boston healthcare System, West Roxbury, Massachusetts ⁷Denver Veterans Affairs Medical Center, Denver, Colorado ⁸Perry Point VA Medical Center, Perry Point, Maryland ⁹Harvard Medical School, Concord, Massachusetts

Purpose: Performance of a technically sound LITA-LAD anastomosis during CABG is critically important. Studies evaluating outcomes when a resident performs the LITA-LAD anastomosis with large scale, nationally representative, multi-surgeon, and randomized data are lacking. We investigated outcomes after CABG in the REGROUP trial according to which surgeon performed the LITA-LAD anastomosis.

Methods: This was a pre-planned subanalysis of the REGROUP trial (ClinicalTrials.gov number NCT01850082) which randomized patients undergoing isolated on-pump CABG to endoscopic versus open vein harvest at 16 VA medical centers between 2014 and 2017. Data on whether a resident or an attending performed the LITA-LAD anastomosis was prospectively collected at the time of surgery. The primary outcome was major cardiac adverse events (MACE), defined as a composite of all deaths, myocardial infarction, or repeat revascularization. SYNTAX score was calculated for all patients and stratified by terciles (low < 23, intermediate 23-32, high >32). There were 1084 patients total available for analysis (344 patients (31.8%) and 740 patients (68.2%) with a resident and attending as primary surgeon, respectively). The median follow-up was 4.7 years (interquartile range 3.84-5.45).

Results: Residents (when compared to attending surgeons, respectively) performed the LITA-LAD anastomosis in patients with lower STS-PROM scores (median 0.6 vs. 0.7, P = .01) and proportionally fewer patients with high SYNTAX scores (22.2% vs. 37.4%, P < .001). However, the median number of grafts performed by residents and attendings were similar (3.2 vs. 3.1, P = .063). Resident procedures were less likely to utilize transit time flow (24.1% vs. 39.9%, P < .001), and were also associated with similar cross-clamp times (median 72 vs. 70 min, P = .02). At 4.7 years of median follow-up, there were no significant differences in MACE (22.4% [77 events] for trainees and 22.8% [169 events] for attendings, P = .87). A Kaplan-Meier survival analysis comparing residents to attendings as primary surgeons showed hazard ratios (HR) of 0.95 (95% CI: 0.73 to 1.25) for MACE, 1.0 (95% CI: 0.73 to 1.49) for all-cause mortality, 1.0 (95% CI: 0.62 to 1.64) for myocardial infarction, and 0.99 (95% CI: 0.65 to 1.5) for repeat revascularization (Figure 1a). Importantly, there were no significant differences when MACE or its components were stratified by SYNTAX tercile (Figure 1b).

Conclusion: In this multi-center, nationally representative, multi-surgeon, randomized REGROUP trial subanalysis using prospectively collected data, residents performed LITA-LAD anastomosis with a clinical trend towards operating in patients with lower SYNTAX scores, but outcomes were not different based on SYNTAX scores. Our data suggests that the current training paradigm for cardiac surgery in the VA results in appropriate patient selection, supporting continuous performance of critical portions of CABG by residents.

Identify the source of the funding for this research project: None