Impact of Resident-Led Operating on Outcomes in Adult Cardiac Surgery: A 10-Year Propensity-Matched Analysis.

OBJECTIVES: Cardiac surgical training in today's day and age faces significant challenges, including declining operative volumes, increasing case complexity, transcatheter interventions, and shorter training programmes, amongst others. Although it is important to train tomorrow's cardiac surgeons, it is important to ensure that high-quality surgical training does not compromise patient safety. This study assessed the impact of resident-led operating on outcomes in adult cardiac surgical procedures defined by the Joint Committee on Surgical Training in the UK cardiac surgical training curriculum. METHODS: Data for all adult cardiac surgeries performed between 2015 and 2024 were retrieved from our institutional database and the UK National Mortality Database was queried for the relevant patients. All index procedures (isolated CABG, isolated AVR, and combined CABG + AVR) were identified. 1:1 propensity-score matching of resident-led and consultant-led cases was performed. In-hospital outcomes (mortality, established postoperative complications, and the duration of postoperative hospitalisation) as well as post-discharge outcomes (long-term) survival were compared. RESULTS: A total of 11,372 index procedures were undertaken at our institution during the study period. Propensity-score matching yielded 4,259 matched pairs. Groups R (resident-led cases) and C (consultant-led cases) had similar demographics, preoperative cardiac function, functional status, medical history, and operative risk scores. Consultants performed more combined procedures (18.2% vs 14.3%, P < .001). Cardiopulmonary bypass and aortic cross-clamp times were longer in Group R than in Group C (94 vs 89 min and 60 vs 56 min, both P < .001). In-hospital mortality was similar between groups. Higher deep sternal wound infection rates (1.2% [95% CI, 0.9-1.5] vs 0.7% [0.5-1.0], P = .033) and longer hospitalisations (7 [IQR 6-10] vs 6d [5-9], P < .001) were seen in Group R. There was no statistically significant difference in long-term survival between groups (Group C HR 0.97 [95% CI, 0.88-1.07], P = .564). CONCLUSIONS: Supervised resident-led operating is safe in well-selected cases. Comparable mortality and morbidity suggest that well-supervised training does not compromise outcomes. These results support structured progressive autonomy in cardiac surgery training, and continued resident involvement in cardiac surgery to promote their development and surgical progression.