Is Training in Cardiac Surgery Dangerous to the Patients? A Retrospective Study with 1071 Propensity Score-Matched Subjects

Objectives: Unfortunately, training in cardiac surgery is in many centers still considered to increase the perioperative risk because of the technical complexity of performed procedures. This superstition prolongs the residency and makes the training more difficult. The aim of this study was to check, if a resident as main operator is a real risk factor.

Methods: We analyzed all patients who underwent elective isolated aortic valve replacement (AVR), elective isolated coronary artery bypass grafting (CABG) or both from January 2008 to June 2016 in our institution. Redo- and off-pump surgery, EF < 30% and other concomitant procedures were exclusion criteria. After this selection, we included 3,077 patients in our study. Within this group 357 (11.6%) had been operated by residents and 2,720 (88.4%) by consultants.

We performed a propensity score matching using the nearest neighbor method with ratio 1:2, considering the most important preoperative conditions like age, sex and comorbidities as well as the type of performed surgery. In this way, a total of 357 patients operated by residents were matched to 714 patients who were operated by consultants and had similar risk profiles. The standarized mean differences were highly reduced after matching. We analyzed preoperative conditions, surgical data, postoperative adverse events and the 30-day mortality and we compared them between analyzed subgroups.

Results: The preoperative profiles and types of performed surgery did not differ between the groups. The procedure lasted longer if the residents operated (172 ± 40 vs. 132 ± 43 minutes for consultants, p < 0.001). Similarly, longer among the residents’ patients were the cardiopulmonary bypass time (87 ± 29 vs. 66 ± 26 minutes for consultants, p < 0.001) and the aortic cross-clamp time (57 ± 21 vs. 42 ± 19 minutes for consultants, p < 0.001). There were no significant differences between the groups regarding postoperative adverse events.

Median time of mechanical ventilation was 15 hours within the group operated by residents and 14 hours if the consultants performed the surgery (p = 0.346). Median intensive care unit length of stay was the same for both groups (1 day, p = 0.118).

The 30-day mortality was very similar for both groups (1.7% among the residents' patients vs. 2% among the consultants’ patients, p = 0.75).

Conclusion: Training in cardiac surgery is safe and selected elective patients can be operated by residents without increased risk of perioperative mortality and complications.