Safety of a cardiac surgical training program over a twelve-year period

INTRODUCTION: The operative results of cardiac surgical procedures performed either by staff surgeons or trainees were compared to evaluate whether there is any additional risk to the patient in operations carried out by trainees.

PATIENT AND METHODS: Between 1994 and 2006, 13 197 operations were done by 9 staff surgeons (S-group) and were compared to 1925 operations performed by 10 residents during their training program (R-group). In both groups, subgroups were defined in which patients either underwent isolated coronary artery bypass grafting (CABG) or aortic valve replacement (AVR). Isolated CABG was performed in 8725 cases (S-group) and 1706 cases (R-group). AVR was done in 1273 patients (S-group) and 191 patients in the R-group. The groups were compared with regard to length of surgery, in-hospital mortality, reexploration for bleeding, sternal wound complications, preoperative infarction and stroke. For overall risk stratification, the EuroSCORE was used. In the CABG groups, the KCH score was additionally used.

RESULTS: Overall mortality was 0.5 % in the R-group (predicted mortality: 4.6 +/- 5 %) compared to 2 % in the S-group (predicted mortality: 6.9 +/- 7.9 %). Analyzing the CABG groups, mortality was 0.7 % in the R-group (predicted: 1.9 +/- 1.8 %) compared to 1.3 % in the S-group (predicted: 2.1 +/- 2.3 %). The rate of reexplorations for bleeding was 1 % in the R-group compared to 2 % in the S-group ( P < 0.05). Duration of extracorporeal circulation (ECC) was slightly longer in the R-group (95.8 +/- 33 compared to 85.5 +/- 28.3 minutes) without reaching statistical significance ( P > 0.05). Analyzing the AVR group, there was a 0.6 % mortality in the R-group (predicted: 6.5 +/- 6 %) compared to 3.1 % in the S-group (predicted: 8.8 +/- 8 %). Again, there were significantly more reexplorations for bleeding in the S-group (0.6 % vs. 2.8 %, P < 0.05). Time on ECC was longer in the R-group (101.6 +/- 21.6 vs. 96.6 +/- 35 minutes) with a resulting longer cross-clamping time (71.2 +/- 17.1 vs. 63.5 +/- 21 minutes).

CONCLUSIONS: Training in cardiac surgery can be done with excellent results and no additional risk to the patients. Both groups performed similarly and had outcomes better than the predicted mortality. Training was mostly performed with isolated coronary artery bypass grafting procedures, resulting in only a small surgical spectrum at the time of board certification.