Operative mortality and morbidity after the Ross procedure: a 26- year learning curve

The role of the Ross procedure in adults is controversial. In comparison with prosthetic aortic valve replacement (AVR), its greater technical complexity has been associated with increased operative mortality. The aim of the present study was to evaluate the impact of surgeon experience on early outcomes after the Ross procedure.

Using multivariate analysis, the details of 530 consecutive patients who had undergone the Ross operation between 1987 and 2013 were analyzed according to whether they had surgery during the early, middle, or later phases of a single-surgeon experience.

Cumulative sum analysis revealed an initial learning curve of around 100 cases. There was no operative mortality in the later phase of 305 cases, compared to 3% (3/100) in the early phase and 2.4% (3/125) in the middle phase (p = 0.04). The mean EuroSCORE did not change significantly across the series (2.3-3.2%, p = 0.4), whereas the patient mix did change. A decreasing incidence of comorbidities, such as acute endocarditis (from 10% to 1.3%, p = 0.02), congestive heart failure (from 8% to 2.3%, p = 0.03) and coronary artery disease (from 11% to 2.3%, p = 0.001), occurred in parallel with increasing case complexity. Isolated Ross procedures were performed less frequently later in the series (42%, n = 130) compared to the early stage (68%, n = 68; p < 0.001), and concomitant ascending aortic or arch replacement were significantly more likely to be performed in later patients (33.8%, n = 103) compared to early-stage patients (1%, n = 1; p <0.001). Independent predictors of operative mortality and major morbidity included endocarditis (odds ratio (OR) 5.2, 95% confidence interval (CI) 1.3-21.5), diabetes (OR 4.3, 95% CI 1.2-15.5), and concomitant coronary artery bypass (OR 3.2, 95% CI 1.2-8.4). Residents or junior faculty performed the surgery in 26 cases, under direct supervision and without patient mortality.

The Ross procedure can be performed with early outcomes comparable to that of isolated AVR in adults. The impact of the learning curve on patient outcomes may be mitigated by careful patient selection and experienced supervision.