Importance: More than half of US cardiothoracic surgeons are older than 55 years, and the association between surgeon number of years in practice and surgical outcomes remains unclear.
Objective: To assess the association between cardiac surgeons' time in practice and operative outcomes for coronary artery bypass grafting (CABG) and valve surgery.
Design, setting, and participants: Cross-sectional analysis performed of surgeon-level outcomes data from the 2014-2016 New York State Cardiac Data Reporting System across the 38 New York cardiac surgery centers. Years in practice were characterized as early career (<10 years) and late career (≥10 years). Participants were 120 cardiothoracic surgeons who performed CABG and 112 cardiothoracic surgeons who performed valve procedures between 2014 and 2016. Data were analyzed in April 2020. Surgeons who trained outside of the United States or had unclear training history were excluded.
Main outcomes and measures: Risk-adjusted operative mortality rate (RAMR). Mortality was defined as all-cause death within 30 days of surgery or within the index hospitalization, whichever was longer. Risk adjustment was performed by a multivariable risk model developed by the New York State Department of Public Health. Restricted cubic spline curve identified the association between risk-adjusted mortality rate and surgeon number of years in practice. Linear regression models adjusted for surgeons' annual case volumes.
Results: A total of 112 CABG surgeons and 120 valve surgeons performed 39 436 CABG and 18 596 valve procedures between 2014 and 2016. The median number of surgeon years in practice was 20.0 (interquartile range [IQR], 12.0-28.5) years. The median surgeon annual case volume was 160.0 (IQR, 92.5-245.0) for CABG procedures and 104.0 (IQR, 43.0-210.0) for valve procedures. The median RAMR was 1.3% (IQR, 0.2%-2.2%) for CABG procedures and 3.1% (IQR, 1.7%-5.1%) for valve procedures. Surgeons with less than 10 years of practice had higher RAMR for valve procedures compared with surgeons with more than 10 years of practice (4.0 [IQR, 1.5-7.7] vs 2.9 [IQR, 1.7-4.7]; P = .20), but the finding was not statistically signficant. The RAMR for surgeons with less than 10 years of practice was similar compared with surgeons with more than 10 years of practice for CABG procedures (1.3 [IQR, 0.3-2.1] vs 1.3 [IQR, 0.0-2.2]; P = .73). A lower number of years in practice was significantly associated with higher RAMR for valve procedures (RAMR estimates for linear term: -1.144; 95% CI, -1.955 to -0.332; P = .006; quadratic term: 0.059; 95% CI, 0.015 to 1.102; P = .008; and cubic term: -0.001; 95% CI, -0.002 to 0.000; P = .01). This association was not observed for CABG.
Conclusions and relevance: In this cross-sectional study, compared with late-career cardiac surgeons, early-career cardiac surgeons were associated with worse risk-adjusted outcomes for valve operations but not for CABG. This finding suggests certain competence deficiency for valve surgery early after finishing training in cardiac surgery.