Validation of the TRI-SCORE in patients undergoing surgery for isolated tricuspid regurgitation

Estimation of peri-procedural risk in patients with tricuspid regurgitation (TR) undergoing isolated tricuspid valve surgery (ITVS) is of paramount importance. The TRI-SCORE is a new surgical risk scale specifically developed for this purpose, which ranged from 0 to 12 points and included eight parameters: right-sided heart failure signs, daily dose of furosemide ≥125 mg, glomerular filtration rate <30 mL/min, elevated bilirubin (with a value of 2 points), age ≥70 years, New York Heart Association Class III-IV, left ventricular ejection fraction <60% and moderate/severe right ventricular dysfunction (with a value of 1 point). The objective of the study was to evaluate the performance of the TRI-SCORE in an independent cohort of patients undergoing ITVS.A retrospective observational study was performed in four centres, including consecutive adult patients undergoing ITVS for TR between 2005 and 2022. The TRI-SCORE and the traditional risk scores used in cardiac surgery (Logistic EuroScore (Log-ES) and EuroScore-II (ES-II)) were applied for each patient, and discrimination and calibration of the three scores were evaluated in the entire cohort.A total of 252 patients were included. The mean age was 61.5±11.2 years, 164 (65.1%) patients were female, and TR mechanism was functional in 160 (63.5%) patients. The observed in-hospital mortality was 10.3%. The estimated mortality by the Log-ES, ES-II and TRI-SCORE was 8.7±7.3%, 4.7±5.3% and 11.0±16.6%, respectively. Patients with a TRI-SCORE ≤4 and >4 had an in-hospital mortality of 1.3% and 25.0%, p=0.001, respectively. The discriminatory capacity of the TRI-SCORE had a C-statistic of 0.87 (0.81-0.92), which was significantly higher than both the Log-ES (0.65 (0.54-0.75)) and ES-II (0.67 (0.58-0.79)), p=0.001 (for both comparisons).This external validation of the TRI-SCORE demonstrated good performance to predict in-hospital mortality in patients undergoing ITVS, which was significantly better than the Log-ES and ES-II, which underestimated the observed mortality. These results support the widespread use of this score as a clinical tool.