Abstract 12957: Prognostic Significance of Elevated Troponin I After Valve Heart Surgery
Background: Cardiac troponin is associated with adverse events after cardiac surgery. However, the ideal time for its measurement after surgery is unclear. In order to predict its relationship to in-hospital mortality, independent value of troponin I levels, at different time points during the first 72 hours after valvular heart surgery, was evaluated.
Methods: All patients, who underwent valvular heart surgery at a single tertiary center from 2011 to 2015, were included. Troponin levels were grouped according to the time after surgery as: within the first 12h, 12 to 24h, 24 to 48h, and 48h to 72h. Univariable and multivariable logistic regression models were built to define the role of troponin measurements during the first 72 hours after surgery as predictors of in hospital mortality. The multivariable models were adjusted for EuroSCORE (ES). Model discrimination was compared using the area under the ROC curve.
Results: 1854 patients were included (48% women, mean age 56±15 years), with a mean ES of 4.5±3.6. The mortality was 14% (267 patients). Individuals who died during the admission had higher troponin values during the first 72 hours (p<0.001), the largest difference occurred after 24 hours (figure, left panel). After adjustment for the ES, all troponin measurements remained significantly associated with mortality (OR per 10 unit change in troponin: 1.3 before 12h; 1.4 at 12 to 24 hours, 1.8 at 24 to 48 hours; and 2.0 at 48 to 72 hours (p<0.001 for all). While initial post-operative troponin measurement did not significantly improved the discrimination for mortality (p=0.17 for troponin before 12 hours), the discrimination improved for late troponin measurements (figure, right panel).
Conclusion: Postoperative troponin is independently associated with in-hospital mortality in individuals who underwent valvular heart surgery. Among the troponin values, those collected after 24 hours are more accurate independent preditors of in-hospital mortality.
Author Disclosures: R.C. Aragao: None. R.O. Sampaio: None. M.S. Bittencourt: None. A.S. Lopes: None. T.A. Accorsi: None. V.E. Rosa: None. J.R. Fernandes: None. G.S. Spina: None. L.A. Hajjar: None. F. Tarasoutchi: None.
- © 2015 by American Heart Association, Inc.