Prediction of Surgical Outcome after Aortic Valve Replacement

BACKGROUND: Aortic valve replacement has some major adverse outcomes. For these, the predictors need identification. METHODS: This was a retrospective file study of 1000 consecutive patients who underwent AVR for degenerative aortic valve disease. Twenty-five preoperative and 5 peroperative factors were screened by a univariate Fisher-exact analysis. The predictors were identified in a second step by logistic regression multivariate analysis. RESULTS: Five hundred thirty patients were male. The mean age was 75 (71-77) years and 610 also underwent CABG. For hospital mortality, need for urgent aortic valve replacement (p < 0.001) was the dominant predictor. Need for digitalis (p = 0.002) and age > 80 (p = 0.005) followed. For postoperative congestive heart failure, need for urgent aortic valve replacement was also dominant (p <0.001). Atrial fibrillation (p = 0.001,) and ejection fraction < 50% (p = 0.055) were less important. For ventricular arrhythmia, previous infarction (p = 0.025) and ejection fraction < 50% (p = 0.032) were identified. For bleeding, concomitant CABG (p = 0.046) and chronic obstructive pulmonary disease were identified. For thromboembolic events only an ejection fraction < 50% (p = 0.027) was identified. CONCLUSIONS: Need for urgent aortic valve replacement is the dominant predictor for postoperative mortality and congestive heart failure. Once a degenerative aortic valve disease becomes symptomatic, prompt referral could prevent the development for need for urgent surgery, with all its adverse postoperative consequences.