Study objective: To determine the optimal strategy for the timing of aortic valve replacement in patients with chronic, severe aortic regurgitation with left ventricular dysfunction.
Design: Decision analysis comparing early surgery (timed at the onset of left ventricular dysfunction) with delayed surgery (timed at the onset of symptoms) using data from the literature and expert opinion for variables in a representative case scenario (40-year-old man with bicuspid aortic valve disease).
Setting: Tertiary care center doing valve replacement surgery.
Measurements and main results: The early-surgery approach was preferred based on quality-adjusted life years. Sensitivity analysis showed that the result was not affected by the following variables within their derived ranges: rate of symptom development after onset of left ventricular dysfunction for the delayed-surgery approach, perioperative mortality for both approaches, and occurrence of major nonfatal stroke or congestive heart failure for both approaches. Although the decision was sensitive to the yearly postoperative mortality rates, the delayed-surgery operative mortality rate had to be almost as low as the early-surgery rate to change the preference to the delayed-surgery approach. The preference could also change if survival were much more important to the patient in the first five years than after five years or if the patient disliked living on anticoagulants enough to value a year on anticoagulants as worth only 80% of a year not on anticoagulants.
Conclusion: This decision analysis provides quantitative support for the impression that patients similar to the case scenario do better with surgery timed at the onset of ventricular dysfunction than with surgery delayed until symptoms develop. It thus supports the practice of following these patients noninvasively in order to time surgery.