Immediate Versus Delayed Hip Arthroscopy for Femoroacetabular Impingement: An Expected Value Decision Analysis.

introductionHip arthroscopy is an increasingly used surgical procedure for both intra- and extra-articular hip pathologies, including femoroacetabular impingement (FAI). Although the arthroscopic approach is known to be preferable to open, the optimal timing of such intervention is unclear. The purpose of this study was to carry out an expected value decision analysis of immediate versus delayed hip arthroscopy for FAI. Its hypothesis is immediate hip arthroscopy is the preferable treatment option.

methodsAn expected value decision analysis was implemented to systematize the decision-making process between immediate and delayed hip arthroscopies. A decision tree was created with options for immediate and delayed surgeries with utilities characterizing each state obtained from surveying 70 patients. Fold-back analysis was then carried out, calculating expected values by multiplying the utility of each health outcome by the probability of that outcome. Corresponding expected values were then summed to "fold back" the decision tree one layer at a time. This was repeated until overall expected values (0 to 100) for immediate and delayed hip arthroscopies resulted with the higher value indicating the preferable option.

resultsFold-back analysis demonstrated that immediate hip arthroscopy is the preferred treatment for FAI over delayed with expected values of 78.27 and 72.63, respectively. Restoration of good function after hip arthroscopy was the most notable contributor to this difference. Immediate hip arthroscopy remained superior even as vast adjustments to preoperative physical function were made in one-way sensitivity analysis. Complications of hip arthroscopy leading to total hip arthroplasty were the least notable contributors to overall expected values. DISCUSSION: This study confirms that immediate surgery is the preferred option when using decision-making analysis combining patient-reported utilities of health outcomes and the probabilities of those outcomes from the literature. This is consistent across a range of estimates of poor function in both the delayed and immediate surgery arms.