Allograft versus autograft decision for anterior cruciate ligament reconstruction: an expected-value decision analysis evaluating hypothetical patients

Abstract

Purpose: The purpose of this study was to determine the optimal decision between autograft and allograft for patients undergoing anterior cruciate ligament (ACL) reconstruction.

Methods: An expected-value decision analysis with sensitivity analysis was performed to systematically quantify the clinical decision. We evaluated 100 randomly selected individuals aged 16 to 70 years with regard to the following variables: age, sex, activity level (International Knee Documentation Committee form), and visual analog scale regarding potential outcome preferences. Patients with prior ACL injury were excluded. A decision tree was constructed (allograft v autograft potential outcomes), and a literature review determined probabilities of potential outcomes. Statistical fold-back analysis calculated optimal treatment. Sensitivity analysis determined the effect of changing the outcome probabilities on the decision.

Results: Of the subjects, 88 met the study inclusion criteria. The mean age was 44 years (range, 16 to 66 years), 67% of subjects were female, and the mean activity level was moderate. The expected value for autograft reconstruction was 11.22 versus 8.42 for allograft. Increasing the probability of complications associated with autograft (sensitivity analysis) decreased the expected value of autograft reconstruction. Significant limitations include that (1) decision analysis does not investigate actual patients in whom discussion of graft options between doctor and patient highly influences the decision and (2) patient decision largely depends on the information provided.

Conclusions: Decision analysis shows that autograft is preferred over allograft for ACL surgical reconstruction.

Clinical relevance: Patients' aversion to allograft tissue in general, and specific aversion to risk of disease transmission, results in a decision for ACL autograft, independent of expected outcomes.