Operative Treatment Has Greater Expected Value Than Nonoperative Treatment for First-Time Patellar Dislocations: A Meta-analysis and Expected Value Decision Analysis.

PURPOSE: To use expected value decision analysis to determine the optimal treatment for first time patellar dislocations.

METHODS

A meta-analysis according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines and expected-value decision analysis were performed. A decision tree addressing the clinical question (operative vs nonoperative) was created, and a meta-analysis was performed to assess the probability of outcomes after operative compared with nonoperative treatment. In total, 70 patients were assessed regarding potential outcome preferences to a hypothetical injury. An expected-value decision analysis was performed to systematically quantify the clinical decision. Statistical fold back analysis calculated optimal treatment, and a sensitivity analysis was performed to determine the effect of changing redislocation rates on the expected value.

RESULTS

Forty-five participants (mean age 20 years [range 12-33], 58% male, 71% athletes) met inclusion criteria. Meta-analysis of 10 randomized controlled trials with 624 patients revealed the probability of a "well" outcome was significantly greater for operative treatment (59.3%, 95% confidence interval [CI] 53.7%-64.7%) than nonoperative treatment (44.7%, 95% CI 39.0%-50.5%). Subsequent redislocation rates with operative treatment were significantly lower (29.8%, 95% CI 24.5%-35.17%) compared with nonoperative treatment (44.7%, 95% CI 39.0%-50.5%). Medial patellofemoral ligament reconstruction showed an 88% probability of a "well" outcome, and a 3% redislocation rate. Operative management had a greater chance of a well outcome (risk ratio 1.43; 95% CI 1.12-1.83, P = .005). The overall expected value for operative treatment was 6.09 versus 4.96 for nonoperative treatment. Secondary analysis of 27 articles for solely medial patellofemoral ligament reconstruction showed an expected value of 7.80 for operative treatment.

CONCLUSIONS

Meta-analysis showed more frequent favorable outcomes and lower subsequent dislocation rates with operative treatment. Decision analysis on the basis of healthy patient responses to hypothetical scenarios shows that operative treatment has a greater expected value for first time patellar dislocations than nonoperative treatment.

LEVEL OF EVIDENCE: Level IV, economic study.