The purpose of this study was to compare risks for revision and short-term complications after total joint arthroplasty (TJA) in matched cohorts of morbidly obese patients, receiving and not receiving prior bariatric surgery.

Patients undergoing elective TJA between 1997 and 2011 were identified in a New York Statewide database, analyzing total knee arthroplasty (TKA) and total hip arthroplasty (THA) separately. Propensity scores were used to match morbidly obese patients receiving and not receiving bariatric surgery prior to TJA. Cox proportional hazard modeling assessed revision risk. Logistic regression evaluated odds for complications.

For TKA, 2636 bariatric surgery patients were matched to 2636 morbidly obese patients. For THA, 792 bariatric surgery patients were matched to 792 morbidly obese patients. Matching balanced all covariates. Bariatric surgery reduced co-morbidities prior to TJA (TKA P < .0001; THA P < .005). Risks for in-hospital complications were lower for THA and TKA patients receiving prior bariatric surgery (odds ratio [OR] 0.25, P < .001; and OR = 0.69, P = .021, respectively). Risks for 90-day complications were lower for TKA (OR 0.61, P = .002). Revision risks were not different for either THA (P = .634) or TKA (P = .431), nor was THA dislocation risk (P = 1.000).

After accounting for relevant selection biases, bariatric surgery prior to TJA was associated with reduced co-morbidity burden at the time of TJA and with reduced post-TJA complications. However, bariatric surgery did not reduce the risk for revision surgery for either TKA or THA.