Posterior versus lateral surgical approach for total hip arthroplasty in adults with osteoarthritis

Background Osteoarthritis (OA) of the hip is a progressive condition that has no cure and often requires a total hip arthroplasty (THA). The principal methods for THA are the posterior and direct lateral approaches. The posterior approach is considered to be easy to perform, however, increased rates of dislocation have been reported. The direct lateral approach facilitates cup positioning which may decrease rates of hip dislocation and diminishes the risk of injury to the sciatic nerve. However, there is an increased risk of limp. Dislocation of a hip prosthesis is a clinically important complication after THA, in terms of morbidity implications and costs. Objectives To determine the risks of prosthesis dislocation, postoperative Trendelenburg gait and sciatic nerve palsy after a posterior approach, compared to a direct lateral approach, for adult patients undergoing THA for primary OA and to update the previous review made in 2003. Search strategy MEDLINE, EMBASE, CINAHL and Cochrane databases were searched and updated, from the previous search of 2002, to Oct 13, 2005. No language restrictions were applied. Selection criteria Published trials comparing posterior and direct lateral surgical approaches to THA in participants 18 years and older with a diagnosis of primary hip OA. Data collection and analysis Retrieved articles were assessed independently by the two reviewers for their methodological quality. Main results Four prospective cohort studies involving 241 participants met the inclusion criteria. The primary outcome, dislocation, was reported in two studies. No significant difference between posterior and direct lateral surgical approach was found [1/77 (1.3%) versus 3/72 (4.2%); relative risk (RR) 0.35; 95% confidence intervals (CI) 0.04 to 3.22]. The presence of postoperative Trendelenburg gait was not significantly different between these surgical approaches. The risk of nerve palsy or injury (all nerves taken together) was significantly higher among the direct lateral approaches [1/43 (2%) versus 10/49 (20%); RR 0.16, 95% CI 0.03 to 0.83]. However, there were no significant differences when comparing this risk nerve by nerve for both approaches, in particular for the sciatic nerve. Of the other outcomes considered only the average range of internal rotation in extension of the hip was significantly higher (weighted mean difference 16 degrees, 95% CI 8 to 23) in the posterior approach group (mean 35 degrees, standard deviation 13 degrees) compared to the direct lateral approach (mean 19 degrees, standard deviation 13 degrees). Authors' conclusions The quality and quantity of information extracted from the trials performed to date are insufficient to make any firm conclusion on the optimum choice of surgical approach in adult patients undergoing primary THA for OA.