Hip resurfacing versus total hip arthroplasty: a systematic review comparing standardized outcomes.

BACKGROUND: Metal-on-metal hip resurfacing was developed for younger, active patients as an alternative to THA, but it remains controversial. Study heterogeneity, inconsistent outcome definitions, and unstandardized outcome measures challenge our ability to compare arthroplasty outcomes studies. QUESTIONS/PURPOSES: We asked how early revisions or reoperations (within 5 years of surgery) and overall revisions, adverse events, and postoperative component malalignment compare among studies of metal-on-metal hip resurfacing with THA among patients with hip osteoarthritis. Secondarily, we compared the revision frequency identified in the systematic review with revisions reported in four major joint replacement registries. METHODS: We conducted a systematic review of English language studies published after 1996. Adverse events of interest included rates of early failure, time to revision, revision, reoperation, dislocation, infection/sepsis, femoral neck fracture, mortality, and postoperative component alignment. Revision rates were compared with those from four national joint replacement registries. Results were reported as adverse event rates per 1000 person-years stratified by device market status (in use and discontinued). Comparisons between event rates of metal-on-metal hip resurfacing and THA are made using a quasilikelihood generalized linear model. We identified 7421 abstracts, screened and reviewed 384 full-text articles, and included 236. The most common study designs were prospective cohort studies (46.6%; n = 110) and retrospective studies (36%; n = 85). Few randomized controlled trials were included (7.2%; n = 17). RESULTS: The average time to revision was 3.0 years for metal-on-metal hip resurfacing (95% CI, 2.95-3.1) versus 7.8 for THA (95% CI, 7.2-8.3). For all devices, revisions and reoperations were more frequent with metal-on-metal hip resurfacing than THA based on point estimates and CIs: 10.7 (95% CI, 10.1-11.3) versus 7.1 (95% CI, 6.7-7.6; p = 0.068), and 7.9 (95% CI, 5.4-11.3) versus 1.8 (95% CI, 1.3-2.2; p = 0.084) per 1000 person-years, respectively. This difference was consistent with three of four national joint replacement registries, but overall national joint replacement registries revision rates were lower than those reported in the literature. Dislocations were more frequent with THA than metal-on-metal hip resurfacing: 4.4 (95% CI, 4.2-4.6) versus 0.9 (95% CI, 0.6-1.2; p = 0.008) per 1000 person-years, respectively. Adverse event rates change when discontinued devices were included. CONCLUSIONS: Revisions and reoperations are more frequent and occur earlier with metal-on-metal hip resurfacing, except when discontinued devices are removed from the analyses. Results from the literature may be misleading without consistent definitions, standardized outcome metrics, and accounting for device market status. This is important when clinicians are assessing and communicating patient risk and when selecting which device is most appropriate for individual patients.

Cost-utility of metal-on-metal hip resurfacing compared to conventional total hip replacement in young active patients with osteoarthritis.

BACKGROUND: Metal-on-metal hip resurfacing arthroplasty (MoM HRA) has emerged as an alternative to total hip arthroplasty (THA) for younger active patients with osteoarthritis (OA). Birmingham hip resurfacing is the most common MoM HRA in Alberta, and is therefore compared with conventional THA. OBJECTIVE: The objective of this study was to estimate the expected cost-utility of MoM HRA versus THA, in younger patients with OA, using a decision analytic model with a 15-year time horizon. METHODS: A probabilistic Markov decision analytic model was constructed to estimate the expected cost per quality-adjusted life-year (QALY) of MoM HRA versus THA from a health care payer perspective. The base case considered patients with OA aged 50 years; men comprised 65.9% of the cohort. Sensitivity analyses evaluated cohort age, utility values, failure probabilities, and treatment costs. Data were derived from the Hip Improvement Project and the Hip and Knee Replacement Pilot databases in Alberta, the 2010 National Joint Replacement Registry of the Australian Orthopaedic Association, and the literature. RESULTS: In the base case, THA was dominated by MoM HRA (incremental mean costs of -$583 and incremental mean QALYs of 0.079). In subgroup analyses, THA remained dominated when cohort age was 40 years instead of 50 years or when only men were assessed. THA dominated when the cohort age was 60 years or when only women were assessed. Results were sensitive to utilities, surgery costs, and MoM HRA revision and conversion probabilities. At a willingness-to-pay of Can $50,000/QALY, there was a 58% probability that MoM HRA is cost-effective. CONCLUSIONS: The results show that, on average, MoM HRA was preferred to THA for younger and male patients, but THA is still a reasonable option if the patient or clinician prefers given the small absolute differences between the options and the confidence ellipses around the cost-effectiveness estimates.