Abstract
Background
Since the development of modern total hip replacement (THR) more than 50 years ago, thousands of devices have been developed in attempt to improve patient outcomes and prolong implant survival. Modern THR devices are often broadly classified according to their method of fixation; cemented, uncemented or hybrid (typically an uncemented acetabular component with a cemented stem). Due to early failures of THR in young active patients, the concept of hip resurfacing was revisited in the 1990's and numerous prostheses were developed to serve this patient cohort, some with excellent clinical results. Experience with metal-on-metal (MoM) bearing related issues particularly involving the ASR (DePuy Synthes, Warsaw, Indiana) precipitated a fall in the use of hip resurfacing (HR) prostheses in Australia from a peak of 30.2% in 2004 to 4.3% in 2015. The effects of poorly performing prostheses and what is now recognised as suboptimal patient selection are reflected in the AOANJRR cumulative percent revision (CPR) data which demonstrates 13.2% revision at 15 years for all resurfacing hip replacements combined; with 11 different types of hip resurfacing prostheses recorded for patients less than 55 years of age and a primary diagnosis of OA. When this data is restricted to only those prostheses currently used in Australia (BHR; Smith and Nephew, Birmingham, UK & ADEPT; MatOrtho Ltd, Surrey, UK) there is a CPR of 9.5% at 15 years for all patients. Despite these CPR results, recognition is emerging of the important distinction between MoM THR and resurfacing.
Furthermore, in light of current consensus for patient selection and the surgical indications for resurfacing, a gender analysis demonstrates a CPR for females of 14.5% at 10 years compared to 3.7% for males. Similar difference for head size >50mm with 6% CPR at 10 years compared to 17.6% for head size <50mm (HR=2.15; 1.76, 2.63; p<0.001). Leading to renewed interest in resurfacing particularly in the young, active male. In addition to registry based CPR data, several studies have concluded that a true difference in mortality rates between HR and other forms of THR exists independent of age, sex or other confounding factors. We hypothesised that a difference in adjusted mortality rates between HR and other forms of THR may also be present in the Australian population.
We undertook an ad hoc data report request to the AOANJRR. The data set provided was deidentified for patient, surgeon and institution and included all HR and conventional THR procedures performed for the diagnosis of primary osteoarthritis recorded in the Registry since inception in 1999. We requested mortality and yearly cumulative percent survival (CPS) of patients for primary HR and THR with sub-group analysis by the mode of fixation.
There were 12,910 hip resurfacings (79% male) compared to 234,484 conventional THR (46.8% male) over the study period. When adjusted for age and gender over the 15 years of available data, there was a statistically significant difference in cumulative percent survival (CPS) between conventional THR and hip resurfacing (HR 1.66 (1.52, 1.82; p<0.001)) and between cemented THR and hip resurfacing (HR 1.96 (1.78, 2.43; p<0.001)); between uncemented THR and hip resurfacing (HR 1.58 (1.45, 1.73; p<0.001)); and between hybrid THR and hip resurfacing (HR 1.82 (1.66, 1.99; p<0.001)). When adjusted for age, gender and ASA over the 3 years data available, there was no statistically significant difference in CPS between hip resurfacing and any individual fixation type of THR.
Discussion
The results demonstrate a statistically significant adjusted survival advantage for hip resurfacing compared to conventional THR and between fixation methods for THR. These findings are consistent with previous studies. While a difference in adjusted mortality rate appears to exist, we are yet to definitively determine the complex interplay of causative factors that may contribute to it.
