Abstract
INTRODUCTION
Metal-on-metal hip resurfacing (MoMHRA) requires a new standardized radiographic evaluation protocol. Evaluation of the acetabular component is similar to total hip arthroplasty but the femoral component requires different criteria since there is no component in the femoral canal and the metallic femoral implant overlies the junctions between bone-cement and cement-prosthesis. Lucencies around the metaphyseal HRA femoral stem can be described with the femoral zonal system into 3 peg-zones (Amstutz' et al) but this doesn't account for bony changes of the femoral neck away from the stem. This study proposes a new femoral zonal system for radiographic HRA assessment. We tested the efficacy of radiographs in identifying a problem by reviewing 711 radiographs of resurfaced hips and correlating radiographic features to outcome.
METHODS
611 in-situ HRA (one surgeon) with minimum two radiographs at >12 months postoperatively and 100 revised HRA (55 referred) were assessed for component positioning, reactive lines±cortical thickening±cancellous condensation (borderline) and lucent lines±osteolysis±bone resorption (sinister). Findings around the acetabular implant were classified in six zones: Zones I-III equally distributed acetabular zones (DeLee-Charnley); Zone IV, V and VI situated in the iliac, pubic and ischial bone respectively. Findings around the proximal femur are defined with a new zonal system, dividing the implant-cement-bone interfaces and the femoral neck into 7 areas. Zones 1,7 at the superior and inferior part of the femoral neck-head, zones 2,3 at the proximal and distal halves of the superior aspect of the stem, zone 4 at the tip, zones 5,6 at the distal and proximal inferior aspects of the stem). Radiological findings and zones were correlated with gender, size, survival, Harris Hip Scores (HHS), metal ions, and adverse soft tissue reactions (ALTR).
RESULTS
Radiological changes were found in 260 cases (36.7%), 151 sinister (21.2%) and 110 borderline (15.5%). 82% of revisions had sinister findings versus 11.3% of in-situ HRA (p<0.001). Of the 52 revised cases with ALTR, 2 had normal Xrays, 3 borderline and 47 sinister (90.4%). Females had a higher incidence of sinister findings (28.9%) compared to males (16.6%)(p<0.001). Sinister radiographs were significantly associated with lower HHS (mean 85 versus 98)(p<0.001), smaller sizes (median 48 versus 50)(p<0.001) and ASR design (58.8%). Metal ions were significantly higher with sinister (mean Cr:17.6μg/l-Co:17.8μg/l)(p<0.001) and borderline findings (mean Cr:3.34μg/l-Co:3.01μg/l)(p<0.01) compared to normal radiographs (mean Cr:2.61μg/l-Co:2.26μg/l). A higher number of pathological zones was correlated with risk of revision(p=0.006), ALTR(p<0.001), female gender(p<0.001), smaller size(p<0.001), and lower HHS(p<0.001). Pathological findings in 34 zones had a sensitivity of 50% and specificity of 94.2% (AUC=0.898) in detecting hips with ALTR (risk ratio=19.6).
DISCUSSION
Radiographic assessment of HRA provides valuable information: 82% of problematic hips had sinister changes and 90.4% of cases with proven ALTR. In order to accurately evaluate progressive radiographic changes in HRA, specific zones have been established around the HRA femoral component. There was a high correlation between radiographic findings, clinical outcome and metal in levels. However, the absence of sinister radiologic findings does not eliminate a problematic HRA and further investigations (ion levels, imaging) would be indicated in symptomatic patients.
