In 2010, a widely used metal-on-metal hip implant was voluntarily recalled from the market by the manufacturer. Our lab performed detailed retrieval analysis of 548 of the explanted devices and 165 periprosthetic tissue specimens. We reported a high degree of variability in the multiple measurements that we performed on the components, including volumetric wear. Other studies using semi-quantitative tissue grading methods to describe the histology of the periprosthetic tissues from metal-on-metal hips have looked for relationships between component wear and histological features. Grammatopoulos et al found higher ALVAL scores in 45 hip resurfacing patients with pseudotumors compared to 11 without pseudotumors and a moderately positive correlation between wear and the histological rankings for tissue necrosis and lymphocytic response. We examined correlations between the component wear measurements from retrieval analysis and the tissue features in this large group of one metal-on-metal hip design. Tissues were processed routinely into paraffin sections, stained with H&E and characterized with light microscopic semi-quantitative scores for ALVAL features and for the number of macrophages and lymphocytes, extent of metal debris and the estimated percentage of necrosis per slide. Spearman correlation analysis was conducted to evaluate any correlations between retrieval wear measurements and histological findings.Introduction
Materials and Methods
Expectations for ceramic-on-metal (COM) bearings included (i) optimal lubrication due to smoother ceramic heads (ii), reduction of metal ions due to elimination of CoCr heads, and (iii) ‘differential hardness’ reducing adhesive wear and squeaking (Firkins 2001, Williams 2007). Additional benefits included (iv) use of heads larger than for ceramic-on-ceramic (COC), (v) reduction in taper corrosion and (vi) simulator studies clearly demonstrated metal ions and wear both reduced compared to MOM (Firkins 2001, Williams 2007, Ishida 2007). However, contemporary ‘3rd body wear’ paradigms focused only on metal debris size range 0.025–0.035um (Firkins 2001). Thus, neglected was the effect of hip impingement, provoking release of large metal particles sized 20–200um (Clarke 2013). In this study, we compared COM retrievals using hypotheses that adverse COM cases would demonstrate a combination of (a) steeply inclined cups, (b) liner “edge-loading”, (c) Ti6Al4V contamination on ceramic, and (d) evidence of 3rd-body CoCr wear by large particles. As a case example, this 51-year old female had her metal-polyethylene (MPE) bearing revised to COM in June 2011. She reported no symptoms 1-year post-op, but scans revealed a palpable mass in the inguinal region of left hip. By March 2013 the patient reported mild pain in her hip, which progressed to severe by April 2014. Scans showed a solid and cystic iliopsoas bursitis while cup position had changed from 43o to 73o inclination. Revision was performed in June 2014, her joint tissues were found extensively stained due to metal contamination, and histology described formation of a large pseudotumor. Analysis of retrieved components was by interferometry, SEM and EDS. Detailed maps were made of wear areas in heads and cups and volumetric wear was determined by CMM techniques. This adverse COM example revealed large diametral mismatch (595um) compared to COM controls (75–115um). The ceramic head had a broad polar stripe of CoCr contamination, roughness 0.1–0.3um high. Equatorial ceramic areas showed arrays of thin metal smears that demonstrated elemental Ti and Al. The CoCr liner revealed wear area into cup rim, as “edge loading”, and also featured a focal rim-defect over 18o circumferential arc. Liner scratches were 20um wide and larger, and wear-rate of CoCr liner averaged approximately 50mm3 per year. In contrast, ceramic head had minimal wear. Our study highlights the underappreciated risk of impingement by metallic prosthetic components. Prior studies of ceramic heads showed black metallic smears. With COM we can anticipate that the broad polar smear will be CoCr alloy (wear of liner on head). However, Ti6Al4V smearing on ceramic heads is a notable signpost indicating impingement by the Ti6Al4V acetabular shell. The femoral neck (Ti6Al4V: CoCr), may also be damaged. Release of large metal particles, 1500-times larger than prior predictions, provoke a particularly adverse ‘3rd body wear’ (Halim, 2015). Such cases confirm our four hypotheses, that COM bearings will then fail in a way similar to MOM. In contrast, COC bearings are immune to such impingement and 3rd-body metal damage.
3-D finite element model of a resurfaced femoral head was composed. Five configurations of cement layer were analyzed and the transient heat transfer analysis during cement polymerization was performed. Peak temperature at the bone-cement interface temperature was lower than 40 oC when there was no or 1.5 mm cement penetration but reached 54 oC and 74 oC with 6 mm penetration and 6 mm penetration plus a cement –filled cyst of 1 cm3, respectively. With deep cement penetration, and a large cement-filled cyst, the peak temperatures exceeded bone thermal osteonecrosis at 55 oC. To evaluate using a finite element analysis model, the possibility of bone thermal necrosis secondary to cement in resurfacing arthroplasty of the hip. With deep cement penetration, and the presence of a large cement-filled cyst, the peak temperatures were in the range of bone thermal osteonecrosis 55 oC. Cementing technique in resurfacing arthroplasty should strive to strike a balance between fixation and avoiding bone thermal necrosis by excessive cement penetration. This information could explain why femoral head cysts >
1cm are a risk factor for femoral loosening after resurfacing arthroplasty and excessive cement penetration could lead to femoral neck fracture. 3-D finite element model of a hemispherical resurfaced femoral head was composed of a metal shell with a diameter of 46 mm. Five configurations of cement layer were analyzed a) no penetration into the bone, b) 1.5 mm penetration, c) 6 mm penetration, d) 6 mm penetration and a 1 cm3 cement filled cyst, and e) 6 mm penetration and 2 cm3 cement-filled cyst. The transient heat transfer analysis during cement polymerization was performed in a series of time steps. The temperature within the bone and cement was lower than 40 oC when there was no or 1.5 mm cement penetration into the femoral head. In contrast, the peak temperature at the bone-cement interface reached 54 oC and 74 oC and 63 oC with 6 mm penetration and 6 mm penetration plus a cement –filled cyst of 1 cm3, respectively.
