Serum metal ion levels in patients with failed total shoulder arthroplasty.

BACKGROUND

Shoulder arthroplasty systems are transitioning toward modular prosthesis options with the goal of reducing complication profiles and increasing range of motion. Modularity may increase the potential for fretting, corrosion, and subsequent release of metal ions. The purpose of this study was to identify associations between implant designs, metallurgy, patient characteristics, and serum metal ion levels in patients undergoing revision shoulder arthroplasty.

METHODS

Fifty-one patients who underwent revision shoulder arthroplasty were retrospectively reviewed based on prospectively collected data. All patients had serum metal ion levels (titanium, cobalt, chromium) measured in parts per billion (ppb) before their revision arthroplasty. Trends and correlations to serum ion levels were examined based on implant materials, number of modular components, and patient characteristics.

RESULTS

Twenty-one patients had anatomic total shoulder arthroplasty (aTSA), and 30 had reverse total shoulder arthroplasty (rTSA). The average age at primary arthroplasty was 64.7 years (49.0-84.0). The average term of implant was 4.83 years (0.18-21.0). Forty-six patients (90%) had titanium-based humeral stems. Twenty-one (100%) aTSA humeral heads were composed of cobalt chrome. Five patients with aTSA had hybrid titanium and polyethylene glenoid components. In those with rTSA, 30 (100%) had glenoid components composed of titanium baseplates and cobalt chrome glenospheres. Titanium ion levels were significantly higher in patients with greater than 7 total modular components (12.6 vs. 6.09 ppb, P = .008). Titanium ion levels were significantly higher in patients with greater than 4 modular components when excluding screws (12.1 vs. 5.06, P = .038). The number of modular components did not have a statistically significant effect on cobalt or chromium serum ion levels. Patients with intraoperative metallosis during revision surgery had titanium ion levels 10.51 units higher on average (P = .009) than those without metallosis. There was no difference in titanium (P = .63), cobalt (P = .29), or chromium (P = .58) levels in those patients with a proven infection compared to those without.

CONCLUSIONS

Increasing modularity in primary TSA should be weighed with the potential for increased serum metal ion levels, particularly titanium. Although the systemic effects of elevated serum titanium remain largely unknown, it may be a predictor for occult corrosion or need for revision. Metallosis poses challenges in the revision setting; however, the implications of increased serum metal ion levels on clinical outcomes after revision shoulder arthroplasty requires further studies.