Over 1 million metal-on-metal (MOM) hip replacements were performed between 2003 and 2010. These prostheses were intended to be more durable than previous models, but have been associated with a higher failure rate. There have also been isolated cases of fulminant cardiomyopathy in patients with very high serum cobalt levels. We screened a cohort of asymptomatic patients with these prostheses for subclinical cardiac abnormalities.Methods
All patients with MOM hip replacements at our centre undergo regular follow-up in the orthopaedic clinic with serum cobalt and chromium levels, and periodic magnetic resonance scanning of the affected joint. We recruited consecutive asymptomatic patients to receive echocardiography and recorded demographic data including age, height, weight, serum cobalt and chromium levels, and date of prosthesis implantation. Echocardiographers were blinded to medical history and laboratory results. The cohort was split into quartiles of serum cobalt. ANOVA, Kruksall-Wallis H test and Chi-squared tests were used to evaluate the differences between groups.Results
Baseline characteristics between each quartile were similar (table 1). Mean duration of implant increased with quartile, but serum cobalt was not correlated with duration of prosthesis (r=0.135, p=0.162). Patients in the highest quartile of serum cobalt levels had larger left ventricular end-diastolic and end-systolic dimensions (50.8 vs. 44.7 mm, p=0.018; 31.6 vs. 25.8 mm, p=0.014 respectively) and indexed end-diastolic and end-systolic volumes by Simpsons method (42.0 vs. 32.8 ml/m2, p<0.001; 16.4 vs. 11.7 ml/m2, p<0.001 respectively) compared with those in the lowest quartile (Figure). They also had larger indexed left atrial volumes (Ln volume 3.2 vs. 2.9, p=0.006). However, there was no difference in wall thickness, ejection fraction, left ventricular outflow tract velocity time integral, mitral valve E and A waves, medial and lateral tissue Doppler magnitude, peak longitudinal strain or strain rate, or time to peak strain.Conclusions
Subjects with the highest serum cobalt levels have larger left ventricular and left atrial chamber sizes than those with the lowest serum levels. However, no difference in even sensitive markers of systolic or diastolic function were found. Left ventricular measurements are part of a range and therefore it is not possible to specify a cut-off for abnormality in these people. Whilst our findings do not suggest a clear relationship between serum ion levels and basic echocardiographic parameters of left ventricular systolic and diastolic dysfunction, we cannot exclude an idiosyncratic response as described in isolated case reports. We therefore advise that patients with MOM hip replacements undergo serial cardiology follow up.