Twenty million North Americans have cobalt-chrome arthroprosthetic components, and 1 million have metal-on-metal hip replacements. 1 Cobalt is a mitochondrial toxin-encephalopathy and cardiomyopathy (cobaltism) may occur from iatrogenic, industrial, dietary, or arthroprosthetic cobalt exposure. 2,3 In unexposed populations, the 95th percentile of cobalt levels in urine and blood are 1 part per billion (ppb) and 0.4 ppb, respectively. 3 Wear and corrosion of cobalt-chrome joint implantations can result in periprosthetic tissue inflammation or necrosis, also known clinically as adverse reactions to metallic debris. 1,4 Periprosthetic cobalt-chrome metallosis is disseminated systemically and may result in arthroprosthetic cobaltism. 1,3,4 Systemic cobalt dissemination can result in brain hypometabolism and atrophy; patients with levels of cobalt in blood as low as 1.1 ppb and in urine as low as 4.1 ppb are reported as having cobalt encephalopathy. 5,6
MethodsThis cohort study analyzed redacted data from 1 orthopedic clinic. The institutional review board at the University of Alaska Anchorage designated the study to be exempt, and patients provided written informed consent for this study. Patients with a suspected hip, knee, or shoulder replacement containing cobalt-chrome were screened with a spot screening of cobalt in urine. Levels of cobalt in blood were then determined for patients with cobalt levels in urine 1 ppb or higher. Patients were designated as cobalt-positive if cobalt levels in urine were 1 ppb or higher or if levels in blood were 0.4 ppb or higher. The reference laboratory test results threshold for reporting cobalt in urine and blood are generally 1 ppb and 0.5 ppb, respectively. Patients with cobalt levels below the reporting threshold in either were assigned half the threshold value. All P values were from 2-sided tests and results were deemed statistically significant at P < .05. Analysis was conducted using Prism version 9.1.1 (GraphPad). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
ResultsA total of 241 patients were screened (mean [SD] age at time of cobalt determination, 68.1 [9.3] years; median duration of exposure to any cobalt-chrome implantation, 11.6 years [range, 2.0-33.2 years]; 117 [48.5%] women); 138 (57%) tested cobalt-positive (mean cobalt level in urine, 12.6 ppb; median, 1.2 ppb). Eleven subjects had no cobalt-chrome implantation (because zirconia and cobaltchrome femoral-heads are indistinguishable radiographically), none of whom were cobalt-positive.Paired cobalt levels in urine and blood (144 patients) correlated significantly (cobalt in blood = 0.25 × cobalt in urine; P < .001). Paired joint-fluid and urine cobalt (57 patients) correlated significantly (joint fluid = 20 × cobalt in urine; P < .001). Patients were classified by type, location, number, and brand of their implantation (Table ). These classes were risk-grouped based on mean cobalt levels in urine (ie, extreme, greater than 20 ppb; high...