The interpretation of metal ion concentrations and their role in clinical management of patients with metal-on-metal implants is still controversial. We questioned whether patients undergoing hip resurfacing with no clinical problems could be differentiated from those with clinical (pain, loss of function) and/or radiographic (component malpositioning, migration, bone loss), problems based on metal ion levels, and if there was a threshold metal level that predicted the need for clinical intervention. Furthermore, we asked if patient and implant factors differed between these functional groups. We retrospectively identified 453 unilateral and 139 bilateral patients with ion measurements at minimum followup of 12 months (mean, 4.3 years; range, 1-12.9 years). Patients were designated as well functioning or poorly functioning based on strict criteria. The acceptable upper levels within the well-functioning group were determined from the 75th percentile plus 1.5x interquartile range. The sensitivity and specificity of these levels to predict clinical problems were calculated. Well-functioning group ions were lower than the poorly functioning group ion levels. The acceptable upper levels were: chromium (Cr) 4.6 mu g/L, cobalt (Co) 4.0 mu g/L unilateral and Cr 7.4 mu g/L, Co 5.0 mu g/L bilateral. The specificity of these levels in predicting poor function was high (95%) and sensitivity was low (25%). There were more males in the well-functioning group and more females and smaller femoral components in the poorly functioning group. Metal levels higher than these proposed safe upper limits can predict problems with metal-on-metal resurfacings and are important parameters in the management of at-risk patients. Level II, diagnostic study. See Guidelines for Authors for a complete description of levels of evidence.