Elevated serum chrome (sCr) and cobalt (sCo) concentrations are associated with local tissue adverse reactions to metal debris following metal-on-metal hip resurfacing (MoM-HR). Serum metal ions <2 µg/l are probably of little clinical relevance and a pragmatic "safe" threshold <5 µg/l has been suggested.The primary aim of this study was to evaluate if a careful selection of patients combined with optimal implant positioning could eliminate cases with "unsafe" serum metal ion levels. A secondary aim was to study the association between different risk factors and having Co and/or Cr levels >5 µg/l. This is a retrospective, single-institution cohort study of 410 consecutive patients operated on with a Birmingham Hip Resurfacing (BHR) implant between 2001 and 2014. 288 of these had a unilateral MoM-HR, pelvic and true lateral radiographs, and a related sCo and sCr sample, and were included in the final analysis. They were allocated to either a presumed "optimal group" consisting of only men aged <60 years old, with femoral head component >48 mm diameter, and with a cup positioned within Lewinnek's safe zones, or a "suboptimal group" consisting of the remaining patients. Fisher's exact test and multiple logistic regression analyses were performed. In the optimal group 48% (47/97) had serum metal ions >2 µg/l and 8% (8/97) >5 µg/l compared to 61% (116/191) and 18% (34/191) in the suboptimal group, p = 0.059 and p = 0.034 respectively. Acetabular cups with an anteversion <5 degrees had the highest odds ratio, 6.5 (95% CI, 3.0-14.3), of having sCo and sCr concentrations exceeding 5 µg/l. A well oriented BHR acetabular component in a presumably "optimal" patient reduces the risk of having elevated serum metal ions but does not eliminate it. Insufficient cup anteversion seems to be the strongest associated factor of elevated serum metals.