Purpose Displaced femoral neck fractures in healthy elderly patients have traditionally been managed with hemiarthroplasty (HA). Recent data suggest that total hip arthroplasty (THA) may be a better alternative. Methods A systematic review of the English literature was conducted. Randomized controlled trials comparing all forms of THA with HA were included. Three authors independently extracted articles and predefined data. Results were pooled using a random effects model. Results Eight trials totalling 986 patients were retrieved. After THA 4 % underwent revision surgery versus 7 % after HA. The one-year mortality was equal in both groups: 13 % (THA) versus 15 % (HA). Dislocation rates were 9 % after THA versus 3 % after HA. Equal rates were found for major (25 % in THA versus 24 % in HA) and minor complications (13 % THA versus 14 % HA). The weighted mean of the Harris hip score was 81 points after THA versus 77 after HA. The subdomain pain of the HHS (weighted mean score after THA was 42 versus 39 points for HA), the rate of patients reporting mild to no pain (75 % after THA versus 56 % after HA) and the score of WOMAC (94 points for THA versus 78 for HA) all favored THA. Quality of life measured with the EQ-5D favored THA (0.69 versus 0.57). Conclusions Total hip arthroplasty for displaced femoral neck fractures in the fit elderly may lead to higher patientbased outcomes but has higher dislocation rates compared with hemiarthroplasty. Further high-quality randomized clinical trails are needed to provide robust evidence and to definitively answer this clinical question.