OBJECTIVE The aim of the study was to compare radiation exposure and image quality between dedicated computed tomography (CT) protocols for preoperative total hip arthroplasty (THA) planning. METHODS Three protocols with automated tube current modulation using 64-slice (n = 177) and 128-slice CT scanners without (n = 129) and with automated tube voltage preselection (n = 84) were compared. RESULTS All 390 CTs were of sufficient quality for THA planning. Mean DLP was 235.0 mGy*cm (effective dose 2.8 mSv). Lowest radiation exposure (2.5 mSv) was seen with automated voltage preselection and the algorithm's selection was 100 kV (90.5% of patients) and 120 kV. Lowest image noise was seen in the highest dose group (3.1 mSv, 128-slice CT fixed tube voltage). A significant difference in cortical bone radiodensity was seen between 100 kV and 120 kV (P < 0.0001). CONCLUSIONS Preoperative pelvic CT for THA planning is possible with very low radiation dose and reliable quality. Automated voltage preselection further decreases the effective dose by 18.2%.