To evaluate the effectiveness of centralizing device, three groups of patients on whom cemented total hip arthroplasty with three types of implant was performed respectively were analyzed and compared; SL stems without centralizing device were used in Group I, Interlok stems with central sleeve in Group II, and MS-30 stems with distal centralizer in Group III .The size of each group was 20. Cement mantle thickness, the amount of displacement of stem tip, and axis alignment of femoral stem on A-P radiograph checked at 2 weeks after operation were assessed. More optimal cement mantle thickness (3~6mm) was achieved in Group II (P=0.046) and Group III (P=0.01) than in Group I. Centralizing effect of the distal femoral stem in the intramedullary canal was significantly better (P<0.05) in Group III ( 4.55% ) than in Group I (11.60%) and Group II (12.35%). Neutral alignment of femoral stem was achieved in 15% of Group I, in 90% of Group II, and in 95% of Group III. Compared to stem without centralizing device, cemented femoral stem with distal centralizer resulted in 1) significantly more centralized distal femoral stem, 2) significantly more neutrally aligned femoral stem,and 3) significantly more optimal cement mantle thickness at zone V. Compared to stem without centralizing device, cemented femoral stem with central sleeve resulted in 1) significantly more neutrally aligned femoral stem, and 2) significantly more optimal cement mantle thickness at zone VI.