<p><b>OBJECTIVE</b>To investigate and the clinical effect of S-ROM modular stem in femur reconstruction in hip revision arthroplasty.</p><p><b>METHODS</b>From January 2008 to January 2016, 21 patients received revision hip arthroplasties using S-ROM stems. There were 5 males and 16 females with an average of 48.33 years old(range, 29 to 73 years old). There were 13 cases caused by aseptic loosening, 4 cases by infection, 2 cases by nonunion of sub-tuberosity osteotomy, 1 case for repeated dislocation, 1 case for traumatic great trochanter fracture. Primary THA reasons:12 cases for DDH(9 cases for Crowe IV), 5 cases for femoral neck fracture, 2 cases for necrosis of femoral head, 2 cases for proximal femoral deformity caused by early infection. The femur bone defects included Paprosky II in 11 cases, IIIa in 9, and IIIb in 1. Harris hip score , pain score and hip flexion were recorded before and after operation. The subjective satisfaction was recorded at the last follow-up.</p><p><b>RESULTS</b>The operation time and blood loss were 189 min(125 to 290 min) and 867 ml (200 to 2 000 ml). At the final followup, the pain score improved from (17.14±9.56) points preoperatively to (41.71±2.03) points (=11.42,=0.00). The function score improved from (24.01±11.02) points preoperatively to (49.95±5.38) points (=9.73,=0.00). Harris hip score improved from (41.15±14.81) points preoperatively to(91.67±5.83) points(=15.33,=0.00). The degree of hip flexion increased from (93.10±27.27) points preoperatively to (121.90±16.62) points at the last follow-up (=4.59,=0.00). The mean subjective satisfaction was 9.48(10 points system), 14 of which were completely satisfactory. The last follow-up hip X-ray showed 21 cases of bone ingrowth, and other femoral stem without loosening or sinking sign significantly. There were 5 cases with bone anchor syndrome around proximal femoral cuff. There were 5 cases of proximal femur fracture, 3 cases of lesser trochanter fracture, 2 cases of greater trochanter fracture intra-operatively. Due to femoral canal stenosis, there were 4 cases of pre-tied wire at lesser trochanter to prevent fractures. There was 1 case of traumatic femoral fracture around stems with a distal oblique fracture, open reduction and locking plate fixation was performed. Other patients had no nerve stretch injury, dislocation, infection and lower limb deep vein thrombosis and other complications at the final follow-up.</p><p><b>CONCLUSIONS</b>S-ROM prosthesis has satisfactory results in hip revision arthroplasty with Paprosky II and III femoral defects. Especially for patients with Crowe IV DDH and other proximal femoral deformities, it is possible to adapt to the medullary cavity morphology. Excellent initial stability, less complications and long-term biological fixation can be achieved with S-ROM in femur revision.</p>