Introduction. The anatomic abnormalities are observed in developmental dysplasia of the hip (DDH) and it is challenging to perform the total hip arthroplasty (THA) for some DDH patients. If acetabular cup was placed at the original acetabular position in patients with high hip dislocation, it may be difficult to perform reduction of hip prosthesis because of soft tissue contracture. The procedures resolving this problem were to use femoral shortening osteotomy, or to place the acetabular cup at a higher cup position than the original hip center. Femoral shortening osteotomy has some concerns about its complicated procedure, time consuming, and risk of non-union. Conversely, implantation of the acetabular cup at the higher cup position may eliminate these shortcomings and this procedure is considered to be preferred if possible. However, the criteria of cases without femoral shortening osteotomy are not clear. In this study, we retrospectively analysed the clinical outcomes of patients performed THAs for high hip dislocation, and clarified the adaptation of THA with or without femoral shortening osteotomy. Methods. We included a total of 65 hip joints from 57 patients who underwent primary THA using Modulus stem for high hip dislocation from November 2007 to December 2015 at our institution. The mean follow up period was 5.2 years (2 – 10 years). The mean age at surgery was 65.4 years (Table 1). Thirty seven hips were classified as Crowe III, and twenty eight hips as Crowe IV based on Crowe classification. We classified patients into two groups based on the use of femoral osteotomy. Then, we compared the surgical time, blood loss, Japanese Orthopaedic Association (JOA) Score as clinical outcomes, preoperative position of the greater trochanter, the cup position, and complications between two groups. The position of the greater trochanter was measured the height of the tip of greater trochanter from the inter teardrop line. The cup center position was assessed by measuring the distance between the cup center and ipsilateral tear drop. Receiver operating characteristic (ROC) curves were plotted for deciding the cut-off value for the height of the greater trochanter. The cut-off value presented the maximum sensitivity and specificity was determined. Results and Discussion. Fifty three THAs were operated without femoral shortening osteotomy, and twelve THAs were performed with femoral shortening osteotomy. The surgical time was significantly longer in the osteotomy group than the non-osteotomy group. The mean height of the tip of the greater trochanter were 53.2±11.4mm in the non-osteotomy group and 92.2±19.7 mm in the osteotomy group (Table 2). The cut-off value of the height of greater trochanter evaluated from the ROC curve analysis was 69.5mm (Fig.1). There were no significant differences in clinical score between two groups. More ratio of revisions and fractures were observed in the osteotomy group with significant differences. Conclusion. There were significant differences in postoperative complications in osteotomy group compare to non-osteotomy group. In cases with a greater trochanter tip height of 69.5 mm or less, it may be considered to avoid femoral shortening osteotomy