To investigate the possible causes of internal fixation failures in intertrochanteric fractures with proximal femoral nail (PFN) and proximal femoral nail anti-rotation (PFNA), as well as the strategy of hip arthroplasty and clinical efficacy.

From January 2010 to December 2017, a retrospective study was done on 41 cases, 15 males and 26 females, aged from 49 to 87 years, average(71±7)years, who received salvage surgery of hip arthroplasty for intramedullary internal fixation failure for intertrochanteric fractures in the ward of bone and joint surgery, the first affiliated hospital of Jinan University. Inclusion criteria: after intramedullary internal fixation for unilateral intertrochanteric fractures; symptoms, signs, and imaging examinations confirmed that the internal fixation failed; the patient had the willingness to receive salvage surgery. Exclusion criteria: the patient was unable to tolerate reoperation; salvage surgery was not hip arthroplasty; confirmed or suspected infection. According to Evans classification, four patient was with type Ⅲ, 14 were type Ⅳ, and 23 were type Ⅴ. Among 41cases, 12 cases were performed with bipolar hemiarthroplasty, 29 cases were with total hip arthroplasty. Seven cases were performed with cemented standard stem, 15 cases were with cementless standard metaphyseal locking stem, 19 cases were with cementless diaphyseal locking stem. The independent sample Student’ t-test was applied in the comparison of surgical related indicators and the preoperative and postoperative scores of hip function. The paired sample Student’ t-test was applied in the comparison of preoperative and postoperative scores of the same group.

The operation time, intraoperative blood loss volume, and postoperative drainage volume in the total hip arthroplasty group were greater than those in the bipolar hemiarthroplasty group; the differences were statistically significant (t =9.548, P<0.01; t=4.374, P<0.05; t=10.048, P<0.01). A total of 41 cases were followed-up from eight to 86 months, (43±16) months on average. Hip joint Harris score was elevated from preoperative (43±6) to (83±4). The Harris scores at the last follow-up of the bipolar hemiarthroplasty group and the total hip arthroplasty group were significantly higher than those before the salvage surgery (t=11.532, P <0.01; t=14.713, P<0.01). The satisfaction rate of hip arthroplasty was 87.8%.

Analyzing the characteristics of PFN/PFNA failure cases and selecting appropriate procedure of total hip arthroplasty or hemiarthroplasty salvage surgery can effectively improve the patients’ symptoms and the function of hip joint, which shows a satisfactory outcome of the mid-term clinical efficacy.