Modified trapdoor procedures using autogenous tricortical iliac graft without preserving the broken cartilage for treatment of osteonecrosis of the femoral head: a prospective cohort study with historical controls

Abstract

Background: The aim of the present study was to investigate clinical and radiological outcomes of autologous tricortical iliac grafting performed through a window created at the femoral head without suturing the opened articular cartilage for the treatment of osteonecrosis of the femoral head (ONFH), called modified trapdoor procedures.

Materials and methods: A total of 59 consecutive patients (67 hips; 36 males and 23 females) with ONFH were included in this study, which was conducted from April 2009 to March 2012. Patients' age ranged from 27 to 46 years old, with a mean age of 36.3 years. Harris hip scores (HHS) were used to evaluate hip function pre- and postoperatively. Anteroposterior and frog-position X-rays and magnetic resonance imaging (MRI) were conducted to assess lesion location, size, and ARCO stage. Clinical failure was defined as score < 80 points or treatment by total hip arthroplasty (THA). Radiographic failure was defined as a > 3 mm of collapse in the hip. This group was retrospectively matched according to the ARCO stage, extent, location, etiology of the lesion, average age, gender, and preoperative Harris hip score to a group of 59 patients (67 hips) who underwent the "light bulb" approach between March 2007 and April 2009.

Results: Mean follow-up was 91.2 ± 13.6 months (range, 75-115 months). Mean HHS was 91.3 ± 4.5, compared with 83.1 ± 4.5 in the "light bulb" cohort at the 6-year follow-up examination (P < 0.001). At the 6-year follow-up, for modified trapdoor procedures, five hips (8.5%) were classified as clinical failure, and three hips underwent total hip arthroplasty; seven hips were classified as (10.4%) radiographic failure. The clinical and radiographic failure of the hips treated with the modified trapdoor procedure was significantly lower compared to the hips treated with the "light bulb" procedure (P < 0.05). Survival of the joint was not significantly related to the location of the femoral head lesion between two groups; however, better clinical and radiographic results were observed in modified trapdoor procedures with size C and the ARCO stage III.

Conclusion: The present study demonstrated superior midterm clinical results in ONFH with the use of autologous tricortical iliac block graft through a femoral head window, without suturing the opened articular cartilage. The femoral head-preserving procedure was superior compared to the "light bulb" procedure treatment in patients with postcollapse osteonecrosis and large lesion.

Keywords: Bone graft; Cartilage; Hip; Osteonecrosis; Outcome; Surgical procedure.

Fig. 1

Bone grafting through a femoral head window. a Exposure of the femoral head without dislocation and creation of a cortical window in the femoral head and removal of all visible necrotic bone. b Autogenous iliac crest struts trimmed into its optimum shape. c Placement of a tricortical iliac bone graft in the groove and fixation with a screw

Fig. 2

A 32-year-old man with osteonecrosis of the femoral head was treated with our modified trapdoor procedure. Anterior–posterior X-ray (a) and frog-position X-ray (b) show femoral head necrosis with segmental collapse. Coronal CT confirmed ONFH with collapse (c). Coronal T1 (d) and STIR (e) showed ONFH with edema. Postoperative radiography (f) showed necrotic bone that had been curetted and replaced with tricortical iliac block graft. Coronal CT (g) showed that necrotic bone had been curetted and replaced with a tricortical iliac block graft. The graft was in accordance with the contour of the femoral head. Anterior–posterior X-ray (h) and frog-position X-ray (i) obtained 1 year postoperatively show that the graft had healed to the host bone without collapse. Anterior–posterior X-ray (j) and frog-position X-ray (k) obtained 4 years postoperatively show that the graft had healed to the host bone, without collapse. Anterior–posterior X-ray (l) and frog-position X-ray (m) obtained 8 years postoperatively show that the contour of the femoral head was intact without collapse. Coronal T1 MR images (n) obtained 4 years postoperatively showed that the contour of the femoral head was intact, with the replacement of the necrotic bone by a viable bone, and normal cartilage at the femoral head. Axial STIR MR images (o) obtained 4 years postoperatively show that the contour of the femoral head remained intact; a portion of the necrotic bone has been replaced with a viable bone