The rationale for head-preserving procedures is the desire to provide sufficient and long-lasting support to necrotic subchondral bone and cartilage [38], in order to prevent collapse and subsequent osteoarthrosis of the joint. A tantalum rod was used to provide direct mechanical support[16]; a bone graft was aimed to replace dead bone with viable bone[12–14]. Various types of stem cells and/or biofactors were used to facilitate bone formation and remodeling, alone or in combination with other methods[15, 19]. Vascularized bone graft is deemed have better effect than nonvascularized bone graft, is technically demanding. Nonvascularized bone graft procedures are currently the most popular methods for preserving the necrotic head and/or deferring THA. Three surgical techniques are common used: (1) the Phemister technique: grafting via a core decompression tract from the greater trochanteric area; (2) “lightbulb” procedure: grafting through a femoral neck or femoral head-neck junction window; (3) trapdoor procedure: grafting through a femoral head window.
Keizer et al.[21] described a cohort of cases that underwent autogenous cancellous bone grafting via a core tract (Phemister technique), after a mean of 7 years of follow-up; 34 of 78 hips (44%) required additional surgery. This procedure is now out of favor, as the necrotic bone was not sufficiently debrided, resulting in insufficient solid support for subchondral bone [17, 20]. In 1998, Mont et al. [18] introduced use of the trapdoor procedure, which involves creating a window at the femoral head. In that study, 20 of 24 Ficat stage III hips (83%) had good or excellent outcomes. This procedure involves disrupting the integrity of the femoral head cartilage, which has been shown to prevent healing. More and more clinicians are choosing to perform bone grafting through a femoral neck or head-neck junction window without disrupting the weight-bearing cartilage (lightbulb procedure). Rosenwasser et al.[23] chiseled out a femoral head-neck junction window and observed a success rate of 81% at 12 years.
All of these procedures tried to maintain the articular cartilage over the necrotic zone. In our study, the clinical survival rate of the procedure without suturing the opened articular cartilage was 92.5% for a minimum of six years, significantly superior to lightbulb procedure. The lightbulb procedure did not damage the articular cartilage. However, it encountered two difficult technical problems. First, it is difficult to completely remove necrotic bone. Actually the necrotic subchondral bone is hard to heal with bone graft, marked collapsed is still observed during conversion to THA [18, 27–31]. Second, we could not perform perfect bone grafting to fill the canal. even if a sophisticated operation to restore the spherical shape of the femoral head is performed, some mild joint incongruence would be present, and it would be impossible to maintain the sphericity without collapse for longer periods. The collapsed cartilage and subchondral bone may serve as floating body and is harm for health. With partly cartilage defect, the function of hip was still well. Femoral head with part chondral lesion detected during arthroscopic surgery still functioned well with minimum 5 years follow-up [34]. These findings indicate that mechanical failure, rather than cartilage degeneration, is the main cause of pain. So we modified the trapdoor procedure. After thorough debridement, an autologous tricortical iliac block graft combined with morselized bone was implanted. Broken cartilage and necrotic subchondral bone was no longer replanted. Compared with previous studies [12–26], this report includes superior results. The rate of clinical success (good or excellent outcome) was 92.5% for 67 hips.
Larger lesions and more advanced linear collapse increase the relative risk of failure. Wang et al.[22] in a retrospective study of 110 patients, who underwent 138 light bulb procedures with a mean follow-up of 25.37 months, demonstrated that “light bulb” technique should not be performed once if subchondral collapse is present. Similarly, Sotereanos et al.[39] showed that the probability of conversion to total hip arthroplasty within an average of 5.5 years after Autogenous Grafts was 38% for stages III and IV hips. The rate of collapse for modified trapdoor procedures was only 18.2% or 9.5% among ONFH hips of ARCO III or size C. When compared with light bulb procedures, modified trapdoor procedures is also a worthwhile procedure in patients with postcollapse osteonecrosis. It doesn't care about the cartilage lesion of ARCO III and thoroughly remove dead bone tissue for size C. Autogenous tricortical iliac provides sufficient support, filles effectively necrotic cavity, and demonstrates incorporation of the graft to the recipient bone of the femoral head.
The first advantage of our technique, compared with the traditional trapdoor procedure, was that a tricortical iliac block graft fixed with one or two screws can more easily heal with host bone and provide good structural support[18, 32]. The autogenous iliac crest transplantation is an effective treatment. However, there is a risk of bone flap loosening. Therefore, screw fixation is needed to increase the stability of the bone flap. The second advantage was our procedure did not involve suturing the opened articular cartilage, so there were no issues related to necrotic subchondral bone healing with graft bone or cartilage–cartilage interface healing. Floating cartilage slices, just as osteochondritis dissecans, may be another reason for unsatisfactory results. Thirdly, use of the anterior approach without dislocation of the hip, results in less damage to posterior structures, with excellent exposure and preservation of the blood supply[40, 41]. Finally, this procedure has the advantages of being minimally invasive, technically simple, short in operative duration, associated with fewer complications, and associated with no adverse effect on the procedure of late arthroplasty.