The most important finding of this study is that medial malleolar triplane osteotomy with AOT from the non-weight-bearing area of the talus for OLTs achieved significant pain reduction and satisfactory ankle function with few complications after a mean follow-up of more than 3 years.
Throughout evolution, to satisfy the needs of flexible ankle joint movement in human upright walking and activities, the talus, located in the center of the ankle, has gradually acquired a unique anatomy in that the majority of its surface (approximately 60%) is covered with articular cartilage [10], and there is little soft tissue (including nutritional arteries) attachment in the body of the talus. On the other hand, the vessel spreading in the talus has failed to achieve a rich blood supply, forming a "watershed" of blood supply between single areas [11]. This anatomical limitation has been implicated in the high risk of posttraumatic osteonecrosis and OLTs [4, 11]. Nondisplaced or mild symptomatic OLTs are often initially treated with a nonoperative approach. Conservative treatments of OLTs include activity modification, protected weight-bearing, rehabilitation, bracing, and nonsteroidal anti-inflammatory drugs. Clinically, a systematic review by Verhagen [12] et al. showed that approximately 45% of patients with OLTs reported successful outcomes with conservative treatment. It is worth noting that the success of nonsurgical treatment for OLTs is mostly in the pediatric population, but it is limited in the adult population, which may be attributable to the relatively active repair ability of articular hyaline cartilage in young people. The operative treatment of OLTs can be broadly divided into two categories: reparative procedures, including bone marrow stimulation (BMS); cell-based therapies; and replacement procedures, including AOT, osteochondral allograft transplantation, periosteum bone transplantation and prosthetic implantation. At present, there is no unified standard for surgical indications for OLTs. It is generally believed that patients with small OLTs less than 15 mm in diameter or less than 150 mm2 in area can obtain satisfactory clinical effects through BMS [4, 13], whereas patients with large OLTs or failed bone marrow stimulation should choose replacement procedures [4, 14]. A previous study [15] showed that the main pathway of BMS for OLTs is through a series of inflammatory reactions and cell differentiation to form new scar tissue (mainly type I collagen) to repair the talus defect area. However, some studies [16, 17] that used MRI to evaluate the imaging results of BMS in the treatment of OLTs found that the lesions could not be completely filled, the quality of scar tissue involved in the repair of articular surface was poor, and it was difficult to achieve complete fusion with primary cartilage. This may be related to the fact that the osteocysts under the cartilage of OLTs are not filled with bone after BMS treatment, resulting in the lack of effective physical support and nutritional supply for the surface tissue. Therefore, we believe that the limitation of osteochondral transplantation is not simply determined by the size of the lesion. In other words, even if the area of an articular cartilage lesion is relatively large (> 150 mm2) and there is no cystic change or the cyst is very shallow, the use of BMS can also obtain sufficient scar tissue filling and repair. In contrast, for OLTs with significant cystic lesions, even if the area of cartilage lesions is not large, the OLT is still suitable for osteochondral transplantation because of the continuous effective physical support and sustainable nutritional supply for the repaired surface tissue through the approach. In the current study, we took a cystic depth of OLTs greater than 6 mm as an indication for AOT. On the other hand, patients with OLT areas greater than 144 mm2 were excluded to avoid the adverse effect of using multiple bone columns for transplantation.
AOT was mainly previously applied to the surgical treatment of osteochondral lesions of the knee joint [18]. The operation is to repair the defect of the articular cartilage surface by transplanting the autogenous bone-cartilage column into the lesions. At the same time, the subchondral bone in the graft fully fills the cavity and finally integrates with the original bone, providing necessary and continuous nutritional and physical support for the cartilage. The goal of AOT is to implant a graft that is similar in both mechanical and biological properties to that of the patient’s native hyaline cartilage. Several retrospective case series have demonstrated positive results with AOT [6, 14, 19, 20]. Imhoff et al. [21] observed significant long-term improvements in mean AOFAS, VAS, and Tegner activity scores in 26 patients. At a mean follow-up time of 7 years, 18 patients indicated they were very satisfied, 4 satisfied, 3 neutral, and 1 moderately unsatisfied with the procedure. Shimozono et al. [20] systematically reviewed 11 mid-term clinical follow-up studies on AOT treatment for OLTs, including 500 cases in ankle joints, with an average follow-up time of 62.8 months. They found that the average AOFAS ankle and hind foot scores increased from 55.1 preoperatively to 86.2 at the last follow-up, with an excellent rate of 87%. This is similar to the clinical results of the current study reported by us. Autologous grafts are most commonly harvested from the ipsilateral knee, specifically from the lateral femoral condyle or the intercondylar notch. The primary concern with osteochondral autograft transfer is donor site morbidity, as reported by several studies [22]. Shimozono et al. [20] also found that the general incidence of complications after AOT was 10.6%, the most important of which was related to the donor site, and the incidence of donor site morbidity was approximately 3.6%. Fraser et al. [23] reported 40 OLT patients in whom the lateral femoral condyle of the same knee joint was used as the donor site for AOT. After an average follow-up of 24 months, they found that approximately 12.5% of the patients had discomfort symptoms at the donor site, including a painful knee joint and chondromalacia patella during high-intensity exercise. In our study, only 1 case of posterior tibial tendon discomfort caused by the use of too many internal fixation screws after AOT was found. The incidence of complications was 4.3%, which was significantly lower than that reported in previous literature, and there were no donor-site-related complications. This may be related to the selection of the vertical surface of the talus as the donor site. This surgery did not increase the operation time and avoided potential discomfort at the donor site. On the other hand, we used the medial cortical bone of the distal tibia to backfill the talus donor site. The autogenous bone had a strong healing ability, and there was no risk of allogeneic infection. The last follow-up CT scan showed that all the donor-site defects achieved good healing, which also reduced the damage to the stress structure of the talus and the potential risk factors for complications.
In this study, the location of OLTs was region IV in 19 patients, region V in 1 patient, and region VII in 3 patients; thus, all the lesions were mainly concentrated in the posterior medial region of the talus where the OLTs usually needed medial malleolar osteotomy for exposure and operation. In general, all kinds of medial malleolar osteotomies not only facilitate the exposure of and operation at the posterior medial area of the talus but also bring additional damage to the medial sensitive area of the ankle joint and the risk of remaining discomfort. Kreuz et al. [24] believed that the total adverse effects of OLT may occur in 3 stages: (1) in the short term, the osteotomy itself can damage adjacent important anatomical structures, such as the posterior tibial tendon, tibial nerve and posterior tibial artery; (2) in the mid-term, the malunion or nonunion caused by poor reduction or fixation of medial malleolus can cause adverse effects; and (3) in the long term, the accelerated degeneration of local articular cartilage can lead to ankle arthritis. Bull et al. [25] reported that the rate of healing-induced deformity after internal malleolar biplane chevron osteotomy was up to 30%. After 2.4 years of follow-up, it was found that approximately 24% of patients had swelling at the osteotomy site, and the residual rate of pain in the internal malleolus was up to 60%. The main reasons were related to poor reduction and internal fixation failure after osteotomy. In the current study, we designed a new medial malleolus osteotomy procedure, a three-plane osteotomy of the medial malleolus, which has the following advantages: (1) By adding the coronal osteotomy of the medial malleolus on the biplane ladder osteotomy, the trauma caused by the osteotomy of the medial malleolus is reduced, the bleeding is reduced, and the adhesion of the posterior medial tendon of the ankle is avoided. (2) By retaining the integrity of the posterior medial bone of the medial malleolus, the risk of injury to important anatomical structures such as the posterior medial neurovascular tendon caused by complete osteotomy of the medial malleolus, such as oblique osteotomy, is avoided. (3) After osteotomy in this way, the osteotomy section at the proximal end of the tibia presents a three-dimensional square concave shape, which is convenient to achieve the complete anastomosis of three planes (the sagittal plane, coronal plane and horizontal plane) under the direct vision when the distal medial malleolus is restored. In addition, the medial malleolus has high stability after reduction, and reduction is not easily lost during fixation. (4) When hollow screws are used to fix the inner malleolus, vertical compression and firm fixation of screws on three sides is allowed, permitting early functional exercise and weight-bearing activities. In the current study, with the triplane osteotomy method we used, malunion or nonunion of the medial malleolus was not observed, and there was no obvious sign of joint degeneration at the medial malleolar osteotomy site. The average time from osteotomy to full weight-bearing activity was 8.1 weeks, earlier than the medial malleolus healing time reported in the literature [8, 26, 27]. During the last follow-up, the ROM of the affected ankle joint was 12.1° dorsiflexion and 44.7° plantar flexion, and the degree of activity was basically restored to the preoperative level, without obvious ankle joint or tendon adhesion, which may be related to the early active rehabilitation training that we allowed.