Reconstruction after Talar Tumor Resection: A Systematic Review

This systematic review investigated the functional outcomes and complications of reconstruction methods after talar tumor resection. A systematic search of PubMed, Embase, and the Cochrane Central Register of Controlled Trials databases identified 156 studies, of which 20 (23 patients) were ultimately included. The mean Musculoskeletal Tumor Society scores in the groups reconstructed using tibiocalcaneal fusion (n = 17), frozen autograft (n = 1), and talar prosthesis (n = 5) were 77.6 (range 66–90), 70, and 90 (range 87–93), respectively. Regarding complications, sensory deficits were observed in one patient (6%) and venous thrombosis in two patients (12%) in the tibiocalcaneal fusion group, while osteoarthritis was observed in one patient (100%) in the frozen autograft group. No complications were observed in the talar prosthesis group. Reconstruction with talar prosthesis seems preferable to conventional tibiocalcaneal fusion after talar tumor resection because it offers better function and fewer complications. However, as this systematic review included only retrospective studies with a small number of patients, its results require re-evaluation in future randomized controlled trials with larger numbers of patients.

To investigate the functional outcomes and complications of reconstruction methods after the resection of tumors originating in the talus, we performed this systematic review of studies that reported the postoperative function and complications of reconstruction after resection.

Materials and Methods
This study followed the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-analyses 2020 statement [31]. The study protocol was registered in the UMIN Clinical Trials Registration as UMIN000047105 (http://www.umin.ac.jp/ctr/ index.htm (accessed on 7 March 2022)). The study was conducted in accordance with the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of Nara Medical University (protocol code 2833 and date of approval 27 November 2020). The requirement for written consent from participants at Nara Medical University was waived because an "opt-out" process was used and the retrospective nature of the study.

Eligibility Criteria
Only studies reporting on the postoperative function and complications after talar tumor resection followed by reconstruction were included. Patients who underwent resection alone for talar tumors without reconstruction or amputation were excluded. We also excluded cases for which the postoperative function and complications were not specified. Patients with talar tumors treated with curettage or partial resection were excluded. The data were extracted for the ankle joint range of motion (ROM) and American Orthopedic Foot and Ankle Society (AOFAS) ankle-hindfoot score, in which 40 points represents pain and 60 points represents ankle-hindfoot function and alignment, MSTS score [32], leg length difference, and complications. Only English and Japanese studies were included, and no restrictions were placed on the year of publication.

Literature Search and Study Selection
The literature was searched on 19 January 2022, using a systematic search strategy in PubMed, Embase, and the Cochrane Central Register of Controlled Trials (Table S1). The reference lists of the retrieved studies were manually searched to identify other relevant studies.

Data Collection and Presentation
Two authors (S.T. and A.K.) independently selected the studies and extracted the data. In cases of disagreement, agreement was reached or a third author was consulted. Data were collected using a data collection sheet and included: author, year of publication, journal name, study type, patient age, patient sex, tumor histological type, reconstruction method, ankle joint ROM, postoperative function of the affected limb (AOFAS and MSTS scores), leg length difference, complications, oncological outcome, and postoperative follow-up period.

Data Summary
Tables 1 and 2 summarize the data extracted from the included studies.    Table 3 summarizes the age, sex, tumor histology, ankle ROM, AOFAS and MSTS scores, leg length difference, complications, oncologic outcomes, and follow-up period for each reconstruction method (tibiocalcaneal fusion, frozen autograft reconstruction, and talar prosthesis reconstruction).

Assessment of Methodological Quality
Two authors (ST and AK) independently assessed the quality of the included studies. In cases of disagreement, agreement was reached or a third author was consulted. The articles included in the final analysis were independently assessed according to the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) tool to assess the quality of non-randomized studies in meta-studies [33].

Assessment of Methodological Quality
Two authors (ST and AK) independently assessed the quality of the included studies. In cases of disagreement, agreement was reached or a third author was consulted. The articles included in the final analysis were independently assessed according to the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) tool to assess the quality of non-randomized studies in meta-studies [33].

Demographic Data and Ratio of Patients Who Underwent Reconstruction with Tibiocalcaneal Fusion, Frozen Autograft, or Talar Prosthesis
A total of 23 patients who underwent reconstruction after the resection of tumors originating in the talus were included. Of them, 17 (74%) underwent reconstruction with tibiocalcaneal fusion, 1 (4%) with a frozen autograft, and 5 (22%) with a talar prosthesis (Tables 1-3).

Methodological Quality of Included Studies
Evaluation of the quality of the individual studies using the RoBANS tool showed an overall moderate risk of bias. All 20 included studies showed that "selection of participants" and "confounding variables" were high, whereas "measurement of exposure," "blinding of outcome," "incomplete outcome data," and "selective outcome reporting" were low.

Discussion
Until 2008, tibiocalcaneal fusion was the only reported reconstruction method after talar tumor resection [7,[17][18][19][20]25,27]. In 2009, Sakayama et al. reported a reconstruction method using frozen autografts [8]. In 2014, Harnroongroj et al. reported a reconstruction method using a talar prosthesis [23]. We analyzed studies that reported on the postoperative function and complications after reconstruction following talar tumor resection and compared the postoperative function and risk of complications among the tibiocalcaneal fusion, frozen autograft, and talar prosthesis groups. The talar prosthesis group had a shorter follow-up period but better postoperative function than the other groups. The risk of complications was higher with tibiocalcaneal fusion and frozen autograft reconstruction than with talar prosthesis reconstruction. Therefore, talar prosthesis reconstruction seems preferable to conventional methods, such as tibiocalcaneal fusion, after talar tumor resection.
This study had several limitations. First, all included studies were retrospective and had a patient-background bias. Because talar prosthesis is a relatively new reconstruction method, its postoperative follow-up period was shorter than those of the other reconstruction methods. Although the talar prosthesis reconstruction method showed better postoperative function than the other reconstruction methods, it will be necessary to analyze long-term follow-up data in the future. Harnroongroj et al. used a talar body prosthesis to replace the talus in 33 patients with talar osteonecrosis, talar fracture, or talar tumor [23]. They reported that the AOFAS score of the 10-20 years (n = 8), 20-30 years (n = 11), and 30-36 years (n = 9) follow-up groups was 78, 76, and 76, respectively, which indicated that talar body prosthesis can maintain good function over a long follow-up period [23]. Second, a higher percentage of patients in the tibiocalcaneal fusion group underwent resection of the peritalar bone and the talar tumor. This may have affected the functional outcomes. Reconstruction with a talar prosthesis cannot be performed in patients for whom the peritalar bone must be resected. Therefore, its use is limited to reconstruction after the resection of malignant tumors localized within the talus or intermediate-grade tumors. In contrast, tibiocalcaneal fusion can be performed even in cases in which the peritalar bone must be resected, and its use is more widely indicated than talar prosthesis reconstruction. RCTs can avoid many of these biases by randomly allocating participants to groups. Since no RCTs were identified, well-designed cohorts and observational studies with strong effects may provide reliable information. Third, patients who underwent resection and reconstruction for talar tumors were rare, resulting in a small number of patients being analyzed (only 23 patients). However, at present, this is all the information we have on reconstruction after talar tumor resection, and the results of this systematic review will be useful to orthopedic surgeons treating tumors.
The talar prosthesis maintains ankle joint motion and leg length [23,36]. This is because the implant is usually modeled from a computed tomography scan of the contralateral talus and is anatomically correct for mortise [36]. In addition, reconstruction with a talar prosthesis has a shorter time to full weight bearing (5 weeks) [36]. Taniguchi et al. [36] replaced 55 talus osteonecroses of 51 patients with a total talar prosthesis. They reported that the Japanese Society for Surgery of the Foot ankle-hindfoot scale [37,38] score was a mean 89.4 (range 76-100) after a mean follow-up of 52.8 months [36]. Thereafter, Morita et al. reported that among 18 patients (19 talus) with more than 10 years of follow-up (median 152 months), the median Japanese Society for Surgery of the Foot ankle-hindfoot scale was 97 (interquartile range 87-99.5) [39].
This study revealed that the risk of complications was higher after tibiocalcaneal fusion and frozen autograft than after talar prosthesis reconstruction. Tenenbaum et al. performed tibiotalocalcaneal fusion with intramedullary nails without bone grafting in 14 patients with talus osteonecrosis and noted complications of fatigue fracture in 1 patient (7%), hardware removal in 3 (21%), and superficial infection in 1 (7%) [35]. Hayashi et al. performed osteoarticular frozen autograft reconstruction in 27 patients with bone tumors, including epiphyses, and reported that grade IV osteoarthritis occurred in 12 patients (44%) and infection in 6 (22%) after a mean follow-up of 94.0 months [40].
Harnroongroj et al. reported that talar body prosthesis replacement of the talus in 33 patients with talar osteonecrosis, talar fracture, and talar tumor resulted in a size discrepancy in 2 patients (6%), infection in 1 (3%), and osteonecrosis of the talar head and neck in 1 (3%) [23]. Taniguchi et al. reported that 22 patients with osteonecrosis of the talus underwent replacement with a talar body prosthesis and were followed up for a mean of 98 (range 18-174) months. Of the 22 patients, 4 (18%) sustained fractures of the talar head and neck that required revision with a total talar prosthesis [41]. Taniguchi et al. [36] replaced 55 taluses of 51 patients with talus osteonecrosis with a total talar prosthesis. After a mean follow-up of 52.8 months, a radiographic evaluation revealed osteosclerosis of the distal tibia in 44%, the navicular bone in 9%, and the calcaneus in 35%; however, no complications required revision surgery [36]. Thereafter, Morita et al. reported that in 18 patients (19 talus) with more than 10 years of follow-up (median 152 months), degenerative changes were observed in the distal tibia in 90%, the navicular bone in 16%, and in the calcaneus in 11%; however, no complications required revision surgery [39]. Thus, they recommend use of a total talar prosthesis rather than a talar body prosthesis [41].
We propose a treatment strategy for talar tumors using Enneking stage (Table 4) [42]. For stage 3 tumors, intermediate-grade tumors with large extraskeletal lesions, pathologic fractures with intra-articular involvement, or complex fractures, we recommend marginal resection followed by total talar prosthesis reconstruction (Table 4). Because the articular cartilage acts as a barrier to tumor invasion, malignant tumors rarely penetrate the articular cartilage into the joint and often spread extraskeletally along the ligamentous and articular capsule attachments [43][44][45][46].
Quan et al. reported that all 11 patients with osteosarcoma but without evidence of intra-articular invasion on preoperative magnetic resonance imaging (MRI) were free of intra-articular invasion upon histological evaluation [43]. If there is no evidence of intra-articular invasion on a preoperative MRI, the tumor can be removed by intra-articular resection [43], followed by total talar prosthesis reconstruction (Stage IA and IIA; Table 4; Figure 2). Malignant talar tumors with intra-articular invasion or extraskeletal involvement that require resection of the peritalar bone cannot be reconstructed with total talar prosthesis and usually require tibiotalocalcaneal fusion (Stage IB and IIB; Table 4). Bone grafting should be used to correct leg length differences. If the talar tumors cannot be removed with negative margins, below-the-knee amputation should be indicated (Stage IB and IIB; Table 4) [1].   (c,d) Magnetic resonance imaging (MRI) showed low-signal-intensity lesions on T1WI and highsignal-intensity lesions on T2WI in the posterior part of the talar body and the extraskeletal area (c: T1WI; d: T2WI). A needle biopsy was performed, and a diagnosis of diffuse large B-cell lymphoma was made. (e,f) After six courses of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone, MRI showed a marked reduction in tumor size (e: T1WI; f: T2WI); however, the patient had severe ankle joint pain due to the talar body collapse that was treated with talar tumor resection and total talar prosthesis reconstruction [36,47]. (g,h) Six years after surgery, osteosclerotic changes were observed in the distal tibia and calcaneus; however, the patient was able to walk unaided without pain. No tumor recurrence was observed (g: dorsiflexed position; h: plantar flexed position).

Conclusions
In conclusion, after talar tumor resection, talar prosthesis reconstruction features better postoperative function than tibiocalcaneal fusion or frozen autograft reconstruction. The risk of complications was higher for tibiocalcaneal fusion or frozen autograft reconstruction than for talar prosthesis reconstruction. Therefore, talar prosthesis reconstruction appears preferable to traditional methods, such as tibiocalcaneal fusion reconstruction after talar tumor resection. However, because this systematic review included only retrospective studies with a small number of patients, its results require re-evaluation in future RCTs with larger numbers of patients.