Is Limb Salvage With Microwave-induced Hyperthermia Better Than Amputation for Osteosarcoma of the Distal Tibia?

Between 2000 and 2015, we treated 106 patients for distal tibia osteosarcoma without metastases. Of those, 52 were treated with microwave-induced hyperthermia (Table 1), and 41 with amputation. A total of 13 patients who would have met our indications for microwave-induced hyperthermia were instead treated with the conventional limb-salvage resection and reconstruction based on the patient’s preferences (which might have been driven by cost, perceived functional demand, or other factors); these patients were not included in this retrospective study. If neither the posterior tibial artery nor the dorsalis pedis artery was salvageable, patients were treated with amputation; these patients (n = 14) were not included in any group in this study. This left 27 patients with amputations available for our study (Table 2) and 79 patients available for the entire study. Patients were considered eligible for microwave-induced hyperthermia if they had an at least 20 mm available distance from the tumor edge to the articular surface, good clinical (such as pain reduction) and imaging responses to neoadjuvant chemotherapy, and no pathologic fracture. Chemonecrosis was assessed using the grading system of Huvos et al. [14]. More than 90% necrosis on the histologic sections was considered a good response to chemotherapy. All patients in this series were available for followup at a minimum of 12 months (mean followup in the hyperthermia group, 79 months, range, 12–158 months; mean followup in the amputation group, 95 months, range, 15–142 months).

All patients had radiographs, CT, MRI, and bone scans. With the numbers available, we found no difference in sex between the amputation group and microwave-induced hyperthermia group (12 males and 15 females versus 30 males and 22 females; p = 0.263) (Table 3). We also found no difference in age (27.5 ± 8.7 years versus 31.2 ± 6.4 years; p = 0.586), tumor grade, tumor stage, and tumor size between the amputation group and the microwave-induced hyperthermia group (Table 3). Of the 79 patients, 54 had a needle biopsy and 32 had an incisional biopsy, including those whose needle biopsy was nondiagnostic. We graded the histologic sections based on the biopsy using Broders’ classification [1], which has four grades according to the rate of differentiation of the tumor cells. We staged patients using the surgical staging systems of the MSTS [6] and the American Joint Committee on Cancer (AJCC) [24]. Nineteen patients had Stage I tumors and 60 had Stage II tumors.

All patients were evaluated by CT and MRI at the end of each chemotherapy regimen preoperatively to define the edge of the tumor, which was determined at the transition of marrow signal from abnormal to normal. Areas of intermediate signal intensity adjacent to the tumor edge were regarded as part of the tumor and should be included in the ablation area. All 79 patients received two cycles of preoperative neoadjuvant chemotherapy based on a standard protocol which was described in a previous study [17].

Patients treated with microwave-induced hyperthermia were evaluated according to the following criteria: (1) assessment of tumor response or progression as assessed by MRI; (2) distance between the ankle cartilage and the tumor as assessed by MRI of 20 mm or more, to obtain a bone width margin of 10 mm and a remaining residual epiphysis of 10 mm, and wide proximal margins on the bone resections [19] (defined as a cuff of 2 cm to 3 cm of normal tissue remaining on all sides of the tumor); and (3) a sufficient amount of epiphysis preserved to allow fixation of the osteotomy junction [21]. Intraoperatively, the adequacy of bone resection was evaluated with frozen section biopsy of a tissue sample obtained from the medullary canal of the residual tibia. For all patients who had amputations, the margins were wide (a cuff of 2 cm to 3 cm of normal tissue remaining on all sides of the tumor). After surgery, the histologic margins were negative in all patients.

All operations were performed by the same two surgeons (QYF and YZ). The microwave-induced hyperthermia machine we used was the FORSEA (Xinhua Company, Nanjing, China) [9, 10], and the microwave generator frequency is 2450 MHz. When microwave-induced hyperthermia was performed (Fig. 1), the main principle was to dissect the tumor with a safe margin as described above, and subsequently perform an en bloc ablation using antenna-guided hyperthermia therapy. The first step was to identify the extent of the tumor and dissect the tumor-bearing bone from surrounding normal tissues with a safe margin (at least 20 mm width). We usually use the original dissection method of double incisions to obtain adequate exposure (Fig. 1F). This step is very important because it is helps to ensure the entire tumor can be killed by microwave-induced hyperthermia. A heat-isolation pad and wet gauze then were placed between the bone tumor and surrounding normal tissues. Then, one to six antennas were placed in different location of the tumor from different angles, matching the suction one-to-one, according to the shape and size of the tumor to ensure the therapeutic range and the tumor edge could be ablated adequately. Heat output was instant when the antennas were placed, and electromagnetic energy then was delivered to the tumor (Fig. 1G). The tumor was ablated with direct heating while normal soft tissues were protected from overheating. The goal of microwave ablation is to create an ablation zone that extends 1 cm beyond the tumor boundary at all points with the core temperature of the tumor reaching 85° to 100°C and the normal tissue temperature remaining less than 40°C for 15 to 20 minutes. During surgery, a circulating cool saline system was used to protect the surrounding normal tissues, and multiple thermocouples were placed in various critical locations to monitor the temperature. To avoid damaging the joint, outlet piping which was connected with a circulating water pump and the thermocouples were specifically placed in the ankle cavity to keep the articular cartilage and its subchondral bone from overheating. All the tissue blocks were evaluated histologically for tumor hyperthermia necrosis and the histologic examination showed that part of the proximal margins were histologically negative and part of the margins were necrotic. After the dead tumor mass was removed or curetted (Fig. 1H), the reconstruction was performed using a mixture of bone chips and bone cement (Fig. 1I–K) [18]. The normal shape of the tibia was restored and prophylactic fixation was performed if necessary (Fig. 1I–K).

Transtibial amputation was performed as common practice [13].The goals and requirements were resecting the bone 2 cm to 3 cm proximal to abnormal bone density, obtaining adequate length of the residual limb, and achieving good soft tissue coverage.

All patients in both groups were given antibiotics for 72 hours after surgery, and they performed bed exercises until wound healing was achieved. A short cast or a brace was used for patients who had microwave-induced hyperthermia until there was radiographic evidence of bone union. Signs of bony union were evaluated by serial sets of plain radiographs [11]. All patients in both groups received postoperative chemotherapy (adriamycin, cisplatin, methotrexate, ifosfamide) [10]).

After discharge from the hospital, clinical and radiographic followups are done every month during the first 6 months, then every 3 months during the next 2 years, and then every 6 months. Chest CT scans were performed to observe pulmonary metastasis every 3 months during the first year and then every 6 months afterward. A bone scan was performed every 6 months during the first year and then every year. All patients have radiographs taken once a year. The MSTS score was used to observe the function of the patients. The status and function of the ankle were specifically assessed clinically and radiologically at followups.

Clinical outcomes were assessed by review of clinic notes, supplemented by phone questionnaires, and email where needed. Local recurrence, metastasis, complications, and death were recorded from the patients’ files. Complications were graded using the classification described by Dindo et al. [3], which graded the complications at five levels. Followup review and data were sorted and analyzed by three of the authors (KH, NB, TY).

All values are expressed as mean ± SD, and all error bars represent the SD of the mean. Student’s t test and one-way ANOVA were used to determine significance. Survival rates were estimated using the Kaplan-Meier method. We compared survival between the two groups using a log-rank test. Chi-square test was used to compare complications between the two groups. The mean, SD, and 95% CI were provided. All statistical tests were two-sided. A probability less than 0.05 was considered statistically significant. Statistical analyses were performed using SPSS Version 17.0 (SPSS Inc, Chicago, IL, USA).