Giant cell tumor of bone is a common primary bone tumor, most of which are benign lesions, but have the tendency to grow aggressively and recur easily after resection, and a few patients may have distant metastases. For the treatment of giant cell tumor of bone, the more widely used method is intracapsular curettage, but this intracapsular curettage method has a very high rate of local recurrence. One of the key factors affecting the local recurrence of tumor is whether the tumor can be completely scraped and the residual cavity treated. At present, the more widely used treatment methods for residual cavity wall include charcoal acid cautery, liquid nitrogen freezing and electrocautery. The methods of filling the residual cavity for reconstruction mainly include allograft bone, artificial bone graft or bone cement filling, among which bone cement filling is considered to be a more ideal method. Because of the heat released by the bone cement during polymerization, as well as the cytotoxic effect of the bone cement monomer are considered to be factors that can effectively deal with the tumor remnant cavity and thus reduce tumor recurrence. It has been suggested that intracapsular curettage supplemented with bone cement filling can achieve marginal resection. However, even with the application of bone cement filling, the recurrence rate is still in the range of 10-15 , and satisfactory clinical results cannot be achieved. Since the shape of the extent of tumor invasion is often irregular, the tumor cavity may have a separation or growth of bone crest, resulting in tumor recurrence as conventional methods may not be able to deal with the tumor remnant cavity thoroughly and effectively. We applied microwave in situ inactivation technology to the treatment of limb bone giant cell tumors in the hope of reducing the local recurrence rate of this tumor. We also compared the clinical results of applying two different microwave inactivation techniques, so as to find a safe and effective method of applying microwaves for the treatment of limb bone giant cell tumors. Data and methods We retrospectively analyzed 21 patients who received in situ microwave inactivation of tumors for giant cell tumors of the limb in our hospital from September 2006 to September 2010. Among them, 6 cases were male and 15 cases were female, aged 19-41 years, with a median age of 25 years. There were 18 primary cases and 3 recurrences (all were local recurrences and no distant metastases were found). The lesion sites were femur in 8 cases, tibia in 5 cases, humerus in 5 cases, radius in 2 cases, and ulna in 1 case. The tumor was accompanied by pathological fracture in 3 cases. All cases underwent preoperative puncture biopsy and were confirmed by preoperative biopsy and postoperative pathology. Two different microwave inactivation techniques were used to treat giant cell tumors of the limb bone. The first method is in situ microwave inactivation and curettage of the tumor, i.e., microwave first and then curettage. The bone involved by the tumor is first separated and the joint capsule of the adjacent joint is incised with a small incision in order to prepare the joint cavity for cooling by applying ice saline infusion. After the tumor is protected by wet gauze, a window is opened on the cortex to reveal the tumor tissue, and the microwave antenna is inserted directly to inactivate the tumor tissue (or directly into the tumor envelope if the tumor completely destroys the cortex), and the inactivation is performed step by step by adjusting the angle and insertion depth of the antenna; the inactivation temperature inside the tumor is monitored to be greater than 80℃, and ice saline is also applied to protect the normal tissue and joint bone around the inactivated bone, and the monitoring temperature After satisfactory inactivation, the inactivated tumor tissue was scraped out of the bone as much as possible (Figure 1), and allograft bone graft was applied to fill the bone defect after tumor scraping, and internal fixation was applied or not according to the size of the lesion. This method was used in a total of 8 cases. The second method is intracapsular scraping of the tumor with microwave-assisted inactivation of the residual cavity after tumor scraping, i.e., scraping followed by microwave. This procedure does not require complete separation to reveal the involved bone, but is the same as the traditional cortical opening for intracapsular tumor scraping. After scraping the tumor tissue as thoroughly as possible, saline is injected into the residual cavity after tumor scraping. If the location does not ensure that the saline can fully fill the residual cavity, the cavity is filled with an appropriate amount of gelatin sponge and then saline is injected, thus ensuring that the residual cavity wall of the tumor is fully in contact with the saline. If the tumor residual cavity is too large, separated or with bone crest formation, the depth and angle of antenna insertion will be adjusted to inactivate the tumor in stages. The microwave was heated until the saline boiled, after which the microwave output was intermittent and the temperature of the tumor residual cavity wall was detected by the thermometer to be greater than 80°C. The inactivation time was 2-6 min, and the average inactivation time was 3 min. The temperature outside the inactivated bone cortex or inside the joint was monitored to be lower than 40°C. After satisfactory inactivation, the four walls of the residual tumor cavity were fully scratched again to completely scrape away the necrotic tissue after inactivation (Figure 2). This method was used in a total of 13 cases. Clinical outcome count data were statistically significant by applying SPSS10.0 chi-square test with P < 0.05. The MSTS score was used for postoperative functional outcome assessment of the affected limb. Discussion Giant cell tumor of bone is currently considered as a benign tumor, and intracapsular scraping of the tumor is a common method of surgical treatment. However, the high rate of local recurrence after intracapsular curettage is the main reason for surgeons to choose the surgical method. The management of the residual tumor cavity after curettage is an important factor in determining tumor recurrence. Most of the literature reports recurrence rates of up to 20 or even 80-90 after scraping alone for giant cell tumors of bone [1-3]. Tumor scraping with cement filling was thought to be effective in reducing the local recurrence rate, but the actual result is often a high recurrence rate even after the application of cement filling, along with a permanent loss of the diseased bone's ability to repair itself after cement filling and a relatively poor long-term functional outcome. This high local recurrence rate has forced some surgeons to use more invasive surgical procedures, such as extensive tumor resection for artificial prosthesis reconstruction [2]. The application of microwave in situ inactivation technology to the treatment of limb bone tumors has been used for more than 30 years, with more satisfactory results especially in the treatment of malignant limb tumors. We have started to use this method in recent years to treat cases of giant cell tumors of the limb bone. The average recurrence rate of all 21 cases treated with microwave therapy reported in this paper was 9.5, which is lower than the local recurrence rate reported in the literature for conventional methods. However, while this microwave in situ inactivation technique can achieve satisfactory local control, it also has significant shortcomings, such as a higher incidence of postoperative fracture and deep infection. In early cases, we used a method similar to that used for malignant limb tumors, applying microwave in situ inactivation of the tumor followed by scraping of the bone graft to reconstruct the bone defect. This method requires local isolation to expose the affected bone, which is more invasive and takes longer to operate; at the same time, because of the need to inactivate the tumor entity, it often requires a higher inactivation output and longer inactivation time due to the slow thermal conductivity of the solid tissue. Since giant cell tumors grow in the epiphysis, the inactivation process will inevitably cause thermal damage to the articular cartilage, which will lead to long-term postoperative joint degeneration. A relatively satisfactory local control of the tumor can be achieved by applying this method, but the incidence of postoperative fractures is high. Even with proper internal fixation, the recurrence rate of distant fractures remains high due to the poor strength of the inactivated bone itself and slow repair. We subsequently improved the surgical approach by focusing on direct microwave inactivation of the tumor and improving the treatment of the residual cavity after microwave-assisted curettage. The surgical trauma of this approach is similar to that of conventional intracapsular scraping of the tumor, and does not require extensive separation to expose the involved bone, while the output power and inactivation time during microwave inactivation are greatly reduced and shortened due to the extremely high microwave thermogenicity of water. Previous experimental studies have shown that all tumor cells can be inactivated in an environment of more than 60°C for 5 min. After the tumor is completely scraped, the residual cavity is injected with saline to ensure that the saline touches every surface of the residual cavity, while the microwave antenna is applied to heat the saline to ensure the inactivation of any surface of the residual cavity by using the uniform consistency of heat conduction in the saline. This method can effectively inactivate the residual cavity and fully reduce the important factors leading to postoperative local recurrence, and the postoperative local recurrence rate is not significantly different from the microwave in situ inactivation technique. It can provide the most ideal reconstruction method of bone graft filling and avoid the postoperative complications caused by microwave inactivation of tumor, and the long-term postoperative follow-up results are very satisfactory. However, this method also has some limitations and is usually only applicable to cases with intact tumor envelope and intact bone shell. Through a retrospective analysis of more than 20 cases of limb bone giant cell tumors treated with microwave therapy, we concluded that microwave inactivation technique is a surgical method to treat limb bone giant cell tumors that can obtain a satisfactory postoperative local recurrence rate. Meanwhile, microwave-assisted residual cavity inactivation by intracapsular scraping of the tumor followed by bone grafting to reconstruct the bone defect, compared with in situ inactivation of the tumor followed by scraping, can achieve satisfactory local control and reduce the incidence of postoperative fracture with better postoperative long-term functional outcome.