Chondroblastoma was defined as a benign tumor, but it could develop local recurrence and metastasis. The local recurrence rate of lesion curettage was 0 ~ 39% [4,11–19]. The risk factors of tumor recurrence have been controversial. There are several risk factors that were considered to be related to tumor recurrence such as age, gender, tumor with aneurysmal bone cyst (ABC), the biologic aggressiveness of tumor, location, epiphyseal plate status, affected epiphyseal plate, therapy method.
The age at diagnosis of CBL was considered as a risk factor for local recurrence, and the most previous studies believed that the younger the patient the greater was the risk for local recurrence. Cong Huang et al. observed that all the patients whose tumors recurred were less than 12 years old. All of their patients were less than 14 years old and were similar to our patients’ age [17]. Meanwhile, R. Suneja et al. mentioned that there were seven local recurrence patients and six of seven were lower than 14 years old. They thought that the local recurrence was associated with younger patients [4]. According to Francesco Muratori et al., at the age of 7 to 42, the age less than 11 was the only significant factor that influenced local recurrence risk. According to our results, all cases with local recurrence were over 11 years old and were less than 6 months from onset to diagnosis. The misdiagnosis of CBL is common, although case with classical features could be diagnosed quickly, but there is few confused cases with giant cell bone tumor (GCT) or ABC may give an overlap clinical presentation [20]. Local pain was the most common symptom and could last more than 1 years and there were a few of patients without any symptom, which will delay the diagnosis [4, 15, 20]. It is not easy to explain our result that all of recurrent patients were more than 11 years old. It may be the bias caused by only three patients less than 11 years old in our group and we were unable to determine the duration of the asymptomatic incubation period prior to the onset of clinical symptoms. There was only one article whose results is similar to ours, that is, advanced age is a risk factor for recurrence [15] and we tended to believe that younger age serves as a risk factor for the recurrence of CBL.
In this study, we also explored the relationship between tumor volume and local recurrence and we found that the larger the tumor volume the greater was the risk for local recurrence (P < 0.05). In the updated analyses, volume over 25cm3 was associated with a statistically significant increase in local recurrence. In the context of other tumors, a larger tumor volume was generally regarded as being associated with a higher likelihood of recurrence [21, 22]. As we all know, there were few studies that mentioned this relationship in CBL. On the one hand, when surgeons tackled large tumors, they were worried that, in comparison to smaller tumors, patients were at a heightened risk of developing limb-length discrepancy (LLD) following extensive curettage of the epiphyseal plate [4, 17, 23]. Notably, 82.1% of our patients' epiphyseal plates were affected, with only a single case of a patient with a closing plate experiencing postoperative LLD and knee varus. It was postulated that with meticulous care, the epiphyseal plate could potentially undergo repair [16]. Prior investigations have established a correlation between the extent of the injured epiphyseal plate and the resistance of long bones to normal growth and development [24, 25]. Meanwhile, several animal experiments have proved that the epiphyseal plate has extremely limited plasticity, when the injured percentage of eccentric lesion is more than 9%, it could cause obvious restriction [24, 26]. Otherwise, according to the Dales and Harris classification, the epiphyseal plates of the distal femur and the proximal tibia were both type B, the blood supply primarily originated from the blood vessels of the adjacent soft tissues, which would sustain minimal damage during the operation. This minimized damage was advantageous for the subsequent repair of the epiphyseal plates [27]. On the contrary, The damage to the epiphyseal plate was considered to be irreversible in a few studies [28]. However, we held the belief that there existed a potential for spontaneous repair of the epiphyseal plate, because most epiphyseal plate were hurt by tumors but only one patient who developed LLD.
On the other hand, the tissue of chondroblastoma was soft and poorly interconnected, leading to the leakage of numerous granular-like tumor tissues during the scraping of larger tumor masses. This leakage prevented complete removal and subsequently resulted in recurrence. To address this issue, we employed alcohol immersion for 3–5 minutes to enhance the brittleness of the tumor tissue, facilitating both inactivation and scraping. Anhydrous ethanol can cause protein denaturation in tumor cells, cytoplasm denaturation in cells, and embolization of small blood vessels supplying the tumor. Additionally, anhydrous alcohol has been proven to have a few adverse effects on surrounding tissues [29]. Concurrently, the application of alcohol to the sclerosed zone surrounding the tumor was advantageous in facilitating the infiltration of progenitor cells into the cavity, thus contributing to the repair of the cavity.
CBL has been proved to be common in distal femur, proximal tibia and proximal humerus, although our results only partially concurred with this observation. In this study, CBL mainly occurred in proximal humerus, distal humerus and proximal tibia. There was few researches that reported that special anatomic site, such as pelvis [30]. The reason why CBL in pelvis was easy to recur was believed that its anatomical position made it difficult to completely expose the tumor and incomplete curettage leaded to recurrence. In this study, there was a patient whose CBL was located in the pelvis. After adequate curettage, the patient has no signs of recurrence at present. Frédéric Sailhan et al. found that CBL far away from the active epiphysis (close to the elbow or far from the knee) has a higher recurrence rate among their patients, which was completely contrary to our results. The CBL of our recurrent patients were located in the proximal humerus, distal femur and proximal tibia respectively. Our study found no significant correlation between the location of CBL and local recurrence.
In our center, the main treatment methods of CBL were curettage, assisted by anhydrous alcohol as adjuvant and bone grafting. Alcohol as adjuvant was considered had a good effect of inhibiting tumor recurrence. Tumors located in the proximal femur and pelvis were reported to be susceptible to recurrence because they were difficult to achieve complete surgical resection [4, 30–32]. Therefore, it is necessary to use adjuvants. High-speed burring and it has been proved that it could obviously reduce local recurrence in CBL curettage [33, 34]. However, Cong Huang et al. expressed concerns that the epiphyseal plate might sustain damage from the heat generated by the burr during the procedure [17]. Cement was also deemed potentially harmful to the epiphyseal plate due to the heat released during its solidification process [20]. Furthermore, the use of other adjuvants, such as phenol or liquid nitrogen, could potentially lead to necrosis [13]. Alcohol has been proved to have a good effect on inactivating tumor cells and has little effect on surrounding tissues [29]. But there were few researches using alcohol alone as the adjuvant to treat the CBL.
Previous studies have suggested that ABC is a risk factor for CBL local recurrence [11, 35, 36]. Whether ABC was the recurrence factor of CBL has always been controversial, and there were also a large number of documents showing that ABC is not related to the recurrence of CBL [19, 37]. According to our results, there was no significant difference between ABC and local recurrence and only 2 patients CBL with ABC, which located in ischium and distal femur respectively and both of them were not developed local recurrence.
Among patients who underwent curettage and artificial bone grafting, three patients developed femoral neck fracture, gluteal sinus tract, and upper limb sinus tract. The bone graft substitutes that were used included calcium phosphate (CaP) and calcium sulfate (CaS) compounds. Generally, the main benefit of CaP ceramics stemmed from their osteoconductive properties, and they could provide some structural support in the form of compression strength. However, they were brittle and their low tensile strength may have contributed to increased postoperative fracture risks [38]. Additionally, CaS compounds were believed to be associated with persistent serous wound drainage, which resulted from the inflammatory response to their resorption. In most cases, this was a sterile drainage that persisted until the calcium sulfate was radiologically absorbed [39]. With proper local wound care, the wounds of two patients with sinus tracts recovered smoothly.
The primary limitation of this study lies in the rarity of the tumor and the relatively small sample size. Additionally, the average follow-up duration was insufficient to adequately assess complications such as secondary osteoarthritis. Furthermore, the retrospective nature of this study may undermine the robustness of our findings. Moreover, this study did not incorporate comparisons with other surgical interventions, such as radiofrequency ablation, to evaluate their impact on recurrence rates.