囊内刮除术与瘤段切除术治疗肱骨近端骨巨细胞瘤的多中心回顾性研究

Multicenter retrospective study on different surgical treatments for giant cell tumor of the proximal humerus

目的探讨肱骨近端骨巨细胞瘤不同手术方式治疗的疗效,分析肿瘤复发及影响肩关节功能的危险因素。方法采用回顾性病例对照方法。纳入2002年1月—2015年6月内蒙古医科大学第二附属医院、中国人民解放军第九六○医院、天津医院、河北医科大学附属第三医院等4个骨肿瘤治疗中心临床资料完整、初次接受手术治疗、随访时间超过36个月的27例肱骨近端骨巨细胞瘤患者,其中男15例,女12例;首诊时年龄18~55 (33.1±12.2)岁。Campanacci分级Ⅱ级17例,Ⅲ级10例。其中17例发生病理性骨折,14例合并移位,3例有骨折线、无移位。14例采用囊内刮除术治疗,其中3例刮除、11例切刮;骨缺损采用自体植骨2例、自体加异体植骨1例,异体骨植骨3例,骨水泥填充8例,其中10例患者采用钢板螺钉内固定。13例采用瘤段切除肩关节重建术治疗,其中半肩关节置换10例,大段异体骨关节移植髓内钉固定3例。术后定期随访。末次随访时,采用国际骨与软组织肿瘤协会(MSTS)评分标准评估不同手术方式患者术后肩关节功能。应用SPSS 22.0统计学软件进行数据分析,比较不同手术方式之间复发和并发症发生率、上肢MSTS评分,对手术方式选择的影响因素和肿瘤复发危险因素进行单因素分析。结果患者术后均获随访36~180 (92.4±38.9)个月。刮除术和瘤段切除术中各有1例局部复发,复发率为1/14和2/13(P=1.000)。4例瘤段切除患者术后发生并发症,总的并发症发生率14.8%(4/27),其中2例大段异体骨吸收(1例伴肿瘤复发)行半肩关节置换术,1例半肩关节置换后肩关节半脱位,1例术后5年假体松动外露再次手术更换骨水泥型假体进行翻修;囊内刮除术的并发症发生率低于瘤段切除术,差异有统计学意义(P=0.041)。有无病理性骨折与手术治疗方案的选择无相关性(r=-0.037, P=1.000),Campanacci分级与手术治疗方案的选择有相关性(r=0.482, P<0.05);有无病理性骨折、不同Campanacci分级、不同手术方式的患者局部复发率差异均无统计学意义(P值均>0.05);刮除术术后肩关节功能明显优于瘤段切除术(P<0.01)。结论肱骨近端骨巨细胞瘤相比较于其他部位容易发生病理性骨折,采用刮除术或瘤段切除术后复发率较低,刮除术后肩关节功能明显优于瘤段切除。由于瘤段切除术后并发症的发生率较高,尤其是瘤段切除大段异体骨关节移植后骨吸收等问题,建议尽可能地选择囊内刮除手术作为肱骨近端骨巨细胞瘤的治疗方案。

Objective To restospectively analyze the treatment and efficacy of giant cell tumor(GCT) in proximal humerus, specify the indications and precautions for different surgical methods and analysis the risk factors for recurrence and shoulder dysfunction. Methods The retrospective case-control study was conducted. From January 2002 to June 2015, the complete clinical data of 27 patients with giant cell tumor of the proximal humerus in the Second Affiliated Hospital of Inner Mongolia Medical University, the 960th Hospial of PLA, Tianjin Hospital and Third Hospital of Medical University were collected. Among them, there were 15 males and 12 females, with age of 18-55(33.1±12.2) years. Twenty-seven patients with initial treatment and follow-up for more than 3 years were included in the study. There were 17 patients with pathological fracture, 14 cases with fracture displacement, 3 cases with pathological fracture but no displacement;Campanacci grade Ⅱ GCT was diagnosed in 17 patients;Campanacci grade Ⅲ GCT in 10 patients, 14 cases of curettage in the initial operation (3 cases of scraping, 11 cases of enlarged scraping), of which 10 patients were treated with plate screw internal fixation. There were 2 cases of autogenous bone grafting, 1 case of autologous plus allogeneic bone grafting, 3 cases of allogeneic bone grafting, 8 cases of bone cement filling to fill the bone defect. Thirteen cases underwent resection and reconstruction of the tumor, including 10 cases of semi-shoulder joint replacement and 3 cases of intramedullary nail fixation with large allograft. The patients were followed up regularly and the effects of upper shoulder joint function were evaluated according to the Musculoskeletal Tumor Society(MSTS) score. Statistical analysis was performed using SPSS 22.0 statistical software. The recurrence rate and complication rate, the upper limb MSTS score, the analysis of the influencing factors of surgical approach selection and the risk factors of tumor recurrence were compared by univariate analysis between the tumor segment resection group and the risk factors for tumor recurrence. Results All the patients were followed up for 36 to 180 (92.4±38.9) months. There was 1 local recurrence in each of the curettage and tumor resection, and the recurrence rate was 1/14 and 2/13. Four patients with tumor resection had postoperative complications, and the total complication rate was 14.8%(4/27). Among them, 2 patients with large segmental allografts underwent semi-shoulder arthroplasty;1 patient with subluxation of shoulder joint;1 patient with prosthesis loosening exposed for 36 months after surgery had to undergo revision surgery with replacement of cemented prosthesis. The incidence of complications of intracapsular curettage was lower than that of tumor resection, and the difference was statistically significant(P=0.041). Single factor analysis showed that there was no correlation between pathological fracture and surgical methods (r=-0.037, P=1.000), Campanacci grading was correlated with surgical methods (r=0.482, P<0.05). There was no correlation between surgical procedure, pathological fracture, Campanacci grading and local recurrence rate(all P values>0.05). The postoperative shoulder function of the patients with scrape reconstruction was significantly better than that of the tumor reconstruction (P<0.01). Conclusions The giant cell tumor of the proximal humerus is prone to pathological fracture compared with other sites. The recurrence rate is lower after curettage or tumor resection, but the latter has high postoperative complications. The shoulder function of the scraping group was significantly better than that of the tumor segment group. Due to the high incidence of complications after resection of the tumor, especially the bone resorption after resection of large segmental allografts. Intracapsular curettage should be selected as a treatment of giant cell tumor in proximal humerus.