[ Next ] [ Prev ] [ Abs ] [ Chi ] [ TOC ] [ Home ]
Chin Med J (Taipei) 1997;59:240-7.
1Department of Orthopedics, China Medical College Hospital, Taichung; and 2Departmnt of Orthopedics and Traumatology, Veterans General Hospital-Taipei, Taipei, Taiwan, R.O.C.
Background. In most of the reported series, almost half of the giant-cell tumors involved the knee region. The characteristics of the lesions were usually benign but often locally aggressive and easily recurrent neoplasms. Some surgeons performed intralesional excision combined with local adjunctive chemical coagulant, and bone grafting or methylmethacrylate cement packing. These methods could both decrease local recurrence and retain the function of joint.
Methods. From January, 1984 to December, 1994, a review was made of the results for eighteen patients who had been managed consecutively at the Veterans General Hospital-Taipei for giant-cell tumor of bone around the knee. Fourteen instances had occurred in the proximal tibia and four, in the distal femur. According to the classification of Campanacci, nine lesions were Stage II and nine, Stage III. Eleven patients had been managed with intralesional excision of the tumor with local adjunctive application; the other seven had en bloc resection and reconstructive procedures.
Results. All patients had been followed for a mean of fiy-six months (range 22 to 125 months). The overall recurrence rate was 11% (2/18). The intralesional excision had 18% (2/11) recurrence; there was no recurrence in the en bloc resection (0/7). The complication rate was 16% (4/18); 9% (1/ 11) for intralesional excision and 42% (3/7) for en bloc resection, respectively. The mean functional score was 28 points (range, 22 to 30) in the intralesional excision group and 21 points (range, 11 to 30) in the en bloc resection group.
Conclusions. En bloc resection with reconstruction had a lower rate of recurrence, but a higher rate of complication and poor functional results. Intralesional excision, combined with a local adjunctive application and packed with bone grafting or methylmethacrylate cement, was an acceptably good method with satisfactory results, which either decreased local recurrence or retained the function of the joint.
[Chin Med J (Taipei) 1997;59:240-7.]
Keywords: en bloc resection, giant-cell tumor, intralesional excision, knee
Received: May 25, 1996.
Accepted: December 11, 1996.
Address reprint requests to: Tain-Hsiung Chen, Department of Orthopedics and Traumatology, Veterans General Hospital-Taipei, No. 201, Sec. 2, Shih-Pai Road, Taipei, Taiwan, R.O.C.
Benign giant-cell tumors account for approximately 19% of benign bone tumors and 9% of all primary bone tumors. In most of the reported series, almost half of the lesions involved the distal end of the femur or proximal end of the tibia [1,2]. These tumors had been characterized as a usual benign, but often locally aggressive, neoplasm. The recommended treatment has ranged from curettage to en bloc resection.
Intralesional excision (curettage) has been associated with high rates of local recurrence, ranging from 27% to 54% [3,4]. En bloc resection has been reported to be a more effective treatment with a lower recurrence rate in the range from 0% to 32%. However, en bloc resection usually requires sacrifice of the articular surface and a complicated reconstructive procedure.
In an attempt to avoid en bloc resection procedure for the treatment of para-articular lesions, some surgeons had performed intralesional excision combined with a local adjunctive application of an agent such as liquid nitrogen, phenol or methyhmethacrylate [5,6].
The purpose of this study is to report 11 years of experience in treatment of giant cell tumor around the knee, comparing recurrence, complications and functional score under different methods of treatment.
Eighteen patients were managed consecutively at Veterans General Hospital-Taipei, for giant-cell tumor of the bone, around the knee, between January, 1984 and December, 1994. The study group included nine female (50%) and nine male patients (50%). The mean age of the patients at the time of operation was 33 years old (range: 18-64). The mean follow-up interval was 56 months (range: 22-125). Medical records and radiographs were all reviewed.
All patients had a functional analysis, according to the evaluation system of Enneking et al. [8], in the most recent follow-up. This system comprises six criteria: pain, function of the extremity, emotional acceptance of any residual deficit, use of mechanical supports, walking ability and gait when a lower extremity is involved. Each category is assigned a score from 0 to 5 points, so that the maximum score is 30 points for the entire functional evaluation. Mann-Whitney U test, Chi-square test, and Fisher exact test were used to analyze the data. The level of significance was p < 0.05.
The staging of giant-cell tumor was classified according to the system of Campanacci [1]. Stage I indicates an intraosseous lesion; Stage II, an intraosseous lesion with cortical thinning; and Stage III, a lesion extending extraosseously.
Different operative procedures were chosen by perceived indication. The indications for a wide resection was extensive tumor (Stage III) invasion, pathological fracture and rapid or large recurrent tumor after intensive curettage [1,7]. Seven patients were managed with an en bloc resection with reconstruction, including a recurrent lesion after curettage when visiting our hospital, and the other eleven patients had had an intralesional excision and local adjunctive application with phenol and alcohol, combined with either bone-grafting (eight patients) or cement packing (three patients).
The en bloc resections consisted of removal of the bone involved with the tumor, and the adjacent tissue, as an unit. The excised normal tissue cuff was not as wide as for conventional malignant tumors. Reconstruction was individualized according to the age of the patients, the location of the tumor, the functional demands, and the preference of the patients. Intercalary allograft was used to perform an arthrodesis of the joint for patients who performed strenous manual labor (Cases 2, 6 and 16). For patients in whom the joint had to be sacrificed but who wanted to have a mobile joint, that was achieved with using of arthroplasty (Cases 11 and 13) or massive osteochondral allograft (Cases 10 and 14).
The intralesional excisions were done in a precise and stepwise manner. After the diagnosis was established histologically with an open biopsy, the tumor was completely exteriorized with unroofing of the cortex, so that all of the cavity occupied by the tumor was visualized. Then, the peripheral tissue was packed with gauze to prevent spillage of tumor tissue during the curettage. The tumor was first removed with a curet until no tumor tissue remained by gross observation. Next, all the inner osseous surface of the tumor cavity was burred with a high-speed burr. Then, the surface of the cavity was cauterized with phenol and alcohol, and these agents were removed from the cavity with copious physiological solution. Finally, the cavity was packed with either methyhmethacrylate or an autograft or allograft. The group of eleven patients included nine Stage II lesions and two Stage III.
In the group of intralesional excision, immobilization by cast should be and was continued for six weeks after operation because of thinning of subchondral bone. For patients with en bloc resection and osteochondral allograft, casting was necessary, too.
Fourteen lesions were in the proximal part of the tibia. Four lesions were in the distal part of the femur. Nine lesions were Campanacci Stage II and nine were Stage III (Table 1). The overall recurrence rate was 11% (2/18). The average time to recurrence was 13 months (range: 12-14).
The group of en bloc resections had seven patients, including four male and three female patients. All seven lesions were Campanacci Stage III. There was no local recurrence or remote metastasis. The mean functional score was 21 points (range: 11-30). Three patients (42%) had complications which led to an additional operation. These complications included one valgus deformity (Case 2), one nonunion (Case 6) and one infection (Case 11). Case 2 received corrective osteotomy for progressive valgus deformity. Case 6 was treated with removal of previous implant and bone graft, and another reconstructive procedures for nonunion. The third patient (Case 11) was already a recurrent case when she consulted this hospital. After a previous operative procedure at an other hospital 13 months previously, recurrence was noted about 12 months later. Infection occurred three months after our operation. Despite repeated debridements, discharge sinus was still noted at a recent follow-up.
The group of intralesional excision had eleven patients, composed of five male and six female patients. Nine lesions were classified as Campanacci Stage II, and two, Stage III. Phenol and alcohol were used in all eleven patients. In eight patients bone grafting was used for fill the osseous defect; the other three patients (Cases 3,5 and 8) had methylmethacrylate to pack the defect. Local recurrence was noted in two of the eleven patients (18%) (Cases 1 and 18). These two patients were classified as Stage II, and the bony defect was packed with bone graft. They received repeated curettage and bone graft, and were free of disease at the latest follow-up visit. No patient had a pulmonary metastasis. The mean functional score was 28 points (range: 22-30). One patient (Case 3) had skin necrosis over the pre-tibial region, but she had no local recurrence. She received debridement, gastrocnemius flap and skin graft. One patient (Case 9) expired after 51 months postoperatively from hypoglycemic shock.
Historically, simple curettage of giant-cell tumors of the long bone have been associated with a high recurrence rate from 27% to 54 % [1,9]. This has led many surgeons to adopt wide excision as the treatment of choice, with a rate of local control increased to more than 90%. However, the functional results were not as good as when the joint was preserved. Intensive curettage combined with packing with methylmethacrylate cement was suggested as a means to decrease the rate of local recurrence (0-29%) [8], with preservation of the function of the joint. The rate of recurrence in this series represents a much lower incidence (18%) than has been reported for most large series of giant-cell tumors of bone.
For giant-cell tumor about the knee, the intralesional technique was preferred for Stage II lesion, after excision; then chemical coagulation was followed by bone grafting. If there were a major loss of bone to prevent supporting body weight, intralesional excision followed by insertion of methylmethacrylate could be effective. In Stage III lesion with simple bulging, or no important structure invasion and an intact subchondral bone or only partial destruction, the intralesional technique was used initially. Two patients of Stage III (Cases 3 and 4) had this method, and there was no recurrence during the periods of follow-up. Sometimes a second looking must be in suspicious cases. The literature also supported the perception that intralesional excision could be repeated if an initial intralesional procedure had failed [14].
Giant-cell tumor with pathological fracture may release tumor cells into the surrounding soft tissue and joint. Some authors have suggested extensive surgery because of fear of tumor contamination of the adjacent tissues [15]. But another author has reported a low risk of local recurrence after curettage and cementing for giant-cell tumor with pathological fracture [14]. Our patient (Case 15), with distal femoral lesion and pathological fracture, was managed with curettage and bone grafting. No recurrence was noted after 30 months of follow-up.
Most recurrences in large series occurred within the first two years after the index treatment of the tumor [4]. Therefore, it is concluded that the duration of follow-up in the present series was adequate to demonstrate that both intralesional excision and en bloc resection are successful methods of treatment for giant-cell tumors. However it was not possible to conclude, from the present study, that the intralesional procedure is as effective as en bloc resection, because the series was not large enough and there were only two recurrences; the difference between the rates of recurrence was not significant (p = 0.117, Table 2).
It is not possible to determine which factor in the intralesional excision made the most important contribution to the low rate of recurrence. The operative techniques included extensive exteriorization of the tumor. The use of high-speed burr and local extension modalities (phenol, liquid nitrogen, and carbon-dioxide lasers) to kill these residual microscopic foci [10,11]. Although local adjunctive application is still controversial, all these techniques should be used step-by-step to prevent local recurrence. For thus large bone defects, there is not enough autogenous graft to pack the whole defect. The autograft and allograft were combined, and it was preferred to put the autograft around the subchondral region to get an earlier solid union in that area. Thus the patient might tolerate early weight bearing
There was no recurrence in patients who had had an en bloc resection; that was an excellent oncological result. The disadvantage of this treatment was its relatively poor functional outcome; major complications included infection, resorption of the graft, collapse of the articular surface, fracture, nonunion and recurrence [12,13]. The functional scores in this group were significantly lower (p = 0.0001) compared with those in the group that had had an intralesional excision. Undoubtedly a factor contributing to this functional loss is that so many of the patients had reconstruction of a joint, with either an allograft or a prosthesis. (Three patients were treated with arthrodesis, which lowered the whole score).
We believed that the intralesional excision is the treatment of choice for Stage I, II, and properly selected Stage III lesions because of superior functional outcome. It is not surprising that, no matter how sophisticated the reconstruction after an en bloc resection, it can never be as effective as the patient's own joint.
Copyright: 1997, Chinese Medical Association (Taipei)