Giant cell tumors (GCT) account for 5-8% of primary bone tumors. About 5% of GCTs occur in flat bones, with the pelvis being the most common. Among the vertebrae, the longest occurrence is in the sacrum, while other vertebrae are less commonly involved (1-4). GCT of the sacrum often occurs in the superior sacrum, which is more locally invasive, has complex vascular-neural anatomy, and is difficult to treat, and there is still no definite ideal solution for the treatment of GCT of the sacrum. In this paper, we investigated the local recurrence rate and complications of sacral GCT treated by excisional scraping with effective bleeding control.
Materials and methods
(A) General data: 43 cases of sacral GCT were admitted to our hospital from July 2000 to July 2007. There were 35 cases with follow-up records. Among them, 18 cases were male and 17 cases were female. The youngest age was 16 years old, the oldest was 61 years old, and the average age was about 32 years old. All patients were routinely examined with preoperative frontal and lateral radiographs, CT or/and MRI to determine the imaging grade of the tumor.
There were 5 cases of GCT occurring in S1-S5, 9 cases of GCT occurring in S1-S4, 12 patients of GCT occurring in S1-S3, 3 cases of GCT occurring in S1-S2, 2 cases of GCT occurring in S2-S5; 1 case of GCT occurring in S3-S5; and 3 cases of GCT occurring in L5-S1.
(B) Surgical method: Among the 35 patients in this group, 8 cases were referred to our hospital after local recurrence of surgery in outside hospitals before coming to our hospital. There were 3 cases with 3 surgeries and 8 cases with 2 surgeries. The treatment of patients with sacral GCT included 31 patients who underwent intra-focal marginal resection (excision and scraping) and 4 patients who underwent extensive marginal resection. Due to the rich blood supply in sacral GCT and the high intraoperative bleeding, intraoperative temporary abdominal aortic blockade was used in 23 sacral GCT cases, including 6 cases in which the abdominal aorta was detached retroperitoneally by anterior surgery and the abdominal aortic flow was temporarily blocked with a shoelace; in 17 cases, the abdominal aorta was implanted with a balloon under preoperative X-ray intervention and the abdominal aortic flow was temporarily blocked intraoperatively. No preoperative vascular block was performed in 12 cases.
(iii) Temporary blockade of abdominal aortic flow.
(1) Internal iliac artery embolization and abdominal aortic balloon placement method: One day before surgery or on the day of surgery, the femoral artery was punctured using the Seldinger puncture method, and the catheter was inserted retrogradely through the femoral artery to the proximal end, and the two sides or unilateral internal iliac arteries were inserted after abdominal aortography to understand the tumor site, nature, extent and blood supply, and gelatin sponge and spring embolus were used as embolic material to block the bilateral or unilateral internal iliac arteries (generally, the internal iliac arteries were inserted after abdominal aortography). The internal iliac artery (usually the side with more tumor invasion) and other target vessels can be embolized. The effect of embolization is determined by reimaging the abdominal aorta. A balloon is placed in the renal artery 1 cm below the bifurcation of the abdominal aorta, and a blocking test is performed and the image is reviewed so that the contrast does not flow distally and does not block the bilateral renal artery flow.
2. Unilateral internal iliac artery ligation and temporary abdominal aortic block technique: The patient was placed in the right lateral position, and a left-sided macrosigmoid incision was made, with three layers of the abdominal muscle incised and the peritoneum pushed medially to reveal the ipsilateral common iliac vessels and internal and external iliac arteries. The internal iliac artery was separated and ligated, and the abdominal aorta was revealed by upward freeing, and a rubber tube was temporarily blocked below the bifurcation of the renal artery with a gauze strip.
(iv) Reconstruction method: Among the 35 cases of sacral GCT, lumbosacral reconstruction was performed using the staple bar system in 29 patients, and reconstruction was not performed in 6 cases that did not affect the stability of the sacroiliac joint.
Results
One patient had a recurrence after an external surgery with a large tumor and surface ulceration and died from a severe infection 2 weeks after reoperation. The remaining 34 patients were followed up from 12 months to 7 years with a mean follow-up of 37 months. 1 case of malignant sacral osteoblastic giant cell tumor died 15 months after surgery. 1 case of sacral osteoblastic giant cell tumor became sarcomatous after surgery and was given local radiotherapy after a second surgery and died 13 months after surgery. 2 patients with sacral osteoblastic giant cell tumor developed pulmonary metastases 2 years after surgery and were given 1 cycle of chemotherapy with adriamycin, cisplatin, and isocyclophosphamide. At 1 year follow-up, there was no significant enlargement of the pulmonary lesions.
Intraoperative bleeding: The mean blood loss was 3278 ml in the vascular block group and 5150 ml in the surgery for giant cell tumor of bone without vascular block. Vascular block significantly reduced intraoperative bleeding, thus it is necessary for preoperative vascular block for sacral giant cell tumor to increase the safety of surgery.
Local recurrence: The local recurrence rate was 30.43% (7/23) in the vascular blocked group and 66.67% (8/12) in the unblocked group. There were 5 patients with 2 recurrences; 10 patients with 1 recurrence.
From Table 1, it can be seen that there was a significant difference in the amount of bleeding in the vascular blocked group compared with the non-vascular blocked group (P=0.006). Considering that intraoperative bleeding is the result of multiple factors, we performed multiple regression analysis on the relevant factors affecting bleeding volume and compared the effects of patients’ gender, age, tumor size and vascular block on intraoperative bleeding volume, and found that tumor size and vascular block were the main influencing factors on the amount of bleeding.
Kaplan-Meier survival curve analysis was performed for the effect of sacral giant cell tumor surgery using vascular blocking technique on long-term tumor recurrence (see Figure 1), and it was found that there was a significant difference in tumor recurrence rate between the vascular blocking group and the non-vascular blocking group (P=0.03), and the difference in local recurrence rate between different blocking methods was not significant, and we can see from the survival curve that most recurrences The overall recurrence rate of the non-vascular blockade group reached 66.67%, and the overall recurrence rate of the vascular blockade group was 30.43%.
Neurological complications: 7 patients with only bilateral sacral 2 nerve roots preserved, 4 patients had partial impairment of bladder control and stool control; 13 patients with bilateral sacral 3 nerve roots preserved, sphincter function was preserved after surgery; 11 patients with unilateral sacral 3 nerve roots preserved, sphincter function was preserved in 7 cases after surgery. 4 patients with one side of sacral 2 and 3 removed, sphincter function was not abnormal after surgery. The four patients who had one side of sacral 2,3 removed did not have abnormal sphincter function after surgery. Only one patient showed motor dysfunction. The unilateral sacral 1 nerve root was removed intraoperatively, and mild postoperative paralysis was observed, but walking with the aid of a cane was possible.
Wound complications: 14 patients had postoperative wound complications (40%), 11 patients required surgical drainage, debridement, and second-stage wound closure, and 2 patients required local flap transfer due to large skin defects. 7 patients had postoperative cerebrospinal fluid leakage of varying degrees, all of which healed with non-operative treatment such as bed tail elevation and antibiotics.
DISCUSSION
(A) Characteristics of sacral osteoblastic giant cell tumor
Giant cell tumors of the sacral bone are rare and few cases have been reported in the literature, with the largest number of cases summarized as 26 reported by Dr. Turcotte of the Mayo Clinic (CORR 1993) (5). The present group of 35 cases of sacral giant cell tumor of bone is the largest number of cases reported internationally. Giant cell tumors of the sacrum are not easily detected in the early stages and grow to a large size compressing the sacral nerve with sciatica and, in severe cases, with urinary and fecal abnormalities. In sacral giant cell tumors, there are few cases of Campanacci grade I. There are significantly more cases of Campanacci grade III osteoblastoma in the sacrum than in the extremities (6-8). The literature reports that giant cell tumors of bone are more common in females, with Turcotte reporting 70% of the 26 cases of GCT of the sacrum in female patients (5). In our group, the number of female patients was 17/35, which differs from most of the literature reports, and it is not known if it is related to the difference in race.
Despite being a benign tumor, GCT of the sacrum is very challenging to treat in sacral giant cell tumors of bone. This is partly because of the anatomical site and partly because giant cell tumors of the sacrum often grow very large before they are detected and thus bleed a lot intraoperatively. Treatment options include radiotherapy, intra-lesional marginal resection, intra-lesional marginal resection combined with radiotherapy, intra-lesional marginal resection with cryosurgery, and extensive marginal resection. The obvious advantage of radiotherapy is to avoid the incidence of complications of surgical treatment. The main disadvantages of radiotherapy are the poor control of tumor growth and the initiation of sarcomatous transformation in about 10% of cases (9-11). None of the 35 patients in our group had received radiotherapy, and only one patient with local recurrence developed sarcomatous transformation. Intra-focal marginal resection avoids injury to nerve roots, the integrity of the pelvic ring, the hip joint and vascular structures. However, the disadvantage of this procedure is the high rate of local recurrence of the tumor. Extensive marginal resection minimizes the recurrence rate, but increases the complication rate of surgical treatment (12-14). In our group, none of the patients had received radiotherapy and the local recurrence after surgery for sacral giant cell tumor of bone was 11/35 (31.4%). This is similar to the local recurrence rate reported by Turcotte (33%), but most of their cases had received radiotherapy at 55 Gy postoperatively (5). There were three patients with two local recurrences in our group; eight patients with one recurrence. The rate of local recurrence is high in sacral giant cell tumors, which are not easy to be resected in their entirety because of the sacral nerve and pelvic organs.
(2) Choice of surgical approach for sacral giant cell tumor of bone
The aggressiveness of giant cell tumor of bone is very strong, and it can never be treated according to the treatment of general benign tumor, and the surgery of giant cell tumor of bone located in sacral area is more difficult and the recurrence rate is high. For patients with lesions located in sacral segment 3 and beyond, extensive marginal resection should be used; while for patients with lesions involving higher sacral segments (S1-2), excision and scraping are more often used. However, in cases of giant cell tumor of bone located in sacrum 1-5, wide marginal resection is used for the distal part of sacrum 2 and the proximal (S1-2) part is scraped.
The local recurrence rate of giant cell tumors of bone occurring in the sacrum is much higher than that of GCTs occurring at other sites, mainly due to the complexity of the anatomic site and the fact that the tumor is usually very large before diagnosis. Giant cell tumors of bone in the sacrum have a high local recurrence rate and poor disease regression. Except for extensive marginal resection of the tumor, the local recurrence rate after all other treatments is in the range of 40-50%. The actual local recurrence rate may be higher, as recurrence may also occur after a longer period of time (11,12).
The recommended treatment for giant cell tumors of bone occurring on the limb is expanded curettage. For giant cell tumors of the sacrum, intralesional marginal resection is performed in most cases due to the desire to preserve the sacral nerve, and the local recurrence rate can be more than 50%, but extensive marginal resection should be performed for sacral tumors in lower planes (below sacral 3), which can preserve rectal and bladder function.
Recent studies have reported that the advantages of treating tumors with intra-focal marginal resection combined with radiotherapy are not statistically different from those of treating them with one treatment alone. This combination treatment increased the rate of complications from surgical treatment and the risk of sarcomatous transformation after local radiotherapy, and did not statistically decrease the rate of local recurrence. There was no statistical difference in the rate of local recurrence after intra-focal marginal resection and radiotherapy. High-dose radiotherapy also did not significantly reduce the rate of local recurrence (12). Extensive marginal resection of the tumor significantly reduces the local recurrence rate but increases the incidence of complications of surgical treatment. The number of cases for cryotherapy as an adjuvant treatment after surgical treatment is too small to be statistically calculated (15).
There is no evidence to support intra-tumor margin resection combined with radiotherapy for sacral giant cell tumor of bone. Extensive marginal resection of the tumor may cause sacral nerve damage, but remains the best means of treatment because of the low rate of local recurrence. In the treatment of giant cell tumor of bone occurring on the sacrum, it is necessary to consider the possible situations such as local recurrence rate after treatment and complication rate of surgical treatment, and choose the treatment plan by weighing the pros and cons.
(C) Intraoperative bleeding control in sacral giant cell tumor of bone
Comparing the mean intraoperative blood loss in each group, the mean blood loss was 3278 ml in the vascular block group and 5150 ml in the non-vascular block group, and there was a statistically significant difference between the two groups (P=0.006). This indicates that preoperative vascular blockade is indeed necessary for sacral giant cell tumors, which can significantly reduce intraoperative bleeding and increase the safety of surgery.
When comparing the recurrence rate of each group, the recurrence rate of the vascular block group was significantly lower than that of the non-vascular block group, and there was a significant difference in the recurrence rate between the two groups (P<0.05), while there was no significant difference in the effect of different blocking methods on the recurrence rate. This may be related to the fact that the intraoperative blood loss in the vascular block group was less and the surgical field was clearly revealed, which was conducive to the complete resection of the tumor. The local recurrence rate of sacral giant cell tumors is much higher than that of giant cell tumors occurring at other sites, mainly due to the complexity of the anatomical site and the fact that the tumors are usually very large before diagnosis. In addition, sacral giant cell tumors are mostly located in high sacral bone, so incision and scraping are more often used during surgery to protect the nerve roots above sacral 3 bilaterally as much as possible. In 26 GCT cases reported by Turcotte (5), the average intraoperative blood loss was 7500 ml, and the local recurrence rate was 53.8% (14/26). The low intraoperative blood loss and clear surgical visualization after vascular block facilitates complete tumor resection. Clinical evidence suggests that controlling intraoperative bleeding is an important measure for complete tumor resection and reducing the recurrence rate. According to our clinical data, we found that the recurrence of sacral giant cell tumor is mostly within 6 months after surgery through Kaplan-Meier curve analysis of recurrence rate of different surgical methods, which suggests that the recurrence of tumor is related to the difficulty of complete removal of residual lesions intraoperatively, and reducing intraoperative bleeding can help improve surgical completeness and cure rate.
At present, there are two main types of traditional methods to control intraoperative bleeding. One is to perform an intraoperative anterior incision, ligate the affected internal iliac artery, and temporarily block the abdominal aorta, which can effectively control intraoperative bleeding. The second type is preoperative angiography and selective unilateral or bilateral embolization of the internal iliac artery and tumor supply vessels. In our experience, in most cases, the intraoperative control of bleeding after embolization alone is unsatisfactory. We applied angiographic technique to embolize the internal iliac artery while prepositioning a balloon in the abdominal aorta, and when intraoperative bleeding is high, the balloon can be temporarily blocked by injecting saline into the balloon, which can have the same effect of controlling bleeding as blocking the abdominal aorta in the antegrade surgery, and at the same time, significantly shorten the operation time. Moreover, the safety of the operation is improved by avoiding the corresponding complications caused by the separation of the blocked vessels in the anterior surgery. The application of intraoperative temporary block of the abdominal aortic balloon for sacral tumors as a new technique is a new attempt to effectively reduce intraoperative bleeding.
(IV) Reconstruction after sacral resection
In the past, sacral reconstruction was not performed for most patients after total or subtotal sacrectomy, and patients had to be bedridden for a long time after surgery, relying on the surgical scar to limit the descent of the spine. Most patients are able to walk upright without significant impact on spinal stability. In recent years, the development of internal fixation devices in the spine has been rapid, and as a result, many surgeons have performed internal fixation to reestablish spinal sacral stability in patients who have undergone total or subtotal sacrectomy (14). Patients can get out of bed early and do not have nerve root symptoms due to spinal instability. However, there are many complications associated with internal fixation in the sacral region, because after sacral resection, there is a large local cavity with no muscle layer on the dorsal side and only a skin flap covering it, thus, local fluid accumulation and even infection can easily occur. If local infection occurs, the internal fixation device must be removed. In our group of 35 cases of sacral giant cell tumor, 29 patients underwent reconstructive internal fixation of the lumbosacral region using the nail bar system. In one case, the internal fixation device had to be removed two months after surgery due to postoperative wound infection. In patients with high sacral resection or bilateral sacroiliac joint resection, internal fixation should be performed to rebuild the stability of the spinal sacrum. However, the possibility of wound infection must be taken into account. In patients with preserved sacrum 1, 2 or intact sacroiliac joints, sacral reconstruction is not necessary. At follow-up, no cases of loosening of the nail bar system were observed. The advantages of using the nail bar system to reconstruct the post-tumor resection defect of the iliac bone are a firm and strong fixation, early postoperative weight-bearing, and functional exercise one week after surgery.
Reference.
1 CDM. Fletcher, KK. Unni and F. Mertens. Pathology and Genetics of Tumors of Soft Tissue and Bone. WHO, 2002, p309-313.
2 F. Schajowicz. Tumors and Tumorlike Lesions of Bone. 2nd edition. Springer-Verlag, 1996, p257-295.
3 KK. Unni. Dahlin’s Bone Tumors. 5th Edition. Lippincott-Raven, 1996, p263-289.
4 Guo W, Xu W, Huvos AG, et al. Comparative frequency of bone sarcomas among different racial groups. Chin Med J (English Ed) , 1999, 112: 1101-1104.
5 Turcotte RE, Sim FH, and Unni KK. Giant Cell Tumor of the Sacrum. Clin Orthop, 1993, 291: 215-221.
6 JC. Cheng, JO. Johnston. Giant-cell tumor of bone: prognosis and treatment of pulmonary metastases. Clin Orthop, 1997, 338: 205-214.
7 M. Campanacci. Bone and Soft Tissue Tumors. 2nd edition. Springer-Verlag, 1999, p99-136.
8 Smith J, Wixon D, and Watson RC. Giant-cell tumor of the sacrum: clinical and radiologic features in 13 patients. J Can Assoc Radiol, 1979, 30:34.
9 Rock MG, Sim FH, Unni KK, et al. Secondary malignant giant-cell tumor of bone: Clinicopathological assessment of nineteen patients. J Bone Joint Surg, 1986, 68A:1073.
10 Dahlin DC and Unni KK: Bone Tumors: General Aspects and Data on 8542 Cases, ed. 4. Springfield, Illinois, Charles C Thomas, 1986, p119-140.
11 McDonald DJ, Sim FH, Mcleod RA, et al. Giant-cell tumor of bone. J Bone Joint Surg, 1986, 68A:235.
12 Leggon RE, Zlotecki R, Reith J, et al. Giant Cell Tumor of the Pelvis and Sacrum. Clin Orthop, 2004, 423:196-207
13 Biagini R, Demitri S, Orsini U, et al. Giant Cell Tumor of the Sacrum. Proceedings of the 10th International Symposium on Limb Salvage 51, 1999.
14 Doita M, Harada T, Iguchi T, rt al. Total Sacrectomy and Reconstruction for Sacral Tumors. Spine, 2003, 28:E296-E301.
15 Marcove RC, Sheth DS, Brien EW, et al. Conservative surgery for giant cell tumors of the sacrum: The role of cryosurgery as a supplement to curettage and partial excision. Cancer, 1994, 74: 1253-1260.