Surgical Treatment of Sacral Neurogenic Tumor: A 10-year Experience with 64 Cases

Objective: To explore the outcomes of surgical treatment of sacral neurogenic tumors

Methods: Between 1 January 2003 and 31 December 2012, data on 64 patients with sacral neurogenic tumors treated with surgery were retrospectively analyzed. The mean age of the 64 cases (35 males and 29 females) was 37.2 years (range, 21-69 years); 38 had neurilemmomas and 26 neurofibromas. Thirty-four of the tumors involved S 1 and S 2 , 11 S 3 or lower, and 19 were single presacral soft tissue masses. Tumors were removed via anterior, posterior or combined anteroposterior approaches. Patients with unstable sacroiliac joints underwent iliolumbar fixation.

Results: Depending on the extent of tumor involvement, one of three surgical approaches was used: a single anterior approach (19 patients), single posterior approach (25 patients), or a combined anteroposterior approach (20 patients). The mean operation time was 3 h (range, 2-6 h) and the mean blood loss 878 mL (range, 400-3120 mL). The mean duration of follow-up was 58.2 months (range, 24-93 months). These surgeries had the following complications. Three patients had massive intraoperative hemorrhage and posterior back pain and discomfort postoperatively. One patient had intraoperative ureteral injuries requiring intraoperative ureteral catheterization. In two patients, the tumor involved the S 1 nerve roots bilaterally, necessitating their removal, which resulted in obvious lower limb motion and sphincteric dysfunction. In 13 patients with unilateral tumor involvement of the nerve roots of S 1 and lower spinal levels, only the contralateral nerve roots of the S1 and lower levels were preserved; eight of these patients had impaired bladder and bowel function. Posterior incisions failed to heal in 10 patients, secondary wound healing occurred in nine of them and one required a gluteus maximus myocutaneous flap. Three patients developed postoperative cerebrospinal fluid leaks that were and alleviated by waist belt compression bandaging and placing them in the Trendelenburg position. Eight patients developed tumor recurrences postoperatively; pathological examination of the tissue excised in the second surgeries revealed malignant changes in the three patients with neurilemmomas. There were no intraoperative deaths. Rod fractures occurred in three of the 18 patients requiring iliolumbar reconstruction.

Conclusions: The clinical characteristics of sacral neurogenic tumors make them easy to diagnose. The approach to resection should be determined by the location and size of the tumor. Patients with huge tumors may lose considerable blood intraoperatively and a have higher risk rate of postoperative complications.