[Treatment of subcutaneous fistula secondary to cerebrospinal fluid leakage in thoracic spinal stenosis cases].

OBJECTIVE: To investigate the treatment strategy for subcutaneous fistula secondary to cerebrospinal fluid leakage (CSFL) in thoracic spinal stenosis (TSS) cases.

METHODS: In the study, 186 CSFL cases diagnosed with TSS and operated in general spine group of Department of Orthopedics, Peking University Third Hospital from January 2005 to December 2014 were retrospectively reviewed, of which eleven had subcutaneous fistula secondary to CSFL and were regularly followed up. Treatment strategy for subcutaneous fistula depended on the severity of CSFL and the recovery rate of thoracic myelopathy. Japanese Orthopedic Association (JOA) score was utilized to evaluate the neurologic status of these patients preoperatively and postoperatively. Statistical analysis was conducted between preoperative and postoperative JOA scores.

RESULTS: All of the 11 patients were regularly followed up for at least 24 months. Six of them had ossification of the posterior longitudinal ligament (OPLL) combined with ossification of ligamentum flavum (OLF), all of them undertook "cave-in" 360° circumferential decompression of the spinal cord with instrumentation. Five cases had OLF only, and received En bloc resection of lamina and OLF and fixation. The follow-up period ranged from 30 months to 131 months, and averaged at (85±34) months. Preoperative symptoms lasted from 3 months to 8 years, and the median was 18 months. Drainages were placed for 2-6 days, and averaged at (4.2±1.1) days. Ten cases appeared with fever during the perioperative period, the maximum body temperature was (37.3-39.7) °C. Prolonged antibiotics were applied in two cases with high fever. Ten cases were treated with conservative methods, CSFL were completely absorbed during the follow-up time, of which compressive dressing was utilized in 8 cases, and punctures combined with compressive dressing were used in 2 cases. For only 1 case, conservative therapy failed and reoperation was required because of neurological deterioration arising from CSF pseudocyst. For these 11 cases, preoperative JOA score arose from (3.8±1.6) preoperatively to (8.9±1.2) at the end of the final follow-up, the recovery rate was 70.8%. No infection of wound or central nerve system were noticed, and neither were unhealing wound.

CONCLUSION: Most TSS cases with subcutaneous fistula secondary to CSFL could be cured by conservative methods, and reoperation is required only if myelopathy caused by cerebrospinal fluid pseudocyst is identified.