Resection of intramedullary spinal cord tumors in children: assessment of long-term motor and sensory deficits.

Object: With modern surgical advances, radical resection of pediatric intramedullary spinal cord tumors (IMSCTs) can be expected to preserve long-term neurological function. Nevertheless, postoperative neurological decline is not uncommon after surgery, and many patients continue to experience long-term dysesthetic symptoms. Preoperative predictors of postoperative neurological decline and sensory syndromes have not been investigated and may serve as a guide for surgical risk stratification.

Methods: Neurological function (as determined using the modified McCormick Scale [mMS]) preoperatively, postoperatively, and 3 months after surgery was retrospectively recorded from patient charts in 164 consecutive patients undergoing resection of IMSCTs. A median 4 years (interquartile range [IQR] 1-8 years) after surgery, long-term motor and sensory symptoms were assessed by telephone interviews and corroborated by subsequent medical visits in 120 available patients. This long-term assessment was retrospectively reviewed for the purposes of this study. The authors reviewed this series to assess long-term motor, sensory, and urinary outcomes and to determine independent risk factors of postoperative neurological decline and long-term sensory dysfunction.

Results: Patients were 8.6 +/- 5.7 years old and presented with a median mMS of 2 (IQR 2-4). Three months after surgery, 38 patients (23%) continued to experience decreased neurological function (1 mMS point) incurred perioperatively. Increasing age (p = 0.028), unilateral symptoms (p = 0.046), and urinary dysfunction at presentation (p = 0.004) independently predicted persistent 3-month perioperative decline. At long-term follow-up (median 4 years), 39 (33%) exhibited improvements in their mMS scores, 13 (30%) had improvement in their urinary dysfunction, and 27 (30%) had resolution of their dysesthesias. Seventy-eight patients (65%) experienced long-term dysesthetic symptoms. Increasing age (p = 0.024), preoperative symptom duration > 12 months (p = 0.027), and worsened postoperative mMS score at hospital discharge (p = 0.013) independently increased the risk of long-term dysesthesias.

Conclusions: In the authors' experience, nearly one third of patients may experience improvement in motor, sensory, and urinary dysfunction years after IMSCT resection, whereas the majority will continue to experience long-term dysesthetic symptoms. Improvement in motor deficits preceded improvement in sensory syndromes, and urinary dysfunction typically resolved much longer after surgery. The risk of persistent perioperative motor decline was increased with older age, unilateral symptoms, preoperative urinary symptoms, and less severe preoperative neurological deficit. The risk of long-term dysesthesias was increased with older age, increased duration of symptoms prior to resection, and greater postoperative neurological deficit.