Cerebellar tonsil reduction for surgical treatment of Chiari malformation type I in children

Abstract

OBJECTIVE The goal of this study was to review the efficacy and safety of different surgical techniques used for treatment of Chiari malformation type I (CM-I) in children. METHODS The authors retrospectively reviewed 437 consecutive children surgically treated for CM-I. Procedures were classified into four groups: bone decompression (posterior fossa decompression [PFD]) and duraplasty (PFD with duraplasty [PFDD]), PFDD with arachnoid dissection (PFDD+AD), PFDD with tonsil coagulation of at least one cerebellar tonsil (PFDD+TC), and PFDD with subpial tonsil resection of at least one tonsil (PFDD+TR). Efficacy was measured as a greater than 50% reduction in the syrinx by length or anteroposterior width, patient-reported improvement in symptoms, and rate of reoperation. Safety was measured as the rate of postoperative complications. RESULTS The mean patient age was 8.4 years (range 3 months to 18 years). In total, 221 (50.6%) patients had syringomyelia. The mean follow-up was 31.1 months (range 3–199 months), and there was no statistically significant difference between groups (p = 0.474). Preoperatively, univariate analysis showed that non-Chiari headache, hydrocephalus, tonsil length, and distance from the opisthion to brainstem were associated with the surgical technique used. Multivariate analysis demonstrated that hydrocephalus was independently associated with PFD+AD (p = 0.028), tonsil length was independently associated with PFD+TC (p = 0.001) and PFD+TR (p = 0.044), and non-Chiari headache was inversely associated with PFD+TR (p = 0.001). In the treatment groups postoperatively, symptoms improved in 57/69 (82.6%) PFDD patients, 20/21 (95.2%) PFDD+AD patients, 79/90 (87.8%) PFDD+TC patients, and 231/257 (89.9%) PFDD+TR patients, and differences between groups were not statistically significant. Similarly, there was no statistically significant difference in postoperative Chicago Chiari Outcome Scale scores between groups (p = 0.174). Syringomyelia improved in 79.8% of PFDD+TC/TR patients versus only 58.7% of PFDD+AD patients (p = 0.003). PFDD+TC/TR remained independently associated with improved syrinx outcomes (p = 0.005) after controlling for which surgeon performed the operation. For those patients whose syrinx did not resolve, no statistically significant differences between surgery groups were observed in the length of follow-up or time to reoperation. Overall, there was no statistically significant difference between groups in postoperative complication rates, including aseptic meningitis and CSF- and wound-related issues, or reoperation rates. CONCLUSIONS In this single-center retrospective series, cerebellar tonsil reduction, by either coagulation or subpial resection, resulted in superior reduction of syringomyelia in pediatric CM-I patients, without increased complications.