Neurological deterioration after foramen magnum decompression for Chiari malformation Type I: old or new pathology?

Object Decompression of the foramen magnum is widely accepted as the procedure of choice for patients with Chiari malformation Type I (CM-I). This study was undertaken to determine the mechanisms responsible for neurological deterioration after foramen magnum decompression and the results of secondary interventions. Methods Between 1987 and 2010, 559 patients with CM-I presented, 107 of whom had already undergone a foramen magnum decompression, which included a syrinx shunt in 27 patients. Forty patients who were neurologically stable did not undergo another operation. Sixty-seven patients with progressive symptoms received a recommendation for surgery, which was refused by 16 patients, while 51 patients underwent a total of 61 secondary operations. Hospital and outpatient records, radiographic studies, and intraoperative images were analyzed. Additional follow-up information was obtained by telephone calls and questionnaires. Short-term results were determined after 3 and 12 months, and long-term outcomes were evaluated using Kaplan-Meier statistics. Results Sixty-one secondary operations were performed after a foramen magnum decompression. Of these 61 operations, 15 involved spinal pathologies not related to the foramen magnum (spinal group), while 46 operations were required for a foramen magnum issue (foramen magnum group). Except for occipital pain and swallowing disturbances, the clinical course was comparable in both groups. In the spinal group, 5 syrinx shunt catheters were removed because of nerve root irritations or spinal cord tethering. Eight patients underwent a total of 10 operations on their cervical spine for radiculopathies or a myelopathy. No permanent surgical morbidity occurred in this group. In the foramen magnum group, 1 patient required a ventriculoperitoneal shunt for hydrocephalus 7 months after decompression. The remaining 45 secondary interventions were foramen magnum revisions, of which 10 were combined with craniocervical fusion. Intraoperatively, arachnoid scarring with obstruction of the foramen of Magendie was the most common finding. Complication rates for foramen magnum revisions were similar to first decompressions, whereas permanent surgical morbidity was higher at 8.9%. Postoperative clinical improvements were marginal in both surgical groups. With the exception of 1 patient who underwent syrinx catheter removal and had a history of postoperative meningitis, all patients in the spinal group were able to be stabilized neurologically. Long-term results in the foramen magnum group revealed clinical stabilizations in 66% for at least 5 years. Conclusions Neurological deterioration in patients after a foramen magnum decompression for CM-I may be related to new spinal pathologies, craniocervical instability, or recurrent CSF flow obstruction at the foramen magnum. Whereas surgery for spinal pathologies is regularly followed by clinical stabilization, the rate of long-term success for foramen magnum revisions was limited to 66% for 5 years due to severe arachnoid scarring in a significant proportion of these patients. Therefore, foramen magnum revisions should be restricted to patients with progressive symptoms.