Revision Chiari surgery in adults: surgical evaluation algorithm and outcomes in a continuity-of-care practice

Abstract

OBJECTIVE

Patients with a prior surgical history for Chiari malformation (CM) as a child or an adult who presents with a recurrence or persistence of symptoms pose a challenging clinical scenario. The authors analyzed preoperative presentations, surgical histories, and outcomes of adult CM revision cases to create an algorithm for the assessment of revision surgery in adult CM patients and contrast this with the care of pediatric CM patients within the context of an integrated continuity-of-care CM practice seeing pediatric and adult patients.

METHODS

The authors retrospectively reviewed the records of patients with CM0, -1, and -1.5 who were 21 years of age or older, had a prior history of CM surgery as either children or adults, and underwent revision surgery as adults at the authors’ institution from December 2009 to February 2020. The cohort was divided into three groups depending on the type of revision surgery performed: posterior fossa decompression with or without duraplasty (PFD/D group), occipitocervical fusion without ventral decompression (OCF group), and ventral decompression in the form of endoscopic endonasal odontoidectomy (VD group).

RESULTS

A total of 50 adult revision cases were divided into 30 PFD/D cases, 13 OCF cases, and 7 VD cases. Forty cases had previously undergone one CM surgery, and 10 had undergone two. Preoperatively, the PFD/D group had significantly fewer cases of diagnosed craniocervical instability (CCI), Ehlers-Danlos syndrome, and dysautonomia than the OCF and VD groups.

A retroflexed odontoid was significantly more common in the VD group than in the PFD/D and OCF groups. Postoperatively, rates of improvement were 60.0% with PFD/D, 84.6% with OCF, and 85.7% with VD, but differences in these rates were not statistically significant.

CONCLUSIONS

In this challenging group of CM patients, possible etiologies warranting additional surgery included residual dorsal compression, persistent syrinx, CCI, and ventral brainstem compression from a significant retroflexed odontoid. Outcomes after revision were often favorable, but careful evaluation, patient selection, and communication with patients are needed in deciding to reoperate. The suggested algorithm can aid decision-making.