Classification of Chiari I malformation based on various cerebrospinal fluid flow patterns at the cranial-vertebral junction and selection of surgical procedure

Abstract

Objective To discuss about the classification of Chiari Ⅰ malformation based on different CSF (cerebrospinal fluid) flow patterns at the cranial-vertebral junction (CVJ) and to investigate its surgical strategies and outcomes. Methods A total of 126 patients with Chiari Ⅰ malformation were admitted to Department of Neurosurgery, Sanbo Brain Hospital, Capital Medical University from January 2008 to December 2015 and retrospectively enrolled into this study. According to the preoperative findings obtained by using cine phase-contrast MRI (cine PC-MRI) and based on the abnormal CSF flow dynamics at the CVJ, all those patients of Chiari Ⅰ malformation were classified into 3 patterns: type Ⅰ (36 cases), type Ⅱ (48 cases) and type Ⅲ (42 cases). Intraoperative ultrasound to detect cerebrospinal fluid again determined the surgical strategy. In type Ⅰ patients, 34 patients underwent subdural decompression, and the remaining 2 patients underwent subarachnoid decompression due to cerebrospinal fluid obstruction. In type Ⅱ patients, 36 patients underwent subarachnoid decompression and the remaining 12 cases underwent subdural decompression. In type Ⅲ patients, 40 cases underwent subarachnoid decompression, and the remaining 2 cases underwent subdural decompression due to unobstructed cerebrospinal fluid flow. Two weeks after surgery, the patients underwent an imaging review and assessment based on cervical spine modified Japanese Orthopaedic Association (mJOA) score. All patients underwent clinical follow-up. The follow-up included the use of the cervical spine mJOA score to assess the improvement of syringomyelia-related symptoms. The Chicago Chiari Outcome Scale (CCOS) was used to assess postoperative neurological recovery. Results All 126 patients had successful surgery. There were no cases of new neurological deficits after operation. Eight cases (6.3%) had fever, and 5 cases (4.0%) had cerebrospinal fluid leakage, which improved after symptomatic treatment. Two weeks after surgery, imaging examination showed that there was no significant improvement in syringomyelia in 2 patients (type Ⅲ, subarachnoid decompression). Further improvement of symptoms was observed after syringomyelia-thoracic shunt. The mJOA score was improved compared with preoperative conditions (12.7±2.0 vs. 10.7±1.6, P<0.01). The follow-up time of 126 patients was 24.8 ± 10.9 months (12 to 96 months). At the last follow-up, the cervial spine mJOA scores of 126 patients were significantly higher than those before surgery (12.8±1.9 vs. 10.7±1.6, P<0.01), but there was no significant difference when compared with two weeks after surgery (P=0.48). There was no significant difference in CCOS score compared with 2 weeks post surgery (14.7±1.5 vs. 14.4±1.5, P=0.576). Conclusions Chiari malformation type Ⅰ could be classified into 3 subtypes according to cerebrospinal fluid dynamic abnormalities in the cranial-cervical junction area. Type Ⅰ is suitable for subdural decompression and type Ⅲ for subarachnoid decompression. In type Ⅱ, specific decompression methods could be chosen based on intraoperative ultrasound through bone window after craniectomy. The postoperative results seem good. Key words: Arnold-Chiari malformation; Hydrodynamics; Neurosurgical procedures; Treatment outcome; Cerebrospinal fluid; Typing