Analysis of cerebrospinal fluid flow dynamics and morphology in Chiari I malformation with cine phase-contrast magnetic resonance imaging

Background

To determine cerebrospinal fluid (CSF) dynamics and morphology in Chiari I malformation (CMI) and assess the response to surgery of the posterior cranial fossa, we examined midsagittal imaging along with anterior cervical 2-3 (AC2-3), posterior cervical 2-3 (PC2-3), and aqueduct CSF flow hydrodynamics in axial imaging by using cine phase-contrast magnetic resonance imaging (PCMR).

Method

We examined 52 patients with CMI, both with and without syringomyelia (SM), pre-/post-surgery, and compared them to 17 healthy volunteers. Statistical analyses included paired t-tests, independent-samples t-tests, binary logistic regression, and crosstab with MedCalc software.

Results

Patients with CMI had significantly shorter clivus length and larger tentorial angle than the healthy controls (P?=?0.004, P?=?0.019, respectively). The AC2-3 cranial/caudal peak velocity (PV), PC2-3 cranial/caudal PV and aqueduct cranial peak PV of patients with CMI were significantly lower than healthy volunteers pre-surgery (P?=?0.034 AC2-3 cranial PV, P?=?0.000002 AC2-3 caudal PV; P?=?0.046 PC2-3 cranial PV, P?=?0.015 PC2-3 caudal PV; P?=?0.022 aqueduct cranial PV) and increased after surgery (P?=?0.024 AC2-3 cranial PV, P?=?0.002 AC2-3 caudal PV; P?=?0.001 PC2-3 cranial PV, P?=?0.032 PC2-3 caudal PV; P?=?0.003 aqueduct cranial PV). The aqueduct caudal PV of patients with CMI was higher than that of healthy controls (P?=?0.004) and decreased post-surgery (P?=?0.012). Patients with pre-surgery PC2-3 cranial PV >2.63 cm/s and aqueduct cranial PV >2.13 cm/s, respectively, experienced primary symptom improvement after surgery.

Conclusions

The innate bony dysontogenesis in patients with CMI contributes to tonsilar ectopia and exacerbates CSF flow obstruction. A pressure gradient that existed between SM and SAS supports the perivascular space theory that is used to explain SM formation. Our findings demonstrate that PCMR maybe a useful tool for predicting patient prognosis.