MR-based evaluation of osseous cervical neural foraminal stenosis (OCNFS) using fast-field-echo resembling CT using restricted echo-spacing (FRACTURE)
摘要
This study aimed to assess the feasibility of fast-field-echo resembling CT using restricted echo-spacing (FRACTURE) for evaluating osseous cervical neural foraminal stenosis (OCNFS).
MethodsA total of 280 cervical foramina from 28 patients with cervical radiculopathy (CR) were analyzed on oblique sagittal CT, FRACTURE, T1WI, and T2WI from bilateral C2/3 to C6/7 segments by three readers. CT served as the reference standard to identify OCNFS. Detection rate and grade of OCNFS were assessed. OCNFS severity was classified into four grades according to Park’s system: Grade 0 (no stenosis), Grade 1 (mild stenosis), Grade 2 (moderate stenosis), and Grade 3 (severe stenosis). Inter-modality agreements of OCNFS grade were evaluated using linear weighted Cohen’s kappa (κ). Sensitivity and specificity were calculated to compare diagnostic performance between FRACTURE and conventional MRI for the diagnosis of OCNFS.
ResultsThe inter-modality agreement of OCNFS grade was substantial between CT and FRACTURE (κ = 0.689–0.715), and moderate between CT and T1WI (κ = 0.427–0.594) or T2WI (κ = 0.462–0.604). Among the 280 cervical neural foramina, 104 were diagnosed as OCNFS using CT. Detection rate of OCNFS was 96.15% (100/104) for FRACTURE, 76.92% (80/104) for T1WI, and 95.19% (99/104) for T2WI. FRACTURE achieved a sensitivity of 96.2% and a specificity of 97.2% for detecting OCNFS. T1WI demonstrated a sensitivity of 76.9% and a specificity of 85.2%, while T2WI exhibited a sensitivity of 95.2% and a specificity of 73.9% for detecting OCNFS.
ConclusionFRACTURE can provide a reliable alternative to CT for evaluating OCNFS.