A novel spinal cord classification system: predict Intraoperative Neuromonitoring Event during correction of congenital kyphosis
10.3760/cma.j.cn121113-20230418-00245
- VernacularTitle:基于MRI的脊髓矢状面分型用于先天性脊柱后凸矫形术中神经电生理监测事件的风险预测
- Author:
Hui XU
1
;
Zhen JIN
;
Junyin QIU
;
Kiram ABDUKAHAR
;
Chen LING
;
Yanjie XU
;
Ziyang TANG
;
Jie LI
;
Zongshan HU
;
Zezhang ZHU
;
Yong QIU
;
Zhen LIU
Author Information
1. 南京大学医学院附属鼓楼医院骨科脊柱外科，南京 210008
- Keywords:
Kyphosis;
Spinal cord;
Magnetic resonance imaging;
Evoked potentials, somatosensory;
Evoked potentials, motor;
Classification
- From:
Chinese Journal of Orthopaedics
2023;43(17):1155-1163
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To propose a novel classification system based on the morphology and relative position of spinal cord in the spinal canal at sagittal T2-MRI, and to investigate the incidence and risk factors of the intraoperative neuromonitoring event (IONME) across these classifications.Methods:From January 2016 to December 2021, a consecutive cohort of 85 patients who underwent surgical correction of congenital kyphosis with pedicle screw/rod constructs were retrospectively reviewed, including 43 males and 42 females, aged 14.6±6.1 years old. According to the morphology and relative location of spinal cord at the apex of the curve on the sagittal-T2 MRI, patients were divided into three groups. Type A (5 cases) is characterized by the spinal cord centrally positioned within the spinal canal, surrounded by discernible cerebrospinal fluid (CSF). Type B (33 cases) depicts the spinal cord abutting the spinal canal's anterior wall, maintaining its intrinsic morphology. In Type C (47 patients), the spinal cord is contorted by the apical vertebral body, devoid of interposing CSF. The global kyphosis (GK) and sagittal deformity ratio (SDAR) of patients were measured before surgery. The incidence of IONME were recorded. All patients included in the study were further divided into the IONME group and the non-IONME group. Potential risk factors were identified using univariate testing. Binary Logistic Regression was used to analyze the independent risk factors for IONM.Results:All of 85 patients were reviewed: 5 (5.9%) Type A; 33 (38.8%) Type B; and 47 (55.3%) Type C spinal cords. Intraoperatively, 27 (31.8%) instances presented with lost trans-cranial motor-evoked potentials (MEPs) and/or somatosensory evoked potentials (SSEPs). Of these, 2 (7.4%) were Type B, and 25 (92.6%) were Type C, reflecting a statistically significant variance in IONME occurrences across types (χ 2=27.15, P<0.001). Notable differences were observed between IONME and non-IONME groups concerning GK, SDAR, and apex location ( t=5.41, P<0.001; t=3.65, P<0.001; χ 2=7.71, P=0.005). Univariate analysis showed that potential risk factors of IONME included Type C spinal cord ( OR=20.46, P<0.001), higher GK ( OR=1.07, P<0.001), SDAR ( OR=1.15, P=0.002) and apical vertebrae located at middle thoracic( OR=4.30, P=0.008). Independent predictors identified on binary Logistics regression modeling included higher GK ( OR=1.05, P=0.015), Type C spinal cord ( OR=6.22, P=0.042) and apex located at middle thoracic ( OR=6.43, P=0.021). Specifically, within Type C, 79% of cases where the apical vertebra was mid-thoracic experienced IONME, contrasting the 42% incidence observed in those with a lower thoracic apex positioning, signifying a notably elevated IONME likelihood for the mid-thoracic region (χ 2=5.16, P=0.023). Conclusion:Risk factors of IONME included Type C spinal cord, higher GK and apex located at middle thoracic during correction of congenital kyphosis. Preoperative MRI spinal cord typing showed great predictive value for IONME.
