Paraspinal muscle morphometry in cervical spondylotic myelopathy and its implications in clinicoradiological outcomes following central corpectomy

Object The objective of this study was to assess the cross-sectional areas (CSAs) of the superficial, deep flexor (DF), and deep extensor (DE) paraspinal muscles in patients with cervical spondylotic myelopathy (CSM), and to evaluate their correlations with functional status and sagittal spinal alignment changes following central corpectomy with fusion and plating. Methods In this retrospective study of 67 patients who underwent central corpectomy with fusion and plating for CSM, the CSAs of the paraspinal muscles were calculated on the preoperative T2-weighted axial MR images and computed as ratios with respect to the corresponding vertebral body areas (VBAs) and as flexor/extensor CSA ratios. These ratios were then compared with those in the normative population and analyzed with respect to various clinicoradiological factors, including pain status, Nurick grade, and segmental angle change at follow-up (SACF). Results The mean CSA values for all muscle groups and the DF/DE ratio were significantly lower in the study cohort compared with an age- and sex-matched normative study group (p < 0.001). Among various independent variables tested in a multivariate regression analysis, increasing age and female sex significantly predicted a lower total extensor CSA/VBA ratio (p < 0.001), while a longer duration of symptoms significantly predicted a greater total flexor/total extensor CSA ratio (p = 0.02). In patients undergoing single-level corpectomy, graft subsidence had a positive correlation with SACF in all patients (p < 0.05), irrespective of the preoperative segmental angle and curvature, while in patients undergoing 2-level corpectomy, graft subsidence demonstrated such a correlation only in the subgroup with lordotic curvatures (p = 0.02). Among the muscle area ratios, the DF/DE ratio demonstrated a negative correlation with SACF in the subgroup with preoperative straight or kyphotic segmental angles (p = 0.04 in the single corpectomy group, p = 0.01 in the 2-level corpectomy group). There was no correlation of any of the muscle ratios with change in Nurick grade. Conclusions Patients with CSM demonstrate significant atrophy in all the flexor and extensor paraspinal muscles, and also suffer a reduction in the protective effect of a strong DF/DE CSA ratio. Worsening of this ratio significantly correlates with greater segmental kyphotic change in some patients. A physiological mechanism based on DF dysfunction is discussed to elucidate these findings that have implications in preventive physiotherapy and rehabilitation of patients with CSM. Considering that the influence of a muscle ratio was significant only in patients with hypolordosis, a subgroup that is known to have facetal ligament laxity, it may also be postulated that ligamentous support supersedes the influence of paraspinal muscles on postoperative sagittal alignment in CSM.