OBJECTIVE In patients with cervical spondylotic myelopathy (CSM), the motor system may undergo progressive functional/structural changes rostral to the lesion, and these changes may be associated with clinical disability. The extent to which these changes have a prognostic value in the clinical recovery after surgical treatment is not yet known. In this study, magnetic resonance spectroscopy (MRS) was used to test 2 primary hypotheses. 1) Based on evidence of corticospinal and spinocerebellar, rubro-, or reticulospinal tract degeneration/dysfunction during chronic spinal cord compression, the authors hypothesized that the metabolic profile of the primary motor cortices (M1s) and cerebellum, respectively, would be altered in patients with CSM, and these alterations would be associated with the extent of the neurological disabilities. 2) Considering that damage and/or plasticity in the remote motor system may contribute to clinical recovery, they hypothesized that M1 and cerebellar metabolic profiles would predict, at least in part, surgical outcome. METHODS The metabolic profile, consisting of N-acetylaspartate (NAA; marker of neuronal integrity), myoinositol (glial marker), choline (cell membrane synthesis and turnover), and glutamate-glutamine (glutamatergic system), of the M1 hand/arm territory in each hemisphere and the cerebellum vermis was investigated prior to surgery in 21 patients exhibiting weakness of the upper extremities and/or gait abnormalities. Age- and sex-matched controls (n = 16) were also evaluated to estimate the pre-CSM metabolic profile of these areas. Correlation and regression analyses were performed between preoperative metabolite levels and clinical status 6 months after surgery. RESULTS Relative to controls, patients exhibited significantly higher levels of choline but no difference in the levels of other metabolites across M1s. Cerebellar metabolite levels were indistinguishable from control levels. Certain metabolites-myo-inositol and choline across M1s, NAA and glutamate-glutamine in the left M1, and myo-inositol and glutamate-glutamine in the cerebellum-were significantly associated with postoperative clinical status. These associations were greatly improved by including preoperative clinical metrics into the models. Likewise, these models improved the predictive value of preoperative clinical metrics alone. CONCLUSIONS These preliminary findings demonstrate relationships between the preoperative metabolic profiles of two remote motor areas and surgical outcome in CSM patients. Including preoperative clinical metrics in the models significantly strengthened the predictive value. Although further studies are needed, this investigation provides an important starting point to understand how the changes upstream from the injury may influence the effect of spinal cord decompression.
Keywords: 1H-MRS; 1H-MRS = proton MRS; 9-HPT = 9-hole peg test; CSM = cervical spondylotic myelopathy; M1 = primary motor cortex; MRS = MR spectroscopy; NAA = N-acetylaspartate; SCI = spinal cord injury; cervical spondylotic myelopathy; clinical outcome; mJOA = modified Japanese Orthopaedic Association; mJOA-LE = mJOA lower-extremity motor domain; mJOA-UE = mJOA upper-extremity sensorimotor domain; neuroinflammation; remote motor system; spinal decompression surgery.