The effects of diabetes on the recovery of motor nerve function after cervical decompression surgery and associated mechanisms

Background To analyse the effect of diabetes on motor nerve function recovery in patients who underwent cervical decompression surgery and to explore the possible mechanisms of this effect through animal experiments. Methods The medical records and follow-up data for patients with cervical spondylosis who underwent cervical spine surgery from January 2021 to December 2023 were retrospectively analysed. The patients were divided into diabetes mellitus (DM) and non-DM groups. The clinical characteristics, preoperative and postoperative Japanese Orthopaedic Association (JOA) scores, JOA recovery rates (JOA-RR), neck disability index (NDI) values, and visual analogue scale (VAS) scores for the two groups were compared, and factors independently associated with motor nerve function recovery were identified via multivariable linear regression analysis. The findings of these analyses were validated in animal experiments involving adult male Sprague‒Dawley (SD) rats with type 2 DM and the same number of healthy SD rats as the control group. Both groups of rats underwent surgery to model incomplete cervical spinal cord injury. The forelimb locomotor assessment scale (FLAS) was used to evaluate the forelimb movement of the rats in the two groups on the 1 ^st , 7 ^th , 14 ^th , and 30 ^th days after surgery, and motor-evoked potentials (MEPs) were measured. The numbers of neurons and functional changes in the axons and other organelles of the samples at the site of injury to the cervical spinal cord were determined via electron microscopy and light microscopy on the day of and 30 days after the operation. Results A total of 129 patients who underwent cervical spine surgery were analysed in this study, including 59 in the DM group and 70 in the non-DM group. The median age, mean preoperative glycosylated haemoglobin (HbA1c) level, mean preoperative glucose level, and mean volume of intraoperative bleeding in the DM group were greater than those in the non-DM group, whereas the JOA, JOA-RR, NDI, VAS neck (VAS (N)), and VAS limbs (VAS (L)) scores within six months after surgery were greater for the non-DM group than for the DM group. Multivariate linear regression suggested that age and the preoperative HbA1c level were independently associated with the postoperative JOA score, the preoperative blood glucose level was independently associated with the postoperative NDI, and the surgical segment and preoperative blood glucose level were independent risk factors for the postoperative VAS (L) score. The animal experiments revealed that both groups of rats began to recover motor nerve function within 7 days after incomplete cervical spinal cord injury; the FLAS score of the non-DM group on the 7 ^th and 14 ^th days was greater than that of the DM group, whereas the FLAS score on the 30 ^th day was not significantly different between the two groups. Compared with those of individuals in the non-DM group, the latency and amplitude of upper-extremity MEPs were both lower for individuals in the DM group. The latencies of the lower-extremity MEPs were similar, but their amplitudes were lower for the DM group than for the non-DM group. Under light microscopy, the individuals in the DM group presented more severe spinal cord tissue necrosis, a more severely scattered central canal, and fewer neurons in the affected area than did the individuals in the non-DM group. Under high-resolution electron microscopy, in the DM group, the axons exhibited lamellar separation and tissue swelling, and the tissue structure was incomplete on the 30 ^th day after surgery. Conclusions The poor recovery of motor nerve function observed in DM patients following cervical decompression may be related to neuronal necrosis and loss and axonal degeneration caused by the high-glucose environment in these individuals.