Incidence of C5 Palsy: Meta-Analysis and Potential Etiology.

Abstract

Among the proposed hypotheses for C5 palsy, the most acceptable etiologies have been the tethering effect of the spinal nerve or reperfusion injury of the spinal cord. We performed a meta-analysis to determine a potential etiology of C5 palsy. The PubMed, Embase, Web of Science, and Cochrane Library databases were searched up to 2017 for relevant studies of the incidence of C5 palsy after cervical decompression surgery. Relevant incident estimates of C5 palsy stratified by the surgical approach and underlying diagnosis were calculated using an appropriate meta-analysis. A total of 107 studies were included for our meta-analysis. The pooled incidence of C5 palsy was 0.00 (95% confidence interval [CI], 0.00-0.01) for anterior decompression in patients with radiculopathy, 0.04 (95% CI, 0.03-0.05) for anterior decompression and 0.07 (95% CI, 0.06-0.08) for posterior decompression in patients with myelopathy. A two-tailed t test with unequal variance accepted the null hypothesis of no differences in the incidence between anterior and posterior decompression in patients with myelopathy (P = 0.999) and rejected the hypothesis of no differences in the incidence between the patients with radiculopathy and myelopathy in anterior decompression (P < 0.001). With no significant difference found in the incidence of C5 palsy between anterior and posterior cervical decompression in patients with myelopathy, the tethering effect of the spinal nerve might not be a plausible etiology for C5 palsy. In contrast, we found a significantly greater incidence of C5 palsy in patients with myelopathy compared with those with radiculopathy undergoing anterior decompression, which might support the reperfusion injury of the spinal cord as the etiology of C5 palsy.