Predictors of neurologic recovery in acute central cervical cord injury with only upper extremity impairment

Study Design. A prospective study of 22 patients with the syndrome of acute central cervical spinal cord injury with motor and sensory impairment involving only upper extremities (ACCSCI-U) was done from admission to the 2-year follow-up. Objectives. To evaluate the course of neurologic function and to identify predictors of a good neurologic recovery in patients with ACCSCI-U. Summary of Background Data. Little is known regarding,the time-course profile of neurologic and functional recovery and the prognostic factors for neurologic recovery in patients with ACCSCI-U. Methods, Between 1997 and 1998, 22 new cases of ACCSCI-U were identified (15 men and 7 women) with a mean age of 45.9 years (range, 13-75 years). They were all treated nonsurgically. Neurologic impairment, as evaluated by the American Spinal Injury Association (ASIA) motor and sensory scores, was assessed on patient arrival at the emergency department and followed prospectively at intervals from admission to the 2-year follow-up. The magnetic resonance imaging (MRI) studies were performed acutely (1 hours to 2 days after injury), subacutely (3-21 days after injury), and chronically (3-9 months after injury). Logistic regression analysis was used to identify the variables that were associated with neurologic improvement. Results. No patient sustained cervical fractures. The mean sagittal diameter of the cervical spinal canals (14.8 mm) was smaller than that of normal subjects in Japan. Of these 22 patients, seven (32%) had radiologic evidence of developmental cervical canal stenosis and two (9%) had evidence of ossification of posterior longitudinal ligament. Thirteen (59%) had cervical spondylosis: of these, four (31%) had cervical stenosis. Nonsurgical treatment resulted in nearly full neurologic recovery within approximately 6 weeks after the injury. None of the patients went on to need surgery. Their average ASIA score of motor, light touch sensation, and pin sensation was increased from 82.2, 104:5, and 104.3 at admission to 99.3, 111.2, and 111.0, respectively, at 2 years after injury. Logistic regression analysis revealed that the best predictor associated with a better neurologic recovery of motor, light touch sensation, and pin sensation was the absence of MRI findings of abnormal signal intensity in the spinal cord, with an odds ratio of 64 (P<0.006), 13 (P<0.04),and 15 (P<0.03), respectively. The results also revealed that percent deficit improvement became significant at 1 week after injury to predict neurologic recovery at 2-year follow-up points. Conclusions. In patients with ACCSCI-U, a favorable neurologic prognosis can be predicted following nonsurgical treatment. Most recovery occurred by 6 weeks, and patients with severe initial neurologic damage and old age had poorer recovery. Touch and pin sensation recovered at the same rate, whereas motor recovered more quickly. The absence of abnormal MRI signal intensity in the spinal cord and a good early neurologic improvement were the significant predictors of long-term improvement in neurologic function.