Changes of improvement in upper limb function predict surgical outcome after laminoplasty in 1 year in patients with cervical spondylotic myelopathy: a retrospective study.

BACKGROUND

Cervical spondylotic myelopathy preoperative prognostic factors include age, preoperative severity, and disease duration. However, there are no reports on the relationship between changes in physical function during hospitalization and postoperative course, and in recent years, the length of hospital stay has shortened. We aimed to investigate whether changes in physical function during hospitalization can predict the postoperative outcome.

METHODS

We recruited 104 patients who underwent laminoplasty for cervical spondylotic myelopathy by the same surgeon. Physical functions, including Simple Test for Evaluating Hand Function (STEF), grip strength, timed up and go test, 10-m walk, and time to stand on one leg, were assessed at admission and discharge. Patients with the Japanese Orthopaedic Association (JOA) score improvement rate of 50% or more were defined as the improved group. Decision tree analysis was investigated factor for identifying improvement in the JOA score. According to this analysis, we divided into two groups using age. Then, we conducted a logistic regression analysis to identify factors that improve the JOA score.

RESULTS

The improved and non-improved groups had 31 and 73 patients, respectively. The improved group was younger (p = 0.003) and had better improved Δgrip strength (p = 0.001) and ΔSTEF (p < .0007). Age was significantly positively correlated with disease duration (r = 0.4881, p =  < .001). Disease duration exhibited a significant negative correlation with the JOA score improvement rate (r = - 0.2127, p = 0.031). Based on the decision tree analysis results, age was the first branching variable, with 15% of patients ≥ 67 years showing JOA score improvement. This was followed by ΔSTEF as the second branching factor. ΔSTEF was selected as the factor associated with JOA improvement in patients ≥ 67 years (odds ratio (OR) 0.95, 95% confidence interval (CI) 0.90-0.99, p = .047); in patients < 67 years, Δgrip strength was identified (OR 0.53, CI 0.33‒0.85, p = .0086).

CONCLUSIONS

In the improved group, upper limb function improved more than lower limb function from the early postoperative period. Upper limb function changes during hospitalization were associated with outcomes one year postoperatively. Improvement factors in upper extremity function differed by age, with changes in grip strength in patients < 67 years and STEF in patients ≥ 67 years, reflecting the outcome at one year postoperatively.