Variables affecting postsurgical prognosis of thoracic myelopathy caused by ossification of the ligamentum flavum.

BACKGROUND CONTEXT: Ossification of the ligamentum flavum (OLF) may result in thoracic myelopathy (TM) because of narrowing of the spinal canal. Because symptoms vary and are subjective, diagnosis of TM caused by OLF is sometimes difficult when based on symptoms and physical examination. Posterior decompression is indicated in patients with TM caused by OLF because it is believed that surgery is the most effective treatment. However, surgical outcomes vary. We are unaware of reports of objective presurgical diagnostic parameters, such as neurologic and radiologic findings, relating to the postsurgical prognosis in patients with TM caused by OLF.

PURPOSE: To determine which presurgical and surgical variables were most closely related to postsurgical prognosis of TM caused by OLF.

STUDY

DESIGN

Retrospective review of the records of the cohort of patients who had undergone surgery from 1988 through 2008 at the University of Toyama Hospital for TM caused by OLF.

PATIENT SAMPLE: Forty-one patients who had surgery for TM caused by OLF that was progressive, severe, or both and for which the diagnosis was based on clinical, radiologic, and pathologic evaluations.

OUTCOME MEASURES: Relationship between the highest follow-up Japanese Orthopaedic Association (JOA) score for neurologic evaluation and of Hirabayashi's formula to indicate the extent of normalization after surgery with respect to the following 10 variables: age at surgery; sex; duration of presurgical symptoms; complications of diabetes mellitus; complications of hypertension; presence of presurgical hyperreflexia in either or both of the patellar tendon reflex and the Achilles tendon reflex; presurgical impairment of joint position sense in the big toes; number of levels affected by OLF; concurrent spinal lesions including ossification of the posterior longitudinal ligament; and intramedullary change of the spinal cord seen on magnetic resonance imaging (MRI).

METHODS

Multiple linear analyses were used to evaluate the variables related to postsurgical recovery.

RESULTS

Presurgical impairment of joint position sense in the big toe was the most important predictor of the highest postsurgical JOA score and of the highest percentage recovery rate. The number of affected OLF levels also predicted the postsurgical highest JOA score, but not statistically significantly so. Age at surgery, sex, and duration of symptoms presurgically did not affect postsurgical recovery. Complications of diabetes mellitus or of hypertension did not affect percentage recovery rate. The difference between recovery rate in patients with or without concurrent spinal lesions was not significant. Presurgical hyperreflexia was not correlated with recovery. Postsurgical JOA scores and percentage recovery rates of scores in patients whose presurgical MRIs had shown intramedullary signal change were not statistically significantly different from those whose MRIs had not shown signal change.

CONCLUSIONS

An excellent postoperative prognosis is not always possible in patients with TM caused by OLF. It may be important to check for impairment of joint position sense in the big toe, the number of levels affected by OLF, and presurgical intramedullary signal change on MRI before continuing to surgery.