Correlation of cord signal change with physical examination findings in patients with cervical myelopathy.

STUDY

DESIGN

Retrospective case series.

OBJECTIVE

To determine whether cord signal change (CSC) visualized on magnetic resonance imaging (MRI) correlates with level-specific physical examination findings as well as other signs of cervical myelopathy.

SUMMARY OF BACKGROUND DATA: Although CSC is often used as a marker for severe cervical spine pathology, it is not known whether CSC detected on MRI actually translates clinically into level-specific findings detected on physical examination.

METHODS

A consecutive series of patients with CSC evident on MRI operated on by a single surgeon from 2010 to 2012 were retrospectively analyzed. Patients' preoperative reflex examination (biceps, brachioradialis, and triceps) including abnormal reflexes (Hoffman sign, inverted radial reflex, clonus, and Babinski) were recorded. Patients were deemed to have an examination consistent with the level of CSC if they had normal reflexes cranial to the level of CSC, were hypo-reflexic at the affected level, and hyper-reflexic caudal to the level of CSC.

RESULTS

Forty-three patients with CSC were identified during the study period (Table 1). Isolated T2 CSC was present in 35 patients, and concomitant T1 and T2 CSC was present in 8 patients. Interestingly, the reflex examination correlated poorly with the cranio-caudad level of CSC, with only 11 of 43 patients (26%) having a concordant examination. In patients with CSC, 16% had clonus, 67% had Hoffman sign, 44% had Romberg sign, and 60% had a gait abnormality.

CONCLUSION

CSC visualized on MRI correlates poorly with the upper extremity reflex examination in patients with cervical myelopathy. Of the pathological reflexes, Hoffman sign has the strongest association with CSC, but still was only positive in 67% of cases. More sensitive clinical measures need to be developed to more accurately associate CSC detected on MRI to the clinical severity of cervical spondylotic myelopathy.