Experimental subacute spinal cord compression: correlation of serial S100B and NSE serum measurements, histopathological changes, and outcome.

Objective: To correlate serial measurements of serum S100B and neuron-specific enolase (NSE) with histopathological changes of the spinal cord and to assess their prognostic significance in a set-up of experimental spinal cord compression.

Methods: The thoracic cords of 22 rabbits were increasingly compressed and decompressed once paresis had developed. After decompression, outcome was rated as favorable or unfavorable. Following sacrifice of the animals, the cord was analyzed microscopically and morphometrically. Serum S100B and NSE were measured daily, and levels were correlated with initial degree of paresis, outcome after decompression, and histopathological changes of the cord.

Results: Regardless of the initial degree of paresis, animals with favorable outcome had significantly higher cell counts than animals with unfavorable outcome. The time course of S100B values following decompression was correlated with outcome. Animals with favorable outcome had either always normal levels or levels that were initially increased but normalized within 2 days. The values of animals with unfavorable outcome were elevated throughout (P<0.0001). No correlation was found between NSE levels and outcome.

Conclusions: The initial degree of paresis is not a prognostic factor to predict outcome. Despite timely decompression, pronounced structural lesions of the cord may develop, resulting in an unfavorable outcome. In cases with favorable outcome, sufficient tissue is preserved to maintain function regardless of the initial extent of paresis. This different reaction of the cord may be followed indirectly with serial measurements of S100B serum levels. Thus, S100B is a reliable biochemical marker allowing for prediction of outcome. NSE does not have this prognostic significance.