Quantification of sensory and motor neurological function following spinal cord injury

Defining a spinal cord injury (SCI) by accurately classifying the patterns of neurological deficit in a manner that facilitates communication concerning treatment and prognosis is essential in spinal cord injury rehabilitation (Cohen et al., 1998; Ditunno et al., 1994). However, just as the development of methods and instruments for taking physical measurement advanced medical research in the 16th century (Angeletti, 1995; Janssens, 1991), moving beyond classification toward measurement of neurological impairment will increase the precision of prediction of outcomes and evaluation of treatments (Lucas et al., 1979). As Medical sciences continue to advance toward restorative and regenerative treatments, (Ramer et al., 2000; Girardi et al., 2000) these advances must be accompanied by methods of measurement that can reliably differentiate subtle changes in function. The International Standards for Neurological Classification of Spinal Cord Injury (Maynard et al., 1997) contain motor and sensory rating scales that can be calibrated using item response theory (IRT) methods to develop measures of neurological function. The neurological scales are subsequently utilized to predict self care and mobility functional ability and change during rehabilitation. Refined, calibrated IRT measures of upper and lower extremity motor ability are stronger predictors of patients' self care (R2 = .545) and mobility (R2 = .497) status at discharge than the conventional functional ability motor raw scores of self care (R2 = .440) and mobility (R2 = .357) respectively. The admission light touch and pin prick sensory measures were strong predictors of self care (R2 = .266) and mobility (R2 = .324) functional status at discharge. However the sensory raw scores were equally strong predictors of self care (R2 = .269) and mobility status at discharge (R 2 = .336). Pin prick sensation was more strongly related to the functional status than was light touch. The strength of prediction of the sensory scales however, does not hold in the presence of the motor scales. When the motor and sensory measures are combined in a single analysis, the motor measures are the stronger predictors. The combination of motor and sensory measures does not substantially increase the variance accounted for in predicting self care (R2 = .550), or mobility (R 2 = .509) status at discharge over the use of motor measures alone. None of the measures or raw scores were effective predictors of change in functional ability during medical rehabilitation. This investigation indicates that the motor and sensory rating scales in the Standards have potential for predicting functional outcome of rehabilitation, especially if the former are calibrated. However, the limitations of the self care and mobility measures need to be resolved for future work to be productive.