Evaluation of the thoraco-laryngeal reflex ('slap test') as an aid to the diagnosis of cervical spinal cord and brainstem disease in horses.

A study was conducted over a 12 month period to assess the specificity and sensitivity of the 'slap test', using endoscopic evaluation, in the detection of cervical spinal cord and caudal brainstem lesions in horses. Fifteen ataxic horses were subjected to the 'slap test' and subsequently examined post mortem. Twelve out of the 15 had histopathological lesions consistent with their clinical signs. Thirteen horses with no history of neurological dysfunction and no histopathological evidence of cervical spinal cord or brainstem disease were used as controls. The laryngeal adductory responses exhibited by all horses were filmed and later scored independently by 3 assessors. The proportion of animals diagnosed with cervical spinal cord and/or brainstem disease, defined by histopathological criteria, was found to be statistically similar to the proportion with abnormal 'slap test' responses, using the McNemar chi-Square test. Despite statistical significance between proportions, sensitivity of the 'slap test' was low, 50% for the left side on both days and 58% for the right side. Specificity was higher, 69% (Day 1) and 75% (Day 2) for the left side and 75% (Day 1) and 69% (Day 2) for the right side. In contrast to this, conventional neurological examination was found to be 100% sensitive and 81% specific in the detection of lesions of histopathological significance in the cervical spinal cord/caudal brainstem. Agreement between scores for the 'slap test' from the same assessor on different days was good, with values for kappa of 0.59 to 0.85. In contrast, agreement between assessors on the 'slap test' score was poor, with kappa 0.35.(ABSTRACT TRUNCATED AT 250 WORDS)