Does choice of imaging modality affect outcome in dogs with thoracolumbar spinal conditions?

Abstract

A retrospective cross-sectional study was done to look for differences in outcome in canine spinal patients that had myelography compared to those that had magnetic resonance imaging. Medical records of dogs with spinal conditions in the period January 2004 to December 2007 were reviewed. Data on patient age, gender, breed, size, neurolocalisation, rate of onset, imaging modality, time taken to image, type of treatment, neurological grade at admission and discharge, length and cost of hospitalisation and status at discharge were collected. Only dogs with neurological grade 3 to 6 with signs referable to the thoracolumbar spine were included. Of 107 dogs that met the inclusion criteria, 66 (62%) had myelography and 41 (38%) had magnetic resonance imaging. Using multivariable analyses, non-chondrodystrophoid breed, increasing age and higher neurological grade at admission were found to be associated negatively with survival. Neurological grade 5 at admission was found to be associated positively with likelihood of neurological improvement. Male gender, higher neurological grade at admission and medical treatment were associated negatively with length of hospitalisation. magnetic resonance imaging, surgical treatment and period of hospitalisation were associated positively with total cost of hospitalisation. No significant association was found between type of imaging and any patient outcome variables except cost of hospitalisation, which was higher for dogs having magnetic resonance imaging. Although magnetic resonance imaging may be considered advantageous compared to myelography because it is non-invasive and provides superior anatomical detail for surgical guidance, no beneficial effect on outcome of dogs with non-ambulatory thoracolumbar spinal disease was found.