Repeatability of Three-Dimensional Computed Tomographic Volumetric Measurement of Lumbosacral Intervertebral Foramina in German Shepherd Dogs.

Abstract

This article determines the repeatability of a recently reported method of volumetric measurement of the lateral intervertebral neurovascular foramina at the lumbosacral junction in German shepherd dogs. Batch files including the intervertebral neurovascular foramina were derived from previously obtained computed tomography studies of the extended lumbosacral junction of 20 German shepherd dogs and converted into volume datasets. Three observers independently performed five measurements of the left and right lumbosacral intervertebral neurovascular foramina in each dog, using an Extended Brilliance Workstation (Phillips, The Netherlands) to generate a volume of the lumbosacral intervertebral foramina in cubic millimetres, as described by Worth and colleagues in 2017. The inter-observer repeatability of the mean of the five-volume measurements (40 foramina) was assessed using the Bland-Altman limits of agreement method. One observer, blinded to the previous measurements, repeated the volumetric analysis on 20 lumbosacral intervertebral foramina for an assessment of intra-observer repeatability using the same statistical methods. This method of volumetric analysis showed good intra- and inter-observer repeatability with 95% of paired comparisons falling within two standard deviations of the mean difference between them. This method of measuring the volume of the lumbosacral intervertebral neurovascular foramina is repeatedly reliable and may be a useful tool when testing the effects of motion and disease on the lumbosacral junction, and could help guide surgical intervention when foraminal narrowing is implicated in the clinical signs.