Radiographic evaluation of lumbar intervertebral disc height index: An intra and inter-rater agreement and reliability study

Abstract

PurposeTo evaluate intra- and inter-rater agreement and reliability of seven reported disc height index (DHI) measurement methods on standing lateral X-ray of lumbar spine. MethodsThe adult patients who had standing lateral X-ray of lumbar spine were recruited. Seven methods were used to measure DHI of each lumbar intervertebral disc level, including a ratio of sum of anterior and posterior disc height (DH) to disc diameter (Method 1), a ratio of middle DH to mid-vertebral body height (Method 2), a ratio of middle DH to disc diameter (Method 3), a ratio of the mean of anterior, middle, and posterior DH to the sagittal diameter of the proximal vertebral body (Method 4), a ratio of DH to vertebral height which cross the centre of adjacent vertebral bodies (Method 5), a ratio of the mean of anterior, middle, and posterior DH to the mean of proximal and distal vertebral body height (Method 6), and a ratio of the sum of anterior and posterior DH to the sum of superior and inferior disc depth (Method 7). Two raters conducted the measurements (one medical student (SS) and the other an experienced spine surgeon (XC)). Bland and Altmańs Limits of Agreement (LOA) with standard difference were calculated to examine intra- and inter-rater agreements between two out of seven methods for DHI. Intra-class correlations (ICC) with 95% confidence intervals were calculated to assess intra- and inter-rater reliability. ResultsThe intra-rater reliability in DHI measurements for 288 participants were ICCs from 0.807 (0.794, 0.812) to 0.922 (0.913, 0.946) by rater 1 (SS) and from 0.827 (0.802, 0.841) to 0.918 (0.806, 0.823) by rater 2 (XC). Method 2, 3, and 5 on all segmental levels had bias (95 % CI does not include zero) or/and out of the acceptable cut-off proportion (>50 %). A total of 609 outliers in 9174 segmental levels’ LOA range. Inter-rater reliability was good-to-excellent in all but method 2 (0.736 (0.712, 0.759)) and method 5 (0.634 (0.598, 0.667)). ICCs of related lines to good-to-excellent reliability methods was excellent in all but only indirect lines in method 1 and 4 (ICCs lie in the range from 0.8 to 0.9). ConclusionFollowing a structured protocol, intra- and inter-rater reliability was good-to-excellent for most DHI measurement methods on X-ray. However, the complicated methods (more indirect lines) and IVD degeneration (nucleus pulposus degeneration and disc herniation) potentially affected the agreement on inter-rater measurements. Method 7 is the best reproducible method to measure disc height index for all intervertebral disc segmental levels with a good-to-excellent intra- and inter-rater reliability and agreement.