Keywords:
Musculoskeletal spine, CT, Comparative studies, Connective tissue disorders, Metabolic disorders
Authors:
C. Dan Lantsman1, A. Herman2, J.-J. Verlaan3, M. Stern1, R. Mader4, I. Eshed1; 1Ramat Gan/IL, 2Ashdod/IL, 3Utrecht/NL, 4Afula/IL
DOI:
10.1594/ecr2018/C-0320
Methods and materials
Abdominal CT examinations of 43 DISH (diagnosed by Resnick radiographic criteria),
31 AS (diagnosed by Modified New-York Criteria [11]) and 42 age-and gender matched (to DISH) control subjects (Males: 29/29/27; mean age: 71.74/56.1/72.71 years,
respectively) were evaluated retrospectively and compared for VAT and SAT adipose tissue surface areas on axial mid-L3,
L4,
L5 levels.
Visceral and subcutaneous fat boundaries were registered manually by a single reader on each mid-vertebral level in which adipose tissue volume was then computed as shown in Figure 1.
Adipose tissue area was calculated by division of the registered adipose tissue volume by slice thickness.
A subgroup of 10% of all examinations in the three groups was re-evaluated by a second reader for reproducibility evaluation.
Comparisons between the groups (DISH,AS,controls) were done by univariate analysis of variance (ANOVA).
Multivariate analysis was done by linear regression models.
Interobserver agreement was calculated using Intraclass Correlation Coefficient (ICC) by the analysis of variance (ANOVA) two-way random for absolute agreement.