Sacroiliac Indicis Increase the Specificity of Bone Scintigraphy in the Diagnosis of Sacroiliitis

Abstract

Objective: Bone scintigraphy is a highly sensitive method in the evaluation of sacroiliitis. Aim of this study is firstly to evaluate interobserver variation of partial and whole sacroiliac indicis, secondly investigation of clinical importance of these indicis in the diagnosis of sacroiliitis. Methods: Fourty-six subjects (24 female: 35.4 +/- 11.9; 22 male: 43.1 +/- 12.4) without sacroiliitis 45 subjects with low back pain (33 female: 43.3+/-11.5, 11 male: 35.5+/-17.2) were included in the study. For right (R) and left (L) whole indices (WSI) irregular region of interest (ROI), for partial indices superior (S) and inferior (I) rectangular ROI were used. For background activity, rectangular ROI was drawn from the sacral region. Indices were calculated from ratio of average counts of sacroiliac and background regions. Two independent observers calculated sacroiliac indices. Interobserver agreement was evaluated by Pearson analysis. Results: There was no significant interobserver difference (p>0.05). Significant correlation existed between all calculated indices. Among 45 patients with suspicion of sacroiliitis 15 had final diagnosis of sacroiliitis and all of the Tc-99m methilenediphosphonate planar and SPECT bone scintigraphy results of these patients were concordant with sacroiliitis. There were 8 false positive results in other 30 patients. Seven of these eight patients had normal index values. If the scintigraphy would be evaluated in conjuction with indicis the specificity would increase from 73% to 97% but sensitivity decreases from 100% to 80%. There was significant correlation between the observers calculated indicis (p<0.001). Conclusion: Superior and inferior sacroiliac index values can be used with confidence. If we use sacroiliac index values to confirm positive results; index values can increase the specificity of bone scintigraphy.