Clinical value of the threshold derived from bone single-photon emission computed tomography/computed tomography in differentiating sternoclavicular arthritis

• Published in: Hellenic journal of nuclear medicine Vol. 25; no. 1; pp. 32 - 37
• Main Authors: Tsuchitani, Tatsuya, Kitajima, Kazuhiro, Azuma, Naoto, Tamura, Masao, Takahashi, Yoshiyuki, Kotoura, Noriko
• Format: Journal Article
• Language: English
• Published: Greece 01-01-2022
• Subjects: Arthritis - diagnostic imaging; Bone and Bones; Humans; Retrospective Studies; Single Photon Emission Computed Tomography Computed Tomography; Tomography, Emission-Computed, Single-Photon - methods
• DOI: 10.1967/s002449912434
• ISSN: 1790-5427

With the recent improvements in the quantitative accuracy of single-photon emission computed tomography (SPECT)/ computed tomography (CT), the value of using standardized uptake value (SUV) in bone SPECT/CT for quantitative assessment has been reported.We established a threshold for inflamed and normal areas of the sternoclavicular joint and examined the clinical value of bone SPECT/CT.
The threshold between the inflamed and normal areas of the sternoclavicular joint was initially calculated. The diagnostic performance of the calculated threshold was subsequently compared with the visual assessment of the whole-body image. The clinical value of the threshold was examined in cases of ambiguous visual assessment and a sub-analysis with pustuloticarthro-osteitis (PAO) patients was done.
The threshold between the inflamed and the normal area in the 93 sternoclavicular joints of 51 patients was 4.46. The area under the ROC curve (AUC), accuracy, sensitivity, and specificity of SUVmax for differentiating sternoclavicular arthritis were 0.92, 0.86, 0.88, and 0.85, respectively. Similarly, the AUC of visual assessment were 0.87, and the difference was not significant (P=0.11). In 25 patients with PAO, the AUC, accuracy, sensitivity, and specificity of SUVmax were 0.94, 0.90, 0.96, and 0.84, respectively with a significant higher AUC of visual assessment (0.82, P=0.032). Furthermore, for cases where there was ambiguous uptake upon visual assessment, the accuracy, sensitivity, and specificity of SUVmax were 0.84, 1.00, and 0.71, respectively, which was useful to judge regarding the initiation of treatment.
Quantitative assessment using SUVmax and the threshold found using bone SPECT/CT for the presence of sternoclavicular arthritis is clinically useful and can be a useful tool for the initiation of treatment, especially in PAO patients.