Inter-reader agreement and additive benefit of SPECT or SPECT/CT modality with [99mTc]Tc-pertechnetate scintigraphy imaging for the assessment of thyroid nodules in a tertiary care center

Aim/introduction In central Europe, up to 76% of population exhibit thyroid nodules (TN). The combination of ultrasound findings and thyroid scan imaging are supposed to ensure a cost-effective and reliable prediction of malignancy risk. However, real-world impact of ultrasound findings, i.e. TI-RADS, has raised concerns due to its highly limited reliability and reproducibility. The recent advancements and increasing access of SPECT/CT scanners might contribute to thyroid management. This study aimed to elucidate additive value of SPECT(-CT) imaging in initial evaluation of complex TN cases, particularly on multinodular goiter and difficult-to-characterize solitary nodules. Materials and methods A total of 61 patients (19 males and 42 females with a mean age of 56 (± 17)) with incidental TN were retrospectively enrolled in this study between November 2022 and July 2023, who underwent both conventional planar thyroid scan and SPECT(-CT) imaging. Three readers with varying experience level selected a target nodule within thyroid gland and evaluated its functionality, ACR TI-RADS categorization and additive benefit of SPECT(-/CT) imaging. Kappa-value for inter-reader agreement (IRA) was interpreted as follows: < 0.01–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial agreement; and 0.81–1.00; almost perfect agreement. Results Of the target TNs, 33 (54%) were located in the left thyroid lobe, 5 (8%) in the isthmus and the remaining target TNs (38%) in the right lobe. The median target TN size was 20 mm (10–59). The target nodule match rate among readers was almost perfect (k-value 0.81; 95% CI: 0.72–0.90). The IRA regarding ACR TI-RADS was moderate with k-value of 0.47 (95% CI: 0.40–0.54). The IRA regarding patient referral was substantial among readers with k-value of 0.60, while k-value was even higher at pairwise comparison of experienced readers (k-value: 0.64). IRA regarding the target TN functionality among readers was good with k-value of 0.68 (95%CI: 0.59–0.77). IRA involving the less experienced reader revealed only slightly lower reliability scores, i.e. a k-value of 0.61. The correlation with ground truth results revealed comparable results among experienced and less-experienced readers. Conclusion Increasing clinical experience of physicians appeared to have a direct correlation with a higher referral rate of SPECT(-CT) imaging, although the additive clinical benefit was only evident for a minority of patients. Hence, SPECT(-CT) imaging might be more suitable for tertiary care centers, as its additive benefit at primary care level would have only negligible effect for patients.