RPS 608 - Thyroid gland

Purpose:

To compare the thyroid imaging reporting and data system (TIRADS) of classifying thyroid nodules with the findings on fine-needle aspiration cytology (FNAC) reported using the Bethesda system.

Methods and materials:

A prospective analysis of 250 patients was performed comparing thyroid nodule ultrasound findings based on the TIRADS classification to the FNAC report based on the Bethesda classification. TIRADS 1 and biopsy-proven malignancy were excluded. Benign-appearing nodules were reported as TIRADS 2 and 3. Indeterminate or suspected follicular lesions were reported as TIRADS 4 and malignant-appearing nodules were reported as TIRADS 5. All the nodules were performed to FNAC and TIRADS findings were compared to Bethesda classification.

Results:

Of the 250 patients, 137 were TIRADS 2, 48 were TIRADS 3, 26 were TIRADS 4, and 39 were TIRADS 5. The probability of a malignant FNAC (Bethesda V-VI) in TIRADS 2, 3, 4, and 5 classes were 0%, 6.9%, 18.9%, and 92.3%, respectively. The benign FNAC (Bethesda II) in TIRADS 2 was 100%, while for TIRADS 3, 4, and 5 were 74.5%, 25.8%, and 7.6%, respectively. Of the 39 patients that were TIRADS 5, 36 patients had a biopsy-proven cancer (92.3% concordance), but 3 were benign (false-positive sonographic impression). Overall concordance rate with FNAC was 94% and the sensitivity, specificity, and negative predictive value were 89%, 90.5%, and 96%, respectively.

Conclusion:

Our study shows a fairly good correlation of thyroid ultrasound reporting using the TIRADS classification with the Bethesda classification of FNAC.

Limitations:

Being a single-centre study can be seen as a limitation.

Ethics committee approval

This prospective study was approved by our institutional ethics committee. Informed consent was obtained from all individual participants included in the study.

Funding:

No funding was received for this work.

Purpose:

To assess the efficiency of the ACR TI-RADS 2017 in selecting the thyroid nodules on ultrasound for FNAC and its correlation with Bethesda cytopathology classification.

Methods and materials:

A prospective study was conducted of 1,000 thyroid nodules evaluated on ultrasound in the radiology department from 2015-2019 at our tertiary hospital. Selection criteria included all discrete thyroid nodules where ACR TI-RADS was applied and the patients had an ultrasound-guided fine-needle aspiration cytology (FNAC) done at our institute as per TI-RADS criteria. TIRADS 1 nodules were excluded from the study. Cytopathology Bethesda classification findings were compared with TI-RADS score on ultrasound.

Results:

Of the 1,000 nodules, 398 nodules were TIRADS 2 (39.8%), 256 nodules were TIRADS 3 (25.6%), 152 nodules were TIRADS 4 (15.2%), and 194 nodules were TIRADS 5 (19.4%). Of the 400 TIRADS 2 nodules, 98 % were benign (Bethesda II). The remaining 2 % were Bethesda III. TIRADS 3 were benign in 72% and malignant in 28% cases. TIRADS 4 nodules were benign in 24% and malignant in 76% cases. TIRADS 5 nodules were malignant in 98% cases (2% false-positive). Overall concordance of ACR TI-RADS and FNAC Bethesda classification was 92% for benign nodules and 95% for malignant nodules. The sensitivity, specificity, and negative predictive value were 90%, 93%, and 96%, respectively.

Conclusion:

Ultrasound ACR TI-RADS 2017 scoring is an efficient system for selecting thyroid nodules on ultrasound for FNAC and has a fairly good correlation with Bethesda cytopathology classification.

Limitations:

Ultrasound being an operator dependent modality, the skill and experience of the operator with thyroid imaging may impact the results.

Ethics committee approval

Ethics committee approval from MIER (Medanta Institute of Education and Scientific Research) was obtained.

Funding:

No funding was received for this work.

Purpose:

To compare the diagnostic accuracy of conventional ultrasound-based thyroid imaging reporting and data system (TIRADS) and contrast-enhanced ultrasound (CEUS) qualitative parameters in the differentiation of benign and malignant thyroid nodules using histopathology as a gold standard.

Methods and materials:

This prospective IRB approved study comprised 60 patients with thyroid nodules at primary presentation. In patients with multiple nodules, the single most suspicious nodule was interrogated. All patients were examined with conventional ultrasound and classification of nodules was performed following ACR TIRADS, followed by CEUS using 2.4 ml of intravenous second-generation contrast agent. Nodules were categorised as benign or malignant based upon both TIRADS and CEUS. The CEUS qualitative parameters evaluated were enhancement degree, homogeneity, margins, order, and wash-out pattern. The diagnosis obtained by each technique were compared with histopathology as a gold standard. Statistical analysis was done to calculate sensitivity, specificity, NPV, and PPV.

Results:

In the series of nodules evaluated, 28 were benign and 32 were malignant on histopathology. TIRADS classified 29 nodules as benign and 31 as malignant. Whereas on CEUS, 27 were classified as benign and 33 as malignant. For TIRADS, sensitivity was 75%, specificity was 75%, PPV was 77.4%, and NPV was 72.4%. For CEUS, sensitivity was 84.4%, specificity was 78.6%, PPV was 81.8%, and NPV was 81.5%.

Conclusion:

CEUS qualitative parameters were found to have superior diagnostic accuracy to conventional ultrasound-based TIRADS in the differential diagnosis of thyroid nodules. Therefore, CEUS is a valuable supplementary technique in the definitive diagnosis of thyroid nodules and should be exploited further.

Limitations:

A single-centre study.

Ethics committee approval

Informed consent from all patients. Animal board approval n/a.

Funding:

No funding was received for this work.

Purpose:

There is a limited consensus on the duration of surveillance neck ultrasound (US) in hemithyroidectomy-only differentiated thyroid cancer (DTC) patients.
Our aim was to determine follow-up US findings, when available, and identify rates of contralateral malignancy in patients with DTC at hemithyroidectomy.

Methods and materials:

A retrospective observational study of all patients who underwent hemithyroidectomy between 01/12/13–31/01/15 in the Greater Glasgow and Clyde (GG&C) healthboard.

Results:

49 patients had DTC identified following hemithyroidectomy (46) or isthmusectomy (3). Based on largest/worst prognostic cancer if multifocal, subtypes were papillary (36), follicular (8), and Hurthle cell (5) with a mean diameter of 23 mm (range 0.3-75 mm). 20 were papillary microcarcinomas (PMC) and 16 of these were incidental. 13 patients had multifocal lobar disease.

36/49 proceeded to initial completion thyroidectomy. 30 completions would have been recommended/reasonable with the current guidelines. Further malignancy was found in 16/36 (44%), all PMC.

The remaining 13 patients had US follow-up over 0.5-5.5 years. In one, a contralateral U3/Thy3f nodule on 1st surveillance US 2 years postoperatively had PMC at completion. A further case had a contralateral U2 nodule upgraded to U3/Thy3a at 3rd annual US/FNA, which was benign on completion. Otherwise, there was normal appearances or sonographically benign nodules with no lymphadenopathy demonstrated.

Conclusion:

In patients with hemithyroidectomy DTC not meeting criteria for initial completion, US/FNA follow-up prompted later completion in 2/13, with one case of completion malignancy, PMC. Otherwise, US findings were limited.

Overall completion malignancy rate was 45%, all PMC.

PMC is often undetectable at US and the role of repeated post-hemithyroidectomy US surveillance is of doubtful benefit in this group.

Limitations:

A small sample size. Future collaborating studies with other Scottish/UK healthboards would be beneficial.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

The efficacy and safety of CNB for initially detected thyroid nodules remain unclear. The purpose of this study was to evaluate the diagnostic performance of CNB as a first-line diagnostic method for diagnosing thyroid neoplasms and compare against those of FNA via a propensity score analysis.

Methods and materials:

This study included 2,898 thyroid nodules from 2,662 patients from three institutions. Adjustments for differences in baseline characteristics were done by using a propensity score analysis. The rates of non-diagnostic and inconclusive results, diagnostic accuracy for diagnosing malignancy, neoplastic follicular patterns lesions (NFPLs), and thyroid neoplasms were compared. Subgroup analyses were performed according to the K-TIRADS categories.

Results:

After 1:1 matching, 753 nodules with CNB and 753 nodules with FNA were enrolled. The non-diagnostic and inconclusive results were significantly lower in the CNB group than the FNA group (all, P<0.001). For diagnosing malignancy, the diagnostic performance of CNB did not significantly differ from FNA. For diagnosing NFPLs, CNB showed significantly higher sensitivity (P<0.001) and AUC (P<0.001) than FNA. On subgroup analysis according to K-TIRADS category, CNB showed significantly higher sensitivity in the low suspicion category (P<0.001) for diagnosing malignancy. For diagnosing NFPLs, CNB showed significantly higher sensitivity and AUCs than FNA in all K-TIRADS categories (all, P<0.001). Regarding thyroid neoplasms, CNB showed significantly higher sensitivity than FNA in low and intermediate suspicion categories (all, P<0.001). The complication rate was similar between groups in matched cohorts.

Conclusion:

CNB might be useful and safe for initially detected thyroid nodules, especially in the low and intermediate suspicion category.

Limitations:

It was a retrospective observation study.

Ethics committee approval:

This study was approved by the IRB and requirement for written informed consent was waived.

Funding:

No funding was received for this work.