Pyeong Hwa Kim1, Hee Mang Yoon2, Jisun Hwang3, Jin Seong Lee1, Ah Young Jung1, Young Ah Cho1, Jung Hwan Baek1. 1. Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 86 Asanbyeongwon-Gil, Songpa-Gu, Seoul, 05505, Republic of Korea. 2. Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 86 Asanbyeongwon-Gil, Songpa-Gu, Seoul, 05505, Republic of Korea. espoirhm@gmail.com. 3. Department of Radiology, Dongtan Sacred Heart Hospital, Hallym University Medical Center, Hwaseong, Republic of Korea.
Abstract
OBJECTIVES: To evaluate the diagnostic performance of adult-based "American College of Radiology- Thyroid Imaging Reporting And Data System" (ACR-TIRADS) and "American Thyroid Association" (ATA) in the pediatric population. METHODS: MEDLINE/PubMed, EMBASE, Cochrane Library, and Web of Science databases were searched for articles investigating the diagnostic performance of each stratification system (ACR-TIRADS or ATA) and evaluated them according to three aspects: (a) the risk of malignancy in each category; (b) diagnostic performance using the classic indicators (sensitivity, specificity); and (c) diagnostic performance regarding fine needle aspiration/biopsy recommendation. In addition to pathologic diagnosis, we allowed imaging follow-up as the reference standard for benign nodules. RESULTS: Eight articles (1036 thyroid nodules) were included. For ACR-TIRADS, the pooled risk of malignancy in category was as follows: category 5 (59.3%); 4 (20.7%); 3 (11.0%); 2 (6.0%), and 1 (5.5%). For nodules of high suspicion of malignancy (category 4 or 5), the pooled sensitivity and specificity were 0.84 and 0.64. For ATA, the pooled risk of malignancy was as follows: category 5 (55.4%); 4 (34.2%); 3 (12.2%); and 2 (7.5%). For nodules of high suspicion of malignancy (category 4 or 5), the pooled sensitivity and specificity were 0.90 and 0.50. For category 5 nodules, the pooled specificity was significantly higher in ACR-TIRADS (p = 0.02). For ACR-TIRADS, the missed malignancy rate was 21.7% and the unnecessary biopsy rate was 62.7%. Information was not sufficient for this calculation with ATA. CONCLUSIONS: The diagnostic performance of ACR-TIRADS and ATA in the pediatric population was somewhat modest. Large studies are mandatory for further validation and future amendments. KEY POINTS: • The pooled sensitivity and specificity for highly suspicious nodules (category 4 or 5) for ACR-TIRADS were 0.84 and 0.64, and for ATA were 0.90 and 0.50, respectively. • When applying ACR-TIRADS for children, the pooled missed malignancy rate (21.7%) and unnecessary biopsy rates (62.7%) are still reasonably high. Insufficient information was available to perform these calculations for the ATA system. • Current risk stratification systems, especially ACR-TIRADS, require modification by focusing more on increasing the sensitivity and decreasing the missed malignancy rate. Lowering size cut-off for biopsy would be a reasonable option.
OBJECTIVES: To evaluate the diagnostic performance of adult-based "American College of Radiology- Thyroid Imaging Reporting And Data System" (ACR-TIRADS) and "American Thyroid Association" (ATA) in the pediatric population. METHODS: MEDLINE/PubMed, EMBASE, Cochrane Library, and Web of Science databases were searched for articles investigating the diagnostic performance of each stratification system (ACR-TIRADS or ATA) and evaluated them according to three aspects: (a) the risk of malignancy in each category; (b) diagnostic performance using the classic indicators (sensitivity, specificity); and (c) diagnostic performance regarding fine needle aspiration/biopsy recommendation. In addition to pathologic diagnosis, we allowed imaging follow-up as the reference standard for benign nodules. RESULTS: Eight articles (1036 thyroid nodules) were included. For ACR-TIRADS, the pooled risk of malignancy in category was as follows: category 5 (59.3%); 4 (20.7%); 3 (11.0%); 2 (6.0%), and 1 (5.5%). For nodules of high suspicion of malignancy (category 4 or 5), the pooled sensitivity and specificity were 0.84 and 0.64. For ATA, the pooled risk of malignancy was as follows: category 5 (55.4%); 4 (34.2%); 3 (12.2%); and 2 (7.5%). For nodules of high suspicion of malignancy (category 4 or 5), the pooled sensitivity and specificity were 0.90 and 0.50. For category 5 nodules, the pooled specificity was significantly higher in ACR-TIRADS (p = 0.02). For ACR-TIRADS, the missed malignancy rate was 21.7% and the unnecessary biopsy rate was 62.7%. Information was not sufficient for this calculation with ATA. CONCLUSIONS: The diagnostic performance of ACR-TIRADS and ATA in the pediatric population was somewhat modest. Large studies are mandatory for further validation and future amendments. KEY POINTS: • The pooled sensitivity and specificity for highly suspicious nodules (category 4 or 5) for ACR-TIRADS were 0.84 and 0.64, and for ATA were 0.90 and 0.50, respectively. • When applying ACR-TIRADS for children, the pooled missed malignancy rate (21.7%) and unnecessary biopsy rates (62.7%) are still reasonably high. Insufficient information was available to perform these calculations for the ATA system. • Current risk stratification systems, especially ACR-TIRADS, require modification by focusing more on increasing the sensitivity and decreasing the missed malignancy rate. Lowering size cut-off for biopsy would be a reasonable option.