Prevalence and Outcomes of Thyroid Nodules with Atypia of Undetermined Significance (AUS) Cytology in a Tertiary Government Hospital: A Ten-year Review (2013 to 2022)

Background and Objective. Thyroid cancer is one of the most common malignancies reported in the Philippine Cancer Registry. Fine-needle aspiration biopsy (FNAB) plays a major role in the early detection of malignant thyroid nodules and is interpreted using the Bethesda System for Reporting Thyroid Cytopathology (TBSRTC). However, Bethesda Category III, Atypia of Undetermined Significance/Follicular Lesion of Undetermined Significance (AUS/ FLUS), remains a diagnostically challenging category for both clinicians and pathologists. This category represents an intermediate interpretation between benign and malignant lesions and is intended to be used as a diagnosis of last resort. This study aimed to determine whether the frequency of AUS/FLUS reporting in our institution falls within the recommended threshold of 7% and to describe the clinicopathologic characteristics of these lesions, including the reasons cited by pathologists for assigning this diagnostic category.

Methods. A retrospective descriptive review of medical records and pathology logbooks was conducted using available records from the Philippine General Hospital (PGH) and the Pathology Research Laboratory (PRL) Unit of the University of the Philippines College of Medicine (UPCM).

Results. The overall frequency of Bethesda Category III (AUS/FLUS) cases in PGH and PRL was 4.27%, which is within the recommended threshold of less than 7% specified by the Bethesda System for Reporting Thyroid Cytopathology. There was no significant difference in the proportion of Category III cases between PGH (6.56%) and PRL (1.59%). In contrast, the proportion of Category I (non-diagnostic) cases differed markedly between PGH (49.78%) and PRL (4.15%). Repeat biopsies of Category III cases were infrequent (15.38% in PGH and 18.64% in PRL). Among benign definitive diagnoses, thyroid follicular nodular disease was the most common, whereas conventional (classic) papillary thyroid carcinoma was the most frequent malignant diagnosis, followed by encapsulated angioinvasive follicular variant papillary thyroid carcinoma. The calculated risk of malignancy was high (67% in PGH, 73.9% in PRL, and 68.9% overall), although these estimates were limited by the small proportion of cases with traceable definitive surgical specimens in the PGH database. The most frequently cited reason for assigning Bethesda Category III was "focal/mild atypia," although this finding was often not further described in the pathology reports. An increasing trend in the practice of intradepartmental case review ("passing around" cases) was also observed over the study period.

Conclusions. The proportion of Bethesda Category III cases among thyroid FNAB specimens in PGH and PRL was within the recommended range. More accurate estimates of repeat biopsy rates and the risk of malignancy may be achieved through the development of a larger database that incorporates patient registries from outside the institution, thereby allowing follow-up of patients who receive definitive management elsewhere. Adoption of the 3rd edition of the Bethesda System for Reporting Thyroid Cytopathology is likewise recommended, as it provides a more detailed characterization of the atypia encountered in AUS cases.