Improving our departmental reporting of thyroid cytology specimens against national guidelines: a two-cycle retrospective audit (2020)

Type of publication:
Journal article

*Halliday E.; *Ahsan S.F.; *Harrison E.; *Harrison K.; *Sansom H.

Cytopathology : official journal of the British Society for Clinical Cytology; November 2020, Volume31, Issue6, Pages 514-524

OBJECTIVE: In the United Kingdom, guidelines from the Royal College of Pathologists (RCPath) facilitate consistent and reproducible reporting and classification of fine needle aspiration cytology (FNAC) thyroid specimens. The aim was to audit our department against RCPath guidelines to refine and improve our reporting process.
METHOD(S): Two-cycle retrospective observational audit of all patients undergoing thyroid FNAC over a two-year period (one year for each cycle). Final histology was correlated. The positive predictive value (PPV) for malignant neoplastic lesions was calculated; for Thy1, Thy1c, Thy2 and Thy2c all cases without final histology were assumed to be benign, while for Thy3a, Thy3f, Thy4 and Thy5 samples the PPV calculation was based only on those cytology samples with corresponding histology. 'False positive' and 'false negative' cases were reviewed.
RESULT(S): 288 cytology samples were included in the first cycle; 96 (33.3%) had corresponding histology. 287 samples were included in the second cycle; 119 (41.5%) had follow-up histology. The rate of non-diagnostic samples (Thy1/1c) decreased from 39.6% to 30.0%. The PPV for malignant neoplastic lesions was Thy1/1c 2.6%, Thy2/2c 0.0%, Thy3a 40.0%, Thy3f 19.4%, Thy4 75.0%, Thy5 100.0% (first cycle); Thy1/1c 4.7%, Thy2/2c 0.7%, Thy3a 13.3%, Thy3f, 7.7%, Thy4, 50.0%, Thy5 100.0% (second cycle).
CONCLUSION(S): Our department was able to reduce the rate of non-diagnostic FNAC samples and improve the diagnostic accuracy of FNAC. Auditing local outcomes helps refine and improve the reporting process. Review of 'false positive' and 'false negative' cases helps examine potential pitfalls of cytology.