ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2011) 25 P347

Retrospective analysis of thyroid fine needle aspiration (FNA) in a tertiary hospital

Alia Nasir, Subash Sivaraman, Hala Alsafadi, Steve Ferryman, Paul Mathews, Yen Ching Yeo, Tom Goodfellow, Nick Hedley, Kim Brett, Barbara Smith, Hisham Mehanna, Martin O Weickert & Sailesh Sankar

Wisdem Centre, Coventry and Warwickshire, Coventry, UK.

Aim: In this study we have compared the adequacy rates of FNA samples between more experienced and less experienced operators in thyroid FNA.

Introduction: The prevalence of thyroid nodules is estimated to be about 4–7%, increasing up to 50% in the adult population. US guided FNA has become the preliminary investigation of choice for thyroid nodules, as it is widely available and has been proven to increase diagnostic efficacy.

In 2004, NICE recommended the setup of ‘one-stop clinics’ for patients with neck lumps.

Material and methods: Data was collected retrospectively from 180 patients who underwent ultrasound scan (US) FNA of thyroid over a period of 21 months, from January 2008 to October 2009; either through One-stop clinic or routine referrals from clinics.

Results were analysed using SPSS version 16.

Results: Of 312 FNA were performed in 180 (156 females) patients, 143 were reported as inadequate sample/indeterminate.

Logistic regression analysis using diagnostic (Thy 2–5) versus non-diagnostic (Thy1) samples as the dependent variable and age, sex, TSH and operator as covariates showed a significantly higher rate of Thy1 samples when FNA was performed by less experienced investigators (59 vs 39%, P=0.031), although only 5% of the variation was explained by the total model.

FNAC had sensitivity of 72.2% and specificity of 55.9% when compared with histopathology in the 58 patients who underwent thyroidectomy. Positive predictive value of possible/definite malignancy was 46.4%. Negative predictive value of non-neoplastic lesion was 79.1%.

Conclusion: This study showed a relatively high proportion of inadequacy of FNA sampling. Although there was a higher success rate in more experienced operators, the contribution of the operator to total variation was small, this on its own account will not account for the results. Other factors including case selection, size and nature of nodule may explain our findings and this needs further evaluation.