THE ACCURACY OF THYROID PALPATION COMPARED TO POINT-OF-CARE ULTRASOUND (POCUS) OF THE THYROID IN THE DETECTION OF CLINICALLY RELEVANT NODULES AMONG ADULT FILIPINOS LIVING IN A COMMUNITY SETTING

Abstract

INTRODUCTION
Thyroid palpation is one of the initial steps in accessing any suspicious nodule of the thyroid and is part of the general physical examination of the neck. It's accuracy is affected by patient factors such as weight and neck circumference, in addition to the clinicians’ experience. This study will determine the accuracy of thyroid palpation compared to point-of-care ultrasound (POCUS) in detecting clinically relevant nodules, defined as more than 1.0 cm based on the American Thyroid Association 2015 recommendation.

METHODOLOGY
The study method was cross-sectional, enrolling 290 Filipino adults, yielding a 99% statistical power with a 0.4 kappa agreement. Inclusion criteria are as follows: >18 years old, no previously known thyroid disease, residing in a 1st class municipality in Batangas City, Philippines. Pregnant and lactating women, those taking thionamides or levothyroxine, with previous thyroid surgery or neck radiation were excluded from the study. Included participants were subjected to the modified rose method of thyroid palpation by two board-certified endocrinologists with at least 10 years of experience followed by POCUS of the thyroid conducted by another endocrinologist with specialized training in thyroid ultrasonography. All examiners were blinded from each other’s results.

RESULTS
The overall agreement between thyroid palpation and POCUS was moderate, with a Kappa value of 0.516 (p<0.001). The prevalence of clinically relevant thyroid nodules among adult Filipinos with no known thyroid disease as determined by POCUS was 48.77% (95% CI: 42.34 - 55.23). Regarding nodule size, there was a non-significant increase in the odds of being palpated for nodules between 1.0-1.5 cm as compared to those less than 1.0 cm (OR=1.46, 95% CI: 0.68-3.09, p = 0.323). However, nodules greater than 1.5 cm exhibited a substantially higher likelihood of being palpated (OR=6.42, 95% CI: 3.31-12.79, p <0.001). The agreement for thyroid palpation performed by both endocrinologists was found to be moderate, with a Kappa value of 0.552 (p <.001). When compared to POCUS, thyroid palpation had a sensitivity of 89.19% (95% CI: 81.68-93.85) and a specificity of 60.71% (95% CI: 51.01-69.64) across all locations. In addition, thyroid palpation had a positive predictive value of 85.71% (95% CI: 77.63-91.21) with a 68% negative predictive value (95% CI: 58.44–76.26). The positive likelihood ratio was 2.27 (95% CI: 1.42-3.62) while the negative likelihood ratio was 0.18 (95% CI: 0.09-0.37). Lastly, the proportion of accurate diagnosis—true positives and negatives—out of all cases was 81.37% (95% CI: 72.73- 87.74) when thyroid palpation was compared to POCUS.

CONCLUSION
Thyroid palpation, when performed by experienced endocrinologists, yields a high sensitivity of detecting clinically relevant nodules. However, thyroid palpation also missed 10.8 % and incorrectly identified 39.3% of clinically relevant nodules. This study provides evidence to support the use of POCUS as part of routine outpatient evaluation of the thyroid to improve the accuracy of detecting clinically relevant thyroid nodules.