RESUMEN
Thyroid fine-needle aspiration (FNA) represents an ideal management tool for thyroid nodules. ThinPrep is routinely used in preparation of a variety of nongynecologic cytology specimens, including FNA. The authors investigated ThinPrep to determine if its diagnostic accuracy is sufficient to triage patients presenting with a thyroid nodule. ThinPrep (TP) slides from four separate study categories were reviewed in a blind fashion. Twenty-five cases of papillary carcinoma, follicular lesion, Hashimoto's thyroiditis and multinodular goiter were examined retrospectively. Diagnostic accuracy of TP for each diagnostic category was determined relative to the final FNA diagnosis. Of 100 total study cases, 46 (46%) had noncorrelation. Twenty five (25%) of the study cases had noncorrelation as a result of insufficient cellularity. The diagnostic accuracy of TP for papillary carcinoma was 64%, for follicular lesion 24%, for Hashimoto's thyroiditis 72% and for multinodular goiter 56%. Cytologic features of papillary carcinoma, Hashimoto's thryoiditis, and multinodular goiter are preserved in TP slides. Cytologic features of follicular lesion are less predictable in TP slides. When all cases of noncorrelation were examined, we concluded that insufficient or marginal cellularity most often accounts for the discrepancy between TP and CS diagnoses in each study category. If techniques can be established to improve cellularity of TP slides, particularly in follicular lesions, we believe that sufficient diagnostic accuracy can be achieved to result in appropriate patient triage. Additional studies exploring methods to improve TP cellularity are needed before TP can be used as the sole diagnostic test for thyroid FNA.
