Colloid goiter was the most common diagnosis in benign lesions [Figure 1], followed by Hashimoto's thyroiditis [Figure 2]. The malignant lesions showed follicular neoplasm (4) [Figure 3], papillary carcinoma (2) [Figures ​[Figures44–6], medullary carcinoma (1) [Figure 7], and anaplastic carcinoma (1) [Figure 8] as shown in Table 2.

| Fig. 1 Photomicrograph of smear showing clusters of follicular epithelial cells against colloid background (Pap, ×400)

| Fig. 2 Photomicrograph of smears from Hashimoto's thyroiditis showing dispersed lymphocytes, tingible body macrophages, and clusters of follicular epithelial cells (Giemsa, ×100)

| Fig. 3 Photomicrograph of smears from follicular neoplasm thyroid showing follicular cells in microfollicular arrangement with coarse chromatin (Giemsa, ×100)

| Fig. 4 Photomicrograph of smears from papillary carcinoma thyroid showing papillae and sheets of follicular epithelial cells (Giemsa, ×100)

| Fig. 6 Photomicrograph of papillary carcinoma thyroid showing ground glass nuclei with nuclear grooving. Tissue section (H and E, ×400)

| Fig. 7 Photomicrograph of smears from medullary carcinoma showing numerous isolated cells with eccentrically placed nuclei giving plasmacytoid appearance

| Fig. 8 Photomicrograph of smears from anaplastic carcinoma thyroid showing dispersed isolated tumor cells with large, hyperchromatic, and irregularly shaped nuclei (Pap, ×400)

Table 2

| Fig. 8 Photomicrograph of smears from anaplastic carcinoma thyroid showing dispersed isolated tumor cells with large, hyperchromatic, and irregularly shaped nuclei (Pap, ×400)

| Fig. 5 Photomicrograph of smears from papillary carcinoma thyroid showing large papillae showing nuclear overlapping, overcrowding with intracytoplasmic inclusions (Giemsa, ×400)

We compared the diagnosis on FNAC based on Bethesda system, with the corresponding histopathological diagnosis. Out of 238 cases of FNAC, in 105 cases, histopathological correlation was available. Cytohistological concordance was found in 102 cases (97.1%) whereas three cases were discordant. Out of these discordant cases, one false positive case was of follicular neoplasm, which on histopathological examination was found to be colloid goiter (1), whereas two false negative cases reported as benign on cytology turned out to be follicular carcinoma (1) and papillary carcinoma (1) on histopathological examination as shown in Table 3.

Table 3

In the present study, the sensitivity, specificity, and diagnostic accuracy of FNAC in diagnosing thyroid lesions using BSRTC system was found to be 75%, 98.9%, and 97.1%, respectively.

DISCUSSION

FNAC of thyroid swelling is an accurate and relatively precise tool for the diagnosis of thyroid lesions. The 6-tiered BSRTC system is very useful for triaging patient with thyroid nodules for clinical management.

In our study, age of patient ranged from 7 years to 78 years with maximum number of patients are the third to fourth decades.[10] Thyroid nodules are more common in females than in males in our study which is similar to other studies.

The inadequacy rate in our study was 5.88%. Ali suggested that the rate of ND test should be below 10%.[11] All thyroid FNAs must be technically adequate with well-represented and well-prepared smears for interpretation.

An interpretation of an inflammatory process such as thyroiditis does not require a minimum number of follicle cells. An interpretation of a colloid nodule in which there is abundant, thick colloid present does not require a minimum number of follicle cells.[12] In solid nodules producing a follicular cell population with less abundant colloid, minimum number of 5–6 groups with at least ten cells is recommended, preferable on a single slide.

The diagnosis of a benign thyroid lesion should be made only on adequate samples, whereas the presence of atypical cells or cellular pattern, always should be addressed regardless of cellularity and it precludes the interpretation of an inadequate sample.[13] Thyroid cyst with little to no follicular cells should be interpreted as “cyst fluid only” under the heading “ND” and not unsatisfactory.

Diagnostic accuracy of thyroid FNA is highly variable. The majority of diagnostic failures are due to ND samples or pathologist issuing diagnosis on samples with inadequate material. Efforts toward improvement in proficiency in FNA sampling and specimen preparation should be given high priority, as in our Institute, all the FNAC, whether USG guided or not, are performed by cytopathologists. It is a fact that the evaluation of sample adequacy is affected by many factors in addition to cellularity. Clinical information, especially physical examination and imaging studies is important in determining sample adequacy.

US-guided FNA lowers the rates of ND aspirations by allowing sampling of the cellular portions of predominantly cystic nodules and thus increased accuracy can be achieved by the use of USG-FNA in the evaluation of thyroid nodules in reducing the need of unnecessary surgery.[2]

According to Jo et al., most of these studies were carried out in tertiary care centers for thyroid lesions and thus they do not accurately reflect the general population and so it is likely that thyroid malignancies are overly represented in their cohort study, whereas in our study, benign lesions are high due to the fact that institution which despite of being tertiary care center caters the patients on referral basis and also to whom who comes directly without referral and this is reason of patient coming to our tertiary teaching hospital represents general population.[4] Another factor, which can be attributed that in our setup, many institutes provide free of cost services of USG and FNAC, etc., for below poverty line patients which gives easy access to the large population to our tertiary care centers.

The correlation of cytological and histopathological diagnoses is an important quality assurance method as it allows cytopathologists to calculate their false positive and false negative results. The false negative rate (FNR) is defined as the percentage of cases with benign cytology but proven to be malignant after histopathological examination. The false positive rate is defined as percentage of cases with a malignant FNA proven to be benign on histological examination.

There were two patients, who had false negative diagnosis, turned out to be papillary carcinoma and follicular carcinoma and one false positive case of follicular neoplasm, which on histological examination was found to be colloid goiter. Cytological examination in the first false negative case diagnosed as colloid goiter revealed moderate cellularity with small clumps of thyroid follicular cells arranged in poorly cohesive groups with moderate colloid favored the diagnosis which later on histopathological examination turned out to be follicular carcinoma whereas in the second false negative case, cytological smear showed low cellularity of follicular cells and cystic macrophages with abundant colloid which was misinterpreted and the diagnosis of benign lesion was made. This was due to the aspiration from the cystic area and not from the cellular area, whereas one false positive case reported as follicular neoplasm was actually colloid goiter. This was because in this case aspiration was probably done from the hypercellular areas of colloid nodule which led to overdiagnosis.

Cytological differentiation between follicular neoplasm and nodular goiter is often difficult.[14] In addition, an overlap of cytomorphological features among reactive follicular cells, Hurthle cell lesions and malignancies because the presence of nuclear grooves and even pseudo-inclusions is not pathognomonic of papillary carcinoma. No reliable cytologic feature distinguish follicular adenoma from follicular carcinoma and distinction between the two is based solely on demonstration of histologic evidence of capsular/and or vascular invasion.[12]

Given that the purpose of thyroid FNA is to provide clinically useful information regarding the need for surgery so that FNA sample must be adequate enough for an interpretation that yields a low FNR. Proper sampling from representative site is important for an accurate diagnosis. Most of the authors recommend preparation of 4–5 smears from different areas of nodules with two to five passes are a reasonable number of passes to ensure an adequate sample. However, certainly for cystic lesions, there is no justification to recommend a different number of passes. Cystic change in thyroid lesion is a common diagnostic pitfall in cytology. USG-guided FNAC results in better sample yield leading to the reduced inadequacy of sample and enhancing overall accuracy. Hence, multiple passes to cover all areas of interest, and USG-guided aspiration reduces sampling error.[15] Errors caused by misinterpretation of cytologic finding in adequate samples are uncommon.

Using BSRTC in a well-defined and consistent manner, especially in gray zone areas of Thyroid cytopathology, allow appropriate triage to FNA versus surgery base. Low sample size is the limitation of the study and in future, larger studies could be performed categorizing the lesion on the basis of Bethesda system.

CONCLUSION

FNAC is rapid, simple, safe, and cost-effective diagnostic modality in the investigation of thyroid disease with high sensitivity, specificity, and accuracy. It can be used as an excellent first-line method for investigating the nature of lesion. Bethesda system is very useful standardized system for reporting thyroid cytopathology, as it is simplified, systematic which provides better communication between cytopathologist and clinician leading to more consistent management approaches.

| Fig. 1 Photomicrograph of smear showing clusters of follicular epithelial cells against colloid background (Pap, ×400)

| Fig. 2 Photomicrograph of smears from Hashimoto's thyroiditis showing dispersed lymphocytes, tingible body macrophages, and clusters of follicular epithelial cells (Giemsa, ×100)

| Fig. 3 Photomicrograph of smears from follicular neoplasm thyroid showing follicular cells in microfollicular arrangement with coarse chromatin (Giemsa, ×100)

| Fig. 4 Photomicrograph of smears from papillary carcinoma thyroid showing papillae and sheets of follicular epithelial cells (Giemsa, ×100)

| Fig. 6 Photomicrograph of papillary carcinoma thyroid showing ground glass nuclei with nuclear grooving. Tissue section (H and E, ×400)

| Fig. 7 Photomicrograph of smears from medullary carcinoma showing numerous isolated cells with eccentrically placed nuclei giving plasmacytoid appearance

| Fig. 8 Photomicrograph of smears from anaplastic carcinoma thyroid showing dispersed isolated tumor cells with large, hyperchromatic, and irregularly shaped nuclei (Pap, ×400)

| Fig. 5 Photomicrograph of smears from papillary carcinoma thyroid showing large papillae showing nuclear overlapping, overcrowding with intracytoplasmic inclusions (Giemsa, ×400)

REFERENCES