Correlation of fine needle aspiration cytology with histopathology for thyroid swellings in a tertiary care hospital in South India аспирационной цитологии пациентов с разрастанием больнице третичного уровня

Fine needle aspiration cytology (FNAC) is documented to be a good tool in the diagnosis of thyroid lesions. It has good diagnostic accuracy and is an inexpensive investigation. Although a good instrument it does have its own limitations especially when it comes to distinguishing a follicular adenoma from follicular carcinoma thyroid. Objectives of the study was the evaluation of FNAC in the diagnosis of thyroid lesions by determining its sensitivity, specificity, diagnostic accuracy, positive predictive value and negative predictive value. Material and methods. This is a retrospective study of 580 cases of thyroid swelling admitted to our hospital and undergone thyroidectomy over a period of 5 years from 2015 to 2020. After thyroidectomy, FNAC reports (Bethesda system for reporting) of those patients were collected and were compared and analyzed with the histopathology reports. Category 1 and category 3 cases were excluded from calculation due to uncertainty in reporting. Data were analyzed to calculate the sensitivity, specificity, diagnostic accuracy, positive predictive value and negative predictive value of FNAC. Results. Our study included a total of 580 cases of thyroid swellings. The male to female ratio was 1:18.2 and the median age was 43 years. On FNAC, 428 cases (73.7 %) were benign, 35 cases (6 %) were indeterminate, 39 cases (6.7 %) were follicular neoplasm, 30 cases (5.1 %) were suspicious for malignancy and 34 (5.8 %) were malignant. 14 (2.4 %) cases were unsatisfactory. Cytology report was then compared with histopathological reports. FNAC had a sensitivity of 84.2 %, specificity of 97.2 % and a diagnostic accuracy of 94.5 %. The positive predictive value and negative predictive value were 88.3 and 96 % respectively. Conclusions. FNAC is a more specific than sensitive when distinguishing between benign and malignant lesions of thyroid. It has good diagnostic accuracy and hence should be used in all tertiary care hospitals prior to surgery in diagnosing thyroid lesions.


Introduction
Thyroid swellings are quite common in the general population and predominantly affect women as compared to men. Based on a study conducted in south India, thyroid swellings can be clinically detected in about 12 percent of adult population [1]. There are numerous etiologies for thyroid swellings and this is commonly classified as either benign or malignant. It is important to distinguish between the two since the modality of treatment varies considerably. Hence, the aim of investigations is not only to determine the cause of thyroid swelling but also to rule out malignancy [2]. Approximately 5 % of individuals with thyroid swellings are found to have malignant nodules [3]. In India, 5-6 % of thyroid cancers occur in women and 2 % are seen in men [4].
It is essential to have a good diagnostic tool in order to diagnose patients with malignancy. Not all patients with thyroid swelling require surgery and furthermore a good diagnostic tool will help surgeons decide the extent of surgery that needs to be performed. Hence, these tests play a crucial role in deciding the treatment of patients [5]. Thyroid gland being a relatively superficial swelling makes it amiable to clinical examination, fine needle aspiration cytology (FNAC) and even trucut biopsy in few specialized centers. Trucut biopsy of thyroid gland is not routinely performed since it increases chances of damage to structures like the recurrent laryngeal nerve and thyroid vasculature. Hence, FNAC is more commonly utilized for diagnosis. It is a simple, quick and relatively cheap procedure easily done in the Out-Patient Department. It is a good investigation to determine the etiology of thyroid swelling and differentiate a benign from a malignant thyroid lesion [6]. Published data shows that the accuracy of FNAC diagnosing a malignant thyroid lesion is more than 90 % [7,8].

Material and Methods
This is a retrospective study of 580 diagnosed cases of thyroid swelling that presented to the general surgery, surgical oncology and Ear, Nose and Throat Out-Patient Departments of Father Muller hospital in the period between January 2015 and December 2019. All case notes were retrieved using our hospital computer data system. Age, sex, cytological and histological diagnosis were reviewed. Diagnosis of cytological smears was done using Bethesda system of reporting. According to this system, reports were classified as follows: category 1 -non-diagnostic or unsatisfactory, category 2 -benign, category 3 -atypia of undetermined significance / follicular lesion of undetermined significance, category 4 -follicular neoplasm, category 5 -suspicious for malignancy, category 6 -malignant.
FNAC results were then compared with histopathology reports to calculate the sensitivity, specificity, positive and negative predictive value of FNAC in diagnosing thyroid lesions. Category 1 and Category 3 cases were excluded from calculations since it's difficult to classify them as benign or malignant based on FNAC.

Results
A total of 580 patients were included in our study. Total number of males was 90 and the total number of females was 490, with female to male ratio being 18.3 : 1. In our study the age ranged from 16 to 78 years, with the median age being 43 years.
Hence, FNAC achieved a sensitivity of 84.2 %, specificity of 97.1 %, and a total accuracy of 94.8 %. The positive predictive value and negative predictive value were 88.3 and 96 % respectively.

Discussion
Numerous studies have demonstrated that among various modalities for diagnosing thyroid swellings, FNAC stands out to be a safe, accurate and less expensive tool [13]. Although a good tool it does have few limitations. If the sampling of the specimen is inadequate diagnosis becomes very difficult. Another major limitation is its inability to distinguish a benign follicular swelling from a malignant one. As with all investigation FNAC can show false positive and negative results mainly with small swellings and when thyroid nodules are associated with inflammation or degeneration. For example, distinguishing a nodular colloid goiter from a follicular neoplasm may be impossible [14]. Thyroid swellings are quite common in India with 12 % of the population presenting with clinically palpable masses [1].
Thyroid swellings are commonly seen in females but the incidence of thyroid cancer in females is less as compared to males. Numerous studies have shown that the female to male ratio for thyroid swellings is 5 : 1 [15][16][17]. Male patients were found to have malignancy rates ranging from 19 to 26 % [15,17]. In our study there were larger number of females as compared to males with a female to male ratio of 18.3:1. The median age was 43 and the ages ranged from 16 to 78 years.
In our study cases diagnosed with thyroid malignancy ranged from 20 to 77 years with the median age being 44. There were larger number of benign cases in our study as compared to malignant and this was similar to other published literature [18,19]. The most common benign lesion encountered in our study was nodular colloid goiter while the most common malignant lesion was papillary carcinoma of the thyroid followed by follicular carcinoma. We encounter only one case of anaplastic carcinoma in our study.
As stated earlier, inadequate sample size makes diagnosis by FNAC a difficult task. In published data inadequate sample size ranges between 2 to 20 % [20]. In our study this rate was 2.4 %. Percentage of cases showing follicular lesion of undetermined significance and atypia of undetermined significance were 1.8 and 4.1 % respectively. This is comparable with published literature where the percentage ranges between 2-6 % [21]. Sinna et al. [18] found a much higher percentage of follicular lesion of undetermined significance on FNAC which is 13.5 %. They attributed this due to varied impression of FNAC by different cytopathologist especially when diagnosing category 3 cases on FNAC.
In our study FNAC has a sensitivity of 84.2 %, specificity of 97.2 % and a diagnostic accuracy of 94.5 %. The positive predictive value was 88.3 % and the negative predictive value was 96 %. Sensitivity, specificity and diagnostic accuracy were comparable to other studies done by Basharat et al. [22] and Kessler et al. [23]. We calculated the FP rate and FN rate for our study. FN rate is the percentage of patients found to have malignant thyroid lesions on histopathology while their FNAC earlier was reported as benign. In our study FN rate was 15.7 %. Our value was relatively higher as compared to a series of studies where FN rate was reported to range from 1.5 to 11.5 % [24]. The FP rate indicates that a patient with malignant FNAC result was found on histological examination to have benign lesion. The FP rate in our study was 2.8 % which was comparable to published literature [24,25].

Conclusion
FNAC is a more specific than sensitive when distinguishing between benign and malignant lesions. It has good diagnostic accuracy and hence should be used in all tertiary care hospitals prior to surgery to diagnose thyroid lesions.