ORIGINAL RESEARCH


https://doi.org/10.5005/jp-journals-10002-1458
World Journal of Endocrine Surgery
Volume 15 | Issue 2 | Year 2023

A 12-year Experience with 298 Fine Needle Aspirations of Thyroid Nodules in a Single Institution: Analysis of Bethesda System of Thyroid Cytopathology Reporting and Cytohistopathological Correlation


Bodhireddy S Reddy1, Shabnam Karangadan2https://orcid.org/0000-0003-1826-2933, Sarada C Devi3, Anuradha Boyareddigari4

1Department of Pathology, Rangaraya Medical College, Kakinada, Andhra Pradesh, India

2Department of Pathology, Lal Path Labs Ltd, Kochi, Kerala, India

3,4Department of Pathology, Sri Venkateswara Medical College, Tirupati, Andhra Pradesh, India

Corresponding Author: Shabnam Karangadan, Lal Path Labs Ltd, Kochi, Kerala, India, Phone: +91 9845869698, e-mail: shabnamk126@gmail.com

Received on: 06 December 2023; Accepted on: 21 September 2023; Published on: 30 October 2023

ABSTRACT

Context: Fine needle aspiration cytology (FNAC) has become a critical step in the evaluation of thyroid nodules and distinguishing neoplastic from nonneoplastic nodules. The Bethesda system for reporting thyroid cytopathology (TBSRTC) helps in a universal diagnostic terminology for the clinicians to decide on the appropriate management and reduce the risk of unnecessary surgeries.

Aims: To assess the utility of the Bethesda system of thyroid cytopathology reporting and efficacy of FNAC with histopathological correlation of subsequent surgically resected specimens.

Materials and methods: The present study includes 3,212 cases with thyroid nodules that underwent FNAC over a 12-year period. The cases were categorized as per the Bethesda system and correlated with clinical features, hormonal profile, and histopathological examination (HPE).

Results: Out of 3,212 cases, the most common Bethesda system category was benign (88.7%) followed by follicular neoplasm (FN) or suspicious for an FN (SFN) (7.3%). Cytohistopathological correlation was done in 298 cases and discrepancy was noted in 28 cases (9.4%). Statistical analysis showed that FNAC with the Bethesda system of reporting has a sensitivity of 83.7%, specificity of 95.4%, positive predictive value (PPV) of 92.8%, negative predictive value (NPV) of 89.3%, and diagnostic accuracy of 90.6%.

Conclusion: Fine needle aspiration cytology (FNAC) is a reliable diagnostic tool in the evaluation of thyroid nodules as indicated by high sensitivity, high specificity, and low rate of surgical intervention in this large comprehensive study. The inclusion of the Bethesda system has provided a more precise cytological diagnosis to effectively guide clinicians for further management.

How to cite this article: Reddy BS, Karangadan S, Devi SC, et al. A 12-year Experience with 298 Fine Needle Aspirations of Thyroid Nodules in a Single Institution: Analysis of Bethesda System of Thyroid Cytopathology Reporting and Cytohistopathological Correlation. World J Endoc Surg 2023;15(2):40–45.

Source of support: Nil

Conflict of interest: None

Keywords: Bethesda system, Cytohistopathological correlation, Fine needle aspiration cytology, Thyroid cytopathology

KEY MESSAGE

The FNAC is a simple and accurate method to triage patients with thyroid nodules, thereby reducing the rate of unnecessary surgeries. The addition of the Bethesda system has further standardized thyroid cytopathology reporting and improved communication between pathologists and clinicians.

INTRODUCTION

Thyroid swelling is a frequent problem mainly in young females, resulting in pressure symptoms, and cosmetic deformity. The prevalence of palpable thyroid nodules ranges from 4 to 7% in adults and 0.2–1.5% in children. In endemic goiter areas and when aided by ultrasound (USG), these values are much higher.1

Most lumps are benign, with cancer occurring in about 5–20% of cases.2 Identifying this small number of malignant tumors from the numerous benign nodules is a challenge in routine clinical practice.

In recent years, fine needle aspiration cytology (FNAC) has become a valuable step in the assessment of neck masses. The strategy for the management of thyroid swelling involves a detailed history taking, thorough clinical examination, thyroid function test (TFT), USG, FNAC, scintiscan, and estimation of tumor markers.3 FNA thyroid is a fast, cost-effective, and minimally invasive screening technique. It is immensely valuable especially in differentiating neoplastic from nonneoplastic nodules and reducing the rate of needless surgery.4

To streamline the multiple terminologies involved in thyroid FNAC reporting, the Bethesda system for reporting thyroid cytopathology (TBSRTC) was introduced at the “Thyroid Fine Needle Aspiration State of the Science Conference” held in Bethesda, Maryland.5 This helped unify the nomenclature and define the morphologic criteria and its corresponding malignancy risk.

This study was undertaken to evaluate the utility of TBSRTC, evaluate the efficacy of FNAC in the preoperative diagnosis of thyroid lesions, and correlate cytology findings with histopathology so that unnecessary thyroidectomies in benign conditions can be avoided.

MATERIALS AND METHODS

A total of 3,212 cases of palpable thyroid nodules that underwent FNAC over a 12-year period (August 2001 to July 2012) were analyzed both prospectively and retrospectively. Out of this, for 298 cases surgically resected specimens were received and were included in the study. Clinical details like age, sex, pressure symptoms, toxic symptoms, TFTs, site, and size of the lesion were retrieved from the clinical records.

The FNAC was performed in each case following informed consent from the patient. In special circumstances, FNAC was done under USG guidance whenever the lesion was difficult to localize. Smears prepared with the material were fixed in 95% isopropyl alcohol and stained with hematoxylin and eosin (H&E). Based on the cytological interpretation of the smears the cases were categorized into the following groups as per TBSRTC.

Surgically resected specimens were formalin-fixed, and sections from the lesion along with adjacent normal thyroid were paraffin-embedded and stained by H&E for histopathological study to assess the size, consistency, cut section, number of lesions, and nature of the lesion.

The statistical analysis for the correlation between cytological and histopathological diagnosis and efficacy of FNAC was determined using the methodology of Galen and Gambino.6

Sensitivity:

Specificity:

Positive predictive value (PPV):

Negative predictive value (NPV):

Diagnostic accuracy:

Discordance rate:

(TP = true positive, FP = false positive, TN = true negative, FN = false negative).

RESULTS

During the period of August 2001 to July 2012, 3,212 FNAs of the thyroid were done. Among these, histopathological correlation was done in 298 cases. According to TBSRTC, 2,850 (88.7%) were classified as benign, 25 (0.8%) as AUS/FLUS, 234 (7.3%) as FN/SFN, 17 (0.5%) as suspicious for malignancy, 38 (1.2%) as malignancy, and 48(1.5%) as nondiagnostic or unsatisfactory for diagnosis on cytology (Fig. 1). The histopathology diagnosis has been illustrated in Table 1.

Table 1: Distribution of thyroid lesions by histopathology
Histological diagnosis No. of cases
Nonneoplastic lesions 175 (58.7%)
Colloid cyst 3 (1%)
Colloid goiter 5 (1.7%)
Multinodular goiter 138 (46.3%)
Hashimoto thyroiditis 29 (9.7%)
Neoplastic lesions 123 (41.3%)
Follicular adenoma 81 (27.3%)
Hyalinizing trabecular adenoma 1 (0.3%)
Papillary carcinoma 35 (11.7%)
Follicular carcinoma 6 (2%)
Total 298

Fig. 1: Cytological diagnoses of thyroid lesions as per Bethesda system (n = 3,212)

The age of 298 patients ranged from 13 to 73 years. Most of the cases, that is, 98 cases (32.9%) occurred in the 31–40 years age-group. Amongst the malignant lesions, most of the papillary carcinoma cases were in the 5th decade (34.3%) and follicular carcinoma cases were in the older age group of the 6th decade (50%). Out of 298 cases, 273 (91.6%) were female and 25 (8.4%) were male with clear female preponderance across all entities.

Among 298 cases the site of the lesion was available for 131 cases. Out of 131 cases, 78 (59.5%) were in the right lobe, 29 (22.2%) in the left lobe, and 24(18.3%) in both lobes of the thyroid. In contrast, 10 (76.9%) out of 13 cases of Hashimoto’s thyroiditis showed involvement of both lobes. In most of the multinodular goiter cases, 72 (52.2%) were >5 cm in size, and amongst neoplastic lesions, most (76; 61.8%) were in the range of 2–5 cm.

Out of 134 (45%), patients presented with neck swelling with pressure symptoms like dysphagia, dyspnea, and hoarseness of voice. Of these, the majority of malignant cases (21/41; 51.2%) and Hashimoto’s thyroiditis cases (16/29; 55.2%) showed pressure symptoms. Toxic symptoms like palpitations, tremors, and exophthalmos were seen in 11 (3.7%) patients, which included nine cases (6.5%) of multinodular goiter, one case (3.5%) of Hashimoto’s thyroiditis, and one case (2.9%) of papillary carcinoma thyroid. Among 298 cases, thyroid hormone profile was available for 30 cases. Out of 22 (73.3%) were in euthyroid, six (20%) were in hypothyroid, and two (6.7%) were in hyperthyroid state.

Out of 298 cases, 111 (37.2%) were clinically diagnosed as solitary thyroid nodule (STN), 108 (36.2%) as nonneoplastic lesions, 57 (19.1%) as neoplastic, and the diagnosis was not made in 22 (7.4%) cases. Among 111 cases of STN, 59 (53.1%) were nonneoplastic lesions, 45 (40.3%) were follicular adenoma, five (4.3%) were papillary carcinoma, two (1.6%) were follicular carcinoma, and one (0.7%) was hyalinizing trabecular adenoma on histopathology. Out of 108 nonneoplastic lesions only 77 (71.4%) and out of 57 neoplastic lesions, 31 (54.4%) were confirmed by histopathology (Table 2).

Table 2: Correlation of clinical diagnosis with histopathological diagnosis
Clinical diagnosis Histopathological diagnosis Total
CC CG MNG HT FA HTA PCT FCT
Nonneoplastic Colloid cyst (CC) 1 - 2 - 1 - - - 4
Colloid goiter (CG) - 2 7 1 1 - 1 - 12
Multinodular goiter (MNG) - - 49 14 17 - 9 1 90
Thyroiditis - - - - 1 - - - 1
Thyrotoxicosis - - 1 - - - - - 1
STN 1 2 47 8 45 1 5 2 111
Neoplastic Adenoma thyroid - - 15 1 10 - 5 1 32
Ca thyroid - - 4 1 - - 10 - 15
PCT - - 2 1 1 - 4 - 8
FCT 1 - 1 - - - - - 2
No definite diagnosis - 1 10 3 5 - 1 2 22
Total 3 5 138 29 81 1 35 6 298

HT, Hashimoto thyroiditis; FA, follicular adenoma; HTA, hyalinizing trabecular adenoma; PCT, papillary carcinoma thyroid; FCT, follicular carcinoma thyroid

Cytohistopathological correlation of 298 cases revealed that 187 (62.7%) lesions were diagnosed as benign, 73 (24.5%) as FN/SFN, 13 (4.4%) as suspicious for malignancy and 25 (8.4%) as malignancy by FNA. From 187 lesions diagnosed as benign on cytology, five (2.7%) emerged as papillary carcinoma, one (0.5%) follicular carcinoma, and 14 (7.5%) follicular adenomas. Rest [167 (89.3%) cases] were benign on histopathology as well. Out of 73 lesions diagnosed as FN/SFN by FNA, four (5.5%) were diagnosed as multinodular goiter and the rest 69 (94.5%) cases proved to be FNs on histopathology. Of 13 cases reported as suspicious for malignancy, nine (69.2%) were papillary carcinoma, and four (30.8%) were follicular adenoma on histology. Of the 25 cases diagnosed as malignancy, four (16%) were benign lesions, and the rest 21 (84%) were papillary carcinoma on histopathological examination (HPE) (Table 3 and Figs 234).

Table 3: Correlation of Bethesda categories of FNA reporting and histopathological diagnosis
Cytological diagnosis Histopathological diagnosis Malignancy rate (%)
CC CG MNG HT FA HTA PCT FCT Total
Nondiagnostic or unsatisfactory 0 0%
Benign 3 5 131 28 14 5 1 187 3.2%
AUS or FLUS 0 0%
FN or SFN 4 63 1 5 73 6.8%
Suspicious for malignancy 4 9 13 69.2%
Malignancy 3 1 21 - 25 84%
Total 3 5 138 29 81 1 35 6 298

Figs 2A to C: Hashimoto’s thyroiditis. (A) Specimen showing tan colored nodule; (B) Cytology smear showing Hürthle cell clusters and lymphoid aggregates (H&E, 400x); (C) Histopathology section showing lymphoid follicles and oxyphilic change of epithelium (H&E, 400x)

Figs 3A to C: Follicular adenoma. (A) Specimen showing a well-encapsulated nodule; (B) Cytology smear showing repetitive pattern (H&E, 100x); (C) Histopathology section showing microfollicular pattern (H&E, 100x)

Figs 4A to C: Papillary carcinoma. (A) Specimen showing grey–white tumor with a granular texture; (B) Cytology smear showing intranuclear pseudoinclusions (H&E, 400x); (C) Histopathology section showing multiple branching true papillae (H&E, 400x)

The malignancy rate noted for the benign category on FNAC was 3.2, 6.8% for FN/SFN, 69.2% for suspicious malignancy, and 84% malignancy category.

A false negative diagnosis is a cytological diagnosis of a nonneoplasm that proved to be a neoplasm. A false positive diagnosis is a cytological diagnosis of a neoplasm (requiring surgical treatment) that was confirmed nonneoplasm. Sensitivity, specificity, PPV, NPV, and accuracy of FNA relative to final histological diagnoses were calculated (Table 4).

Table 4: Statistical analysis for neoplastic lesions
Test being evaluated (FNAC) Reference standard test (histopathology)
Positive Negative
Positive + suspicious 103 (true positives) 8 (false positives)
Negative 20 (false negatives) 167 (true negatives)

DISCUSSION

Fine needle aspiration (FNA) of the thyroid was first demonstrated by Bibbo and Wilbur.1 FNAC is a simple and accurate method to triage patients with thyroid swellings to determine further intervention. With FNA the incidence of thyroid malignancy has doubled at surgery with an associated 50% reduction in the number of thyroidectomies performed.1,4

The FNA has some limitations too in terms of specimen adequacy, pathologist expertise, and overlapping cytological features.4 Introduction of the Bethesda system of reporting has allowed a more precise cytological diagnosis which in turn reduced confusion amongst clinicians in interpretation of the cytology report.

In the present study, 3,212 FNAs were done during an 11-year period. According to the Bethesda system, the majority (88.7%) were classified as benign followed by FN/SFN (7.3%) on cytology. This was in concordance with other studies (Table 5).7-12

Table 5: Comparison of cytological diagnoses of different thyroid lesions
Cytological diagnosis Present study Upadhyaya et al.7 Reddy et al.8 Nandedkar et al.9 Mufti and Molah10 Ugurluoglu et al.11 Kaur and Gupta12
Nondiagnostic or unsatisfactory 1.5% 2.8% 3.7% 4.29% 11.6% 12% 6.67%
Benign 88.7% 61.5% 89.2% 82.67% 77.6% 72% 84.67%
AUS/FLUS 0.8% 0 0.002% 0.82% 0.8% 2.6% 1.33%
FN/SFN 7.3% 11.9% 2% 9.07% 4% 3.2% 2.67%
Suspicious for malignancy 0.5% 4.6% 0.6% 1.15% 2.4% 4.2% 1.33%
Malignancy 1.2% 19.3% 4.1% 1.98% 3.6% 6% 3.33%
Sample size 3,212 109 484 606 250 1096 150

Out of the 3,212 cases, the surgically resected specimen was available in 298 cases. Among these nonneoplastic lesions were 58.7% and neoplastic were 41.3%. Multinodular goiter (78.9%) was the most common nonneoplastic lesion as in studies by Nandedkar et al.9 (58%) and Mufti and Molah4 (65.5%). The most common malignant lesion was papillary carcinoma which was in concordance with studies by Reddy et al.8 and Mufti and Molah.10 Neoplastic lesions were found to be more common than nonneoplastic in the study of Ko et al.4 This is not in concordance with our study and may be due to the variation in the geographical distribution of lesions (Table 6).

Table 6: Comparison of distribution of lesions of thyroid
Histological diagnosis Present study Reddy et al.8 Nandedkar et al.9 Mufti and Molah10 Ko et al.4
Nonneoplastic lesions 175 (58.7%) 43 (79.6%) 138 (80.7%) 58 (69.1%) 64 (30.9%)
Colloid cyst 3 (1%) - 13 (7.6%) -
Colloid goiter 5 (1.7%) - 30 (17.5%) 9 (10.7%) -
Multinodular goiter 138 (46.3%) 80 (46.8%) 38 (45.3%) 61 (29.4%)
Hashimoto’s thyroiditis 29 (9.7%) - 8 (4.7%) 10 (11.9%) 2 (1%)
Others 7 (4.1%) 1 (1.2%) 1 (0.5%)
Neoplastic lesions 123 (41.3%) 11 (20.4%) 33 (19.3%) 26 (30.9%) 143 (69.1%)
Follicular adenoma 81 (27.3%) 2 (3.7%) 12 (7%) 6 (7.1%) 41 (19.8%)
Hyalinizing trabecular adenoma 1 (0.3%) - - -
Papillary carcinoma 35 (11.7%) 8 (14.8%) 6 (3.5%) 13 (15.5%) 98 (47.4%)
Follicular carcinoma 6 (2%) 10 (5.9%) 2 (2.4%) 4 (1.9%)
Medullary carcinoma - 1 (1.9%) 4 (2.3%) 1 (1.2%) -
Anaplastic carcinoma - - 1 (1.2%) -
Lymphoma - 1 (0.6%) 3 (3.5%) -
Sample size 298 54 171 84 207

The age of the patients in the study ranged from 13 to 73 years with a mean of 39.4 years. The majority of the patients (32.9%) presented in their 4th decade of life and were females (91.6%) as in the other studies.3,8,9,11,12 In this study, most lesions (59.5%) were in the right lobe, whereas, Hashimoto thyroiditis involved both lobes (76.9%). The fact that the right lobe is more commonly involved was also noted by other studies.2,3

In terms of clinical presentation, swelling was the most common symptom as was noted in most of the other studies.2,3,9 Along with this 45% of the patients presented with pressure symptoms like dysphagia, dyspnea, and hoarseness of voice and 3.7% of the patients presented with toxic symptoms like palpitations, tremors, and exophthalmos.

In the current study thyroid profile data was available in 30 cases (10.1%). Most were in the euthyroid state (73.3%) followed by hypothyroid (20%) and hyperthyroid (6.7%) state. Similar findings were noted by Sengupta et al.3 and Bamanikar et al.2

In this study, cytohistological concordance was achieved in 270 cases (90.6%), which was similar to studies by Reddy et al.8 (90.8%) and Bamanikar et al.2 (94.3%). False-positive diagnosis of neoplasm was made in 2.7% of the cases. Four cases were reported as papillary carcinoma and four cases as FNs on cytology. Subsequent histology revealed nodular goiter and one case presumed as papillary carcinoma showed the features of Hashimoto thyroiditis. Incorrect diagnosis could be due to too much emphasis on cellularity and overlapping cytological features.

False negatives were noted in 6.7% of cases. Out of 20 cases that were reported benign on cytology, 14 cases exhibited features of follicular adenoma, five cases of papillary carcinoma, and one case of follicular carcinoma on HPE. Cytohistological discordance could be attributed to sampling error and cystic degeneration.

False positives and false negatives were 1.2 and 1.8%, respectively in the study of Nandedkar et al.9 and 1.8 and 21.5%, respectively in the study of Ko et al.4 Present study was similar to the study of Ko et al.4 as it was a larger study than one by Nandedkar et al.,9 hence, more reliable.

In this study, sensitivity was 83.7%, specificity of 95.4%, PPV of 92.8%, NPV of 89.3%, diagnostic accuracy of 90.6%, and discordance rate of 9.4% for identification of neoplasms on cytology.

Sensitivity, specificity, diagnostic accuracy, NPV, and PPV of the present study were nearer and comparable with other studies of Upadhyaya et al.,7 Reddy et al.,8 Nandedkar et al.,9 Ugurluoglu et al.,11 and Ko et al.4 (Table 7). Sensitivity was slightly lower in studies by Reddy et al.8 and Bamanikar et al.2 This could be attributed to the small sample size. USG guidance for FNA allows for better characterization and precise sampling of the lesion. Thereby significantly improves sensitivity and specificity.1

Table 7: Comparison of efficacy of FNAC in diagnosing neoplasms
Study Sample size Sensitivity Specificity Diagnostic accuracy NPV PPV
Reddy et al.8 54 72.72% 95.3% 90.7% 93.1% 80%
Nandedkar et al.9 171 85.7% 98.6% 97.1% 98.0% 90.0%
Ugurluoglu et al.11 1,096 93% 79% 88% 85% 90%
Bamanikar et al.2 104 50% 100% 94.2% 93.8% 100%
Ko et al.4 207 78.4% 98.2% 84.4% 66.3% 99%
Upadhyaya et al.7 109 75% 100% 93.6% 92% 100%
Present study 298 83.7% 95.4% 90.6% 89.3% 92.8%

CONCLUSION

The FNAC of thyroid swelling provides a reliable preoperative diagnosis compared to other diagnostic methods. Its favorable influence further aided by USG guidance, on the management of thyroid lesions, is best displayed by the low rate of surgical treatment in this study. Bethesda system has standardized thyroid cytopathology reporting and improved communication between cytopathologists and clinicians, thereby leading to more consistent management. A benign diagnosis on FNAC should be wary of false negative results occurring and these patients have to be followed up, especially in cases with strong clinical suspicion of malignancy.

ORCID

Shabnam Karangadan https://orcid.org/0000-0003-1826-2933

REFERENCES

1. Bibbo M, Wilbur DC. Comprehensive Cytopathology, 4th edition. Philadelphia: Elsevier; 2015.

2. Bamanikar S, Soraisham P, Jadhav S, et al. Cyto-histology and clinical correlation of thyroid gland lesions: a 3-year study in a tertiary hospital. Clin Canc Investig J 2014;3(3):208–212. DOI: 10.4103/2278-0513.132112

3. Sengupta A, Pal R, Kar S, et al. Clinico-pathological correlates of incidentally revealed thyroid swellings in Bihar, India. J Pharm Bioall Sci 2012;4(1):51–55. DOI: 10.4103/0975-7406.92730

4. Ko HM, Jhu IK, Yang SH, et al. Clinico pathologic analysis of fine needle aspiration cytology of the thyroid. Acta Cytolog 2003;47(5):727–732. DOI: 10.1159/000326596

5. Cibas ES, Ali SZ. NCI Thyroid FNA State of the Science Conference. The Bethesda system for reporting thyroid cytopathology. Am J Clin Pathol 2009;132(5):658–665. DOI: 10.1309/AJCPPHLWMI3JV4LA

6. Galen RS, Gambino SR. Beyond normality-the predictive value and efficiency of medical diagnosis. New York: Wiley; 1975.

7. Upadhyaya P, Dhakal S, Adhikari P, et al. Histopathological review of diagnostic categories of the Bethesda system for reporting thyroid cytopathology – an institutional experience of 5 years. J Cytol 2019;36(1):48–52. DOI: 10.4103/JOC.JOC_64_18

8. Reddy P, Prakash A, Giriyan SS. Evaluation of Bethesda system for reporting thyroid cytology with histopathological correlation. Int J Res Med Sci 2018;6(1):247–252. DOI: 10.18203/2320-6012.ijrms20175728

9. Nandedkar SS, Dixit M, Malukani K, et al. Evaluation of thyroid lesions by fine –needle aspiration cytology according to Bethesda system and its histopathological correlation. Int J Appl Basic Med Res 2018;8(2):76–82. DOI: 10.4103/ijabmr.IJABMR_169_17

10. Mufti ST, Molah R. The Bethesda system for reporting thyroid cytopathology: a five-year retrospective review of one centre experience. Int J Health Sci (Qassim) 2012;6(2):159–173. DOI: 10.12816/0005991

11. Ugurluoglu C, Dobur F, Karabagli P, et al. Fine needle aspiration biopsy of thyroid nodules: cytologic and histopathologic correlation of 1096 patients. Int J Clin Exper Pathol 2015;8(11):14800–14805. PMID: 26823807.

12. Kaur A, Gupta A. FNAC of thyroid lesions based on Bethesda system: a retrospective cytological study in a tertiary care centre in North India. JK Sci 2021;23(4):175–179.

________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.