Marcadores radiológicos prequirúrgicos en pacientes intervenidos por nódulos tiroideos. Sistematización actualizada y clasificación estandarizada

Résumé

Introduction. When diagnosing a thyroid nodule (TN) we always start with an ultrasound study, without taking the patients symptoms into account. The finding will then determine the following strategy. In most cases, this leads to unnecessary additional examinations that can be inconclusive about the malignancy of the TN. When this occurs, we end up performing a diagnostic surgical intervention, with its associated morbidity. Lately, systematical and standardized thyroid cervical ultrasonography subject to malignancy, has led to multiple radiologic classifications being published in order to define nodule’s nature more precisely, which are used daily worldwide. The primary outcome of this study is to learn the diagnostic ability of the four most used radiologic classifications in the clinical practice as described in previous studies. These classifications are the following: The American Thyroid Association (ATA), The American College of Radiology (ACR-TIRADS); The Korean Thyroid Imaging Reporting and Data System (K-TIRADS) and The European Thyroid Imaging Reporting and Data System (EU-TIRADS). Materials and methods. We included all patients on which thyroid surgery was performed in the Department of General and Digestive Surgery of the Hospital Universitario Ramón y Cajal who had a TN diagnosed by ultrasound, disregarding the reason of surgery, between January 2020 and December 2021. We gathered the radiological ultrasonography characteristics of the TN and assigned each of them a category in each of the classifications analyzed, as well as the recommendation of cytologic study based on the size of the TN. We then compared these results with the preoperative cytology and with the pathologic results of the definitive specimen. The postoperative complications were registered with a 6-month follow-up. Results. After two years, we included 216 patients in this study who were put through thyroid surgery. Of these, 72.2% were women. The median of age was 59 (47.69). Only 4.6% of them had a previous cervical surgery. The preoperative TSH was 1.82 (0.88, 2.5) mUI/L. 63% of the interventions were hemithyroidectomies, 32.9% total thyroidectomies and 4.1% completion thyroidectomy. 28.7% of the patients received an associated cervical lymphadenectomy. We analyzed the intraoperative and postoperative complications registered in the patient’s clinical history. 10.2% had an intraoperative adverse event, all of them related to the loss of nervous signal of the recurrent laryngeal nerve (RLN) during intraoperative monitorization. The rate of postoperative complications was 26.4%. Transitory hypoparathyroidism, hypoparathyroidism persistent at 6 months (both calculated in patients submitted to total thyroidectomy or a completion thyroidectomy, transitory RLN palsy, persistent RLN palsy at 6 months and hematoma, were respectively 32.5%, 10%, 6.9%, 2.8% and 2.3%. 1.9% of patients were reoperated in the immediate postoperative period of hematoma or wound infection. We identified the radiological and anatomopathological characteristics of 259 TN, most of them solid (50.2%), isoechoic (54.4%), with round edged (79.5%), without calcifications (65.6%) and without halo (83.8%). The median size of the TN was of 22 (13,37) mm. The distribution of the TN cytologic study was the following: Bethesda I (4%), Bethesda II (29%), Bethesda III (19%), Bethesda IV (24%), Bethesda V (6%), Bethesda VI (18%). After studying the specimen, 38.6% of the TN were diagnosed as nodular hyperplasia and 23.2% as follicular adenoma. Papillary thyroid carcinoma (PTC) was diagnosed in 45 (17.4%) TN and papillary thyroid microcarcinoma (PTMC) in 26 (10%) TN. The median size of the TN as measured by the pathologists was 6 (15,39) mm. Lymphadenectomy was associated in case of confirmed malignancy and it demonstrated the presence of lymphatic metastasis in 32 patients (62.7%). The percentage of malignant histology in the different groups was Bethesda I (37%), Bethesda II (12.9%), Bethesda III (26.8%), Bethesda IV (36%), Bethesda V (69.2%) and Bethesda VI (97.4%). We obtained the value for sensibility (SE), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) of each ultrasonography category in all four classifications included in this study, based on the anatomopathological result of the specimen. When analyzing each individual category, the ATA classification had the best SE in the high-risk category (0.8) as opposed to ACR-TIRADS 5 (0.43), K-TIRADS 5 (0.52) o EU-TIRADS 5 (0.71). However, when adding TN size into the equation, SE grew equally in all classifications: ACR-TIRADS 5 (0.81), K-TIRADS 5 (0.82), EU-TIRADS 5 (0.71) y ATA (0.82). The PPV of all high-risk categories was ACR-TIRADS 5 (0.67), K-TIRADS 5 (0.6), EUTIRADS 5 (0.55) y ATA (0.46). When comparing classifications between them, the ATA and EU-TIRADS classified TN in a similar way. When analyzing the consistency in TN punction indication based on size, K-TIRADS and ATA (0.750) had the best results, although these were globally satisfactory. We analyzed SE, SP, PPV and NPV by subgroups based on the Bethesda category and we found that in the indetermined cytology group (Bethesda III) the ACR-TIRADS 5 had the highest values (SE = 1, SP= 0.16, PPV= 0.44 y PNV= 1). ACR-TIRADS also had the best values in the category for high suspicion or Bethesda IV (SE = 0.71, SP= 0, PPV= 0.62 y PNV= 1). In the Bethesda VI category, K-TIRADS 5 had a SE of 0.9, higher than that of the ATA (SE = 0.86), ACR-TIRADS (SE = 87) y EU-TIRADS (SE = 0.84). 25 TN did not meet cytologic criteria in any of the analyzed classifications. In this sample we obtained 52% of PTMC and 8% PTC. We demonstrated the presence of metastasis in 16% of patients. Of the 25 patients who did not have indication of Fine Needle Aspiration Biopsy (FNAB), 4 (16%) had some sort of postoperative complication and only one of these patients had a malignant TN. Conclusions. - The radiologic ultrasonography classifications have a diagnostic capacity when are compared with the definitive anatomopathological result in our cohort. - The correlation between these classifications is satisfactory, although when classifying TN and indicating FNA the ACR-TIRADS differs the most. - The link between the preoperative anatomopathological results and the ultrasonography findings differs according to the Bethesda classification. - The morbidity of our sample is slightly higher than that registered in other studies. - The malignancy rate in our sample was similar to that found in other studies, except for Bethesda II in which we had a higher rate. - Thyroid ultrasound could be useful in reducing the number of unnecessary thyroid surgeries and its associated morbidity.