Artificial Intelligence for Pre-operative Diagnosis of Malignant Thyroid Nodules Based on Sonographic Features and Cytology Category.

BACKGROUND: Current diagnosis and classification of thyroid nodules are susceptible to subjective factors. Despite widespread use of ultrasonography (USG) and fine needle aspiration cytology (FNAC) to assess thyroid nodules, the interpretation of results is nuanced and requires specialist endocrine surgery input. Using readily available pre-operative data, the aims of this study were to develop artificial intelligence (AI) models to classify nodules into likely benign or malignant and to compare the diagnostic performance of the models. METHODS: Patients undergoing surgery for thyroid nodules between 2010 and 2020 were recruited from our institution's database into training and testing groups. Demographics, serum TSH level, cytology, ultrasonography features and histopathology data were extracted. The training group USG images were re-reviewed by a study radiologist experienced in thyroid USG, who reported the relevant features and supplemented with data extracted from existing reports to reduce sampling bias. Testing group USG features were extracted solely from existing reports to reflect real-life practice of a non-thyroid specialist. We developed four AI models based on classification algorithms (k-Nearest Neighbour, Support Vector Machine, Decision Tree, Naïve Bayes) and evaluated their diagnostic performance of thyroid malignancy. RESULTS: In the training group (n = 857), 75% were female and 27% of cases were malignant. The testing group (n = 198) consisted of 77% females and 17% malignant cases. Mean age was 54.7 ± 16.2 years for the training group and 50.1 ± 17.4 years for the testing group. Following validation with the testing group, support vector machine classifier was found to perform best in predicting final histopathology with an accuracy of 89%, sensitivity 89%, specificity 83%, F-score 94% and AUROC 0.86. CONCLUSION: We have developed a first of its kind, pilot AI model that can accurately predict malignancy in thyroid nodules using USG features, FNAC, demographics and serum TSH. There is potential for a model like this to be used as a decision support tool in under-resourced areas as well as by non-thyroid specialists.

Impact of dedicated multidisciplinary service on patient selection and outcomes for surgical treatment of primary aldosteronism.

BACKGROUND: Primary aldosteronism is the most common surgically curable cause of endocrine hypertension. Management of the unilateral subtype of primary aldosteronism with adrenalectomy requires multidisciplinary input. It is unclear if a dedicated endocrine hypertension service confers better outcomes compared to standard care offered by individual clinicians. METHODS: In this retrospective study, patients from the Monash University Endocrine Surgery Database were divided into either the endocrine hypertension service group, where patients were managed by a dedicated multidisciplinary team, or the standard group, where patients were managed by individual clinicians. The comparisons included patient selection for surgery, perioperative blood pressure control, and surgical cure rate. RESULTS: Despite similar perioperative blood pressure, patients in the endocrine hypertension service group (n = 41) were on fewer antihypertensive medications (1 vs 2, P = .011) compared to the standard group (n = 55). A larger proportion of patients in the endocrine hypertension service group had either bilateral adrenal nodules or no adrenal lesions on computed tomography (41% vs 18%, P = .013). Patients in the standard group had larger adrenal lesions on computed tomography (median 15 mm vs 10 mm, P = .032). Postoperatively, the biochemical cure rate was higher in the endocrine hypertension service group at 6 to 12 months (97% vs 76%, P = .021). CONCLUSION: Patients managed by endocrine hypertension service were more likely to be diagnosed with surgically curable primary aldosteronism without a unilateral adrenal adenoma on imaging, required fewer medications for perioperative blood pressure control, and experienced superior postoperative outcomes. Referral to a dedicated endocrine hypertension service is recommended for patients with primary aldosteronism who wish to pursue a surgical cure.

Preoperative serum calcitonin may improve initial surgery for medullary thyroid cancer in patients with indeterminate cytology.

BACKGROUND: Medullary thyroid cancer (MTC) is rare, with poorer outcomes than differentiated thyroid cancer. We aimed to identify areas for improvement in the pre-operative evaluation of patients with possible MTC in a high-volume endocrine surgery unit in accordance with current practice guidelines. We hypothesised that the selective use of serum calcitonin (sCT) as a biomarker for possible MTC could guide the extent of initial surgical management. METHODS: We recruited MTC patients between 2000 and 2020 from the Monash University Endocrine Surgery Unit database. Demographics, tumour characteristics, pre-operative evaluation, operative management, and outcomes were analysed. RESULTS: Of 1454 thyroid cancer patients, 43 (3%) had MTC. Pre-operatively, 36 (84%) patients with MTC confirmed on cytology (28, 65%), elevated sCT (6, 14%) or RET mutation (2, 4%). Of these 36 patients, 31 (86%) had optimal extent of thyroidectomy and lymph node dissection (LND). Five (14%) had less than total thyroidectomy due to nerve injury. Thirty-four patients had compartmental LND. In the 12 (27%) patients with indeterminate or non-diagnostic cytology, 5 had elevated sCT and were managed as above. None of the remaining seven had LND, thus potentially suboptimal surgery. CONCLUSION: Our findings reflect the rarity of MTC, and the challenges of pre-operative diagnosis. The addition of sCT may improve surgical planning in patients with indeterminate cytology.

Hypotension in Posterior Retroperitoneoscopic Versus Transperitoneal Laparoscopic Adrenalectomy.

INTRODUCTION: Despite preoperative optimization, hemodynamic instability can be a major challenge during adrenalectomy. Even brief episodes of intraoperative hypotension can be associated with ischemia-reperfusion injury. This study aimed to compare intraoperative hemodynamic parameters between posterior retroperitoneoscopic adrenalectomy (PRA) and transperitoneal laparoscopic adrenalectomy (TPA). METHODS: This is a retrospective study of patients undergoing PRA and TPA without conversion or concomitant intraabdominal pathology from 2008 to 2019. The primary outcome was intraoperative hypotension defined by mean arterial pressure <60 mm Hg or the need for ≥1 intravenous vasopressors at least 30 min after anesthetic induction. RESULTS: Overall, 108 patients met the inclusion criteria; 33 (30.6%) had pheochromocytoma, 26 (24.1%) had aldosterone excess, 8 (7.4%) had corticosteroid excess, and 41 (38.0%) had nonfunctioning adrenal tumors. Of these, 68 (63.0%) underwent PRA and 40 (37.0%) underwent TPA. Age, sex, body mass index, preinduction blood pressure, number of preoperative antihypertensives, and histopathological diagnosis were similar in the two groups. Tumor size was greater in the TPA group. The presence of pheochromocytoma was an independent risk factor for hypotension. Multivariate analysis revealed that PRA was associated with a higher risk of experiencing a mean arterial pressure <60 mm Hg (odds ratio 4.44, 95% confidence interval 1.27-15.54, P = 0.02) and the need for ≥1 intravenous vasopressors (odds ratio 9.97, 95% confidence interval 3.34-29.78, P < 0.001) compared with TPA. CONCLUSIONS: Although PRA offers several advantages over TPA, it carries a greater risk of intraoperative hypotension. A prospective trial is required to validate these findings. Nevertheless, institution of risk reduction strategies is encouraged to be considered for individuals undergoing PRA.

Preoperative mineralocorticoid receptor antagonist reduces postoperative hyperkalaemia in patients with Conn syndrome.

BACKGROUND: The preoperative use of mineralocorticoid receptor antagonists (MRA) in patients with unilateral forms of primary aldosteronism (PA) is not standardized. The current Endocrine Society Guidelines do not specifically recommend MRA treatment before surgery. It is unclear whether preoperative MRA can optimize perioperative blood pressure and potassium control, and reduce the incidence of postoperative hyperkalaemia. OBJECTIVE: This study aimed to investigate the effect of MRA on the incidence of postoperative hyperkalaemia in addition to perioperative blood pressure and potassium concentration in patients undergoing unilateral adrenalectomy for the treatment of PA. DESIGN: Retrospective cohort study. SETTING: Tertiary referral centres, Victoria, Australia. PATIENTS: A total of 96 patients who were diagnosed with unilateral forms of PA: 73 patients ('MRA' group) received preoperative MRA while 23 patients ('No-MRA' group) did not. RESULTS: The prevalence of postoperative hyperkalaemia was significantly higher in the 'No-MRA' group at 2-4 weeks after surgery, compared to the 'MRA' group (35% vs. 11%, p = .014). In a logistic regression, the use of MRA significantly predicted a lower incidence of postoperative hyperkalaemia after adjusting for age, sex, baseline aldosterone-to-renin ratio, potassium and preoperative eGFR. Before surgery, patients in the 'MRA' group had normalized blood pressure and potassium concentration requiring fewer antihypertensive medications and no potassium supplements. CONCLUSION: Preoperative MRA use was associated with optimal perioperative blood pressure and normalized serum potassium in addition to a lower incidence of postoperative hyperkalaemia. MRA should be considered standard treatment for patients awaiting surgery for PA.

Cancer in thyroglossal duct cysts.

BACKGROUND: Thyroglossal duct cysts (TDC) account for 7% of midline neck swellings. TDC carcinoma (TDCC) is rare, reported in 1% of all TDCs. We aimed to describe the incidence of TDCC, the accuracy of fine needle aspiration cytology (FNAC), and to identify suspicious, predictive ultrasound (USG) features. METHOD: A cross sectional study of TDC patients in the Monash University Endocrine Surgery Database from 2001 to 2021. RESULTS: Eighty-six patients had surgery for TDC, of median age 48 and a female preponderance (62%). Preoperative USG was used in 85% and FNAC in 57%. There were six cases (7%) of TDCC with papillary thyroid cancer (PTC). USG in five TDCC cases showed a solid nodule within the cyst wall. FNAC was undertaken in five TDCC cases from the solid nodule; malignant in one and suspicious for malignancy in two. Two TDCC patients had concurrent total thyroidectomy (TT), and three subsequently had a TT. Multifocal thyroid PTC was found in two patients. One patient had a recurrence in the lateral lymph nodes, nine years after excision of the TDCC and TT. All patients are alive and well with no distant metastases (median follow up 11 years). CONCLUSION: The 7% incidence of TDCC is higher than the usually reported 1% in the literature. We recommend a preoperative USG to evaluate the TDC, thyroid gland and cervical lymph nodes, and an FNAC to target the solid component of the TDC, and favour total thyroidectomy for all patients with TDCC.

Less extensive surgery for low-risk papillary thyroid cancers post 2015 American Thyroid Association guidelines in an Australian tertiary centre.

INTRODUCTION: The 2015 American Thyroid Association guidelines (ATA15) consider hemithyroidectomy (HT) a viable treatment option for low-risk papillary thyroid cancers (PTCs) between 1 and 4 cm. We aimed to examine the impact of ATA15 in a high-volume Australian endocrine surgery unit. METHODS: A retrospective study of all patients undergoing thyroidectomy from January 2010 to December 2019. INCLUSION CRITERIA: PTC histopathology, Bethesda V-VI, size 1-4 cm, and absence of clinical evidence of lymph node or distant metastases pre-operatively. Primary outcome was rate of HT before and after ATA15. RESULTS: Of 5408 thyroidectomy patients, 339 (6.3%) met the inclusion criteria - 186 (54.9%) pre-ATA15 (2010-2015) and 153 (45.1%) post-ATA15 (2016-2019). The patient groups were similar; there were no significant differences between groups in age, sex, tumour size, proportion with Bethesda VI cytology, compressive symptoms, or thyrotoxicosis. Post-ATA15, there was a significant increase in HT rate from 5.4% to 19.6% (P = 0.0001). However, there was no corresponding increase in completion thyroidectomy (CT) rate (50.0% versus 27.6%, P = 0.2). The proportion managed with prophylactic central neck dissection (pCND) fell from 80.5% to 10.8% (P < 0.0001). Pre-ATA15, the only factor significantly associated with HT was Bethesda V. In contrast, post-ATA15, HT was more likely in patients with younger age, smaller tumours, and Bethesda V. CONCLUSION: After the release of 2015 ATA guidelines, we observed a significant increase in HT rate and a significant decrease in pCND rate for low-risk PTCs in our specialised thyroid cancer unit. This reflects a growing clinician uptake of a more conservative approach as recommended by ATA15.

Investigation of recurrent laryngeal palsy rates for potential associations during thyroidectomy.

BACKGROUND: There are many clinical associations and potential mechanisms of injury resulting in recurrent laryngeal nerve palsy (RLNP) after thyroidectomy. One possible cause of RLNP is focal intralaryngeal compression of the recurrent laryngeal nerve (RLN), which may be associated with the tracheal tube (TT). Therefore, we examined current RLNP rates to investigate potential associations, including intralaryngeal, airway, anaesthetic and anthropometric factors. METHODS: We analysed 1003 patients undergoing thyroid surgery at The Alfred from 2010 to 2017, who had anatomically intact RLNs at the conclusion of thyroidectomy. All included patients underwent pre- and post-operative flexible nasendoscopy. The primary outcome was RLNP rate. We analysed potential associated factors including age, sex, operative time, surgical indication, pathology, American Society of Anaesthesiologists Physical Status, Mallampati scores, body mass index, intubation grade, TT size and specimen weight. The independent risk factors were identified by logistic regression analysis. RESULTS: Overall, RLNP occurred in 83 patients (8.3%) of which one was permanent (0.1%). On univariate analysis, RLNP was associated with male sex (P = 0.02), and duration of surgery (P = 0.002). On multivariate analysis, both male sex (P = 0.047) and duration of surgery (P = 0.04) remained significant. Further, factors postulated to cause intralaryngeal compression of the RLN, including TT size, body mass index, intubation grade and Mallampati score, were not significantly associated with RLNP. CONCLUSION: Our study showed a RLNP rate of 8.3%, and associations with longer operative duration, and male sex. Potential intralaryngeal factors were not identified.

Quantifying the differences in surgical management of patients with definitive and indeterminate thyroid nodule cytology.

INTRODUCTION: Thyroid nodules are increasingly common. Despite being an essential pre-operative diagnostic tool, up to 30% of fine needle aspirate cytology (FNAC) yields a non-definitive diagnosis. This study aimed to quantify differences in surgical management of patients with definitive and indeterminate thyroid nodule cytology, and determine if clinical factors can improve cytological diagnosis. MATERIALS AND METHODS: Patients who underwent thyroidectomy for nodules from 2001 to 2015 were recruited. Those with benign and malignant preoperative cytology were included in the "definitive diagnosis" (DC) group; patients with all other preoperative cytology results were included in the "indeterminate diagnosis" (IC) group. We compared demographics and procedures between these groups. Clinical factors and demographics were also compared between patients with benign and malignant histology in the IC group. RESULTS: A total of 3821 cases were included. A significantly larger proportion of the IC patients had a hemithyroidectomy (IC 69% vs. DC 39%, p < 0.001) initially, and also had a significantly higher rate of two-stage surgery compared to the DC group (IC 17% vs. DC 11%, p < 0.001). Patients in the DC group were twice as likely to undergo concurrent central lymph node dissection for papillary and medullary cancers than the IC group (p < 0.001). Overall, up to 60% of IC patients had been over- or under-treated at initial surgery. The clinical factors examined were not significantly associated with higher risk of malignancy in IC patients. CONCLUSION: This study highlights the potential for improved preoperative diagnosis to streamline decision making for surgical management of patients with thyroid nodules.

Anti-Thyroid Antibodies and TSH as Potential Markers of Thyroid Carcinoma and Aggressive Behavior in Patients with Indeterminate Fine-Needle Aspiration Cytology.

BACKGROUND: Indeterminate fine-needle aspiration cytology (FNAC) imposes challenges in the management of thyroid nodules. This study aimed to examine whether preoperative anti-thyroid antibodies (Abs) and TSH are indicators of thyroid malignancy and aggressive behavior in patients with indeterminate FNAC. METHODS: This was a retrospective study of thyroidectomy patients from 2008 to 2016. We analyzed Abs and TSH levels, FNAC, and histopathology. Serum antibody levels were categorized as 'Undetectable', 'In-range' if detectable but within normal range, and 'Elevated' if above upper limit of normal. 'Detectable' levels referred to 'In-range' and 'Elevated' combined. RESULTS: There were 531 patients included. Of 402 patients with preoperative FNAC, 104 (25.9%) had indeterminate cytology (Bethesda III-V). Of these, 39 (37.5%) were malignant and 65 (62.5%) benign on histopathology. In the setting of indeterminate FNAC, an increased risk of malignancy was associated with 'Elevated' thyroglobulin antibodies (TgAb) (OR 7.25, 95% CI 1.13-77.15, P = 0.01) and 'Elevated' thyroid peroxidase antibodies (TPOAb) (OR 6.79, 95% CI 1.23-45.88, P = 0.008). Similarly, while still 'In-range', TSH ≥ 1 mIU/L was associated with an increased risk of malignancy (OR 3.23, 95% CI 1.14-9.33, P = 0.01). In all patients with malignancy, the mean tumor size was 8 mm larger in those with TSH ≥ 1 mIU/L (P = 0.03); furthermore, in PTC patients, 'Detectable' TgAb conferred a 4 × risk of lymph node metastasis (95% CI 1.03-13.77, P = 0.02). CONCLUSION: In this cohort, in indeterminate FNAC patients, Abs and TSH were associated with an increased risk of malignancy. Additionally, TgAb and TSH were potential markers of aggressive biology. As such, they may be diagnostic and prognostic adjuncts.

Routine Preoperative Laryngoscopy for Thyroid Surgery Is Not Necessary Without Risk Factors.

Background: Routine preoperative vocal cord (VC) assessment with laryngoscopy in patients undergoing thyroidectomy allows clear documentation of baseline VC function, aids in surgical planning in patients with palsies, and facilitates interpretation of intraoperative neuromonitoring (IONM) findings. We aimed to determine the incidence of preoperative vocal cord palsy (VCP); to evaluate the associated risk factors for preoperative VCP; and to calculate the cost-savings potential of implementing a selective approach. Methods: Patients with a pre-thyroidectomy VC assessment by fiberoptic laryngoscopy were retrospectively recruited from the Monash University Endocrine Surgery Unit database from 2000 to 2018. Cases with preoperative VCP were reviewed for potential contributing factors and compared with a non-palsy cohort. Results: Of the 5987 patients who had preoperative laryngoscopy, VCP was documented in 41 (0.68%) patients. Four clinical parameters were found to be potential indicators of VCP, including: age (p < 0.001), nodule ≥3.5 cm recorded on ultrasound imaging (p = 0.01), presence of voice symptoms (p < 0.001), and previous neck surgery (p < 0.001). Malignant cytology (p = 0.5) and exposure to head and neck irradiation were not different between the groups. Utilizing these risk factors, 2354 (39%) patients had at least one feature that may raise suspicion for preoperative VCP. By performing preoperative laryngoscopy only on this subset of patients, the potential cost savings exceeds 400 Australian Dollars per patient. Conclusions: Using this large dataset, we have established that a VCP is rare in the absence of a large nodule, hoarseness, or previous neck surgery. Therefore, in the era of IONM, we support a selective approach to preoperative laryngoscopy by using the aforementioned criteria.

Temporal analysis of thyroid cancer management in a Melbourne tertiary centre.

BACKGROUND: The American Thyroid Association (ATA) management guidelines for thyroid cancer were revised in 2009. The aim of this study was to determine if management of thyroid cancer in our institution has changed in accordance with the introduction of the revised ATA guidelines (ATA2009 ), and to compare the characteristics and management of thyroid cancer in a Melbourne endocrine surgery unit over a 7-year period. METHODS: All patients treated by the Monash University Endocrine Surgery Unit for thyroid cancer between 2007 and 2013 were divided into two groups - the pre-ATA2009 group (2007-2010) and the post-ATA2009 group (2011-2013). Comparisons were made of the demographics, cytology, pathology, surgical outcome and adjuvant therapy using t-test and chi-squared tests. RESULTS: There were 333 patients in the pre-ATA2009 group and 342 patients in the post-ATA2009 group. Fewer non-diagnostic fine-needle aspiration cytology results were identified in the post-ATA2009 group (4% versus 0.9%; P = 0.01), while the rates of other fine-needle aspiration cytology categories were similar. There was a reduction in the use of radioactive iodine ablation in the post-ATA2009 group, both in the proportion of patients being treated (66% versus 48%; P < 0.001) and the dosages used (mean 96 mCi versus 80 mCi; P < 0.01), despite similar tumour size in both groups. CONCLUSION: The key changes in practice thought to be attributable to the 2009 revised ATA guidelines were the reduction in the use and dosage of radioactive iodine in the management of differentiated thyroid cancer.

Adrenal incidentaloma follow-up is influenced by patient, radiologic, and medical provider factors: A review of 804 cases.

BACKGROUND: The majority of adrenal incidentalomas are benign, although some are large, functional, or malignant and may require surgery. Therefore all require follow-up. This study aimed to determine the pattern of adrenal incidentaloma follow-up in a level 1 trauma center, focusing on the factors that influence whether follow-up is facilitated. METHODS: Patients with computed tomography-detected adrenal incidentalomas between January 2010 and September 2015 were included. A keyword search identified case files, which were reviewed for demographic characteristics, managing unit, computed tomography indication and findings, and follow-up arrangements. Statistical analysis was performed using Stata SE Version 14. RESULTS: A total of 38,848 chest and abdominal computed tomographic scans were performed in the study period, revealing 804 patients with adrenal incidentalomas who met inclusion criteria (mean age 65, 58% male). The mean size of adrenal incidentaloma was 23 mm. Follow-up was organized in 30% of cases and was more likely to occur in younger patients (mean age 62 vs 66, P < .001); in larger lesions (mean size 26 mm vs 21 mm, P < .001); if the computed tomographic scan suggested follow-up (P < .001); or if the computed tomography report suggested a diagnosis (P < .001). Follow-up arrangements were most likely to be made by the trauma unit (39%, P = .01). CONCLUSION: This study highlights that adrenal incidentalomas follow-up is often overlooked, and that follow-up is influenced by patient, radiologic, and medical provider factors. An adrenal lesion follow-up protocol may improve follow-up rates but requires further analysis.