Search

Objectives: The presence of nodal metastases in patients with papillary thyroid carcinoma (PTC) has both staging and treatment implications. However, lymph nodes are often not removed during thyroidectomy. Prior work has demonstrated the capability of artificial intelligence (AI) to predict the presence of nodal metastases in PTC based on the primary tumor histopathology alone. This study aimed to replicate these results with multi-institutional data. Methods: Cases of conventional PTC were identified from the records of 2 large academic institutions. Only patients with complete pathology data, including at least 3 sampled lymph nodes, were included in the study. Tumors were designated "positive" if they had at least 5 positive lymph node metastases. First, algorithms were trained separately on each institution's data and tested independently on the other institution's data. Then, the data sets were combined and new algorithms were developed and tested. The primary tumors were randomized into 2 groups, one to train the algorithm and another to test it. A low level of supervision was used to train the algorithm. Board-certified pathologists annotated the slides. HALO-AI convolutional neural network and image software was used to perform training and testing. Receiver operator characteristic curves and the Youden J statistic were used for primary analysis. Results: There were 420 cases used in analyses, 45% of which were negative. The best performing single institution algorithm had an area under the curve (AUC) of 0.64 with a sensitivity and specificity of 65% and 61% respectively, when tested on the other institution's data. The best performing combined institution algorithm had an AUC of 0.84 with a sensitivity and specificity of 68% and 91% respectively. Conclusion: A convolutional neural network can produce an accurate and robust algorithm that is capable of predicting nodal metastases from primary PTC histopathology alone even in the setting of multi-institutional data. [ABSTRACT FROM AUTHOR]

Objective(s): To review the experience of 3 hospitals with airway management during surgery for substernal goiter and identify preoperative factors that predict the need for advanced airway management techniques. Methods: A retrospective chart review between 2009 and 2017 of patients with substernal goiter treated surgically at 1 of 3 hospitals was performed. Results: Of the 179 patients included in the study, 114 (63.7%) were female, the mean age was 55.1 years (range 20-87). Direct laryngoscopy or videolaryngoscopy was successful in 162 patients (90.5%), with fiberoptic intubation used for the remaining 17 patients. Thirty-one patients (17.4%) required >1 intubation attempt; these patients had larger thyroids (201.3 g, 95% CI 155.3-247.2 g) than those intubated with 1 attempt (144.7 g, 95% CI 127.4-161.9 g, P =.009). Those who required >1 attempt had higher BMI (38.3, 95% CI 34.0-42.6 vs. 32.9, 95% CI 31.5-34.3, P =.02). Mallampati score was found to be a predictor of >1 attempt, though tracheal compression and tracheal shift were not found to be predictors of >1 attempt, nor was the lowest thyroid extent. BMI was the only independent factor on multivariable logistic regression of needing >1 attempt (odds ratio 1.056, 95% CI 1.011-1.103, P =.015). Conclusions: The majority of patients undergoing surgery for substernal goiter can be intubated routinely without the need for fiberoptic intubation. Thyroid-specific factors such as lowest thyroid extent and mass effect of the gland on the trachea do not appear to be associated with difficult intubation, whereas classic patient factors associated with difficulty intubation are. Level of evidence: VI [ABSTRACT FROM AUTHOR]