Efficacy of machine learning to identify clinical factors influencing levothyroxine dosage after total thyroidectomy.

BACKGROUND: We employed Machine Learning (ML) to evaluate potential additional clinical factors influencing replacement dosage requirements of levothyroxine. METHOD: This was a retrospective study of patients who underwent total or completion thyroidectomy with benign pathology. Patients who achieved an euthyroid state were included in three different ML models. RESULTS: Of the 487 patients included, mean age was 54.1 ± 14.1 years, 86.0% were females, 39.0% were White, 53.0% Black, 2.7% Hispanic, 1.4% Asian, and 3.9% Other. The Extreme Gradient Boosting (XGBoost) model achieved the highest accuracy at 61.0% in predicting adequate dosage compared to 47.0% based on 1.6 mcg/kg/day (p < 0.05). The Poisson regression indicated non-Caucasian race (p < 0.05), routine alcohol use (estimate = 0.03, p = 0.02), and osteoarthritis (estimate = -0.10, p < 0.001) in addition to known factors such as age (estimate = -0.003, p < 0.001), sex (female, estimate = -0.06, p < 0.001), and weight (estimate = 0.01, p < 0.001) were associated with the dosing of levothyroxine. CONCLUSIONS: Along with weight, sex, age, and BMI, ML algorithms indicated that race, ethnicity, lifestyle and comorbidity factors also may impact levothyroxine dosing in post-thyroidectomy patients with benign conditions.

Clinical Factors Predictive of Lymph Node Metastasis in Thyroid Cancer Patients: A Multivariate Analysis.

BACKGROUND: Early-stage thyroid cancers have excellent survival. However, lymph node metastases (LNM) confer a worse prognosis and are not always known preoperatively. Therefore, investigation on the clinical and histological factors predictive of LNM in thyroid cancers was conducted to tailor the extent of surgery and radioactive iodine therapy. STUDY DESIGN: Multivariate logistic regressions were performed based on retrospective data from thyroid cancer patients seen between 2013 and 2020 at a single institution. RESULTS: Among 913 patients, mean age was 49.4 years, 76.5% were female, 58.3% were White, 21.2% were Black, and 27.9% had LNM. In the multivariate analyses in which the outcome was LNM, White (odds ratio [OR] 1.74, 95% CI 0.98 to 3.15, p = 0.064) and Hispanic patients (OR 2.36, 95% CI 0.97 to 5.77, p = 0.059) trended toward higher risk of LNM compared to Black patients, whereas age (OR 0.98, 95% CI 0.97 to 1.00, p = 0.008) showed protective effect. Tumor size (OR 1.04, 95% CI 1.01 to 1.07, p = 0.007), extrathyroidal extension (OR 2.46, 95% CI 1.53 to 3.97, p < 0.001), lymphovascular invasion (OR 6.30, 95% CI 3.68 to 11.14, p < 0.001), and multifocality (OR 1.47, 95% CI 1.01 to 2.12, p = 0.042) were associated with higher risk of LNM. In another model with outcome as >5 LNM, tumor size (OR 1.07, 95% CI 1.03 to 1.11, p = 0.001), age (OR 0.95, 95% CI 0.93 to 0.97, p < 0.001), extrathyroidal extension (OR 3.20, 95% CI 1.83 to 5.61, p < 0.001), and lymphovascular invasion (OR 6.82, 95% CI 3.87 to 12.17, p < 0.001) remained significant predictors. CONCLUSION: Our analyses demonstrated and confirmed that age, tumor size, extrathyroidal extension, and lymphovascular invasion are independent predictors of significant LNM, thereby conferring higher risk of recurrence. Risk of LNM based on these patient characteristics should be considered when planning an operative approach.

Comparing the rate and extent of malignancy in surgically excised thyroid nodules across race and ethnicity.

BACKGROUND: Few studies have compared the features of thyroid cancer among races and ethnicities. We hypothesized that race and ethnicity may influence the frequency and features of thyroid malignancy in thyroid nodules. METHOD: This was a retrospective chart review of patients between 2013 and 2020 who underwent thyroidectomy. RESULTS: In the analysis of 2737 patients, thyroid cancer was less prevalent among Blacks (24.0% vs Whites 52.1%, Hispanics 58.7%, Asians 71.7%, and Others 57.9%, p < 0.001). Thyroid cancer in Blacks was less likely to have extrathyroidal extension (9.7% vs Whites 18.6%, Hispanics 25.8%, Asians 18.2%, and Others 17.8%, p = 0.01), overall nodal involvement (12.4% vs Whites 31.1%, Hispanics 37.5%, Asians 36.3%, and Others 30.1%, p < 0.01), and lateral neck metastasis (4.4% vs Whites 10.8%, Hispanics 6.3%, Asians 13.2%, and Others 9.6%, p = 0.02). CONCLUSIONS: Race and ethnicity may play important roles in the risk of malignancy as well as in the extent of thyroid cancer.