ABSTRACT
BACKGROUND: After thyroidectomy, patients require Levothyroxine (LT4). It may take years of dose adjustments to achieve euthyroidism. During this time, patients encounter undesirable symptoms associated with hypo- or hyper-thyroidism. Currently, providers adjust LT4 dose by clinical estimation, and no algorithm exists. The objective of this study was to build a decision tree that could estimate LT4 dose adjustments and reduce the time to euthyroidism. METHODS: We performed a retrospective cohort analysis on 320 patients who underwent total or completion thyroidectomy at our institution. All patients required one or more LT4 dose adjustments from their initial postoperative dose before attaining euthyroidism. Using the Classification and Regression Tree algorithm, we built various decision trees from patient characteristics, estimating the dose adjustment to reach euthyroidism. RESULTS: The most accurate decision tree used thyroid-stimulating hormone values at first dose adjustment (mean absolute error = 13.0 µg). In comparison, the expert provider and naïve system had a mean absolute error of 11.7 µg and 17.2 µg, respectively. In the evaluation dataset, the decision tree correctly predicted the dose adjustment within the smallest LT4 dose increment (12.5 µg) 79 of 106 times (75%, confidence interval = 65%-82%). In comparison, expert provider estimation correctly predicted the dose adjustment 76 of 106 times (72%, confidence interval = 62%-80%). CONCLUSIONS: A decision tree predicts the correct LT4 dose adjustment with an accuracy exceeding that of a completely naïve system and comparable to that of an expert provider. It can assist providers inexperienced with LT4 dose adjustment.
Subject(s)
Decision Trees , Drug Dosage Calculations , Hormone Replacement Therapy/methods , Thyroidectomy/adverse effects , Thyroxine/administration & dosage , Adult , Aged , Female , Humans , Hyperthyroxinemia/blood , Hyperthyroxinemia/etiology , Hyperthyroxinemia/prevention & control , Hypothyroidism/blood , Hypothyroidism/drug therapy , Hypothyroidism/etiology , Machine Learning , Male , Middle Aged , Postoperative Care/methods , Retrospective Studies , Thyrotropin/blood , Thyroxine/adverse effectsABSTRACT
BACKGROUND: Levothyroxine (LT4) is one of the most prescribed drugs in the United States; however, many patients started on LT4 after thyroidectomy suffer from symptoms of hyper- or hypo-thyroidism before achieving euthyroidism. This study aims to describe the time required for dose adjustment before achieving euthyroidism and identify predictors of prolonged dose adjustment (PDA+) after thyroidectomy. METHODS: This is a single institution retrospective cohort study of patients who achieved euthyroidism with LT4 therapy between 2008 and 2017 after total or completion thyroidectomy for benign disease. Patients who needed at least three dose adjustments (top quartile) were considered PDA+. Binomial logistic regression was used to identify predictors of PDA+. RESULTS: The 605 patients in this study achieved euthyroidism in a median of 116 d (standard deviation 124.9) and one dose adjustment (standard deviation 1.3). The 508 PDA- patients achieved euthyroidism in a median of 101 d and one dose adjustment. The 97 PDA+ patients achieved euthyroidism in a median of 271 d and three dose adjustments. Iron supplementation (odds ratio = 4.4, 95% confidence interval = 1.4-13.5, P = 0.010) and multivitamin with mineral supplementation (odds ratio = 2.4, 95% confidence interval = 1.3-4.3, P = 0.004) were independently associated with PDA+. Age, gender, preoperative thyroid disease, and comorbidities did not independently predict PDA+. CONCLUSIONS: After thyroidectomy, achieving euthyroidism can take nearly 4 mo. Iron and mineral supplementation are associated with PDA+. This information can inform the preoperative counseling of patients and suggests that this may expedite achieving euthyroidism.
Subject(s)
Hormone Replacement Therapy/methods , Hyperthyroidism/chemically induced , Hypothyroidism/drug therapy , Thyroidectomy/adverse effects , Thyroxine/administration & dosage , Adult , Aged , Dietary Supplements/adverse effects , Dose-Response Relationship, Drug , Female , Hormone Replacement Therapy/adverse effects , Hormone Replacement Therapy/statistics & numerical data , Humans , Hyperthyroidism/blood , Hypothyroidism/blood , Hypothyroidism/etiology , Male , Middle Aged , Prospective Studies , Retrospective Studies , Thyroxine/adverse effects , Thyroxine/blood , Time FactorsABSTRACT
BACKGROUND: Patients often struggle to attain euthyroidism after thyroidectomy, and multiple dosing schemes have been proposed to supplant the standard weight-based approach for initial levothyroxine dosing after thyroidectomy. The objectives of this study were to review the literature for existing levothyroxine dosing schemes and compare estimation accuracies with novel schemes developed with machine learning. METHODS: This study retrospectively analyzed 598 patients who attained euthyroidism after total or completion thyroidectomy for benign disease. A scoping review identified existing levothyroxine dosing schemes. Thirteen machine learning algorithms estimated euthyroid dose. Using 10-fold cross-validation, we compared schemes by the proportion of patients having a predicted dose within 12.5 µg/day of their euthyroid dose. RESULTS: Of 264 reviewed articles, 7 articles proposed retrospectively implementable dosing schemes. A novel Poisson regression model proved most accurate, correctly predicting 64.8% of doses. Incorporating 7 variables, Poisson regression was significantly more accurate than the best scheme in the literature (body mass index/weight based) that correctly predicted 60.9% of doses (Pâ¯=â¯.031). Standard weight-based dosing (1.6 µg/kg/day) correctly predicted 51.3% of doses, and the least effective scheme (age/sex/weight based) correctly predicted 27.4% of doses. CONCLUSION: Using readily available variables, a novel Poisson regression dosing scheme outperforms other machine learning algorithms and all existing schemes in estimating levothyroxine dose.
