Machine Learning-Based Prediction of Digoxin Toxicity in Heart Failure: A Multicenter Retrospective Study.

Digoxin toxicity (plasma digoxin concentration ≥0.9 ng/mL) is associated with worsening heart failure (HF). Decision tree (DT) analysis, a machine learning method, has a flowchart-like model where users can easily predict the risk of adverse drug reactions. The present study aimed to construct a flowchart using DT analysis that can be used by medical staff to predict digoxin toxicity. We conducted a multicenter retrospective study involving 333 adult patients with HF who received oral digoxin treatment. In this study, we employed a chi-squared automatic interaction detection algorithm to construct DT models. The dependent variable was set as the plasma digoxin concentration (≥ 0.9 ng/mL) in the trough during the steady state, and factors with p < 0.2 in the univariate analysis were set as the explanatory variables. Multivariate logistic regression analysis was conducted to validate the DT model. The accuracy and misclassification rates of the model were evaluated. In the DT analysis, patients with creatinine clearance <32 mL/min, daily digoxin dose ≥1.6 µg/kg, and left ventricular ejection fraction ≥50% showed a high incidence of digoxin toxicity (91.8%; 45/49). Multivariate logistic regression analysis revealed that creatinine clearance <32 mL/min and daily digoxin dose ≥1.6 µg/kg were independent risk factors. The accuracy and misclassification rates of the DT model were 88.2 and 46.2 ± 2.7%, respectively. Although the flowchart created in this study needs further validation, it is straightforward and potentially useful for medical staff in determining the initial dose of digoxin in patients with HF.

Effect of polypharmacy on plasma bepridil concentration in patients with heart failure: a multicenter retrospective study.

BACKGROUND: Polypharmacy, defined as the concurrent use of over six drugs, is common in the treatment of heart failure (HF); however, unpredictable drug interactions with bepridil may occur. In this study, we have elucidated the influence of polypharmacy on plasma bepridil concentrations in patients with HF. METHODS: We conducted a multicenter retrospective study involving 359 adult patients with HF who received oral bepridil. Because QT prolongation is an adverse effect following plasma bepridil concentrations ≥800 ng/mL, the risk factors for patients achieving these concentrations at steady state were elucidated via multivariate logistic regression. The correlation between bepridil dose and plasma concentration was examined. The effect of polypharmacy on the value of the concentration-to-dose (C/D) ratio was investigated. RESULTS: A significant relationship was observed between bepridil dose and plasma concentration (p <  0.001), and the intensity of the correlation was moderate (r = 0.503). Based on multivariate logistic regression, the adjusted odds ratios for a daily dose of bepridil ≥1.6 mg/kg, polypharmacy, and concomitant of aprindine, a cytochrome P450 2D6 inhibitor, were 6.82 (95% coefficient interval: 2.104-22.132, p = 0.001), 2.96 (95% coefficient interval: 1.014-8.643, p = 0.047), and 8.63 (95% coefficient interval: 1.684-44.215, p = 0.010), respectively. Despite the moderate correlation in non-polypharmacy, the correlation was not observed in polypharmacy. Therefore, inhibiting metabolism, along with other mechanisms, may contribute to the polypharmacy-induced increase in plasma bepridil concentrations. Moreover, the C/D ratios in the groups receiving 6-9 and 10≤ concomitant drugs were 1.28- and 1.70-fold higher than in those receiving <6 drugs, respectively. CONCLUSIONS: Plasma bepridil concentrations may be influenced by polypharmacy. Moreover, the plasma bepridil concentration increased in correlation with the number of concomitant drugs used. Although the mechanism of this increase could not be determined, plasma bepridil concentrations should be periodically monitored for safe use in patients with HF. TRIAL REGISTRATION: Retrospectively registered.

Developing a clinical prediction rule to identify patients with lumbar disc herniation who demonstrate short-term improvement with mechanical lumbar traction.

OBJECTIVE: To develop a clinical prediction rule (CPR) that predicts treatment responses to mechanical lumbar traction (MLT) among patients with lumbar disc herniation (LDH). METHOD: This study was an uncontrolled prospective cohort study. The subjects included 103 patients diagnosed with LDH for which they underwent conservative therapy. The subjects received MLT for 2 weeks, and the application of any other medication was left at the discretion of the attending physician. The initial evaluation was performed prior to the initiation of treatment. The independent variables from the initial evaluation were imaging diagnosis, Oswestry Disability Index (ODI), Fear-Avoidance Beliefs Questionnaire score, visual analog scale, medical interview, physical examination. The patients whose ODI after 2 weeks of treatment improved by ≥50% of that at the initial evaluation were defined as responders. RESULTS: Of the 103 subjects, 24 were responders, and the five predictors selected for the CPR were limited lumbar extension range of motion, low-level fear-avoidance beliefs regarding work, no segmental hypomobility in the lumbar spine, short duration of symptoms, and sudden onset of symptoms. For the patients with at least three of the five predictors, the probability of their ODI greatly improving increased from 23.3% to 48.7% compared with the patients without these predictors (positive likelihood ratio, 3.13). CONCLUSION: Five factors were selected for the CPR to predict whether patients with LDH would demonstrate short-term improvement following conservative therapy with MLT.