Development and validation of a risk-prediction nomogram for in-hospital mortality in adults poisoned with drugs and nonpharmaceutical agents: An observational study

Abstract

Acute poisoning with drugs and nonpharmaceutical agents represents an important challenge in the emergency department (ED).The objective is to create and validate a risk-prediction nomogram for use in the ED to predict the risk of in-hospital mortality in adults from acute poisoning with drugs and nonpharmaceutical agents.This was a prospective cohort study involving adults with acute poisoning from drugs and nonpharmaceutical agents admitted to a tertiary referral center for toxicology between January and December 2015 (derivation cohort) and between January and June 2016 (validation cohort). We used a program to generate nomograms based on binary logistic regression predictive models. We included variables that had significant associations with death. Using regression coefficients, we calculated scores for each variable, and estimated the event probability. Model validation was performed using bootstrap to quantify our modeling strategy and using receiver operator characteristic (ROC) analysis. The nomogram was tested on a separate validation cohort using ROC analysis and goodness-of-fit tests.Data from 315 patients aged 18 to 91 years were analyzed (n = 180 in the derivation cohort; n = 135 in the validation cohort). In the final model, the following variables were significantly associated with mortality: age, laboratory test results (lactate, potassium, MB isoenzyme of creatine kinase), electrocardiogram parameters (QTc interval), and echocardiography findings (E wave velocity deceleration time). Sex was also included to use the same model for men and women. The resulting nomogram showed excellent survival/mortality discrimination (area under the curve [AUC] 0.976, 95% confidence interval [CI] 0.954-0.998, P < 0.0001 for the derivation cohort; AUC 0.957, 95% CI 0.892-1, P < 0.0001 for the validation cohort).This nomogram provides more precise, rapid, and simple risk-analysis information for individual patients acutely exposed to drugs and nonpharmaceutical agents, and accurately estimates the probability of in-hospital death, exclusively using the results of objective tests available in the ED.

Figure 1

Patient flow diagram.

Figure 2

Risk-prediction nomogram for mortality in acute poisoning with drugs and nonpharmaceutical agents incorporating age (y), sex (1 male, 0 female), QTc interval (msec), DT (msec), initial CKMB (ng/mL), initial lactate levels (mmol/L), and K⁺ levels (mmol/L). CKMB = MB isoenzyme of creatine kinase, DT = the E wave velocity deceleration time, QTc = corrected QT interval.

Figure 3

Risk-prediction nomogram in a patient with acute poisoning from a toxic alcohol that did not survive. A line is drawn downward from the value of each category to the score line. The points are then added to determine the total score, and a line is drawn upward to find the risk of mortality. Death probability estimation: K⁺ (mmol/L): 6.3 − score = 10; initial lactate (mmol/L): 0.9 − score = 0.2; initial CKMB (ng/mL): 7.94 − score = 0.4; DT (msec): 278 – score = 6.3; QTc (msec): 427.36 – score =3.5; sex: 1 (male) – score = 0; age (y): 59 – score =5.1. Total score = 25.5, with a death probability of 0.68. CKMB = MB isoenzyme of creatine kinase, DT = the E wave velocity deceleration time, QTc = corrected QT interval.

Figure 4

Risk-prediction nomogram in a patient with acute poisoning from toxic alcohol that survived. Death probability estimation: K⁺ (mmol/L): 3.7 − score=5.8; initial lactate (mmol/L): 1.3 − score=0.5; initial CKMB (ng/mL): 7.3 − score=0.4; DT (msec): 251 – score=5.6; QTc (msec): 258.7 – score= 2.1; sex: 1 (male) – score=0; age (y): 68 – score=5.8. The total score = 20.2, with a death probability of 0.004.

Figure 5

Receiver operating characteristic curves validate the discriminatory power of the nomogram predictive for mortality in the validation cohort. Areas under the curves: nomogram probability – 0.949 (95% confidence interval 0.879–1.000, P < 0.001); predicted probability using the developed model–0.957 (95% confidence interval 0.892–1.000, P < 0.001).