Mortality risk factors for patients with cardiotoxic exposures treated with high-dose insulin: analysis of the National Poison Data System®.

Abstract

Introduction: High-dose insulin/glucose is an inotrope, vasodilator, and standard therapy for beta-adrenoceptor and calcium channel blocker poisoning, yet no large database studies have examined its use. This study sought to describe high-dose insulin use in the United States using the National Poison Data System^®. Determining mortality risk factors was the primary aim.

Methods: We identified all National Poison Data System® cases in which "High dose insulin/glucose" therapy was Recommended or Performed from 2019 to 2021, the first three years National Poison Data System^® allowed specific coding for high-dose insulin. We developed logistic regression models to determine clinical factors associated with death in patients receiving high-dose insulin. We also evaluated methylthioninium chloride (methylene blue) use as a refractory vasoplegia marker.

Results: High-dose insulin was used in 1,856 patients, primarily for exposures to calcium channel blockers (n = 1,116 [60%]) and beta-adrenoceptor blockers (n = 985 [53%]), with the most common drugs being amlodipine (n = 677 [61%]) and metoprolol (n = 371 [38%]). Death occurred in 431 [23%] patients; amlodipine was the most common cardiotoxicant in fatal exposures (n = 202 [47%]). Calcium channel blocker exposure was significantly associated with death compared to beta-adrenoceptor blockers (odds ratio 2.2; 95% CI: 1.6-3.8). Exposure to verapamil, compared to amlodipine or diltiazem, was associated with death (odds ratio 1.7; 95% CI: 1.0-2.7). Increasing age, hyperglycemia, heart block, and concomitant treatment with mechanical ventilation or vasopressors were all associated with death. Methylthioninium chloride was more commonly used in patients with amlodipine exposures (110/677 [16%]) than with verapamil or diltiazem (7/325 [2%]; P <0.001).

Discussion: Among patients treated with high-dose insulin, amlodipine-exposed patients were more commonly treated with methylthioninium chloride, suggesting they experienced more refractory vasoplegia. As high-dose insulin is a vasodilator, more data are needed to better define the role for high-dose insulin in amlodipine poisoning.

Conclusion: In this study of patients treated with high-dose insulin, exposure to calcium channel blockers was more lethal than beta-adrenoceptor blocker poisoning. Amlodipine was the most common cardiotoxicant in patients who lived or died, while verapamil was the most lethal cardiotoxicant.