Cardiac Arrest Following the Administration of Intravenous Diphenhydramine for Sedation to an Infant With Congenital Heart Disease

Abstract

Diphenhydramine (Benadryl) is a first-generation antihistamine that is used primarily to treat allergic reactions including anaphylaxis, urticaria, and allergic rhinitis. Despite its availability as an over-the-counter medication, toxicity may occur with its use especially when administered in large doses or via the intravenous route. We present a 3-month-old infant with Trisomy 21 who suffered a cardiac arrest immediately following administration of a single 1.25 mg/kg dose of intravenous diphenhydramine, prescribed for sedation in the Pediatric ICU setting. The potential cardiovascular and respiratory effects of diphenhydramine are presented, previous reports of life-threatening adverse effects reviewed, and options to limit these effects discussed.

Introduction

Diphenhydramine is a first-generation antihistamine that works by competitively blocking the binding of histamine to H₁ receptors in smooth muscle, endothelium, and the brain.¹ It was first synthesized in 1943 by George Rieveschl at the University of Cincinnati, while investigating medications for the treatment of muscle spasms. In 1946, it became the first prescription anti-histamine approved by the US Food and Drug Administration (FDA). During the 1980s, the FDA approved diphenhydramine as an over-the-counter medication.

As a competitive antagonist of histamine at the H₁ receptor, diphenhydramine reduces the end-organ effects of histamine including allergic symptoms. In addition to acting as a histamine antagonist, diphenhydramine also provides antimuscarinic, antiemetic, local anesthetic, and sedative effects. Clinically, diphenhydramine is primarily used for allergic reactions including anaphylaxis, urticaria, and allergic rhinitis.^2,3 Other applications include the treatment of common cold symptoms, motion sickness, nausea, and insomnia.³ Despite its availability as an over-the-counter medication, toxicity may occur related to its use. We present the case of an infant with trisomy 21 who suffered a cardiac arrest immediately following the intravenous administration of a single dose of diphenhydramine, which was prescribed to produce sedation in the PICU setting. The potential cardiovascular effects of diphenhydramine are presented, previous reports of life-threatening adverse effects reviewed, and options to limit these effects discussed.

Case Report

The patient was a 3-month-old female (4 kg) with trisomy 21, an unrepaired atrioventricular septal defect, and chronic respiratory insufficiency due to tracheomalacia. Given her multiple medical issues, she had spent much of her life in the hospital, requiring care for cardiac, respiratory, and airway concerns. At 4 weeks of age, she presented to the ED and was hospitalized for 36 days while being treated for a coronavirus infection and respiratory failure. She was then discharged home on the following medications: chlorothiazide (0.7 mL orally, twice a day) and furosemide (10 mg/mL orally, twice a day, for 180 days) A week after discharge she presented to the ED again with lethargy and breathing difficulties. The parents reported that her siblings had upper respiratory symptoms and that the patient had been having increased work of breathing with accessory muscle use for 3 days at home. She required endotracheal intubation and mechanical ventilation and was admitted to the cardiothoracic ICU with acute respiratory failure secondary to rhinovirus bronchiolitis.

She required endotracheal intubation for 11 days and was then extubated to non-invasive ventilator support with bilevel positive airway pressure. She subsequently required a second course of endotracheal intubation for 5 to 6 days. Chronic medications included furosemide and hydrochlorothiazide to avoid fluid overload. On day 41 she underwent pulmonary artery banding to prevent overcirculation of the pulmonary bed related to the atrioventricular septal defect. The procedure was well tolerated, and she returned to the cardiothoracic ICU for postoperative care. Following surgery, her trachea was extubated, and she was transitioned to non-invasive support with bilevel positive airway pressure and shorter periods of high flow nasal cannula. During the postoperative period, the patient was not receiving any vasoactive medications. A postoperative electrocardiogram did not reveal arrhythmias or conduction delays (i.e., QT and QTc of 248 ms and 394 ms, respectively). Analgesia was provided by nurse-controlled analgesia with fentanyl and intermittent intravenous doses of midazolam.

The morning after surgery on postoperative day 1, she remained irritable and restless with nasal flaring and grunting despite a dose of ketorolac (0.5 mg/kg), 2 to 3 doses of fentanyl (total of 2 mcg/kg), and 2 additional IV doses of midazolam (0.05 mg/kg and 0.1 mg/kg). During a period of agitation, her heart rate increased to 180 to 200 beats/min with a blood pressure of 120/80 mm Hg. Diphenhydramine was prescribed to provide sedation and treat pruritus thought to be related to the fentanyl infusion. The patient was administered 5 mg (1.25 mg/kg) of a commercially available product intravenously over 1 minute to the best of our knowledge, since it was given as an IV bolus. The diphenhydramine was given via a peripherally placed catheter. Within 1 to 2 minutes of the diphenhydramine administration, acute cardiac decompensation occurred. The patient became apneic and unresponsive. The heart rate rapidly decreased from 180 to 200 beats/min to 60 to 80 beats/min with widening of the QRS complex that progressed to asystole. Standard resuscitation protocols were initiated with endotracheal intubation, chest compressions, and the administration of epinephrine. Cardiopulmonary resuscitation was required for 19 minutes with multiple doses of epinephrine to achieve return of spontaneous circulation. No other medications were administered during the resuscitation. During resuscitation, cerebral oxygen saturations were 50% to 60%, arterial line blood pressure was 50 to 60 mm Hg, and the end-tidal carbon dioxide was >10 mm Hg. She had no prior history of cardiac events and had never received diphenhydramine previously.

Within minutes of the resuscitation, the infant's neurologic status stabilized. After resuscitation, IV midazolam was used to provide sedation. Subsequent central nervous system imaging showed no evidence of hypoxic-ischemic damage. On day 138 of hospitalization, she underwent surgery for complete repair of her atrioventricular septal defect. Although she had a protracted hospital course related to her respiratory status and severe tracheomalacia, she was eventually discharged home and has done well without further cardiorespiratory events.

Discussion

The clinical uses of diphenhydramine relate primarily to the treatment of symptoms related to allergic phenomena including urticaria, rhinorrhea, or pruritus.³ Although not a first-line medication, diphenhydramine and other first-generation antihistamines have also found anecdotal and sporadic use to produce sedation for patients in the pediatric critical care setting.^4,5 Despite its relatively wide therapeutic window, adverse effects may occur with the therapeutic use of diphenhydramine in the recommended dosages. While not labeled for routine use in children younger than 2 years, it is indicated in the prevention or treatment of pruritus or allergic reactions when under the care of a health care professional. In this case, 5 mg/kg/24 hr is recommended in infants and children.

The potential cardiovascular effects that have been reported in both the adult and pediatric populations include anticholinergic effects with tachycardia, bundle branch block, atrioventricular dissociation, atrioventricular block, widening of the QRS complex, and arrhythmias related to alterations in repolarization and prolongation of the QT interval.^6–9 These have been reported following acute and chronic use of both IV and oral dosage formulations and following overdose as well as normal dosing.

Diphenhydramine can result in QRS widening due to blockade of fast sodium channels and prolongation of phase 1 of the cardiac action potential, thereby increasing the duration of depolarization. QT interval prolongation can also occur especially with larger doses of diphenhydramine, which can result in lethal ventricular arrhythmias including torsade de pointes. Prolongation of the QT interval results from the blockade of potassium channels with lengthening of phase 3 repolarization of the cardiac action potential.^7–9 Cardiac conduction effects have been noted with other antihistamines. Terfenadine, a second-generation, nonsedating antihistamine drug, was removed from the United States market owing to prolongation of the QT interval. The impact of these effects on cardiac conduction and arrhythmogenesis has been highlighted by previous anecdotal reports of adverse cardiovascular events following the ingestion or administration of toxic doses of diphenhydramine (Table).^7–21 These reports have involved potentially high blood concentrations related to excessive dosing. However, we were unable to identify any reports of these cardiac effects following the administration of a single, age-appropriate (mg/kg) dose, particularly in the pediatric population.

Our investigation did not reveal any deviation from these guidelines that would indicate that a high plasma concentration was the cause of the event. Given the acuity of the event and the need for resuscitation, blood was not obtained to assay plasma concentrations. A review of the patient's medication history revealed that the only medications administered perioperatively included the sedatives and analgesics listed in the case report, diuretics (i.e., furosemide and hydrochlorothiazide) to control volume overload, and vancomycin for prophylaxis of surgical site infections prior to the surgical procedure.

The Naranjo algorithm (Adverse Drug Reaction Probability Scale was completed with a score of 6, indicating a probable adverse drug interaction (Supplemental Table S).²² Although the administration of diphenhydramine was temporally associated with widening of the QRS and cardiac arrest in our patient, we cannot definitely prove a causal relationship, especially in the setting of a patient recovering from surgery for congenital heart disease, where arrhythmias may be common. The administration time and dose of diphenhydramine were within our institutional guidelines. Proof of a causal relationship between a specific medication and postoperative events are difficult given the likelihood that such events may be multifactorial. Previous reports linking diphenhydramine to arrhythmias have generally involved excessive or large doses. Our intent in publishing this case report is to bring attention to the anecdotal association of diphenhydramine and a potentially lethal ventricular arrhythmia. Future work is required to definitively identify a causal relationship and further refine criteria for use of this medication in this clinical scenario.

ABBREVIATIONS

ECG

electrocardiogram

ED

emergency department

FDA

US Food and Drug Administration

ICU

intensive care unit

IV

intravenous

PICU

pediatric intensive care unit

QRS

ECG indication of ventricular contraction from the Q, R, and S waves

QT

ECG interval from the QRS complex to the end of the T wave

QTc

corrected QT