TOXICOLOGY/ORIGINAL RESEARCH
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity: A Randomized, Controlled Pilot Study in a Swine Model Shawn M. Varney, MD*; Vikhyat S. Bebarta MD; Susan M. Boudreau, RN, BSN; Toni E. Vargas, PA-C; Maria Castaneda, MS; Lee A. Zarzabal, MS *Corresponding Author. E-mail: [email protected].
Study objective: Diphenhydramine is a moderately lipophilic antihistamine with sodium channel blockade properties. It is consumed recreationally for mild hallucinogenic and hypnotic effects and causes dysrhythmias, seizures, and death with overdose. Intravenous lipid emulsion is a novel agent used to treat lipophilic drug overdose. Two case reports describe clinical improvement with intravenous lipid emulsion after diphenhydramine toxicity, but no prospective studies have been reported. Our objective is to determine whether intravenous lipid emulsion improved hypotension compared with sodium bicarbonate for severe diphenhydramine toxicity in a model of critically ill swine. Methods: Twenty-four swine weighing 45 to 55 kg were infused with diphenhydramine at 1 mg/kg per minute until the mean arterial pressure reached 60% of baseline. Subjects were randomized to receive intravenous lipid emulsion (bolus of 7 mL/kg and then 0.25 mL/kg per minute) or sodium bicarbonate (2 mEq/kg plus an equal volume of normal saline solution). We measured pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, QRS interval, and serum diphenhydramine level. Twelve animals per group provided a power of 0.8 and a of .05 to detect a 50% difference in mean arterial pressure. We assessed differences between groups with a repeated-measures linear model (MIXED) and Kaplan-Meier estimation methods. We compared systolic blood pressure, mean arterial pressure, and cardiac output with repeated measures ANOVA. Results: Baseline weight, hemodynamic parameters, QRS interval, time to hypotension, and diphenhydramine dose required to achieve hypotension were similar between groups. After hypotension was reached, there was no overall difference between intravenous lipid emulsion and sodium bicarbonate groups for cardiac output or QRS intervals; however, there were transient differences in mean arterial pressure and systolic blood pressure, favoring intravenous lipid emulsion (difference: mean arterial pressure, sodium bicarbonate versus intravenous lipid emulsion –20.7 [95% confidence interval –31.6 to –9.8]; systolic blood pressure, sodium bicarbonate versus intravenous lipid emulsion –24.8 [95% confidence interval –37.6 to –12.1]). Time to death was similar. One intravenous lipid emulsion and 2 sodium bicarbonate pigs survived. End-of-study mean total serum diphenhydramine levels were similar. The mean lipid layer diphenhydramine level was 6.8 mg/mL (SD 3.1 mg/mL) and mean aqueous layer level 8.6 mg/mL (SD 5.5 mg/mL). Conclusion: In our study of diphenhydramine-induced hypotensive swine, we found no difference in hypotension, QRS widening, or diphenhydramine levels in aqueous layers between intravenous lipid emulsion and sodium bicarbonate. [Ann Emerg Med. 2015;-:1-10.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2015 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2015.05.028
INTRODUCTION Background Diphenhydramine is a moderately lipophilic (octanol:water partition coefficient 3.40) first-generation antihistamine with sodium channel blockade properties similar to those of tricyclic antidepressants.1 It is used recreationally for its mild hallucinogenic and hypnotic effects, and its abuse has recently increased.2,3 Patients who overdose typically present with an anticholinergic toxidrome. More important, in severe cases, seizures, ventricular dysrhythmias, and death occur.4-7 Severe Volume
-,
no.
-
:
-
2015
intoxication with QRS widening on the ECG is usually treated with sodium bicarbonate or hypertonic saline solution. Intravenous lipid emulsion, or lipid therapy, is a medication with established efficacy for severe local anesthetic toxicity and has been used to treat intoxication by other lipophilic agents.8 Some studies in small, lean animals have reported that intravenous lipid emulsion improves hypotension and prolongs survival compared with standard therapy for drug intoxication, whereas other studies show no effect.9-11 To our knowledge, thus far only Annals of Emergency Medicine 1
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Editor’s Capsule Summary
What is already known on this topic Intravenous lipid emulsion is a novel agent used to treat lipophilic drug overdose. Two case reports describe clinical improvement with intravenous lipid emulsion after diphenhydramine toxicity, but no prospective studies have been reported. What question this study addressed Does intravenous lipid emulsion produce improved effects on hypotension compared with sodium bicarbonate for severe diphenhydramine toxicity in a model of critically ill swine? What this study adds to our knowledge Several relevant endpoints (pulse rate, systolic and mean arterial blood pressure, cardiac output, QRS interval, serum diphenhydramine level, and mortality) showed no difference between the intravenous lipid emulsion and sodium bicarbonate groups. How this is relevant to clinical practice Intravenous lipid emulsion is an emerging therapy frequently used for cardiovascular toxicity. These animal data indicate that intravenous lipid emulsion is unlikely to have a meaningful effect on diphenhydramine cardiovascular toxicity.
case reports have been published on its use in humans, and often multiple therapies were used in each case. In 2 cases, intravenous lipid emulsion was used for diphenhydramine toxicity. In both cases, multiple therapies had been administered and the intravenous lipid emulsion was administered late in the resuscitation; therefore, the effects of each intervention could not be determined.4,12 To our knowledge, there are no preclinical or clinical prospective studies evaluating intravenous lipid emulsion as a treatment for diphenhydramine toxicity. According to the heterogeneity of overdose cases and infrequent presentations in single centers, a prospective clinical trial is not feasible. Therefore, a randomized, controlled study in a model of diphenhydramine-induced cardiotoxicity in critically ill, large animals may provide evidence on the effectiveness of intravenous lipid emulsion. Goal of This Investigation Our objective was to determine whether intravenous lipid emulsion was more effective in improving hypotension 2 Annals of Emergency Medicine
compared with sodium bicarbonate for treatment of diphenhydramine overdose in a model of critically ill swine. MATERIALS AND METHODS Study Design and Setting This investigation was a nonblinded, randomized, parallelgroup-design study approved by the Institutional Animal Care and Use Committee of the Wilford Hall Clinical Research Division. A flowchart of this experimental design is shown in Figure 1. All animal experiments complied with the regulations and guidelines of the Animal Welfare Act and the American Association for Accreditation of Laboratory Animal Care. Animals were housed and the study was conducted in the Animal Care Facility. Selection of Participants Before the experiment, the animals were randomly assigned with an online randomization tool (http://www. randomization.com) to one of 2 intervention groups: sodium bicarbonate, the standard drug for sodium channel blockade; or intravenous lipid emulsion, the study drug. Our methods were similar to those of an intravenous lipid emulsion study for amitriptyline cardiotoxicity.13 Twenty-four healthy female Yorkshire swine (Sus scrofa) weighing 45 to 55 kg were sedated with intramuscular ketamine 10 mg/kg and anesthetized with isoflurane through nose cone. After intubation, they were maintained on isoflurane (1% to 3%) and oxygen (FiO2 of 0.4 to 0.45). Temperature was maintained for all animals between 36.8 C (98.2 F) and 38 C (100.4 F) with heating adjuncts, including a warmed induction room and operating room, warm intravenous fluids at all times, a bed warmer during the procedure, and a warming blanket. Isoflurane was titrated to 1% to 2% to mitigate isoflurane-induced hypotension, and arterial pCO2 was maintained between 38 and 42 mm Hg. Interventions Instrumentation included placement of an endotracheal tube for ventilation. Invasive hemodynamic variables were measured with 8-French Swan-Ganz continuous cardiac output, mixed SvO2 monitoring catheters (model 746HF8) and the Edwards Vigilance II monitor (Edwards Lifesciences, Irvine, CA). Measurements included continuous cardiac output, systemic vascular resistance, mixed SvO2, central venous pressure, pulmonary artery pressure, and core temperature. The catheter ports were flushed with saline solution, and the catheter was placed through cut-down in the right internal jugular vein. Aortic pressure was measured continuously through the femoral artery with a 2-French Millar catheter (Millar Instruments, Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 1. Study timeline with interventions. Time (T) points “zT-60 min” and “zT-30 min” are only estimates and differ for each animal. The study times of interest began when the animals reached hypotension (time 0 minutes) and received treatment. T, Hypotension time. MAP, Mean arterial pressure.
Houston, TX). An 8.5-French introducer (Arrow, Reading, PA) was placed in the carotid artery for laboratory sampling, and another was placed in the femoral vein for medication administration. All animals received a bolus of 15 mL/kg of warmed normal saline solution intravenously during instrumentation. Baseline values were obtained for serum electrolytes, CBC count, and diphenhydramine levels, as well as arterial blood gases for pH, pCO2, bicarbonate, and lactate. Venous samples were obtained at hypotension and at subsequent 15-minute intervals to measure changes in serum electrolyte and lactate levels. Additionally, a baseline ECG was performed to document QRS- and QTc-interval measurements. We used several animals to refine the model, including control animals that received intravenous lipid emulsion alone without adverse effects. For our experiment, diphenhydramine (Amerisource Bergen, Chesterbrook, PA) at 5 mg/mL was infused intravenously at 1 mg/kg per minute through a central venous catheter until hypotension was reached. We defined hypotension as a decrease in mean arterial pressure to 60% of baseline value. A normal saline solution bolus of 10 mL/kg was then administered to the animals. Before the experiment, the swine were randomized and assigned (12 to each group) to either the experimental group (intravenous lipid emulsion bolus of 7 mL/kg, followed by an infusion of 0.25 mL/kg per minute until the end of the study) or to the standard treatment group (8.4% sodium bicarbonate bolus of 2 mEq/kg, plus an equivalent volume of normal saline solution). We determined these doses of intravenous lipid emulsion and sodium bicarbonate according to previously published studies for sodium channel blockade toxicity in rabbits, rats, and our protocol Volume
-,
no.
-
:
-
2015
development animals because no studies in large animals had been reported with diphenhydramine toxicity, to our knowledge.9-11 A control arm using normal saline solution alone was not used in this study because of Institutional Animal Care and Use Committee recommendations. The standard therapy arm (sodium bicarbonate administration) served as the control because critically ill patients presenting to the emergency department with diphenhydramine toxicity would likely be treated with sodium bicarbonate as a standard management approach. The Institutional Animal Care and Use Committee granted permission to administer diphenhydramine to 5 pigs until hypotension was reached and then to resuscitate using only normal saline solution to study the effect. Despite 2 boluses of intravenous fluids at 7 mL/kg (representing the intravenous lipid emulsion bolus) and additional boluses (10 mL/kg), all 5 swine died. We used 15 protocol development animals to determine the toxic doses of diphenhydramine and effective dose of intravenous lipid emulsion, as well as to test the effects of intravenous lipid emulsion on the animal model. To determine the appropriate diphenhydramine dose to cause hypotension, we used previous animal studies.9 With an infusion of diphenhydramine at 1 mg/kg per minute, animals achieved hypotension and subsequent death. Less toxic regimens (0.25 to 0.5 mg/kg per minute) resulted in long periods to reach hypotension. To identify an effective intravenous lipid emulsion dose, we initially used the dose recommended for humans (bolus of 1.5 mL/kg, followed by infusion at 0.25 mL/kg per minute) and increased stepwise to 7 mL/kg when no response in blood pressure was observed.8 Although studies have described higher doses in rabbit and rat models (up to Annals of Emergency Medicine 3
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
12 and 15 mL/kg, respectively), even at 7 mL/kg the animals’ blood was milky white. To determine the effect of intravenous lipid emulsion alone on the swine model, we administered 7 mL/kg to animals that had received no diphenhydramine. We noted no skin flush or alteration in hemodynamic or respiratory parameters. The animals survived without consequence. Necropsy was performed and showed no pulmonary emboli or other pulmonary toxicity. We have published use of the intravenous lipid emulsion dose and effects of intravenous lipid emulsion in control animals in a previous study.13
Methods of Measurement and Outcome Measures Hemodynamic parameters (pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, systemic vascular resistance, SVO2, and central venous pressure) were measured continuously throughout the study and 60 minutes after standard or study drug administration. Laboratory data, ECG (Zoll M Series CCT; ZOLL Medical Corporation, Chelmsford, MA), and serum diphenhydramine levels (measured by high-performance liquid chromatography and fluorescent detection) were recorded every 15 minutes from study initiation to study end. Diphenhydramine levels were determined in the total serum sample, fatty layer, and aqueous layer. A simplified summary of the methodology for determining diphenhydramine levels in total serum, the lipid-rich fraction, and the lipid-poor fraction consisted of 4 steps, described in detail below: (1) preparation of the lipid-rich and lipid-poor fractions; (2) extraction of diphenhydramine; (3) high-pressure liquid chromatography analysis of diphenhydramine with fluorescent detection; and (4) specificity, linearity, reproducibility (precision), and accuracy assessments. Lipid-rich and lipid-poor fractions were prepared by centrifuging 1-mL samples of each pig’s serum at 14,500 revolutions per minute for 20 minutes in a walk-in cold room at 2 C (35.6 F) to 8 C (46.4 F). After the lipidpoor layer was removed with a Pasteur pipet, 1 mL of water was added to the remaining fatty layer, vortex mixed for 10 to 20 seconds, and then transferred with a Pasteur pipet to a separate tube. Serum diphenhydramine calibration curves were prepared at 0, 6.25, 12.5, 25.0, 50.0, and 100.0 mg/mL. Sodium hydroxide and an extraction solvent were added to each calibrator and test sample and shaken. Samples were then centrifuged at 3,000 revolutions per minute for 20 minutes at 5 C (41 F) (SD 3 C [37.4 F]). The top organic layer was transferred to separate tubes and evaporated in a TurboVap LV evaporator (Biotage, Uppsala, Sweden) at 50 C (122 F) 4 Annals of Emergency Medicine
(SD 3 C [37.4 F]) under a gentle stream of nitrogen. The samples were then reconstituted in 1.0 mL of the highpressure liquid chromatography mobile phase and transferred to high-pressure liquid chromatography injection vials. High-pressure liquid chromatography analysis was performed on a Waters Alliance high-pressure liquid chromatography machine (Waters 2695 Separations Module; Waters, Milford, MA) equipped with a Phenomenex Luna C18 chromatographic column (Phenomenex, Torrance, CA) and a Waters 474 Fluorescent detector. The highpressure liquid chromatography mobile phase consisted of 2.5 mM ammonium acetate in methanol-water, 50:50, volume/volume, adjusted to pH 4.0. The fluorescence detector excitation and emission wavelengths were 230 and 560 nm, respectively. Injection volumes were 50.0 mL and chromatographic flow rates were 0.2 mL/min. The method was specific in that no interferences were detected in serum samples from 6 pigs not receiving diphenhydramine and in any of the predose samples. The serum diphenhydramine standards curves were linear from 6.25 to 100.0 mg/mL (R2¼0.9987; n¼4). Recoveries of serum diphenhydramine over the concentration range of 6.25 to 100.0 mg/mL were 97.8%. Within-run precision of serum diphenhydramine at 12.5, 25.0, and 50.0 mg/mL (n¼10) was 2.62%, 1.28%, and 0.8%, respectively. Between-day precision of the 12.5, 25.0, and 50.0 mg/mL calibrators was 14.2%, 6.78%, and 2.94% (n¼10), respectively. Accuracies of the 6.25, 12.5, 25.0, 50.0, and 100.0 mg/mL calibrators were 100.5%, 97.8%, 100.3%, 106.4%, and 98.1%, respectively. At the end of the study, any surviving animals were euthanized with intravenous administration of sodium pentobarbital 100 mg/kg. We defined death as asystole when mean arterial pressure was less than 20 mm Hg because some animals may have had a rhythm and no measurable blood pressure, or only a small pulse pressure.
Primary Data Analysis The study’s primary outcome was an improvement in mean arterial pressure of greater than 50% of trough values after administration of the study drug (sodium bicarbonate or intravenous lipid emulsion). Baseline hemodynamic parameters were assessed for treatment group differences with a parametric t test. A secondary outcome was posthypotension reversal of the QRS-interval widening on ECG. The preliminary assessment of the primary outcome was assessed with a repeated-measures linear model with adjustment for treatment group, time, and the treatmentby-time interaction, with a first-order autoregressive covariance structure assumed.14 The RM ANOVA also Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
appropriately managed missing data. In accordance with the preliminary assessment of a significant treatment-bytime interaction, a post hoc analysis was conducted to evaluate treatment differences during the follow-up period up to the point of 15 minutes posthypotension, at which time only 1 subject remained in the intravenous lipid emulsion group. Post hoc comparisons were adjusted for multiple testing with the Sidak correction method, and after adjustment, no significant differences between treatment groups were detected.15 Secondary analysis was performed to assess the relationship between survival time from the point of hypotension to treatment, using the Kaplan-Meier method and log-rank test. It was determined that equal sample sizes of 12 animals per group would achieve a power of 0.8 to detect a 50% difference in mean arterial pressure with a 2-sided test at significance level of .05 (NCSS PASS 2002, Kaysville, UT). Differences in response (eg, improvement in diphenhydramine-induced hypotension) between the 2 treatments were assessed with repeated-measures ANOVA adjusted for treatment, time, and the interaction of treatment by time. We compared mean values at time zero between groups for pulse rate, mean arterial pressure, cardiac output, serum lactate levels, QRS interval, diphenhydramine dose administered, and serum diphenhydramine levels. All statistical testing was 2-sided with a significance level of 5% and was performed with SAS (version 9.3; SAS Institute, Inc., Cary, NC). All graphic presentations were created with R (version 3.0.2). RESULTS Characteristics of Study Subjects Twenty-four swine were randomly assigned to intravenous lipid emulsion and sodium bicarbonate groups (12 in each arm). At baseline, the groups had similar weights (intravenous lipid emulsion 48.4 kg [SD 3.3]; sodium bicarbonate 49.4 kg [SD 3]). In addition, there was no important difference for each group in regard to mean pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, SvO2, and QRS intervals (Table 1). Main Results Intravenous lipid emulsion and sodium bicarbonate groups required similar mean amounts of diphenhydramine to reach hypotension (mean intravenous lipid emulsion 31.2 mg/kg [SD 8.3]; mean sodium bicarbonate 33.3 mg/ kg [SD 8.3]). The total mean amounts of diphenhydramine infused were 1,510 mg in the intravenous lipid emulsion group and 1,645 mg in the sodium bicarbonate group. The intravenous lipid emulsion group reached hypotension at a Volume
-,
no.
-
:
-
2015
mean 32 minutes 13 seconds [SD 9:41]; the sodium bicarbonate group, 34 minutes 8 seconds [SD 8:36]. The time required to reach hypotension varied for each animal and is shown in Figure 1 as “zT-60 min,” or approximately 60 minutes before hypotension. The time when hypotension was achieved and standard or study drug was administered was time 0 and the starting point for observation of study drug effect. There was no overall difference between intravenous lipid emulsion and sodium bicarbonate groups for mean arterial pressure, pulse rate, systolic blood pressure, cardiac output, or QRS interval (Table 2). Transient differences were observed in mean arterial pressure and systolic blood pressure at 4 minutes posthypotension for approximately 1 minute (difference: mean arterial pressure, sodium bicarbonate versus intravenous lipid emulsion –20.7 [95% confidence interval {CI} –31.6 to –9.8]; systolic blood pressure, sodium bicarbonate versus intravenous lipid emulsion –24.8 [95% CI –37.6 to –12.1]) (Figure 2). For graphic representation of data on individual swine in the sodium bicarbonate and intravenous lipid emulsion groups for mean arterial pressure, systolic blood pressure, and pulse rate, see Figures E1 through E6, available online at http:// www.annemergmed.com. We found no important difference between the 2 groups in mean QRS-interval measurements between baseline and hypotension in sodium bicarbonate versus intravenous lipid emulsion: mean difference 5.8 ms (95% CI –19.82 to 31.48 ms). Subjects in the intravenous lipid emulsion group died at a mean 12:33 minutes (SD 9:29) after time zero; the sodium bicarbonate group died at a mean 07:48 minutes (SD 2:44) and were not statistically different. One intravenous lipid emulsion pig (8.3%) and 2 sodium bicarbonate pigs (16.7%) survived (Figure 3). The mean QRS intervals were similar between groups when baseline and end-of-study values were compared in sodium bicarbonate versus intravenous lipid emulsion: 3 ms (95% CI –37.4 to 43.4 ms) (Table 3). LIMITATIONS Swine physiology does not specifically mimic that of humans, but swine cardiac physiology is similar.16,17 Swine models have been used for intoxicated models and to evaluate therapies, including intravenous lipid emulsion.13,18,19 Another limitation of an animal model is potential adverse events that humans may not manifest. Previous studies using a swine model have not produced a hypersensitivity syndrome, and yet intravenous lipid emulsion has been effective.13,20-23 In our study, we Annals of Emergency Medicine 5
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity Table 1. Mean values of hemodynamic parameters in each group before receiving diphenhydramine infusion.* Parameter MAP, mm Hg Pulse rate, beats/min Systolic blood pressure, mm Hg Cardiac output, L/min Percentage of oxygen saturation QRS interval, ms
Sodium Bicarbonate 94 85.8 112.6 4.9 72.3 82
(10) (21.5) (11.4) (1.1) (7.8) (34.1)
Difference (95% CI)†
Intravenous Lipid Emulsion 95 79.9 115.8 4.6 73.3 72.7
1 5.9 3.2 0.3 0.9 9.3
(8.2) (9.3) (9.6) (0.9) (5.1) (11.5)
(8.7 to 6.7) (8.1 to 20.0) (12.1 to 5.8) (0.6 to 1.1) (6.5 to 4.7) (12.2 to 30.9)
*Data are presented as mean (SD). † t Test.
observed transient skin flushing in 1 intravenous lipid emulsion pig. This pig died approximately 6 minutes after intravenous lipid emulsion infusion, similar to other animals. We administered diphenhydramine intravenously instead of orally, as would occur after overdose. However, this approach is used in experimental models and allowed us to control the rate and amount of diphenhydramine in each animal. Another limitation is that we studied 1 intravenous lipid emulsion dose. A different dose of intravenous lipid emulsion bolus or infusion may have produced different results. We used the dose of 7 mL/kg in accordance with previous studies.13 Another limitation is a potential analytic error in measuring serum diphenhydramine levels. The diphenhydramine levels were determined with traditional methods and compared with known, calibrated standards, using high-performance liquid chromatography and fluorescent detection techniques. Samples were measured multiple times and found to be accurate, reproducible, and consistent and were also similar to diphenhydramine levels reported in human cases. Finally, we did not include a placebo arm. However, before we began the experiments, we refined the protocol using several animals, as in our previous study.13 In 5 pigs, we found that treatment with saline solution alone for diphenhydramine toxicity was uniformly fatal, whereas administration of intravenous lipid emulsion alone without diphenhydramine resulted in 100% survival. As a result, in
discussion and in compliance with our Institutional Animal Care and Use Committee to reduce animal use, we did not use an experimental control arm.
DISCUSSION We found no difference in mean arterial pressure, QRSinterval measurement, or survival for intravenous lipid emulsion compared with sodium bicarbonate. The inability to find a difference between them in rescuing the diphenhydramine cardiotoxic subjects may be due to multiple factors. The first may be the intravenous lipid emulsion dose. We selected 7 mL/kg intravenous lipid emulsion, which produced milky white blood in the animals. Perhaps a larger dose would have produced a difference in outcomes. Second, the efficacy, if any, of intravenous lipid emulsion and sodium bicarbonate was the same under the conditions of this experiment. Diphenhydramine levels were not different in fatty versus aqueous serum samples. Third, the swine model may produce specific effects, causing concern that it is an inadequate model for lipid investigation.19 We conducted preliminary work in our protocol development animals in which we did not detect adverse effects.13 In our combined experience of studying the effects of intravenous lipid emulsion in more than 40 swine receiving intravenous lipid emulsion in 2 studies, we observed cutaneous flushing only twice and death once. Furthermore, other investigators have reported treating various toxins in the swine model.18,19,21,22 Finally, the animals may have been critically ill and simply not
Table 2. Mean values of hemodynamic parameters at hypotension.* Parameter MAP, mm Hg Pulse rate, beats/min Systolic blood pressure, mm Hg Cardiac output, L/min Percentage of oxygen saturation QRS interval
Sodium Bicarbonate 58.3 78.8 67.3 4.9 68.1 99
(5.5) (13.6) (7) (0.9) (10.5) (27.1)
Difference (95% CI)†
Intravenous Lipid Emulsion 58.2 89.4 67.4 4.5 67.8 83.8
(4.3) (11.5) (5.9) (0.9) (9.5) (20.5)
0.2 10.7 0.2 0.4 0.3 15.2
(4 to 4.3) (21.3 to 0) (5.7 to 5.3) (0.4 to 1.2) (8.2 to 8.7) (5.2 to 35.5)
*Data are presented as mean (SD). † t Test.
6 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 2. Hemodynamic variables (MAP, systolic blood pressure, and pulse rate) in diphenhydramine-poisoned animals for sodium bicarbonate and intravenous lipid emulsion groups on abbreviated timeline, emphasizing the 9 minutes posthypotension. A, Mean arterial pressure. B, Mean systolic blood pressure. C, Mean pulse rate. Data are reported as mean with SD bars. DPH, Diphenhydramine. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 7
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 3. Kaplan-Meier survival analysis comparing sodium bicarbonate and intravenous lipid emulsion groups of diphenhydraminepoisoned animals.
recoverable regardless of intervention. There may be other explanations as well. We measured diphenhydramine levels in all study animals. To our knowledge, there are no other published studies using lipid emulsion for diphenhydramine toxicity in which diphenhydramine levels were measured. In a review of human adult and pediatric patients, Nine and Rund24 noted diphenhydramine monointoxication deaths. They reported that the mean diphenhydramine level for adults was 16.14 mg/L (range 0.87 to 48.5 mg/L), and for pediatric patients it was 6.35 mg/L (range 0.69 to 13.7 mg/L). These values are similar to ours. We found that diphenhydramine levels were not higher in the serum lipid layer. Previous authors have reported that drugs of moderate to high lipophilicity may be more absorbed by intravenous lipid emulsion.12,23,25 In our previous work, we found that intravenous lipid emulsion
did not absorb amitriptyline, a tricyclic antidepressant.13 Similar to that in our study with amitriptyline, intravenous lipid emulsion in this study did not act as a “lipid sink” to absorb free diphenhydramine in the serum. If it had, then higher levels of diphenhydramine should have been detected in the serum lipid layer after centrifugation compared with the aqueous layer. Thus, intravenous lipid emulsion is ineffective in absorbing diphenhydramine, intravenous lipid emulsion works by different mechanisms for diphenhydramine, or both. To our knowledge, no other published studies using intravenous lipid emulsion for diphenhydramine toxicity have been reported. Although there are animal studies using intravenous lipid emulsion for other toxic agents or drugs, they involved small animals and did not measure drug levels.9-11 There are human case reports of diphenhydramine and intravenous lipid emulsion, but limitations of these
Table 3. Mean values of QRS interval and diphenhydramine levels at death.* Parameter QRS interval, ms DPH level, total, mg/mL DPH level, aqueous, mg/mL‡ DPH level, fatty layer, mg/mL‡
Sodium Bicarbonate 108.8 (31.1) 12.8 (7) — —
Difference (95% CI)†
Intravenous Lipid Emulsion 104.7 15.8 8.6 6.8
(31.9) (6.1) (5.5) (3.1)
4.1 (30.6 to 38.9) 3 (8.55 to 2.61)
—, Not measured. *Data are presented as mean (SD). † t Test. ‡ No animal received both intravenous lipid emulsion and sodium bicarbonate.
8 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
cases include timing of the intravenous lipid emulsion administration (late in resuscitation, as would be expected in current practice), the intravenous lipid emulsion dose used, and other uncontrollable or unknown factors. According to our results, intravenous lipid emulsion was not better than sodium bicarbonate for diphenhydramine toxicity. Future studies are needed with varied doses of intravenous lipid emulsion in large-animal critical care translational models. Moreover, the next step would be a randomized, controlled, blinded trial in humans comparing intravenous lipid emulsion and sodium bicarbonate to determine whether the former is effective in recovery from cardiotoxicity caused by lipophilic agents. However, heterogeneity of these clinical cases would make this difficult. In addition, case reports of successful treatment with intravenous lipid emulsion in toxicity caused by lipophilic agents may suffer from publication bias because negative-result reports are less likely to be published, making the denominator of all cases unknown.26 In our study of diphenhydramine-induced hypotension in swine, we did not detect a difference in improving hypotension or survival in animals treated with intravenous lipid emulsion compared with sodium bicarbonate. Other measures, such as QRS widening or drug concentrations in the blood aqueous layers in diphenhydramine-induced cardiotoxicity in swine, were also not improved by intravenous lipid emulsion. Supervising editor: Richard C. Dart, MD, PhD Author affiliations: From the Department of Emergency Medicine, University of Texas Health Science Center San Antonio, San Antonio, TX (Varney); Medical Toxicology (Bebarta) and the Department of Emergency Medicine (Castaneda), San Antonio Military Medical Center, San Antonio, TX; the Department of Emergency Medicine, San Antonio Military Medical Center, San Antonio, TX (Boudreau); the Office of the Chief Scientist/59th MDW, Wilford Hall Ambulatory Surgical Center, San Antonio, TX (Vargas, Zarzabal); and Medpro Technologies, San Antonio, TX (Zarzabal). Author contributions: SMV and VSB conceived and designed the trial, supervised the conduct of the study and data collection, and drafted the article. SMV obtained research funding. SMV, VSB, SMB, TEV, and MC performed the study. LAZ provided statistical advice on study design, analyzed the data, and prepared the figures. All authors contributed substantially to article revision. SMV takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist and provided the following details: The study was funded by an Emergency Medicine Basic Research Skills grant from the Emergency Medicine Foundation and by a Volume
-,
no.
-
:
-
2015
grant from the US Air Force Office of the Surgeon General (SG5, FWH20100054A). Publication dates: Received for publication February 24, 2015. Revisions received April 9, 2015; April 28, 2015; and May 21, 2015. Accepted for publication May 28, 2015. Presented as a poster at the ACEP Scientific Assembly, October 2012, Denver, CO. The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the US Air Force, Department of Defense, or the US government.
REFERENCES 1. Sangster J. Octanol-Water Partition Coefficients: Fundamentals and Physical Chemistry. West Sussex, England: Wiley; 1997. 2. Botch-Jones SR, Johnson R, Kleinschmidt K, et al. Diphenhydramine’s role in death investigations: an examination of diphenhydramine prevalence in 2 US geographical areas. Am J Forensic Med Pathol. 2014;35:181-185. 3. Erbe S, Bschor T. Diphenhydramine addiction and detoxification. A systematic review and case report. Psychiatr Prax. 2013;40:248-251. 4. Abdi A, Rose E, Levine M. Diphenhydramine overdose with intraventricular conduction delay treated with hypertonic sodium bicarbonate and IV lipid emulsion. West J Emerg Med. 2014;15:855-858. 5. Eckes L, Tsokos M, Herre S, et al. Toxicological identification of diphenhydramine (DPH) in suicide. Forensic Sci Med Pathol. 2013;9:145-153. 6. Cole JB, Stellpflug SJ, Gross EA, et al. Wide complex tachycardia in a pediatric diphenhydramine overdose treated with sodium bicarbonate. Pediatr Emerg Care. 2011;27:1175-1177. 7. Jang DH, Manini AF, Trueger NS, et al. Status epilepticus and widecomplex tachycardia secondary to diphenhydramine overdose. Clin Toxicol (Phila). 2010;48:945-948. 8. Association of Anaesthetists of Great Britain and Ireland. AAGBI safety guideline: Management of severe local anaesthetic toxicity. Available at: http://www.aagbi.org/sites/default/files/la_toxicity_2010_0.pdf. Published December 2010. Accessed April 20, 2015. 9. Harvey M, Cave G. Intralipid outperforms sodium bicarbonate in a rabbit model of clomipramine toxicity. Ann Emerg Med. 2007; 2:178-185. 10. Bania T, Chu J. Hemodynamic effect of Intralipid in amitriptyline toxicity. Acad Emerg Med. 2006;13:S177. 11. Yoav G, Odelia G, Shaltiel C. A lipid emulsion reduces mortality from clomipramine overdose in rats. Vet Hum Toxicol. 2002;44:30. 12. Abdelmalek D, Schwarz ES, Sampson C, et al. Life-threatening diphenhydramine toxicity presenting with seizures and a wide complex tachycardia improved with intravenous fat emulsion. Am J Ther. 2014;21:542-544. 13. Varney SM, Bebarta VS, Vargas TE, et al. Intravenous lipid emulsion therapy does not improve hypotension compared to sodium bicarbonate for tricyclic antidepressant toxicity: a randomized, controlled pilot study in a swine model. Acad Emerg Med. 2014;21:1212-1219. 14. Wang Z, Goonewardene LA. The use of MIXED models in the analysis of animal experiments with repeated measures data. Can J Anim Sci. 2004;84:1-11. 15. Westfall P, Tobias R, Wolfinger R. Multiple Comparisons and Multiple Tests Using SAS. 2nd ed. Cary, NC: SAS Press; 2011. 16. Suzuki Y, Yeung AC, Ikeno F. The representative porcine model for human cardiovascular disease. J Biomed Biotechnol. 2011:1-10. 17. White F, Roth D, Bloor C. The pig as a model for myocardial ischemia and exercise. Lab Anim Sci. 1986;36:351-356.
Annals of Emergency Medicine 9
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity 18. Bebarta VS, Tanen DA, Boudreau S, et al. Intravenous cobinamide versus hydroxocobalamin for acute treatment of severe cyanide poisoning in a swine (Sus scrofa) model. Ann Emerg Med. 2014;64:612-619. 19. Offerman SR, Barry JD, Richardson WH, et al. Subcutaneous Crotaline Fab antivenom for the treatment of rattlesnake envenomation in a porcine model. Clin Toxicol (Phila). 2009;47:61-68. 20. Weinberg G, Rubinstein I. Pig in a poke: species specificity in modeling lipid resuscitation. Anesth Analg. 2012;114:907-909. 21. de Queiroz Siqueira M, Chassard D, Musard H, et al. Resuscitation with lipid, epinephrine, or both in levobupivacaine-induced cardiac toxicity in newborn piglets. Br J Anaesth. 2014;112:729-734. 22. Mauch J, Jurado OM, Spielmann N, et al. Resuscitation strategies from bupivacaine-induced cardiac arrest. Paediatr Anaesth. 2012;22: 124-129.
10 Annals of Emergency Medicine
23. Heinonen JA, Litonius E, Backman JT, et al. Intravenous lipid emulsion entraps amitriptyline into plasma and can lower its brain concentration—an experimental intoxication study in pigs. Basic Clin Pharmacol Toxicol. 2013;113:193-200. 24. Nine JS, Rund CR. Fatality from diphenhydramine monointoxication: a case report and review of the infant, pediatric, and adult literature. Am J Forensic Med Pathol. 2006;27:36-41. 25. French D, Smolline C, Ruan W, et al. Partition constant and volume of distribution as predictors of clinical efficacy of lipid rescue for toxicological emergencies. Clin Toxicol (Phila). 2011;49: 801-809. 26. Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014;7:264. Available at http://www. biomedcentral.com/1756-0500/7/264. Accessed June 1, 2013.
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E1. Mean arterial pressure of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E2. Mean arterial pressure of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 10.e1
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E3. Systolic blood pressure of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E4. Systolic blood pressure of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension.
10.e2 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E5. Pulse rate of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E6. Mean pulse rate of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 10.e3
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity: A Randomized, Controlled Pilot Study in a Swine Model Shawn M. Varney, MD*; Vikhyat S. Bebarta MD; Susan M. Boudreau, RN, BSN; Toni E. Vargas, PA-C; Maria Castaneda, MS; Lee A. Zarzabal, MS *Corresponding Author. E-mail: [email protected].
Study objective: Diphenhydramine is a moderately lipophilic antihistamine with sodium channel blockade properties. It is consumed recreationally for mild hallucinogenic and hypnotic effects and causes dysrhythmias, seizures, and death with overdose. Intravenous lipid emulsion is a novel agent used to treat lipophilic drug overdose. Two case reports describe clinical improvement with intravenous lipid emulsion after diphenhydramine toxicity, but no prospective studies have been reported. Our objective is to determine whether intravenous lipid emulsion improved hypotension compared with sodium bicarbonate for severe diphenhydramine toxicity in a model of critically ill swine. Methods: Twenty-four swine weighing 45 to 55 kg were infused with diphenhydramine at 1 mg/kg per minute until the mean arterial pressure reached 60% of baseline. Subjects were randomized to receive intravenous lipid emulsion (bolus of 7 mL/kg and then 0.25 mL/kg per minute) or sodium bicarbonate (2 mEq/kg plus an equal volume of normal saline solution). We measured pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, QRS interval, and serum diphenhydramine level. Twelve animals per group provided a power of 0.8 and a of .05 to detect a 50% difference in mean arterial pressure. We assessed differences between groups with a repeated-measures linear model (MIXED) and Kaplan-Meier estimation methods. We compared systolic blood pressure, mean arterial pressure, and cardiac output with repeated measures ANOVA. Results: Baseline weight, hemodynamic parameters, QRS interval, time to hypotension, and diphenhydramine dose required to achieve hypotension were similar between groups. After hypotension was reached, there was no overall difference between intravenous lipid emulsion and sodium bicarbonate groups for cardiac output or QRS intervals; however, there were transient differences in mean arterial pressure and systolic blood pressure, favoring intravenous lipid emulsion (difference: mean arterial pressure, sodium bicarbonate versus intravenous lipid emulsion –20.7 [95% confidence interval –31.6 to –9.8]; systolic blood pressure, sodium bicarbonate versus intravenous lipid emulsion –24.8 [95% confidence interval –37.6 to –12.1]). Time to death was similar. One intravenous lipid emulsion and 2 sodium bicarbonate pigs survived. End-of-study mean total serum diphenhydramine levels were similar. The mean lipid layer diphenhydramine level was 6.8 mg/mL (SD 3.1 mg/mL) and mean aqueous layer level 8.6 mg/mL (SD 5.5 mg/mL). Conclusion: In our study of diphenhydramine-induced hypotensive swine, we found no difference in hypotension, QRS widening, or diphenhydramine levels in aqueous layers between intravenous lipid emulsion and sodium bicarbonate. [Ann Emerg Med. 2015;-:1-10.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2015 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2015.05.028
INTRODUCTION Background Diphenhydramine is a moderately lipophilic (octanol:water partition coefficient 3.40) first-generation antihistamine with sodium channel blockade properties similar to those of tricyclic antidepressants.1 It is used recreationally for its mild hallucinogenic and hypnotic effects, and its abuse has recently increased.2,3 Patients who overdose typically present with an anticholinergic toxidrome. More important, in severe cases, seizures, ventricular dysrhythmias, and death occur.4-7 Severe Volume
-,
no.
-
:
-
2015
intoxication with QRS widening on the ECG is usually treated with sodium bicarbonate or hypertonic saline solution. Intravenous lipid emulsion, or lipid therapy, is a medication with established efficacy for severe local anesthetic toxicity and has been used to treat intoxication by other lipophilic agents.8 Some studies in small, lean animals have reported that intravenous lipid emulsion improves hypotension and prolongs survival compared with standard therapy for drug intoxication, whereas other studies show no effect.9-11 To our knowledge, thus far only Annals of Emergency Medicine 1
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Editor’s Capsule Summary
What is already known on this topic Intravenous lipid emulsion is a novel agent used to treat lipophilic drug overdose. Two case reports describe clinical improvement with intravenous lipid emulsion after diphenhydramine toxicity, but no prospective studies have been reported. What question this study addressed Does intravenous lipid emulsion produce improved effects on hypotension compared with sodium bicarbonate for severe diphenhydramine toxicity in a model of critically ill swine? What this study adds to our knowledge Several relevant endpoints (pulse rate, systolic and mean arterial blood pressure, cardiac output, QRS interval, serum diphenhydramine level, and mortality) showed no difference between the intravenous lipid emulsion and sodium bicarbonate groups. How this is relevant to clinical practice Intravenous lipid emulsion is an emerging therapy frequently used for cardiovascular toxicity. These animal data indicate that intravenous lipid emulsion is unlikely to have a meaningful effect on diphenhydramine cardiovascular toxicity.
case reports have been published on its use in humans, and often multiple therapies were used in each case. In 2 cases, intravenous lipid emulsion was used for diphenhydramine toxicity. In both cases, multiple therapies had been administered and the intravenous lipid emulsion was administered late in the resuscitation; therefore, the effects of each intervention could not be determined.4,12 To our knowledge, there are no preclinical or clinical prospective studies evaluating intravenous lipid emulsion as a treatment for diphenhydramine toxicity. According to the heterogeneity of overdose cases and infrequent presentations in single centers, a prospective clinical trial is not feasible. Therefore, a randomized, controlled study in a model of diphenhydramine-induced cardiotoxicity in critically ill, large animals may provide evidence on the effectiveness of intravenous lipid emulsion. Goal of This Investigation Our objective was to determine whether intravenous lipid emulsion was more effective in improving hypotension 2 Annals of Emergency Medicine
compared with sodium bicarbonate for treatment of diphenhydramine overdose in a model of critically ill swine. MATERIALS AND METHODS Study Design and Setting This investigation was a nonblinded, randomized, parallelgroup-design study approved by the Institutional Animal Care and Use Committee of the Wilford Hall Clinical Research Division. A flowchart of this experimental design is shown in Figure 1. All animal experiments complied with the regulations and guidelines of the Animal Welfare Act and the American Association for Accreditation of Laboratory Animal Care. Animals were housed and the study was conducted in the Animal Care Facility. Selection of Participants Before the experiment, the animals were randomly assigned with an online randomization tool (http://www. randomization.com) to one of 2 intervention groups: sodium bicarbonate, the standard drug for sodium channel blockade; or intravenous lipid emulsion, the study drug. Our methods were similar to those of an intravenous lipid emulsion study for amitriptyline cardiotoxicity.13 Twenty-four healthy female Yorkshire swine (Sus scrofa) weighing 45 to 55 kg were sedated with intramuscular ketamine 10 mg/kg and anesthetized with isoflurane through nose cone. After intubation, they were maintained on isoflurane (1% to 3%) and oxygen (FiO2 of 0.4 to 0.45). Temperature was maintained for all animals between 36.8 C (98.2 F) and 38 C (100.4 F) with heating adjuncts, including a warmed induction room and operating room, warm intravenous fluids at all times, a bed warmer during the procedure, and a warming blanket. Isoflurane was titrated to 1% to 2% to mitigate isoflurane-induced hypotension, and arterial pCO2 was maintained between 38 and 42 mm Hg. Interventions Instrumentation included placement of an endotracheal tube for ventilation. Invasive hemodynamic variables were measured with 8-French Swan-Ganz continuous cardiac output, mixed SvO2 monitoring catheters (model 746HF8) and the Edwards Vigilance II monitor (Edwards Lifesciences, Irvine, CA). Measurements included continuous cardiac output, systemic vascular resistance, mixed SvO2, central venous pressure, pulmonary artery pressure, and core temperature. The catheter ports were flushed with saline solution, and the catheter was placed through cut-down in the right internal jugular vein. Aortic pressure was measured continuously through the femoral artery with a 2-French Millar catheter (Millar Instruments, Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 1. Study timeline with interventions. Time (T) points “zT-60 min” and “zT-30 min” are only estimates and differ for each animal. The study times of interest began when the animals reached hypotension (time 0 minutes) and received treatment. T, Hypotension time. MAP, Mean arterial pressure.
Houston, TX). An 8.5-French introducer (Arrow, Reading, PA) was placed in the carotid artery for laboratory sampling, and another was placed in the femoral vein for medication administration. All animals received a bolus of 15 mL/kg of warmed normal saline solution intravenously during instrumentation. Baseline values were obtained for serum electrolytes, CBC count, and diphenhydramine levels, as well as arterial blood gases for pH, pCO2, bicarbonate, and lactate. Venous samples were obtained at hypotension and at subsequent 15-minute intervals to measure changes in serum electrolyte and lactate levels. Additionally, a baseline ECG was performed to document QRS- and QTc-interval measurements. We used several animals to refine the model, including control animals that received intravenous lipid emulsion alone without adverse effects. For our experiment, diphenhydramine (Amerisource Bergen, Chesterbrook, PA) at 5 mg/mL was infused intravenously at 1 mg/kg per minute through a central venous catheter until hypotension was reached. We defined hypotension as a decrease in mean arterial pressure to 60% of baseline value. A normal saline solution bolus of 10 mL/kg was then administered to the animals. Before the experiment, the swine were randomized and assigned (12 to each group) to either the experimental group (intravenous lipid emulsion bolus of 7 mL/kg, followed by an infusion of 0.25 mL/kg per minute until the end of the study) or to the standard treatment group (8.4% sodium bicarbonate bolus of 2 mEq/kg, plus an equivalent volume of normal saline solution). We determined these doses of intravenous lipid emulsion and sodium bicarbonate according to previously published studies for sodium channel blockade toxicity in rabbits, rats, and our protocol Volume
-,
no.
-
:
-
2015
development animals because no studies in large animals had been reported with diphenhydramine toxicity, to our knowledge.9-11 A control arm using normal saline solution alone was not used in this study because of Institutional Animal Care and Use Committee recommendations. The standard therapy arm (sodium bicarbonate administration) served as the control because critically ill patients presenting to the emergency department with diphenhydramine toxicity would likely be treated with sodium bicarbonate as a standard management approach. The Institutional Animal Care and Use Committee granted permission to administer diphenhydramine to 5 pigs until hypotension was reached and then to resuscitate using only normal saline solution to study the effect. Despite 2 boluses of intravenous fluids at 7 mL/kg (representing the intravenous lipid emulsion bolus) and additional boluses (10 mL/kg), all 5 swine died. We used 15 protocol development animals to determine the toxic doses of diphenhydramine and effective dose of intravenous lipid emulsion, as well as to test the effects of intravenous lipid emulsion on the animal model. To determine the appropriate diphenhydramine dose to cause hypotension, we used previous animal studies.9 With an infusion of diphenhydramine at 1 mg/kg per minute, animals achieved hypotension and subsequent death. Less toxic regimens (0.25 to 0.5 mg/kg per minute) resulted in long periods to reach hypotension. To identify an effective intravenous lipid emulsion dose, we initially used the dose recommended for humans (bolus of 1.5 mL/kg, followed by infusion at 0.25 mL/kg per minute) and increased stepwise to 7 mL/kg when no response in blood pressure was observed.8 Although studies have described higher doses in rabbit and rat models (up to Annals of Emergency Medicine 3
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
12 and 15 mL/kg, respectively), even at 7 mL/kg the animals’ blood was milky white. To determine the effect of intravenous lipid emulsion alone on the swine model, we administered 7 mL/kg to animals that had received no diphenhydramine. We noted no skin flush or alteration in hemodynamic or respiratory parameters. The animals survived without consequence. Necropsy was performed and showed no pulmonary emboli or other pulmonary toxicity. We have published use of the intravenous lipid emulsion dose and effects of intravenous lipid emulsion in control animals in a previous study.13
Methods of Measurement and Outcome Measures Hemodynamic parameters (pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, systemic vascular resistance, SVO2, and central venous pressure) were measured continuously throughout the study and 60 minutes after standard or study drug administration. Laboratory data, ECG (Zoll M Series CCT; ZOLL Medical Corporation, Chelmsford, MA), and serum diphenhydramine levels (measured by high-performance liquid chromatography and fluorescent detection) were recorded every 15 minutes from study initiation to study end. Diphenhydramine levels were determined in the total serum sample, fatty layer, and aqueous layer. A simplified summary of the methodology for determining diphenhydramine levels in total serum, the lipid-rich fraction, and the lipid-poor fraction consisted of 4 steps, described in detail below: (1) preparation of the lipid-rich and lipid-poor fractions; (2) extraction of diphenhydramine; (3) high-pressure liquid chromatography analysis of diphenhydramine with fluorescent detection; and (4) specificity, linearity, reproducibility (precision), and accuracy assessments. Lipid-rich and lipid-poor fractions were prepared by centrifuging 1-mL samples of each pig’s serum at 14,500 revolutions per minute for 20 minutes in a walk-in cold room at 2 C (35.6 F) to 8 C (46.4 F). After the lipidpoor layer was removed with a Pasteur pipet, 1 mL of water was added to the remaining fatty layer, vortex mixed for 10 to 20 seconds, and then transferred with a Pasteur pipet to a separate tube. Serum diphenhydramine calibration curves were prepared at 0, 6.25, 12.5, 25.0, 50.0, and 100.0 mg/mL. Sodium hydroxide and an extraction solvent were added to each calibrator and test sample and shaken. Samples were then centrifuged at 3,000 revolutions per minute for 20 minutes at 5 C (41 F) (SD 3 C [37.4 F]). The top organic layer was transferred to separate tubes and evaporated in a TurboVap LV evaporator (Biotage, Uppsala, Sweden) at 50 C (122 F) 4 Annals of Emergency Medicine
(SD 3 C [37.4 F]) under a gentle stream of nitrogen. The samples were then reconstituted in 1.0 mL of the highpressure liquid chromatography mobile phase and transferred to high-pressure liquid chromatography injection vials. High-pressure liquid chromatography analysis was performed on a Waters Alliance high-pressure liquid chromatography machine (Waters 2695 Separations Module; Waters, Milford, MA) equipped with a Phenomenex Luna C18 chromatographic column (Phenomenex, Torrance, CA) and a Waters 474 Fluorescent detector. The highpressure liquid chromatography mobile phase consisted of 2.5 mM ammonium acetate in methanol-water, 50:50, volume/volume, adjusted to pH 4.0. The fluorescence detector excitation and emission wavelengths were 230 and 560 nm, respectively. Injection volumes were 50.0 mL and chromatographic flow rates were 0.2 mL/min. The method was specific in that no interferences were detected in serum samples from 6 pigs not receiving diphenhydramine and in any of the predose samples. The serum diphenhydramine standards curves were linear from 6.25 to 100.0 mg/mL (R2¼0.9987; n¼4). Recoveries of serum diphenhydramine over the concentration range of 6.25 to 100.0 mg/mL were 97.8%. Within-run precision of serum diphenhydramine at 12.5, 25.0, and 50.0 mg/mL (n¼10) was 2.62%, 1.28%, and 0.8%, respectively. Between-day precision of the 12.5, 25.0, and 50.0 mg/mL calibrators was 14.2%, 6.78%, and 2.94% (n¼10), respectively. Accuracies of the 6.25, 12.5, 25.0, 50.0, and 100.0 mg/mL calibrators were 100.5%, 97.8%, 100.3%, 106.4%, and 98.1%, respectively. At the end of the study, any surviving animals were euthanized with intravenous administration of sodium pentobarbital 100 mg/kg. We defined death as asystole when mean arterial pressure was less than 20 mm Hg because some animals may have had a rhythm and no measurable blood pressure, or only a small pulse pressure.
Primary Data Analysis The study’s primary outcome was an improvement in mean arterial pressure of greater than 50% of trough values after administration of the study drug (sodium bicarbonate or intravenous lipid emulsion). Baseline hemodynamic parameters were assessed for treatment group differences with a parametric t test. A secondary outcome was posthypotension reversal of the QRS-interval widening on ECG. The preliminary assessment of the primary outcome was assessed with a repeated-measures linear model with adjustment for treatment group, time, and the treatmentby-time interaction, with a first-order autoregressive covariance structure assumed.14 The RM ANOVA also Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
appropriately managed missing data. In accordance with the preliminary assessment of a significant treatment-bytime interaction, a post hoc analysis was conducted to evaluate treatment differences during the follow-up period up to the point of 15 minutes posthypotension, at which time only 1 subject remained in the intravenous lipid emulsion group. Post hoc comparisons were adjusted for multiple testing with the Sidak correction method, and after adjustment, no significant differences between treatment groups were detected.15 Secondary analysis was performed to assess the relationship between survival time from the point of hypotension to treatment, using the Kaplan-Meier method and log-rank test. It was determined that equal sample sizes of 12 animals per group would achieve a power of 0.8 to detect a 50% difference in mean arterial pressure with a 2-sided test at significance level of .05 (NCSS PASS 2002, Kaysville, UT). Differences in response (eg, improvement in diphenhydramine-induced hypotension) between the 2 treatments were assessed with repeated-measures ANOVA adjusted for treatment, time, and the interaction of treatment by time. We compared mean values at time zero between groups for pulse rate, mean arterial pressure, cardiac output, serum lactate levels, QRS interval, diphenhydramine dose administered, and serum diphenhydramine levels. All statistical testing was 2-sided with a significance level of 5% and was performed with SAS (version 9.3; SAS Institute, Inc., Cary, NC). All graphic presentations were created with R (version 3.0.2). RESULTS Characteristics of Study Subjects Twenty-four swine were randomly assigned to intravenous lipid emulsion and sodium bicarbonate groups (12 in each arm). At baseline, the groups had similar weights (intravenous lipid emulsion 48.4 kg [SD 3.3]; sodium bicarbonate 49.4 kg [SD 3]). In addition, there was no important difference for each group in regard to mean pulse rate, systolic blood pressure, mean arterial pressure, cardiac output, SvO2, and QRS intervals (Table 1). Main Results Intravenous lipid emulsion and sodium bicarbonate groups required similar mean amounts of diphenhydramine to reach hypotension (mean intravenous lipid emulsion 31.2 mg/kg [SD 8.3]; mean sodium bicarbonate 33.3 mg/ kg [SD 8.3]). The total mean amounts of diphenhydramine infused were 1,510 mg in the intravenous lipid emulsion group and 1,645 mg in the sodium bicarbonate group. The intravenous lipid emulsion group reached hypotension at a Volume
-,
no.
-
:
-
2015
mean 32 minutes 13 seconds [SD 9:41]; the sodium bicarbonate group, 34 minutes 8 seconds [SD 8:36]. The time required to reach hypotension varied for each animal and is shown in Figure 1 as “zT-60 min,” or approximately 60 minutes before hypotension. The time when hypotension was achieved and standard or study drug was administered was time 0 and the starting point for observation of study drug effect. There was no overall difference between intravenous lipid emulsion and sodium bicarbonate groups for mean arterial pressure, pulse rate, systolic blood pressure, cardiac output, or QRS interval (Table 2). Transient differences were observed in mean arterial pressure and systolic blood pressure at 4 minutes posthypotension for approximately 1 minute (difference: mean arterial pressure, sodium bicarbonate versus intravenous lipid emulsion –20.7 [95% confidence interval {CI} –31.6 to –9.8]; systolic blood pressure, sodium bicarbonate versus intravenous lipid emulsion –24.8 [95% CI –37.6 to –12.1]) (Figure 2). For graphic representation of data on individual swine in the sodium bicarbonate and intravenous lipid emulsion groups for mean arterial pressure, systolic blood pressure, and pulse rate, see Figures E1 through E6, available online at http:// www.annemergmed.com. We found no important difference between the 2 groups in mean QRS-interval measurements between baseline and hypotension in sodium bicarbonate versus intravenous lipid emulsion: mean difference 5.8 ms (95% CI –19.82 to 31.48 ms). Subjects in the intravenous lipid emulsion group died at a mean 12:33 minutes (SD 9:29) after time zero; the sodium bicarbonate group died at a mean 07:48 minutes (SD 2:44) and were not statistically different. One intravenous lipid emulsion pig (8.3%) and 2 sodium bicarbonate pigs (16.7%) survived (Figure 3). The mean QRS intervals were similar between groups when baseline and end-of-study values were compared in sodium bicarbonate versus intravenous lipid emulsion: 3 ms (95% CI –37.4 to 43.4 ms) (Table 3). LIMITATIONS Swine physiology does not specifically mimic that of humans, but swine cardiac physiology is similar.16,17 Swine models have been used for intoxicated models and to evaluate therapies, including intravenous lipid emulsion.13,18,19 Another limitation of an animal model is potential adverse events that humans may not manifest. Previous studies using a swine model have not produced a hypersensitivity syndrome, and yet intravenous lipid emulsion has been effective.13,20-23 In our study, we Annals of Emergency Medicine 5
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity Table 1. Mean values of hemodynamic parameters in each group before receiving diphenhydramine infusion.* Parameter MAP, mm Hg Pulse rate, beats/min Systolic blood pressure, mm Hg Cardiac output, L/min Percentage of oxygen saturation QRS interval, ms
Sodium Bicarbonate 94 85.8 112.6 4.9 72.3 82
(10) (21.5) (11.4) (1.1) (7.8) (34.1)
Difference (95% CI)†
Intravenous Lipid Emulsion 95 79.9 115.8 4.6 73.3 72.7
1 5.9 3.2 0.3 0.9 9.3
(8.2) (9.3) (9.6) (0.9) (5.1) (11.5)
(8.7 to 6.7) (8.1 to 20.0) (12.1 to 5.8) (0.6 to 1.1) (6.5 to 4.7) (12.2 to 30.9)
*Data are presented as mean (SD). † t Test.
observed transient skin flushing in 1 intravenous lipid emulsion pig. This pig died approximately 6 minutes after intravenous lipid emulsion infusion, similar to other animals. We administered diphenhydramine intravenously instead of orally, as would occur after overdose. However, this approach is used in experimental models and allowed us to control the rate and amount of diphenhydramine in each animal. Another limitation is that we studied 1 intravenous lipid emulsion dose. A different dose of intravenous lipid emulsion bolus or infusion may have produced different results. We used the dose of 7 mL/kg in accordance with previous studies.13 Another limitation is a potential analytic error in measuring serum diphenhydramine levels. The diphenhydramine levels were determined with traditional methods and compared with known, calibrated standards, using high-performance liquid chromatography and fluorescent detection techniques. Samples were measured multiple times and found to be accurate, reproducible, and consistent and were also similar to diphenhydramine levels reported in human cases. Finally, we did not include a placebo arm. However, before we began the experiments, we refined the protocol using several animals, as in our previous study.13 In 5 pigs, we found that treatment with saline solution alone for diphenhydramine toxicity was uniformly fatal, whereas administration of intravenous lipid emulsion alone without diphenhydramine resulted in 100% survival. As a result, in
discussion and in compliance with our Institutional Animal Care and Use Committee to reduce animal use, we did not use an experimental control arm.
DISCUSSION We found no difference in mean arterial pressure, QRSinterval measurement, or survival for intravenous lipid emulsion compared with sodium bicarbonate. The inability to find a difference between them in rescuing the diphenhydramine cardiotoxic subjects may be due to multiple factors. The first may be the intravenous lipid emulsion dose. We selected 7 mL/kg intravenous lipid emulsion, which produced milky white blood in the animals. Perhaps a larger dose would have produced a difference in outcomes. Second, the efficacy, if any, of intravenous lipid emulsion and sodium bicarbonate was the same under the conditions of this experiment. Diphenhydramine levels were not different in fatty versus aqueous serum samples. Third, the swine model may produce specific effects, causing concern that it is an inadequate model for lipid investigation.19 We conducted preliminary work in our protocol development animals in which we did not detect adverse effects.13 In our combined experience of studying the effects of intravenous lipid emulsion in more than 40 swine receiving intravenous lipid emulsion in 2 studies, we observed cutaneous flushing only twice and death once. Furthermore, other investigators have reported treating various toxins in the swine model.18,19,21,22 Finally, the animals may have been critically ill and simply not
Table 2. Mean values of hemodynamic parameters at hypotension.* Parameter MAP, mm Hg Pulse rate, beats/min Systolic blood pressure, mm Hg Cardiac output, L/min Percentage of oxygen saturation QRS interval
Sodium Bicarbonate 58.3 78.8 67.3 4.9 68.1 99
(5.5) (13.6) (7) (0.9) (10.5) (27.1)
Difference (95% CI)†
Intravenous Lipid Emulsion 58.2 89.4 67.4 4.5 67.8 83.8
(4.3) (11.5) (5.9) (0.9) (9.5) (20.5)
0.2 10.7 0.2 0.4 0.3 15.2
(4 to 4.3) (21.3 to 0) (5.7 to 5.3) (0.4 to 1.2) (8.2 to 8.7) (5.2 to 35.5)
*Data are presented as mean (SD). † t Test.
6 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 2. Hemodynamic variables (MAP, systolic blood pressure, and pulse rate) in diphenhydramine-poisoned animals for sodium bicarbonate and intravenous lipid emulsion groups on abbreviated timeline, emphasizing the 9 minutes posthypotension. A, Mean arterial pressure. B, Mean systolic blood pressure. C, Mean pulse rate. Data are reported as mean with SD bars. DPH, Diphenhydramine. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 7
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure 3. Kaplan-Meier survival analysis comparing sodium bicarbonate and intravenous lipid emulsion groups of diphenhydraminepoisoned animals.
recoverable regardless of intervention. There may be other explanations as well. We measured diphenhydramine levels in all study animals. To our knowledge, there are no other published studies using lipid emulsion for diphenhydramine toxicity in which diphenhydramine levels were measured. In a review of human adult and pediatric patients, Nine and Rund24 noted diphenhydramine monointoxication deaths. They reported that the mean diphenhydramine level for adults was 16.14 mg/L (range 0.87 to 48.5 mg/L), and for pediatric patients it was 6.35 mg/L (range 0.69 to 13.7 mg/L). These values are similar to ours. We found that diphenhydramine levels were not higher in the serum lipid layer. Previous authors have reported that drugs of moderate to high lipophilicity may be more absorbed by intravenous lipid emulsion.12,23,25 In our previous work, we found that intravenous lipid emulsion
did not absorb amitriptyline, a tricyclic antidepressant.13 Similar to that in our study with amitriptyline, intravenous lipid emulsion in this study did not act as a “lipid sink” to absorb free diphenhydramine in the serum. If it had, then higher levels of diphenhydramine should have been detected in the serum lipid layer after centrifugation compared with the aqueous layer. Thus, intravenous lipid emulsion is ineffective in absorbing diphenhydramine, intravenous lipid emulsion works by different mechanisms for diphenhydramine, or both. To our knowledge, no other published studies using intravenous lipid emulsion for diphenhydramine toxicity have been reported. Although there are animal studies using intravenous lipid emulsion for other toxic agents or drugs, they involved small animals and did not measure drug levels.9-11 There are human case reports of diphenhydramine and intravenous lipid emulsion, but limitations of these
Table 3. Mean values of QRS interval and diphenhydramine levels at death.* Parameter QRS interval, ms DPH level, total, mg/mL DPH level, aqueous, mg/mL‡ DPH level, fatty layer, mg/mL‡
Sodium Bicarbonate 108.8 (31.1) 12.8 (7) — —
Difference (95% CI)†
Intravenous Lipid Emulsion 104.7 15.8 8.6 6.8
(31.9) (6.1) (5.5) (3.1)
4.1 (30.6 to 38.9) 3 (8.55 to 2.61)
—, Not measured. *Data are presented as mean (SD). † t Test. ‡ No animal received both intravenous lipid emulsion and sodium bicarbonate.
8 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
cases include timing of the intravenous lipid emulsion administration (late in resuscitation, as would be expected in current practice), the intravenous lipid emulsion dose used, and other uncontrollable or unknown factors. According to our results, intravenous lipid emulsion was not better than sodium bicarbonate for diphenhydramine toxicity. Future studies are needed with varied doses of intravenous lipid emulsion in large-animal critical care translational models. Moreover, the next step would be a randomized, controlled, blinded trial in humans comparing intravenous lipid emulsion and sodium bicarbonate to determine whether the former is effective in recovery from cardiotoxicity caused by lipophilic agents. However, heterogeneity of these clinical cases would make this difficult. In addition, case reports of successful treatment with intravenous lipid emulsion in toxicity caused by lipophilic agents may suffer from publication bias because negative-result reports are less likely to be published, making the denominator of all cases unknown.26 In our study of diphenhydramine-induced hypotension in swine, we did not detect a difference in improving hypotension or survival in animals treated with intravenous lipid emulsion compared with sodium bicarbonate. Other measures, such as QRS widening or drug concentrations in the blood aqueous layers in diphenhydramine-induced cardiotoxicity in swine, were also not improved by intravenous lipid emulsion. Supervising editor: Richard C. Dart, MD, PhD Author affiliations: From the Department of Emergency Medicine, University of Texas Health Science Center San Antonio, San Antonio, TX (Varney); Medical Toxicology (Bebarta) and the Department of Emergency Medicine (Castaneda), San Antonio Military Medical Center, San Antonio, TX; the Department of Emergency Medicine, San Antonio Military Medical Center, San Antonio, TX (Boudreau); the Office of the Chief Scientist/59th MDW, Wilford Hall Ambulatory Surgical Center, San Antonio, TX (Vargas, Zarzabal); and Medpro Technologies, San Antonio, TX (Zarzabal). Author contributions: SMV and VSB conceived and designed the trial, supervised the conduct of the study and data collection, and drafted the article. SMV obtained research funding. SMV, VSB, SMB, TEV, and MC performed the study. LAZ provided statistical advice on study design, analyzed the data, and prepared the figures. All authors contributed substantially to article revision. SMV takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist and provided the following details: The study was funded by an Emergency Medicine Basic Research Skills grant from the Emergency Medicine Foundation and by a Volume
-,
no.
-
:
-
2015
grant from the US Air Force Office of the Surgeon General (SG5, FWH20100054A). Publication dates: Received for publication February 24, 2015. Revisions received April 9, 2015; April 28, 2015; and May 21, 2015. Accepted for publication May 28, 2015. Presented as a poster at the ACEP Scientific Assembly, October 2012, Denver, CO. The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the US Air Force, Department of Defense, or the US government.
REFERENCES 1. Sangster J. Octanol-Water Partition Coefficients: Fundamentals and Physical Chemistry. West Sussex, England: Wiley; 1997. 2. Botch-Jones SR, Johnson R, Kleinschmidt K, et al. Diphenhydramine’s role in death investigations: an examination of diphenhydramine prevalence in 2 US geographical areas. Am J Forensic Med Pathol. 2014;35:181-185. 3. Erbe S, Bschor T. Diphenhydramine addiction and detoxification. A systematic review and case report. Psychiatr Prax. 2013;40:248-251. 4. Abdi A, Rose E, Levine M. Diphenhydramine overdose with intraventricular conduction delay treated with hypertonic sodium bicarbonate and IV lipid emulsion. West J Emerg Med. 2014;15:855-858. 5. Eckes L, Tsokos M, Herre S, et al. Toxicological identification of diphenhydramine (DPH) in suicide. Forensic Sci Med Pathol. 2013;9:145-153. 6. Cole JB, Stellpflug SJ, Gross EA, et al. Wide complex tachycardia in a pediatric diphenhydramine overdose treated with sodium bicarbonate. Pediatr Emerg Care. 2011;27:1175-1177. 7. Jang DH, Manini AF, Trueger NS, et al. Status epilepticus and widecomplex tachycardia secondary to diphenhydramine overdose. Clin Toxicol (Phila). 2010;48:945-948. 8. Association of Anaesthetists of Great Britain and Ireland. AAGBI safety guideline: Management of severe local anaesthetic toxicity. Available at: http://www.aagbi.org/sites/default/files/la_toxicity_2010_0.pdf. Published December 2010. Accessed April 20, 2015. 9. Harvey M, Cave G. Intralipid outperforms sodium bicarbonate in a rabbit model of clomipramine toxicity. Ann Emerg Med. 2007; 2:178-185. 10. Bania T, Chu J. Hemodynamic effect of Intralipid in amitriptyline toxicity. Acad Emerg Med. 2006;13:S177. 11. Yoav G, Odelia G, Shaltiel C. A lipid emulsion reduces mortality from clomipramine overdose in rats. Vet Hum Toxicol. 2002;44:30. 12. Abdelmalek D, Schwarz ES, Sampson C, et al. Life-threatening diphenhydramine toxicity presenting with seizures and a wide complex tachycardia improved with intravenous fat emulsion. Am J Ther. 2014;21:542-544. 13. Varney SM, Bebarta VS, Vargas TE, et al. Intravenous lipid emulsion therapy does not improve hypotension compared to sodium bicarbonate for tricyclic antidepressant toxicity: a randomized, controlled pilot study in a swine model. Acad Emerg Med. 2014;21:1212-1219. 14. Wang Z, Goonewardene LA. The use of MIXED models in the analysis of animal experiments with repeated measures data. Can J Anim Sci. 2004;84:1-11. 15. Westfall P, Tobias R, Wolfinger R. Multiple Comparisons and Multiple Tests Using SAS. 2nd ed. Cary, NC: SAS Press; 2011. 16. Suzuki Y, Yeung AC, Ikeno F. The representative porcine model for human cardiovascular disease. J Biomed Biotechnol. 2011:1-10. 17. White F, Roth D, Bloor C. The pig as a model for myocardial ischemia and exercise. Lab Anim Sci. 1986;36:351-356.
Annals of Emergency Medicine 9
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity 18. Bebarta VS, Tanen DA, Boudreau S, et al. Intravenous cobinamide versus hydroxocobalamin for acute treatment of severe cyanide poisoning in a swine (Sus scrofa) model. Ann Emerg Med. 2014;64:612-619. 19. Offerman SR, Barry JD, Richardson WH, et al. Subcutaneous Crotaline Fab antivenom for the treatment of rattlesnake envenomation in a porcine model. Clin Toxicol (Phila). 2009;47:61-68. 20. Weinberg G, Rubinstein I. Pig in a poke: species specificity in modeling lipid resuscitation. Anesth Analg. 2012;114:907-909. 21. de Queiroz Siqueira M, Chassard D, Musard H, et al. Resuscitation with lipid, epinephrine, or both in levobupivacaine-induced cardiac toxicity in newborn piglets. Br J Anaesth. 2014;112:729-734. 22. Mauch J, Jurado OM, Spielmann N, et al. Resuscitation strategies from bupivacaine-induced cardiac arrest. Paediatr Anaesth. 2012;22: 124-129.
10 Annals of Emergency Medicine
23. Heinonen JA, Litonius E, Backman JT, et al. Intravenous lipid emulsion entraps amitriptyline into plasma and can lower its brain concentration—an experimental intoxication study in pigs. Basic Clin Pharmacol Toxicol. 2013;113:193-200. 24. Nine JS, Rund CR. Fatality from diphenhydramine monointoxication: a case report and review of the infant, pediatric, and adult literature. Am J Forensic Med Pathol. 2006;27:36-41. 25. French D, Smolline C, Ruan W, et al. Partition constant and volume of distribution as predictors of clinical efficacy of lipid rescue for toxicological emergencies. Clin Toxicol (Phila). 2011;49: 801-809. 26. Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014;7:264. Available at http://www. biomedcentral.com/1756-0500/7/264. Accessed June 1, 2013.
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E1. Mean arterial pressure of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E2. Mean arterial pressure of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 10.e1
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E3. Systolic blood pressure of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 9 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E4. Systolic blood pressure of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension.
10.e2 Annals of Emergency Medicine
Volume
-,
no.
-
:
-
2015
Varney et al
Intravenous Lipid Emulsion Therapy for Severe Diphenhydramine Toxicity
Figure E5. Pulse rate of all pigs treated with sodium bicarbonate, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Sodium bicarbonate was administered at hypotension.
Figure E6. Mean pulse rate of all pigs treated with intravenous lipid emulsion, graphed over time from the beginning of the study to 10 minutes after hypotension (time to hypotension is abbreviated). Intravenous lipid emulsion was administered at hypotension. Volume
-,
no.
-
:
-
2015
Annals of Emergency Medicine 10.e3
