Original Contributions

doi:10.1016/j.jemermed.2010.02.030 The Journal of Emergency Medicine, Vol. 41, No. 5, pp. 453 459, 2011 Copyright 2011 Elsevier Inc. Printed in the USA. All rights reserved 0736-4679/$ see front matter Original Contributions EFFECT OF THE ADDITION OF VASOPRESSIN OR VASOPRESSIN PLUS NITROGLYCERIN TO EPINEPHRINE ON ARTERIAL BLOOD PRESSURE DURING CARDIOPULMONARY RESUSCITATION IN HUMANS Laurent Ducros, MD,* Eric Vicaut, MD, PHD, Christian Soleil, MD,* Morgan Le Guen, MD,* Papa Gueye, MD, PHD,* Thomas Poussant, MD,* Alexandre Mebazaa, MD, PHD,* Didier Payen, MD, PHD,* and Patrick Plaisance, MD, PHD *Department of Anesthesiology and Critical Care, Clinical Research Unit, and Department of Emergency Medicine, Lariboisière University Hospital, Paris, France Reprint Address: Laurent Ducros, MD, Department of Anesthesiology and Critical Care, Lariboisière University Hospital, 2 Rue Ambroise Pare, Paris 75010, France e Abstract Background: Infusion of a vasopressor during cardiopulmonary resuscitation (CPR) in humans increases end decompression (diastolic) arterial blood pressure, and consequently increases vital organ perfusion pressure and survival. Several vasoactive drugs have been tested alone or in combination, but their hemodynamic effects have not been investigated clinically in humans. Study Objective: We tested the hypothesis that epinephrine (1 mg) co-administered with vasopressin (40 IU) nitroglycerin (300 g) results in higher diastolic blood pressure than epinephrine alone. Study Design: A prospective, randomized, double-blinded controlled trial in the prehospital setting. The study included 48 patients with witnessed cardiac arrest. Patients received either epinephrine alone (E alone) or epinephrine plus vasopressin (E V) or epinephrine plus vasopressin plus nitroglycerin (E V N). A femoral arterial catheter was inserted for arterial pressure measurement. Outcome Measures: The primary end point was diastolic blood pressure during CPR, 15 min after the first drug administration (T 15 min). Results: After exclusions, a total of 44 patients were enrolled. Diastolic blood pressures (mm Hg) at T 15 min were not statistically different between groups (median [interquartile range]: 20 [10], 15 [6], and 15 [13] for E alone, E V, and E V N, This study was supported by Advanced Circulatory Systems Incorporated, formerly CPRx, Minneapolis, Minnesota. respectively. The rate of return of spontaneous circulation was 63% (n 10) in the epinephrine group, 43% (n 6) in the epinephrine plus vasopressin group, and 36% (n 5) in the triple therapy group (NS). Conclusions: Addition of vasopressin or vasopressin plus nitroglycerin to epinephrine did not increase perfusion blood pressure compared to epinephrine alone in humans in cardiac arrest, suggesting the absence of benefit in using these drug combination(s). 2011 Elsevier Inc. e Keywords epinephrine; vasopressin; nitroglycerin; out-ofhospital CPR; blood pressure; cardiac arrest INTRODUCTION Vasopressor infusion during cardiopulmonary resuscitation (CPR) in humans in cardiac arrest increases the tone of both venous and arterial vessels, venous return, arterial blood pressure, and consequently, vital organ perfusion pressure. It is recognized that during CPR, maintenance of an adequate diastolic arterial blood pressure (the pressure at the end of the chest decompression) preserves vital organ perfusion and is important for survival (1,2). Poor outcomes after cardiac arrest have raised the question of the optimal pharmacological approach to augment circulation during CPR. High-dose RECEIVED: 7 April 2009; FINAL SUBMISSION RECEIVED: 30 August 2009; ACCEPTED: 17 February 2010 453

454 L. Ducros et al. epinephrine improves return of spontaneous circulation (ROSC) and hospital admission but there is no difference in long-term survival or neurologic outcome at hospital discharge compared to low-dose epinephrine (3). The use of natural or synthesized vasopressors other than epinephrine has been the subject of many investigations. Despite encouraging results in animal studies and in human clinical trials, vasopressin alone is reported to have no benefit in improving survival to hospital discharge when compared to epinephrine in the treatment of patients with in-hospital and out-of-hospital cardiac arrest (4 10). The combination of vasopressin plus epinephrine raises arterial blood pressure in animals but is associated with a decrease in cerebral and endocardial blood flow (11,12). Based upon the hypothesis that this drug combination causes an increase in vasoconstriction in cerebral and endocardial vascular beds, animal studies have been performed with the addition of nitroglycerin as a nitric oxide releasing agent; improved vital organ blood flow was found with the combination of nitroglycerin plus vasopressin plus epinephrine, vs. epinephrine alone (13,14). However, the hemodynamic effect of the three drug combinations (epinephrine alone, epinephrine plus vasopressin, or epinephrine plus vasopressin plus nitroglycerin) remains unknown in humans in cardiac arrest. In the present study, we tested the hypothesis that the addition of vasopressin to epinephrine, with or without nitroglycerin, would result in higher diastolic arterial blood pressure than epinephrine alone during CPR for human cardiac arrest. MATERIALS AND METHODS This study is a prospective, randomized, double-blinded, and controlled clinical trial performed on an intention-totreat basis. It was approved by the local Institutional Review Board (IRB; the Consultative Council for the Protection of Persons Volunteering for Biomedical Research). The IRB waived the requirement for informed consent due to clinical circumstances; signed consent was obtained from patients who eventually were discharged from the hospital. The study period was August 2001 to August 2004. This study was performed in a single out-of-hospital setting in Paris, France. The prehospital emergency response chain of survival in Paris is two-tiered and combines basic life support (BLS) with firefighters and advanced cardiac life support (ACLS) with out-of-hospital emergency medical services, as previously described (15). The following patients were included: adults with a witnessed out-of-hospital cardiac arrest who presented with ventricular fibrillation, pulseless electrical activity, or asystole and requiring ACLS, and for whom BLS with active compression-decompression CPR (CardioPump, Ambu, Le Haillan, France) was already started by firefighters before the arrival of the medical team, at a compression rate of 100.min 1 and a compression:ventilation ratio of 15:2, according to the 2000 guidelines (16). Active compression-decompression is routinely applied by firefighters in Paris. Of note, vasopressors can be administered only by the medical team. Patients with the following criteria were excluded: unwitnessed cardiac arrest, spontaneous palpable carotid or femoral pulse restored before administration of a vasopressor, lack of intravenous access, pregnancy, traumatic injuries and anatomic abnormalities that prevented safe femoral artery cannulation. Patients with presumed irreversible death or known terminal illness at the beginning of ACLS also were not included for ethical consideration. Figure 1 presents the timeline of the study. If all inclusion criteria were met, patients underwent randomization. Treatment assignments to the study drugs were randomly generated by a computer and placed in sealed numbered envelopes. Time intervals were noted based upon witnesses declaration and dispatching center recordings. Upon arrival at the scene, the prehospital medical team started ACLS that included tracheal intubation connected to an impedance threshold device (Advanced Circulatory Systems, Minneapolis, MN) and ventilation with 100% oxygen. Immediately after the venous line was placed and after randomization, the study patients were split into three groups. Each group received a simultaneous intravenous (i.v.) administration of one of three study drugs every 5 min at T 0, T 5, and T 10, except in the case of ROSC. Epinephrine (E) administration was not blinded, whereas vasopressin (V), nitroglycerin (N), or placebos were given blindly. All drugs were exclusively injected intravenously, followed by 20 ml of normal saline. The study drugs (vasopressin or nitroglycerin) were prepared every day in advance by the pharmacist of our hospital in plastic syringes according to local sterilization procedure to ensure drug stability and safety 24 h after removal from the glass ampules. The three study groups were: 1) group E alone received open E (1 mg) and two syringes (1 ml each) of placebo; 2) group E V received open E (1 mg) V (40 IU) placebo; 3) group E V N received open E (1 mg) V (40 IU) N (300 g). The dose of vasopressin (40 IU) was the same as that chosen in previously published human studies (6,7,9). The three successive administrations of study drugs every 5 min and the dose of nitroglycerin were adapted from two previous animal studies (pigs) in cardiac arrest (14,17). Additional interventions such as the administration of sodium bicarbonate, atropine, lidocaine, or amiodarone and fibrinolysis were used at the discretion of the physician managing the CPR. After the 15 min study period, all patients still in cardiac arrest (regardless of their study group) received 1 mg of epi-

Arterial Blood Pressure during CPR 455 Figure 1. Diagram presenting the timeline of the study with the number of patients with ROSC (return of spontaneous circulation) and the number of patients with successful arterial blood pressure measurement (art. meas.) in each studied group. Group E epinephrine alone; group E V epinephrine plus vasopressin; group E V N epinephrine plus vasopressin plus nitroglycerin. nephrine every 3 min until ROSC was achieved or ACLS was discontinued. During the first minutes of the study period, after the insertion of the i.v. line and the first drug administration, a femoral arterial catheter (Seldicath, 5F, 12 cm, Plastimed, Le Plessis-Bouchard, France) was inserted and connected with fluid-flushed tubing to transducers (Sorensen Transpac III, Abbott Systems, Chicago, IL) and a monitoring system (Propaq Encore, Physiocontrol, Welch Allyn, Mareuil les Meaux, France) to continuously measure diastolic arterial blood pressure during decompression and systolic blood pressure during compression. Transducers were calibrated and taped to the midaxillary line of the patient. End-tidal carbon dioxide (ETCO 2 ) was also continuously measured with a capnometer (Normocap, Datex-Ohmeda, Madison, WI). Diastolic and systolic blood pressure and ETCO 2 of patients who were still in cardiac arrest 10 min after the initiation of study drug administration were recorded three times: at 10, 15, and 20 min after the initiation of study drug administration (T 10, T 15, T 20, respectively). If catheterization was not successful, patients were analyzed for outcome data (as study drugs were given) and not for arterial pressure. The primary end point was the diastolic (end decompression) arterial pressure during CPR, 15 min after the initiation of study drug infusion, as expected from a previous animal study (14). Secondary end points were systolic (end compression) arterial pressure and ETCO 2 15 min after the initiation of study drug infusion, the time-course of the diastolic and systolic arterial pressure and ETCO 2 between T 10 and T 20, and the rates of ROSC, hospital admission, and hospital discharge. For all parameters studied, comparisons of the three treatments were performed using the non-parametric Kruskall-Wallis test. It was decided a priori to make individual comparisons among treatments only if the global test was significant. At the time the protocol was written, we did not have the knowledge of the standard deviation value for diastolic blood pressure after the administration of the study drugs in humans. We chose a sample size of n 16 patients per group to have an 80% power to detect an effect size equal to 1 (i.e., a difference between one group and the others equal to one standard deviation of diastolic blood pressure) with a two-sided significance level fixed at 5%. Under the hypothesis that some patients would recover spontaneous cardiac activity before T 15, the required sample size would not be reached for comparison of diastolic blood pressure among groups during CPR. However, complete exclusion of these patients might have induced a bias. To limit the consequences of their exclusion in the primary end point analysis, and to retain data on the hemodynamic effect of the three treatments, we added a sensitivity analysis with a test based on ranks that included patients who recovered cardiac activity, who were attributed an arbitrarily high value.

456 L. Ducros et al. RESULTS There were 48 patients screened for eligibility for the study (Figure 1). Four patients were excluded for ethical considerations because life support was discontinued within minutes of starting ACLS (2 patients with known terminal illness and 2 with presumed irreversible death). Among the 44 patients who were enrolled in the study and who received study drugs, 19 patients presented with ROSC before T 15, and 25 were still in cardiac arrest at T 15, among whom arterial catheterization and arterial blood pressure measurements were successful in 19 patients (Figure 1). No patients received fibrinolysis, serum bicarbonate, or other drugs that could have interfered with hemodynamic parameters except for patients presenting with ventricular fibrillation who received amiodarone according to guidelines. Baseline characteristics were similar among the three treatment groups (Table 1). The mean time intervals between collapse and basic life support, or collapse and the first drug administration, were similar among groups. In addition, baseline characteristics of the subgroup of patients analyzed at T 15 for primary end point were similar to those of the entire group of patients. The mean time interval between first drug injection and ROSC was significantly longer in the E V N group compared to the E V group. Diastolic blood pressure during CPR, 15 min after the first administration of study drugs, was similar among the three groups (Figure 2): 20[10], 15[6], and 15[13] for E alone, E V, and E V N, respectively (p 0.68; in mm Hg, median [interquartile range]). There was also no difference among groups for systolic arterial blood pressure and ETCO 2 at this time point. The time course of diastolic and systolic blood pressure and ETCO 2 between T 10 and T 20 was similar among the three groups. Of note, ETCO 2 values trended higher in the E V N group at T 10. Concerning outcome, the rate of ROSC at T 15 was 10/16 (63%) in group E alone, 6/14 (43%) in group E V, and 5/14 (36%) in group E V N(Table 1). Although the rates of ROSC, hospital admission, and hospital discharge seemed favorable after E alone, there were no statistically significant differences when compared with the two other groups. Two patients survived to hospital discharge without neurologic impairment. They received bystander CPR and belonged to the E-alone group. To eliminate the bias due to the resuscitation of 19 patients before 15 min, a test based on ranks was employed with an Table 1. Baseline Characteristics and Outcomes of the Patients in the Three Studied Groups* Group E Alone (n 16) Group E V (n 14) Group E V N (n 14) Total (n 44) Patients with Remaining Cardiac Arrest at T15 (n 25) Age year, mean SE 60 4 56 4 52 4 56 2 56 3 Male/female, n 13/3 12/2 13/1 38/6 23/2 CPR by bystander, n 7 4 5 16 8 Suspected cause of cardiac arrest, n Myocardial infarction 8 7 9 24 12 Dysrhythmia 1 0 1 2 0 Pulmonary embolism 2 2 0 4 1 Asphyxia 1 0 1 2 1 Others 4 5 3 12 11 Initial cardiac rhythm, n Ventricular fibrillation 0 2 4 6 4 Pulseless electrical activity 8 3 3 14 4 Asystole 8 9 7 24 17 Time from collapse to BLS 8 6 11 2 9 2 9 1 10 2 (min; mean SD ) Time from collapse to T0 20 2 23 3 23 3 22 2 24 2 (min; mean SD) Time from call for help to T0 15 1 19 1 18 2 17 1 18 1 (min; mean SD) Time from T0 to ROSC (min; mean SD) 8 2 5 1 14 3 9 1 25 0 ROSC, n 10 6 5 21 Hospital admission, n 8 5 4 17 Hospital discharge, n 2 0 0 2 * Group E alone (epinephrine alone), group E V (epinephrine plus vasopressin), group E V N (epinephrine plus vasopressin plus nitroglycerin), both for the entire group of screened patients and for the subgroup of patients who remained in cardiac arrest 15 min after the first drug administration (T 15). p 0.048 between E alone and E V groups. p 0.017 between E V and E V N groups. CPR cardiopulmonary resuscitation; BLS basic life support; ROSC return of spontaneous circulation; T0: time of the first drug administration.

Arterial Blood Pressure during CPR 457 Figure 2. Evolution of diastolic arterial blood pressure, systolic arterial pressure, and end-tidal CO 2 (ETCO 2 ) between 10 and 20 min after the first drug administration in patients with continuing cardiac arrest and for whom arterial catheterization was achieved (Median and quartiles [10 th ;90 th percentiles]). The diastolic arterial blood pressure, systolic arterial blood pressure, and ETCO 2 were similar at each time point among the three groups. Group E epinephrine alone; group E V epinephrine plus vasopressin; group E V N epinephrine plus vasopressin plus nitroglycerin. arbitrarily high value attributed to those patients. There were no statistically significant differences in diastolic blood pressure values among the three groups at T 15. DISCUSSION To the best of our knowledge, these data are the first human data reported that examine the hemodynamic consequences of various drug regimens in a prehospital setting in patients in cardiac arrest. We aimed to determine the optimal drug combination that would give the highest diastolic arterial blood pressure during CPR, for potentially better outcomes. The present study was designed to detect a large difference in diastolic blood pressure between epinephrinealone and epinephrine-plus-vasopressin groups. Contrary to the hypothesis, vasopressin did not confer a large hemodynamic benefit in this patient population. Surprisingly, arterial blood pressure values tended to be higher in the epinephrine-alone group, which was consistent with the finding of a higher ROSC rate in this group. The comparison between epinephrine alone and vasopressin alone was deliberately not tested. In a large clinical study comparing epinephrine and vasopressin in an out-of-hospital setting, there was no survival difference found between groups (9). The post hoc analysis of the subset of patients with asystole who received vasopressin showed significant improvement to hospital discharge but not neurologically intact survival. Moreover, all of those patients received epinephrine after vasopressin infusion. These negative results have been confirmed by others (8,10). It has been emphasized in the most recent guidelines that there is insufficient evidence to support or refute the use of vasopressin for any cardiac arrest rhythm, or eventually for patients in asystole or pulseless electrical activity (18,19). The emphasis on the combination of epinephrine and vasopressin seemed more relevant for this study. In a retrospective analysis of out-of-hospital cardiac arrest, it was observed that the addition of vasopressin to epinephrine was associated with an increased rate of ROSC among the group presenting with asystole (20). However, the results from a large French multicenter study recently have been published; there was no difference in the outcomes of patients receiving either epinephrine alone (1 mg) or the combination of epinephrine (1 mg) and vasopressin (40 IU) as the two first drug administrations (21). Addition of nitroglycerin to vasopressin did not alter arterial blood pressure during CPR, as was previously described in animals (14). Moreover, the higher ETCO 2 values in this group suggests that circulation may have been enhanced with the addition of nitroglycerin to vasopressin and epinephrine, supporting the physiological hypothesis previously tested in animals (13,14). On the contrary, however, there was a trend toward lower resuscitation rates, with or without nitroglycerin, when vasopressin was co-administered with epinephrine. This could be attributed to a detrimental effect from administering vasopressin three times within 10 min whereas the usual approach is one 40-IU dose. However, in a pig model of cardiac arrest, the three-time vasopressin administration was, on the contrary, associated with both higher arterial blood pressure during CPR and a higher ROSC rate (17).

458 L. Ducros et al. Limitations There are some limitations to this study. First, blood pressure measured at the aortic root is commonly the most relevant diastolic pressure for vital organ perfusion pressure, however, femoral artery blood pressure is correlated to the aortic pressure and usually used in clinical trials in humans in cardiac arrest (22,23). In many experimental and clinical studies, arterial pressure is the main determinant of vital organ perfusion pressure and has been shown to be determinant of survival (2). Second, the calculated sample size was not reached for two reasons: 1) Arterial line catheterization was unsuccessful in 6/25 patients; 2) The number of patients for whom the full hemodynamic measures were performed was lower than expected. This is related to the much shorter time to ROSC in our study (roughly 9 min) compared to a previous study (19 min) that also used the impedance threshold valve (23). The design of the present study was based upon the results of the earlier study, and it was anticipated that a greater number of patients would remain in cardiac arrest for the primary end point analysis. These differences cannot be explained. CONCLUSION In conclusion, the present study shows that the addition of vasopressin or vasopressin plus nitroglycerin to epinephrine does not increase diastolic blood pressure during cardiopulmonary resuscitation 15 min after drug administration in humans with prolonged cardiac arrest. Conversely, epinephrine administration alone induced higher, though non-significant, values of arterial blood pressure, ROSC, and survival rates. These results suggest the absence of benefit in outcome from the drug combinations studied when compared to epinephrine alone. REFERENCES 1. Paradis N, Martin G, Rivers E, et al. Coronary perfusion pressure and the return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990;263:1106 13. 2. Sanders A, Ewy G, Taft T. Prognostic and therapeutic importance of aortic diastolic pressure in resuscitation from cardiac arrest. Crit Care Med 1984;12:871 3. 3. Gueugniaud PY, Mols P, Goldstein P, et al. A comparison of repeated high doses and repeated standard doses of epinephrine for cardiac arrest outside the hospital. European Epinephrine Study Group. N Engl J Med 1998;339:1595 601. 4. Lindner KH, Brinkmann A, Pfenninger EG, Lurie KG, Goertz A, Lindner IM. Effect of vasopressin on hemodynamic variables, organ blood flow, and acid-base status in a pig model of cardiopulmonary resuscitation. Anesth Analg 1993;77:427 35. 5. Wenzel V, Lindner KH, Prengel AW, et al. Vasopressin improves vital organ blood flow after prolonged cardiac arrest with postcountershock pulseless electrical activity in pigs. Crit Care Med 1999;27:486 92. 6. Lindner KH, Prengel AW, Brinkmann A, Strohmenger HU, Lindner IM, Lurie KG. Vasopressin administration in refractory cardiac arrest. Ann Intern Med 1996;124:1061 4. 7. Lindner KH, Dirks B, Strohmenger HU, Prengel AW, Lindner IM, Lurie KG. Randomised comparison of epinephrine and vasopressin in patients with out-of-hospital ventricular fibrillation. Lancet 1997;349:535 7. 8. Stiell IG, Hebert PC, Wells GA, et al. Vasopressin versus epinephrine for inhospital cardiac arrest: a randomised controlled trial. Lancet 2001;358:105 9. 9. Wenzel V, Krismer AC, Arntz HR, Sitter H, Stadlbauer KH, Lindner KH. A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation. N Engl J Med 2004; 350:105 13. 10. Aung K, Htay T. Vasopressin for cardiac arrest: a systematic review and meta-analysis. Arch Intern Med 2005;165:17 24. 11. Wenzel V, Linder KH, Augenstein S, Prengel AW, Strohmenger HU. Vasopressin combined with epinephrine decreases cerebral perfusion compared with vasopressin alone during cardiopulmonary resuscitation in pigs. Stroke 1998;29:1462 7; discussion 1467 8. 12. Mulligan KA, McKnite SH, Lindner KH, Lindstrom PJ, Detloff B, Lurie KG. Synergistic effects of vasopressin plus epinephrine during cardiopulmonary resuscitation. Resuscitation 1997;35:265 71. 13. Wenzel V, Lindner KH, Mayer H, Lurie KG, Prengel AW. Vasopressin combined with nitroglycerin increases endocardial perfusion during cardiopulmonary resuscitation in pigs. Resuscitation 1998;38:13 7. 14. Lurie KG, Voelckel WG, Iskos DN, et al. Combination drug therapy with vasopressin, adrenaline (epinephrine) and nitroglycerin improves vital organ blood flow in a porcine model of ventricular fibrillation. Resuscitation 2002;54:187 94. 15. Plaisance P, Adnet F, Vicaut E, et al. Benefit of active compressiondecompression cardiopulmonary resuscitation as a prehospital advanced cardiac life support. A randomized multicenter study. Circulation 1997;95:955 61. 16. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: section 6: pharmacology II: agents to optimize cardiac output and blood pressure. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation 2000;102(8 Suppl):I129 35. 17. Mayr VD, Wenzel V, Voelckel WG, et al. Developing a vasopressor combination in a pig model of adult asphyxial cardiac arrest. Circulation 2001;104:1651 6. 18. International Liaison Committee on Resuscitation. 2005 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations. Part 4: Advanced life support. Resuscitation 2005;67:213 47. 19. ECC Committee, Subcommittees and Task Forces of the American Heart Association. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112(24 Suppl):IV1 203. 20. Guyette FX, Guimond GE, Hostler D, Callaway CW. Vasopressin administered with epinephrine is associated with a return of a pulse in out-of-hospital cardiac arrest. Resuscitation 2004;63:277 82. 21. Gueugniaud PY, David J, Chanzy E, et al. Vasopressin and epinephrine vs. epinephrine alone in cardiopulmonary resuscitation. N Engl J Med 2008;359:21 30. 22. Rivers E, Lozon J, Enriquez E, et al. Simultaneous radial, femoral, and aortic arterial pressures during human cardiopulmonary resuscitation. Crit Care Med 1993;21:878 83. 23. Plaisance P, Lurie KG, Payen D. Inspiratory impedance during active compression-decompression cardiopulmonary resuscitation: a randomized evaluation in patients in cardiac arrest. Circulation 2000;101:989 94.

Arterial Blood Pressure during CPR 459 ARTICLE SUMMARY 1. Why is this topic important? Maintenance of adequate diastolic blood pressure is an important issue for survival during advanced cardiac life support. Addition of vasopressin alone or vasopressin plus nitroglycerin to epinephrine improves vital organ perfusion in animals, but the hemodynamic effects of these drug combinations are unknown in humans. 2. What does this study attempt to show? This study attempts to test the hypothesis that the addition of vasopressin or vasopressin plus nitroglycerin to epinephrine would result in higher diastolic arterial blood pressure during cardiopulmonary resuscitation compared to epinephrine alone, and consequently, to higher vital organ perfusion. 3. What are the key findings? Contrary to the hypothesis, arterial blood pressure values during cardiopulmonary resuscitation tended to be higher in the epinephrine alone group compared to the two other groups. In addition, we found more return of spontaneous circulation in this group. 4. How is patient care impacted? These results suggest the absence of benefit in outcome from the combination of epinephrine plus vasopressin, with or without nitroglycerin, compared to epinephrine alone, in advanced cardiac life support.