Outcome after cardiopulmonary resuscitation (CPR) is poor. The optimal approach to chest compressions and vasoactive drug and ventilatory management in CPR has been the focus of much research1 and has been incorporated into updated guidelines.2 Although positive end-expiratory pressure (PEEP) has been studied in many settings in the critically ill3 and in connection with ventricular fibrillation,4 the role of PEEP in asphyxial cardiac arrest has received limited attention. Administration of PEEP may increase cardiac output by lessening left ventricular afterload or by reversing atelectasis-induced increases in pulmonary vascular resistance (PVR), thereby reducing right ventricular afterload. In addition, PEEP may increase oxygenation by lessening the development of pulmonary edema,3 decreasing atelectasis, and improving ventilation/perfusion matching. Conversely, PEEP may potentially reduce cardiac output and organ perfusion by reducing venous return.
We hypothesized that administration of PEEP would improve outcome after CPR in a rodent model of experimental cardiac arrest.
Animal Selection and Care
Adult male Sprague-Dawley rats (340–400 g; Charles-River, Quebec) were used in all experiments. Institutional Ethics approval was obtained from the Animal Care Committee at the Research Institute at the Hospital for Sick Children. All animal work conformed to the guidelines of the Canadian Council for Animal Care.
Anesthesia and Ventilation
Anesthesia was induced with intraperitoneal ketamine (60 mg/kg) and xylazine (5 mg/kg). The depth of anesthesia was confirmed by the absence of response to paw clamp. A tracheostomy was performed with a 14° cannula. Teflon catheters (24°, Becton Dickinson, Sandy, UT) were inserted into the penile vein and into the aorta (via the femoral artery). Electrocardiogram recordings were made with subcutaneous limb leads. The animals’ lungs were ventilated with the following settings: tidal volume (VT, 8 mL/kg), respiratory rate (RR, 38/min), PEEP (1 cm H20), and Fio2 0.21. A small animal ventilator (Model 683; Harvard Apparatus, South Natick, MA) was used. A fluid bolus of lactated Ringer’s solution (5 mL/kg) was administered over 15 min. Anesthesia was maintained with a continuous infusion of ketamine and xylazine (20 and 1 mg · kg−1 · h−1, respectively). Core temperature was regulated with a homeothermic blanket control unit (Model 50-7079-F; Harvard Apparatus). The animals were then allowed 20 min for stabilization.
Asphyxial cardiac arrest was induced by disconnecting the tracheostomy tube from the ventilator. Cardiac arrest was defined as a mean arterial blood pressure of <10 mm Hg. The asphyxial time interval was defined as the period between ventilator disconnection and the commencement of resuscitative efforts. Resuscitation was commenced 1 min after the onset of cardiac arrest5,6 by restarting ventilation (VT 8 mL/kg, RR 38/min, with PEEP and Fio2 as below— Series 1). Administration of a single dose of IV epinephrine (Series 1: 30 μg/kg, Series 2: 10 μg/kg; see below) was followed by manual anteroposterior compression of the thorax (approximately 200/min).7 Successful initial resuscitation was defined as return of spontaneous circulation (ROSC) when the mean arterial blood pressure was sustained at ≥50 mm Hg. Survival, the main study end point, was defined as the presence of spontaneous circulation for 60 or 120 min after initial resuscitation. Resuscitative efforts were discontinued if ROSC did not occur within 10 min of chest compressions. Mean arterial blood pressure, body temperature, and airway pressures were monitored throughout. Arterial blood gases were measured before cardiac arrest and at 10, 30, and 60 min after resuscitation. Ventilation was for 60 min (Series 1) or 120 min (Series 2) or until death (defined here as mean arterial blood pressure <10 mm Hg). At the end of the experiment, all animals were killed using an anesthetic overdose.
Two series of experiments were performed:
- Series 1. Asphyxial cardiac arrest followed by resuscitation with epinephrine 30 μg/kg, Fio2 1.0, and ventilation as above, and randomized to either 5 cm H2O PEEP or to zero PEEP. To assess whether the higher oxygenation achieved with PEEP explained the results, an additional group was studied in which CPR was completed using 5 cm H2O PEEP and Fio2 = 0.21 (instead of 1.0). Finally, to determine the effect of PEEP on functional residual capacity (FRC), additional groups of animals were studied in which the FRC was measured 15 min after CPR with PEEP of 5 or 0 cm H2O (Fio2 1.0 only).8
- Series 2. Asphyxial cardiac arrest was followed by resuscitation with epinephrine 10 μg/kg (vs 30 μg/kg as in Series 1), Fio2 = 1.0, ventilated as above but extended to 120 min, and randomized to 5 cm H2O PEEP or to zero PEEP.
The sonographer was blinded to the allocation of PEEP. Echocardiography was performed using a Hewlett-Packard Sonos 5500 echocardiographic system (M2424A, Andover, MA) with a 7.5 MHz transducer set to a sweep speed of 150 cm/s, as previously described.5,6 Examination was performed in the supine position, with all two-dimensional images and M-Mode tracings acquired at chordal level from a short-axis view. The transducer was gently applied to the left parasternal border to obtain a short-axis view of the left (distal to the mitral valve leaflets) and right ventricles at a frame rate of at least 113 Hz and a depth of 2 cm. Data were obtained at baseline and at 1, 2, 3, 4, 5, 10, 30, and 60 min after resuscitation. Images were stored electronically for subsequent blinded analysis. From M-Mode images, off-line measurements were made of left ventricular cavity dimensions at end-systole and end-diastole. Fractional shortening (FS) was calculated as:
where LVEDA and LVESD indicate left ventricular end-diastolic and end-systolic diameters, respectively. From two-dimensional images, left ventricular cavity areas were calculated from a short-axis view of the heart just distal to the mitral valve. Left ventricular end-diastolic area (LVEDA) was determined at the peak of ventricular filling by tracing along the endocardial border of the chamber, which adopts a circular configuration. Left ventricular end-systolic area was determined after ventricular emptying by tracing along the endocardial border of the chamber in a similar fashion. Three measurements were taken for each variable and were averaged.
Lung Edema and Functional Residual Capacity
Lung edema was assessed using lung wet/dry weight ratio, as previously described.5 FRC was measured at end-expiration by direct volume displacement at the completion of the experiment, as previously described.8
All data were analyzed using commercial software (Sigma Stat, Version 2.0; Jandel Corporation, San Raphael, CA). Survival data were compared using Kaplan–Meier analysis. Fischer’s exact test or χ2 tests was used to compare proportions, and parametric data were analyzed by analysis of variance and post hoc Student–Newman–Keuls tests. Nonparametric data are expressed by repeated measures analysis of variance for ranks. Significance was set at P <0.05. Nonparametric data are expressed as median and interquartile range, and parametric data as mean ± sd.
All baseline variables (weight, Pao2, Paco2, and pH) were comparable among groups in all series.
High-Dose Epinephrine (Series 1, Epinephrine 30 μg/kg)
Survival and ROSC
Survival after cardiac arrest was better with PEEP (Fio2 1.0 and 0.21) versus no PEEP (7/9 vs 6/6 vs 0/9, P < 0.01; Fig. 1). All animals attained initial ROSC. The time required for ROSC was similar with PEEP (5 cm H2O) versus no PEEP (33 ± 7 vs 36 ± 7 vs 46 ± 17 s; P = 0.08).
PEEP, Oxygenation, and Respiratory Mechanics
PEEP was associated with higher FRC (Fio2 1.0) (Fig. 2, Panel A). PEEP was also associated with smaller increases in peak airway pressure (Fig. 2, Panel B) after CPR and with lower wet/dry lung weight ratio and was not affected by level of inspired oxygen (Fig. 2, Panel C). The Pao2 achieved with 5 cm H2O PEEP (and Fio2 0.21) was comparable to that achieved with zero PEEP (and Fio2 1.0) and lower than that achieved with 5 cm H2O PEEP (and Fio2 1.0) (66 ± 10 and 89 ± 16 mm Hg vs 424 ± 84 mm Hg; P = 0.001; Fig. 2, Panel D). Arterial pH decreased immediately after cardiac arrest and ROSC and returned to near normal levels at 30 min with PEEP 5 cm H2O, Fio2 1.0 (7.23 ± 0.11) versus zero PEEP (7.1 ± 0.01, P = 0.14). Arterial base excess also decreased after cardiac arrest and ROSC with PEEP 5 cm H2O/Fio2 1.0 (−9.23 ± 3.4) versus zero PEEP (−14.5 ± 0.35; P = 0.86). Paco2 increased after arrest and ROSC and then remained slightly higher than normal in both groups. HCO3 decreased significantly after arrest and ROSC, with minimal recovery in both groups.
The left ventricular end-diastolic area (LVEDA) was similar in both groups at baseline and decreased in both groups after cardiac arrest, but to a lower level in the zero PEEP group (P < 0.05 versus baseline; Fig. 3, Panel A). The FS was similar at baseline, increased in both groups immediately after arrest, and decreased thereafter, remaining below baseline (P < 0.05 over time, both groups). There were no between-group differences in FS at any time (Fig. 3, Panel B).
Conventional-Dose Epinephrine (Series 2, Epinephrine 10 μg/kg)
The administration of PEEP did not affect outcome when 10 μg/kg of epinephrine was used during CPR after asphyxial cardiac arrest. All animals (5 cm H2O PEEP and zero PEEP) survived to 120 min after CPR.
The current study explores the mechanisms of PEEP-mediated effects during CPR after asphyxial cardiac arrest. After high-dose epinephrine, PEEP improved survival and diastolic function but did not alter systolic function or hemodynamics. Lung volumes were increased and pulmonary edema reduced. Although arterial oxygenation improved after administration of PEEP, improvements in survival were not mediated by this mechanism because the rank order of survival was 5 cm H2O PEEP (Fio2 1.0) ≈ 5 cm H2O PEEP (Fio2 0.21) > zero PEEP (Fio2 1.0), whereas the rank order of oxygenation (Pao2) was 5 cm H2O PEEP (Fio2 1.0) > 5 cm H2O PEEP (Fio2 0.21) ≈ zero PEEP (Fio2 1.0).
This is the first study, to our knowledge, to evaluate the effect of PEEP and high versus low doses of epinephrine in the setting of asphyxial cardiac arrest. PEEP has been studied in models of cerebral perfusion4 during ventricular fibrillation and using active compression-decompression CPR, in which it improved indices of organ perfusion but not survival.9 Constant flow insufflation during CPR also improves systolic arterial blood pressure and carotid flow, but not survival.10 Thus the current study suggests that PEEP may be fundamentally different from other ventilation adjuncts in CPR and might be explored as an experimental option in other cardiac arrest models.
Rationale for Epinephrine Use
Epinephrine is a mainstay of cardiovascular therapy in advanced cardiac life support.2 Higher doses of epinephrine, compared with lower doses, have been associated with improved initial resuscitation and survival to hospital admission but not to hospital discharge.11 There are significant concerns regarding the potential toxicity of epinephrine at higher doses, specifically the potential direct cause of cardiac dysfunction and death after initially successful resuscitation.11 These concerns are supported by laboratory studies in which epinephrine has been associated with myocyte injury and pulmonary edema, and death.5,6 These concerns have led to the study, and adoption, of a “pure” vasoconstrictor (e.g., vasopressin) as part of most CPR algorithms2; nonetheless, although alternatives are available, epinephrine remains the initial vasopressor of choice.2 IV epinephrine achieves a plasma peak and maximal cardiovascular effect within 2 min,12 but the majority of deaths secondary to the medication itself occur after several half lives, suggesting that death is not due to the ongoing effect of exogenous epinephrine but rather to injury incurred early in the resuscitation effort.12
Cardiopulmonary Interactions During CPR
Cardiac filling is fundamental to the generation of a cardiac output, and therefore to the success of resuscitation and subsequent cardiac performance. Cardiac arrest may be accompanied by atelectasis13 and an increase in PVR that is aggravated by conventional CPR techniques. Because the pulmonary and systemic circulations exist in series, a critical increase in PVR (made more significant by impaired flow) could reduce already impaired systemic perfusion, making salvage unlikely or impossible. Reversal of elevated PVR may therefore improve resuscitation potential.
PVR is affected by alterations in minute volume, airway pressure, lung volume, pH, and arterial oxygen content. In this study, we chose to investigate the effect of PEEP, which can potentially affect all of these variables. In the study, where epinephrine was used in “higher” doses (30 μg/kg), the PEEP improved survival and cardiac performance, increased FRC, and reduced pulmonary edema, which in turn reduced airway pressure (reflecting increase compliance), and increased arterial oxygen content and delivery. A survival benefit was not seen with epinephrine 10 μg/kg, however. We speculate that PEEP, by initially reducing venous return and right ventricular output, may protect against the previously demonstrated abrupt dose-related pulmonary venous hypertension and pulmonary edema associated with epinephrine dosing in experimental asphyxial cardiac arrest.6 We confirmed, by using PEEP with lower Fio2, that the beneficial effect was not due to the effect on oxygenation. Finally, it is possible that PEEP reduced left ventricular afterload as well as preload.
Ventilator Strategies During CPR
Suboptimal ventilator strategies may adversely affect the outcome of resuscitative efforts.14 However, the ideal ventilator settings (e.g., Vt, RR, Pao2, and airway pressure) during CPR are not known. There are complex interactions between intrathoracic pressures, right and left ventricular function, arterial oxygenation and carbon dioxide elimination, venous return, cardiac output, and microvascular flow.15 Although arterial oxygenation can be reasonably achieved by chest compressions without independent ventilation in early treatment of witnessed cardiac arrest (usually ventricular fibrillation), this may not be true of prolonged arrests or arrest due to asphyxia. Indeed, rapid improvements in tissue oxygenation are, in contrast to ventricular fibrillation, critically important in the early treatment of asphyxial cardiac arrest.16 Tissue hypoxia in this context may be due to the preceding event (e.g., airway obstruction), as well as reduced perfusion during CPR and pulmonary factors (e.g., pulmonary edema, hemorrhage, and atelectasis).
Respiratory system compliance has been shown to decrease after CPR.9 However, many clinicians have reservations regarding development of “auto-PEEP” during CPR, based on fears that cardiac output and mean arterial blood pressure could be adversely affected by increased intrathoracic pressure,4,17 particularly in the setting of hypovolemia.18 In the laboratory setting, improved cardiac filling and organ perfusion compared with conventional intermittent positive pressure ventilation, albeit at lower Vt than in the control group, have been convincingly demonstrated.19 In that study, lower airway pressures were associated with improved survival. Notably, animals in that study did not have any evidence of atelectasis, pulmonary hemorrhage, or pulmonary edema.19 In a porcine model of ventricular fibrillation, PEEP increased oxygenation, improved static lung compliance, and resulted in higher coronary perfusion and systolic aortic pressures when added to CPR using an automatic cardiac defibrillator.9
Limitations of the Study
This study has several potential limitations. First, the model investigated asphyxial cardiac arrest only, and the results cannot automatically be extrapolated to arrests from nonasphyxial causes, such as ventricular fibrillation, in which postresuscitation myocardial function is different from that after asphyxia.7 Second, the benefit was seen only in a brief cardiac arrest and when “high-dose” epinephrine was administered; this would limit extrapolation to only a small fraction of cardiac arrests. Third, the animals did not have preexisting cardiac or respiratory disease, both of which are common in patients who suffer hypoxic cardiac arrest, and thus postresuscitation organ dysfunction patterns may differ. Fourth, there are important interspecies differences in cardiovascular physiology (rodent heart rates are significantly faster than heart rates of humans, dogs, and pigs) as well as in lung and chest wall compliance and in sensitivity to positive airway pressure; indeed, the equivalent of 5 cm H2O of PEEP as used in this model has not been determined in humans. Last, the ventilation protocol applied during CPR was continued into the postresuscitation phase and until the end of the protocol, and therefore does not distinguish whether survival was improved by the intra- or post-CPR intervention.
Use of PEEP during CPR was associated with improved survival in a small animal model of asphyxial cardiac arrest in which high doses of epinephrine were used. The effects appeared primarily to be related to the mechanical effects of PEEP on the respiratory and cardiovascular systems and are not due to the increase in systemic oxygenation.
1.Gazmuri RJ, Nadkarni VM, Nolan JP, Arntz HR, Billi JE, Bossaert L, Deakin CD, Finn J, Hammill WW, Handley AJ, Hazinski MF, Hickey RW, Jacobs I, Jauch EC, Kloeck WG, Mattes MH, Montgomery WH, Morley P, Morrison LJ, Nichol G, O’Connor RE, Perlman J, Richmond S, Sayre M, Shuster M, Timerman S, Weil MH, Weisfeldt ML, Zaritsky A, Zideman DA. Scientific knowledge gaps and clinical research priorities for cardiopulmonary resuscitation and emergency cardiovascular care identified during the 2005 International Consensus Conference on ECC [corrected] and CPR science with treatment recommendations: a consensus statement from the International Liaison Committee on Resuscitation (American Heart Association, Australian Resuscitation Council, European Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa, and the New Zealand Resuscitation Council); the American Heart Association Emergency Cardiovascular Care Committee; the Stroke Council; and the Cardiovascular Nursing Council. Circulation 2007;116:2501–12
2.2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112:IV1–IV203
3.Levy MM. Optimal peep in ARDS. Changing concepts and current controversies. Crit Care Clin 2002;18:15–33, v–vi
4.Hodgkin BC, Lambrew CT, Lawrence FH III, Angelakos ET. Effects of PEEP and of increased frequency of ventilation during CPR. Crit Care Med 1980;8:123–6
5.McCaul CL, McNamara P, Englelberts D, Slorach C, Hornberger LK, Kavanagh BP. The effect of global hypoxia on myocardial function after successful cardiopulmonary resuscitation in a laboratory model. Resuscitation 2006;68:267–75
6.McCaul CL, McNamara PJ, Englelberts D, Wilson GJ, Romaschin A, Redington AN, Kavanagh BP. Epinephrine increases mortality after brief asphyxial cardiac arrest in an in vivo rat model. Anesth Analg 2006;102:542–8
7.Kamohara T, Weil MH, Tang W, Sun S, Yamaguchi H, Klouche K, Bisera J. A comparison of myocardial function after primary cardiac and primary asphyxial cardiac arrest. Am J Respir Crit Care Med 2001;164:1221–4
8.Duggan M, McCaul CL, McNamara PJ, Englelberts D, Ackerley C, Kavanagh BP. Atelectasis causes vascular leak and lethal right ventricular failure in uninjured rat lungs. Am J Respir Crit Care Med 2003;167:1633–40
9.Moser B, Voelckel W, Gardetto A, Sumann G, Wenzel V. One night in a snowbank: a case report of severe hypothermia and cardiac arrest. Resuscitation 2005;65:365–8
10.Brochard L, Boussignac G, Adnot S, Bertrand C, Isabey D, Harf A. Efficacy of cardiopulmonary resuscitation using intratracheal insufflation. Am J Respir Crit Care Med 1996;154:1323–9
11.Vandycke C, Martens P. High dose versus standard dose epinephrine in cardiac arrest—a meta-analysis. Resuscitation 2000;45:161–6
12.Mielke LL, Frank C, Lanzinger MJ, Wilhelm MG, Entholzner EK, Hargasser SR, Hipp RF. Plasma catecholamine levels following tracheal and intravenous epinephrine administration in swine. Resuscitation 1998;36:187–92
13.Markstaller K, Rudolph A, Karmrodt J, Gervais HW, Goetz R, Becher A, David M, Kempski OS, Kauczor HU, Dick WF, Eberle B. Effect of chest compressions only during experimental basic life support on alveolar collapse and recruitment. Resuscitation 2008;79:125–32
14.Aufderheide TP, Lurie KG. Death by hyperventilation: a common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med 2004;32:S345–S351
15.Pinsky MR. Heart-lung interactions. Curr Opin Crit Care 2007;13:528–31
16.Lopez-Herce J, Garcia C, Dominguez P, Carrillo A, Rodriguez-Nunez A, Calvo C, Delgado MA. Characteristics and outcome of cardiorespiratory arrest in children. Resuscitation 2004;63:311–20
17.Hevesi ZG, Thrush DN, Downs JB, Smith RA. Cardiopulmonary resuscitation: effect of CPAP on gas exchange during chest compressions. Anesthesiology 1999;90:1078–83
18.Ido Y, Goto H, Lavin MJ, Robinson JD, Mangold JV, Arakawa K. Effects of positive end-expiratory pressure on carotid blood flow during closed-chest cardiopulmonary resuscitation in dogs. Anesth Analg 1982;61:557–60
19.Srinivasan V, Nadkarni VM, Yannopoulos D, Marino BS, Sigurdsson G, McKnite SH, Zook M, Benditt DG, Lurie KG. Rapid induction of cerebral hypothermia is enhanced with active compression-decompression plus inspiratory impedance threshold device cardiopulmonary resuscitation in a porcine model of cardiac arrest. J Am Coll Cardiol 2006;47:835–41