Resuscitation 56 (2003) 9 /13 www.elsevier.com/locate/resuscitation
Induced hypothermia using large volume, ice-cold intravenous fluid in comatose survivors of out-of-hospital cardiac arrest: a preliminary report Stephen Bernard a,b, , Michael Buist a, Orlando Monteiro a, Karen Smith b a
b
The Intensive Care Unit, Dandenong Hospital, David St, Dandenong, Victoria 3175, Australia Department of Epidemiology and Preventive Medicine, Monash University, St Kilda Rd, Prahran 3181, Victoria, Australia Received 1 July 2002; accepted 26 July 2002
Abstract Study hypothesis: Recent studies have shown that induced hypothermia for twelve to twenty four hours improves outcome in patients who are resuscitated from out-of-hospital cardiac arrest. These studies used surface cooling, but this technique provided for relatively slow decreases in core temperature. Results from animal models suggest that further improvements in outcome may be possible if hypothermia is induced earlier after resuscitation from cardiac arrest. We hypothesized that a rapid infusion of large volume (30 ml/kg), ice-cold (4 8C) intravenous fluid would be a safe, rapid and inexpensive technique to induce mild hypothermia in comatose survivors of out-of-hospital cardiac arrest. Methods: We enrolled 22 patients who were comatose following resuscitation from out-of-hospital cardiac arrest. After initial evaluation in the Emergency Department (ED), a large volume (30 ml/kg) of icecold (4 8C) lactated Ringers solution was infused intravenously over 30 min. Data on vital signs, arterial blood gas, electrolyte and hematological was collected immediately before and after the infusion. Results: The rapid infusion of large volume, ice-cold crystalloid fluid resulted in a significant decrease in median core temperature from 35.5 to 33.8 8C. There were also significant improvements in mean arterial blood pressure, renal function and acid /base analysis. No patient developed pulmonary odema. Conclusion: A rapid infusion of large volume, ice-cold crystalloid fluid is an inexpensive and effective method of inducing mild hypothermia in comatose survivors of out-of-hospital cardiac arrest, and is associated with beneficial haemodynamic, renal and acid /base effects. Further studies of this technique are warranted. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cardiac arrest; Induced hypothermia; Intravenous fluid
Resumo Hipo´tesis de estudio : Estudios recientes han demostrado la hipotermia inducida por doce a veinticuatro horas mejora el resultado en pacientes que han sido resucitados de un paro cardı´aco extrahospitalario. Estos estudios usaron enfriamiento superficial, pero esta te´cnica produce disminuciones relativamente lentas en la temperatura central. Los resultados en los modelos animales sugieren que pueden ser posibles mayores mejorı´as en el resultado si la hipotermia es inducida despue´s de la resucitacio´n del paro cardı´aco. Planteamos la hipo´tesis que la ra´pida infusio´n de grandes volu´menes (30 ml/kg), de fluidos intravenosos helados (4 8C) serı´a una te´cnica segura, ra´pida y poco costosa para inducir hipotermia leve en sobrevivientes comatosos de paro cardı´aco extrahospitalario. Me´todos : Enrolamos 22 pacientes que estaban comatosos despue´s de ser resucitados de un paro cardı´aco extrahospitalario. Despue´s de la evaluacio´n inicial en el departamento de emergencias (ED), se les infundio´ en 30 minutos un gran volumen (30 ml/kg) de solucio´n Ringer lactato helado (4 8C).Se recogieron datos de signos vitales, gases en sangre arterial, electrolitos y hematologı´a inmediatamente antes y despue´s de la infusio´n. Resultados : La ra´pida infusio´n de grandes volu´menes de fluidos cristaloides helados produjo la ra´pida disminucio´n en la temperatura central de 35.5 a 33.8 8C. Tambie´n se produjeron mejorı´as significativas en la presio´n arterial media, funcio´n renal y ana´lisis a´cido base. Ningu´n paciente desarrollo´ edema pulmonar. Conclusio´n : Una ra´pida
Corresponding author. Fax: /61-3-9554-8378 E-mail address:
[email protected] (S. Bernard). 0300-9572/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 9 5 7 2 ( 0 2 ) 0 0 2 7 6 - 9
10
S. Bernard et al. / Resuscitation 56 (2003) 9 /13
infusio´n de grandes volu´menes de fluidos cristaloides helados es un medio ra´pido y barato para inducir hipotermia leve en los sobrevivientes comatosos de un paro cardı´aco extrahospitalario, y esta´ asociado con efectos hemodina´micas, renales y a´cido base beneficiosos. Se garantizan estudios ulteriores de esta te´cnica. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Palavras chave: Paro cardı´aco; Hipotermia inducida; Fluidos intravenosos
Resumen Hipo´tesis de estudio : Estudios recientes han demostrado la hipotermia inducida por doce a veinticuatro horas mejora el resultado en pacientes que han sido resucitados de un paro cardı´aco extrahospitalario. Estos estudios usaron enfriamiento superficial, pero esta te´cnica produce disminuciones relativamente lentas en la temperatura central. Los resultados en los modelos animales sugieren que pueden ser posibles mayores mejorı´as en el resultado si la hipotermia es inducida despue´s de la resucitacio´n del paro cardı´aco. Planteamos la hipo´tesis que la ra´pida infusio´n de grandes volu´menes (30 ml/kg), de fluidos intravenosos helados (4 8C) serı´a una te´cnica segura, ra´pida y poco costosa para inducir hipotermia leve en sobrevivientes comatosos de paro cardı´aco extrahospitalario. Me´todos : Enrolamos 22 pacientes que estaban comatosos despue´s de ser resucitados de un paro cardı´aco extrahospitalario. Despue´s de la evaluacio´n inicial en el departamento de emergencias (ED), se les infundio´ en 30 minutos un gran volumen (30ml/kg) de solucio´n Ringer lactato helado (4 8C). Se recogieron datos de signos vitales, gases en sangre arterial, electrolitos y hematologı´a inmediatamente antes y despue´s de la infusio´n. Resultados : La ra´pida infusio´n de grandes volu´menes de fluidos cristaloides helados produjo la ra´pida disminucio´n en la temperatura central de 35.5 a 33.8 8C. Tambie´n se produjeron mejorı´as significativas en la presio´n arterial media, funcio´n renal y ana´lisis a´cido base. Ningu´n paciente desarrollo´ edema pulmonar. Conclusio´n : Una ra´pida infusio´n de grandes volu´menes de fluidos cristaloides helados es un medio ra´pido y barato para inducir hipotermia leve en los sobrevivientes comatosos de un paro cardı´aco extrahospitalario, y esta´ asociado con efectos hemodina´micas, renales y a´cido base beneficiosos. Se garantizan estudios ulteriores de esta te´cnica. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Palabras clave: Paro cardı´aco; Hipotermia inducida; Fluidos intravenosos
1. Introduction Out-of-hospital cardiac arrest is common and associated with a poor prognosis, with less than 3% of patients surviving to good outcome at hospital discharge [1]. Patients who are initially successfully resuscitated by emergency medical services often remain unconscious because of severe anoxic brain injury [2]. Recently, clinical trials of induced hypothermia (IH) have shown improved outcome in comatose survivors of out-ofhospital cardiac arrest [3 /5]. These studies used surface cooling with ice packs [3,4] or forced cold air cooling [5], however, these techniques of surface cooling had significant limitations. Firstly, these are relatively slow methods of decreasing core temperature, with a 0.9 8C/ h decrease in core temperature using ice packs [3,4] and 0.3 8C/h using forced cold air cooling [4]. Secondly, the covering of the patient with ice packs or cooling blankets during resuscitation is inconvenient for medical and nursing staff. Finally, the use of ice packs or refrigerated units (for forced air cooling) limit the use of these techniques to the hospital environment. Since there is evidence from animal studies that outcome is improved if cooling is initiated during resuscitation or immediately after return of spontaneous circulation [6,7], the development of a technique to induce hypothermia which is feasible in the out-ofhospital setting may further improve outcome. We hypothesized that a rapid intravenous infusion of large
volume (30 ml/kg), ice-cold (4 8C) lactated Ringers solution would be an effective alternative to surface cooling for the induction of mild hypothermia. We report our preliminary experience with this technique.
2. Materials and methods The study was undertaken in the Emergency Department (ED) of Dandenong Hospital in Melbourne, Victoria, Australia. Patients were eligible for inclusion in the study if they arrived unconscious at the ED after resuscitation by paramedics from out-of-hospital cardiac arrest. Exclusion criteria included children (B/18 years), possible pregnancy (femalesB/50 years), coma possibly due to cerebrovascular accident or head trauma, cardiogenic shock (systolic blood pressureB/ 90 mmHg despite adrenaline (epinephrine) infusion, clinical signs of pulmonary oedema, patients with a diagnosed terminal illness and spontaneous hypothermia (B/34 8C at initial measurement). Date on the time from collapse to ambulance call, call to ambulance arrival, and ambulance arrival to return of spontaneous circulation and initial cardiac rhythm was recorded from the ambulance Patient Care Record. Routine management in the ED included mechanical ventilation with 100% oxygen, resuscitation of the circulation and maintenance of a mean arterial blood pressure of 90 /100 mmHg using either adrenaline or
S. Bernard et al. / Resuscitation 56 (2003) 9 /13
glyceryl tri-nitrate infusions as required. Following assessment of neurological status and confirmation of coma, vecuronium 20 mg was administered intravenously. Midazolam 1/5 mg was also administered intravenously. Monitoring included continuous electrocardiogram and invasive blood pressure. A central venous catheter was inserted if thrombolytic therapy was not indicated. A bladder catheter incorporating a temperature probe was inserted. Routine investigations included a 12 lead electrocardiogram, chest X-ray, electrolyte, haematology, cardiac enzymes, and arterial blood gas analysis. Immediately following this initial evaluation and the drawing of blood for laboratory analysis, an intravenous infusion of 30 ml/kg lactated Ringers solution was administered over 30 min via either a peripheral cannula or femoral venous catheter, using a pressure bag and standard intravenous infusion equipment. The lactated Ringers solution was stored at a controlled temperature (4 8C) in the ED blood refrigerator until required. At the conclusion of the infusion, vital signs were recorded and blood sent for arterial blood gas, electrolyte and haematological analysis. Arterial blood gases were corrected for temperature. After this data collection, ice packs were placed on the patient to decrease core temperature to 33 8C. Patients were admitted to the Intensive Care Unit (ICU), where hypothermia at 33 8C was maintained for 12 h, with active re-warming over 6 h, following a previously described protocol [3,4]. Statistical analysis was undertaken using the commercial statistical package Stata [8]. Differences between the pre-fluid and post-fluid variables were analyzed using the Wilcoxon signed-rank test. A value of P B/ 0.05 was considered significant. The study was approved by the Institutional Ethics Committee. Informed, written consent was obtained from the next-of-kin after arrival at hospital.
11
Table 1 Patient characteristics Number of patients Age Cause of cardiac arrest
Initial cardiac rhythm
Collapse to call EMS ,# Call EMS to arrival ,# Arrival EMS to ROSC ,# Total collapse to ROSC ,# ROSC to infusion commenced ,# Survival (initial rhythm VF) Survival (initial rhythm not VF)
22 70 (57, 75) Presumed cardiac (21) Overdose of tricyclic antidepressant (1) Ventricular fibrillation (14) Asystole (4) Pulseless electrical activity (4) 2.0 (1.0, 3.8) 8.0 (6.0, 10.0) 16.0 (7.8, 21.0) 26.0 (20.3, 34.5) 73.0 (50.3, 99.3) 8/14 2/8
Values given as median (interquartile range). # Time in minutes. VF, ventricular fibrillation; ROSC, return of spontaneous circulation; EMS, emergency medical service.
There were no adverse effects of the rapid infusion of this volume of intravenous crystalloid fluid. In particular, although central venous pressure increased and the pO2 decreased, no patient developed pulmonary oedema on either clinical or chest X-ray criteria. There were 14 patients with an initial rhythm of ventricular fibrillation of whom eight (57%) were discharged alive from hospital. Two of eight patients with an initial cardiac rhythm other than ventricular fibrillation also survived to hospital discharge. All deaths occurred subsequent to re-warming in the setting of severe neurological injury between 18 and 168 h and these were considered to be unrelated to the initial fluid therapy.
4. Discussion 3. Results There were 22 patients enrolled in the study between March 2000 and May 2002. The demographics of the patients are shown in Table 1. The mean total collapse to return of spontaneous circulation time of 27 min (range 12/48 min) is consistent with our previously published clinical studies of induced hypothermia [3,4]. The cardiac arrests were witnessed in all 14 patients with an initial cardiac rhythm of ventricular fibrillation and four of the remaining eight patients. The vital signs and laboratory results before and after the infusion of cold fluid are shown in Table 2. In addition to the decrease in core temperature, there were significant improvements in mean arterial blood pressure, renal function and acid /base measurements.
Recent clinical studies suggest that IH improves outcome in patients with anoxic brain injury following resuscitation from out-of-hospital cardiac arrest and is not associated with adverse side-effects [3 /5]. In a preliminary study, we used IH in 22 adult patients with coma following resuscitation from pre-hospital cardiac arrest [3]. Hypothermia (33 8C) was induced in the ED using surface cooling with ice packs and maintained for 12 h in the ICU. Compared with 22 historical controls, outcome was significantly improved and there were no additional complications. In a subsequent randomized, controlled trial, we compared IH in 43 patients with normothermia in 34 patients [4]. Hypothermia was also induced in the ED using surface cooling with ice packs and maintained for 12 h in the ICU. Outcome at hospital discharge was
12
S. Bernard et al. / Resuscitation 56 (2003) 9 /13
Table 2 Vital signs and laboratory investigations immediately pre and post fluid infusion
Temperature, 8C (bladder) MAP, mmHg Pulse/min CVP, mmHg pH PO2, mmHg PCO2, mmHg Bicarbonate, mmol/l Sodium, mmol/l Potassium, mmol/l Creatinine, mmol/l Hematocrit, %
Pre fluid Median (IQR)
Post fluid Median (IQR)
Difference (95% CI)
P value
35.5 (35.1, 35.9) 83 (74, 100) 86 (66, 106) 7 (5, 10) 7.28 (7.18, 7.34) 386 (217, 454) 39 (33, 42) 18 (16, 21) 137 (136, 141) 4.0 (3.6, 4.3) 130 (116, 150) 0.38 (0.36, 0.43)
33.8 (33.4, 34.4) 103 (82, 113) 80.5 (67, 93) 12 (8, 15) 7.30 (7.25, 7.40) 187 (89, 391) 37 (32, 41) 21 (16, 22) 138 (137, 141) 4.0 (3.3, 4.2) 105 (88, 125) 0.38 (0.33, 0.41)
1.6 ( 2.0, 1.1) 10 (2, 14) 5 ( 15, 4) 4 ( 1.0, 10) 0.04 (0.01, 0.10) 79 ( 257, 34) 3 ( 8, 0) 1.0 (0, 3) 0 ( 2, 1.0) 0.1 ( 0.4, 0.3) 20 ( 36, 13) 0.01 ( 0.02, 0.01)
B 0.001 0.012 0.091 0.050 0.014 0.025 0.049 0.009 0.586 0.986 0.002 0.134
Median difference, confidence intervals calculated using the binomial exact method. IQR, interquartile range; CI, confidence intervals.
significantly improved, with 21/43 (49%) of the hypothermia patients discharged to home or rehabilitation compared with 9/34 (26%) of the normothermic patients. After correction for baseline differences, the odds ratio for improved outcome using hypothermia was 5.25 (95% confidence intervals 1.47 /18.76). Surface cooling using ice-packs in both studies was associated with a rate of core temperature decrease of 0.9 8C/h. In a multi-center European study, 273 patients who were resuscitated from out-of-hospital cardiac arrest were randomized to hypothermia (32 /34 8C for 24 h) using surface cooling with a cold air blanket [5]. The outcome at six months after hospital discharge in the IH group was significantly improved, with 75/136 (55%) of the hypothermia patients having a favorable neurological outcome compared with 54/137 (39%) of the normothermic patients. However, surface cooling using a cold air blanket was associated with a temperature decrease of only 0.3 8C/h. Other studies of surface cooling in surgical patients [9,10] have confirmed that surface cooling is a relatively slow technique to decrease core temperature. Since there is evidence in animal studies of significantly reduced anoxic neurological injury if hypothermia is induced immediately after cardiac arrest [6,7], the development of techniques for the rapid induction of hypothermia may further improve neurological outcome after resuscitation from cardiac arrest. One alternative to surface cooling is the use of a large volume of ice-cold intravenous crystalloid fluid. This has been studied in elective surgical patients [11] and healthy volunteers [12,13]. In a study of elective surgical patients, Baumgardner et al., administered small volume (5 ml/kg) of 5% cold (1 /4 8C) albumin solution intravenously [11]. Core temperature decreased by 0.6 8C (9/0.1 8C) following rapid (/100 ml/min) infusion, but this decrease was less (0.4 8C) following slow intravenous
infusion. The effect of large volume, ice-cold fluid intravenous infusion for the induction of moderate hypothermia has also been studied in anaesthetized volunteers by Rajek et al. who infused 40 ml/kg 0.9% saline solution over 30 min in nine subjects [12]. Using 4 8C fluid, the mean decrease in core temperature was 2.5 8C (9/0.4), which was 0.4 8C more than expected if the fluid was distributed evenly. In a similar study, Frank et al. examined age related differences in core cooling using large volume (40 ml/kg) cold intravenous fluid in eight young (18 /23 years) and eight older volunteers (55 /71 years) [13]. The older subjects had a greater decrease in core temperature at the end of the infusion (35.0 8C compared with 35.9 8C), possibly because of decreased thresholds for vasoconstriction, heat production and catecholamine release. However, the results of the above studies undertaken in elective surgical patients or healthy volunteers may not be applicable to patients who have been resuscitated from prolonged cardiac arrest with possible severe myocardial dysfunction. The infusion of a large volume of intravenous fluid in such patients would be expected to cause pulmonary oedema, however, this did not occur in any patient in this study. Our preliminary report suggests that mild hypothermia may be rapidly and safely induced using 30 ml/kg of ice-cold fluid in patients who have been resuscitated from out-of-hospital cardiac arrest. The decrease in core temperature of 1.6 8C is likely to provide significant cerebral protection if administered shortly after resuscitation from cardiac arrest [3,4]. In addition, the increase in mean arterial blood pressure might also be beneficial in patients with anoxic encephalopathy [14]. The significant decrease in systemic oxygen pressure is not clinically important, since this decrease would lead to a minimal change in oxygen content of arterial blood. The small patient numbers in this preliminary report pre-
S. Bernard et al. / Resuscitation 56 (2003) 9 /13
clude an assessment of the effect of this therapy on outcome, however the finding that 8/14 patients (57%) with an initial rhythm of ventricular fibrillation were discharged home from hospital is encouraging. This study has a number of limitations. There was no control group to compare changes in core temperature, blood pressure, renal function and acid /base if largevolume, ice-cold fluid was not given. However, in previous studies [3,4], we had not observed spontaneous decreases in temperature without active cooling efforts. Whilst improvements in blood pressure, renal function and acid /base may occur with recovery of the circulation and oxygenation, we have not previously noted such rapid changes occurring during the time of this study (mean 73 min after return of the circulation). The small sample size of this study precludes any assessment of this technique on outcome. Nevertheless, we consider the results sufficiently encouraging and propose that further studies be undertaken in the ED and out-of-hospital setting. For the latter, paramedics could administer large volume, ice-cold fluid immediately after resuscitation of patients from cardiac arrest and, if this is shown to be feasible and safe, further studies comparing this approach with surface cooling in the ED could be undertaken.
References [1] Bernard SA. Outcome from prehospital cardiac arrest in Melbourne, Australia. Emerg Med 1998;10:25 /9. [2] Edgren E, Hedstrand U, Kelsey S, et al. Assessment of neurological prognosis of comatose survivors of cardiac arrest. Lancet 1994;343:1055 /9.
13
[3] Bernard SA, Jones BM, Horne MK. A clinical trial of induced hypothermia in comatose survivors of prehospital cardiac arrest. Ann Emerg Med 1997;30:146 /53. [4] Bernard SA, Gray T, Buist MD, Jones BM, Silvester W, Gutteridge G, Smith K. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med 2002;346:557 /63. [5] The Hypothermia After Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurological outcome after cardiac arrest. N Engl J Med 2002; 346:549 /556. [6] Stertz F, Safar P, Tisherman SA, et al. Mild hypothermic cardiopulmonary resuscitation improves outcome after cardiac arrest in dogs. Crit Care Med 1991;19:379 /89. [7] Kuboyama K, Safar P, Radovsky A, Tisherman SA, Stezoski SW, Alexander H. Delay in cooling negates the beneficial effect of mild resuscitative cerebral hypothermia after cardiac arrest in dogs: A prospective, randomized study. Crit Care Med 1993;21:1348 /58. [8] StataCorp. Stata statistical software. In: Release 6, College Station, Stata Corporation; 1999. [9] Plattner O, Kurz A, Sessler DI, Ikeda T, Christensen R, Marder D, Clough D. Efficacy of intraoperative cooling methods. Anesthesiology 1997;87:1089 /95. [10] Theard MA, Tempelhoff R, Crowder RM, Cheng MA, Todorov A, Dacey RG. Convection versus conduction cooling for induction of mild hypothermia during neurovascular procedures in adults. J Neurosurg Anesthesiol 1997;9:250 /5. [11] Baumgardner JE, Baranov D, Smith DS, Zager EL. The effectiveness of rapidly infused intravenous fluids for inducing moderate hypothermia in neurosurgical patients. Anesth Analg 1999;89:163 /9. [12] Rajek A, Grief R, Sessler DI, Baumgardner J, Laciny S, Bastanmehr H. Core cooling by central venous infusion of icecold (4 degrees Celcius or 20 degrees Celcius) fluid: isolation of core and peripheral thermal compartments. Anaesthesiology 2000;93:629 /37. [13] Frank SM, Raja SN, Bulcao C, Goldstein DS. Age-related thermoregulatory differences in core cooling in humans. Am J Physiol Integr Comp Physiol 2000;279:349 /54. [14] Safar P, Kochanek P. Cerebral blood flow promotion after prolonged cardiac arrest. Crit Care Med 2000;28:3104 /6.
