Resuscitation 51 (2001) 309–315

Sodium bicarbonate improves the chance of resuscitation after 10minutes of cardiac arrest in dogs

Elaine C.M. Leong, Jason C. Bendall, Anita C. Boyd, Rosemarie Einstein *Department of Pharmacology, Uni�ersity of Sydney, Sydney, NSW 2006, Australia

Received 19 April 2001; received in revised form 13 June 2001; accepted 13 June 2001

Abstract

The likelihood of successful defibrillation and resuscitation decreases as the duration of cardiac arrest increases. Prolongedcardiac arrest is also associated with the development of acidosis. These experiments were designed to determine whetheradministration of sodium bicarbonate and/or adrenaline in combination with a brief period of cardiopulmonary resuscitation(CPR) prior to defibrillation would improve the outcome of prolonged cardiac arrest in dogs. Ventricular fibrillation (VF) wasinduced by a.c. shock in anaesthetised dogs. After 10 min of VF, animals received either immediate defibrillation (followed bytreatment with bicarbonate or control) or immediate treatment with bicarbonate or saline (followed by defibrillation). Treatmentwith bicarbonate was associated with increased rates of restoration of spontaneous circulation. This was achieved with fewershocks and in a shorter time. Coronary perfusion pressure was significantly higher in NaHCO3-treated animals than in controlanimals. There were smaller decreases in venous pH in NaHCO3-treated animals than in controls. The best outcome in this studywas achieved when defibrillation was delayed for approximately 2 min, during which time NaHCO3 and adrenaline wereadministered with CPR. The results of the present study indicate that in prolonged arrests bicarbonate therapy and a period ofperfusion prior to defibrillation may increase survival. © 2001 Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Ventricular fibrillation; Defibrillation; Acidosis; Bicarbonate; Cardiac arrest; Resuscitation

www.elsevier.com/locate/resuscitation

1. Introduction

In cardiac arrests that occur out-of-hospital, it is notuncommon for a significant time to elapse betweendiscovery of the cardiac arrest and arrival ofparamedics. Approximately two-thirds of these patientspresent with the potentially reversible rhythm of ven-tricular fibrillation (VF) [1]. The likelihood of successfuldefibrillation and resuscitation decreases with durationof VF [2]. Although early defibrillation is associatedwith a good prognosis following brief arrest, immediatedefibrillation in prolonged arrest commonly results inasystole or electromechanical dissociation, both ofwhich are associated with poor outcome [1,3].

The inevitable development of metabolic and respira-tory acidosis that has been demonstrated in studies inanimals [4] and humans [5] is one of the key physiolog-ical differences between brief and prolonged cardiac

arrest and could contribute to different patient out-comes. Acidosis influences a variety of cellular electro-physiological parameters including resting membranepotential, threshold potential and conduction velocities.These effects are likely to contribute to the pro-arrhyth-mic influence of acidosis [6]. In addition, hypoxia andacidosis produce reversible myocardial depression [1,7],the major cause being a decrease in the sensitivity ofcontractile proteins to calcium [6]. The systemic re-sponse to vasopressors may be decreased in the pres-ence of acidosis [4,8] and this may also affect theoutcome of resuscitation.

In experiments in animals, Vukmir et al. [4] showedthat, after prolonged (15 min) arrest, there was animproved survival rate in animals that were treatedwith sodium bicarbonate and adrenaline. In these stud-ies, however, sodium bicarbonate and adrenaline wereonly administered after immediate defibrillation hadbeen unsuccessful. Other workers have suggested that abrief period of myocardial perfusion before counter-shock improves cardiac resuscitation outcome fromcardiac arrest in dogs [9] and humans [10].

* Corresponding author.E-mail address: einstein@med.usyd.edu.au (R. Einstein).

0300-9572/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved.PII: S 0 3 0 0 -9572 (01 )00421 -X

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315310

Myocardial damage is associated with the cumulativeelectrical energy of defibrillation shocks [11,12]. There-fore, any procedure that decreases the number of defi-brillation attempts needed to restore circulation wouldbe clinically desirable. The experiments described in thisreport were designed to determine whether administra-tion of sodium bicarbonate and/or adrenaline in combi-nation with a brief period of cardiopulmonaryresuscitation (CPR) prior to delivery of the first countershock would improve the chance of successful defibril-lation after 10 min of cardiac arrest in dogs.

2. Material and methods

The experiments were conducted in 24 adult mongreldogs (16–23 kg). The University of Sydney AnimalCare and Ethics Committee approved the procedures.Animals were anaesthetised, intubated and allowed tobreathe spontaneously. Anaesthesia induction was withsodium thiopentone (17 mg/kg intravenously (i.v.)) andmaintained with halothane (1–2%) in 100% 02. Electro-cardiography was monitored continuously (79DPolygraph; Grass Medical Instruments) via subcuta-neous electrodes. Femoral veins and arteries were can-nulated bilaterally for drug administration, bloodsampling and continuous blood pressure recording(Statham P23AC transducer, 79D Polygraph; GrassMedical Instruments). Central venous pressure wasrecorded continuously (Statham P23D transducer, 79DPolygraph; Grass Medical Instruments), via a catheterin the right external jugular vein, advanced to the rightatrium. The coronary perfusion pressure (CoPP) wascalculated as the difference between aortic diastolicpressure and the right atrial pressure. Anaesthesia wasdiscontinued prior to the initiation of cardiac arrest.The animals breathed 100% O2 for 1 min and thenroom air until the corneal reflex could be elicited. VFwas then induced by applying a 110V a.c. shock (50Hz) to transthoracic subcutaneous chest electrodes for2 s, repeated if required. Induction of arrest was fol-lowed by a non-intervention time of 10 min. Animalswere then randomly assigned to receive CPR and drugtreatment prior to defibrillation (n=12) or immediatedefibrillation followed by drug treatment (n=12).

Closed-chest CPR was performed continuously witha Thumper® Cardio-Pulmonary Resuscitator Model1005 (Michigan Instruments). The compression padwas placed on the lateral chest wall of the animal anda force (30–40 kg) was applied (approximately 5 cmcompression depth) at 60 compressions/min, with a50% duty cycle and a compression:ventilation (100%oxygen) ratio of 5:1. Ventilation was provided at amaximum airway pressure of 25 mmHg. Restoration ofspontaneous circulation (ROSC) was defined as a sys-tolic blood pressure of 60 mmHg, sustained for a

minimum of 10 min. Survival was defined as ROSC for30 min.

2.1. Immediate treatment

Following 10 min of VF, animals were randomlyassigned to receive adrenaline (0.1 mg/kg i.v.) in combi-nation with either sodium bicarbonate (2 mmol/kg i.v.)or NaCl (2 ml/kg, 0.9% i.v.) at the commencement ofthe resuscitation procedure. After drug administrationand 2 min of CPR, a 200 J defibrillation shock wasdelivered (LIFEPAK 10 Monitor Defibrillator; Physio-Control Corporation). If VF was not terminated bydefibrillation, lignocaine (2 mg/kg i.v.) was adminis-tered and defibrillation repeated (360 J). If ROSC didnot occur, a further dose of the treatment (eithersodium bicarbonate (1 mmol/kg i.v.) or NaCl (1 ml/kg,0.9% i.v.)) was administered. Resuscitation continuedwith Advanced Life Support (ALS) if required. Admin-istration of sodium bicarbonate or sodium chloride wasblinded during the resuscitation period.

2.2. Immediate defibrillation

Following 10 min of VF, resuscitation commencedwith delivery of a 200 J defibrillation shock. If ROSCdid not occur, animals were randomly assigned toreceive sodium bicarbonate (2 mmol/kg i.v.) andadrenaline (0.1 mg/kg i.v.) or adrenaline (0.1 mg/kgi.v.) alone. CPR was performed for 60 s followed by a360 J defibrillation shock. If ROSC was not attained, afurther dose of NaHCO3 (1 mmol/kg) was administeredto the sodium bicarbonate group. If ROSC had notoccurred, resuscitation continued with standard ALS.

2.3. Ad�anced Life Support

CPR was performed continuously, except duringdefibrillation. ALS comprised of adrenaline (0.1 mg/kg)administered every 5 min for the duration of the arrest.Further defibrillation shocks (360 J) were delivered toterminate VF as required. Lignocaine (2 mg/kg) wasadministered for VF refractory to treatment and/oradrenaline and defibrillation. This dose was repeatedonce if required. Atropine (0.04 mg/kg) was adminis-tered to animals with pulseless, organized rhythms orasystole. Prior to defibrillation all drugs were circulatedfor approximately 60 s using mechanical CPR. Resusci-tation was continued until ROSC occurred or for atotal resuscitation period of 30 min.

CPR was discontinued when a perfusing rhythm wasobtained. Lignocaine (2 mg/min) and adrenaline infu-sions (as required) were administered following ROSCto treat ventricular arrhythmias and maintain systolicblood pressure �80 mmHg, respectively.

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315 311

Blood samples were taken at pre-arrest and at 5 minafter initiation of resuscitation and analysed for pH andPCO2

(Radiometer ABL, System 625).

2.4. Statistical analysis

All results are expressed as mean�standard error ofthe mean. The outcome of resuscitation (ROSC andsurvival) was compared by Fisher’s exact test. Thenumber of defibrillation attempts and the coronaryperfusion pressures were compared with the Mann–Whitney U-test and unpaired Student t-test (two-tailed), respectively. Blood gas analysis variables (pHand PCO2

) were compared by repeated-measures analy-sis of variance. Differences were considered to be sig-nificant when P�0.05.

3. Results

Control values for the immediate treatment and im-mediate defibrillation groups are shown in Table 1.With the exception of resting heart rates, where theNaHCO3 immediate defibrillation group had a signifi-cantly lower rate, there were no statistically significantdifferences in pre-arrest variables between or withingroups. Data from one animal was not included in theresults due to technical difficulties during resuscitation.Ventricular fibrillation was successfully induced in allanimals with a.c. shocks. All animals remained in VFfor the 10 min non-intervention period.

Both ROSC and survival rate were higher inNaHCO3-treated animals than in control animals. Thisdifference was significant in the immediate treatmentgroup (P�0.05); however, it just failed to reach statis-tical significance in the immediate defibrillation group(P=0.06) (Table 2). Furthermore, for animals thatwere resuscitated, NaHCO3-treated animals attainedROSC in a shorter time than control animals (Table 2).

NaHCO3-treated animals also required significantlyfewer defibrillation shocks during the resuscitation pe-riod than control animals (P�0.05) (Table 2). For theanimals in which ROSC was achieved, the averagenumber of shocks required in bicarbonate-treated ani-mals was 1.8. The immediate defibrillation and immedi-ate treatment control animals that achieved ROSCrequired four and five shocks respectively. Coronaryperfusion pressure (measured 5 min after commence-ment of resuscitation or, for those animals achievingROSC, immediately prior to ROSC) was significantlyhigher in NaHCO3-treated animals than in control ani-mals (P�0.05) (Table 2).

At 5 min after resuscitation was commenced, thevenous pH in control animals was significantly less thanpre-arrest values (P�0.05) (Table 3). There weresmaller decreases in venous pH in NaHCO3-treatedanimals so that, at 5 min, the pH was not significantlydifferent from pre-arrest values (Table 3). Venous PCO2

of control animals was not significantly different after 5min of resuscitation compared with pre-arrest values(Table 3). However, in the NaHCO3-treated animals,venous PCO2

increased significantly (P�0.05) in thisperiod (Table 3).

4. Discussion

Sudden cardiac death is most commonly associatedwith ventricular fibrillation (VF). The majority of theseevents occur out-of-hospital [13,14] and therefore thereis a significant interval while awaiting the arrival ofemergency medical services. Current resuscitationguidelines recommend immediate defibrillation as thetreatment of choice for ventricular fibrillation [15]. Themajor determinant of resuscitation success is the time todefibrillation [16], so that prompt defibrillation substan-tially improves the likelihood of success in animals [2]and survival in humans following out-of-hospital car-diac arrest [17].

Table 1Pre-arrest values for haemodynamic parameters and blood variables in dogs assigned to receive immediate treatment or immediate defibrillationafter 10 min of cardiac arresta

Immediate defibrillationVariable Immediate treatment

NaHCO3Control (n=6)NaHCO3 (n=6)Control (n=6)(n=5)

124�7MAP (mmHg) 126�6 122�5 114�9142�11Heart rate (b.p.m.) 163�17 162�6 137�2*7.4�0.0 7.5�0.0Arterial pH 7.4�0.1 7.3�0.0

34�135�632�4PaCO2(mmHg) 27�2

146�1 143�1Sodium (mmol/l) 146�1 143�3Potassium (mmol/l) 4.3�0.1 4.0�0.1 4.3�0.3 4.5�0.2Calcium (mmol/l) 1.1�0.04 1.1�0.04 1.1�0.04 1.1�0.07

a Within these groups, animals were randomly assigned to either control or sodium bicarbonate treatment groups. *P�0.05 compared withrespective control.

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315312

Table 2Outcome of resuscitation in dogs after immediate treatment or immediate defibrillationa

Variable Immediate defibrillationImmediate treatment

NaHCO3 (n=6) Control (n=6)Control (n=6) NaHCO3 (n=5)

ROSC 1 5* 1 47�1 30 5�2Time to ROSC (min) 144* 10 4Survival

1.7�0.3* 5.5�0.5Defibrillation attempts 2.4�0.8*4.5�0.823�6* 8�19�2 29�3*CoPP (mmHg)

a Within these groups, animals were randomly assigned to either control or sodium bicarbonate treatment groups. * P�0.05 compared withrespective control.

Vukmir et al. [4] demonstrated significantly higherROSC and survival benefit with immediate defibrilla-tion followed by bicarbonate administration in pro-longed (15 min) arrest, although bicarbonate treatmentdid not influence the outcome following brief (5 min)arrest. The results of the present study with 10 minnon-intervention confirm those of Vukmir et al. [4] anddemonstrate increases in ROSC and survival in animalstreated with bicarbonate. The same trend was seen inimmediate defibrillation group treated with bicarbon-ate; however, the failure to reach statistical significancecould have been associated with the small group size.Treatment of animals with sodium bicarbonate andadrenaline prior to defibrillation was associated withsignificant increases in ROSC and survival when com-pared with adrenaline alone. Niemann et al. [9] previ-ously demonstrated a survival benefit in dogs treatedwith adrenaline prior to defibrillation following 7.5 minof VF; thus, differences in ROSC and survival in thisstudy are likely to be attributable to sodium bicarbon-ate treatment.

There is biochemical, histological and clinical evi-dence that the cumulative energy delivered to the my-ocardium during defibrillation attempts is associatedwith myocardial damage [11,12,18], which decreases thelikelihood of successful resuscitation. Furthermore, re-peated defibrillation attempts may decrease activity ofpacemakers by increasing parasympathetic tone [19],thereby further reducing the likelihood of attainingROSC.

Although Vukmir et al. [4] did not demonstrate anydifferences in the number of defibrillation attemptsbetween bicarbonate-treated and control animals, thisstudy demonstrates that significantly fewer countershocks were required during resuscitation in bicarbon-ate-treated animals compared with the control animals.As ROSC was attained more frequently in bicarbonate-treated animals, fewer shocks would be expected duringthe resuscitation period. However, where ROSC wasachieved, the number of shocks required was less than50% of the number required in the corresponding con-

trol animals. This indicates that VF either persisted orrecurred more frequently during the resuscitation pe-riod in control animals.

During cardiac arrest, the absence of adequate car-diac output leads to tissue hypoxia, cellular anaerobicmetabolism and the depletion of high-energy phos-phates [20]. Ultimately, oxidative phosphorylation failsresulting in accumulation of H+ ions that inhibit anaer-obic glycolysis and reduce production of ATP [21].Furthermore, lactate is produced during anaerobic gly-colysis, exacerbating acidosis. Failure of ventilationresults in the accumulation of CO2, thereby worseningcellular acidosis. Acidosis decreases cardiac contractil-ity [7,1], may decrease CoPP [22] and attenuate theeffect of vasopressors [4,8], may cause arrhythmias [6]and is associated with neurological damage [23].

The use of bicarbonate in cardiac arrest has beencriticised because of potential adverse effects includingsystemic alkalosis, paradoxical intracellular acidosis,hypercapnia (due to CO2 production) and hyperna-traemia [24]. It was not meaningful to assess the venouspH immediately following 10 min VF (i.e. commence-ment of resuscitation) due to the absence of tissueblood flow that occurs during VF. The pH measured at5 min was considered to be reflective of the cumulativemetabolic changes during 10 min of VF and during thefirst 5 min of resuscitation. Acidosis, still present in thismodel of cardiac arrest 5 min after resuscitation, wasreversed by treatment with sodium bicarbonate, so thatthe venous pH at this time was decreased significantlyfrom pre-arrest values only in control groups. Further-more, the dose of bicarbonate used (3 mmol/kg) wassufficient to restore pH without causing alkalosis or asignificant increase in pH from the pre-arrest values.Although there was no means of determining whetherintracellular acidosis occurred, PCO2

increased signifi-cantly in bicarbonate-treated animals. These changes inPCO2

have been shown to be transient following thereturn of ROSC [4]. In vitro studies reporting intracel-lular acidosis have been used to argue against bicarbon-ate therapy. However, their validity has been

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315 313

Table 3Venous blood pH and PCO2

5 min after initiation of resuscitation in dogs after 10 min of cardiac arresta

Variable Immediate treatmentTime Immediate defibrillation

Control (n=6) NaHCO3 (n=6) Control (n=6) NaHCO3 (n=5)

pH Pre-arrest 7.28�0.02 7.26�0.02 7.24�0.02 7.22�0.035 min 6.90�0.07* 7.15�0.09 7.03�0.03* 7.17�0.05

55�3 60�3Pre-arrest 59�4PCO256�5

56�135 min 82�7* 57�6 79�8*

a Animals received either immediate treatment or immediate defibrillation. Within these groups, animals were randomly assigned to eithercontrol or sodium bicarbonate treatment groups. * P�0.05 compared with pre-arrest values.

questioned and they may not be able to be extrapolatedto the in vivo setting [25]. In any event, the occurrenceof intracellular acidosis (if present) may be an accept-able adverse effect, considering that bicarbonate ther-apy was associated with significant increases in ROSCand survival. In out-of-hospital arrest in humans,Warner et al. [1] found that the ability to restore a pulsein the field was a major prognostic indicator for sur-vival. Therefore, therapies that increase the likelihoodof attaining ROSC are also likely to be useful.

The coronary perfusion pressure of bicarbonate-treated animals (�20 mmHg) was significantly higherthan that in control animals. Wenzel et al. [8] demon-strated that only the first dose of adrenaline was effec-tive at improving CoPP in a pig model of cardiacarrest; therefore, as the duration of arrest increases, theeffectiveness of adrenaline as a vasopressor decreases.Furthermore, the effectiveness of adrenaline as a vaso-pressor has been shown to decrease in the presence ofacidosis [4,8]. The higher CoPP in bicarbonate-treatedanimals may be due to the increased effectiveness ofadrenaline in the less acidotic conditions. One of thebest predictors of ROSC during CPR, in both animalsand humans, has been shown to be a CoPP of between20 and 30 mmHg [8]. Thus, the higher rate of ROSC inbicarbonate-treated animals was likely to be due, atleast in part, to the improved CoPP.

Not only is the effectiveness of adrenaline decreasedfollowing the first dose, but Lerman and Engelstein [26]showed that as the duration of VF increased, thedefibrillation threshold also increased and, therefore,successful defibrillation was correspondingly less likelyto be achieved. Thus, when circulation is not readilyrestored, the prognosis is poor. In the current study, itappeared that ROSC was attained earlier in the bicar-bonate-treated animals than in controls.

The best outcome in this study was achieved whendefibrillation was delayed for approximately 2 min,during which time NaHCO3 and adrenaline were ad-ministered with CPR, despite the fact that this in-creased the time in VF to approximately 12 min. Cobbet al [10] indicated that, after arrest lasting longer than

4 min, a brief period of CPR prior to defibrillationimproved the outcome. Sodium bicarbonate reversedthe acidosis associated with cardiac arrest and increasedCoPP with out any obvious adverse effects that im-pacted on outcome. Prolonged cardiac arrest that oc-curs out-of-hospital presents unique challenges.Significant acidosis is likely to be present by the timeparamedics arrive and the opportunity for immediatedefibrillation to restore a perfusing rhythm has usuallypassed. Current guidelines advocate consideration ofpotentially reversible causes, followed by a standardapproach of cluster shocks, adrenaline and CPR [15],regardless of the period of arrest. There has been onlyone randomised controlled clinical trial investigatingbuffer therapy in the setting of out-of-hospital cardiacarrest [27]. While this study found no beneficial effectof buffer therapy, there are good reasons why furtherinvestigations of immediate treatment with sodium bi-carbonate (and adrenaline) prior to defibrillation inprolonged (�10 min) cardiac arrest in humans areneeded [28]. Neurological outcome was not determinedin the present study so that the likely impact of thisregime on brain resuscitation remains to be determined.It is likely, however, that obtaining ROSC in the field isthe first and critical step in obtaining a favourableresuscitation outcome. The results of the present studyindicate that in prolonged arrests bicarbonate therapyand a period of perfusion prior to defibrillation mayincrease survival in this group of patients in whom theprognosis is very poor.

Acknowledgements

This study was supported in part by a grant donatedby the Laerdal Foundation for Acute Medicine, Sta-vangar, Norway. Michigan Instruments (GrandRapids, MI, USA) generously supplied Thumper® andAcute Care Systems (Sydney, Australia) suppliedLIFEPAK-10®. The authors gratefully acknowledge theexpert technical assistance of Desmond Richardason,Tina Stasinopoulos and Donald Davidson, and the

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315314

extensive out-of-hospital insight provided by StephenIrons (EMT-P).

References

[1] Warner LL, Hoffman JR, Baraff LJ. Prognostic significane offield response in out-of-hospital ventricular fibrillation. Chest1985;87:22–8.

[2] Yakaitis RW, Ewy GA, Otto CW, Taren DL, Moon TE. Influ-ence of time and therapy on ventricular fibrillation in dogs. CritCare Med 1980;8:157–63.

[3] Hargarten KN, Stueven HA, Waite EM, Olson DW, Mateer JR,Aufderheide TP, Darrin JC. Prehospital experience with defibril-lation of course ventricular fibrillation — a ten year review. AnnEmerg Med 1999;19:157–62.

[4] Vukmir RB, Bircher NG, Radovsky A, Safar P. Sodium bicar-bonate may improve outcome in dogs with prolonged cardiacarrest. Crit Care Med 1995;23:515–22.

[5] Levy MM. An evidence-based evaluation of the use of sodiumbicarbonate during cardiopulmonary resuscitation. Crit CareClin 1988;14:457–83.

[6] Orchard CH, Cingolani HE. Acidosis and arrhythmias in cardiacmuscle. Cardiovasc Res 1994;28:1312–9.

[7] Teplinsky K, O’Toole M, Olman M, Walley KR, Wood LDH.Effect of lactic acidosis on canine haemodynamics and leftventricular function. Am J Physiol 1990;258:1193–9.

[8] Wenzel V, Linder KH, Krismer AC, Miller EA, Voelckel WG,Lingnau W. Repeated administration of vasopressin but notepinephrine maintains coronary perfusion pressure after earlyand late administration during prolonged cardiopulmonary re-suscitation in pigs. Circulation 1999;99:1379–84.

[9] Niemann JT, Cairns CB, Sharma J, Lewis R. Treatment ofprolonged ventricular fibrillation — immediate countershockversus high dose epinephrine and CPR preceding countershock.Circulation 1992;85:281–7.

[10] Cobb LA, Fahrenbruch CE, Walsh TR, Copass MK, OlsufkaM, Breskin M, Hallstrom AP. Influence of cardiopulmonaryresuscitation prior to defibrillation in patients with out-of-hospi-tal ventricular fibrillation. J Am Med Assoc 1999;281:1182–8.

[11] Gaba DM, Talner NS. Myocardial damage following thranstho-racic direct current countershock in newborn piglets. PaediatrCardiol 1982;2:281–8.

[12] Kerber RE, Carter J, Klein S, Grayzel J, Kennedy J. Open chestdefibrillation during cardiac surgery: energy and current requir-ment. Am J Cardiol 1980;46:393–6.

[13] Herlitz J, Ekstrom L, Wennerblom B, Axelsson A, Bang A,Holmberg S. Effect of bystander initiated cardiopulmonary re-suscitation on ventricular fibrillation and survival after witnessed

cardiac arrest outside hospital. Br Heart J 1994;72:408–12.[14] Martens P, Vandekerckhove Y, The Belgian CPCR Study

Group. Optimal defibrillation stategy and follow-up of out-of-hospital cardiac arrest. Resuscitation 1996;31:25–32.

[15] de Latorre F, Nolan J, Robertson C, Chamberlain D, Baskett P.European resuscitation guidelines 2000 for adult advanced lifesupport. Resuscitation 2001;48:211–21.

[16] Haynes BE, Mendoza A, McNeil M, Schroeder J, Smiley DR. Astatewide early defibrillation initiative including lay persons andoutcome reporting. J Am Med Assoc 1999;266:545–7.

[17] Kellerman AL, Hackman BB, Somes G, Kreth TK, Nail L,Dobyns P. Impact of first-responder defibrillation in an urbanemergency medical services system. J Am Med Assoc1994;270:1708–1713.

[18] Brown CG, Griffith RF, Ligten PV, Hoekstra J, Nejman G,Mitchell L, Dzwonczyk R. Median frequency — a new parame-ter for prediciting defibrillation success rate. Ann Emerg Med1991;20:787–9.

[19] Strohmenger H, Lindner KH, Prengel AW, Pfenninger EG,Bothner U, Lurie KG. Effects of epinephrine and vasopressin onmedian fibrillation frequency and defibrillation success in aporcine model of cardiopulmonary resuscitation. Resuscitation1996;31:65–73.

[20] Kern KB, Garewal HS, Sanders AB, Janas W, Nelson J, SloanD, Tacker WA, Ewy GA. Depletion of myocardial adenosinetriphosphate during prolonged untreated ventricular fibrillation:effect on defibrillation success. Resuscitation 1990;20:221–9.

[21] Halperin ML, Cheema-Dhadli S, Halperin FA, Kamel KS.Rationale for the use of sodium bicarbonate in a patient withlactic acidosis due to a poor cardiac output. Nephron1994;66:258–61.

[22] Narins RG, Cohen JJ. Bicarbonate therapy for organic acidosis:the case for its continued use. Ann Intern Med 1987;106:615–8.

[23] Abramson NS, Safar P, Detre KM, Kelsey SF, Monroe J,Reinmuth O, Snyder JV. Neurological recovery after cardiacarrest: effect of duration of ischaemia. Crit Care Med1985;13:930–1.

[24] Gazmuri RJ. Buffer treatment for cardiac resuscitation: puttingthe cart before the horse. Crit Care Med 1999;27:875–6.

[25] Levraut J, Labib Y, Chave S, Payan P, Raucoules-Aime M,Grimaud D. Effect of sodium bicarbonate on intracellular pHunder different buffering conditions. Kidney Int 1996;49:1262–7.

[26] Lerman BB, Engelstein EE. Increased defibrillation thresholddue to ventricular fibrillation duration. J Electrocardiol1995;28:21–4.

[27] Dybvik T, Strand T, Steen PA. Buffer therapy during out-of-hospital cardiopulmonary resuscitation. Resuscitation1995;29:89–95.

[28] Koster RW. Correction of acidosis during cardio-pulmonaryresuscitation. Resuscitation 1995;29:87–8.

Portuguese Abstract and Keywords

A probabilidade de desfibrilhacao e reanimacao eficazes diminui com o aumento de duracao da paragem cardıaca. A paragemcardıaca prolongada tambem esta associada ao desenvolvimento de acidose. Estas experiencias foram projectadas para determinarse a administracao de bicarbonato de sodio e/ou adrenalina em combinacao com um breve perıodo de reanimacao cardiopul-monar (CPR) antes da desfibrilhacao melhorava o prognostico da paragem cardıaca prolongada em caes. A fibrilhacaoventricular (VF) era induzida por choque com corrente alterna em caes anestesiados. Apos 10 min de FV, os animais recebiamou desfibrilhacao imediata (seguida de tratamento com bicarbonato ou controle) ou tratamento imediato com bicarbonato ousoro fisiologico (seguido de desfibrilhacao). O tratamento com bicarbonato correlacionou-se com aumento da taxa de recuperacaode circulacao espontanea, o que foi conseguido com menos choques e num perıodo de tempo mais curto. A pressao de perfusaocoronaria era significativamente mais alta em animais tratados com bicarbonato que nos animais de controle. Houve menordescida dos valores de pH venoso nos animais tratados com bicarbonato que nos controle. O melhor prognostico neste estudo

E.C.M. Leong et al. / Resuscitation 51 (2001) 309–315 315

foi conseguido quando a desfibrilhacao foi atrasada cerca de 2 min, tempo durante o qual, se fez SBV e foram administradosbicarbonato e adrenalina. Os resultados deste estudo indicam que nas paragens prolongadas o tratamento com bicarbonato e umperıodo de perfusao antes da desfibrilhacao podem aumentar a sobrevida.

Pala�ras cha�e: Fibrilhacao ventricular; Desfibrilhacao; Acidose; Bicarbonato; Paragem cardıaca; Reanimacao