EDITOR:

Augmenting conventional resuscitative efforts with infusion of lipid emulsions has resulted in successful resuscitation from intractable local anaesthetic-induced cardiac arrest in a number of recent case reports. These follow animal studies demonstrating efficacy for lipid infusion in local anaesthetic-induced cardio-toxicity [Reference Weinberg, Ripper, Feinstein and Hoffman1] and other lipid-soluble drug toxidromes [Reference Harvey and Cave2]. Two mechanisms of action have been forwarded as the basis for the observed beneficial effects of lipid infusion. Weinberg’s ‘lipid sink’ hypothesis [Reference Weinberg3] proposes reduced tissue binding by re-establishing equilibrium in a plasma-lipid phase. Additional investigators have proposed a metabolic stimulant effect with enhanced myocardial lipid utilization resulting in augmented cardiac performance. The premise of the latter garners potential benefit in non-lipophilic-toxin-associated cardiac arrest. Accordingly, we decided to explore the hypothesis that lipid infusion might result in improved resuscitation outcomes in an intact animal model of asphyxial cardiac arrest.

With the approval of the Ruakura Animal Ethics Committee (Ruakura Animal Research Centre, Hamilton, New Zealand), 16 sedated (ketamine 50 mg kg⁻¹, xylazine 4 mg kg⁻¹ via intramuscular injection) adult (age 95–135 days) New Zealand White rabbits were studied. Animal care and husbandry was in accord with regulations set by the institutional Ethics Committee. All animals underwent arterial (connected in standard fashion to Hewlett-Packard 78834A neonatal monitor; Hewlett-Packard, Palo Alto, CA, USA) and venous cannulation, continuous electrocardiogram (ECG) monitoring, and formal tracheostomy with tracheal intubation under direct vision. After completion of invasive procedures during which time animals breathed room air, control animals received an intravenous (i.v.) infusion of 3 mL kg⁻¹ 0.9% saline solution over a 2-min period. Test animals received 3 mL kg⁻¹ Intralipid 20% over an identical period. All fluids were pre-warmed to 37°C prior to delivery.

Respiratory arrest was induced via i.v. administration of vecuronium 0.1 mg kg⁻¹. Ensuing pulseless electrical activity (PEA) was defined by the development of mean arterial pressure (MAP) ⩽20 mmHg in the presence of potentially perfusing cardiac rhythm on ECG. Two minutes after the onset of PEA, we began resuscitative measures with manual external chest compressions at 160–180 compressions per minute, and pressure-controlled mechanical ventilation (100% oxygen at 10 cmH₂O pressure-controlled ventilation at 12 breaths per minute) via a Nuffield series 200 paediatric ventilator (Penlon Ltd, Abington, England). Five minutes after the commencement of resuscitative efforts, epinephrine 1 mL kg⁻¹ 1:10 000 was administered, and repeated 1 min later if spontaneous circulation had not been restored. External chest compressions were continued to 15 min in animals failing to achieve return of spontaneous circulation (ROSC).

ROSC was defined by maintenance of MAP >35 mmHg (equivalent to 50% referenced MAP) in the absence of external chest compressions for a period of greater than 1 min. On achieving ROSC, animals were monitored to 15 min with acquisition of haemodynamic parameters (pulse rate, MAP) at 2.5-min intervals. External chest compressions were not re-instituted in animals achieving ROSC but later exhibiting a second period of cardiac arrest. At the termination of the monitoring interval, all animals were killed via i.v. administration of 3 mL (300 mg mL⁻¹) pentobarbitone.

Time to PEA was 244 ± 74 s in the saline group and 231 ± 76 s in the Intralipid-treated group. Cardiac rhythm at PEA in the saline group was: sinus tachycardia in two animals, 2 : 1 atrio-ventricular block in three animals, sinus bradycardia in two animals and junctional bradycardia in one animal. Cardiac rhythm at PEA in the Intralipid-treated group was: sinus tachycardia in four animals, 2 : 1 atrio-ventricular block in two animals and sinus bradycardia in two animals. Four animals in the saline group (n = 8), and seven animals in the Intralipid group (n = 8) exhibited ROSC and survived to protocol termination (P = 0.106). No difference in time to ROSC in surviving animals was observed. Of the four animals achieving ROSC in the saline group, one animal did so prior to epinephrine administration. Similarly, one animal in the Intralipid-treated group achieved ROSC prior to epinephrine administration.

These data demonstrate no statistically significant difference in ROSC for Intralipid-pre-treated rabbits undergoing resuscitation from asphyxial PEA. The study hypothesis is however not convincingly rejected given the limitations of the experiment. Primarily, due to the preliminary nature of the study and limited data on which to base sample size estimation, the study is insufficiently powered to detect lesser differences in resuscitation outcome. Furthermore, administration of Intralipid as pre-treatment rather than intra-arrest limits generalization of study findings to any clinically encountered arrest situations. These data may however be gainfully employed to power future studies investigating the potential utility of lipid infusion in non-lipophilic-toxin-induced cardiac arrest models.

The prospect of improved outcome in PEA secondary to hypoxic cardiac stunning via augmentng myocardial energetics through administration of preferential myocardial substrates is intriguing. Support for lipid administration in animal models of cardiac arrest exists. Augmented contractile performance associated with increased myocardial high-energy phosphate content has been demonstrated in isolated rabbit hearts following myocardial stunning when Intralipid was administered during reperfusion [Reference Van de Velde, DeWolff, Leather and Wouters4]. Furthermore, Intralipid infusion both into bypass perfusate, and i.v. following return of coronary flow, has been shown to improve haemodynamic parameters in dogs undergoing cardiopulmonary bypass following ischaemic cardiac arrest [Reference Jones, Tibbs, McDonald, Lowe and Hewitt5]. More recently, Stehr and colleagues [Reference Stehr, Ziegeler and Pexa6] have demonstrated an independent positive inotropic effect for lipid emulsion in bupivacaine-induced cardiac depression in rat hearts. Given that myocardial substrate preference under conditions of increased lactate concentration is known to shift toward oxidation of free fatty acids [Reference Liedtke, DeMaison, Eggleston, Cohen and Nellis7], improved outcomes with lipid infusion in cardiac arrest secondary to generalized myocardial ischaemia may not be unexpected. Further studies addressing the beneficial mechanism of action of lipid emulsions in local anaesthetic and non-lipophilic toxin-induced arrest models are warranted.