2.1

Study design

This study was a retrospective analytic study based on a secondary analysis of two experimental porcine studies on VF-induced cardiac arrest. One study aimed to compare the effects of the timing of the first epinephrine administration, while the other focused on the interval of epinephrine administration. Unstratified block random allocation was used to assign study subjects to each group in both experimental studies. The study protocol was approved by the Institutional Animal Care and Use Committee of the study institution. All experiments were performed in accordance with the tenets of the Laboratory Animal Act of the Korean Ministry of Food and Drug Safety. The results are reported in accordance with Animal Research: Reporting of In Vivo Experiments guidelines.

2.2

Animal preparation

The study involved 20 healthy female pigs, approximately 3 months old and weighing 41–59 kg. The subjects were acquired from a local farm, acclimatized, and fasted overnight. They were sedated using intramuscular injections of tiletamine/zolazepam hypochloride (Zoletil®; Virbac, France, 2–4 mg/kg) and xylazine (Rompun®; Bayer Korea, South Korea, 2 mg/kg). Endotracheal intubation was performed using a 7.0 mm endotracheal tube, and a mechanical ventilator was applied at a tidal volume of 8 mL/kg and 15 breaths/min to maintain normocapnia and oxygenation. Anesthesia was maintained with continuous inhalation of isoflurane (2–5%). Central venous catheter was inserted through internal jugular vein for VF induction.

Two burr-hole trephinations were performed 1 cm anterior to the coronal sutures. In the first burr hole, a micro-tip pressure catheter (Millar Instruments, USA) was inserted to measure the intracranial pressure (ICP), and a laser Doppler probe (OxyFloTM; Oxford Optronix, United Kingdom) was inserted for continuous measurement of cortical cerebral microvascular flow (CBF). An additional micro-tip pressure catheter was inserted into the femoral artery and placed in the abdominal aorta to measure the aortic blood pressure. The heart rate was continuously monitored throughout the experiment using electrocardiography, oxygen saturation using pulse oximetry, body temperature, and end-tidal carbon dioxide (ETCO ₂ ). All physiological parameters were recorded using a digital recording system (PowerLab; AD Instruments, Australia) and processed using data processing software (LabChart; AD Instruments, Australia).

2.3

Experimental protocol

We categorized the study subjects into Short and Long NF groups based on the protocol of each experiment. In both experiments, VF was induced by transmitting low-voltage electricity near the right ventricular wall using a guidewire. In the Short NF group, the cardiac arrest state without any treatment (NF interval) was maintained for 4 min, and mechanical chest compressions using LUCAS 2 (Stryker Physio-Control; Sweden) along with bag-valve mask ventilation (Basic life support-BLS phase) were delivered for 6 min after the NF interval. Defibrillation for shockable rhythms was initiated every 2 min in the BLS phase (from 8 min since VF induced) and continued until ROSC. A 1 mg epinephrine bolus was administered every 4 min, starting from 10 min since VF induction (advanced cardiovascular life support-ACLS phase). In the Long NF group, the NF interval was 6 min, and after 4 min of BLS phase, defibrillation and epinephrine administration were initiated. Epinephrine was administered every 5 min after VF in the Long NF group. Otherwise, resuscitation process was the same in both experiments; After the termination of resuscitation, the subject was sacrificed if ROSC sustained over 20 min or was not achieved after 24 min of CPR ( Fig. 1 ).

Experimental protocol for two porcine cardiac arrest models.

NF, no-flow; VF, ventricular fibrillation; BLS, basic life support; ACLS, advanced cardiovascular life support.

2.4

Measurements

Physiological parameters were continuously measured throughout the experiment. Cerebral perfusion pressure (CePP) was calculated as the difference between the MAP and ICP. Relative CBF during CPR was presented as the relative ratio of the median cortical CBF to baseline ((measurement – baseline value)/baseline value). Baseline value was defined as median cortical CBF during the first 2 min interval of CPR (BLS first phase) for each subject. Baseline physiological parameters were extracted for 2 min and arterial blood gas analysis was conducted before VF induction.

2.5

Outcomes

The primary outcome was the mean difference in MAP (mmHg) between 2 min of resuscitation phases before and after epinephrine administration. Secondary outcomes were the mean differences in CePP (mmHg), ETCO ₂ (mmHg), and relative CBF.

2.6

Statistical analysis

The sample size was not calculated as this was a secondary analysis based on two porcine experimental studies. Study subjects were selected based on the similarity of the resuscitation protocol, except for no flow interval. The independent variable was duration of NF (4 min for Short NF and 6 min for Long NF), and dependent variables included CePP, MAP, ETCO2, and CBF. One study subject that achieved ROSC within 14 min from VF induction was excluded. Median value and interquartile ranges of each physiological parameter for each subject were calculated for every 2 min interval during baseline and resuscitation phases. The Wilcoxon rank-sum test was used to compare physiological parameters between the study groups. A linear mixed-effect model was employed to evaluate the difference between the study groups, considering the group-time interaction (study group * resuscitation phase) in median values of parameters. In this model, a random effect was applied to each study subject, and fixed effects for study group and resuscitation periods were considered. Considering the effect on individual subjects, the mean differences in each parameter were calculated by subtracting the value of the previous phase from that of the current phase. All analyses were performed using R version 3.5 (R Foundation for Statistical Computing; Vienna, Austria). Statistical significance was set at p < 0.05, and 95% confidence intervals were used.