This protocol demonstrates the use of an implantable cardioverter-defibrillator (ICD) to reliably induce ventricular arrhythmias, perform internal cardioversion, and obtain intracardiac ECG recordings in a porcine model.
Method Article
This protocol demonstrates the use of an implantable cardioverter-defibrillator (ICD) to reliably induce ventricular arrhythmias, perform internal cardioversion, and obtain intracardiac ECG recordings in a porcine model.
This study presents a reproducible porcine cardiac arrest model utilizing an implantable cardioverter-defibrillator (ICD) for reliable induction and cardioversion of ventricular fibrillation (VF), as well as intracardiac ECG monitoring during resuscitation. Large animal models are vital for translational research in cardiac arrest; however, conventional VF induction and cardioversion techniques -- primarily using external electrodes and defibrillation -- are often hampered by motion artifacts, inconsistent conversion rates, and high animal resource use. This protocol describes repeated cycles of VF induction and cardioversion within the same animal, providing precise control while minimizing physiological stress and animal numbers, in accordance with the 3R principles (Replacement, Reduction, and Refinement). The ICD-based method allows for accurate rhythm monitoring by intracardiac ECG throughout resuscitation, with improved artifact resistance compared to surface ECG. Eleven pigs underwent repeated VF induction and cardioversion, with high success rates and recovery of spontaneous circulation. The model enables standardized and reliable data acquisition for cardiac arrest studies, ensuring experimental consistency and facilitating reproducible investigation of pathophysiology and resuscitation strategies while minimizing animal use.
Pigs and humans share significant anatomical and physiological similarities, making porcine models essential for resuscitation research1. Techniques to induce ventricular fibrillation (VF) and cardiac arrest include: direct current (DC) application by spinal needles inserted into the myocardium using a 9 V battery, or sinus generator2; transcutaneous or subcutaneous alternating current (AC)3,4; intramyocardial application with a pacer electrode using either direct current (DC) (9 V battery), alternating current (AC; oscilloscope/function generator), or an external pacemaker5,6,7,8. While all these methods effectively induce VF, conversion to sinus rhythm (SR) is typically performed through external transcutaneous application of a DC shock and return of spontaneous circulation (ROSC) is often suboptimal. Moreover, external shocks impose significant physiological and tissue stress and potential injury9,10, limiting the number of repeat interventions per animal. Consequently, VF studies often require a large number of animals to detect statistically significant differences, which conflicts with ethical imperatives to reduce animal use.
Previous research in humans has shown that VF can be induced using an implantable cardioverter-defibrillator (ICD) system11. To our knowledge, repeated VF inductions and conversions using an ICD system have not been studied in a porcine model. An electrocardiogram (ECG) is used to detect the conversion from ventricular fibrillation to sinus rhythm. During cardiopulmonary resuscitation (CPR), high-quality standard surface ECG may be unreliable or challenging to obtain due to motion artifacts and artifacts related to the electrode-skin interface and the electrical properties of the electrodes12,13,14.
In line with the 3R principles of animal research (Replacement, Reduction, and Refinement)8, and our ongoing research on flow detection during cardiac arrest15,16,17, we aimed to develop a predictable and reproducible porcine cardiac arrest model for repeated VF induction and conversion to SR within a single animal. Additionally, we aimed to use the ICD device to obtain intracardiac ECG readings during CPR. By reducing variability and improving CPR success, this method addresses key limitations of existing models, minimizes animal use, and facilitates more reliable cardiac arrest research. This refined approach ensures consistency and offers significant advantages over traditional techniques, enabling researchers to achieve reliable results while adhering to ethical standards in experimental design. The model presented is technically advanced and requires access to an ICD and a programming unit, as well as fluoroscopy or ultrasound for guiding pacemaker lead insertion. Due to the size of the ICD, pacemaker leads, and vascular introducers, the model is applicable to pigs weighing at least 25 kg.
The experiments described in this protocol were approved by the Norwegian Animal Research Authority (FOTS-ID 25415) and performed in accordance with EU Directive 2010/63/EU on the Protection of Animals Used for Scientific Purposes. Eleven Norwegian Landrace pigs (Sus scrofa domesticus), comprising 10 males and one female, with a mean age of 56 ± 1.5 days and a mean weight of 30 ± 3 kg, were used. The experimental setup is illustrated in Figure 1.
1. Animal retrieval, anesthesia, and monitoring
2. Instrumentation
3. Pacemaker/defibrillator programming
4. Establish non-invasive continuous carotid Doppler monitoring
5. Conduct cardiac arrest studies
6. Euthanasia and disposal of waste
Intentionally timed induction of VF and subsequent cardioversion to SR with ROSC was considered a positive outcome of the method. Failure to either induce VF or cardiovert was considered a negative outcome. Using the VF fibber mode consistently induced VF or VT in all animals, as desired (Table 1). Further, using the defibrillator mode consistently cardioverted all pigs. Both induction of VF and cardioversion were remotely controlled from the programming unit; thus, no external sinus generator or defibrillator was needed. All animals were repeatedly cardioverted to sinus rhythm with ROSC several times throughout each experiment (Table 1)15,17. The resting interval between each VF sequence was a minimum of 5 min.The success rate for achieving ROSC after induced VF was 92% across all sequences in all animals (Table 1). VF requiring two DC shocks for conversion to ROSC occurred twice in one animal. In two animals that had an intentionally large myocardial infarction induced by the injection of microbeads in the coronary arteries before the induction of VF15, VF induction led to PEA. In one animal, DC shock resulted in immediate transition to asystole. Standard CPR was subsequently initiated in all three animals, but none achieved ROSC. In this series of 11 porcine experiments utilizing ICD-based induction and cardioversion of ventricular fibrillation, each animal successfully underwent repeated cycles of VF and resuscitation. The ICD protocol consistently yielded high cardioversion success rates with minimal physiological stress and rapid recovery of spontaneous circulation. Intracardiac ECG monitoring provided artifact-free rhythm documentation throughout resuscitation, enabling precise detection of arrhythmia and interventions. These results demonstrate the technique's reliability, reproducibility, and potential for detailed outcome analysis, such as measuring conversion rates, recovery intervals, and ECG signal quality, thereby supporting robust experimental investigation of resuscitation strategies

Figure 1: Experimental setup. (A) The experimental animals were intubated with an endotracheal tube in the prone position and connected to a ventilator. In the supine position, under ultrasound guidance, the following vascular access was established: (B) A PiCCO arterial line was inserted into the femoral artery, (C) a central venous catheter was placed in the right external jugular vein, and (D) a vascular sheath introducer was inserted into the right internal jugular vein. Additionally, a suprapubic catheter was inserted into the bladder (not shown here), and (E) a RescueDoppler carotid flow probe was placed over the left carotid artery. (F) Through the vascular sheath, a pacemaker electrode was advanced into the right ventricle and its position verified using echocardiography or fluoroscopy. (G) An implantable cardioverter-defibrillator unit (ICD) was inserted in a subcutaneous pocket under the left clavicle and connected to the pacemaker electrode. (H) The animals were connected to a multimodal monitor recording ECG, tail plethysmography (SpO₂), arterial blood pressure, central venous pressure, and end-tidal CO₂. (I) The RescueDoppler sensor was connected to a RescueDoppler recorder, providing continuous carotid flow tracings. (J) The ICD device was connected through telemetry to an ICD programmer. Figure created using BioRender. Please click here to view a larger version of this figure.
| Pacemaker-induced arrhythmia | CPR procedures performed | Arrhythmia conversion rate (%) | |||
| Pig* | Ventricular fibrillation (n) | Ventricular tachycardia (n) | Manual compression (n) | Mechanical compression (n) | |
| 1 | 12 | 4 | 6 | 2 | 91.6 |
| 2 | 14 | 7 | 0 | 5 | 100 |
| 3 | 9 | 7 | 0 | 9 | 100 |
| 4 | 8 | 4 | 5 | 5 | 100 |
| 5 | 18 | 0 | 12 | 5 | 94.4 |
| 6 | 8 | 0 | 8 | 0 | 87.5 |
| 7 | 15 | 8 | 9 | 6 | 100 |
| 8 | 18 | 0 | 9 | 9 | 94.4 |
| 9 | 14 | 2 | 9 | 5 | 100 |
| 10 | 27 | 0 | 13 | 11 | 100 |
| 11 | 19 | 0 | 12 | 4 | 84.2 |
Table 1: Summary of ICD-induced VF/VT and subsequent cardioversion. *Eleven pigs underwent experiments. On average, 18 arrhythmias, including ventricular fibrillation (VF) and ventricular tachycardia (VT)- were induced per animal using an implantable cardioverter-defibrillator (ICD). The mean VT/VF duration was 0.75 min. CPR was performed in selected VF episodes according to earlier study protocols15,16,17. Animals were then converted to sinus rhythm (SR) by direct current (DC) shock via the ICD. Each animal was stabilized for 5 min before the next arrhythmia was induced.
To our knowledge, the presented model utilizing an implantable defibrillator to both induce and cardiovert ventricular fibrillation (VF) in pigs has not previously been described in the literature. VF is commonly induced using either sinusoidal waveform generators or direct current from an external battery through epicardial electrodes. Cardioversion is typically performed using external defibrillators. While these methods may reliably induce VF, failure to cardiovert animals using external defibrillation has been reported in up to 45% of cases2.
ECG monitoring during CPR is challenging due to compression-related artifacts that may obscure the underlying rhythm18. However, an implantable defibrillator with intracardiac ECG enables continuous rhythm monitoring during chest compressions, allowing for more accurate real-time rhythm assessment and reliable post-compression analysis.
In contrast, this model reliably cardioverted all animals, with multiple successful conversions per experiment. This enabled the repeated use of individual animals across experimental protocols, thereby reducing the number of animals sacrificed and improving overall resource efficiency. Additionally, the model eliminated the need for external defibrillation patches, freeing space on the thorax and improving access during instrumentation.
The model is unique in its ability to study the physiological effects of short, complete circulatory arrests in intact animals, where cardiac restart is achieved by a gentle, single internal shock. This feature enables controlled investigations of ischemic tolerance in vital organs such as the kidneys and gastrointestinal tract, redistribution of blood flow after global ischemia, and the interplay between endogenous sympathetic activation, ischemic stress responses, and the administration of adrenergic drugs. The model, therefore, provides a valuable experimental platform for exploring organ-specific and systemic responses to cardiac arrest, causing whole-body ischemia under highly standardized conditions. Beyond electrode placement, maintaining physiological stability is crucial for reproducibility. Continuous monitoring of temperature, ventilation, and anesthesia depth ensures normothermia, adequate oxygenation, and stable hemodynamics throughout the experiment.
The presented representative results demonstrate the reliability and reproducibility of the ICD-controlled VF induction and cardioversion in a controlled experimental setting. The ability to induce and terminate VF remotely enabled rapid and precise timing, with minimal external interference, which is valuable for studying cardiac arrest physiology and resuscitation dynamics. The outcome analysis focused on the consistency of VF induction, the number of successful cardioversions, and the stability of ROSC.
In this study, we adhered to the 3Rs principle by generating statistically robust data while minimizing the number of animals used. Owing to the high rate of successful resuscitations, ventricular fibrillation and cardioversion could be induced multiple times in each subject. By employing repeated-measures statistics within subjects and mixed-effects models accounting for both fixed and random effects, the study could be conducted using only eleven animals.
Limitations
The primary limitation of the model lies in the requirement for specialized equipment and operator expertise. A pacemaker generator, transvenous pacing lead, programming unit, and a trained operator are all essential. Accurate placement of the pacing lead is critical and requires imaging guidance, typically by transthoracic echocardiography or fluoroscopy.
Consequently, implementation of this model may be challenging in resource-limited settings. However, once established, it offers a highly effective and reproducible method to induce and terminate VF, minimizing the occurrence of non-convertible fibrillation and enhancing experimental control. This model is broadly relevant to research groups conducting in vivo resuscitation studies in large animal models.
This porcine model was developed to assess the feasibility of automated carotid Doppler monitoring during cardiac resuscitation, and not for testing clinical resuscitation strategies. The study focused on acute physiological responses and was conducted as a terminal protocol, which excluded long-term follow-up or histological evaluation. Additionally, we used only 11 animals in this study. Although multiple repeated experiments were performed in each animal, inter-individual variability may not have been fully captured. Furthermore, the study did not include a control group using traditional methods. Although such methods were referenced as having a high failure rate, the lack of direct comparison within the same experimental framework limits the ability to demonstrate the relative advantages of the proposed technique. Finally, the model requires access to specialized equipment such as an ICD, programmer, and Doppler ultrasound, which may limit its applicability in resource-constrained settings. Additionally, the surgical protocol is technically demanding, involving multiple vascular access points, bladder stoma creation, and ICD implantation, which may present a steep learning curve for inexperienced operators.
In this experimental setup, intracardiac ECG recordings could not be exported from the programming unit; therefore, data interpretation relied solely on visual monitoring. Future studies should incorporate a method enabling direct export of ECG traces to facilitate quantitative analysis.
Charlotte Björk Ingul is a consultant for Cimon Medical, the company that owns the technology associated with RescueDoppler. The remaining authors declare no conflicts of interest.
We used Perplexity and Microsoft Copilot to assist with English language proofreading. All content was reviewed and approved by the authors. This project has received funding from Nord University, Norwegian University of Science and Technology, and the Norwegian Research Council.
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| Abbott Durata 7120 ICD Single Coil Lead | Abbott, Abbot Park, IL | 7120 | |
| Abbott Gallant VR ICD | Abbott, Abbot Park, IL | CDVRA500Q | |
| Abbott Merlin Programmer | Abbott, Abbot Park, IL | 3650 | |
| Central venous catheter Certofix trio | B. Braun Medical, Melsungen, Germany | 4167408-07 | |
| Endotracheal tube Portex #7.0 | Smiths Medical, Rockland, MA | 100/150/070 | |
| Exagon VET (Pentobarbital 400 mg/mL) | VetViva Richter, Wels, Austria | N/A | |
| Heparin 5000 IU/mL | Panpharma, Luitré, France | ETI3M807-1 | |
| Ketamine 50 mg/mL | Abcur AB, Helsingborg, Sweden | LB-12355-01 | |
| Midazolam 5 mg/mL | B. Braun Medical, Melsungen, Germany | 753-12612876-0621 | |
| Morphine | Abcur AB, Helsingborg, Sweden | LB-122184-01 | |
| Normal Saline flush solution | Fresenius Kabi, Uppsala, Sweden | B202382-05 | |
| PentoCur (Thiopental) | Abcur AB, Helsingborg, Sweden | LB-122564-01 | |
| Periphal IV catheter 20G Vasofix | B. Braun Medical, Melsungen, Germany | 4268113B | |
| PiCCO Monitoring kit (pressure transducer, flush system and thermistor) | Pulsion, Feldkirchen, Germany | PV8215 | |
| PiCCO catheter 5 Fr 20 cm | Pulsion, Feldkirchen, Germany | PV2015L20-A | |
| Pressure transducer and flush kit TruWave | Edwards Lifescience, Irvine, CA | PX272 | |
| RescueDoppler | Cimon Medical, Trondheim, Norway | N/A | Proprietary, not commercially available |
| Ringer acetate | Fresenius Kabi, Uppsala, Sweden | B203150-02 | |
| Urinary cather 12 Ch | Various | N/A | |
| Vascular sheath introducer Prelude 7.0 Fr | Merit Medical, South Jordan, UT | PSI-7F-11-035-18G | |
| Vue Ultrasound Gel | Optimum medical, Leeds, United Kingdim | 1157 |
Request permission to reuse the text or figures of this JoVE article
Request Permission