Maintenance of pig brain function under extracorporeal pulsatile circulatory control (EPCC)

**Background:** Selective vascular access to the brain is desirable for metabolic tracer, pharmacological, and other studies to characterize neural properties in isolation from somatic influences. However, current methods for artificial control of cerebral circulation often abolish pulsatility-dependent vascular signaling or neural network phenomena such as the electrocorticogram (ECoG), even while preserving individual neuronal activity. The authors aimed to develop a system that mechanically renders cerebral hemodynamics fully regulable to replicate or modify native pig brain perfusion, preserving high-order network activity as measured by ECoG and depth electrode recordings.

**Methods:** The study was approved by the Institutional Animal Care and Use Committee of UT Southwestern Medical Center. Domestic farm pigs were used. Preliminary studies were conducted in 4 anesthetized juvenile pigs (20–25 kg) to develop surgical and perfusion techniques. Six additional pigs (30–60 kg) were exsanguinated to provide heterologous blood for EPCC. The main results were obtained from two female pigs (subject 1: 25 kg; subject 2: 35 kg). Anesthesia included isoflurane replaced by intravenous ketamine (10–20 mg/kg/hr), xylazine, and guaifenesin during neurophysiological recording. A craniectomy (approx. 5 cm rostral-caudal × 4 cm transverse) was performed, and two linear ECoG strips (4 electrodes each, 1 cm spacing) and two depth electrode arrays (8 platinum sites each, 0.5 cm spacing) were placed. A brain probe monitored temperature, intracranial pressure, and tissue oxygen saturation. Neurophysiological recordings were sampled at 1000 Hz with a band-pass filter of 0.5–300 Hz. Power spectra were calculated for delta (1–4 Hz), theta (5–7 Hz), alpha (8–12 Hz), beta (13–30 Hz), and gamma (31–50 Hz) using Welch’s method. The EPCC system included a centrifugal pump (BVP-ZX, Harvard Biosciences) controlled by custom LabVIEW software that used native pressure waveforms as input. The circuit included an oxygenator, heat exchanger, Doppler flow probes, and fiber optic pressure catheters. In subject 1, the ascending aorta was cannulated and clamped proximally and distally to isolate cerebral circulation. In subject 2, the brachiocephalic artery was cannulated and clamped. Venous return was from the superior vena cava. EPCC was maintained for up to 5 hours. Blood gases and electrolytes were monitored continuously and with periodic samples. After euthanasia, brains were examined histologically with Nissl/luxol fast blue and luxol fast blue/periodic acid Schiff stains.

**Key Results:** In subject 1 (aortic isolation), native aortic pressure averaged systolic 44.6 ± 2.5 mmHg and diastolic 26.9 ± 0.7 mmHg (MAP 32.8 ± 1.3 mmHg, heart rate 92.8 BPM). Under EPCC, aortic pressure was 42.0 ± 0.15/34.8 ± 0.16 mmHg (MAP 37.2 mmHg). Carotid pressure under EPCC was 16.4 ± 0.09/14.8 ± 0.05 mmHg (MAP 15.3 mmHg), compared to native carotid pressure of 19.1 ± 0.8/16.4 ± 0.5 mmHg (MAP 17.3 ± 0.5 mmHg). In subject 2 (brachiocephalic isolation), native brachiocephalic pressure averaged 91.7 ± 0.94/54.9 ± 0.47 mmHg (MAP 67.2 ± 0.61 mmHg, heart rate 74.3 BPM). Under EPCC, brachiocephalic pressure was 95.8 ± 0.63/59.7 ± 0.13 mmHg (MAP 71.7 mmHg). Carotid pressure under EPCC was 85.0 ± 0.71/55.9 ± 0.19 mmHg (MAP 65.6 mmHg), compared to native carotid pressure of 87.6 ± 0.69/56.9 ± 0.37 mmHg (MAP 67.1 ± 0.46 mmHg). Brachiocephalic flow under EPCC was 9.01 mL per beat. Neurophysiological recordings showed little to no change in absolute power spectra across all frequency bands after EPCC compared to native conditions. Brain tissue oxygenation was significantly higher under EPCC (p < 0.0001), likely due to supplemental oxygen. Intracranial pressure and temperature remained stable. Histological examination of the cerebral cortex showed preserved cellular structure and layer organization with no abnormalities.

**Clinical Implications:** This study demonstrates that EPCC can maintain near-native brain function and structure for at least 5 hours in pigs, enabling the investigation of neural activity and circulatory manipulation independently of the rest of the body. The preservation of ECoG and depth electrode recordings, including gamma activity sensitive to metabolic dysfunction, suggests that EPCC does not impair high-order network activity. This method could facilitate metabolic tracer, pharmacological, and physiological studies that require isolation of the cerebral circulation. The brachiocephalic approach provided better preservation of carotid pressure than aortic isolation. Limitations include the short duration (5 hours), use of ketamine anesthesia, and the need for further validation in longer studies and with additional physiological manipulations.