Development of a computed tomography perfusion protocol to support large animal resuscitation research

The influence of hemorrhagic shock on brain perfusion in a swine model of raised intracranial pressure

PURPOSE

In patients with hemorrhagic shock and an intracranial space occupying lesion (SOL), brain perfusion is severely compromised due to raised intracranial pressure (rICP), significantly worsening outcomes. This study aims to develop a swine model of a SOL with rICP and shock and characterize the effect on brain perfusion.

METHODS

Ten male swine were divided into two groups- normal ICP (nICP) and rICP. rICP animals had an intracranial Fogarty balloon catheter inserted, which was infused with saline to simulate a SOL. Animals underwent hemorrhage to systolic blood pressures (SBP) of 60, 40, and 20mmHg. Cerebral blood flow (CBF) and cerebral blood volume (CBV) were measured using CT perfusion.

RESULTS

The CBF/Mean arterial pressure (MAP) and CBV/MAP curves were modeled using non-linear regression, with both groups demonstrating a sigmoid relation. In both the CBF/MAP and CBV/MAP curves, animals with rICP had loss of autoregulation at a higher MAP compared to nICP. The curves were an excellent fit for CBF (nICP R2 = 0.95; rICP R2 = 0.77) and CBV (nICP R2 = 0.96; rICP R2 = 0.78).

CONCLUSIONS

This study aids in quantifying the compounding insult of raised ICP and hemorrhage with regard to brain perfusion. Raised ICP results in autoregulatory failure at a higher MAP compared to animals with nICP. These results can help inform future studies that should be aimed at evaluating novel interventions for this complex clinical scenario.

Thoracic Endovascular Aortic RepairAcutely Augments Left Ventricular Biomechanics in An Animal Model: A Mechanism for Postoperative Heart Failure and Hypertension

BACKGROUND

Thoracic aortic stent grafts are thought to decrease aortic compliance and may contribute to hypertension and heart failure after thoracic endovascular aortic repair (TEVAR). Left ventricular (LV) biomechanics immediately after TEVAR, however, have not been quantified. Pressure-volume (PV) loop analysis provides gold-standard LV functional information. The aim of this study is to use an LV PV loop catheter and analysis to characterize the LV biomechanics before and acutely after TEVAR.

METHODS

Anesthetized Yorkshire swine (N = 6) were percutaneously instrumented with an LV PV loop catheter. A 20 mm × 10 cm stent graft was deployed distal to the left subclavian via the femoral artery under fluoroscopy. Cardiac biomechanics were assessed before and after TEVAR. As a sensitivity analysis, inferior vena cava occlusion with PV loop assessment was performed pre and post-TEVAR in 1 animal to obtain preload and afterload-independent end-systolic and end-diastolic PV relationships (ESPVR and EDPVR).

RESULTS

All animals underwent successful instrumentation and TEVAR. Post-TEVAR, all 6 animals had higher mean LV ESP (106 vs. 118 mm Hg, P = 0.04), with no change in the EDPVR. inferior vena cava occlusion also moved the ESPVR curve upward and leftward, indicating increased LV work per unit time. There was no augmentation of EDPVR following TEVAR (P > 0.05). Postmortem exams in all animals revealed appropriate stent placement and no technical complications.

CONCLUSIONS

TEVAR was associated with an acute increase in LV end-systolic pressure and shift in the ESPVR, indicating increased ventricular work. This data provides potential mechanistic insights into the development of post-TEVAR hypertension and heart failure. Future stent graft innovation should focus on minimizing the changes in cardiac physiology.

High flow cooled air can decrease brain temperature without injuring the snout or brain in Swine

Introduction

Targeted temperature management plays an important role in the treatment of myriad critical illnesses. Non-invasive, quick-onset options for isolated brain temperature control remain lacking. The goal of this study was to assess the safety and efficacy of a novel intranasal high flow cooled air device using a large animal model.

Methods

Yorkshire swine were instrumented with temperature probes in the rectum, brain, ear, and snout, and a novel intranasal cooled air device was applied to half. The primary outcome was effectiveness of brain and snout cooling by the study device. Secondary outcomes included tympanic cooling and absence of rectal cooling, CT, CT perfusion, and histologic evidence of injury in the short- and medium-term.

Results

10 animals (54.7 kgs.+/- 19.4 SD) underwent non-survival evaluation, 3 underwent delayed evaluation (56.1 kgs.+/-6.4). From baseline to cooling period, intranasal (38.1 °C vs. 34.9 °C, p = 0.01), intracranial (38.6 °C vs. 37.2 °C, p = 0.01) and tympanic temperatures (38.3 °C vs. 37.5 °C, p = 0.01) decreased, while rectal temperature remained unchanged.After cooling, nasal temperature (34.9 °C vs. 37.6 °C, p = 0.004) and intracranial temperatures (37.2 °C vs. 37.9 °C, p = 0.01) increased. Rectal temperatures remained unchanged. A mixed effects model showed association between temperature and study period (p<0.0001), temperature and probe location (p = 0.002), with interaction between study period and probe location (p<0.0001). There was no evidence of injury to the snout or CT cerebral perfusion metrics, or in histologic end points.

Conclusion

This novel, non-invasive, intranasal high flow cooled air device provides isolated brain and head cooling without any evidence injury in the short or medium term.

Exploring Intra-arterial Contrast Administration for Intraoperative Imaging Using a Swine Model

Intraoperative computed tomography (CT) imaging with endovascular delivery of intra-arterial (IA) contrast could potentially provide higher attenuation with lower contrast volumes than intravenous (IV) administration. We aimed to compare IA and IV contrast use for organ-specific CT abdominal imaging. Five anesthetized swine had external jugular and brachial artery access with ascending aortic pigtail placement. An IV protocol was 100 mL at 5 mL/sec over 20 sec vs 50 mL of IA contrast at 5 mL/sec over 10 sec. Region-of-interest markers were applied to anatomical regions to measure attenuation (HU) over time. IA and IV contrast protocols achieved adequate aortic opacification (IA, 455 ± 289 vs IV, 450 ± 114 HU). The IA contrast aortic attenuation curve reached peak attenuation compared with IV contrast (IA, 8 vs 23 sec; P < .001). Time to peak attenuation was similar between IA and IV contrast in the portal vein (IA, 38 vs IV, 42 sec, P = .25). IA administration achieved a superior contrast-to-noise ratio (CNR) in less time compared with IV (R2 = .94; P < .001). IA contrast achieved adequate opacification with less bolus broadening and a superior CNR compared with IV contrast while using a smaller contrast volume for directed organ-directed imaging.

Lower Extremity Extracorporeal Distal Revascularization in a Swine Model of Prolonged Extremity Ischemia

BACKGROUND

Acute arterial occlusion of the lower extremity is a time-dependent emergency that requires prompt revascularization. Lower extremity extracorporeal distal revascularization (LEEDR) is a technique that can be initiated bedside when definitive therapy is delayed. The aim of this study is to evaluate this technique in a swine model of prolonged extremity ischemia.

METHODS

Anesthetized swine underwent right femoral and left posterior tibial artery cannulation, left iliac venous flow monitoring (mL/min), and continuous left anterior compartment pressure (CP) monitoring (mm Hg). The iliac artery was clamped for 6 hr. LEEDR animals underwent 5 hr of extracorporeal femoral-to-tibial blood flow at 150 mL/min; controls had no intervention. At 6 hr, LEEDR was discontinued, iliac flow restored, and anterior CP monitored for 3 hr.

RESULTS

Baseline characteristics were similar across both the groups. Iliac clamping saw an expected fall in iliac venous flow (258 ± 30 to 82 ± 19; P < 0.001). LEEDR resulted in a rise in iliac venous flow (82 ± 20 to 181 ± 16; P < 0.001); control arm flow remained reduced (71 ± 8; P < 0.001). Once inflow was restored, venous flow returned to baseline. Revascularization provoked a higher peak CP in the control arm versus in the LEEDR group (25 ± 5 vs. 6 ± 1; P = 0.02).

CONCLUSIONS

An extracorporeal circuit can temporarily revascularize an extremity in a swine model of prolonged ischemia, mitigating reperfusion injury and maintaining normal CPs. This concept should undergo further evaluation as a bedside tool to mitigate extremity ischemia prior to definitive revascularization.

Partial vs Full Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) in a Swine Model of Raised Intracranial Pressure and Hemorrhagic Shock

BACKGROUND

Partial resuscitative endovascular balloon occlusion of the aorta (pREBOA) is a potential method to mitigate the ischemia observed in full REBOA (fREBOA). However, the effect of pREBOA on cerebral perfusion in the setting of raised intracranial pressure (rICP) is unknown. The aim was to evaluate the effects of no REBOA (nREBOA) vs pREBOA vs fREBOA on cerebral perfusion in a swine model of rICP and hemorrhagic shock.

STUDY DESIGN

Anesthetized swine (n = 18) underwent instrumentation. Controlled hemorrhage was performed over 30 minutes. rICP was achieved using an intracranial Fogarty catheter inflated to achieve an ICP of 20 mmHg. Animals underwent intervention for 30 minutes, followed by resuscitation. The primary outcome was cerebral perfusion measured by ICP (millimeters of mercury), cerebral perfusion pressure (CPP; millimeters of mercury), and cerebral blood flow (CBF; milliliters per minute per 100 g) derived from CT perfusion. The secondary outcomes included hemodynamics and lactate (millimoles per liter).

RESULTS

The peak ICP of pREBOA animals (22.7 ± 2.5) was significantly lower than nREBOA and fREBOA. pREBOA CPP was significantly higher compared with nREBOA and fREBOA during resuscitation. The pREBOA CBF was greater during intervention and resuscitation compared with nREBOA (p < 0.001). Systolic blood pressure was similar between pREBOA and fREBOA, and coronary perfusion was significantly greater in pREBOA. fREBOA had significantly higher lactate during the intervention (9.3 ± 1.3) and resuscitation (8.9 ± 3.5) compared with nREBOA and pREBOA.

CONCLUSION

pREBOA produced greater cerebral perfusion, as demonstrated by more favorable CPP, CBF, and ICP values. fREBOA was associated with metabolic derangement and diminished pressure during resuscitation. pREBOA is superior to fREBOA in a swine model and should be considered over fREBOA for aortic occlusion.

Characterization of cerebral blood flow during open cardiac massage in swine: Effect of volume status

Introduction: Patients in cardiac arrest treated with resuscitative thoracotomy and open cardiac massage (OCM) have high rates of mortality with poor neurological outcomes. The aim of this study is to quantitate cerebral perfusion during OCM using computed tomography perfusion (CTP) imaging in a swine model of normo- and hypovolemia.

Methods: Anesthetized swine underwent instrumentation with right atrial and aortic pressure catheters. A catheter placed in the ascending aorta was used to administer iodinated contrast and CTP imaging acquired. Cerebral blood flow (CBF; ml/100 g of brain) and time to peak (TTP; s) were measured. Animals were then euthanized by exsanguination (hypovolemic group) or potassium chloride injection (normovolemic group) and subjected to a clamshell thoracotomy, aortic cross clamping, OCM, and repeated CTP. Data pertaining to peak coronary perfusion pressure (pCoPP; mmHg) were collected and % CoPP > 15 mmHg (% CoPP; s) calculated post hoc.

Results: Normovolemic animals (n = 5) achieved superior pCoPP compared to the hypovolemic animals (n = 5) pCoPP (39.3 vs. 12.3, p < 0.001) and % CoPP (14.5 ± 1.9 vs. 30.9 ± 6.5, p < 0.001). CTP acquisition was successful and TTP elongated from spontaneous circulation, normovolemia to hypovolemia (5.7 vs. 10.8 vs. 14.8, p = 0.01). CBF during OCM was similar between hypovolemic and normovolemic groups (7.5 ± 8.1 vs. 4.9 ± 6.0, p = 0.73) which was significantly lower than baseline values (51.9 ± 12.1, p < 0.001).

Conclusion: OCM in normovolemia generates superior coronary hemodynamics compared to hypovolemia. Despite this, neither generates adequate CBF as measured by CTP, compared to baseline. To improve the rate of neurologically intact survivors, novel resuscitative techniques need to be investigated that specifically target cerebral perfusion as existing techniques are inadequate.

Whole Blood Selective Aortic Arch Perfusion for Exsanguination Cardiac Arrest: Assessing Myocardial Tolerance to the Duration of Cardiac Arrest

INTRODUCTION

Selective aortic arch perfusion (SAAP) is an endovascular technique that consists of aortic occlusion with perfusion of the coronary and cerebral circulation. It been shown to facilitate return of spontaneous circulation (ROSC) after exanguination cardiac arrest (ECA), but it is not known how long arrest may last before the myocardium can no longer be durably recovered. The aim of this study is to assess the myocardial tolerance to exsanguination cardiac arrest before successful ROSC with SAAP.

METHODS

Male adult swine (n = 24) were anesthetized, instrumented, and hemorrhaged to arrest. Animals were randomized into three groups: 5, 10, and 15 min of cardiac arrest before resuscitation with SAAP. Following ROSC, animals were observed for 60 min in a critical care environment. Primary outcomes were ROSC, and survival at 1-h post-ROSC.

RESULTS

Shorter cardiac arrest time was associated with higher ROSC rate and better 1-h survival. ROSC was obtained for 100% (8/8) of the 5-min ECA group, 75% (6/8) of the 10-min group, 43% (3/7) of the 15-min group (P = 0.04). One-hour post-ROSC survival was 75%, 50%, and 14% in 5-, 10-, and 15-min groups, respectively (P = 0.02). One-hour survivors in the 5-min group required less norepinephrine (1.31 mg ± 0.83 mg) compared with 10-SAAP (0.76 mg ± 0.24 mg), P = 0.008.

CONCLUSION

Whole blood SAAP can accomplish ROSC at high rates even after 10 min of unsupported cardiac arrest secondary to hemorrhage, with some viability beyond to 15 min. This is promising as a tool for ECA, but requires additional optimization and clinical trials.Animal Use Protocol, IACUC: 0919015.

Characterizing Brain Perfusion in a Swine Model of Raised Intracranial Pressure

INTRODUCTION

Perfusion of the brain is critical, but this can be compromised due to focal space occupying lesions (SOL). SOLs can raise intracranial pressure (ICP), resulting in reduced cerebral blood flow (CBF). Most gyrencephalic models of brain injury focus on parenchymal injury, with few models of acutely elevated ICP. We hypothesized that we could employ a SOL technique to develop a titratable ICP model and sought to quantitate the resulting decrease in brain perfusion.

METHODS

Six swine were anesthetized and instrumented. A Fogarty balloon catheter was inserted intracranially. Blood CO₂ partial pressure was maintained between 35 and 45 mmHg. The Fogarty balloon was infused with normal saline at 1 mL/min to ICP targets of 10, 20, 30, and 40 mmHg. CBF (mL/100 g/min) were assessed at each ICP level using computed tomography perfusion (CTP). Data are presented as the mean ± standard deviation with all pressures measured in mmHg. CBF values were compared between baseline and each ICP level using analysis of variance.

RESULTS

Baseline ICP was 5 ± 2 and systolic blood pressure was 106 ± 7. Balloon volumes (mL) required to achieve each incremental ICP level were 2.4 ± 0.5, 4.9 ± 1.7, 7.6 ± 1.6, and 9.9 ± 1.7. CBF decreased with each raised ICP level, with CBF being significantly less than baseline at ICP values of 30 (56.1 ± 34.7 versus 20.6 ± 11.0, P < 0.05) and 40 (56.1 ± 34.7 versus 6.5 ± 10.6, P < 0.05).

CONCLUSIONS

An intracranial balloon catheter can be used to increase ICP, delivering a proportionate reduction in CBF. This model can be used in the future studies to examine adjuncts that manipulate intracranial pressure and their effect on brain perfusion.

The Underlying Cardiovascular Mechanisms of Resuscitation and Injury of REBOA and Partial REBOA

Introduction: Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) is used for aortic control in hemorrhagic shock despite little quantification of its mechanism of resuscitation or cardiac injury. The goal of this study was to use pressure-volume (PV) loop analysis and direct coronary blood flow measurements to describe the physiologic changes associated with the clinical use of REBOA. Methods: Swine underwent surgical and vascular access to measure left ventricular PV loops and left coronary flow in hemorrhagic shock and subsequent placement of occlusive REBOA, partial REBOA, and no REBOA. PV loop characteristics and coronary flow are compared graphically with PV loops and coronary waveforms, and quantitatively with measures of the end systolic and end pressure volume relationship, and coronary flow parameters, with accounting for multiple comparisons. Results: Hemorrhagic shock was induced in five male swine (mean 53.6 ± 3.6 kg) as demonstrated by reduction of stroke work (baseline: 3.1 vs. shock: 1.2 L*mmHg, p < 0.01) and end systolic pressure (ESP; 109.8 vs. 59.6 mmHg, p < 0.01). ESP increased with full REBOA (178.4 mmHg; p < 0.01), but only moderately with partial REBOA (103.0 mmHg, p < 0.01 compared to shock). End systolic elastance was augmented from baseline to shock (1.01 vs. 0.39 ml/mmHg, p < 0.01) as well as shock compared to REBOA (4.50 ml/mmHg, p < 0.01) and partial REBOA (3.22 ml/mmHg, p = 0.01). Percent time in antegrade coronary flow decreased in shock (94%-71.8%, p < 0.01) but was rescued with REBOA. Peak flow increased with REBOA (271 vs. shock: 93 ml/min, p < 0.01) as did total flow (peak: 2136, baseline: 424 ml/min, p < 0.01). REBOA did not augment the end diastolic pressure volume relationship. Conclusion: REBOA increases afterload to facilitate resuscitation, but the penalty is supraphysiologic coronary flows and imposed increase in LV contractility to maintain cardiac output. Partial REBOA balances the increased afterload with improved aortic system compliance to prevent injury.

A technical and data analytic approach to pressure-volume loops over numerous cardiac cycles

Cardiac pressure-volume (PV) loop analysis is the reference standard for studying the cardiovascular implications of clinical perturbations (eg, heart failure, aortic occlusion, hypovolemia) and is a benchmark for comparisons with noninvasive alternatives (eg, ultrasound, magnetic resonance imaging). Historically, most PV loop analyses were of individual cardiac cycles for which the options to analyze PV loops using off-the-shelf software were limited, and home-grown analysis software often lacked peer review or code-sharing. Our aim was to describe a start-to-finish implementation of PV loops for determination of hemodynamic parameters in swine, to provide technical advice for vascular access and proceduralization, and to describe data capture, curation, preprocessing, and analysis of raw PV time data. We have provided a novel data analytic method to programmatically analyze raw PV loop data beyond single cardiac cycles and real, raw swine PV loop data and the accompanying MATLAB (MathWorks, Inc, Natick, Mass) code as an example of how to process and analyze raw data directly.