Endovascular Perfusion Augmentation for Critical Care Decreases Vasopressor Requirements while Maintaining Renal Perfusion

INTRODUCTION

Shock that is refractory to standard resuscitation (vasopres- sors/crystalloid) is a challenging clinical scenario. The search for adjuncts has been an intense area of research to improve the morbidity and mortality for patients with these conditions. Adjuncts in septic shock such as thiamine, steroids, and ascorbic acid have been studied with debated efficacy, adjuncts for ischemia-reperfusion like valproic acid and hydrogen sulfide have been investigated in animal models, and post-cardiac arrest syndrome mimics an inflammatory shock model with therapeutic cooling as an adjunct (1–4). Supra-celiac aortic cross clamp for surgical aortic reconstruction, return of spontaneous circulation after cardiopulmonary resuscitation (CPR), and Resuscitative Endovascular Balloon Occlusion of the Aorta (REBOA) for trauma results in a severe ischemia-reperfusion injury (IRI) (5–7). While there has been intense focus on decreasing the physiologic burden with decreased cross clamp times, improved quality of CPR, and modified REBOA deflation schedules (intermittent or partial), less effort has been spent on exploration and translation of novel adjuncts in management of the resulting ischemia reperfusion injury (8–10).

Endovascular Perfusion Augmentation for Critical Care (EPACC) is a novel mechanical automated endovascular resuscitation adjunct for the treatment of hyperdynamic distributive shock (11). In early translational experience, inflation of a balloon catheter in the descending thoracic aorta facilitated proximal hemodynamic homeostasis and because the balloon is operating at low inflation volumes (<2 mL), this can be achieved with minimal degradation in distal tissue perfusion (12). Because balloon control is automated, the EPACC catheter must work in concert with an automated fluid and drug delivery platform to execute rule-based algorithms using realtime physiologic inputs. When combined, EPACC and automated standardized critical care integrates mechanical and pharmacologic pressure augmentation with fluid administration to achieve resuscitation endpoints.

In the initial proof-of-concept study, endovascular pressure augmentation produced excellent proximal hemodynamic homeostasis, but this came at the expense of reduced distal organ perfusion (11). In retrospect, this initial approach relied too heavily on balloon support instead of a balance of balloon pressure augmentation in combination with crystalloid, leaving the animals intravascularly depleted. In that study, the triggers for further intervention and the interplay of balloon support along with crystalloid and vasopressor administration were not sufficiently nuanced. As such, there was a clear need to optimize the EPACC platform to be less fluid restrictive.

These initial results raised fundamental questions about the ideal parameters for EPACC and its overall viability as a resuscitation strategy. What balloon volume range supports balanced resuscitation requirements and downstream organ perfusion? And overall, does EPACC provide an inherent physiologic benefit beyond immediate hemodynamic support? In an effort to address these complex questions and further explore the utility of this concept, we designed a study with various algorithmic approaches to EPACC combined with automated standardized critical care to manage post-REBOA IRI in a translational model. We hypothesize that EPACC would promote improved hemodynamic stability and decrease crystalloid/vasopressor requirements compared with a standardized resuscitation algorithm. Additionally, a lower threshold for fluid administration and vasopressor uptitration during EPACC would result in less ischemic debt compared with a higher threshold (similar to the original approach) for triggering standard resuscitation techniques.

METHODS Overview

The Institutional Animal Care and Use Committee at Wake Forest Baptist Medical Center approved this study. All animal care and use were in strict compliance with the Guide for the Care and Use of Laboratory Animals in a facility accredited by AAALAC.

Animal preparation

Healthy adult castrate male and non-pregnant female Yorkshire-cross swine (Oak Hill Genetics, Ewing, Ill) between 65 kg and 75 kg were premedicated with 5 mg/kg to 7 mg/kg intramuscular tiletamine/zolazepam (Telazol, Zoetic, Olot, Spain). After isoflurane induction and endotracheal intubation, animals were maintained with ∼2% isoflurane, 100% oxygen at 2L/min mixed with room air at 1.5 L/min, and ventilated to maintain end-tidal carbon dioxide at 35 mm Hg to 45 mm Hg. All animals received an initial 1L intravenous (IV) crystalloid bolus during setup and maintenance IV fluids (Plasma-Lyte A, Baxter Healthcare Corporation, Deerfield, Ill). To offset the vasodilatory effects of isoflurane, a norepinephrine infusion was titrated before experimentation to achieve a target mean arterial pressure (MAP) >60 mm Hg. An underbody warmer was used to maintain core body temperature between 35.8°C and 37.8°C. After a laparotomy and placement of a cystostomy tube, a splenectomy was performed to minimize hemodynamic variation from autotransfusion. Perivascular flow probes were placed around the supraceliac aorta, left carotid artery, and the left renal artery (Transonic, Ithaca, NY).

Vascular instrumentation

A 7F arterial sheath was placed in the right common femoral artery for distal arterial blood pressure measurement and a 12F arterial sheath was placed in the left common femoral artery to introduce the aortic balloon catheter. A 9F venous catheter was placed in the left femoral vein for hemorrhage, transfusion, and resuscitation. Bilateral jugular veins were cannulated for maintenance fluid/drug administration and central venous pressure measurements. A 5F sheath was placed in the left axillary artery for proximal blood pressure measurements. The right brachial artery was cannulated with a 5F sheath for frequent lab draws. A 7F custom compliant aortic balloon catheter was positioned just distal to the aortic flow probe in the distal descending thoracic aorta. The abdomen was closed to minimize insensible fluid losses. IV heparin was administered (50 units/kg bolus and 10 units/kg/h continuous rate) to achieve an activated clotting time of ∼100 s based on a rate developed from prior studies.

Data collection

Physiologic measurements of proximal and distal aortic pressure, central venous pressure, and aortic/renal/carotid blood flow were collected in real time with a multichannel data acquisition system (Powerlab, AD Instruments, Colorado Springs, Colo). Blood chemistries, arterial blood gases, urine, and serum were collected at start of hemorrhage (T0), end of wean for occlusion (T75), end of endovascular support (T270), and end of experiment (T330). Additional time points were obtained: T29, T45, T54, T65, T90, T105, T120, T150, T180, T210, T240, and T300.

A necropsy was performed with collection of bilateral kidney specimens. For each renal tissue section, 10 random fields of renal cortex containing at least 1 to 2 glomerulus were assessed at 20 times magnification by a blinded veterinary pathologist and scored on a 0 to 4 point scale for tubular lesions. Normal-0, Mild-1: very mild vacuolation of cytoplasm, occasional swelling of cells, blebbing apical portion of the cytoplasm. Brush borders are still visible in most areas. Moderate-2: more marked swelling, vacuolization and blebbing of cells, presence of amorphous eosinophilic material in the lumen, loss of brush borders. Marked-3: cellular degeneration and cell sloughing in the tubular lumen. Severe-4: shrunken cells with hypereosinophilic cytoplasm and pyknotic nuclei, cellular casts. See supplemental Figure 1, https://links.lww.com/SHK/B416 for an example of lesions that were seen in the histologic specimens for each score.

Automated care platform

The automated resuscitation platform consisted of a custom microprocessor that received continuous sidestream physiologic data. The unit wirelessly transmitted instructions based on predefined algorithms to three peripheral devices: a custom automated balloon syringe pump, a multichannel peristaltic infusion pump for maintenance fluids and drugs (Ismatec, Cole-Palmer, Vernon Hills, Ill), and a peristaltic pump for blood transfusion and crystalloid boluses (Masterflex, Cole-Palmer, Vernon Hills, Ill).

Initial injury phase

Animals underwent a controlled 30% total blood volume hemorrhage over 30 min and collected in citrated bags. Then the aortic balloon was inflated to complete occlusion (absence of distal aortic flow) for 30 min. Fifteen minutes into occlusion, a 20-min infusion of calcium gluconate was initiated to prevent hypocalcemia from citrated autologous whole blood retransfusion. At T55, animals were transfused back to 95% of baseline blood volume over 18 min.

Critical care phase

At the initiation of the critical care phase (T60) as blood was being actively transfused, balloon support was weaned to maintain a proximal mean pressure of at least 65 mm Hg over 15 min. After the wean, animals were randomized: standardized critical care (SCC), EPACC with high threshold (EPACC-High), and EPACC with low threshold (EPACC-Low). For all groups, we defined 60 mm Hg to 70 mm Hg as the target MAP range for the critical care interventions (Fig. 1).

F1Fig. 1:

The experimental flow and design. EPACC indicates endovascular perfusion augmentation for critical care.

Animals in the EPACC groups received automated endovascular support based on custom closed-loop adaptive feedback algorithms to control the volume of the aortic balloon catheter. The aortic balloon support worked in conjunction with the automated intravenous crystalloid boluses and vasopressor titration for the 3.5 h of the intervention phase. At T270, the balloon was removed, followed by one additional hour of SCC alone.

Animals in the SCC group underwent balloon removal at T75. Automated crystalloid boluses and norepinephrine titration ensued based on a protocol to maintain normotension throughout the critical care phase, ending at T330.

Standardized critical care

A custom algorithm was designed based on heuristics of critical care utilizing mean arterial pressure, central venous pressure, and fluid responsiveness (delta MAP of 5 mm Hg) to test bolus as input variables. The algorithm had the option to provide weight-based crystalloid boluses (5 mL/kg) and titration of norepinephrine for vasoactive support (Supplemental Figure 2, https://links.lww.com/SHK/B417).

EPACC balloon control algorithm

In earlier versions, EPACC control inputs were based on proximal blood pressure and direct distal aortic flow measurements. Since distal aortic flow is difficult to measure clinically, the present study utilized a calculated percent gradient defined as (proximal MAP — distal MAP) /proximal MAP) × 100. The goal of the endovascular balloon algorithm is to maintain proximal normotension (proximal MAP > 62.5 mm Hg and < 67.5 mm Hg) by dynamically adjusting the balloon volume, serving as a mechanical resistance in the aorta. The gradient across the balloon was used as a threshold (low vs. high) for the addition of vasopressors and crystalloid in the same manner that SCC was applied for hypotension. Two differing gradient parameters in this study (EPACC-Low: 17% and EPACC-High: 33%) represented the threshold values at which fluid boluses were administered or vasopressors were uptitrated. The higher gradient represented more balloon support (less distal blood flow) and vice versa. An increasing EPACC gradient therefore served as a surrogate marker for declining hemodynamics, analogous to progressive proximal hypotension in the SCC group.

Data analysis

Data analysis was performed with Python Version 3.7 (Python Software Foundation, Wilmington Delaware) and R statistical software (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are presented as medians with interquartile ranges. Kurskal-Wallis one-way analysis of variance was used for data that were not normally distributed with reflex Dunn test to perform pairwise testing to identify statistical differences. Dichotomous and categorical variables were analyzed by chi-square statistics. Statistical significance was set at P < 0.05.

Exclusion criteria includedapre-procedure white blood cell count > 25 109/L, expiration before the end of study as defined by a proximal MAP <20 mmHg for 5 min during any time point or technical deviations from the established protocol. Additionally, animals were excluded if a norepinephrine dose rate of greater than 0.1 mcg/kg/min was required to maintain normotension for more than 10 min prior to experimentation or if the norepinephrine dose rate to maintain normo- tension immediately prior to the beginning of the experiment was greater than 0.06 mcg/kg/min, signifying abnormal baseline hemodynamics.

Based on observations from the prior EPACC study, the primary outcome was to detect a difference in terminal creatinine with secondary outcomes being maintenance of proximal normotension with a subsequent decrease in resuscitation requirements (crystalloid and vasopressors). Power analysis was designed to detect an effect size ofa25% increase in creatinine at the end of the study between the SCC versus the EPACC groups. Using a baseline creatinine of 1.68 mg/dL ± 0.25 in the SCC group, and a planned one-way ANOVA, a power of 90% and an alpha of 0.05, a total sample size of six animals per group will be needed.

RESULTS Baseline characteristics

15 animals were included in the analysis. The median weight for the SCC was 70.9 kg (interquartile range: 65.4 kg-78.8 kg), EPACC-Low was 69.8 kg (interquartile range: 67.9 kg-73.9 kg), and EPACC-High 73.6 kg (interquartile range: 71.5 kg-74 kg) and no differences between the groups P = 0.81. There were no differences in baseline characteristics or initial laboratory parameters between any of the groups (Table 1). At the end of hemorrhage all groups experienced a similar decrease in proximal mean arterial pressure, P = 0.81. During the injury phase the proportion of time with distal mean arterial pressure <62.5 mm Hg was similar, ranging from 95.0% to 98.3% (P = 0.87).

Table 1 - Baseline characteristics and standardized injury metrics Results (median with interquartile range) Standardized critical care EPACC low threshold EPACC high threshold P value Number of animals 5 5 5 Number of excluded animals 3 0 3 0.267 Females 3 3 3 1.0 Weight (kg) 70.9 (65.4–78.8) 69.8 (67.9–73.9) 73.6 (71.5–74) 0.810 Initial pH 7.426 (7.4–7.43) 7.414 (7.41–7.48) 7.467 (7.42–7.49) 0.453 Initial Hemoglobin (g/dL) 9.9 (9.9–10.5) 10.2 (9.5–10.2) 9.9 (9.5–9.9) 0.529 Initial WBC (109/L) 16.99 (16.29–20.07) 20.47 (15.75–20.61) 14.15 (14.04–15.26) 0.651 Initial creatinine (mg/dL) 1.8 (1.5–1.8) 1.5 (1.4–1.8) 1.5 (1.5–1.6) 0.554 Initial blood urea nitrogen (mg/dL) 9 (8–9) 7 (6–8) 6 (6–8) 0.059 Initial potassium (mmol/L) 3.7 (3.7–3.7) 3.5 (3.4–3.7) 3.8 (3.7–3.8) 0.087 Initial glucose (mg/dL) 95 (89–102) 112 (106–115) 103 (96–105) 0.160 Baseline proximal mean arterial pressure (mm Hg) 63.2 (62.34–67.72) 59.81 (59.81–68.22) 62.23 (59.36–69.23) 0.970 Baseline aortic flow (mL/min/kg) 38.05 (36.18–43.65) 41.59 (40.11–42.01) 46.70 (45.81–48.18) 0.093 Injury phase: proportion of time with distal hypotension (<62.5 mm Hg) 98.3% (83%-100%) 96.7% (97%-100%) 95.0% (88%-100%) 0.870 Injury phase: proximal mean arterial pressure at end of hemorrhage phase (mm Hg) 30.15 (28.73–41.64) 27.69 (26.81–36.17) 36.66 (31.61–33.50) 0.403

EPACC indicates endovascular perfusion augmentation for critical care; WBC, white blood cell count.

Our power analysis predicted an effect size of 25% and need for six animals per group, but only five animals at the end of the study met criteria for inclusion in each group. Three animals in the SCC (37.5%) and EPACC-High (37.5%) group died during setup or were excluded during instrumentation based on predetermined criteria; this was slightly higher than expected death rate (14%) and no difference between groups, P = 0.25. Also the observed effect size on the primary endpoint was 17% and based on an 80% power and significance level of 0.05, 112 animals per group would have been required for adequate power, which was not feasible nor justifiable. Due to these results during the interim analysis of the results and based on the number of animals needed to detect difference the study was terminated with only five animals per group.

Resuscitation

During the intervention phase with balloon support, the percentage of time with active balloon support (balloon volume > 0 mL) was significantly greater in the EPACC-High group compared with the EPACC-Low (96.27% vs. 69.51%; P < 0.01). During this same period, the EPACC-High groups had an average of 1.19 adjustments per minute over the intervention phase compared with the EPACC-Low group with an average of 1.41 adjustments per minute, P = 0.42. During the intervention phase with balloon support, the percentage of time spent at proximal normotension (> = 60 mm Hg for consecutive 1 min bins) was significantly greater in the EPACC groups (EPACC-High: 99.5% vs. EPACC-Low: 97.6%; P = 0.06) compared with SCC (82.4%; P = 0.001 and P = 0.04 respectively) (Table 2).

Table 2 - Hemodynamic results during the study period Result (median with interquartile range) Standardized critical care EPACC low threshold EPACC high threshold P value S: SCC H: highL: low Distal mean arterial pressure during intervention phase (mm Hg) 61.05 (56.36–63.16) 54.39 (49.8–58.82) 45.23 (42.31–49.92) S vs. H: <0.001S vs. L: <0.001H vs L: <0.001 Distal aortic flow during intervention phase (mL/min/kg) 57.00 (49.28–70.44) 48.01 (37.96–60.23) 44.92 (35.39–54.12) S vs. H: <0.001S vs. L: <0.001H vs L: 0.029 Proximal mean arterial pressure during intervention phase (mm Hg) 64.25 (61.83–66.43) 66.27 (64.90–67.72) 64.87 (64.12–65.71) S vs. H: 0.031S vs. L: <0.001H vs L: <0.001 Proximal carotid flow during intervention phase (mL/min/kg) 3.80 (3.29–4.31) 4.55 (3.75–5.77) 4.49 (3.99–5.09) S vs. H: <0.001S vs. L: <0.001H vs L: 0.168 Percent time spent at proximal nomortension (>60 mm Hg) during intervention phase 82.4% (81%-90%) 97.6% (98%-99%) 99.5% (99.5%-100%) S vs. H: 0.001S vs. L: 0.038H vs L: 0.056 Renal artery flow during intervention phase (mL/min/kg) 1.77 (1.34–2.18) 2.29 (2.04–2.81) 1.25 (0.91–1.90) S vs. H: <0.001S vs. L: <0.001H vs L: <0.001 Ratio of renal to distal aortic flow (%) 2.78% (1.78–3.65) 3.94% (3.07–5.49%) 2.97% (2.01–3.92) S vs. H: 0.046S vs. L: <0.001H vs. L: <0.001 Total urine output during intervention phase (mL/kg/h) 2.85 (2.43–3.18) 2.92 (2.38–3.06) 1.13 (0.91–1.42) S vs. H: 0.020S vs. L: 0.472H vs L: 0.012 Cumulative pressor dose at end of intervention phase (mcg/kg) 59.45 (31.32–61.61) 16.23 (14.56–29.42) 13.72 (12.42–16.87) S vs. H: 0.013S vs. L: 0.049H vs L: 0.218 Cumulative pressor dose at end of experiment (mcg/kg) 80.89 (41.3–87.64) 27.20 (25.15–49.15) 24.07 (19.52–24.23) S vs. H: 0.009S vs. L: 0.103H vs L: 0.102 Cumulative number of boluses at end of intervention 18 (18–23) 12 (9–19) 12 (10–14) 0.146 Pathologic renal injury scoring 2.32 (2.16–2.41)N = 10 2.32 (2.03–2.60)N = 10 2.48 (2.11–2.69)N = 10 0.971

EPACC indicates endovascular perfusion augmentation for critical care; SCC, standardized critical care.

During the intervention phase, EPACC animals required significantly less intravenous norepinephrine (EPACC-Low: 16.23 mcg/kg vs. EPACC-High: 13.72 mcg/kg; P = 0.218) compared with SCC (59.45 mcg/kg, P = 0.049 and P = 0.013). There was no statistical difference between the number of weight based crystalloid boluses each group received, P = 0.15 (Table 2 and Fig. 2).

F2Fig. 2:

Lactate, blood levels at the start of experiment (0), end of injury (75), end of intervention care (270), and end of critical care phase (330) (A). Creatinine, blood levels at start of experiment (0), end of injury (75), end of intervention phase (270), and end of critical care phase (330) (B). Cumulative norepinephrine dose at end of intervention phase, ( ∗ ) P value of 0.009, ( ∗∗ ) P value of 0.049 (C). Cumulative number of weight based (5 mL/ kg) crystalloid boluses at end of intervention phase (D). All plots are presented as box and whisker plots represented by median with interquartile range as the box and 1st and 4th quartile as the whiskers with outliers as dots and the mean as diamonds. EPACC indicates endovascular perfusion augmentation for critical care; EPACC-High, high threshold; EPACC-Low, low threshold; SCC, standardized critical care.

Laboratory values

There was no statistical difference between all groups (SCC, EPACC-High, and EPACC-Low) for lactate or creatinine at the start of experiment (T0), end of injury (T75), end of intervention (T270), end of experiment (T330), or peak values observed during the study (Table 3 and Fig. 2).

Table 3 - Laboratory results Result (median with interquartile range) Timepoint (min) Standardized critical care EPACC low threshold EPACC high threshold P value Lactate, blood (mmol/L) 0 2.36 (2.33–2.83) 3.12 (2.62–3.24) 2.67 (2.56–3.21) 0.264 75 9.11 (8.86–9.49) 8.56 (8.42–9.54) 8.84 (8.73–9.21) 0.914 270 4.67 (4.35–5.02) 4.87 (4.79–5.26) 5.48 (3.91–5.52) 0.677 330 4.91 (4.51–5.04) 4.74 (4.55–5.03) 5.30 (3.58–5.38) 0.961 Peak 9.77 (8.86–10.17) 9.87 (9.39–10.07) 9.77 (8.86–10.17) 0.992 Creatinine, blood (mg/dL) 0 1.8 (1.5–1.8) 1.5 (1.4–1.8) 1.5 (1.5–1.6) 0.554 75 2.1 (2–2.4) 1.9 (1.9–2.4) 2.0 (2–2.2) 0.534 270 2.0 (1.9–2.1) 1.8 (1.7–1.8) 2.2 (2–2.6) 0.110 330 2.1 (2.1–2.2) 1.8 (1.7–1.9) 2.0 (2.0–2.3) 0.341 Peak 2.2 (2.1–2.4) 2.0 (1.9–2.4) 2.2 (2.2–2.6) 0.533

EPACC indicates endovascular perfusion augmentation for critical care.


Hemodynamics

Consecutive 10 min bins were sampled during the intervention phase T60 to T270 for analysis of hemodynamics (Fig. 2). During the intervention phase, median distal blood pressure was decreased for both EPACC-High and EPACC-Low compared with SCC (45.23 mm Hg and 54.39 mm Hg vs. 61.05 mm Hg; P < 0.001 and P < 0.001). Distal aortic flow was similarly decreased for both EPACC-High and EPACC-Low compared with SCC (44.92 mL/min/kg and 48.01 mL/min/kg vs. 57.00 mL/kg/min; P < 0.001 and P < 0.001). When examining the proximal hemodynamics during intervention phase there was a significant increase in proximal mean arterial pressure in EPACC (EPACC-Low: 66.27 mm Hg; EPACC- High: 64.87 mm Hg) compared with SCC (64.25 mm Hg, P < 0.001 and P = 0.03). There was also an associated increase in carotid flow during intervention phase for EPACC (EPACC- Low: 4.55 mL/min/kg; EPACC-High: 4.49 mL/min/kg) compared with SCC (3.80 mL/min/kg, P < 0.001 and P < 0.001) (Table 2 and Fig. 3).

F3Fig. 3:

Proximal mean aortic blood pressure (A), weight-based carotid flow (B), distal mean aortic blood pressure (C), and weightbased aortic flow (D). All graphs are presented as averages of 10 min blocks with displayed values representing the median and interquartile range as error bars. The dotted line represents endovascular aortic balloon removal in SCC. The long dashed line represents endovascular aortic balloon removal in EPACC groups. EPACC indicates endovascular perfusion augmentation for critical care; EPACC-High, high threshold; EPACC-Low, low threshold; SCC, standardized critical care.

Renal indices

The weight-based median renal blood flow in the SCC group was 1.77 mL/min/kg during the intervention phase. The weightbased median renal blood flow was statistically significantly decreased in the EPACC-High group (1.25 mL/min/kg) compared with SCC (P = 0.001). In contrast, the EPACC-Low group median renal blood flow (2.29 mL/min/kg) was statistically increased compared with SCC with a P < 0.001. The median ratio of renal flow to distal aortic flow for SCC was 2.78% (1.78%-3.65%), EPACC-High was 2.97% (2.01%-3.92%), and EPACC-Low 3.94% (3.07%-5.49%) and were statistically different between all groups, P < 0.05 (Fig. 4). There was a statistically significant decrease in median hourly urine output during critical care in EPACC-High (1.13 mL/kg/h) compared with SCC (2.85 mL/kg/h) and EPACC-Low (3.0 mL/kg/h) with a P value of 0.02 and 0.01 respectively. Bilateral renal pathologic injury (10 samples per animal) scoring from post-mortem did not demonstrate a statistical difference, P = 0.97 (Table 2).

F4Fig. 4:

The median weight-based renal blood flow over time in 10 min blocks with interquartile range portrayed with error bars. The dotted line represents the removal of the endovascular aortic balloon in the control group. The long dashed line represents endovascular aortic balloon removal in the EPACC groups (A). The median weight-based renal blood flow during intervention phase with 1 min bins with interquartile range, (∗) P value of <0.001 (B). All plots are presented as box and whisker plots represented by median with interquartile range as the box and 1st and 4th quartiles as the whiskers with outliers as dots and the mean as diamonds. EPACC indicates endovascular perfusion augmentation for critical care; EPACC-High, high threshold; EPACC-Low, low threshold; SCC, standardized critical care.

DISCUSSION

In a swine model of hemorrhage and post-REBOA IRI, EPACC mechanical pressure augmentation combined with an automated crystalloid and drug delivery platform maintained tight hemodynamic targets while decreasing vasopressor requirements. The decrease in vasopressor dose that was observed in the EPACC-High cohort, likely resulted from the heavy reliance on the aortic balloon for support of proximal blood pressure. However, this reliance on a high level of balloon support (EPACC-High) for proximal hemodynamic augmentation did come at a cost to downstream tissue beds as evidenced by decreased renal perfusion and urine output. In contrast, EPACC-Low resulted in improved renal perfusion when compared with the standardized critical care animals despite the presence of an endovascular balloon acting as a mechanical resistor upstream of the renal arteries. Moreover, this increase in renal perfusion occurred in the EPACC-Low group despite the fact that overall aortic flow beyond the balloon was lower compared with SCC.

While other markers of renal perfusion (serum creatinine and renal histologic injury score) were equivalent, the short duration of this experiment may not have been sufficient to elucidate differences that may manifest 12 to 24 h into a resuscitation. Importantly, we did note that the higher EPACC threshold for critical care interventions (EPACC-High) had a negative impact on both downstream aortic and renal blood flow, as well as urine output. Taken together, these data suggest that mechanical pressure augmentation using a balloon catheter as an adjunct to conventional resuscitation efforts may be beneficial within a range of parameters, but is likely detrimental when used too aggressively. It would seem that in this particular model of ischemic injury and hemorrhagic shock, the amount of hemodynamic support provided by EPACC-Low was able to strike the balance between optimizing upstream and downstream tissue perfusion.

The exact mechanism for these findings is not entirely clear. It is conceivable that EPACC-Low increased renal blood flow by enabling significantly less norepinephrine usage, thus preventing detrimental vasoconstriction of the renal microvascular bed. While EPACC inherently reduced the blood pressure below the level of the balloon, it is possible that this small blood pressure decrement across the balloon in the EPACC- Low group was within a range that still permitted renal autoregulation to maintain ultrafiltration and metabolite clearance. This concept of adaptive endoluminal resistance at the macrovascular level may have important implications for the initial phases of post-REBOA weaning and resuscitation (13).

Historically, the armamentarium for the management of severe shock has relied heavily upon vasopressors and crystalloid to improve oxygen delivery to vitals organs. Yet, the use of vasopressors and large volume crystalloid administration in the context of refractory shock is often counterproductive, with complications including pulmonary edema and peripheral limb ischemia (14–16). EPACC represents a novel, adjunctive approach to the management of post ischemia-reperfusion injury and may enable a balanced resuscitative effort. This type of endovascular resuscitation is novel in the context of traumatic injury and particularly as a follow-on intervention after REBOA. It is conceivable that other distributive shock states such as SIRS, septic shock, anaphylactic shock, neurogenic shock, and post cardiac arrest syndrome may benefit from this approach as well (17–19).

By maintaining stable hemodynamics through the use of automation, EPACC combined with automated critical care enables the provider to focus on more cognitively demanding aspects of complex patient care, as opposed to continually adjusting the aortic balloon. This approach may potentially allow providers to participate in the care of multiple critically ill patients simultaneously or essential hemodynamic support in the setting of prolonged transport times for patients that otherwise may not be stable enough to transport, particularly in the austere environments of military trauma (20).

Despite these initial results, many questions are still unanswered about the clinical applicability of EPACC. First, the optimal duration of EPACC remains unclear based on the timeframe and design of this study. It is conceivable that beyond a certain amount of time, intra-aortic balloon may prove detrimental as a nidus for thrombus formation and distal thromboembolism precluding real-world viability. It is also possible that shorter amounts of dynamic balloon support as a dynamic “post REBOA wean” to overcome the most severe phases of IRI after an endovascular resuscitation with REBOA would actually be advantageous. Additionally, it is still unclear if the early observations of improved renal blood flow using EPACC will translate into improved outcomes at later time points? Furthermore, does EPACC provide a durable physiologic benefit or does it delay the manifestations of an immutable physiologic burden? Longer duration studies at 12 to 24 h are needed to further elucidate the role of EPACC as a resuscitative adjunct and define the criteria for use to understand the risks versus benefits.

Additional lines of inquiry center arise from the implementation of the critical care algorithms that control fluid and vasopressor administration. The automation of critical care allowed for the reduction of provider bias during experimentation but at present, this algorithm does not encompass the full range of therapeutic capabilities or even accurately approximate an expert consensus on the best method of IRI resuscitation. The current algorithms have resulted in the administration of large volume crystalloid fluids, which may not reflect modern resuscitation practices. Further refinement of our automated critical care approach using fluids and vasopressors may result in improved outcomes independent of EPACC support, particularly with the use of alternative vasopressor agents such as vasopressin (21, 22). On the other hand, the large volume fluid requirements in this study may simply reflect the severity of this particular injury model. As such, our model may have resulted in an ischemic burden that is too severe and potentially non-survivable, which would lack clinical relevance.

The main limitation of this study was that it was underpowered to detect differences in our primary endpoint due to smaller than anticipated effect size. Additionally, inherent variability of physiologic responses in translational animal models may have confounded the detection of significant differences for the primary and other secondary endpoints. Unfortunately, the cost and ethical constraints regarding large animal research preclude larger studies that would be needed to assess for more subtle differences.

Future studies will seek to elucidate the optimal duration and guardrails for EPACC therapy to maximize benefits while limiting additional ischemic burden or patient risk. Additionally, refinement of our automated fluid and drug delivery critical care platform is ongoing. Finally, key insights from these studies will inform the conduct of a survival study to understand longer term physiologic outcomes. Such steps are essential for translation of these concepts into clinical practice.

CONCLUSION

Computer-controlled, low-level hemodynamic support with a balloon catheter in the descending thoracic aorta (EPACC), when combined with conventional approaches to critical care treatment (vasopressors and crystalloid), shows promise as a resuscitative adjunct in the treatment of profound hyperdynamic distributive shock after ischemia reperfusion injury. A small amount of EPACC balloon support in concert with conventional resuscitation agents can optimize perfusion to vascular beds both upstream and downstream of the balloon and maintain proximal blood pressure homeostasis in the face of severe IRI. Further evaluation is warranted to establish refined parameters for implementation and to clarify longer term outcomes for this method of adaptive macrovascular resistance.

ACKNOWLEDGMENTS

The authors acknowledge the veterinary management team at the Animal Resource Program at Wake Forest Baptist Medical Center.

REFERENCES 1. Jentzer JC, Vallabhajosyula S, Khanna AK, Chawla LS, Busse LW, Kashani KB. Management of refractory vasodilatory shock. Chest 154 (2):416–426, 2018. 2. Causey MW, Miller S, Hoffer Z, Hempel J, Stallings JD, Jin G, Alam H, Martin M. Beneficial effects of histone deacetylase inhibition with severe hemorrhage and ischemia-reperfusion injury. J Surg Res 184 (1):533–540, 2013. 3. Zhang M, Shan H, Wang T, Liu W, Wang Y, Wang L, Zhang L, Chang P, Dong W, Chen X, et al. Dynamic change of hydrogen sulfide after traumatic brain injury and its effect in mice. Neurochem Res 38 (4):714–725, 2013. 4. Nolan JP, Neumar RW, Adrie C, Aibiki M, Berg RA, Bottiger BW, Callaway C, Clark RS, Geocadin RG, Jauch EC, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication. A Scientific Statement from the International Liaison Committee on Resuscitation; the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clin

留言 (0)

沒有登入
gif