Main

Cardiac arrest is a model of whole body ischemia reperfusion injury (1–3). In the pediatric population, asphyxial cardiac arrest is one of the commonest causes of cardiac arrest, and is becoming a more frequently recognized cause in adults (4–6). Acute post arrest reperfusion injury is characterized by similar functional and biochemical changes, which occur after reperfusion injury. Post resuscitation myocardial dysfunction, inflammatory cascade activation, tissue injury, and reactive oxygen species activation are hallmarks of reperfusion injury (7–12). Postconditioning (Post C) performed immediately upon reperfusion, by pharmacological or mechanical methods ameliorates reperfusion injury (13–15). Postconditioning is accomplished by brief interruptions in reperfusion of the target organ, (ischemia Post C) or by ischemia of remote organs (remote Post C). Ischemia or remote postconditioning can only be performed under controlled circumstances owing to its initiation immediately upon reperfusion. Clearly, it would be desirable if a method were available which did not have such rigorous time constraints Whole body periodic acceleration (pGz) might serve such a purpose. The application of pGz, sinusoidal head to foot motion produces pulsatile shear stress to the vascular endothelium. This causes release of endothelial derived vasoactive and cardioprotective substances, such as endothelial derived nitric oxide (eNO), prostaglandins, tissue plasminogen activator (t-PA), and adrenomedullin into the circulation (3,16–19). These substances have been found to be cardioprotective in several ischemia reperfusion models (20–33).

The purpose of this study was to demonstrate the acute benefits of pGz as a means of delayed post C in CPR after asphyxia induced shock, e.g. institution of pGz 15 min after ROSC. These benefits include improved acute post resuscitation myocardial stunning and regional blood flows as well as decreased post resuscitation inflammatory cytokine surge and neutrophil infiltration.

METHODS

Animal preparation.

The Institutional Animal Care and Use Committee of Mount Sinai Medical Center of Greater Miami approved all animal studies in compliance with the Animal Welfare Act. Twenty male Yorkshire juvenile pigs weighing 4 ± 0.7 kg (age 2–3 mos) were used in this study. The animals were initially anesthetized with ketamine (10 mg/kg, intramuscularly) and maintained in a surgical plane of anesthesia with intravenous propofol. An airway was established by direct laryngoscopy and intubation carried out with a 5.0 mm cuffed endotracheal tube. The animals were paralyzed with pancuronium bromide (0.1 mg/kg) and ventilated with a volume ventilator (Puritan Bennett, Pleasanton, CA) at 10 mL/kg tidal volume to achieve normal arterial carbon dioxide tension. FiO2 was maintained at 0.21. Intravascular catheters were placed into the femoral artery to measure systemic blood pressure by connecting the fluid filled catheter to a pressure transducer (Transpac, Abbott Critical Care Systems, North Chicago, IL). A right atrial catheter was placed via the left external jugular vein for administration of fluids and drugs and the measurement of right atrial pressure. Arterial blood gases and electrolytes were measured using a blood gas analyzer (Rapid Lab TM348, Bayer Diagnostics, Tarrytown, NY). A left ventricular catheter was placed via the right carotid artery, and location confirmed by typical left ventricular waveforms, echocardiography and post mortem, for microsphere injections. An 5.5 F balloon-tipped flow-directed thermodilution pulmonary arterial catheter (OpticomTM, Abbot Laboratories, Chicago, IL) was inserted via the right jugular vein and floated into the pulmonary artery under direct pressure monitoring for measurements of pulmonary arterial pressure, right atrial pressure, core body temperature and cardiac output. Placement was confirmed by the typical wave form of the pulmonary artery, and further confirmed at necropsy. The cardiac output was determined by thermodilution in triplicate using ice-cold saline. Body temperature was maintained between 37 and 39°C by means of heating pad. A standard 3-lead electrocardiogram (ECG) configuration was continuously monitored.

Platform design.

The motion platform, which imparts whole body periodic acceleration, has been previously described (34,35). It consists of a wooden platform powered by a linear displacement motor attached to an amplifier that is controlled by a sine wave generator (Non Invasive Monitoring Systems, Inc., North Bay Village, FL). The platform moves sinusoidally in a head-to-foot direction. The following link depicts a video of the pGz platform and motion: http://www.miamiheartresearch.org/pgzmotion/.

Experimental design.

After surgical preparations, the paralyzed animals were observed for 30 min as baseline (BL). Asphyxia was induced by disconnecting the endotracheal tube from the mechanical ventilator, and occluding the endotracheal tube. Additionally, the animals were paralyzed to avoid gasping. After 2 min of reaching a mean blood pressure of 30 mm Hg or below (Hypoxia Phase), the occlusion of the endotracheal tube was relieved and the animals underwent closed chest compression/ventilation CPR using a pneumatic piston device (Thumper, Michigan Instrument, Grand Rapids, MI). The compression force was adjusted to decrease the anteroposterior diameter of the chest by 25–30% with between 60 and 70 lb/in2; the ventilation was set at 10 mL/kg with a compression ventilation ratio of 30:2. After return of spontaneous circulation (ROSC), e.g. an unassisted systolic blood pressure of >50 mm Hg for at least 15 min, animals were randomly assigned to receive standard care, with conventional mechanical ventilatory support (CMV), to maintain normal blood gases (Control) or CMV and pGz (pGz). Fifteen minutes after ROSC, pGz was applied at 180 cpm with G at ± 0.4 and termed delayed postconditioning. Both groups received FiO2 0.21 during CPR and after ROSC. No further pharmacological intervention was given during CPR and after ROSC (Fig. 1). All animals were continuously monitored for 3 h and received propofol (0.05 mg/kg/min) after ROSC, and then humanely euthanized while still under anesthesia with 100 mg/kg of sodium pentobarbital, consistent with the recommendation of the Panel on Euthanasia of the American Veterinary Medical Association. Tissues were sampled from vital organs for microsphere analysis and snap frozen at −40°C, and stored for future analysis.

Figure 1
figure 1

Schematic of the three phase asphyxial model. Hypoxia phase: asphyxia is induced and after loss of aortic pulsation (LOAP defined as aortic pulse pressure of <2 mm Hg), 3 mins on nonintervention are observed. CPR phase: 3 mins of CPR is performed with a THUMPER device or until return of circulation. Reperfusion phase: all animals remained on conventional mechanical ventilatory support to maintain normal arterial blood gases. In one group (pGz), pGz was begun 15 mins after return of circulation. No pharmacologic intervention is performed during CPR or after ROSC.

Full size image

Echocardiography.

Echocardiography was performed with a Hewlett-Packard SONOS 2000 echocardiographic system with a 3.5/2.7 MHz transducer and recorded on VCR tapes. In each animal, two-dimensional short-axis view and M-mode tracings were performed according to the recommendation of the American Society of Echocardiography and as previously described by our laboratory (8,36,37). All measurements were performed by an investigator blinded to the treatment assignment. Wall motion was assessed by the wall motion score index (WMSI) that was obtained by the sum of wall motion scores divided by the number of visualized segments. In this scoring system, higher scores indicate more severe wall motions abnormalities (1 = normal, 2 = hypo kinesis, 3 = akinesis, 4 = dyskinesis, and 5 = aneurysm) (38,39).

Biochemical analysis.

Biochemical indices of reperfusion injury were obtained. Troponin I was measured as biochemical evidence of myocardial injury. Troponin I levels were measured with a pig cardiac Troponin I enzyme immunoassay kit (Life Diagnostics, Inc., West Cheater, PA). Markers of inflammatory activation, i.e. tumor necrosis alpha (TNF-α), and IL-6 (IL-6) were measured by a porcine specific ELISA kit (R&D Systems, Minneapolis, MN). A marker of tissue oxidative stress, myeloperoxidase activity (MPO) in the heart and lung was measured as previously described (40). MPO activity in each sample was determined by measuring the change in absorbance at 460 nm. One unit of MPO activity is the amount of enzyme that will reduce 1 μM peroxide per minute.

Regional blood flow.

Regional blood flow measurements were performed as previously reported by our laboratory (8,41,42). A bolus of 6 × 106 colored microspheres (Dye-Trak, Triton Technologies, San Diego, CA) suspended in 3 mL of a carrier solution of 0.9% NaCl and 0.01% Tween-80 was injected into the left ventricle. The reference blood sample was collected from the abdominal aorta. Blood flow was calculated as the tissue-derived numbers of microspheres referenced to arterial blood derived numbers of microspheres multiplied by the reference arterial blood withdrawal rate normalized per gram of tissue weight.

Statistical analyses.

All values are reported as means ± SD and where appropriate range and median values are reported. Continuous variables were evaluated by analysis of variance for repeated measures. For variables with significant differences, post hoc analysis was done using Tukey HSD for equal or unequal sample size. Comparison of discrete variables was evaluated by Fisher's exact test. Statistical analysis was performed using STATISTICA (StatSoft Inc., Tulsa OK). Statistical significance was taken as p < 0.05.

RESULTS

Survival and electrocardiography.

Time under anesthesia, and surgical procedure were not different between groups. The median time from endotracheal tube occlusion to reach 2 min of mean blood pressure of 30 mm Hg or less was 10 min (range, 8–12 min) (Hypoxia Time) and was not significantly different between the Control and pGz groups. Figure 1 shows the schematic of the protocol.

All animals had severe bradycardia upon initiation of CPR; none were in ventricular fibrillation. Mean heart rate immediately before start of CPR was 32 ± 10 bpm, and 42 ± 20 bpm for Control and pGz, respectively (p = NS). All animals had ROSC. There were no significant differences in the CPR time between groups, pGz (3 ± 1 min) and Control (3.5 ± 0.5 min).

Arterial blood gases and hemodynamics.

Tables 1 and 2 list the arterial blood gases and hemodynamics in both groups over time. There were no significant differences between the Control and pGz groups in arterial blood gas variables across time. Both groups had a significant increase in arterial PaCO2, a decrease in Pao2 and pH during hypoxia compared with BL but there was no significant difference between groups. There was a rise of serum potassium compared with BL in both groups. In Control and pGz animals during CPR, K+ was 6.3 ± 1.2, 5.9 ± 1.5 meq/dL, respectively (p = 0.07). There were no significant differences between groups for mean arterial blood pressure, coronary perfusion pressure and cardiac index. Pulmonary artery pressure significantly increased from BL in both groups during CPR and 30 min after ROSC. Control animals had an increase from 25 ± 2 mm Hg at BL to 35 ± 3 and 42 ± 2 mm Hg during CPR and 30 min after ROSC. pGz animals also had an increase from 24 ± 3 at BL to 37 ± 5 mm Hg during CPR. Fifteen minutes after ROSC, when pGz was applied for delayed postconditioning, mean pulmonary arterial pressure was 42 ± 2 mm Hg in the Control group and 23 ± 4 in the pGz group (p < 0.01).

Table 1 Arterial blood gases and potassium levels
Full size table
Table 2 Hemodynamic data
Full size table

Echocardiography.

Myocardial stunning was present in all animals in both groups after ROSC. Ejection fraction (EF), fractional shortening (FS) and wall motion score index were severely impaired in Control animals at both 30 and 120 min after ROSC. Delayed postconditioning with pGz animals had significantly less myocardial dysfunction compared with Control group at 30 and 120 mins after ROSC (Fig. 2).

Figure 2
figure 2

Echocardiographic measurements at BL and after 30 mins and 120 min after ROSC. A, Percent EF. B, Percent FS. C, WMSI. †p < 0.05 values compared with BL values. ‡p < 0.05 values Control (□) vs pGz (▪).

Full size image

Biochemical indices of reperfusion.

Three hours after ROSC, the surge in inflammatory markers, e.g. IL-6 (IL-6) and tumor necrosis alpha (TNF-α) was attenuated in the pGz animals compared with the Controls (p < 0.05) (Fig. 3). Oxidative stress and neutrophil infiltration as evidenced by myeloperoxidase activity was also attenuated 3 h after ROSC in both heart and lungs of pGz treated animals (p < 0.01) (Fig. 4). Troponin I, a marker of cardiac tissue injury, was attenuated in pGz treated animals 3 h after ROSC compared with Control (p < 0.01) (Fig. 5).

Figure 3
figure 3

Serum levels of tumor necrosis alpha (TNF-α) (A) IL-6 (IL-6) (B) at BL and after 3 h ROSC. †p < 0.05 values compared with BL values. ‡p < 0.05 values Control (♦) vs pGz (•).

Full size image
Figure 4
figure 4

Myeloperoxidase activity in heart and lung tissue at completion of the experiment. ‡p < 0.01 values Control (□) vs pGz (▪).

Full size image
Figure 5
figure 5

Serum troponin I levels at BL and after 3 h ROSC. †p < 0.01 values compared with BL values. ‡p < 0.01 values Control (□) vs pGz (▪).

Full size image

Regional blood flows.

Thirty minutes after ROSC, regional blood flow in the cerebral cortex and cardiac ventricles was significantly increased in pGz animals compared with Control (p < 0.01). This increase persisted in the ventricles 2 h after ROSC, but not in cerebral cortex where it declines to 44 ± 8% of BL. There were no significant differences between pGz and Control groups in cerebellar or renal regional blood flows (Fig. 6).

Figure 6
figure 6

Regional blood flows in: A, Cerebral cortex, B, Cerebellum. C, Combined right and left ventricles. D, Kidney as percent change from BL values for each organ. ‡p < 0.05 values Control (□) vs pGz (▪).

Full size image

DISCUSSION

Unlike classical postconditioning whereby the postconditioning stimulus is performed immediately at reperfusion, we applied pGz 15 min after ROSC. This intervention, which we term ‘delayed postconditioning' because of the delayed institution of its application, still renders pGz an effective strategy to reduce injury induced by whole body ischemia reperfusion.

Classical postconditioning is defined as a series of brief interruptions in reperfusion applied at the immediate onset of reperfusion (43,44). Na et al. in cats (44a) and Zhao et al. in canines (45) first demonstrated this phenomenon in models of coronary artery occlusion. Na et al. found postconditioning to prevent reperfusion induced venticular fibrillation (44a). Zhao et al. found reduced myocardial infact size and neutrophil accumulations (45). They found reduced myocardial infarct size and neutrophil accumulations. The mechanisms involved in postconditioning have been the subject of numerous investigations. Important signaling mechanisms have been identified during ischemia reperfusion, the p42/44 ERK1/2 and PI3-Akt pathways collectively termed RISK pathway. These are innate cellular survival kinase pathways which when stimulated inhibit proapoptotic pathways, stimulate eNOS and activate KATP channels (46–48). Endothelial derived NO (eNO) inhibits the mPTP (mitochondrial transitional pore) and attenuates necrosis and apoptosis (49,50). Classical ischemic postconditioning activates components of the RISK pathway (PI3-Akt and ERK1/2) (43,49–53).

Endothelial derived NO and prostaglandins are critical during and after whole body ischemia reperfusion injury. Inhibition of eNO or prostaglandins before whole body ischemia reperfusion produces poorer survival and functional outcome (3,54,55). Our laboratory has shown that pGz through increased pulsatile shear stress increases eNO, prostaglandin E2, and adrenomedullin (17–19). These have all been found to be cardioprotective in various models of ischemia reperfusion. Thus, pGz postconditioning after asphyxial shock may exert its beneficial effects through these endothelial derived mediators.

In contrast to classical postconditioning, pGz in asphyxia-induced shock was applied 15 min after ROSC. This time lapse was done to better mimic a usual clinical scenario of asphyxic cardiac arrest, whereby 15 min might elapse after ROSC before the patient being transported to a hospital facility. Even this delayed pGz application, beneficial effects occur as evidenced by attenuated and improved functional as well as biochemical indices of reperfusion injury. These findings with pGz postconditioning contrast to those reported by other investigators of a loss of cardioprotection when the postconditioning stimuli are delayed even a few minutes (15). These differences can possibly be accounted for by taking into consideration that classical postconditioning stimuli when invoked may alter the upstream signaling RISK pathway or others. pGz postconditioning effect involves more downstream signaling, such as eNOS, prostaglandins, and adrenomedullin.

There are limitations to the present work, which need to be acknowledged. Our asphyxia cardiac arrest model was performed on healthy anesthetized swine and therefore we cannot extrapolate our conclusions to diseased states, or infer any potentially confounding effects of anesthesia. Both groups received the same anesthetic and dosing. We also did not determine whether or not the cardioprotection afforded by delayed pGz (15 min after ROSC) can still be obtained if pGz were to be applied beyond this time period, or even enhanced if applied before this time period. The acute period of observation in our study allows us to draw conclusions on the acute cardiovascular and biochemical changes since longer periods of study would be required to determine whether any additional gains could be obtained on neurologic outcomes. The present study demonstrates that pGz applied 15 min after ROSC from asphyxia induced shock (delayed postconditioning) decreases acute post resuscitation myocardial stunning, improves regional blood flow to the heart and transiently to cerebral cortex, decreases indices of tissue damage, reduces inflammatory cytokine surge as well as oxidative stress markers and neutrophil infiltration compared with standard supportive interventions.