In vivo measurement of lung capillary-alveolar macromolecule permeability by saturation bronchoalveolar lavage

Cardiopulmonary responses during the cooling and the extracorporeal life support rewarming phases in a porcine model of accidental deep hypothermic cardiac arrest

Background

This study aimed to assess cardiac and pulmonary pathophysiological responses during cooling and extracorporeal life support (ECLS) rewarming in a porcine model of deep hypothermic cardiac arrest (DHCA). In addition, we evaluated whether providing a lower flow rate of ECLS during the rewarming phase might attenuate cardiopulmonary injuries.

Methods

Twenty pigs were cannulated for ECLS, cooled until DHCA occurred and subjected to 30 min of cardiac arrest. In order to assess the physiological impact of ECLS on cardiac output we measured flow in the pulmonary artery using Doppler echocardiography as well as a modified thermodilution technique using the Swan-Ganz catheter (injection site in the right ventricle). The animals were randomized into two groups during rewarming: a group with a low blood flow rate of 1.5 L/min (LF group) and a group with a normal flow rate of 3.0 L/min (NF group). The ECLS temperature was adjusted to 5 °C above the central core. Cardiac output, hemodynamics and pulmonary function parameters were evaluated.

Results

During the cooling phase, cardiac output, heart rhythm and blood pressure decreased continuously. Pulmonary artery pressure tended to increase at 32 °C compared to the initial value (20.2 ± 1.7 mmHg vs. 29.1 ± 5.6 mmHg, p = 0.09). During rewarming, arterial blood pressure was higher in the NF than in the LF group at 20° and 25 °C (p = 0.003 and 0.05, respectively). After rewarming to 35 °C, cardiac output was 3.9 ± 0.5 L/min in the NF group vs. 2.7 ± 0.5 L/min in LF group (p = 0.06). At the end of rewarming under ECLS cardiac output was inversely proportional to the ECLS flow rate. Moreover, the ECLS flow rate did not significantly change pulmonary vascular resistance.

Discussion

Using a newly developed experimental model of DHCA treated by ECLS, we assessed the cardiac and pulmonary pathophysiological response during the cooling phase and the ECLS rewarming phase. Despite lower metabolic need during hypothermia, a low ECLS blood flow rate during rewarming did not improved cardiopulmonary injuries after rewarming.

Conclusion

A low ECLS flow rate during the rewarming phase did not attenuate pulmonary lesions, increased blood lactate level and tended to decrease cardiac output after rewarming. A normal ECLS flow rate did not increase pulmonary vascular resistance compared to a low flow rate. This experimental model on pigs contributes a number of pathophysiological findings relevant to the rewarming strategy for patients who have undergone accidental DHCA.

Deep Hypothermic Cardiac Arrest Treated by Extracorporeal Life Support in a Porcine Model: Does the Rewarming Method Matter?

OBJECTIVES

Extracorporeal life support (ECLS) is the reference rewarming technique of accidental deep hypothermic cardiac arrest (DHCA). This study was designed to examine the impact of different rewarming blood flow rates and temperature setting of ECLS on cardiopulmonary lesions after DHCA in a porcine model of accidental hypothermia.

METHODS

Twenty-four pigs were cannulated for ECLS, cooled until DHCA occurred, and subjected to 30 minutes of cardiac arrest. During the rewarming phase, we compared a low blood flow rate of 1.5 L/min versus a high flow rate of 3.0 L/min as well as two-temperature-setting rewarming strategies: a temperature during ECLS adjusted to 5°C above the central core temperature versus 38°C maintained throughout the rewarming phase. Cardiac output, hemodynamics and pulmonary function parameters were evaluated. Biologic markers of ischemia-reperfusion injuries were analyzed at baseline and at the end of the experiment.

RESULTS

DHCA occurred at 21.2 ± 2°C. There was a trend for better cardiac output in groups with high blood flow (p = 0.053), with no interaction between ECLS flow and temperature (p = 0.63), a trend toward lower pulmonary vascular resistance (PVR; p = 0.075) and a significant decrease in arterial PVR in groups with high blood flow (p = 0.013) with no interaction (p = 0.47 and p = 0.60 for PVR and arterial PVR, respectively). Serum interleukin-6, tumor necrosis factor-α, receptor for advanced glycation end products (RAGE), and neuron-specific enolase were significantly increased between baseline and endpoint. The increase in the serum RAGE concentration was higher in the 38°C rewarming temperature groups compared to 5°C above adjusted temperature. There were no other significant differences in biomarkers.

CONCLUSIONS

We developed a porcine model of DHCA treated by ECLS. Our data suggest that cardiac output tended to improve with a high-flow-rate rewarming strategy while a high-temperature delta between core temperature and ECLS increased the RAGE markers of lung injury.

Increased cardiac index due to terbutaline treatment aggravates capillary-alveolar macromolecular leakage in oleic acid lung injury in dogs

Introduction

We assessed the in vivo effects of terbutaline, a beta2-agonist assumed to reduce microvascular permeability in acute lung injury.

Methods

We used a recently developed broncho-alveolar lavage (BAL) technique to repeatedly measure (every 15 min. for 4 hours) the time-course of capillary-alveolar leakage of a macromolecule (fluorescein-labeled dextran) in 19 oleic acid (OA) lung injured dogs. BAL was performed in a closed lung sampling site, using a bronchoscope fitted with an inflatable cuff. Fluorescein-labeled Dextran (FITC-D70) was continuously infused and its concentration measured in plasma and BAL fluid. A two-compartment model (blood and alveoli) was used to calculate KAB, the transport rate coefficient of FITC-D70 from blood to alveoli. KAB was estimated every 15 minutes over 4 hours. Terbutaline intra-venous perfusion was started 90 min. after the onset of the injury and then continuously infused until the end of the experiment.

Results

In the non-treated injured group, the capillary-alveolar leakage of FITC-D70 reached a peak within 30 minutes after the OA injury. Thereafter the FITC-D70 leakage decreased gradually until the end of the experiment. Terbutaline infusion, started 90 min after injury, interrupted the recovery with an aggravation in FITC-D70 leakage.

Conclusions

As cardiac index increased with terbutaline infusion, we speculate that terbutaline recruits leaky capillaries and increases FITC-D70 leakage after OA injury. These findings suggest that therapies inducing an increase in cardiac output and a decrease in pulmonary vascular resistances have the potential to heighten the early increase in protein transport from plasma to alveoli within the acutely injured lung.

The effect of lysozyme on elastase-mediated injury

Previous studies by this laboratory demonstrated that lysozyme is increased in human pulmonary emphysema, and that it preferentially binds to elastic fibers, which undergo degradation in this disease. In the current investigation, the relationship between lysozyme and elastic fiber injury was further examined, both in vitro and in vivo. The effect of exogenously administered egg-white lysozyme on pancreatic elastase-induced injury was determined using a biosynthetically radiolabeled extracellular matrix preparation mainly composed of elastic fibers. Although matrix treated with lysozyme showed attachment of the protein to elastic fibers, there was no significant increase in elastolysis compared with untreated controls following exposure to either 1 microg/ml or 100 ng/ml of pancreatic elastase. However, lysozyme did impair the ability of hyaluronan (HA) to prevent elastase injury to elastic fibers. Matrix samples sequentially treated with lysozyme and HA, then incubated with 1 microg/ml or 100 ng/ml of pancreatic elastase, showed significantly increased elastolysis compared with those treated with HA alone. Since HA is closely associated with elastic fibers in vivo, the ability of lysozyme to enhance elastolysis was further tested in an animal model of emphysema induced by intratracheal administration of porcine pancreatic elastase. Animals exposed to aerosolized lysozyme prior to elastase administration showed significantly increased airspace enlargement. The mean linear intercept of the lysozyme-treated animals was 123 microm compared with 75 microm for controls receiving aerosolized water (P < 0.0001). These findings suggest that lysozyme may not be an innocuous component of the inflammatory response associated with pulmonary emphysema, but may actually play a role in the pathogenesis of the disease.