Stefan Kreyer

Effects of spontaneous breathing during airway pressure release ventilation on intestinal blood flow in experimental lung injury.

Background In critical illness, the gut is susceptible to hypoperfusion and hypoxia. Positive-pressure ventilation can affect systemic hemodynamics and regional blood flow distribution, with potentially deleterious effects on the intestinal circulation. The authors hypothesized that spontaneous breathing (SB) with airway pressure release ventilation (APRV) provides better systemic and intestinal blood flow than APRV without SB. Methods Twelve pigs with oleic acid–induced lung injury received APRV with and without SB. When SB was abolished, either the tidal volume or the ventilator rate was increased to maintain pH and arterial carbon dioxide tension constant as compared to APRV with SB. Systemic hemodynamics were determined by double indicator dilution. Blood flow to the intestinal mucosa–submucosa and muscularis–serosa was measured using colored microspheres. Results Systemic blood flow increased during APRV with SB. During APRV with SB, mucosal–submucosal blood flow (ml · g−1 · min−1) was 0.39 ± 0.21 in the stomach, 0.76 ± 0.35 in the duodenum, 0.71 ± 0.35 in the jejunum, 0.71 ± 0.59 in the ileum, and 0.63 ± 0.21 in the colon. During APRV without SB and high tidal volumes, it decreased to 0.19 ± 0.03 in the stomach, 0.42 ± 0.21 in the duodenum, 0.37 ± 0.10 in the jejunum, 0.3 ± 0.14 in the ileum, and 0.41 ± 0.14 in the colon (P < 0.001, respectively). During APRV without SB and low tidal volumes, the respective mucosal–submucosal blood flows decreased to 0.24 ± 0.10 (P < 0.01), 0.54 ± 0.21 (P < 0.05), 0.48 ± 0.17 (P < 0.01), 0.43 ± 0.21 (P < 0.01), and 0.50 ± 0.17 (P < 0.001) as compared to APRV with SB. Muscularis–serosal perfusion decreased during full ventilatory support with high tidal volumes in comparison with APRV with SB. Conclusion Maintaining SB during APRV was associated with better systemic and intestinal blood flows. Improvements were more pronounced in the mucosal–submucosal layer.

Effects of spontaneous breathing during airway pressure release ventilation on cerebral and spinal cord perfusion in experimental acute lung injury.

Background Systemic-blood flow, cerebral-blood flow, and spinal cord blood flow can be affected by mechanical ventilation. We investigated the effect of spontaneous breathing on cerebral and spinal blood flow during airway pressure release ventilation (APRV) with and without spontaneous breathing. Methods Twelve pigs with oleic-acid-induced lung injury were ventilated with APRV with or without spontaneous breathing in random order. Without spontaneous breathing, either the upper airway pressure limit of mechanical ventilation or the ventilator rate was increased to maintain pH and PaCO2 constant. Systemic hemodynamic parameters were determined by the double indicator dilution method, cerebral and spinal cord blood flow was measured with colored microspheres. Statistics: ANOVA+Newmann-Keuls-test. Results As compared with APRV without spontaneous breathing and high tidal volume (VT) spontaneous breathing during APRV showed higher systemic blood flow and perfusion of the basal ganglia, frontal lobe, hippocampus, brain stem, temporal lobe, thalamus (all P<0.001), cerebellum, spinal cord (all P<0.01), and the central cortical region (P<0.05). During APRV without spontaneous breathing and low VT blood flow was lower in the basal ganglia, frontal lobe, hippocampus (all P<0.01), and temporal lobe (P<0.05) whereas perfusion of the thalamus, central cortical region, brain stem, cerebellum, and spinal cord were not different compared with APRV with spontaneous breathing. Conclusions In parallel with higher systemic blood flow regional cerebral and spinal cord blood flow were also higher when spontaneous breathing was maintained during APRV. The higher regional blood flow by maintaining spontaneous breathing was more pronounced when compared with full ventilatory support using high VT.

Evaluation of the Cytosorb™ Hemoadsorptive Column in a Pig Model of Severe Smoke and Burn Injury.

Introduction: Host inflammatory response to any form of tissue injury, including burn, trauma, or shock, has been well documented. After significant burns, cytokines can increase substantially within the first 24 h after injury and may contribute to subsequent organ failure. Hemoadsorption by cytokine-adsorbing columns may attenuate this maladaptive response, thereby improving outcomes. The aim of this study was to investigate the feasibility, technical safety, and efficacy of cytokine and myoglobin removal by early use of a cytokine absorbing column (CytoSorb) in a porcine model of smoke inhalation and burn injury. Methods: Anesthetized female Yorkshire pigs (n = 15) were injured by wood bark smoke inhalation and a 40% total body surface area deep burn and observed for 72 h or death. The animals were randomized to hemoadsorption treatment (n = 9) or a sham group (n = 6) before injury. A 6-h hemoadsorption or sham session was performed on days one, two, and three. Serum cytokines (IL-1b, IL-6, IL-8, IL-10, TNF-alpha) and myoglobin were measured systemically, locally in bronchoalveolar lavage fluid and also in circulating blood before and after the adsorbing column to evaluate single pass clearance by the device. Results: Hemoadsorption caused significant removal of IL-1b, IL-6, IL-10, and myoglobin across the device mainly during the first run, ranging from 22% for IL-6 to 29% for IL-1b and 41% removal rates for myoglobin after 15 min of treatment. Systemic cytokine or myoglobin serum concentrations did not change. Conclusions: In a porcine model of smoke and burn injury, hemoadsorption using the CytoSorb cartridge did not result in significant systemic or pulmonary reductions in the measured cytokines or myoglobin despite efficient transmembrane reductions. Further investigations are needed to optimize the efficiency of mediator clearance to affect both circulating levels and clinically relevant outcomes.

Modular extracorporeal life support: effects of ultrafiltrate recirculation on the performance of an extracorporeal carbon dioxide removal device

The combination of extracorporeal carbon dioxide removal (ECCO2R) and hemofiltration is a possible therapeutic strategy for patients needing both lung and renal support. We tested the effects of the recirculation of ultrafiltrate on membrane lung (ML) CO2 removal (VCO2ML). Three conscious, spontaneously breathing sheep were connected to a commercially produced ECCO2R device (Hemolung; Alung Technologies, Pittsburgh, PA) with a blood flow of 250 ml/min and a gas flow of 10 L/min. A hemofilter (NxStage, NxStage Medical, Lawrence, MA) was interposed in series after the ML. Ultrafiltrate flow was generated and recirculated before the ML. We tested four ultrafiltrate flows (0, 50, 100, and 150 ml/min) for 25 min each, eight times per animal, resulting in 24 randomized test repetitions. We recorded VCO2ML, hemodynamics and ventilatory variables, and natural lung CO2 transfer (VCO2NL) and collected arterial and circuitry blood samples. VCO2ML was unchanged by application of ultrafiltrate recirculation (40.5 ± 4.0, 39.7 ± 4.2, 39.8 ± 4.2, and 39.2 ± 4.1 ml/min, respectively, at ultrafiltrate flow of 0, 50, 100, and 150 ml/min). Minute ventilation, respiratory rate, VCO2NL, and arterial blood analyses were not affected by ultrafiltrate recirculation. In the tested configuration, ultrafiltrate recirculation did not affect VCO2ML. This modular technology may provide a suitable platform for coupling CO2 removal with various forms of blood purification.

Enhanced Extracorporeal CO2 Removal by Regional Blood Acidification: Effect of Infusion of Three Metabolizable Acids.

Acidification of blood entering a membrane lung (ML) with lactic acid enhances CO2 removal (VCO2ML). We compared the effects of infusion of acetic, citric, and lactic acids on VCO2ML. Three sheep were connected to a custom-made circuit, consisting of a Hemolung device (Alung Technologies, Pittsburgh, PA), a hemofilter (NxStage, NxStage Medical, Lawrence, MA), and a peristaltic pump recirculating ultrafiltrate before the ML. Blood flow was set at 250 ml/min, gas flow (GF) at 10 L/min, and recirculating ultrafiltrate flow at 100 ml/min. Acetic (4.4 M), citric (0.4 M), or lactic (4.4 M) acids were infused in the ultrafiltrate at 1.5 mEq/min, for 2 hours each, in randomized fashion. VCO2ML was measured by the Hemolung built-in capnometer. Circuit and arterial blood gas samples were collected at baseline and during acid infusion. Hemodynamics and ventilation were monitored. Acetic, citric, or lactic acids similarly enhanced VCO2ML (+35%), from 37.4 ± 3.6 to 50.6 ± 7.4, 49.8 ± 5.6, and 52.0 ± 8.2 ml/min, respectively. Acids similarly decreased pH, increased pCO2, and reduced HCO3− of the post-acid extracorporeal blood sample. No significant effects on arterial gas values, ventilation, or hemodynamics were observed. In conclusion, it is possible to increase VCO2ML by more than one-third using any one of the three metabolizable acids.

[Lung protective ventilation - protective effect of adequate supported spontaneous breathing].

Based on available data, it can be suggested that spontaneous breathing during ventilator support has not to be suppressed even in patients with severe pulmonary dysfunction if no contraindications are present. Experimental data do not support the contention that spontaneous breathing aggravates ventilator-induced lung injury. During spontaneous breathing increase in PTP is maximal in the depended lung areas in adjunct to the diaphragm and causes recruitment of initially atelectatic lung areas thereby avoiding cyclic alveolar collapse and reopening. This should result in a lung protective effect of adequate supported spontaneous breathing. Clinical data supported this belief demonstrating improvement in pulmonary gas exchange, systemic blood flow, and oxygen supply to the tissue and a decrease in days on ventilator support and duration of stay in the intensive care unit.

Early Utilization of Extracorporeal CO2 Removal for Treatment of Acute Respiratory Distress Syndrome Due to Smoke Inhalation and Burns in Sheep.

Introduction: In thermally injured patients, inhalation injury is often associated with acute respiratory distress syndrome (ARDS), and is an independent predictor of increased morbidity and mortality. Extracorporeal CO2 removal (ECCO2R) therapy offers new possibilities in protective mechanical ventilation in ARDS patients. We performed an early application of ECCO2R in mild-to-moderate ARDS in sheep ventilated in BiPAP mode. Our aim was to investigate its effect on severity of the lung injury. Material and Methods: Non-pregnant farm-bred ewes (n = 15) were anesthetized and injured by a combination of wood-bark smoke inhalation and a 40% total body surface area full-thickness burn, and were observed for 72 h or death. The animals were randomized to a Hemolung group (n = 7) or a Control group (n = 8) at time of ARDS onset. ECCO2R was performed in the Hemolung group after onset of ARDS. Histopathology, CT scans, systemic and pulmonary variables, and CO2 removal were examined. Results: Early application of ECCO2R therapy with Hemolung in spontaneously breathing sheep decreased PaCO2 significantly, while the device removed about 70 mL of CO2 per minute. This did not result in lower minute ventilation in the Hemolung group. Lungpathology and CT scans did not show a difference between groups. Conclusion: In an ovine model of ARDS due to smoke inhalation and burn injury, early institution of ECCO2R in spontaneously breathing animals was effective in removing CO2 and in reducing PaCO2. However, it had no effect on reducing the severity of lung injury or mortality. Further studies are necessary to detail the interaction between extracorporeal CO2 removal and pulmonary physiology.

Extracorporeal blood purification in burns: A review

A prolonged and fulminant inflammatory state, with high levels of pro- and anti-inflammatory mediators, is seen after extensive thermal injury. Blood purification techniques including plasma exchange, continuous venovenous hemofiltration, and adsorbing membranes have the potential to modulate this response, thereby improving outcomes. This article describes the scientific rationale behind blood purification in burns and offers a review of literature regarding its potential application in this patient cohort.

The effect of pumpless extracorporeal CO2 removal on regional perfusion of the brain in experimental acute lung injury.

Background: Lung-protective mechanical ventilation with low tidal volumes (VT) is often associated with hypercapnia (HC), which may be unacceptable in patients with brain injury. CO2 removal using a percutaneous extracorporeal lung assist (pECLA) enables normocapnia despite low VT, but its effects on regional cerebral blood flow (rCBF) remain ambiguous. We hypothesized that reversal of HC by pECLA impairs rCBF in a porcine lung injury model. Methods: Lung injury was induced in 9 anesthetized pigs by hydrochloric acid aspiration. rCBF and systemic hemodynamics were measured by colored microsphere technique and transpulmonary-thermodilution during a randomized sequence of 4 experimental situations: pECLA shunt-on (1) with HC and (2) without HC, pECLA shunt-off (3) with HC and (4) without HC. Results: HC increased rCBF (P<0.05). CO2 removal with pECLA resulting in normocapnia, decreased rCBF to levels comparable to those without pECLA and normocapnia. HC resulted in increased cardiac output (+25.5%). Cardiac output was highest during HC with pECLA shunt (+44.9%). During pECLA with CO2 removal, cardiac output (+38.1%) decreased compared with pECLA without CO2 removal, but stayed higher than during normocapnia/no pECLA shunt (P<0.05). Conclusions: In this animal model, mechanical ventilation with low VT was associated with HC and increased rCBF. CO2 removal by pECLA restored normocapnia, reduced rCBF to levels of normocapnia, but required a higher systemic blood flow for the perfusion of the pECLA device. If these results could be transferred to patients, extracorporeal CO2 removal might be an option for treatment of combined lung and brain injury in condition of a sufficient cardiac flow reserve.

Weight loss of respiratory muscles during mechanical ventilation.

failure. We would like to comment on the documented weight loss of the respiratory muscles. The resting period of 48 h on mechanical ventilation seems relatively short to produce a mass reduction of as much as 20–30% of muscle tissue itself. The question arises, therefore, whether other components of the respiratory muscle samples than muscle tissue itself could have contributed to the weight loss during mechanical ventilaton. If we look at the methodology being used to determine muscle weights, we find that the authors removed the samples before the animals were killed by exsanguination. Therefore, besides muscle tissue, the pieces also contained blood components. Since metabolic demands of respiratory muscles decrease to some extent during rest [2], altered blood flow to the respiratory muscles may have contributed to the different weights of samples in mechanically ventilated and spontaneously breathing animals. Results from our own laboratory support this theory. In our experimental model using pigs with oleic-acid-induced lung injury, we measured diaphragmatic blood flow with colored microspheres. Twelve pigs breathed spontaneously at ambient airway pressure (FIO2 0.35, respiratory rate 47±3/min) (mean±SE) and then were paralyzed and mechanically ventilated (FIO2 0.35, respiratory rate 53±3/min, VT 6.5±0.4 ml/kg, PEEP 5±0 cmH2O, peak inspiratory pressure 15±1 cmH2O). Both spontaneous breathing and mechanical ventilation resulted in hypercapnia (PaCO2 59±4 mmHg and 58±3 mmHg, respectively) at unchanged arterial pH (7.32±0.02 and 7.35±0.03, respectively). Mechanical ventilation significantly improved oxygenation (PaO2/FIO2 269±24 mmHg) as compared to spontaneous breathing (PaO2/FIO2 174±12 mmHg) (P<0.001, students t-test) in these severely injured lungs. Diaphragmatic muscle perfusion was approximately 80% lower during mechanical ventilation (0.10±0.03 ml/g wet tissue weight/min) than during spontaneous breathing (0.50±0.09 ml/g wet tissue weight/min) (P<0.001, students t-test). Thus, the mass reduction during mechanical ventilation found in the study by Capdevila et al. [1] may be at least in part due to the reduced blood flow to the respiratory muscles as compared to the spontaneously breathing rabbits.