Lung Protective Ventilation With Extracorporeal Life Support In Piglets With Acute Lung Injury

BackgroundAcute lung injury (ALI) and acute respiratory distress syndrome (ARDS) are the most severe forms of acute and persistent hypoxemic respiratory failure (PHRF) in adults and children. ALI and ARDS in children have very high mortality and morbidity in recent domestic multicenter clinical investigations. Pathogenesis and phathophysiology of ALI/ARDS involve variable insults as pulmonary or extra-pulmonary origin, and severe alveolar-to-vascular permeability, leading to bilateral infiltration, edema, intra-pulmonary shunting and ventilation-perfusion mismatching. Clinically it is characterized as refractory hypoxaemia requiring aggressive ventilation and intensive care to survive. Currently, its death rate is more than 50% in Chinese pediatric intensive care unit (PICU). Although various interventions with lung protective ventilation strategy are implemented and tend to be effective, no any single therapy claims cost-effective in pediatric ALI/ARDS yet. Very often, combined or alternative therapies such as lung tidal volume restriction, and alveolar recruitment including high frequency oscillation (HFOV), fluid restriction, prone position, inhaled nitric oxide (iNO), and exogenous surfactant (Surf), are considered appropriate. These treatment modalities depend on effective gas exchange and adequate pulmonary perfusion to improve oxygenation. When there is dysfunction at any level of ventilation and perfusion due to sever injury in the lungs, PHRF and ALI occur. As ALI is early phase of ARDS, it is obvious that early intervention with effective and adequate therapy is vital in bringing up optimal response and outcome prediction in the very sick children.Extracorporeal membrane oxygenation (ECMO) is a unique therapy for life support in those who have impaired respiratory and circulatory function. It improves oxygenation with minimum ventilation requirement, thus enabling lung rest for reparation. Technology of ECMO involves an extracorporeal circuit conducting deoxygenated venous blood flow, after re-oxygenation and warming, to systemic circulation, either through artery or vein depending on whethere there is a heart or lung failure. It removes CO₂ out of, while fresh O₂ is provided to, the circulation, through convection of gas and blood flow in the oxygenator, or as an artificial lung. ECMO treatment also has adverse effects as it provokes a systemic inflammatory response as reflected by neutropenia, activation of polymorphonuclearcytes, release of proinflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IL-8, activation of complement, and a capillary leak syndrome with systemic and pulmonary edema, in addition to hemolysis of damaged red blood cell in the circuit. This may lead to secondary organ injury. Therefore we consider a modulation or down-regulation of the inflammatory process in the lungs during ECMO should alleviate lung injury and improve the prognosis of children with PHRF and ALL In our previous, as well as others, studies, iNO and/or Surf tend to be effective in improving oxygenation and inhibiting pulmonary inflammation in experimental ALI/ARDS induced by intravenous endotoxin, oleic acid, or by intra-tracheal or intra-abdominal bacteria. iNO is capable of selectively dilating intrapulmonary resistant vessels, reducing pulmonary artery hypertension, improving hypoxemia by reducing intrapulmonary shunt, and optimizing ventilation-perfusion matching. It also reveals anti-inflammatory capability by inactivating nuclear transcription factor (NF)-κΒand its downstream proinflammatory cytokine synthesis. iNO may also decrease the expression of adhesion molecules, preventing neutrophil adhesion and migration in the injured lungs. Pulmonary surfactant, a mixture of phospholipids and specific proteins produced by the typeâ…¡alveolar epithelial cells, is important in maintaining alveolar expansion during breath. The rationale for the use of exogenous surfactant in the treatment of patients with ALI/ARDS is not only to recover the function of surfactant, but also to inhibit the stimulated production of superoxide anions, to suppress the stimulated secretion and synthesis of proinflammatory cytokines, such as TNF-α, IL-6, IL-8, and also to inhibit granulocyte activation. Previous studies from this lab have demonstrated that a combined administration of iNO and Surf has better therapeutic effects than either therapy alone. It is of interest to know whether this combined use may exert similar effects of anti-inflammation in ECMO, with special emphasis on endotoxin-induced ALI, and any benefit in facilitating lung repair during recovery from ECMO. Objectives1. To observe ECMO effects on lung mechanics, gas exchange, and hemodynamics and the inflammation induced by ECMO in ALI piglets between 0 and 24 h, and to detect the lung pathology and biochemical injury in piglets survived at 168 h.2. To investigate the effects of a combined use of iNO and surfactant on mitigation of the lung inflammatory injury induced by ECMO in healthy and ALI piglets, and to observe the effects on lung reparation in the survived animals.MethodsAfter sedated intramuscularly with ketamine, twenty-three piglets, male, 4-5 week-old, body weight 9-14 kg, received i.v. infusion of LPS (18-20μg/kg) within one h, followed by mechanical ventilation with a standard tidal volume of 7-9 ml/kg for 4-8 h. ALI was defined as PaO₂/FiO₂≤300 mmHg, dynamic lung compliance (Cdyn) decreased by more than 30% of its baseline level. This moment was regarded as treatment time 0 h. At 0, 12, and 24 h three animals were immediately sacrificed. Their lungs showed that LPS induced diffuse alveolar damage represented by alveolar atelectasis, and leukocyte sequestration. The other 20 ALI animals were randomly allocated to four groups (n=5) and defined as: VENT group, animals treated with PCV ventilation; VENOS group, animals treated with PCV ventilation, inhalation of 10 ppm NO, 50 mg of surfactant phospholipids/kg body weight via the endotracheal tube; ECMO group, animals treated with PCV ventilation, ECMO; ENOS group, animals treated with PCV ventilation, ECMO, iNO and surfactant as VENOS group. Additionally, five healthy piglets were used as a normal control group (NENOS group). Animals in NENOS group were treated with PCV ventilation, ECMO, iNO and surfactant. During the experiment, Ringer's lactate solution was i.v. infused to keep normal blood pressure, and 1.4% bicarbonate sodium in Ringer's solution was given to overcome metabolic acidosis. FiO₂ was adjusted to maintain PaO₂ greater than 60 mm Hg, and PaCO₂ was maintained between 35 and 45 mm Hg by varying the respiratory rate (RR) in the non-ECMO-treated animals. Continuous infusion of heparin maintained the activated clotting time (ACT) at 180 to 220 s in ECMO groups. The ECMO flow was kept 70-80 ml/kg/min. The initial sweep gas flow of oxygenator was set at 2 L/min and titrated to keep PaCO₂ between 35 to 45 mm Hg. Animals were treated with above settings from 0 to 24 h and fed between 24 h and 168 h. Arterial blood gas, Cdyn, systematic hemodynamics, airway pressure, minute ventilation volume, RR of ventilator, and ECMO parameters were monitored at the baseline, establishment of ALI, and each h during the treatment. Blood samples were collected at baseline, establishment of ALI, 4, 8, 16, 24, 48, and 168 h of the treatment. At 168 h, animals were sacrificed by overdose of 10% potassium chloride and lung tissues and bronchoalveolar lavage fluid (BALF) were collected. Total proteins (TP), total phospholipids (TPL), disaturated phosphatidylcholine (DSPC) were measured with biochemical methods and minimum and maximum surface tension (γ_min andγ_max) of TPL in BALF were measured using pulsating bubble technique. Commercial available kits were used to measure the levels of NOx^-, IL-8, IL-6, myeloperoxidase (MPO) malondialdehyde (MDA), glutathione (GSH), and total nitric oxide synthase activity (tNOS). The expression of IL-8, IL-6, keratinocyte growth factor (KGF), hepatocyte growth factor (HGF), vascular endothelial grow factor (VEGF), vascular endothelial grow factor receptor 2 (VEGFR-2), collagenâ…¢, inducible nitric oxide synthase (iNOS), and endothelial nitric oxide synthase (eNOS) mRNA in lung tissues were measured by real-time polymerase chain reaction (real-time PCR) at the end of the experiment.Results1. ALI model: Piglets appeared transient purple plague, poor response, the increase of blood temperature and heart rate, tachypnea, and the decrease of systemic blood pressure, Cdyn and PaO₂ and accompanied with the decrease of peripheral WBC 0.5-1 hour since LPS infused. It took 4-8 h to result in ALI. The lung pathology showed that LPS induced diffuse alveolar damage represented by alveolar bleeding, atelectasis, leukocyte sequestration.2. Hemodynamics during treatment: During the treatment, the systemic hemodynamics of ECMO and ENOS groups was stable. Levels of MABP, dp/dtmax, and SVR in ECMO group were 19%, 25%, and 36% higher than those in VENT group respectively.3. Lung mechanics during treatment: There was a reducing trend of RR in the ECMO and ENOS groups. At 8 and 16 h, RR of ECMO group was significantly lower than that of VENT and VENOS groups (p<0.01). At 24 h, RR of VENOS was obviously higher in comparison to that of ECMO group (p<0.05). At 8 and 16 h, minute ventilation volume (MV) of ECMO group was significantly lower than that of VENT and VENOS groups (p < 0.01). Minute ventilation volume of 8 h in ENOS group was significantly lower compared to that in VENT group (p < 0.05).4. Gas exchange during treatment: There was a continuous and significant improvement in PaO₂/FiO₂ in ECMO and ENOS groups during ECMO treatment and it was improved in both VENT and VENOS groups as well.5. Phospholipids, surface tension, white cell counts (WCC), and TP in BALF: ENOS and NENOS groups had higher DSPC than that of VENT and VENOS groups (p < 0.05) and had lower minimum surface tension of TPL in BALF than that of VENT group (p < 0.05). ECMO group had higher DSPC than that of VENT (p < 0.05). WCC in ENOS and NENOS groups was lower than that in VENT group (p < 0.05). There was not significant difference in TP levels among the groups.6. NOx^- concentration: At 0 h, plasma NOx^- of VENT and ECMO groups was significantly higher than that of NENOS group (p < 0.05). At 4 h, VENT group had significantly higher NOx^- than ENOS and NENOS groups (p < 0.05). At 168 h, BALF NOx^- of VENT group was higher than that of VENOS, ECMO, and NENOS groups (p< 0.01).7. IL-6 concentration: During the treatment, plasma IL-6 in all groups increased and reached its peak level at 24 h. At 48 and 168 h, this level was decreased in all groups.8. IL-8 concentration: During the treatment, plasma IL-8 in all groups increased and reached its peak level at 24 h. At 16 h, plasma IL-8 of VENT group was significantly higher than that in ENOS and NENOS (p < 0.05). At 24 h, VENT, VENOS, and ECMO had higher plasma IL-8 than that in ENOS and NENOS (p < 0.05). There was not significant difference in BALF IL-8 among the groups.9. The mRNA expression of IL-6, IL-8, iNOS, and eNOS in the lung tissue: There were no significant differences in mRNA expression of IL-6 and eNOS among the groups. IL-8 expression in VENT was significantly higher than that in VENOS, ENOS, and NENOS (p < 0.01). ECMO had higher IL-8 expression than that in ENOS and NENOS (p < 0.05). VENT had significantly higher iNOS expression than that in ECMO, ENOS, and NENOS (p < 0.01).10. The mRNA expression of VEGF, VEGFR-2, KGF, HGF, and collagenâ…¢in lung tissues: There were no significant differences in mRNA expression of VEGF, KGF, and collagenâ…¢among the groups. Expression of VEGFR-2 in NENOS was significantly higher than that in VENT (p < 0.05). ENOS and NENOS had higher expression of HGF than in VENT (p < 0.05).11. MPO, MDA, GSH, tNOS, NOx^-, and W/D in lung tissue: There were not significant differences in the levels of MPO, GSH, tNOS, and NOx^- in lung tissue among the groups. MDA in ENOS and NENOS was significantly lower than that in VENT (p < 0.05), and VENT group had higher W/D than that in NENOS group (p < 0.05).12. Lung histopathology: There was prominent neutrophil infiltration in VENT. VENOS and ECMO had moderate pathological changes. There was modest neutrophil infiltration in ENOS and NENOS. Volume density of alveolar aeration in ENOS and NENOS was significantly higher than in VENT (p < 0.05).Conclusions1. ALI was successful established by intravenous administration of LPS in young piglets, which enabled assessment of therapeutic efficacy and safety of ECMO and investigation of mechanicsm of lung injury and repair in the recovery.2. ECMO initiated a systemic inflammatory response and caused lung injury in these animals.3. ECMO improved lung mechanics, oxygenation, and hemodynamic condition of ALI piglets in acute phase and alleviated the lung inflammatory response induced by ventilator treatment associated with differential expression of iNOS and endogenous NO metabolites in the lungs.4. Combined use of iNO and surfactant mitigated the inflammatory response provoked by ECMO as reflected by plasma IL-8 production in the acute phase and the lung expression of IL-8 in the recovery phase as well as altered MDA production. This modality upgraded the expression of HGF and facilitated reparation of alveolar epithelial cells in the recovery phase.