Conventional Mechanical Ventilation Is Associated with Bronchoalveolar Lavage-induced Activation of Polymorphonuclear Leukocytes: A Possible Mechanism to Explain the Systemic Consequences of Ventilator-induced Lung Injury in Patients with ARDS

The study was carried out in 26 patients with ARDS recruited in the intensive care units of the University Hospitals of Bari (Italy) and Geneva (Switzerland). Bronchoalveolar lavage (BAL) samples obtained from these patients were the remaining material collected from a randomized trial in which a total of 37 patients were recruited. 15In that trial, a lung-protective strategy was found to decrease BAL fluid and blood cytokines compared with a conventional ventilatory strategy.

The protocol was approved by the Institutional Review Boards and informed consent was obtained. Patients were cared for by attending physicians who were not involved in the protocol, and all decisions regarding patient care management were made at these physician's discretion.

Protocol, inclusion, exclusion, and withdrawal criteria have been previously described. 15Briefly, patients were sedated, paralyzed, and ventilated (Servo 300; Siemens-Elema, Stockholm, Sweden). A volume-pressure (V-P) curve on zero end-expiratory pressure was measured. The V-P curve of the respiratory system of patients with early ARDS has a characteristic sigmoidal shape with lower (LIP) and upper (UIP) inflection points thought to approximate the pressure required to initiate recruitment of collapsed alveoli, and the pressure at which overdistension of some lung units occurs, respectively. 15,17After the V-P curve was measured, PEEP was restored to its previous level, and BAL and plasma samples were collected 20–30 min later (entry). Patients were then randomly assigned either to a control strategy using a conventional ventilatory strategy (V^Tand PEEP targeted to maintain the arterial carbon dioxide tension [Paco²] between 35 and 40 mmHg and PEEP to obtain the greatest improvement in arterial oxygen saturation [Sao²] without worsening hemodynamics) or to a lung-protective strategy (V^Tset to obtain a value of plateau pressure [Pplat] < pressure at the UIP regardless of Paco², and PEEP 2–3 cm H²O above the pressure at LIP). 15All measurements were repeated 36–40 h (36 h) after randomization. 15 

Bronchoalveolar lavage was performed using a telescoping catheter (Ballard, Draper, UT) with two aliquots of 40–50 ml sterile isotonic saline. Lavage with a third aliquot was performed if there was less than 30–40 ml recovered fluid from the first 100 ml. The first aliquot was discarded, and the remaining BAL fluid was rapidly filtered through sterile gauze and then spun at 4°C at 400 g  for 15 min. The supernatant was centrifuged at 80,000 g  for 30 min at 4°C to remove the surfactant-rich fraction and then concentrated 10-fold on a 5,000 molecular weight cutoff filter (Amicon, Beverly, MA) under nitrogen. The concentrated supernatant was then frozen at −70°C for later analysis.

Cytokine concentrations from the two groups of patients were analyzed by ELISA in Geneva and have been previously reported. 15The remaining samples were thawed in Toronto where this study was carried out in blinded fashion. BAL fluid from each individual patient analyzed separately and all experiments were performed in duplicate.

Neutrophils were isolated from heparinized whole blood drawn by venipuncture from normal volunteers. Isolation was performed using 3% dextran sedimentation and discontinuous plasma-Percoll (Amersham Pharmacia Biotech, Inc., Baie d'Urfe, Quebec, Canada) gradients as described previously. 18The separation procedure required 2 h. The neutrophils were resuspended at a cell density of 9 × 10⁶cells/ml in endotoxin-free DMEM (Life Technologies, Burlington, Ontario, Canada) and were used immediately after isolation. Neutrophil viability exceeded 97% as assessed by trypan blue exclusion.

Oxidant production and surface expression of CD18, CD63, and L-selectin were measured using flow cytometry. Neutrophils were incubated in the patient's BAL fluid at a density of 8 × 10⁶cells/ml for 2 h at 37°C. For measurement of oxidant production, cells were then incubated with 5 μm dihydrorhodamine (DhR; Molecular Probes, Inc., Eugene, OR) in HEPES-buffered saline containing Ca²⁺and Mg²⁺. After a 5-min incubation, neutrophils were fixed with 1% paraformaldehyde. In separate experiments, neutrophils were incubated in BAL fluid, fixed with 1% paraformaldehyde, and washed twice. Neutrophils were then labeled with mouse antihuman FITC-CD18 antibody or mouse antihuman L-selectin phycoerythrin-labeled antibodies (Serotec Ltd., Oxford, England) for 30 min and washed twice in PBS. Neutrophils were also labeled with mouse antihuman CD63 antibody (Serotec Ltd.) for 30 min, washed, and resuspended in PBS with FITC-labeled secondary antibody (1:500). After a 30-min incubation, neutrophils were washed and resuspended in PBS. Stained cells were analyzed on a FACScan (Becton Dickinson, Palo Alto, CA) using FL1 detector (488-nm excitation and 530-nm emission wavelengths). Cells were gated on the forward and right-angle light scatters to exclude debris and cell clumps. Typically, 1 × 10⁵cells were analyzed per condition, and the specific fluorescence was determined by subtracting the fluorescence after staining with secondary antibodies in the absence of primary antibody. Values are expressed as relative fluorescence index by dividing the linear fluorescence of the experimental groups by the values obtained from the unstimulated control cells.

Extracellular release of elastase from neutrophils was quantified by measuring the degradation of elastin in cell supernatants using an Elastase Assay Kit (EnzChek; Molecular Probes). Elastin was prepared by reductive alkylation of soluble bovine neck ligament elastin and labeled with BODIPY®FL dye such that the conjugate's fluorescence is quenched. Upon digestion by elastase, the fluorescence is revealed. The resulting increase in fluorescence was monitored with a fluorescence microplate reader (CytoFluor 2300; Millipore, Bedford, MA) at 530 nm.

Plasma concentration of interleukin 6 (IL-6) was measured using an enzyme-linked immunoabsorbent assay (Medgenix, Fleures, Belgium).

To evaluate differences over time within each group, repeated-measures analysis of variance (ANOVA) by Bonferroni method was used. To evaluate differences between the two groups, the Fisher exact test for categorical variables, the t  test with unequal variance for continuous variables, and the Mann-Whitney rank-sum test for ordinal variables were used.

All tests of significance were two-tailed, and P < 0.05 was considered as significant. Data are presented as mean ± SD.

Baseline characteristics and underlying conditions responsible for ARDS in the patients included in the study are presented in table 1. Sepsis, pneumonia, and multiple trauma were the underlying conditions responsible for ARDS.

Values of ventilator settings and of arterial blood gases immediately (2 to 3 h) after randomization are presented in table 2. In the control group, PEEP levels were lower, while tidal volume and Pplat were higher than in the lung-protective-strategy group. Pao²did not differ between groups, although Fio²was higher in the control group. A significant increase in Paco²(permissive hypercapnia) was observed in the lung-protective-strategy group.

The cell pellets obtained from the BAL fluids were analyzed for cell distribution, and there was an increased percentage of polymorphonuclear leukocytes in the control group, but not in the protective group. 15 

Bronchoalveolar lavage fluid obtained from the two groups of patients at entry resulted in a similar degree of stimulation of neutrophil oxidant production as determined by oxidation of dihydrorhodamine. BAL fluid obtained from the control group 36 h after mechanical ventilation resulted in significantly increased oxidant production by neutrophils as compared with the values at entry. In contrast, incubation of neutrophils with BAL fluid obtained from the lung-protective-strategy ventilation group at 36 h did not induce any further change in oxidant production. The value of oxidant production in the lung-protective-strategy group was significantly lower (P < 0.001) than in the control group at 36 h (fig. 1).

Bronchoalveolar lavage fluid obtained from both groups at entry had similar effects on the expression of neutrophil surface markers of CD18, CD63, and L-selectin. BAL fluid obtained from patients ventilated with the control strategy for 36 h resulted in significantly increased expression of CD18 and CD63 (fig. 2), and enhanced shedding of L-selectin compared with the BAL fluid obtained at entry (fig. 3). In contrast, these variables remained unchanged when neutrophils were incubated with BAL fluid obtained from the lung-protective-strategy-ventilated patients at 36 h.

We also examined whether incubation of neutrophils with the BAL fluid would increase extracellular release of elastase, a potent tissue-damaging proteolytic enzyme contained in primary granules and released by activated neutrophils. 19BAL fluid obtained from both groups at entry had similar effects on elastase release by neutrophils. BAL fluid obtained from the control ventilatory strategy at 36 h induced a significant increase in neutrophil elastase release (fig. 4). In contrast, the level of elastase release remained low when neutrophils were incubated with BAL fluid obtained form the lung-protective-strategy patients at 36 h.

The percentage change of neutrophil elastase release stimulated by BAL obtained at 36 h and entry was correlated with the change of plasma concentration (36 h − entry) of IL-6 (fig. 5).

Several studies have reported that neutrophil elastase plays a key role in developing MSOF in patients with head injury 20and at the early stage of acute pancreatitis. 21We examined the relation between the number of failing organs in these ARDS patients and neutrophil elastase release (percent of baseline) stimulated by BAL fluid at 36 h, and a significant correlation was observed (fig. 6).

Acute respiratory distress syndrome is the most severe form of acute lung injury affecting both medical and surgical patients. It is characterized by a protein- and cell-rich pulmonary edema fluid due to increased vascular permeability in the lung. Neutrophils constitute up to 70–80% of the cells in BAL fluid of ARDS patients compared with approximately only 1–3% in normal subjects. 22Neutrophil-derived proteases have been suggested to be potent mediators involved in the increased lung permeability. 23,24The results of the current study demonstrate that BAL fluid obtained from patients ventilated with a conventional ventilatory strategy activates neutrophils and induces release of the proteolytic enzyme elastase to a greater extent than BAL from patients ventilated with a protective ventilatory strategy. Furthermore, the high plasma concentrations of IL-6, a pivotal inflammatory cytokine in the pathogenesis of MODS, 25,26correlates well with increased release of neutrophil elastase. These results suggest that the activation of neutrophils and the subsequent release of proteases may be an important mechanism of ventilator-induced lung injury during conventional strategies that overdistend the lung and/or allow repeated alveolar recruitment and derecruitment. 8,27These observations may at least in part explain the decreased incidence of MODS 6and the improved survival observed with protective ventilatory strategies. 6,17 

In response to an inflammatory stimulus, there is a rapid and often massive transmigration of neutrophils from the blood across the endothelium and alveolar epithelium into the alveolar space. Neutrophils are a pivotal component of the innate immune response and essential in host defense against invading microorganisms. 28However, under certain pathologic conditions, such as ARDS, 29chronic bronchitis, 30asthma, 31and cystic fibrosis, 32these inflammatory cells can cause cellular injury by a number of mechanisms, including excessive production of reactive oxygen and nitrogen species and release of proteolytic enzymes.

In ARDS, the initial acute pulmonary inflammatory response may become systemic and self-propagating even after removal of the initiatory events. This can lead to diffuse endothelial injury, organ hypoperfusion (or malperfusion), and ultimately, MODS. 14The persistence of the inflammatory response in ARDS has been associated with poor prognosis. This concept is supported by clinical studies showing that (1) the concentration of inflammatory cytokines in the BAL and plasma on entry is higher in nonsurvivors than in survivors 2; (2) plasma concentrations of inflammatory mediators remain persistently elevated in nonsurvivors compared with survivors in whom the concentration of the mediators decreased over time 1; and (3) the severity of lung injury is correlated with the extent of neutrophil influx and the accumulation of neutrophil-derived cytotoxic products in the alveolar space. 33By way of contrast, the inflammatory response is remarkably different in an uncomplicated pneumonia in which inflammation proceeds in a more regular fashion and is resolved when inciting causes, such as bacteria, are removed. 34The regulatory mechanisms that limit inflammatory damage and facilitate repair of injured tissues may be lost in ARDS, 35and/or insults subsequent to the predisposing factors causing ARDS may continue to actively provoke the inflammatory response. 13,36Efforts to modulate this inflammatory response have met with mixed success, 37,38and thus, it is important to understand the mechanisms underlying this response.

The potential role of neutrophils as major effector cells in the generation of ventilator-induced lung injury has been clearly demonstrated by several experimental studies. 16,39In animals with intact circulating neutrophils, conventional mechanical ventilation increases alveolar-endothelial barrier permeability, impairs oxygenation, and alters lung morphology. Importantly, this damage is abrogated in animals with neutrophil depletion. 16,39 

We found that the release of elastase was correlated with the degree of SIRS and MODS. Such “ventilator-induced” neutrophil activation and the consequent release of elastase may represent the underlying mechanisms by which injurious ventilatory strategies spread inflammatory mediators from the lung into the systemic circulation, possibly leading to MODS. In general, the mechanisms by which mechanical ventilation induces neutrophil activation may include direct physical stress by mechanical ventilation as well as cytotoxic mediators released in the lung. In addition to optimizing ventilatory strategies, several therapies targeting inflammatory mediators have been evaluated in patients with ARDS. Our study may provide a rationale for future therapies directed at controlling the inflammatory response, perhaps with modulators directed at neutrophil function.