Higher Mean Airway Pressure Research Articles

High-frequency mechanical ventilation principles and practices in the era of lung-protective ventilation strategies.

The term high-frequency ventilation is used to describe a heterogeneous group of ventilation modes that are characterized by high respiratory frequencies and low tidal volumes. The increasing understanding of the pathogenesis of VILI, including concepts such as volutrauma and atelectrauma, has led to a renewed interest in the role of HFV in lung-protective ventilation strategies. Inherent to many modes of HFV are low tidal volumes and small pressure swings during the respiratory cycle, which allow for higher mean airway pressures than those safely achieved with CMV. This has the potential to reduce lung injury by limiting volutrauma, whereas maintaining bigger lung volumes at end-expiration may reduce atelectrauma. Of the various forms of HFV, HFO is the only mode with an active expiration phase. This characteristic, combined with superior gas conditioning, may make HFO a promising ventilatory strategy for adults. Although a significant amount of data exists in the literature to support the application of HFO in infants and children who have acute respiratory failure, clinical data on the use of HFO in adults is only now emerging. Early studies of applying HFO in ARDS patients have demonstrated its safety and benefit in terms of oxygenation. Additionally, limited data exist on the comparison between HFO and CMV in this patient population; however, encouraging preliminary results have been reported. The optimum strategy for the application of HFV, including the timing of HFV initiation, remains unclear.

Ventilation à haute fréquence chez le nouveau-né

Neonatal high-frequency ventilation, which has been used for 15 years, allows for CO 2 elimination while using low tidal volume, generated by small pressure variations. The main expected benefit is a decrease in baro / volutrauma, and the possibility of using high mean airway pressures. Elective use of this technique for the treatment of respiratory distress syndrome in premature infants may reduce the rate of occurrence of prolonged respiratory courses toward bronchopulmonary dysplasia. Despite several animal experiments supporting this, randomized clinical trials gave inconsistent results. In the most recent trial which showed benefits of using high-frequency ventilation, those benefits were actually modest. This highlights the fact that bronchopulmonary dysplasia is a prematurity-related multifactorial disease, rather than a simple ventilator-related disease. High-frequency ventilation, however, is among the most useful techniques in neonatal intensive care. This technique indeed allows the treatment of severe cases of respiratory failure, such as those resulting from congenital diaphragmatic hernia, or to restore gas exchanges when conventional ventilation has failed. High-frequency ventilation has contributed to the observed decrease in the number of neonates who require extracorporeal membrane oxygenation.

The use of high-frequency oscillatory ventilation in adults with acute lung injury.

The use of HFOV in adults is still in its infancy. There is, however, much promise to support further study of this ventilatory modality. Rescue case series have shown that HFOV is effective in improving gas exchange and appears safe in this group of extremely ill patients. In addition, as evidence continues to mount regarding the importance of VILI and its mechanisms, HFOV provides a theoretically attractive alternative to conventional lung-protective ventilatory modes. When HFOV is used in adults, it should be in conjunction with an effort to recruit atelectatic lung units by employing higher mean airway pressures and weaning FIO2 before P(AW). HFOV could be used as one of a number of new therapies for the patient failing to oxygenate on CMV. Its routine use to prevent VILI cannot be recommended at this time, as no data are available. Further clinical studies potentially leading to a large randomized controlled trial of HFOV versus best conventional therapy appear worth pursuing.

Lung injury in neonates: Causes, strategies for prevention, and long-term consequences

HMD, demonstrating that ventilator strategy can contribute to the progression of HMD. The protective ventilatory strategy used by Meredith et al16 also included an attempt to maximally recruit the lung by using a high mean airway pressure. Although this approach improved gas exchange, the animals developed clinical signs of cardiac compromise. The problems with cardiac output impairment were avoided if, after the lung was recruited, the mean airway pressure was gradually decreased. As a guide, Kinsella et al17 used the chest radiograph and oxygenation (PaO2/partial pressure of oxygen in the alveoli [PAO2]) to

Cerebral blood flow during partial liquid ventilation in surfactant-deficient lungs under varying ventilation strategies.

OBJECTIVE: To test the hypothesis that cerebral and other regional organ blood flow would be maintained during partial liquid ventilation (PLV) in an animal model of acute lung injury during different ventilation strategies. DESIGN: A prospective, randomized study. SETTING: Animal research facility. SUBJECTS: Sixteen piglets, 2 to 4 wks of age. INTERVENTIONS: Severe lung injury was induced in infant piglets by repeated saline lavage and high tidal volume ventilation. Animals were then randomized to either conventional volume-controlled ventilation or PLV. MEASUREMENTS AND MAIN RESULTS: Organ blood flow was determined in both groups using radiolabeled microspheres under four conditions: high mean airway pressure, Paw; high Paco(2), high Paw; normal Paco(2); low Paw, high Paco(2); low Paw, normal Paco(2). There were no differences in cerebral blood flow during conventional ventilation and PLV, regardless of ventilation strategy. CONCLUSIONS: These results suggest in an acute lung injury model, PLV does not affect cerebral blood flow or other regional organ blood flow over a range of airway pressures.

Low superior vena cava flow and intraventricular haemorrhage in preterm infants

OBJECTIVESTo document the incidence, timing, degree, and associations of systemic hypoperfusion in the preterm infant and to explore the temporal relation between low systemic blood flow and the development of...

Conventional ventilation versus high-frequency oscillation: Hemodynamic effects in newborn babies

We conducted a prospective study to assess the hemodynamic effects of conventional mechanical ventilation (CMV) compared with high-frequency oscillation (HFO) in newborn babies with respiratory distress syndrome. A total of 18 consecutive term and preterm infants were examined by two-dimensional M-mode and pulsed Doppler echocardiography. Five patients had to be excluded, three of them because of increasing cardiovascular support after initiation of HFO. The remaining 13 infants (seven males, six females) had a median gestational age of 33 wks (range, 25-40) and a birth weight of 2350 g (range, 790-3600). Patients entered the study at 21 hrs (range, 5-69) of life, receiving total maintenance fluid of 90 mL/kg/day (range, 60-120). Five babies (38%) needed continuous inotropic support. HFO was used as a rescue therapy in infants who failed with CMV. In all 13 patients, HFO significantly impaired cardiac performance compared with CMV by decreasing aortic velocity-time integral: median, 10.2 cm (range, 6.0-14.6) vs. 8.3 cm (range, 5.3-12.4; p<.002); stroke volume: median, 3.8 mL (range, 1.6-6.8) vs. 3.2 mL (range, 1.3-5.9; p<.002); and cardiac index: 281 mL/min/kg of body weight (range, 177-579) vs. 200 mL/min/kg of body weight (range, 156-591; p<.002). Fractional shortening was also significantly reduced: median, 0.31% (range, 0.24-0.44) vs. 0.29% (range, 0.20-0.34; p<.01), because of a significantly smaller left ventricular diastolic diameter during HFO: median, 1.4 cm (range, 1.0-1.9) vs. 1.4 cm (range, 0.9-1.8; p<.05), with a median difference of -0.07 cm (range, -0.4-0.2). HFO also causes a significant decrease in heart rate-corrected left ventricular ejection time: median, 0.25 sec (range, 0.23-0.28) vs. 0.23 sec (range, 0.21-0.26; p < .02) and heart rate-corrected velocity of circumferential fiber shortening (Vcfc): median, 1.3 circ/sec (range, 1.0-1.6) vs. 1.2 circ/sec (range, 0.9-1.4; p<.05). Left ventricular end-systolic wall stress (LVESWS; g/cm2) remained stable. The correlation between Vcfc and LVESWS did not show any significance (CMV, r2 = .2; HFO, r2 = .09). The regression line between Vcfc and LVESWS showed a higher y-intercept and steeper slope during CMV than during HFO. Heart rate, mean arterial pressure, and left ventricular systolic diameter remained unchanged. In newborn babies, HFO significantly decreased left ventricular cardiac output caused by reduced left ventricular filling and HFO decreased contractility at higher mean airway pressures than with CMV.

Different ventilation strategies affect lung function but do not increase tumor necrosis factor-alpha and prostacyclin production in lavaged rat lungs in vivo.

Using an in vivo animal model of surfactant deficiency, the authors compared the effect of different ventilation strategies on oxygenation and inflammatory mediator release from the lung parenchyma. In adult rats that were mechanically ventilated with 100% oxygen, acute lung injury was induced by repeated lung lavage to obtain an arterial oxygen partial pressure < 85 mmHg (peak pressure/positive end-expiratory pressure [PEEP] = 26/6 cm H2O). Animals were then randomly assigned to receive either exogenous surfactant therapy, partial liquid ventilation, ventilation with high PEEP (16 cm H2O), ventilation with low PEEP (8 cm H2O), or ventilation with an increase in peak inspiratory pressure (to 32 cm H2O; PEEP = 6 cm H2O). Two groups of healthy nonlavaged rats were ventilated at a peak pressure/PEEP of 32/6 and 32/0 cm H2O, respectively. Blood gases were measured. Prostacyclin (PGI2) and tumor necrosis factor-alpha (TNF-alpha) concentrations in serum and bronchoalveolar lavage fluid (BALF) as well as protein concentration in BALF were determined after 90 and 240 min and compared with mechanically ventilated and spontaneously breathing controls. Surfactant, partial liquid ventilation, and high PEEP improved oxygenation and reduced BALF protein levels. Ventilation with high PEEP at high mean airway pressure levels increased BALF PGI2 levels, whereas there was no difference in BALF TNF-alpha levels between groups. Serum PGI2 and TNF-alpha levels did not increase as a result of mechanical ventilation when compared with those of spontaneously breathing controls. Although alveolar protein concentration and oxygenation markedly differed with different ventilation strategies in this model of acute lung injury, there were no indications of ventilation-induced systemic PGI2 and TNF-alpha release, nor of pulmonary TNF-alpha release. Mechanical ventilation at high mean airway pressure levels increased PGI2 levels in the bronchoalveolar lavage-accessible space.

High-frequency oscillatory ventilation (HFOV) in bronchiolitis patients

Respiratory syncytial virus (RSV) is nowadays the leading cause of bronchiolitis and viral pneumonia in children. Although the course is often benign, some children need prolonged hospitalisation and mechanical ventilation or even ECMO if conventional mechanical ventilation (CMV) fails. HFOV is currently considered to be contraindicated in obstructive airway disease with prolonged time constants due to the risk of dynanmic airtrapping. This could give rise to circulatory and ventilatory compromise and barotrauma. Nevertheless, bronchiolitis patients are sometimes put on HFOV after deterioration on CMV. We report 9 patients with RSV bronchiolitis and pulmonary overdistention (small airway disease) successfully treated with HFOV after deterioration on CMV. Although marked hyperinflation was present in all our patients prior to transition, no airleaks developed during HFOV. In one patient the oxygenation index (OI) increased after start of HFOV. Nitric oxide was added and oxygenation improved immediately. All patients survived without residual lung disease. In distinct to current opinion, we showed that small airway disease can safely and successfully be managed with HFOV. Ventilatory strategy should be directed to open up the small airways and keep them open with sufficiently high mean airway pressures (`the open airway strategy' similar to the `open lung strategy' in diffuse alveolar disease). Permissive hypercapnia may be used to reduce pressure swings as much as possible, leading to less shear stress on lung tissue, without influencing airway recruitment. Further dynamic airtrapping can be prevented with the use of longer expiratory than inspiratory times and with prevention of spontaneous breathing. An increasing OI at 48 h may be an indicator of failure of HFOV and alternative treatments should be considered. NO might be such an option to avoid ECMO.

Factors associated with high-frequency oscillatory ventilation response in infants with respiratory failure.

A prospective, clinical study was conducted to investigate the factors associated with the effectiveness and outcome of high-frequency oscillatory ventilation (HFOV) in infants with respiratory failure unresponsive to optimal conventional respiratory and pharmacological therapy. A high-lung volume strategy was used in cases with diffuse lung disease. A total of 30 treatments in 24 patients (including 14 premature infants), aged 1 to 127 days, was included for analysis. By 6 to 8 hours, > 20% improvement in oxygenation (good response) was noted in 57% of treatments. < 20% improvement (poor response) in 20%, 23% showed delayed response 10 to 24 hours after initiation of HFOV. Thirty-five percent of treatments with initial good response deteriorated later because of severe underlying problems or occurrence of ventilatory complications. Seventeen treatments (57%) were weaned successfully to conventional mechanical ventilation (CMV). Poor response was associated with a higher oxygenation index prior to HFOV and a severe underlying problem. Failure to be successfully weaned to CMV was associated with a higher mean airway pressure and a severe underlying condition. Premature infants and infants with the persistent pulmonary hypertension of newborn usually had a good response to HFOV. From results of this study, it is concluded that patients with severe lung disease and a severe underlying problem responded poorly to HFOV. An initial good response did not predict successful weaning.

Low Upper Body and Cerebral Blood Flow in the Preterm Infant. † 1003

AIM: Low blood flow to the preterm upper body and brain is common in the first 24 hours and is associated with a high risk of subsequent intraventricular hemorrhage. We hypothesised low flow would be associated with early large ductal shunts and higher mean airway pressure (MAP). This study aimed to analyse how these and other clinical and echocardiographic factors relate to these low flow states.

Experimental studies on lung mechanics, gas exchange and oxygen delivery under open lung conditions Mechanical ventilation with decelerating versus constant inspiratory flow

The treatment of patients with acute respiratory failure poses an ever‐present clinical challenge since conventional ventilation sometimes causes further impairment of lung function. Thus, it is important to identify characteristics of different ventilatory patterns and ventilatory strategies that affect airway pressures, lung volumes, pulmonary gas exchange and hemodynamics. In order to mimic acute respiratory failure in the patient, we have used an experimental animal injury model, characterized by highly unstable alveoli.The basic prerequisite for this thesis was a positive end‐expiratory pressure (PEEP) sufficient to keep the lung open throughout the whole ventilatory cycle. Different inspiratory flow patterns were studied by using a successful recruitment procedure together with ventilation performed using PEEP levels below, at, and above the inflection‐point pressure. To elucidate the effect of prolonged inspiration time, sufficient PEEP was supplied to prevent end‐expiratory collapse already during the reference mode with an inspiration/expiration ratio (I:E ratio) of 1:1. When an I:E ratio of 2:1 or above was used, both studies (one with mean airway pressure constant and the other with total PEEP constant) showed impaired hemodynamics, but displayed unaffected arterial oxygen tension. Decelerating inspiratory flow delivers the major part of the tidal volume during early inspiration under sustained constant inspiratory pressure. This reduces serial dead space and improves alveolar ventilation. In all studies, decelerating inspiratory flow showed increased carbon dioxide elimination, i.e. improved alveolar ventilation, but decelerating inspiratory flow was found to be no better for alveolar recruitment or in oxygenation.In an attempt to explain radiological observations when studying the relationship of pulmonary radiological appearance to end‐expiratory pressure, Staubs' concept of quantal alveolar behavior was used. This states that the individual alveolus fills essentially independently of its neighbors and has a propensity to be either air‐filled and expanded, or completely fluid‐filled and collapsed. Stepwise sequential deflation of the lung from full expansion, followed by stepwise inflation of the lung, produced a pressure/attenuation loop. This hysteresis phenomenon implies that the attenuation of the lung relates to the set PEEP, but also to the immediately preceding structural status. For identical end‐expiratory pressures the attenuation on the hysteresis down‐slope invariably had a higher value than that of the up‐slope. Thus, the pressure at which alveolar collapse takes place differs from that of expansion.In conclusion, ventilation with decelerating inspiratory flow revealed lower peak inspiratory and higher mean airway pressure than did constant inspiratory flow. No differences in end‐inspiratory pressure were seen. Decelerating inspiratory flow delivers the major part of the tidal volume during early inspiration, which reduces serial dead space and improves alveolar ventilation. The decelerating inspiratory flow was found to be no better than constant inspiratory flow for alveolar recruitment or oxygenation. When the lung was ventilated with PEEP sufficient to prevent end‐expiratory collapse, an inspiration‐to‐expiration ratio at and above 2:1 impaired hemodynamics. Our results suggest that, in order to obtain and maintain open lung conditions, the lung must be recruited and PEEP reduced to the level just above the alveolar collapse pressure.

Randomised controlled trial of high frequency oscillation (HFO) vs conventional ventilation (CV): effects on extracellular volume (ECV) and fluid balance

Background. Higher mean airway pressures (MAD) are used in HFO(high volume strategy) than in CV. High MADs reduce cardiac output and diuresis. Hypothesis. HFO inhibits the postnatal decrease in ECV by reducing diuresis. Methods. Ventilated preterm infants (GA<30 wks) were randomised to HFO (Infant Star ™) or CV (Sechrist™ model IV-100B). We measured ECV (sucrose dilution), urine output, fluid intake and MAD. Results. HFO-infants had higher MADs for 24 hours (median 6.7 vs 3.7 cmH2O; p=0.01). Postnatal urine output increased and ECV decreased in both groups. Conclusion. HFO did not prevent the postnatal adaptation of extracellular volume. Table

Necrotising enterocolitis (NEC) complicating the use of indomethacin for symptomatic patent ductus arteriosus (PDA). Is there a strong association? 909

The pathophysiology of NEC is incompletely understood and so the search for causative factors continues. It has been shown that indomethacin produces significant reduction in mesenteric blood flow(1,2). We hypothesized that the risk of NEC would be increased in very low birth weight (VLBW) infants receiving indomethacin. With the advocation of prevention of intraventricular hemorrhage with prophylactic indomethacin(3,4), we conducted a case control study to investigate the association of NEC and use of indomethacin for closure of PDA in VLBW infants. There were 187 such infants admitted to the Health Sciences Centre Neonatal Intensive Care Unit between January 1993 and December 1994. Of these, 29 developed NEC compared to 158 who did not. The infants who developed NEC were of lower gestational age mean±SE (27.1±0.3 vs 28.8±0.2 p£0.001); birth weight (879±37 vs 1111±22 p£0.0000); and more developed PDA (53% vs 32% p£0.000). Infants who received indomethacin were over represented in the NEC group (29% vs 11.5% p£0.013). There were no differences in inotropic support; surfactant use; prenatal steroids; tocolytics (indomethacin); and the highest mean airway pressure. Indomethacin treatment of PDA is strongly associated with higher risk of developing NEC. This could be a cause-effect association or an indirect association due to birth weight and gestational age differences.

The safety of a nitric oxide inhalation system with high frequency oscillatory ventilation.

Nitric oxide (NO) inhalation and high frequency oscillatory ventilation (HFOV) has been indicated in infants with severe respiratory failure. The purpose of the present study was to evaluate the safety of an NO inhalation system with HFOV in terms of nitrogen dioxide (NO2) production. The NO inhalation system consisted of a high frequency oscillatory ventilator, a neonatal circuit and a test lung. The NO concentration was changed from 0 to 19 p.p.m. At each level of NO, the oxygen (O2) concentration was changed from 21 to 100%. The NO and NO2 concentrations were measured with a chemiluminescence analyzer using a molybdenum converter. The NO2 concentration was increased when either the O2 or the NO concentration was increased. The interposition of the endotracheal tubes increased NO2 concentrations at 4 p.p.m. NO. The high stroke volume and high mean airway pressure produced a significant increase in NO2 production at 4 p.p.m. NO. The increase in NO2 production was prevented by placing a one-way valve at the joint of the NO gas line to the inspired limb. It was concluded that the NO inhalation system with HFOV can be safely used when a one-way valve is placed at the joint of the NO gas line to the inspired limb and when inhaled NO is at a relatively low concentration.

DEXAMETHASONE FOR BRONCHOPULMONARY DYSPLASIA; EFFECTS OF CLINICAL CHOICES ON OUTCOMES? - A MODERN PARABLE. 1222

In 1990, dexamethasone was introduced into our nursery for the treatment of broncho-pulmonary dysplasia following an adaptation of the Cummings Protocol(NEJM 1989; 320: 1505-10). Due to lack of equipoise amongst the clinicians, this was not introduced as a randomised controlled trial. The purpose of this study was to review the consequences of implementation of this new treatment, and what conclusions might be drawn 5 years later. We reviewed the nursery course of all babies who survived of birthweight 800 grams or less between 1990 and 1993 inclusive, and correlated patient selection for dexamethasone, patterns of neonatal illness and 18 month neuro-developmental outcomes. The significance of differences was assessed using ANOVA, MANOVA, chi square and regression analysis. Of the 64 surviving infants, 37 (steroid group) received dexamethasone for (mean 55.3 days, range 26-137 days, SD 25.3 days), and 27(non-steroid group) did not, (birthweight 682±67, 727±63 grams; gestational age 25.3±1.6, 27.3±2.0 weeks respectively). Neonatologists were significantly more likely to give dexamethasone to babies of shorter gestation (p=0.0001), lower birth weight (p=0.0001), and higher mean airway pressure at 10 days (p=0.0001). There were no significant differences in days of ventilation, oxygen therapy or hospital days between the two groups. At 18 months, major impairment (cerebral palsy, PDI <68, MDI <68, legally blind, sensorineural deafness) occurred more commonly in the steroid group, 14 (39%), than in the non-steroid group, 4 (15%), p=0.04. Comparison of impaired babies to those free of impairment across groups showed that infants with major impairments had significantly greater MAP10D(9.3±3.5, 6.4±3.0, p=0.003), and more prolonged treatment with dexamethasone (58.3±36 vs. 22.6±25 days. p=0.0001).

EFFECT OF MEAN AIRWAY PRESSURE (MAP) ON LUNG VOLUME (LV) IN HIGH FREQUENCY OSCILLATORY VENTILATION (HFO) OF PRETERM INFANTS ▴ 2096

In HFO, the strategy aimed at adequate alveolar recruitment by using a sufficiently high mean airway pressure (high volume strategy) results in improved lung function and presumably a better outcome. Measurement of LV, however, has not yet been available in HFO. We present the first data on 1) the LV of preterm infants undergoing HFO, 2) the relationship between LV and MAP and 3) the time needed for stabilization of LV after modification of MAP.

Effects of Pressure-controlled with Different I

Pressure-controlled (PCV) and pressure-controlled inverse ratio ventilation (PCIRV) have been proposed instead of volume-controlled conventional ratio ventilation (VC) with positive end-expiratory pressure (PEEP) for patients with adult respiratory distress syndrome (ARDS). The advantages advocated with the use of PCIRV are to decrease airway pressures and to improve gas exchange. However, most studies did not compare PCIRV and VC while keeping both the level of ventilation and end-expiratory alveolar pressure (total-PEEP) constant. Nine patients with moderate to severe ARDS (lung injury score 2.83 +/- 0.18) had their lungs ventilated with VC, PCV with a conventional ratio (I:E 1:2; PC 1/2), and PCIRV (I:E 2:1 and 3:1; PC 2/1 and PC 3/1, respectively). Ventilator settings were adjusted to keep tidal volume, respiratory rate, FIo2, and total-PEEP constant in every mode. With each mode, a complete set of ventilatory, hemodynamic, and gas exchange parameters was obtained after 30 min. In PC 3/1, the data obtained could not be strictly compared to the other modes because total-PEEP was higher despite external-PEEP being set at zero. For the other modes (VC, PC 1/2, and PC 2/1), despite differences in peak airway pressures, no difference was noted for end-inspiratory and end-expiratory static airway pressures (which better reflect alveolar pressures) nor for lung and respiratory system compliance. Arterial oxygenation deteriorated slightly with PC 2/1 despite a higher mean airway pressure, whereas alveolar ventilation tended to be slightly, but not significantly, improved (lower PaCo2). A decrease in systolic and mean arterial pressure also was observed with PC 2/1 without other significant hemodynamic change. In this prospective controlled study, no short-term beneficial effect of PCV or PCIRV could be demonstrated over conventional VC with PEEP in patients with ARDS.

High-Frequency Ventilation in Acute Pediatric Respiratory Failure

High-Frequency Ventilation in Acute Pediatric Respiratory Failure

Indomethacin compromises hemodynamics during positive-pressure ventilation, independently of prostanoids.

We examined whether prostanoids contribute to the impaired cardiac function and decrease in regional blood flow induced by increasing mean airway pressure. Using microspheres, we measured cardiac output and major organ blood flow and assayed prostaglandin E2, 6-ketoprostaglandin F1 alpha, and thromboxane B2 in blood at mean airway pressures of 5-25 cmH2O in mechanically ventilated newborn piglets treated with ibuprofen (40 mg/kg, n = 6), indomethacin (0.3 mg/kg, n = 6), or vehicle (n = 6). Blood gases and pH were stable throughout the experiments. Prostanoid levels remained constant with increasing mean airway pressure in vehicle-treated pigs and were unchanged by indomethacin. However, ibuprofen decreased the prostanoid levels at all mean airway pressures studied (P < 0.01). As ventilatory pressure was progressively increased, cardiac output decreased gradually and similarly by 42-45% (P < 0.05) in all groups. At the highest mean airway pressure, blood flow decreased to the kidneys by 37-57%, to the ileum by 58-74%, and to the colon by 53-71% (P < 0.05) in all groups. Cerebral blood flow remained constant at all ventilatory pressures regardless of the treatment. There was no difference in cardiac output and regional hemodynamics between ibuprofen- and vehicle-treated animals. However, after indomethacin, ileal blood flow at the higher ventilatory pressures was 41-46% lower and cerebral blood flow at all mean airway pressures was 14-25% lower than after the other treatments (P < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)