Search

Objectives The objective of this study was to provide an updated review of survival for pediatric hematopoietic stem cell transplantation patients requiring extracorporeal membrane oxygenation therapy as well as characterize the demographics, clinical variables, and complications associated with mortality. Design Retrospective database review of the Extracorporeal Life Support Organization Registry from 1990 to 2019. Setting Extracorporeal membrane oxygenation centers reporting to Extracorporeal Life Support Organization. Patients Patients treated with extracorporeal membrane oxygenation greater than 28 days to 18 years old with International Classification of Diseases Ninth Revision, International Classification of Diseases Tenth Revision, and current procedural terminology codes consistent with hematopoietic stem cell transplantation were included. Interventions None. Measurements and main results Demographics, year of extracorporeal membrane oxygenation run, clinical variables, comorbid diagnoses, and extracorporeal membrane oxygenation complications were assessed in relation to the primary study outcome of survival to hospital discharge. Ninety patients were included in the final analysis. The overall survival rate for the study period was 19%. However, the survival rate in the last decade (2010-2019) improved to 26% (p = 0.01; odds ratio 9.4 [1.2-74.8]). Factors associated with decreased survival included comorbid malignancy, elevated peak inspiratory pressure in conventionally ventilated patients, and pulmonary and metabolic complications on extracorporeal membrane oxygenation. Conclusions Pediatric patients posthematopoietic stem cell transplantation supported with extracorporeal membrane oxygenation have improving survival rates over time. With 26% of patients (16/62) surviving to hospital discharge in the last decade (2010-2019), history of hematopoietic stem cell transplantation may no longer be considered an absolute contraindication to extracorporeal membrane oxygenation. As advancements are made in hematopoietic stem cell transplantation therapies and extracorporeal membrane oxygenation management, the indications for life-saving extracorporeal membrane oxygenation support among patients posthematopoietic stem cell transplantation may expand accordingly.

Objective To describe the physiology of air leak in bronchopleural fistula in mechanically ventilated patients and how understanding of its physiology drives management of positive-pressure ventilation. To provide guidance of lung isolation, mechanical ventilator, pleural catheter, and endobronchial strategies for the management of bronchopleural fistula on mechanical ventilation. Data sources Online search of PubMed and manual review of articles (laboratory and patient studies) was performed. Study selection Articles relevant to bronchopleural fistula, mechanical ventilation in patients with bronchopleural fistula, independent lung ventilation, high-flow ventilatory modes, physiology of persistent air leak, extracorporeal membrane oxygenation, fluid dynamics of bronchopleural fistula airflow, and intrapleural catheter management were selected. Randomized trials, observational studies, case reports, and physiologic studies were included. Data extraction Data from selected studies were qualitatively evaluated for this review. We included data illustrating the physiology of driving pressure across a bronchopleural fistula as well as data, largely from case reports, demonstrating management and outcomes with various ventilator modes, intrapleural catheter techniques, endoscopic placement of occlusion and valve devices, and extracorporeal membrane oxygenation. Themes related to managing persistent air leak with mechanical ventilation were reviewed and extracted. Data synthesis In case reports that demonstrate different approaches to managing patients with bronchopleural fistula requiring mechanical ventilation, common themes emerge. Strategies aimed at decreasing peak inspiratory pressure, using lower tidal volumes, lowering positive end-expiratory pressure, decreasing the inspiratory time, and decreasing the respiratory rate, while minimizing negative intrapleural pressure decreases airflow across the bronchopleural fistula. Conclusions Mechanical ventilation and intrapleural catheter management must be individualized and aimed at reducing air leak. Clinicians should emphasize reducing peak inspiratory pressures, reducing positive end-expiratory pressure, and limiting negative intrapleural pressure. In refractory cases, clinicians can consider lung isolation, independent lung ventilation, or extracorporeal membrane oxygenation in appropriate patients as well as definitive management with advanced bronchoscopic placement of valves or occlusion devices.

OBJECTIVES Acute respiratory failure is common in pediatric hematopoietic cell transplant recipients and has a high mortality. However, respiratory prognostic markers have not been adequately evaluated for this population. Our objectives are to assess respiratory support strategies and indices of oxygenation and ventilation in pediatric allogeneic hematopoietic cell transplant patients receiving invasive mechanical ventilation and investigate how these strategies are associated with mortality. DESIGN Retrospective, multicenter investigation. SETTING Twelve U.S. pediatric centers. PATIENTS Pediatric allogeneic hematopoietic cell transplant recipients with respiratory failure. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Two-hundred twenty-two subjects were identified. PICU mortality was 60.4%. Nonsurvivors had higher peak oxygenation index (38.3 [21.3-57.6] vs 15.0 [7.0-30.7]; p < 0.0001) and oxygen saturation index (24.7 [13.8-38.7] vs 10.3 [4.6-21.6]; p < 0.0001), greater days with FIO2 greater than or equal to 0.6 (2.4 [1.0-8.5] vs 0.8 [0.3-1.6]; p < 0.0001), and more days with oxygenation index greater than 18 (1.4 [0-6.0] vs 0 [0-0.3]; p < 0.0001) and oxygen saturation index greater than 11 (2.0 [0.5-8.8] vs 0 [0-1.0]; p < 0.0001). Nonsurvivors had higher maximum peak inspiratory pressures (36.0 cm H2O [32.0-41.0 cm H2O] vs 30.0 cm H2O [27.0-35.0 cm H2O]; p < 0.0001) and more days with peak inspiratory pressure greater than 31 cm H2O (1.0 d [0-4.0 d] vs 0 d [0-1.0 d]; p < 0.0001). Tidal volume per kilogram was not different between survivors and nonsurvivors. CONCLUSIONS In this cohort of pediatric hematopoietic cell transplant recipients with respiratory failure in the PICU, impaired oxygenation and use of elevated ventilator pressures were common and associated with increased mortality.

OBJECTIVE We hypothesized that pretreatment with atorvastatin improves alveolar capillary permeability and hemodynamics and, thus, confers protection against lung injury caused by high-stretch mechanical ventilation. METHODS Twenty-four isolated sets of normal rabbit lungs were utilized. Treated animals received atorvastatin (20 mg/kg body weight/day by mouth) for 3 days before surgery. Lungs were perfused constantly (300 mL/min) and ventilated for 1 hr with pressure-control ventilation at either 23 (high pressure; resulting in tidal volume approximately 22 mL/kg) or 11 (low pressure; tidal volume approximately 10 mL/kg) cm H2O peak inspiratory pressure and positive end-expiratory pressure of 3 cm H2O. Four groups were examined: high pressure-no statin, high pressure-statin pretreatment, low pressure-no statin, and low pressure-statin pretreatment. RESULTS The high-pressure-no statin group sustained more damage than the low-pressure groups. In high-pressure groups, lungs of statin-pretreated vs. no statin-pretreated animals sustained a significantly lower increase in ultrafiltration coefficient (an accurate marker of alveolar capillary permeability; high-pressure-statin pretreatment vs. high-pressure-no statin, -0.013 +/- 0.017 g/min/mm Hg/100g vs. 1.723 +/- 0.495 g/min/mm Hg/100g; p < .001), lower weight gain (i.e., less edema formation; 4.62 +/- 1.50 grams vs. 17.75 +/- 4.71 grams; p = .005), improved hemodynamics (i.e., lower increase in mean pulmonary artery pressure; 0.56 +/- 0.51 mm Hg vs. 5.62 +/- 1.52 mm Hg; p = .04), lower protein concentration in bronchoalveolar lavage fluid (p < .001), and fewer histologic lesions (p = .013). Apoptosis of lung parenchyma cells was not different (p = .97). There was no difference between low-pressure-statin pretreatment and low-pressure-no statin groups regarding these outcomes. CONCLUSION In this model, atorvastatin improves alveolar capillary permeability and hemodynamics and, thus, attenuates lung injury caused by high-stretch mechanical ventilation.

Objective Uncertainties about the numerous degrees of freedom in ventilator settings leave many unanswered questions about the biophysical determinants of lung injury. We investigated whether mechanical ventilation with high air flow could yield lung mechanical stress even in normal animals. Design Prospective, randomized, controlled experimental study. Setting University research laboratory. Subjects Thirty normal male Wistar rats (180-230 g). Interventions Rats were ventilated for 2 hrs with tidal volume of 10 mL/kg and either with normal inspiratory air flow (V') of 10 mL/s (F10) or high V' of 30 mL/s (F30). In the control group, animals did not undergo mechanical ventilation. Because high flow led to elevated respiratory rate (200 breaths/min) and airway peak inspiratory pressure (PIP,aw = 17 cm H2O), two additional groups were established to rule out the potential contribution of these variables: a) normal respiratory rate = 100 breaths/min and V' = 30 mL/sec; and b) PIP,aw = 17 cm H2O and V' = 10 mL/sec. Measurements and main results Lung mechanics and histology (light and electron microscopy), arterial blood gas analysis, and type III procollagen messenger RNA expression in lung tissue were analyzed. Ultrastructural microscopy was similar in control and F10 groups. High air flow led to increased lung plateau and peak pressures, hypoxemia, alveolar hyperinflation and collapse, pulmonary neutrophilic infiltration, and augmented type III procollagen messenger RNA expression compared with control rats. The reduction of respiratory rate did not modify the morphofunctional behavior observed in the presence of increased air flow. Even though the increase in peak pressure yielded mechanical and histologic changes, type III procollagen messenger RNA expression remained unaltered. Conclusions Ventilation with high inspiratory air flow may lead to high tensile and shear stresses resulting in lung functional and morphologic compromise and elevation of type III procollagen messenger RNA expression.

OBJECTIVE Acute respiratory distress syndrome is occasionally seen in newborn infants due to a severe inflammatory process in the lungs that affects capillary-alveolar permeability, epithelial integrity, and type I and II pneumocyte function. The aim of this study was to investigate the effect of a topically applied nuclear factor-kappaB inhibitor (IkappaB kinase-NF-kappaB essential modulator binding domain [IKK-NBD] peptide) on gas exchange, lung function, lung fluids, and inflammation in a piglet model of repeated airway lavage that is characterized by surfactant deficiency, lung edema, and an inflammatory response. DESIGN Prospective, randomized, controlled animal study. SETTING Research laboratory of a university children's hospital. SUBJECTS A total of 24 anesthetized, mechanically ventilated newborn piglets. INTERVENTIONS Repeated airway lavage was carried out until both the Pao2 decreased to approximately 40 mm Hg, while ventilating the piglets with an Fio2 of 0.6, and a peak inspiratory pressure of >/=18 cm H2O was needed to maintain tidal volume at 6 mL/kg. One group of piglets served as a control (n = 8), a second group (S, n = 8) received a porcine surfactant preparation (Curosurf), and a third group received IKK-NBD peptide admixed to surfactant (S+IN, n = 8). MEASUREMENTS AND MAIN RESULTS After 6 hrs of mechanical ventilation after intervention, S+IN group piglets showed decreased extravascular lung water (S+IN vs. S, 20 +/- 3 vs. 28 +/- 10 mL/kg; p < .05) and a lesser protein content in the epithelial lining fluid (S+IN vs. S, 38 +/- 5 vs. 50 +/- 5 mg/L; p < .05). Functional residual capacity (S+IN vs. S, 16.7 +/- 6.3 vs. 12.2 +/- 4.3 mL/kg; p < .05), alveolar volume (S+IN vs. S, 5.4 +/- 1.8 vs. 4.6 +/- 1.5 mL/kg; p < .05), and lung mechanics were improved. Bronchoalveolar lavage showed a lesser percentage of polymorphonuclear leukocytes (S+IN vs. S, 70% +/- 6% vs. 82% +/- 3%; p < .01) and a reduction in the chemokine leukotriene B4 (S+IN vs. S, 2.0 +/- 0.6 vs. 3.5 +/- 1.4 pg/mL; p < .01). CONCLUSIONS A topically applied nuclear factor-kappaB inhibitor improves lung edema and lung volumes and reduces inflammation in this newborn piglet model of airway lavage.

Objective: To examine the isolated and combined effects of positive end-expiratory pressure (PEEP) and partial liquid ventilation (PLV) on the development of ventilator-induced lung injury in an ovine model. Design: Prospective controlled animal study. Setting: University-based cardiovascular animal physiology laboratory. Subjects: Thirty-eight anesthetized supine sheep weighing 22.3 ± 2.2 kg. Interventions: Animals were ventilated for 6 hrs (respiratory rate, 15; FIO 2 , 1.0, inspiratory/expiratory ratio, 1:1) with one of five pressure-controlled strategies, expressed as peak inspiratory pressure (PIP)/PEEP: low-PIP, 25/5 cm H 2 O (n = 8); high-PIP, 50/5 cm H 2 O (n = 8); high-PIP-PLV, 50/5 cm H 2 O-PLV (n = 8); high-PEEP, 50/20 cm H 2 O (n = 7); and high-PEEP-PLV, 50/20 cm H 2 O-PLV (n = 7). Measurements and Main Results: Compared with the low-PIP control, high-PIP ventilation increased airleak, shunt, histologic evidence of lung injury, neutrophil infiltrates, and wet lung weight. Maintaining PEEP at 20 cm H 2 O or adding PLV reduced the development of physiologic shunt and dependent histologic injury indexes. Neither higher PEEP nor PLV reduced the high incidence of barotrauma observed in high-PIP animals. Conclusions: We conclude that application of PLV or PEEP at 20 cm H 2 O may improve gas exchange and afford lung protection from ventilator-induced lung injury during high-pressure mechanical ventilation in this model.

OBJECTIVE To examine the effects of square wave, sinusoidal, and linear inspiratory pressure waveforms during pressure-controlled assist/control ventilation on the firing pattern of pulmonary stretch receptors and phrenic nerve activity. DESIGN Experimental, comparative study. SETTING Research laboratory at a university biomedical center. SUBJECTS Nine anesthetized, endotracheally intubated young cats (2.5-3.4 kg). INTERVENTION With interposed periods of continuous positive airway pressure (0.2 kPa), each cat was exposed to periods of assist/control ventilation with three different pressure waveforms, where the peak inspiratory pressure (0.74 +/- 0.13 kPa), end-expiratory pressure (0.2 +/- 0.02 kPa), and tidal volume (14.9 +/- 5.22 mL/kg) were kept constant. Preset controlled ventilator rate was set below the rate of spontaneous breathing, and the mechanical inflation time equaled the inspiratory time during spontaneous breathing on continuous positive airway pressure. MEASUREMENTS AND MAIN RESULTS Respiratory rate and arterial blood gases did not change between the three pressure waveforms during assist/control ventilation. Peak pulmonary stretch receptor activity was lower and mean phrenic nerve activity higher during continuous positive airway pressure than during assist/control ventilation (p

Objective: To investigate, in a rat model, the role of the Mac1/ICAM-1 pathway and the anti-inflammatory activity of steroid in ventilator-induced lung injury. Design: Prospective, randomized controlled study. Setting: Animal investigation using Wistar rats. Intervention: Rats in three randomly assigned groups of 18, a total of 54 animals, were subject to the following: Two groups received high peak inspiratory pressure (35 cm H 2 O) ventilation after pretreatment with methylprednisolone (high-methylprednisolone group) or pretreatment with methylprednisolone vehicle (high-vehicle group). The third group of animals received low peak inspiratory pressure (7 cm H 2 O) ventilation after pretreatment with methylprednisolone vehicle (low-vehicle group). Except for animals previously killed to establish baseline values, after 40 mins of mechanical ventilation, the animals in each group were killed. Some animals provided histological samples, and the rest received total lung lavage. Measurement: We measured flow cytometry of lavage fluid, cell counts of tissue samples, and pressure-volume curves before and after mechanical ventilation. Results: In the groups that received high peak inspiratory pressure ventilation, both the number of neutrophils that infiltrated the lungs and the expression of Mac-1 and ICAM-1 on neutrophils and macrophages increased significantly more than in the low-vehicle group. Static lung compliance was reduced in the high peak inspiratory pressure groups. In the high peak inspiratory pressure groups, there were significantly fewer neutrophils in samples from the high-methylprednisolone group (0.412 ± 0.1 x 10 5 ) than from the high-vehicle group (1.10 ± 0.1 x 10 5 ; p

OBJECTIVES: To compare carinal pressures vs. proximal airway pressures, and gas exchange efficacy with a constant minute volume, in lung-injured rabbits during conventional mechanical ventilation (CMV) and intratracheal pulmonary ventilation (ITPV); and to evaluate performance of a prototype ITPV gas delivery and continuous airway pressure monitoring system. DESIGN: Prospective controlled study. SETTING: Animal research laboratory at a teaching hospital. SUBJECTS: Sixteen adult female rabbits. INTERVENTIONS: Anesthetized rabbits were tracheostomized with a multilumen endotracheal tube. Anesthesia and muscle relaxation were maintained continuously throughout the study. Proximal airway pressures and carinal pressures were recorded continuously. The injection port of the multilumen endotracheal tube was used for the carinal pressure monitoring. To prevent obstruction of the port, it was flushed with oxygen at a rate of 11 mL/min. CMV was initiated with a pressure-limited, time-cycled ventilator set at an FiO2 of 1.0 and at a flow of 1.0 L/kg/min. The pressure limit of the ventilator was effectively disabled. A normal baseline for arterial blood gases was achieved by adjusting the inspiratory/expiratory time ratios. ITPV was established using a flow of 1.0 L/kg/min through a reverse thrust catheter, at the same baseline and inspiratory/expiratory ratio. Carinal positive end-expiratory pressure was maintained at a constant value of 2 cm H2O by adjusting the expiratory resistance of the ventilator circuit Lung injury was achieved over a 30-min period by three normal saline lavages of 5 mL/kg each. After lung injury, all animals were consecutively ventilated for 1 hr with CMV, for 1 hr with ITPV, and again for 1 hr with CMV. Six rabbits were ventilated at 30 breaths/min (group 1), and ten rabbits were ventilated at 80 breaths/min (group 2). Four rabbits in group 2 were subjected, 1 hr after return to CMV from ITPV, to another session of ITPV, with positive end-expiratory pressure gradually being increased to 4, 6, and 8 cm H2O for 15 mins each. RESULTS: No significant differences were observed in carinal peak inspiratory pressure between CMV and ITPV modes, at both low and high frequencies of breathing, indicating that the inspired tidal volume remained constant during both modes of ventilation. Significant gradients were noted between proximal airway and carinal peak inspiratory pressure during ITPV but not during CMV. Initiation of ITPV, at a flow of 1.0 L/kg/min, required an increase in the ventilator expiratory resistance to maintain a constant level of positive end-expiratory pressure (2 cm H2O) as measured at the carina. During ITPV, the PaCO2 was significantly reduced by 20% at 30 breaths/min (p < .05) and by 22% at 90 breaths/min (p < .01), compared with CMV. Arterial oxygenation was significantly enhanced with a positive end-expiratory pressure of 6 and 8 cm H2O (p < .05 and .001, respectively), compared with a positive end-expiratory pressure of 2 cm H2O during ITPV. All components of the new prototype gas delivery and airway pressure monitoring system functioned without failure, at least for 3 hrs of the CMV, ITPV, and CMV trials. CONCLUSIONS: ITPV in saline-lavaged, lung-injured rabbits at breathing frequencies of 30 and 80 breaths/min, compared with CMV at the same minute ventilation, can improve CO2 exchange. During ITPV, significant pressure gradients can develop between carinal and proximal airway pressures. Continuous carinal pressure monitoring is therefore necessary for the safe clinical application of ITPV. Reliable carinal pressure monitoring can be achieved by adding a small bias flow through the carinal pressure monitoring port. Although ITPV can remove CO2 from injured lungs efficiently, simultaneous addition of positive end-expiratory pressure can further improve arterial oxygenation.

OBJECTIVE To assess whether the delivered nitric oxide (NO) concentration is affected by a change in the ventilatory setting during neonatal mechanical ventilatory support. DESIGN Prospective, experimental study. SETTING Laboratory at Nagoya City University Medical School. INTERVENTIONS This study was performed by using a pressure-limited, time-cycled, ventilatory support with a neonatal circuit and a 50-mL silicone test lung. NO in N2 gas was administrated into the inspiratory limb at a distance of 4 cm, 80 cm, or 160 cm from the Y piece connected to the adapter of an endotracheal tube. The NO concentration was measured every 0.5 sec by a chemiluminescence analyzer at the Y piece. MEASUREMENT AND MAIN RESULTS NO concentrations were compared with each of the ventilatory settings of peak inspiratory pressure (PIP) (10-30 cm H2O), positive end-expiratory pressure (0-10 cm H2O), ventilatory flow (10, 20, 30 L/min), and ventilatory rate (30, 40, 50, 60, 70 breaths/min), respectively. The NO concentration was significantly lower when NO was added at 4 cm than at 80 cm or 160 cm from Y piece at the same ventilatory setting of PIP, positive end-expiratory pressure and ventilatory flow, respectively, (p < .01). Although the NO concentration was increased as the settled PIP level was increased (p < .01 or p < .05), it was not changed when the settled positive end-expiratory pressure level was increased. A decrease was seen in the NO concentration as the settled ventilatory flow was increased (p < .01). Lastly, the NO concentration fluctuated greatly in association with the settled ventilatory rate. CONCLUSION The NO concentration delivered to patients is influenced by the ventilatory setting during neonatal mechanical ventilatory support.

OBJECTIVE: To compare intratracheal pulmonary ventilation (ITPV) with conventional ventilation in a rabbit model of surfactant deficiency. DESIGN: A prospective randomized animal study. SETTING: The Children's National Medical Center Research Animal Facility in Washington, DC. SUBJECTS: Adult male New Zealand white rabbits (n = 20), weighing 1.4-4.2 kg. INTERVENTIONS: After anesthesia and catheter placement, rabbits were tracheotomized, paralyzed, and placed on the conventional ventilator. We determined pulmonary functions at baseline. We washed surfactant out of the lungs by using serial bronchoalveolar lavages. Pulmonary function studies were determined after completion of the bronchoalveolar lavages and were used as an indication of severity of lung injury. Animals were randomized into two groups: We placed ten animals on ITPV, using the ITPV reverse thruster catheter designed by Kolobow and a prototype ITPV ventilator designed at Children's National Medical Center; we placed ten animals on conventional ventilation using the Sechrist iv-100 ventilator. Arterial blood gases were drawn every 15 mins, and the ventilator settings were adjusted to the minimal level that would maintain arterial blood gases in the following ranges: pH 7.35-7.45, PaCO2 30-40 torr (3.995.33 kPa), PaO2 50-70 torr (6.66-9.33 kPa). Animals were ventilated with the randomized ventilation techniques for 4 hrs. MEASUREMENTS AND MAIN RESULTS: Peak inspiratory pressure, mean airway pressure, and positive end-expiratory pressure were measured at the distal end of the endotracheal tube. We recorded these variables plus respiratory rate at baseline and every 30 mins for a total of 4 hrs of ventilation. Lung compliance did not differ between groups at the postlavage study period (ITPV, 0.56+/-0.13 mL/cm H2O/kg; conventional 0.49+/-0.15 mL/cm H2O/kg). At the end of the 4 hr study period, peak inspiratory pressure (ITPV, 26.2+/-4.6 cm H2O; conventional, 32.4+/-5.04 cm H2O, p = .007) and positive end-expiratory pressure (ITPV, 3.9+/-1.96 cm H2O; conventional, 6.3+/-1.42 cm H2O, p = .005) were lower in the ITPV ventilation group. Peak inspiratory pressure was significantly lower in the ITPV group by 2 hrs into the study. CONCLUSION: In this model of surfactant deficiency lung injury, ventilation and oxygenation were achieved at significantly lower ventilator settings using ITPV compared with conventional ventilation. Long-term studies are needed to determine whether this reduction in ventilation is maintained, and if so, if lung injury is reduced.

OBJECTIVE To determine whether intratracheal pulmonary ventilation (ITPV) allows for effective oxygenation and ventilation at lower mean airway pressures and peak inspiratory pressures than conventional ventilation in a piglet model of meconium aspiration syndrome. DESIGN Prospective, interventional study. SETTING The animal research laboratory at Children's National Medical Center, Washington, DC. SUBJECTS Twenty newborn piglets, 2 to 7 days of age, weighing 1.8 to 2.8 kg. INTERVENTION Animals were anesthetized, paralyzed, intubated, and ventilated. Femoral arterial and venous catheters were inserted; 5 mL/kg of 20% meconium in normal saline was instilled into the endotracheal tube. Animals were randomized to either ITPV or conventional ventilation, and settings were adjusted to maintain ideal blood gases, i.e., pH 7.35 to 7.45, PCO2 40 to 45 torr (5.3 to 6 kPa), PO2 80 to 100 torr (10.7 to 13.3 kPa), and SaO2 > or = 90%. Ventilatory settings were adjusted as needed to a maximum of: FIO2 1.0, peak inspiratory pressure 40 cm H2O, positive end-expiratory pressure 5 cm H2O, and respiratory rate 80 breaths/min. MEASUREMENTS AND MAIN RESULTS Arterial blood gases were taken every 30 mins for 4 hrs and ventilatory settings were adjusted to maintain optimal blood gases. Heart rate, mean arterial blood pressure, and arterial saturation were monitored continuously. The animals in the ITPV group had significantly lower peak inspiratory pressure at 1, 2, 3, and 4 hrs after meconium instillation (p < .018) and significantly lower mean airway pressure at 2, 3, and 4 hrs after meconium instillation (p < .03). The mean peak inspiratory pressure in the ITPV animals ranged from 17 +/- 2.7 cm H2O at baseline to 16.6 +/- 5.7 cm H2O at 4 hrs compared with 16.5 +/- 2.7 cm H2O at baseline to 31.8 +/- 9.1 cm H2O at 4 hrs in the conventionally ventilated animals (p < .04). The mean airway pressure ranged from 6.3 +/- 1.1 mm Hg at baseline to 6.8 +/- 2.5 mm Hg at 4 hrs in the ITPV group compared with 5.5 +/- 1.2 mm Hg at baseline to 10.7 +/- 3.4 mm Hg at 4 hrs in the conventional ventilation group (p < .03). The lungs of the ITPV animals were less hemorrhagic and had less pathologic evidence of injury than the lungs of the conventionally ventilated animals. CONCLUSIONS These results indicate that ITPV can be used to effectively ventilate and oxygenate piglets with meconium aspiration syndrome at lower mean airway pressures and peak inspiratory pressures than conventional ventilation. This lower pressure causes less injury to the lungs of the animals.

Objective To summarize our institutional experience concerning mechanical ventilation support and blood gas measurements in the 24-hr period following extracorporeal life support (ECLS) for pediatric acute respiratory failure. Design Descriptive study. Setting A tertiary pediatric referral center. Patients Children aged 1 month to 18 yrs treated with ECLS for acute respiratory failure at University of Michigan Hospitals from November 1982 to June 1993. All patients aged 1 month to 18 yrs who received ECLS for acute respiratory failure were included. Patients who received ECLS primarily for cardiac support or who had a diagnosis of congenital gastrointestinal malformation (i.e. congenital diaphragmatic hernia) were excluded. Interventions ECLS for severe pediatric respiratory failure. Measurements and main results Forty-nine children were treated at our center with ECLS for acute respiratory failure 36 (73%) survived. Ventilator settings immediately after decannulation from ECLS for survivors were as follows: FIO2 0.53 +/- 0.18 (SD); intermittent mandatory ventilation (IMV) 29.6 +/- 1.18 breaths/min, positive end-expiratory pressure 5.3 +/- 1.6 cm H2O, mean airway pressure 12.6 +/- 2.9 cm H2O, and peak inspiratory pressure 31.7 +/- 5.5 cm H2O. Arterial blood gas measurements at decannulation were PaO2 89.4 +/- 30.9 torr (11.9 +/- 4.1 kPa), PaCO2 43.7 +/- 9.7 torr (5.8 +/- 1.3 kPa), and pH 7.39 +/- 0.07. Twenty-four hours after decannulation, ventilator settings and arterial blood gas measurements were as follows: FIO2 0.42 +/- 0.14, IMV 27.4 +/- 13.5 breaths/min, positive end-expiratory pressure 5.2 +/- 1.6 cm H2O, mean airway pressure 12.0 +/- 3.4 cm H2O, peak inspiratory pressure 31.1 +/- 6.5 cm H2O, PaO2 77.0 +/- 16.9 torr (10.3 +/- 2.3 kPa), PaCO2 44.9 +/- 8.4 torr (6.0 +/- 1.1 kPa), and pH 7.40 +/- 0.07. Variables associated with oxygenation status (P[A-a]O2) and mean airway pressure (oxygenation index) improved during the immediate 24-hr period postbypass (p Conclusions Successful decannulation from ECLS for > 24 hrs resulted in long-term survival in 97% (36/37) of children. Ventilator parameters and arterial blood gas measurements during the 24-hr period following bypass have been described for this population. Such conventional support may indicate safe levels of oxygen and mechanical ventilation pressures for the postbypass recovering lung.

Objective: To determine whether reducing FIO 2 during perfluorocarbon-associated gas exchange would cause deterioration of hemodynamics, lung mechanics, or gas exchange in normal piglets. Design: A prospective, controlled animal trial. Setting: Experimental animal laboratory in a university setting. Subjects: Twelve normal anesthetized piglets, 7 to 14 days old, and weighing 3.31±0.75 kg. Interventions: After the induction of anesthesia, tracheostomy and catheterization, piglets were stabilized. They were mechanically ventilated with a tidal volume of 15 mL/kg, inspiratory time of 25%, positive end-expiratory pressure of 4 cm H 2 O, and a respiratory rate of 20 to 28 breaths/min to obtain a baseline PaCO 2 between 34 and 45 torr (4.7 and 6.0 kPa). Each animal was studied during continuous positive-pressure breathing, and during perfluorocarbon-associated gas exchange. They were ventilated at an FIO 2 of 1.0 for 15 mins. FIO 2 was randomly varied among 0.75, 0.5, and 0.3 every 15 mins, then returned to 1.0. At each FIO 2 , measurements of gas exchange, lung mechanics, and hemodynamics were made. After continuous positive-pressure breathing, perfluorocarbon-associated gas exchange was instituted by replacing the gaseous functional residual capacity of the lungs with perfluorooctylbromide. Animals were then ventilated and measurements were taken. Measurements and Main Results: At each FIO 2 , measurements of gas exchange (arterial blood gases and saturation), lung mechanics (mean airway pressure, static end-inspiratory pressure, and peak inspiratory pressure), and hemodynamics (heart rate, and mean arterial, right atrial, pulmonary artery occlusion, and pulmonary arterial pressures) were recorded. In six piglets, cardiac output was measured at each FIO 2 by thermodilution. Cardiac index, indexed oxygen delivery and consumption, and indexed pulmonary vascular resistance were derived using standard formulas. Piglets were well saturated at all FIO 2 settings during continuous positive-pressure breathing. However, during perfluorocarbon-associated gas exchange, arterial saturation decreased to 72% at an FIO 2 of 0.3. Cardiac index and oxygen consumption were not affected by reducing FIO 2 during perfluorocarbon-associated gas exchange, and were not significantly different than during continuous positive-pressure breathing. Oxygen delivery was reduced at an FIO 2 of 0.3 during perfluorocarbon-associated gas exchange, but oxygen consumption remained in the flow independent portion of the curve despite arterial desaturation. Pulmonary arterial pressure was higher during perfluorocarbon-associated gas exchange than during continuous positive-pressure breathing. Pulmonary arterial pressure and indexed pulmonary vascular resistance were significantly higher during perfluorocarbon-associated gas exchange at an FIO 2 of 0.3 than at any other FIO 2 settings. Conclusions: Piglets showed no adverse effects on lung mechanics during perfluorocarbon-associated gas exchange. Hemodynamics were well supported at all FIO 2 settings, and arterial blood was fully oxygenated during perfluorocarbon-associated gas exchange at an FIO 2 of ≥0.5

OBJECTIVES The purposes of this report are to a) describe the University of Michigan experience with venoarterial or venovenous extracorporeal life support for severe pediatric pulmonary rescue therapy, and b) examine survivors and nonsurvivors for differences that might be useful for examination in future, prospective studies. DESIGN Case series report. Phase I study of safety and effectiveness of extracorporeal life support for pediatric respiratory failure. SETTING University of Michigan Medical Center. PATIENTS Non-neonatal pediatric patients treated with extracorporeal life support for severe respiratory failure at the University of Michigan. INTERVENTIONS Extracorporeal life support for pulmonary failure. MEASUREMENTS AND MAIN RESULTS From November 1982 until May 1991, 25 pediatric patients underwent extracorporeal life support for severe pulmonary failure. Twenty patients were treated in the last 36 months. Sixty percent (15/25 patients) survived their life-threatening respiratory illness, were weaned from mechanical ventilation, and were discharged home. The mean patient age was 4.1 yrs, and mechanical ventilation duration before extracorporeal life support was 5.9 days. Mean blood gas data and mechanical ventilation pressures before extracorporeal life support were: peak inspiratory pressure of 48.6 cm H2O, mean airway pressure of 21.9 cm H2O, positive end-expiratory pressure of 9.7 cm H2O, PaCO2 of 43 torr (5.7 kPa), PaO2 of 69 torr (9.1 kPa), estimated alveolar-arterial oxygen gradient of 563 torr (75 kPa), and FIO2 of 0.98. Variables associated with survival included: age of survivors vs. nonsurvivors, 2.1 vs. 7.1 yrs (p less than .02); peak inspiratory pressure of survivors vs. nonsurvivors, 43.1 vs. 57.9 cm H2O (p less than .03); mean airway pressure of survivors vs. nonsurvivors, 18.4 vs. 27.2 cm H2O (p less than .03); and positive end-expiratory pressure of survivors vs. nonsurvivors, 8.1 vs. 12.1 cm H2O (p less than .01). There were no differences detectable in the blood gas values (PaO2, PaCO2, P[A-a]O2) in survivors and nonsurvivors before extracorporeal life support. The number of days mechanical ventilation was used before extracorporeal life support in survivors and in nonsurvivors was similar. CONCLUSION Extracorporeal life support is an effective rescue therapy for pediatric patients with severe respiratory failure (University of Michigan survival rate of 60%).

OBJECTIVE To establish if the pressure indicated in the manometer of an infant ventilator (IV 100B, Sechrist, Anaheim, CA) reflects the true pressure delivered to the proximal airway during mechanical ventilation in the neonatal ICU. DESIGN With approval of our Institutional Research Board, data were collected prospectively. Peak inspiratory pressure and end-expiratory pressure were measured at the "Y" piece of the breathing tubing. Pressure readings from the conventional ventilator's manometer were compared with simultaneously obtained measurements using an electronic monitor. SETTING This study was conducted in a 45-bed neonatal ICU, admitting 700 to 750 newborns per year. PATIENTS Twelve neonates who required mechanical ventilation were included in the study. INTERVENTIONS Specific interventions were not made by study design. Measurements routinely obtained were compared. MEASUREMENTS AND MAIN RESULTS Two hundred seventy-five simultaneous measurements of peak inspiratory pressure and positive end-expiratory pressure were compared. Peak inspiratory pressure values were higher with the electronic monitor in 273 (99%) of 275 measurements and the mean of the differences between the electronic monitor and ventilator's manometer was statistically significant (p less than .001). For positive end-expiratory pressure measurements, values indicated by the electronic monitor were lower in 152 (55%) of 275 determinations, equal in 65 (23%), and higher in 58 (21%) determinations. Percent variations between methods ranged from 0% to 140% for peak inspiratory pressures and from 0% to 500% for positive end-expiratory pressure. CONCLUSIONS These data demonstrate that it is impossible to know the true pressure delivered to the proximal airway of a neonate during mechanical ventilation by observing the ventilator pressure manometer. The manometer readings consistently underestimate the true peak inspiratory pressure values and are very unpredictable regarding positive end-expiratory pressure values. These findings support the use of other methods to monitor the proximal airway pressure besides the ventilator's manometer in the neonatal ICU. Furthermore, mean airway pressure should not be calculated from the pressure readings obtained from the tested ventilator's manometer.

To study the pulmonary microvascular injury produced by ventilation barotrauma, the isolated perfused lungs of 4 to 6-wk-old New Zealand white rabbits were ventilated by one of the following methods: peak inspiratory pressure (PIP) 23 cm H2O, gas flow rate 1.1 L/min (group 1); PIP 27 cm H2O, gas flow rate 6.9 L/min (group 2); PIP 50 cm H2O, gas flow rate 1.9 L/min (group 3); or PIP 53 cm H2O, gas flow rate 8.3 L/min (group 4). Microvascular permeability was assessed using the capillary filtration coefficient (Kfc) before and 5, 30, and 60 min after a 15-min period of ventilation. Baseline Kfc was not significantly different between groups. A significant increase over the baseline Kfc was noted at 60 min in group 2 and in all postventilation Kfc values in groups 3 and 4 (p less than .05). Group 1 Kfc values did not change significantly after ventilation. At all post-ventilation times, values for Kfc were significantly greater in groups 3 and 4 than in group 1 (p less than .05). Group 4 Kfc values were significantly greater than those in group 2 at 5 and 30 min postventilation. These data indicate that high PIP, and to a lesser extent, high gas flow rates cause microvascular injury in the compliant nonadult lung and suggest that the combination of high PIP and high gas flow rates are the most threatening to microvascular integrity.

Objectives A five-element lumped pulmonary model was developed to estimate respiratory mechanics automatically and noninvasively. The model was applied to diagnose obstructed endotracheal tube. Events like bronchospasm and stiff chest wall were also tested to determine the specificity of the diagnosis. Cases with positive end-expiratory pressure were also included in the analysis to see the effects of positive end-expiratory pressure on the model. Design Randomized controlled animal study. Setting University department of anesthesiology. Subjects Ten anesthetized, paralyzed, and mechanically ventilated mongrel dogs (19-45 kg) of either gender. Interventions Two levels of upper airway obstruction were induced in ten dogs by partially constricting the endotracheal tube. Acute bronchial constriction was produced in five dogs by injecting methacholine through a central venous catheter. In the same five dogs, the chest wall was stiffened by wrapping a pressure cuff around the chest. Positive end-expiratory pressure was also applied as a separate event in these five animals. Measurements and main results Airway pressure and flow were continuously recorded at the mouth. Model parameters were iteratively identified until the root mean square error between respiratory impedance (obtained from airway pressure and flow) and model-predicted impedance (calculated using Ohm's law) was minimum. The peak inspiratory pressure increased and the peak expiratory flow rate decreased with increasing levels of partial obstruction. The value of the model parameters R(1) and C(2) increased and C(1) decreased with partial obstructed endotracheal tube, whereas R(1) increased and L and C(2) decreased with bronchospasm. With stiff chest wall, R(2) increased and C(2) decreased. With positive end-expiratory pressure, the L parameter decreased and no significant change in other model parameters was observed. Obstructed endotracheal tube is indicated if R(1) increased > or =30%, C(1) decreased > or =10%, and C(2) increased > or =10% from baseline. The test results using 45 events, including control, three complications, and positive end-expiratory pressure, show that when the model is used to diagnose obstructed endotracheal tube, the method has a sensitivity of 90% and specificity of 97%. Conclusions During an obstructed endotracheal tube, model parameters change such that the event can be diagnosed noninvasively, automatically, and accurately. The model differentiates between upper airway obstruction and complications like bronchospasm and stiff chest wall.

Objective: A functional total liquid ventilator should be simple in design to minimize operating errors and have a low priming volume to minimize the amount of perfluorocarbon needed. Closed system circuits using a membrane oxygenator have partially met these requirements but have high resistance to perfluorocarbon flow and high priming volume. To further this goal, a single piston prototype ventilator with a low priming volume and a new high-efficiency hollow-fiber oxygenator in a circuit with a check valve flow control system was developed. Design: Prospective, controlled animal laboratory study. Setting: Research facility at a university medical center. Subjects: Seven anesthetized, paralyzed, normal New Zealand rabbits Interventions: The prototype oxygenator, consisting of cross-wound silicone hollow fibers with a surface area of 1.5 m 2 with a priming volume of 190 mL, was tested in a bench-top model followed by an in vivo rabbit model. Total liquid ventilation was performed for 3 hrs with 20 mL.kg -1 initial fill volume, 17.5-20 mL.kg -1 tidal volume, respiratory rate of 5 breaths/min, inspiratory/expiratory ratio 1:2, and countercurrent sweep gas of 100% oxygen. Measurements and Main Results: Bench top experiments demonstrated 66-81% elimination of Co 2 and 0.64-0.76 mL.min -1 loss of perfluorocarbon across the fibers. No significant changes in Paco 2 and Pao 2 were observed. Dynamic airway pressures were in a safe range in which ventilator lung injury or airway closure was unlikely (3.6 ± 0.5 and -7.8 ± 0.3 cm H 2 O, respectively, for mean peak inspiratory pressure and mean end expiratory pressure). No leakage of perfluorocarbon was noted in the new silicone fiber gas exchange device. Estimated in vivo perfluorocarbon loss from the device was 1.2 mL.min -1 . Conclusions: These data demonstrate the ability of this novel single-piston, nonporous hollow silicone fiber oxygenator to adequately support gas exchange, allowing successful performance of total liquid ventilation.

OBJECTIVES To study changes in PaCO2 and PaO2 during intratracheal pulmonary ventilation (ITPV) and hybrid intratracheal pulmonary ventilation (h-ITPV) compared with conventional mechanical ventilation (CMV) in a rabbit model of respiratory failure, and to define the technique of h-ITPV that combines conventional mechanical ventilation and ITPV. DESIGN Prospective, interventional study. SUBJECTS Twelve adult New Zealand White rabbits. INTERVENTIONS Surfactant deficiency was induced by saline lavage, and rabbits were randomized to either ITPV or h-ITPV. The study consisted of four phases: phase 0, CMV after saline lavage, ventilator rate 30 breaths/min; phase I, ITPV or h-ITPV initiated at the same pressure and rate as in phase 0; phase II, ITPV or 1.0 L/min h-ITPV bias flow, with peak inspiratory pressure (PIP) decreased and ventilator rate increased to achieve the lowest tidal volume while maintaining adequate gas exchange; and phase III, animals returned to CMV. MEASUREMENTS AND MAIN RESULTS In phase I, no difference in PaCO2 was observed between ITPV, h-ITPV, or CMV. There was a decrease in PaO2 when switching from CMV to ITPV but not to h-ITPV. In phase II, it was possible to decrease PIP (average of 37% for ITPV and 36% for h-ITPV) and tidal volume (average of 64% for ITPV and 53% for h-ITPV) without compromising gas exchange (p < .05). Oxygenation tended to improve from phase 0 to the end of phase II. In phase III, PaCO2 increased (average of 71% for ITPV and 79% for h-ITPV) and pH decreased (p < .05). Normocapnia was achieved using significantly higher PIP and tidal volume, compared with phase 0 (p < .05). CONCLUSIONS ITPV and h-ITPV can effectively ventilate and oxygenate rabbits with surfactant-deficient lungs at tidal volumes and therefore pressures lower than required with CMV. Maximum benefit appears to occur at high ventilator rates. These findings suggest that both modes of ventilation may represent powerful new tools in the management of patients with acute respiratory failure. (Crit Care Med 2000; 28:774-781)

Objective: Perfluorocarbon liquid ventilation has been shown to enhance pulmonary mechanics and gas exchange in the setting of respiratory failure. To optimize the total liquid ventilation process, we developed a volume-limited, time-cycled liquid ventilatory support, consisting of an electrically actuated, microprocessor-controlled, double-cylinder, piston pump with two separate limbs for active inspiration and expiration. Design: Prospective, controlled, animal laboratory study, involving sequential application of conventional gas ventilation, partial ventilation (PLV), and total liquid ventilation (TLV). Setting: Research facility at a university medical center. Subjects: A total of 12 normal adult New Zealand rabbits weighing 3.25 ± 0.1 kg. Interventions: Anesthestized rabbits were supported with gas ventilation for 30 mins (respiratory rate, 20 cycles/min; peak inspiratory pressure, 15 cm H2O; end-expiratory pressure, 5 cm H2O), then PLV was established with perflubron (12 mL/kg). After 15 mins, TLV was instituted (tidal volume, 18 mL/kg; respiratory rate, 7 cycles/min; inspiratory/expiratory ratio, 1:2 cycles/min). After 4 hrs of TLV, PLV was re-established. Measurements and Main Results: Of 12 animals, nine survived the 4-hr TLV period. During TLV, mean values ± SEM were as follows: PaO2, 363 ± 30 torr; PaCO2, 39 ± 1.5 torr; pH, 7.39 ± 0.01; static peak inspiratory pressure, 13.2 ± 0.2 cm H2O; static end-expiratory pressure, 5.5 ± 0.1 cm H2O. No significant changes were observed. When compared with gas ventilation and PLV, significant increases occurred in mean arterial pressure (62.4 ± 3.5 torr vs. 74.0 ± 1.2 torr) and central venous pressure (5.6 ± 0.7 cm H2O vs. 7.8 ± 0.2 cm H2O) (p Conclusions: Total liquid ventilation can be performed successfully utilizing piston pumps with active expiration. Considering the enhanced flow profiles, this device configuration provides advantages over others.

To develop a predictable, dose-dependent, clinically relevant model of severe respiratory failure associated with a 40% total body surface area, full-thickness (third-degree) cutaneous flame burn and smoke inhalation injury in adult sheep.Model development.Research laboratory.Adult female sheep (n = 22).Animals were divided into three groups, determined by the number of smoke breaths administered (24, 36, 48) for a graded inhalation injury. The smoke was insufflated into a tracheostomy with a modified bee smoker at airway temperatures40 degrees C. All animals concurrently received a 40% total body surface area (third-degree) cutaneous flame burn to the body (flanks). After injury, the animals were placed on volume-controlled ventilation to achieve PaO260 mm Hg and PaCO240 mm Hg. Arterial blood gases and ventilator settings were monitored every 6 hrs postinjury for up to 7 days.All animals survived the induction of injury. In the 24 smoke breath/40% total body surface area burn (24/40) group, PaO2/F(IO2) never decreased below 300, and peak inspiratory pressure was consistently14 cm H2O with normal arterial blood gases throughout the observation period. With 36 smoke breaths/40% total body surface area burn (36/40) (n = 7), all animals had PaO2/F(IO2) of200 and peak inspiratory pressure of 26 cm H2O within 40-48 hrs, as 30% died during the study period. With 48 smoke breaths/40% total body surface area burn (48/40) (n = 12), all animals developed respiratory distress syndrome (RDS) in 24-30 hrs, but none survived the experimental period.Development of RDS by smoke and cutaneous flame bum injury depends on smoke inhalation dose. A combination of 36 breaths of smoke and a 40% total body surface area (third-degree) cutaneous flame burn injury can induce severe RDS (PaO2/F(IO2)200) within 40-48 hrs to allow evaluation of various treatment modalities of RDS.

Objective: High-frequency oscillation studies have shown that ventilation at high end-expiratory lung volumes combined with small volume cycles at high rates best preserves exogenous surfactant and gas exchange in lavaged lungs. We investigated whether surfactant composition and gas exchange can also be preserved by conventional modes of mechanical ventilation, which combine high levels of positive end-expiratory pressure (PEEP) with small pressure amplitudes. Design: Prospective, randomized, nonblinded, controlled study. Setting: Research laboratory. Subjects: Thirty male Sprague-Dawley rats. Interventions: Rats were lung-lavaged and treated with exogenous surfactant (100 mg/kg). After 5 mins, four different ventilator settings (Fio 2 = 1.0) were applied for 3 hrs in four groups of rats [peak inspiratory pressure (cm H 2 O); static PEEP (cm H 2 O); inspiratory/expiratory ratio; frequency], as follows: 26/2/1:2/30 (group 2612), 261611:2130 (group 26/6), 20/10/1:2/30 (group 20/ 10-static), and 20/6/7:3/130, creating an auto PEEP of 4 cm H 2 O (group 20/10-auto). Measurements and Main Results: In all groups, Pao 2 increased immediately to prelavage values after surfactant therapy. In group 26/2, Pao 2 deteriorated to postiavage values within 30 mins when PEEP was decreased to 2 cm H 2 O, whereas Pao 2 remained stable for 3 hrs in the other groups. The Paco 2 increased in groups 26/2 and 20/10-static; Paco 2 could not be reduced by increasing ventilation frequency to 130 in group 20/10-static. Groups 26/6 and 20/10-auto remained normocapnic. Bronchoalveolar lavage protein concentration was higher in groups 26/2 and 26/6 compared with groups 20/10-static and 20/10-auto. There was significantly more conversion of surface active large aggregates into nonactive small aggregates in group 2612 compared with groups 20/10-static and 20/10-auto. Conclusions: We conclude that exogenous surfactant composition is preserved by conventional modes of mechanical ventilation that use small pressure amplitudes, and adequate oxygen: ation is maintained by high end-expiratory pressure levels. Effective carbon dioxide removal can be achieved by applying a ventilation mode that creates auto PEEP and not by a mode that applies the same level of PEEP by static PEEP only.

Objective To test the hypothesis that inhaled nitric oxide may be an effective therapeutic agent in meconium aspiration syndrome and may improve oxygenation after pretreatment with surfactant. Design Prospective, interventional study. Setting The animal research laboratory at The Children's National Medical Center. Subjects Eight newborn pigs, 1 to 2 wks of age, 4.1 +/- 0.4 kg, were used for the study. Interventions Animals were anesthetized, paralyzed, intubated, and mechanically ventilated. Catheters were placed in the femoral vein and artery, and in the pulmonary artery. After 1 hr of recovery, 10 mL/kg of 20% meconium in normal saline solution was insufflated into the lungs. Animals were ventilated to maintain arterial blood gases in a normal range (i.e., pH of 7.35 to 7.45, Paco2 of 40 to 45 torr [5.3 to 6.0 kPa], and Pao2 of 70 to 90 torr [9.3 to 12.0 kPa]). Ventilatory settings were increased, as needed, until the following maximum settings were reached: FIO2 of 1.0; peak inspiratory pressure of 40 cm H2O; and intermittent mandatory ventilation of 60 breaths/min. After 2 hrs of conventional ventilation or demonstration of clinically important lung disease by failure to maintain desired blood gases on the maximum ventilatory settings, 4 mL/kg of beractant was given intratracheally. After a short period of stabilization following surfactant therapy, inhaled nitric oxide was administered. Concentrations of 40, 20, and 10 parts per million were given. To assure that there was no additive effect of inhaled nitric oxide, each dose was given for 20 mins, followed by a 15-min normalization period at 0 parts per million. Measurements and main results Physiologic measurements, ventilatory settings, arterial blood gases, and methemoglobin were recorded at each study period. Measurements were taken after each exposure to inhaled nitric oxide and after its discontinuation. Arterial oxygen saturation and Pao2 were significantly lower after meconium aspiration when compared with baseline values. After treatment with surfactant, administration of inhaled nitric oxide improved oxygenation without a significant decrease in pulmonary arterial pressure. Conclusions In this model of meconium aspiration syndrome, short-term exposure to inhaled nitric oxide after treatment with surfactant improved oxygenation secondary to better distribution of inhaled nitric oxide. The increase in oxygenation may be secondary to an improved ventilation/perfusion mismatch, since the primary etiology of hypoxia in this model may be a combination of parenchymal lung disease and pulmonary hypertension.

OBJECTIVE To evaluate the effects of a double-lumen, two-stage endotracheal tube on gas exchanges (ventilatory efficiency) during conventional mechanical ventilation, using a ventilator in rabbits. DESIGN Prospective, randomized, crossover laboratory animal trial. SETTING Research laboratory in the Beijing Children's Hospital. SUBJECTS Five adult New Zealand rabbits, weighing 3.2 to 3.7 kg. INTERVENTIONS A new type of endotracheal tube-the double-lumen two-stage endotracheal tube-was designed and tested for ventilation efficiency in rabbits with normal and injured lungs. The new tube (size 3.0 mm) was made out of two Portex endotracheal tubes (size 3.0 mm) by adhering two vertical cross-sections at the distal end of the two tubes. The new tube and a conventional endotracheal tube of the same size (inner diameter 3.0 mm) were randomly used in pressure control ventilation. Each trial was maintained for 30 mins. MEASUREMENTS AND MAIN RESULTS Effects from the two endotracheal tubes on pulmonary mechanics, hemodynamics, and gas exchange were observed. We measured peak inspiratory pressure, positive end-expiratory pressure (PEEP), intrinsic PEEP, mean airway pressure, and arterial blood and mixed expired gas variables (Pao2, Paco2, pH, and mixed expired gas Pco2). The new endotracheal tube acutely increased CO2 elimination in all animals with normal and injured lungs. Paco2 decreased from 46 +/- 4 to 36 +/- 5 torr (6.1 +/- 0.5 to 4.8 +/- 0.7 kPa; p < .01) in normal lungs and from 48 +/- 5 to 36 +/- 5 (6.4 +/- 0.7 to 4.8 +/- 0.7 kPa; p < .01). Meanwhile, tidal volume fraction decreased from 0.48 +/- 0.07 to 0.35 +/- 0.05 (p < .01) and from 0.56 +/- 0.07 to 0.40 +/- 0.07 (p < .01) in normal and injured lungs, respectively. Intrinsic PEEP of the new tube slightly increased, but there were no significant differences in comparison with the conventional tube. CONCLUSION Compared with the conventional endotracheal tube, the new double-lumen, two-stage endotracheal tube reduced Paco2 by decreasing anatomical deadspace in rabbits with normal and injured lungs under pressure control ventilation, thus enhancing ventilatory efficiency and reducing ventilator-induced injury.

Objective To investigate the relationship between the period of mechanical ventilation before extracorporeal life support and survival in patients with respiratory failure. Design Retrospective review. Setting Surgical intensive care unit at a university medical center. Patients Thirty-six consecutive adult patients with severe respiratory failure managed with extracorporeal life support. Interventions Extracorporeal life support was utilized in 36 acute respiratory failure adult patients with a variety of diagnoses and an estimated mortality rate of > 90%. Management protocols were followed before and during extracorporeal life support. The 36 patients were physiologically similar before extracorporeal life support was initiated: shunt of 48 +/- 17%; F10(2) of 1.0 +/- 0.1; peak inspiratory pressure of 56 +/- 16 cm H2O; positive end-expiratory pressure of 14 +/- 6 cm H2O; and respiratory rate of 23 +/- 10 breaths/ min. Ventilation was utilized for 1 to 17 days before extracorporeal life support. Typical lung rest settings during extracorporeal life support were F10(2) of 0.40, peak inspiratory pressure of 30 cm H2O, positive end-expiratory pressure of 10 cm H2O, and respiratory rate of 6 breaths/min. Death was almost always secondary to end-stage pulmonary failure. Measurements and main results Survival (hospital discharge) in these 36 patients was inversely associated with the number of days of preextracorporeal life support ventilation, with a 50% mortality rate predicted by logistic regression after 5 days of mechanical ventilation. The overall survival rate was 18 (50.0%) of 36 patients. Conclusions In severe acute respiratory failure treated with lung rest and extracorporeal life support, a predicted 50% mortality rate was associated with 5 days of preextracorporeal life support mechanical ventilation.

Objective : To evaluate a portable scavenging system for nitric oxide and its oxides, designed for use on neonatal transport. Design : A prospective evaluation of the nitric oxide scavenging system, using a neonatal transport incubator ventilator and a test lung. Setting : Laboratory of a tertiary care children's hospital. Interventions : The scavenging system was tested, using a neonatal transport Incubator with attached ventilator, ventilator circuit, and a neonatal test lung. Nitric oxide was administrated on the inspiratory limb, and nitric oxide and its oxides were measured in the expiratory gas after passing through the scavenger. Measurements and Main Results : A modified scrubber assembly was filled with 50% activated charcoal and 50% aluminas potassium permanganate pellets (3). Three wire meshes were placed before, in between, and after the two chemicals to facilitate gas flow. Using the maximum FIO 2 , with a nitric oxide concentration of 120 parts per million (ppm), the test lung continuous flow ventilation (FIO 2 of 0.86, peak inspiratory pressure of 30 cm H 2 O, positive end-expiratory pressure of 6 cm H 2 O, and respiratory rate of 60 breaths/min) was performed for 4 hrs with each of four freshly prepared scavenging systems. A fifth scavenging system was tested for a 12-hr period. The mean composition of the exhaled gases for 4 hrs were : nitric oxide 0.01 ± 0.03 (SD) ppm, nitric dioxide 0.06 ± 0.06 ppm, and other oxides 0.05 ± 0.09 ppm. After 12 hrs of 120 ppm of inhaled nitric oxide, the fifth scavenger system had undetectable nitric oxide, nitric dioxide, and other oxides in the exhaled gas. Normal room air contained between 0.0 and 0.03 ppm of nitric oxide, 0.0 and 0.02 ppm of nitric dioxide, and 0.0 and 0.02 ppm of other oxides. Conclusion : Nitric oxide, nitric dioxide, and other dioxides can be safely scavenged by this portable scavenging system, allowing safe administration of nitric oxide free from environmental contamination with nitric oxide and its oxides.

OBJECTIVE Perfluorocarbon liquid ventilation has been shown to have advantages over conventional gas ventilation in premature newborn and lung-injured animals. To simplify the process of liquid ventilation, we adapted an extra-corporeal life-support circuit as a time-cycled, volume-limited liquid ventilator. DESIGN Laboratory study that involved sequential application of gas and liquid ventilation in normal cats and in lung-injured sheep. SETTING A research laboratory at a university medical center. SUBJECTS Eight normal cats weighing 2.7 to 3.8 kg (mean 3.1 +/- 0.5), and four lung-injured young sheep weighing 10.4 to 22.5 kg (mean 15.9 +/- 5.0). INTERVENTIONS Normal cats were supported with traditional gas ventilation for 1 hr (respiratory rate 20 breaths/min, peak inspiratory pressure 12 cm H2O, positive end-expiratory pressure 4 cm H2O, and FIO2 1.0). The lungs were then filled with perfluorocarbon (30 mL/kg) and tidal volume liquid ventilation was instituted, utilizing a newly developed liquid ventilation device. Liquid ventilatory settings were 4 secs for inspiration time, 8 secs for expiration time, 5 breaths/min for respiratory rate, and 15 to 20 mL/kg for tidal volume. Liquid ventilation utilizing this device was also applied to sheep after induction of severe lung injury by right atrial injection of 0.07 mL/kg of oleic acid, followed by saline pulmonary lavage. Extracorporeal life support was instituted to provide a stable model of lung injury. For the first 30 mins of extracorporeal support, all animals were ventilated with gas. Animals were then ventilated with 15 mL/kg of perfluorocarbon over the ensuing 2.5 hrs. MEASUREMENTS AND MAIN RESULTS In normal cats, mean PaO2 values after 1 hr of liquid or gas ventilation were 275 +/- 90 (SD) torr (36.7 +/- 10.4 kPa) in the liquid-ventilated animals and 332 +/- 78 torr (44.3 +/- 10.4 kPa) in the gas-ventilated animals (NS). Mean PaCO2 values were 40.5 +/- 5.7 torr (5.39 +/- 0.31 kPa) in the liquid-ventilated animals and 37.6 +/- 2.3 torr (5.01 +/- 0.31 kPa) in the gas-ventilated animals (NS). Mean arterial pH values were 7.35 +/- 0.07 in the liquid-ventilated animals and 7.34 +/- 0.04 in the gas-ventilated animals (NS). No significant changes in heart rate, mean arterial pressure, lung compliance, or right atrial venous oxygen saturation were observed during liquid ventilation when compared with gas ventilation. In the lung-injured sheep, an increase in physiologic shunt from 15 +/- 7% to 66 +/- 9% was observed with induction of lung injury during gas ventilation. Liquid ventilation resulted in a significant reduction in physiologic shunt to 31 +/- 10% (p < .001). In addition, the extracorporeal blood flow rate required to maintain the PaO2 in the 50 to 80 torr (6.7 to 10.7 kPa) range was substantially and significantly (p < .001) lower during liquid ventilation than during gas ventilation (liquid ventilation 15 +/- 5 vs. gas ventilation 87 +/- 15 mL/min/kg). CONCLUSIONS Liquid ventilation can be performed successfully utilizing this simple adaptation of an extracorporeal life-support circuit. This modification to an existing extracorporeal circuit may allow other centers to apply this new investigational method of ventilation in the laboratory or clinical setting.

To evaluate the outcome in patients with severe adult respiratory distress syndrome (ARDS) managed with limitation of peak inspiratory pressure to 30 to 40 cm H2O, low tidal volumes (4 to 7 mL/kg), spontaneous breathing using synchronized intermittent mandatory ventilation from the start of ventilation, and permissive hypercapnia without the use of bicarbonate to buffer acidosis. Also, to compare hospital mortality rate with that predicted by the Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system and the "ventilator score."A ten-bed general intensive care unit in a university hospital.Prospective, descriptive study.Fifty-three patients with severe ARDS having a lung injury score ofor = 2.5.Data recording.The hospital mortality rate was significantly lower than that predicted by the APACHE II scores (26.4% vs. 53.3%, p = .004), even after correcting the latter for the effect of hypercapnic acidosis (26.4% vs. 51.1%, p = .008). The mortality rate increased with increasing number of organ failures, but was only 43% in patients withor = 4 organ failures, 20.5% withor = 3 organ failures, and 6.6% with only respiratory failure. The mean maximum PaCO2 was 66.5 torr (range 38 to 158 torr [8.87 kPa, range 5.07 to 21.07]), and the mean arterial pH at the same time was 7.23 (range 6.79 to 7.45). There was no correlation between the maximum PaCO2 or the corresponding pH and the total respiratory rate at the same time. No pneumothoraces developed during mechanical ventilation.These results lend further support to the hypothesis that limitation of peak inspiratory pressure and reduction of regional lung overdistention by the use of low tidal volumes with permissive hypercapnia may reduce ventilator-induced lung injury and improve outcome in severe ARDS. This hypothesis is supported by a large body of experimental evidence, which also suggests that ventilator-induced lung injury may result in the release of inflammatory mediators, and thus may have the potential to augment the development of multiple organ dysfunction. However, the hypothesis requires testing in a randomized trial as acute hypercapnia could potentially have some adverse as well as beneficial effects.

Objective The purpose of this report was to examine the Extracorporeal Life Support Organization registry database for predictors of outcome for severe pediatric respiratory failure managed with extracorporeal life support. Design Retrospective cohort study. Setting Extracorporeal Life Support Organization data registry. Patients All nonneonatal pediatric patients who were treated in the United States with extracorporeal life support for severe pediatric respiratory failure reported to the Extracorporeal Life Support Organization registry as of August 1991. Patients with congenital heart disease and congenital gastrointestinal malformations were excluded from study. Interventions Venoarterial or venovenous extracorporeal life support for severe life-threatening pulmonary failures. Measurements and main results As of August 1991, 220 pediatric patients meeting study entry criteria were reported to the Registry having received extracorporeal life support for severe pulmonary failure, since 1982. Forty-six percent (102 of 220 patients) were successfully managed with this technology and survived to hospital discharge. The mean patient age was 36.8 +/- 51.6 months. Fifty-one percent of the patients were male. The mean duration of mechanical ventilation before extracorporeal life support was 6.3 +/- 5.9 days. Mean blood gas and ventilatory measurements obtained before extracorporeal life support were as follows: PaCO2 52 +/- 23 torr (6.9 +/- 3.0 kPa); PaO2 59 +/- 32 torr (7.8 +/- 4.3 kPa); estimated alveolar-arterial oxygen gradient 561 +/- 63.4 torr (74.8 +/- 8.5 kPa); peak airway pressure 49.5 +/- 13.1 cm H2O; mean airway pressure 24.3 +/- 8.2 cm H2O; positive end-expiratory pressure 11.8 +/- 6.3 cm H2O; ventilator rate 58 +/- 64.4 breaths/min; and FIO2 0.98 +/- 0.07. The mean duration of extracorporeal life support for all patients was 247 +/- 164 hrs. For the 102 survivors, the mean time for decannulation from extracorporeal life support to extubation from mechanical ventilation was 6.5 +/- 7.6 days. Stepwise multivariate logistic regression modeling found the following variables to be associated with patient survival: a) patient age, b) days of mechanical ventilation before extracorporeal life support, c) peak inspiratory pressure, d) alveolar-arterial oxygen gradient, and e) extracorporeal life support administered since December 31, 1988 (all p Conclusions Extracorporeal life support may represent an effective rescue therapy for some selected pediatric patients with severe respiratory failure for whom conventional mechanical ventilation support has failed to improve. Predictors of survival for this life-support therapy exist that may be helpful for individual patient prognostication and future prospective study.

OBJECTIVE To assess whether communication capabilities of ventilator-dependent patients are improved by the use of the Passy-Muir unidirectional valve. DESIGN Prospective study. SETTING An 18-bed multidisciplinary intensive care unit (ICU) at the University Hospital, Las Palmas, Spain. PATIENTS Ten chronic ventilator-dependent patients who had undergone tracheostomy and met the following criteria: ability to eliminate tracheobronchial secretions in order to maintain a patent and unobstructed airway, adequate gas exchange while ventilated with an FIO2 of 60 torr [8 kPa]), Paco2 of < 55 torr (7.3 kPa), normal hemodynamics without the need for administration of vasopressors, and normal mental state. Eight patients presented with pulmonary disease, and two presented with neuromuscular disease. INTERVENTIONS AND METHODS Before attaching the Passy-Muir valve, the following procedures were performed: a) suction of tracheal and pharyngeal secretions; b) deflation of the tracheostomy tube cuff; c) increase in the ventilator's tidal volume setting to maintain the inspiratory pressure before cuff's deflation; d) set peak inspiratory pressure alarm and disconnect expiratory volume alarm. The valve was then connected between the tracheostomy tube and the Y-shaped piece of the ventilator's circuit. Respiratory movements, arterial blood gases, peak inspiratory pressure, respiratory rate, quantity of secretions, and changes in sense of smell were monitored during the study. The valve's efficacy was evaluated according to the patient's ability to talk and be understood during the entire respiratory cycle. RESULTS The Passy-Muir valve was effective in improving communication in eight of ten patients who, during its use, presented insignificant cardiorespiratory changes, decreased secretions, and effected considerable improvement in well-being. Its use was impossible in two patients: one with severe pulmonary disease because cuff deflation prevented adequate ventilation, and one patient with neuromuscular disease and laryngopharyngeal dysfunction. CONCLUSIONS The Passy-Muir unidirectional valve allows ventilator-dependent patients to talk and communicate without assistance. Patients felt better and were motivated to participate in their own care.

OBJECTIVES To study the individual and combined effects of surfactant inactivation and mechanical ventilation on pulmonary microvascular permeability and lung compliance. DESIGN Prospective, controlled trial. An isolated, perfused, lung model of surfactant inactivation and mechanical ventilation at 15, 30, and 45 cm H2O peak inspiratory pressure was developed in young (4 to 6 wks) New Zealand white rabbits. SETTING Laboratory of a university-affiliated medical school. MEASUREMENTS AND MAIN RESULTS Isolated, perfused lungs were prepared for measurement of the capillary filtration coefficient before and after one of four interventions: instillation of dioctyl succinate, a surfactant inactivator, without ventilation (group 1); ventilation without dioctyl succinate at 15, 30, or 45 cm H2O peak inspiratory pressure (group 2); ventilation after dioctyl succinate pretreatment at 15, 30, or 45 cm H2O peak inspiratory pressure (group 3); and control lungs without dioctyl succinate or ventilation (group 4). A significant increase in the capillary filtration coefficient was noted after dioctyl succinate treatment alone, after ventilation alone at 45 cm H2O peak inspiratory pressure, and after dioctyl succinate plus ventilation at 15, 30, and 45 cm H2O peak inspiratory pressure. Dioctyl succinate plus ventilation produced a significantly greater increase in the capillary filtration coefficient than ventilation alone at 15 and 45 cm H2O peak inspiratory pressure. CONCLUSIONS These data suggest that ventilation after surfactant inactivation is more injurious to the pulmonary microvasculature than ventilation alone, and that generalized lung overdistention is not the primary mechanism for microvascular injury in the diseased, noncompliant lung. The increases seen in the capillary filtration coefficient in postventilated surfactant inactivated lungs, even at low-ventilation pressures, suggest that low peak inspiratory pressures do not overdistend the dioctyl succinate-treated lung.

BACKGROUND AND METHODS A new mode of jet ventilation, high-frequency two-way jet ventilation, was devised and introduced to increase CO2 elimination. High-frequency two-way jet ventilation was achieved by adding reverse jet pulses inside the trachea through an intratracheal reverse jet system to the expiratory phase of common high-frequency jet ventilation. The ventilatory efficiency and features of high-frequency two-way jet ventilation were investigated and compared with those features of high-frequency jet ventilation in ten dogs in the same experimental condition. Random sample selection and randomized crossover trial were used for comparison between high-frequency two-way jet ventilation and high-frequency jet ventilation. Peak inspiratory pressure, end-expiratory pressure, and the arterial blood gas variables (PaO2, PaCO2, and pH) were measured during the study. RESULTS PaCO2 with high-frequency two-way jet ventilation was about 35% lower than that with high-frequency jet ventilation (from 45 to 29 torr [6.0 to 3.9 kPa], p less than .01). Simultaneously, peak inspiratory pressure and end-expiratory pressure during high-frequency two-way jet ventilation were significantly lower than those same variables measured during high-frequency jet ventilation. End-expiratory pressure of high-frequency two-way jet ventilation was a negative pressure (-2.45 +/- 0.45 cm H2O). The pH of high-frequency two-way jet ventilation was significantly higher than that of high-frequency jet ventilation. CONCLUSIONS Compared with high-frequency jet ventilation, high-frequency two-way jet ventilation demonstrated a ventilatory feature of increasing CO2 elimination and simultaneously decreasing airway pressure.

In order to minimize barotrauma in newborn infants with respiratory failure, peak inspiratory pressures should not exceed those required for adequate gas exchange. We examined whether four commonly used pressure-limited, constant flow ventilators limit pressure reliably during simulated active expiration against the inspiratory stroke of the ventilator. Three machines of each type were tested at 13 different expiratory flow rates (2 to 14 L/min). Flow-dependent pressure overshoot above a dialed pressure limit of 20 cm H2O was observed in all machines. However, the magnitude differed significantly between ventilators from different manufacturers (p = .0009). Pressure overshoot above 20 cm H2O was consistently lowest in the Healthdyne (0.8 cm H2O at 2 L/min, 3.6 cm H2O at 14 L/min) and highest in the Bourns BP200 (3.0 cm H2O at 2 L/min, 15.4 cm H2O at 14 L/min). We conclude that peak inspiratory pressure overshoots on pressure-limited ventilators occur during asynchronous expiration. This shortcoming may contribute to barotrauma in newborn infants who "fight" positive-pressure ventilation.

High-frequency jet ventilation (HFJV) was compared with conventional ventilation (CV) after surfactant displacement with diocytyl sodium sulfosuccinate (OT) via ultrasonic nebulization. After aerosol delivery, dogs were separated into 3 groups and followed for 2 h. Ventilator settings were not changed in group I (CV) and group II (HFJV) after OT delivery. In group III, drive pressure was increased to 40 psi. Adequacy of oxygenation varied directly with peak inspiratory pressure (PIP) rather than airway pressure (Paw) in both HFJV and CV. In all groups, immediately after OT administration PaO2 decreased and there was a slight increase in Paw; PIP was significantly elevated only in groups I and III. Two hours later, PaO2 had returned to baseline in groups I and III, but had not improved significantly in group II. Paw remained the same in all groups. These data demonstrate that in noncompliant lungs, oxygenation is not improved unless a high PIP is used to establish the critical opening pressures needed to rerecruit alveoli. In this noncompliant lung model, HFJV was not effective at low Paw values, and thus offered no apparent advantage over CV.

Clinical features and assisted ventilation (AV) aspects of 10 neonates with early onset group B streptococcus (GBS) septicemia were compared with those of 12 randomly selected newborns with severe respiratory distress syndrome (RDS). Initial chest radiograph in the GBS group was interpreted as RDS in seven of 10 cases. Although 9 of the 10 neonates with early onset GBS were preterm, they were of a significantly higher mean birth weight (p < 0.01) and the mean gestational age (p < 0.005). The duration of rupture of fetal membranes was not significantly different between the two groups. Contrary to neonates with severe RDS, who gradually developed hypoxia necessitating AV, the neonates with early onset GBS required AV because of persistent apnea, shock, and respiratory insufficiency. During the first 24 h of life, neonates with early onset GBS were more acidotic, had significantly higher PaCO2 and lower PaO2 values, and required significantly higher peak inspiratory pressure (PIP) on the respirator (po < 0.01) while on AV, although the oxygen requirement was similar.

We successfully employed inverse ratio ventilation on a 6-year-old multiple trauma victim with severe adult respiratory distress syndrome after conventional ventilation modes using volume ventilation with high positive peak inspiratory pressure and PEEP had failed to improve oxygenation.