Abstract
To determine if decreased respiratory frequency (ventilatory rate) improves indices of lung damage, 17 sets of isolated, perfused rabbit lungs were ventilated with a peak static airway pressure of 30 cm H2O. All lungs were randomized to one of three frequency/peak pulmonary artery pressure combinations: F20P35 (n = 6): ventilatory frequency, 20 breaths/min, and peak pulmonary artery pressure, 35 mm Hg; F3P35 (n = 6), ventilatory frequency, 3 breaths/min, and peak pulmonary artery pressure of 35 mm Hg; or F20P20 (n = 5), ventilatory frequency, 20 breaths/min, and peak pulmonary artery pressure, 20 mm Hg. Mean airway pressure and tidal volume were matched between groups. Mean pulmonary artery pressure and vascular flow were matched between groups F20P35 and F3P35. The F20P35 group showed at least a 4.5-fold greater mean weight gain and a 3-fold greater mean incidence of perivascular hemorrhage than did the comparison groups, all p ⩽ 0.05. F20P35 lungs also displayed more alveolar hemorrhage than did F20P20 lungs (p ⩽ 0.05). We conclude that decreasing respiratory frequency can improve these indices of lung damage, and that limitation of peak pulmonary artery pressure and flow may diminish lung damage for a given ventilatory pattern. Hotchkiss JR, Jr., Blanch L, Murias G, Adams AB, Olson DA, Wangensteen OD, Leo PH, Marini JJ. Effects of decreased respiratory frequency on ventilator-induced lung injury.
Evidence has accumulated that mechanical ventilation with high alveolar pressures, associated with elevated end-inspiratory lung volumes, can cause or worsen lung injury (1-5). Studies by West and colleagues (6-9) and by Bachofen and coworkers (10-12) have demonstrated that elevated intramural capillary pressures can disrupt the blood:air-space barrier, even causing overt capillary rupture, a phenomenon denoted “stress failure.” Guery and colleagues (13) have highlighted the interaction between vascular dynamics, ventilation, and increased pulmonary permeability. Our recent work has indicated that increasing pulmonary blood flow in the setting of ventilation with high transpulmonary pressures can worsen lung injury. Elevation of pulmonary vascular pressure caused by interactions between lung volume and vascular resistance appeared to best explain our results (14-19). The current study was designed to answer a specific question: at matched peak and mean pulmonary artery pressures and matched airway pressures (inspiratory plateau, end-expiratory pressure, and mean airway pressure) does a decrease in ventilatory frequency (respiratory rate) lead to decreased edema formation or histologic alterations? To answer this question three groups of isolated, perfused, ventilated rabbit lungs were studied (Table 1).
| Set Parameter | F20P35 | F3035 | F20P20 | |||
|---|---|---|---|---|---|---|
| Ventilatory frequency, breaths/min | 20 | 3 | 20 | |||
| Peak pulmonary artery pressure, mm Hg | 35 | 35 | 20 | |||
| Airway plateau pressure, cm H2O | 30 | 30 | 30 | |||
| Applied PEEP, cm H2O | 5 | 5 | 5 | |||
| Total time at high lung volume | ||||||
| (minutes at inspiratory pressure) | 15 | 15 | 15 | |||
| Left atrial pressure, mm Hg | 10 | 10 | 10 |
All procedures and techniques were approved by the Animal Care and Use Committee of Regions Hospital, (St. Paul, MN). New Zealand white rabbits weighing 3.3 ± 0.07 kg were anesthetized with an intramuscular injection of a ketamine/acepromazine solution (10 ml of ketamine [100 mg/ml] + 2 ml of acepromazine [10 mg/ml]; Vedco, St. Joseph, MO), followed by 12.0 mg/kg pentobarbital (Abbott Labs, North Chicago, IL) intravenously. Pentobarbital was then supplemented as needed to obtain surgical anesthesia. An endotracheal tube (5.0 mm internal diameter) was inserted via a tracheostomy, and an ear vein was cannulated for heparin administration (700 U/kg).
A midline sternotomy was performed, measurement of ischemic time was begun, and the animal was exsanguinated. The heart-lung block was dissected free, gently ventilated with an Ambu bag, and weighed. The heart and endotracheal tube were also weighed at the end of the experiment to calculate initial lung weight. Perfusion cannulae (7 mm internal diameter) were placed in the pulmonary artery and left atrium; each cannula housed a pressure tap (PE 90). Perfusion was begun at a constant flow of 50 ml/min. The time from the beginning of exsanguination to the beginning of perfusion, including the periods of manual ventilation, was noted. The preparation was suspended from a counterbalanced force transducer (FT03; Grass Instruments, Quincy, MA) to allow continuous monitoring of preparation weight.
The perfusate in the circuit was 500 ml of Krebs-Henseleit solution, to which 5% albumin and 1.43 g HEPES buffer had been added; pH was adjusted to 7.4, and 50 ml of autologous blood were added to serve as an erythrocyte biomarker for histologic analysis; 10 mg of ketorolac tromethamine (Abbott Labs) were added to ameliorate thromboxane-mediated pulmonary hypertension secondary to ischemia and reperfusion. The perfusion circuit and monitoring equipment were modified from that used in our previous work (14) by the addition of an open-ended segment of tubing in parallel with the pulmonary artery cannula (pulmonary artery pressure vent). The peak pulmonary artery pressure attained during inspiration could be set by adjusting the height of the open-ended segment of the pressure vent above the lung. Clamping the pulmonary artery pressure vent converted the system to a constant flow circuit. Left atrial pressure could be controlled by adjusting the height of the left atrial outflow vent above the lung.
Perfusion was discontinued and the lungs of all groups were recruited with continuous positive airway pressure (CPAP) of 25 cm H2O until grossly free of atelectasis (approximately 30 to 40 s). CPAP was decreased to 5 cm H2O, and the endotracheal tube connection was clamped (to prevent loss of recruited volume). The preparation was connected to a ventilator (PB335; Puritan-Bennett, Corp., Lenexa, KS), the connector unclamped, and baseline ventilation begun with pressure control ventilation (PCV) at inspiratory pressure of 15 cm H2O, positive end-expiratory pressure (PEEP) of 5 cm H2O, frequency of 20 breaths/min, inspiratory time fraction of 0.5, and Fi O2 of 0.21. Perfusion was resumed, and the free extremity of the left atrial cannula was positioned so the left atrial pressure was 10 mm Hg. Vascular flow was slowly increased from 50 to 100 ml/min over 5 min; left atrial pressure was maintained at 10 mm Hg. The pulmonary artery pressure vent was kept closed during this portion of the experiment. Ventilation was continued for 10 min, after which the lungs were returned to CPAP 5 cm H2O and allowed to equilibrate.
After perfusate temperature, vascular pressures, and weight had stabilized, pulmonary capillary pressures were determined by simultaneous double occlusion, and the force transducer was calibrated by the addition of a 1 g weight to the preparation. Flow was resumed at 100 ml/min, weight was allowed to stabilize, and the free end of the atrial cannula rapidly raised by 3 cm. Preparation weight was continuously recorded for 7 min. Capillary pressure was again determined by double occlusion, and left atrial pressure returned to 10 mm Hg. Ultrafiltration coefficients (at baseline and at the end of the experimental period) were determined from the weight gain slope between 5 and 7 min after the step increase in left atrial pressure and the difference between prestep and poststep double occlusion pressures. A later time interval (5 to 7 min) was used than in our previous studies because in pilot studies this modification led to greater reproducibility of the Kf measurement. When we were unable to obtain an isogravimetric preparation (e.g., severely injured lungs after injurious ventilation), the preparation was observed until the rate of weight gain was constant, at which time the left atrial pressure was increased. The slope of the 2 min preceding the step increase was then subtracted from the slope recorded from Minutes 5 to 7. Perfusion was interrupted, and a glass syringe was used to recruit the lungs to a transpulmonary pressure of ∼ 25 cm H2O. Inspiratory and expiratory pressure-volume curves were obtained, using volume steps of 10 ml to a maximum transpulmonary pressure of ∼ 30 cm H2O and a minimum transpulmonary pressure (deflation limb) of ∼ 2 cm H2O. The results were plotted for each preparation, and the average compliance over the linear segment of the inflation pressure-volume curve was determined.
The lungs were recruited with CPAP 25 cm H2O, ventilated at baseline settings for 5 min as vascular flow was increased from 50 to 100 ml/min, and returned to CPAP 5 cm H2O. The force transducer was recalibrated with a 20-g weight. Vascular flow was slowly increased to 450 ml/min while maintaining left atrial pressure at 10 mm Hg.
When stable at a perfusion rate of 450 ml/min, preparations were randomized to one of three groups: F20P35 (n = 6), F3P35 (n = 6), or F20P20 (n = 5) by withdrawing a group assignment from a sealed envelope. Lungs in the F20P35 and F20P20 group were ventilated at a frequency of 20 breaths/min and an inspiratory time fraction of 0.5. Lungs in the F3P35 group were ventilated at a frequency of three breaths/min and an inspiratory time fraction of 0.5. All groups were ventilated for 30 min with pressure-control ventilation at an inspiratory pressure of 30 cm H2O, PEEP 5 cm H2O, and Fi O2 of 0.21. Initial expiratory tidal volumes (measured between Minute 2 and Minute 5 of ventilation) were determined from the integrated flow signal of a Fleisch type pneumotachometer placed in series with the tracheostomy tube.
Increasing pulmonary vascular resistance because of lung inflation caused cyclic elevation of pulmonary artery pressure during the inspiratory phase of each breath. In lungs randomized to groups F20P35 and F3P35, the peak inspiratory pulmonary artery pressure thus attained was set at 35 mm Hg. Peak inspiratory pulmonary artery pressures in group F20P20 were limited to 20 mm Hg. The set pump flow was decreased to 200 ml/min in lungs randomized to group F20P20 (to allow consistent limitation of peak inspiratory vascular pressure to 20 mm Hg). Left atrial pressure was adjusted to 10 mm Hg in each group. The pulmonary artery pressure vent was unclamped and ventilation begun. The pulmonary artery pressure vent was then raised to attain the appropriate peak inspiratory pulmonary artery pressure for the assigned group. The position of the pulmonary artery pressure vent was adjusted during the course of the experiment to maintain the desired peak pulmonary artery pressure. Vascular flow through the lung was determined as the difference between set pump flow and pulmonary artery pressure vent flow, as measured by timed collection (the pump was calibrated before each experiment to ensure the accuracy of its flow settings). Weight gain during the experimental period was defined as the weight gain at 30 min minus the weight gain after 1 min of ventilation, allowing for the time required to adjust the pulmonary vascular pressures appropriately.
At the end of the ventilatory period, CPAP 5 cm H2O was resumed. An aliquot of perfusate was obtained for measurement of hemoglobin concentration and immediate postventilation pH and Po 2. The pressure vent was clamped, flow was returned to 100 ml/min, and left atrial pressure was set to 10 mm Hg. The preparation was allowed to stabilize. Capillary pressures were measured by double occlusion, the ultrafiltration coefficient was determined, and air-space pressure-volume curves were obtained. The left hilus was isolated and ligated, and the right lung was placed on CPAP of 25 cm H2O, fixed with 40 ml of 10% buffered formalin injected into the airways, and immersed in the same solution. At the conclusion of the study, one investigator (J.R.H.) was formally blinded to the group assignments of the lungs. Samples (3 mm coronal sections) were systematically taken perpendicular to the cranial–caudal axis from the upper, middle, and lower lung at the measured mid-distance from the cranial and caudal boundaries of each region. These sections included the entire dorsal, ventral, medial, and lateral borders at the cranial:caudal section level.
The three histologic samples obtained from each animal were examined by a pathologist (D.O.) blinded to both treatment group and the experimental protocol. The primary histologic outcomes assessed were the number of extra-alveolar vessels in each section with overt perivascular hemorrhage and the fraction of alveoli within the most damaged regions which contained erythrocytes. First, each section was scanned in its entirety at low power (magnification ×25) and the number of blood vessels with perivascular hemorrhage was counted. When observed, perivascular hemorrhage was confirmed at high power (magnification ×100 to ×200). For each lung, the number of vessels with perivascular hemorrhage in each of the three sections was added to obtain the total number of vessels with perivascular hemorrhage. The region of each section with the greatest extent of alveolar hemorrhage was then located using low power, and the fraction of the alveoli within that region containing any erythrocytes was determined under high power (magnification ×200 to ×400). These fractions were averaged to determine the alveolar hemorrhage for each lung.
All values are reported as mean ± SEM. Between-group comparisons for baseline values, alveolar and perivascular hemorrhage, weight gain, and percent changes in compliance and ultrafiltration coefficient were performed using one-way analysis of variance (ANOVA). Absolute changes in compliance and ultrafiltration coefficient were compared using repeated measures ANOVA. Scheffe's test, appropriate for unequal cell sizes, was performed for multiple comparisons; a corrected alpha of < 0.05 was considered statistically significant (20). After 17 rabbits had been randomized, an interim analysis indicated that further rabbits were not needed in the low pressure group, and the study was terminated.
The initial characteristics of each group, and the baseline ventilatory values at the beginning of injurious ventilation are shown in Table 2. Ischemic time averaged 24 ± 1 min, and it did not vary between groups or throughout the course of the study. There were no significant differences in baseline characteristics or initial ventilatory variables between groups. End-expiratory and end-inspiratory airway flows were zero in all preparations. Postventilation pH (7.30 ± 0.02), temperature (35.3 ± 0.2° C), and hemoglobin concentration (0.7 ± 0.03 g/ dl) did not differ significantly between groups. Postventilation Po 2 was 130 to 150 mm Hg in all preparations, reflecting ventilation with room air.
| Parameter | F20P35 | F3P35 | F20P20 | |||
|---|---|---|---|---|---|---|
| Baseline | ||||||
| Lung weight, g | 14.7 ± 2.7 | 16.0 ± 1.6 | 15.3 ± 0.54 | |||
| Static compliance, ml/cm H2O | 6.8 ± 1 | 5.6 ± 1.3 | 5.8 ± 0.8 | |||
| Kf, ml/min/mm Hg/100 g | 0.16 ± 0.05 | 0.19 ± 0.09 | 0.27 ± 0.12 | |||
| Beginning of ventilation | ||||||
| Plateau pressure, cm H2O | 30.4 ± 0.05 | 30.3 ± 0.1 | 30.1 ± 0.3 | |||
| PEEP, cm H2O | 5.3 ± 0.07 | 5.4 ± 0.05 | 5.3 ± 0.07 | |||
| Mean airway pressure, cm H2O | 17.9 ± 0.04 | 17.8 ± 0.06 | 17.7 ± 0.17 | |||
| Expriatory tidal volume, ml | 32.1 ± 2.5 | 34.5 ± 0.9 | 32.4 ± 3.6 | |||
| After ventilation | ||||||
| Static compliance, ml/cm H2O | 1.9 ± 0.53 | 3.8 ± 0.7 | 4.4 ± 0.8 | |||
| Kf, ml/min/mm Hg/100 g | 0.77 ± 0.24 | 0.31 ± 0.07 | 0.53 ± 0.20 |
The hemodynamic characteristics of each group before, during, and after ventilation are shown in Table 3. There were no unexpected significant differences in hemodynamic characteristics between groups at baseline or during ventilation. Post-ventilation pulmonary artery pressure tended to be higher in the F20P35 group than in the F3P35 group (p = 0.014 at 100 ml/min flow and p = 0.06 at 450 ml/min).
| Parameter | F20P35 | F3P35 | F20P20 | |||
|---|---|---|---|---|---|---|
| Baseline | ||||||
| Ppa, mm Hg, at perfusate flow 100 ml/min | 14.8 ± 0.3 | 14.3 ± 0.6 | 14.3 ± 0.9 | |||
| Pdo, mm Hg | 11.7 ± 0.2 | 11.3 ± 0.1 | 11.1 ± 0.2 | |||
| ΔP preintermediate, %† | 65.3 ± 2.5 | 69.3 ± 2.2 | 72.8 ± 5.3 | |||
| ΔP postintermediate, %† | 34.9 ± 2.6 | 31.0 ± 2.1 | 27 ± 5.3 | |||
| Ppa at perfusate flow 450 ml/min/mm Hg | 22.2 ± 0.5 | 22.33 ± 1 | 22.6 ± 1.6 | |||
| Ppa after overflow vent opened, mm Hg | 16.9 ± 1.0 | 16.2 ± 1.2 | 13.3 ± 0.7 | |||
| Beginning of ventilation | ||||||
| Peak Ppa, mm Hg | 35 | 35 | 20 | |||
| Nadir Ppa, mm Hg | 22.3 ± 0.57 | 20.3 ± 0.6 | 14.8 ± 0.97‡ | |||
| Mean Ppa, mm Hg | 28.7 ± 0.29 | 27.7 ± 0.3 | 17.4 ± 0.49‡ | |||
| Left atrial pressure, mm Hg | 10 | 10 | 10 | |||
| Vascular flow, ml/min/g lung‡ | 27.5 ± 0.29 | 24.3 ± 5.3 | 2.0 ± 0.6‡ | |||
| After ventilation | ||||||
| Ppa at perfusate flow 450 ml/min | 26.2 ± 1.6 | 21.0 ± 1 | 24.2 ± 1.7 | |||
| Ppa, mm Hg, at perfusate flow 100 ml/min | 16.0 ± 0.7 | 13.0 ± 0.3§ | 14.4 ± 0.8 | |||
| Pdo, mm Hg | 11.4 ± 0.4 | 10.8 ± 0.11 | 11.1 ± 0.3 | |||
| ΔP preintermediate, %† | 80 ± 3.4 | 73 ± 3.7 | 74 ± 4.7 | |||
| ΔP postintermediate, %† | 20 ± 3 | 27 ± 3.7 | 26 ± 5 |
None of the baseline characteristics correlated with histologic abnormalities. In groups F20P35 and F3P35, hemodynamic and ventilatory characteristics other than ventilatory frequency did not correlate with indices of damage.
F20P35 lungs suffered more edema formation, as assessed by weight gain during the ventilatory period, than did F3P35 lungs (1.77 ± 0.53 g/g lung tissue versus 0.40 ± 0.08 g/g lung tissue, p = 0.029; (Figure 1). Similarly, F20P35 lungs gained more weight than did F20P20 lungs (1.77 g/g lung tissue versus 0.11 ± 0.03 g/g lung tissue, p = 0.012).
Fig. 1. Lung weight gain, perivascular hemorrhage, and alveolar hemorrhage after injurious ventilation. *p < 0.05 for the comparison with F20P35 group; †p < 0.05 versus both F20P35 and F3P35 groups. All results reported as mean ± SEM.
F20P35 lungs displayed more perivascular hemorrhage than did F3P35 lungs (57 ± 13 versus 19 ± 8 vessels per lung, p = 0.029 (Figure 1). Similarly, F20P35 lungs displayed more perivascular hemorrhage than did F20P20 lungs (57 ± 13 versus 7 ± 3 vessels per lung, p = 0.008). In contrast, there was no difference between the high pulmonary artery pressure groups when alveolar hemorrhage in the most damaged regions was evaluated (61 ± 6 versus 64 ± 5% of alveoli contained erythrocytes, p = NS). Both high pulmonary artery pressure groups displayed significantly greater alveolar hemorrhage than the low pulmonary artery pressure group (33 ± 8% of alveoli contained erythrocytes, p ⩽ 0.029 for both comparisons). Recasting the analysis in terms of the number of erythrocytes contained in each alveolus (intensity of hemorrhage), or the relative area of each section displaying alveolar hemorrhage, did not change these results.
Two preparations in the F20P35 group developed gross air leaks (air-space rupture) during injurious ventilation; none of the lungs in the other two groups developed air leaks during or after injurious ventilation. Changes in compliance did not attain significance between groups when analyzed by repeated- measures ANOVA. However, when considered as percent decreases from baseline, there were significant differences between groups, despite the loss of two lungs in the F20P35 group because of air-space rupture (F20P35: −69 ± 9.3%; F3P35: −24 ± 11.9%; F20P20: −25 ± 8.2%; p = 0.035 for F20P35 versus F3P35 and p = 0.036 for F20P35 versus F20P20) (Figure 2).
Fig. 2. Static compliances before and after injurious ventilation, percent change in static compliance, and ultrafiltration coefficients before and after injurious ventilation. Open bars = before ventilation; closed bars = after injurious ventilation. *p < 0.05 for the comparison to F20P35 group. All results reported as mean ± SEM.
The mean ultrafiltration coefficient rose in all groups; however, between-group comparisons did not attain statistical significance either as absolute values or as percent changes (Figure 2).
Our main findings in this study were that lungs ventilated at low frequencies and high peak pulmonary artery pressures formed less edema and displayed less perivascular hemorrhage than did those ventilated at higher frequencies but identical peak pulmonary artery pressures. In addition, lungs ventilated at high peak pulmonary artery pressures and flows demonstrated more extensive histologic alterations and edema formation than did those subjected to the same ventilatory pattern but lower peak vascular pressures and flows. Lungs subjected to high peak pulmonary artery pressures underwent more alveolar hemorrhage than did those exposed to low peak pulmonary artery pressures, regardless of ventilatory frequency. Several methodologic issues must be addressed.
We did not include a true control group (low vascular pressure, low ventilatory frequency, and low airway plateau pressure) because we have previously found this model to display minimal gravimetric, histologic, and permeability alterations when not subjected to high vascular or air-space pressures in experiments of the same duration as this study (14). Additionally, the current study was not designed to distinguish between the effects of frequency per se and those of an increased total number of high pressure ventilatory cycles. The study was carried out in a system containing few leukocytes, and was not designed to address issues of leukocyte activation or trapping. Finally, this study was designed to address only a very acute phase of lung injury and was not designed to consider the potential for vascular remodeling.
The peak vascular pressures studied and, therefore, the flow rates they engendered, were set arbitrarily. The flow resulting from the high vascular pressure averaged ∼ 400 ml/min, or about two-thirds the normal cardiac output of a rabbit weighing 3 kg (21). The low pressure group was not intended to be physiologic. The addition of a prostacyclin inhibitor could have altered the lung microcirculation and inflammatory response. The interruption of lymphatic drainage prevalent in unventilated isolated, perfused lungs could have accentuated edema formation. Finally, pulmonary vascular pressures in the intact animal vary less during the respiratory cycle than do those in isolated, perfused lung preparations. Despite these limitations, this model does allow determination of the damaging effects caused by extremes of ventilation and perfusion. To extend our findings to the clinical arena will require an intact animal model with tightly controlled hemodynamics.
A dependence of edema formation on minute ventilation was previously noted by Bshouty and Younes (22), although their study differed from ours in four notable ways. First, their study was conducted with a physiologic ventilatory pattern, whereas we employed injurious ventilatory patterns. Second, they elevated vascular pressure via left atrial pressure (thereby increasing pressure along the entire vascular tree, rather than the preintermediate segment of the pulmonary vasculature), whereas we increased pulmonary artery pressure and held left atrial pressure fixed. Third, we used considerably higher vascular flows on a “per gram of lung” basis. Finally, we assessed histologic changes as reflected by lung hemorrhage.
Several mechanisms may underlie the diminution of edema formation and perivascular hemorrhage observed with decreased respiratory frequency. Higher ventilatory frequency could have depleted surfactant, increasing alveolar surface tension and lowering end-inspiratory perivascular pressure in the F20P35 group (23). The increased transvascular pressure gradient across extraalveolar vessels could then favor fluid transudation, vessel disruption, and perivascular hemorrhage. The importance of extraalveolar vessels has been demonstrated in other models of lung injury (24). Alternatively, surfactant film adsorption may have decreased surface tension forces during the later portion of the inspiratory phase in group F3P35, leading to more similar durations of applied perivascular stress between groups F20P35 and F3P35 (25). The larger number of stress cycles imposed on the F20P35 group may then have led to more severe vascular disruption. The importance of an increased number of stress cycles in augmenting structural damage is well known in engineering, particularly in the analysis of materials fatigue, and has been investigated in a variety of biomaterials (26-28). In both cases, the lower pulmonary artery pressures in the F20P20 group could have ameliorated injury by decreasing the transvascular pressure gradient. It is likely that the increased vascular pressures in the F20P35 group also increased both filtration surface area and the hydraulic driving pressure for fluid filtration relative to the F20P20 group (29). However, these effects alone seem unlikely to entirely explain the differences between the two groups. In particular, if the differences in pulmonary artery pressures are assumed to have been transmitted entirely to the filtering segments of the vasculature, the resulting differences in hydraulic driving pressure would account for a relatively small fraction of the differences in weight gain between groups F20P35 and F20P20.
The results of this study indicate that decreased ventilatory frequency lessens edema formation and perivascular hemorrhage, consistent with earlier work in intact, salicylate-treated dogs (30). Other previous work has suggested that repetitive airway opening and closure can contribute to lung injury and mediator production (31, 32). In addition, limitation of peak pulmonary artery pressures (and flows) may substantially moderate the severity of lung injury for a given ventilatory pattern. If confirmed in intact animals, our results would suggest that decreased respiratory frequency or peak pulmonary artery pressure could diminish the severity of ventilator-induced lung injury. Although drawing parallels between animal models and clinical practice is clearly hazardous, it is conceivable that similar mechanisms could be operative in the early postintubation phase of the acute respiratory distress syndrome. If so, the definition of a “lung protective” ventilatory strategy could be modified to include such cofactors as respiratory frequency and pulmonary vascular pressure and flow.
The writers wish to thank Terrie Grove for her excellent technical assistance during this study, and Dr. Cynthia Gross for her statistical assistance.
Supported by HealthPartners Foundation, by SCOR Grant HL-50152 from the National Institutes of Health, and by American Heart Association Scientist Development Grant 9930184N.
| 1. | Tsuno K., Prato P., Kolobow T.Acute lung injury from mechanical ventilation at moderately high airway pressures. J. Appl. Physiol.691990956961 |
| 2. | Parker J. C., Hernandez L. A., Peevy K. J.Mechanisms of ventilator-induced lung injury. Crit. Care Med.211993131143 |
| 3. | Dreyfuss D., Saumon G.Role of tidal volume, FRC, and end-inspiratory volume in the development of pulmonary edema following mechanical ventilation. Am. Rev. Respir. Dis.148199311941203 |
| 4. | Dreyfuss D., Basset G., Soler P., Saumon G.Intermittant positive-pressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am. Rev. Respir. Dis.1321985880884 |
| 5. | Parker J. C., Hernandez L. A., Longenecker G. L., Peavy K., Johnson W.Lung edema caused by high peak inspiratory pressures in dogs. Am. Rev. Respir. Dis.1421990321328 |
| 6. | West J. B., Tsukimoto K., Mathieu-Costello O., Prediletto R.Stress failure in pulmonary capillaries. J. Appl. Physiol.70199117311742 |
| 7. | Fu Z., Costello M. L., Tsukimoto K., Prediletto R., Elliott A. R., Mathieu-Costello O., West J. B.High lung volume increases stress failure in pulmonary capillaries. J. Appl. Physiol.731992123133 |
| 8. | Costello M. L., Mathieu-Costello O. M., West J. B.Stress failure of alveolar epithelial cells studied by scanning electron microscopy. Am. Rev. Respir. Dis.145199214461455 |
| 9. | Tsukimoto K., Mathieu-Costello O., Prediletto R., West J. B.Ultrastructural appearances of pulmonary capillaries at high transmural pressures. J. Appl. Physiol.711991573582 |
| 10. | Wu D. X. Y., Weibel E. R., Bachofen H., Schurch S.Lung lesions in experimental hydrostatic pulmonary edema: an electron microscopic and morphometric study. Exp. Lung Res.211995711730 |
| 11. | Bachofen H., Schurch S., Michel R. P., Weibel E. R.Experimental hydrostatic pulmonary edema in rabbit lungs: morphology. Am. Rev. Respir. Dis.1471993989996 |
| 12. | Bachofen H., Schurch S., Michel R. P., Weibel E. R.Experimental hydrostatic pulmonary edema in rabbit lungs: barrier lesions. Am. Rev. Respir. Dis.14719939971004 |
| 13. | Guery B. P., DeBoisblanc B. P., Fialdes P., Sarphy T. G., Nelson S., Chidiac C., Beucaire G., Summer W. R., Mason C. M.Pulmonary stress injury within physiological ranges of airway and vascular pressures. J. Crit. Care1319985866 |
| 14. | Broccard A. F., Hotchkiss J. R., Kuwayama N., Olson D. A., Jamal S., Wangensteen O. D., Marini J. J.Consequences of vascular flow on lung injury induced by mechanical ventilation. Am. J. Respir. Crit. Care Med.157199819351942 |
| 15. | Hakim T. S., Michel R. P., Chang H. K.Effect of lung inflation on pulmonary vascular resistance by arterial and venous occlusion. J. Appl. Physiol.53198211101115 |
| 16. | Benjamin J. J., Murtagh P. S., Proctor D. F., Menkes H. A., Permutt S.Pulmonary vascular interdependence in excised dog lobes. J. Appl. Physiol.371974887894 |
| 17. | Dawson C. A., Grimm D. J., Linehan J. H.Effects of lung inflation on longitudinal distribution of pulmonary vascular resistance. J. Appl. Physiol.43197710891092 |
| 18. | De Bono E. F., Caro C. G.Effect of lung-inflating pressure on pulmonary blood pressure and flow. Am. J. Physiol.205196311781186 |
| 19. | Smith J. C., Mitzner W.Analysis of pulmonary vascular interdependence in excised dog lobes. J. Appl. Physiol.481980450467 |
| 20. | Snedecor, G. W., and W. G. Cochran, editors. 1967. Statistical Methods, 6th ed. Iowa State University Press, Ames. 128–130, 258–275. |
| 21. | Neutze J. M., Wyler F., Rudolph A. M.Use of radioactive microspheres to assess distribution of cardiac output in rabbits. Am. J. Physiol.2151968486495 |
| 22. | Bshouty Z., Younes M.Effect of breathing pattern and level of ventilation on pulmonary fluid filtration in dog lung. Am. Rev. Respir. Dis.1451992372376 |
| 23. | Hida W., Hildebrandt J.Alveolar surface tension, lung inflation, and hydration affect interstitial pressure [Px(f)]. J. Appl. Physiol.571984262270 |
| 24. | Lamm W. J. E., Luchtel D., Albert R. K.Sites of leakage in three models of acute lung injury. J. Appl. Physiol.64198810791083 |
| 25. | Smith J. C., Stamenovic D.Surface forces in lungs: I. Alveolar surface tension-lung volume relationships. J. Appl. Physiol.60198613411350 |
| 26. | Hashin Z., Rotem A.A cumulative damage theory of fatigue failure. Mater. Sci. Eng.341978147160 |
| 27. | Schectman H., Bader D. L.In vitro fatigue of human tendons. J. Biomech.301997829835 |
| 28. | Pattin C. A., Caler W. E., Carter D. R.Cyclic mechanical property degradation during fatigue loading of cortical bone. J. Biomech.2919966979 |
| 29. | Anglade D., Corboz M., Menaouar A., Parker J. C., Sanou S., Bayat S., Benchetrit G., Grimbert F. A.Blood flow vs. venous pressure effects on filtration coefficient in oleic acid-injured lung. J. Appl. Physiol.84199810111023 |
| 30. | Mascheroni D., Kolobow T., Fumagalli R., Moretti M. P., Chen V., Buckhold D.Acute respiratory failure following pharmacologically induced hyperventilation: an experimental animal study. Intensive Care Med.151988814 |
| 31. | Muscedere J. G., Mullen J. B. M., Gan K., Slutsky A. S.Tidal ventilation at low airway pressures can augment lung injury. Am. J. Respir. Crit. Care Med.149199413271334 |
| 32. | Tremblay L., Valenza F., Ribeiro S. P., Li J., Slutsky A. S.Injurious ventilatory strategies increase cytokines and c-fos m-RNA expression in an isolated rat lung model. J. Clin. Invest.991997944952 |
Lluis Blanch is the recipient of Grant BAE 97/5478 from FIS (Spain).
