Acute lung injury (ALI) and its more severe form, acute respiratory distress syndrome (ARDS) continues to be associated with a relatively high mortality. Although conventional mechanical ventilation (CMV) remains the primary supportive therapy for this group of patients, this intervention has been shown to worsen existing lung injury (i.e., ventilator induced lung injury, VILI). Overstretch of some areas of the lung due to high peak inspiratory pressure (PIP) and/or large tidal volume (VT) can injure lung tissue, as can shear forces created between adjacent regions of inflated and collapsed alveili.1,2 Patients with ALI/ARDS have aerated and relatively healthy units in the nondependent region and collapsed and consolidated units in the dependent region, which renders the lung more vulnerable to mechanical stretch during CMV and results in VILI.3,4
Pulmonary surfactant dysfunction can occur in animals when ventilated with high PIP and/or VT. In healthy rats, a short period of high PIP ventilation resulted in significant increases of plasma proteins and conversion of functionally active, large surfactant aggregates (LA) into inferior functioning small aggregates (SA) in bronchoalveolar lavage fluid (BALF), which resulted in significant pulmonary surfactant dysfunction and correlated well with the severity of VILI.5 In a rabbit model of ALI, ventilatory modes utilizing high VT caused a significant decrease in LA surfactant pool size which contributed to alveolar instability and lung dysfunction.6 Based on these results, it can be hypothesised that ventilatory strategies aimed at minimizing pulmonary surfactant dysfunction induced by CMV would decrease the progressive lung dysfunction typically observed in patients with ALI/ARDS when standard CMV is used.
Tracheal gas insufflation (TGI) is a technique in which fresh gas is introduced into the trachea. It has been suggested as a useful adjunct to CMV in patients with ALI/ARDS. When combined with CMV, TGI can augment ventilation by reducing the dead space of the ventilating system, reducing ventilatory pressures and VT while maintaining constant arterial PaCO2.7-11 In this study, we tested the hypothesis that TGI, ventilating rabbits with limited PIP and VT, would minimize CMV induced alveolar pulmonary surfactant dysfunction and thereby attenuate VILI with lipopolysaccharide-induced ALI.
Continuous TGI was introduced as described elsewhere.12 Ancillary flow was injected into the lower part of the endotracheal tube (ETT) through six capillaries passing through the wall 1.5 cm above the distal end of the tube (Size ID 3.0, Vygon, France). Two other independent capillaries were used for collection of expired gases and measurement of pressure at the tip of the ETT.
Animal preparation and measurements
Sixteen healthy adult, New Zealand white rabbits [weight: (2.34±0.26) kg, age: 6 to 8 weeks] were provided by Beijing Anzhen Hospital, Capital University of Medical Sciences, China. Following anaesthesia with an intramuscular injection of ketamine (23 mg/kg), acepromazine (0.58 mg/kg), and xylazine (0.78 mg/kg), animals were tracheotomized, intubated (ETT see above), and mechanically ventilated (VIP Gold - Bird, Viasys, USA). The ventilator was initially set at a PIP of 9 to13 cmH2O, positive end expiratory pressure (PEEP) of 5 cmH2O, flow rate of 10 L/min to provide a VT of about 10 mL/kg, respiratory frequency of 30 breaths/min to maintain PaCO2 within normal range (35 to 45 mmHg), with inspiratory time of 1.0 second and a fraction of inspired oxygen (FiO2) of 1.0. During the experiment, PEEP, respiratory frequency, Ti, flow rate and FiO2 were kept constant; PIP and VT were adjusted according to experimental protocols (see below). Catheters (5 Fr.) were placed in a jugular vein and carotid artery. All animals were paralysed with pancuronium bromide (0.1 mg/kg), received metabolic substrate (5% glucose: 5 ml·kg-1·h-1) and supplemental paralytic (pan curonium bromide: 0.1 mg·kg-1·h-1) during the protocol. Sodium bicarbonate was given if the pH ≤ 7.25. Following instrumentation and stabilization, pre-injury values of pH, PaO2 and PaCO2 with a blood gas analyser (Nova Statprofile M, USA), dynamic respiratory compliance (Cdyn) and VT (GM 250 Navigator, Newport Medical Instruments, USA) were recorded. Mean expired PaCO2 was measured and physiological dead space to tidal volume ratio (VD/VT) was calculated as previously described.12 In addition, heart rate (HR) and mean systemic arterial pressure (MAP) (Spacelab, USA) were continuously monitored throughout the experiment.
Lung injury was induced by intratracheal instillation of lipopolysaccharide (Echoli. Serum type 055B5, Sigama Company, USA) at a dose of 50 mg/kg, PIP was adjusted to maintain VT at about 10 ml/kg for each animal. Lung injury entry criteria were defined as Cdyn < 50% of baseline value and <0.50 ml·cmH2O-1·kg-1, PaO2 < 200 mmHg and maintained for 30 minutes, which time defined post-injury. Animals were then randomly assigned to one of two groups, TGI or CMV (n=8 for each group) with continuous insufflation rate of 0.5 L/min (FiO2=1.0). Arterial blood gases, ventilator parameters, lung mechanics and VD/VT were measured and recorded at post-injury and then every 30 minutes during 4 hour post injury; PIP was adjusted according to previous blood gas to keep eucapnia (PaCO2 35 to 45 mmHg) and was limited to a maximum of 30 cmH2O, to minimize ventilatory pressure and volume requirements.
Bronchoalveolar lavage and lung tissue collection
At the end of the protocol, animals were euthanized with an overdose of sodium pentobarbital and then disconnected from the ventilator. The right main stem bronchus was clamped and the left lung was lavaged with room temperature citrated normal saline solution at 4×10 ml/kg.13 BALF was pooled and the total volume recorded. More than 85% of the instilled saline was collected from each animal. The BALF was immediately centrifuged for 10 minutes at 200 ×g and 4°C for removing cell debris and was then stored at -70°C for further chemical analysis and surface tension measurement. Unclamped right main bronchus and both lungs were reopened with a continuous positive airway pressure at 5 cmH2O and then perfused via pulmonary arteries with Millonig’s buffer until it ran clear.12 The left main stem was ligated and the left lung was removed and snap frozen in liquid nitrogen and stored at -80°C for subsequent analysis of inflammatory mediators (see below). The right lung was perfused for 30 minutes via the pulmonary artery with 10% buffered formalin at a pressure of 65 cmH2O and then stored in 10% buffered formalin until processed within one week.
Chemical analysis and surface tension measurement of BALF
Aliquots of BALF were thrice extracted with chloroform-methanol (2:1, v/v) to isolate the phosolipids in the chloroform phase and the total phosolipids (TPL) were quantified using a phosphorus assay.13 Total proteins (TP) in the BALF were measured by Bradford protein assay; TPL and TP measures were corrected by the total volume of BALF and body weight and presented as mg/kg. To measure the surfactant aggregate pool sizes, 10 ml of BALF was spun at 40 000 ×g (Survall® RC 5B Plus, Kendro, USA) for 15 minutes at 4°C, yielding a supernatant (small aggregate, SA) surfactant fraction, while the pellet was suspended in 10 ml normal saline (large aggregate, LA) surfactant fraction.14 Total phospholipid phosphorus in the SA and LA were extracted and quantified by the same method for TPL (see above) and values were expressed as SA/LA ratio. The in vitro surface activity of BALF sample was assessed using a pulsating bubble surfactometer (Electronics, USA) as described by Enhorning et al.15 Briefly, aliquots of BALF were lyophilized using a speed vacuum concentrator and a Labconco® (Savant Instrument Inc., USA) and then the pellets were resuspended in normal saline to a concentration of 2.5 mg phospholipid/ml. With this technique, an air bubble was formed in a surfactant suspension and pulsated at 20 pulsations/min between a radius of 0.4 mm and 0.55 mm, with temperature maintained at 37°C. Values of minimum and maximum surface tension (γmin, γmax) were obtained from the average values of 100 pulsations (5 minutes) at a minimum and maximum bubble radius following 10 seconds of adsorption, γ min and γ max were expressed as mN/m.
Histological and morphometric examination of lungs
Representative lung tissue blocks from all right lung lobes were embedded in paraffin. Sections stained with haematoxylin and eosin were examined by microscopy for evidence of lung injury. Lung expansion was quantified by the point counting method and expressed as volume density (VV) of aerated alveolar spaces, using total parenchyma as the reference volume.13 Fifty fields of each lung section were examined from each animal (magnification: ×300), and field to field variability was determined by calculating the coefficient of variation of VV [CV(VV)]. A low value of CV (VV) indicates homogeneity of alveolar aeration.
Inflammatory mediator analysis
Lung myeloperoxidase (MPO) activity and interleukin 8 (IL-8) protein were measured using a colorimetric bioassay and rabbit specific enzyme linked immunosorbent assay respectively.7 Lung MPO and IL-8 were expressed as U/g and ng/g tissue weight respectively.
All values are reported as mean±standard deviation (SD). A paired student t test was used to determine within group differences (i.e. post-injury vs pre-injury), and a two sample, equal variance student t test was used to determine between group differences. Significance was set at P < 0.05. Statistical results were generated using SPSS v11.
Body weight did not differ significantly between groups. HR and MAP were maintained at 180 to 290 beats/min and 65 to 110 mmHg respectively for all animals and no accidental deaths occurred. Physiological data at pre- and post-injury are summarized in Tables 1 and 2. Post-injury values of PIP and VD/VT were significantly higher (all P <0.01), and Cdyn and PaO2 were significantly lower (all P <0.01) in both groups than pre-injury. VT was kept at about 10 ml/kg and PaCO2 was maintained within normal range in both groups. There were no significant group differences for all parameters either at pre-injury or post-injury.
Summarized mean physiological data over 4 hour treatment time are compared in Table 3. PIP, VT and VD/VT over time were significantly lower (P <0.01) and Cdyn and PaO2 over time were significantly greater (P <0.05) in the TGI group than in the CMV group. PaCO2 in both groups was maintained within normal range and did not differ significantly between groups. Further analyses showed that Cdyn and PaO2 over 4 hour treatment time in the CMV group significantly (P <0.05) decreased as compared with their post-injury values and needed vigorous PIP to keep a constant VT (Table 2). VD/VT in the CMV group did not change significantly from post-injury value. PIP, VT and VD/VT decreased by 25%, 35% and 29% (all P <0.05) respectively from post-injury values in the TGI group, but Cdyn and PaO2 did not change significantly (Tables 1 and 2).
Values for TPL, TP, TPL/TP, SA/LA, γmin and γmax in BALF at the end of the 4 hour experiment are shown in Table 4. There was no significant difference in TPL between groups. TP and SA/LA in the TGI group were significantly (P < 0.05) lower, and TPL/TP in the TGI group was significantly (P < 0.05) higher than those in the CMV group. When analysed in pulsating bubble surfactometer, the lower surface activity, γmin of surfactant suspension (but notγmax) was significantly different between groups (P < 0.05).
Lung histological findings included prominent atelectasis, hyaline membranes, oedema, intra-alveolar and interstitial patchy haemorrhage and infiltration of neutrophils in the lungs of animals in the CMV group. In the TGI group, there was improved aeration of alveoli, and hyaline membranes, oedema, haemorrhage and infiltration of neutrophils were less severe than in the CMV group. Results from morphometric and inflammatory analyses are shown in Table 5. Aeration of alveoli was significantly (P <0.05) improved in the TGI group as increased VV and lower value for CV(VV) as compared with CMV group. Lung inflammation was significantly (P <0.05) reduced in the TGI group by lower values for MPO and IL-8 as compared with CMV group.
Several studies have compared TGI and CMV, with consistent findings showing improvement in pulmonary physiology with the application of TGI.7-11,16 The mechanism of reduced ventilatory support seems to be through TGI reducing dead space. This improves ventilation and oxygenation through lower ventilatory pressures and volumes. In agreement with these studies, our present study has shown that TGI can reduce PIP, VT and VD/VT, while maintain eucapnia and oxygenation. Most previous studies have used catheters placed within the ETT to provide a continuous or expiratory phasic flow that defines TGI. It is difficult to provide effective TGI to small lungs (such as small animals and human neonates) because of small internal diameter within ETT and the fragile and vulnerable trachea involved. In the current study, we used a multiple channel ETT with six capillary flow ports through which TGI was applied without interrupting the CMV. This sized tube has been used by others17,18 as well as by ourselves12 and is effective in providing TGI to small lungs. Bench testing has suggested TGI flow rate of 0.5 L/min to be optimal for providing significant CO2 removal and negligible alternation in ventilatory pressures and volumes attributable to the low TGI flow alone.12
We hypothesized that TGI, by providing animals with limited PIP and VT to better ventilation, would minimize CMV induced pulmonary surfactant dysfunction. On analysing the BALF, we found that the TGI group had significantly lowered TP, SA/LA and γmin values and significantly higher TPL/TP value compared with CMV group, indicating alveolar surfactant composition and function were preserved in the TGI group. Various studies have shown that certain aspects of mechanical ventilation can lead to deterioration in the surfactant environment. Mechanical ventilation with high PIP can cause protein-rich oedema and inflammation in healthy rat lungs, which incurs alveolar surfactant inhibition and lung dysfunction.19,20 High VT ventilation promotes the conversion of biophysically active LA to inactive SA in injured lungs of adult rabbits, and the rate of conversion increases with severity of injury.5,6In vitro experiments have confirmed that a combination of the phasic changes in surface area associated with tidal volume ventilation, together with the presence of a serine dependent protease, now known to be a carbooxyesterase, is primarily responsible for the conversion of surfactant LA to SA forms.21 Because animals in the TGI group were ventilated with lower pressure and VT, the disruptive force and therefore the protein-rich oedema was reduced, and the phasic changes of the contact surface for protease within the injured lung are minimized.22,23 To our knowledge, this is the first study to demonstrate the protective effect of TGI in combination with pressure limited ventilation on pulmonary surfactant.
Histologically, the TGI group showed less lung injury as evaluated by atelectasis, hyaline membranes formation, oedema, haemorrhage and neutrophil infiltration. These histological changes were further confirmed by lung morphometric and inflammatory analyses, which showed TGI group had the higher alveolar aeration and lower levels of MPO and IL-8 in their injured lung tissues. This improvement is consistent with preservation of surfactant composition and function (Table 4), as well as with what was seen physiologically with the significant differences in PIP, VT, VD/VT, Cdyn and PaO2 between the TGI and CMV groups (Table 3).
In summary, we found that TGI, by providing lower PIP and VT to improve ventilation, resulted in more favourable alveolar pulmonary surfactant composition and function and less severity of lung injury than conventional mechanical ventilation in this animal model of ALI. Based on this data, we believe that better ventilatory care to patients with ALI/ARDS may be achieved by using TGI in combination with pressure limited ventilation as a lung protective strategy.
1. Dreyfuss D, Saumon G. Ventilator induced lung injury
: lessons from experimental studies. Am J Respir Crit Care Med 1998; 157:294-323.
2. Zhang XR, Du YC, Jiang HY, Xu JY, Xu YJ. Experimental study of acute lung injury
induced by different tidal volume ventilation in rats. Chin Med J 2005; 118:777-780.
3. Moran JL, Bersten AD, Solomon PJ. Meta-analysis of controlled trials of ventilator therapy in acute lung injury
and acute respiratory distress syndrome: an alternative perspective. Intensive Care Med 2005; 31:227-235.
4. Liao PH, Whitehead T, Evans T, Griffiths M. Ventilator-associated lung injury. Lancet 2003; 361:332-340.
5. Zhu GF, Zhou X, Min J, Jin XQ. Pulmonary surfactant
impairment in the development of ventilator induced lung injury
in rats. Chin J Tuberc Respir Dis (Chin) 2001; 24:648-650.
6. Ito Y, Veldhuizen RAW, Yao LJ, Mccaig L, Bartlett AJ, Lewis JF. Ventilation strategies affect surfactant aggregate conversion in acute lung injury
. Am J Respir Crit Care Med 1997; 155:493-499.
7. Kirmse M, Fujino Y, Hromi J, Mang H, Hess D, Kacmarek RM. Pressure-release tracheal gas insufflation
reduces airway pressures in lung-injured sheep maintaining eucapnia. Am J Respir Crit Care Med 1999; 160:1462-1467.
8. Zhan QY, Wang C, Shang MY, Tong ZH, Weng XZ. Efficacy of continuous tracheal gas insufflation
in spontaneously breathing canine with acute lung injury
. Chin Med J 2001; 114:658-660.
9. Hota S, Crooke PS, Adams AB, Hotchkiss JR. Optimal phasic tracheal gas insufflation
timing: an experimental and mathematical analysis. Crit Care Med 2006; 34:1408-1414.
10. Zhan QY, Wang C, Shang MY, Tong ZH, Weng XZ. Effects of continuous tracheal gas insufflation
during biphasic intermittent positive airway pressure ventilation on canine model of acute lung injury
with spontaneous breathing. Natl Med J China (Chin) 2000; 80:54-57.
11. Zhang B, Liu YN. Effects of tracheal gas insufflation
on blood gases and respiratory mechanics in acute hypercapnia rabbits. Chin J Tuberc Respir Dis (Chin) 1999; 22:523-527.
12. Zhu GF, Shaffer TH, Wolfson MR. Continuous tracheal gas insufflation
during partial liquid ventilation in juvenile rabbits with acute lung injury
. J Appl Physiol 2004; 96:1415-1424.
13. Zhu GF, Sun B, Niu SF, Cai YY, Lin K, Lindwall R, et al. Combined surfactant therapy and inhaled nitric oxide in rabbits with oleic-acid induced acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 158:237-243.
14. Lewis JF, Ikegami M, Jobe AH. Altered surfactant function and metabolism in rabbits with acute lung injury
. J Appl Physiol 1990; 69:2303-2310.
15. Enhorning G. Pulsating bubble technique for evaluating pulmonary surfactant
. J Appl Physiol 1977; 43:198-203.
16. Martinez-Perez M, Bernabe F, Pena R, Fernandez R, Nahum A, Blanch L. Effects of expiratory tracheal gas insufflation
in patients with severe head trauma and acute lung injury
. Intensive Care Med 2004; 30:2021-2027.
17. Dassieu G, Broochard L, Benani M, Avenel S, Danan C. Continuous tracheal gas insufflation
in preterm infants with hyaline membrane disease: A prospective randomized trial. Am J Respir Crit Care Med 2000; 162: 826-831.
18. Dyer IR, Esmail M, Findlay G, Mecklenburgh JS, Dingley J. Effect of catheter design on tracheal pressures during tracheal gas insufflation
. Eur J Anaesthesiol 2003; 20:740-744.
19. Veldhuizen RA, Tremblay LN, Govindarajan A, van Rozendaal BA, Haagsman HP, Slutsky AS. Pulmonary surfactant
is altered during mechanical ventilation
of isolated rat lung. Crit Care Med 2000; 28:2545-2551.
20. Malloy JL, Veldhuizen RA, Lewis JF. Effects of ventilation on the surfactant system in sepsis induced lung injury. J Appl Physiol 2000; 88:401-408.
21. Gross NJ. Extracellular metabolism of pulmonary surfactant
: the role a new serine protease. Ann Rev Physiol 1995; 57:135-150.
22. Oliver RE, Rozycki HJ, Greenspan JS, Wolfson MR, Shaffer TH. Tracheal gas insufflation
as a lung-protective strategy: physiologic, histologic, and biochemical markers. Pediatr Crit Care Med 2005; 6:64-69.
23. Kerr CL, Veldhuizen RA, Lewis JF. Effects of high-frequency oscillation on endogenous surfactant in an acute lung injury
model. Am J Respir Crit Care Med 2001; 164:237-242.
Keywords:© 2006 Chinese Medical Association
tracheal gas insufflation; acute lung injury; pulmonary surfactant; ventilator induced lung injury; mechanical ventilation