Changes in the mechanical properties of the respiratory system during the development of interstitial lung edema
© Dellacà et al. 2008
Received: 19 October 2007
Accepted: 12 June 2008
Published: 12 June 2008
Pulmonary edema induces changes in airway and lung tissues mechanical properties that can be measured by low-frequency forced oscillation technique (FOT). It is preceded by interstitial edema which is characterized by the accumulation of extravascular fluid in the interstitial space of the air-blood barrier. Our aim was to investigate the impact of the early stages of the development of interstitial edema on the mechanical properties of the respiratory system.
We studied 17 paralysed and mechanically ventilated closed-chest rats (325–375 g). Total input respiratory system impedance (Zrs) was derived from tracheal flow and pressure signals by applying forced oscillations with frequency components from 0.16 to 18.44 Hz distributed in two forcing signals. In 8 animals interstitial lung edema was induced by intravenous infusion of saline solution (0.75 ml/kg/min) for 4 hours; 9 control animals were studied with the same protocol but without infusion. Zrs was measured at the beginning and every 15 min until the end of the experiment.
In the treated group the lung wet-to-dry weight ratio increased from 4.3 ± 0.72 to 5.23 ± 0.59, with no histological signs of alveolar flooding. Resistance (Rrs) increased in both groups over time, but to a greater extent in the treated group. Reactance (Xrs) did not change in the control group, while it decreased significantly at all frequencies but one in the treated. Significant changes in Rrs and Xrs were observed starting after ~135 min from the beginning of the infusion. By applying a constant phase model to partition airways and tissue mechanical properties, we observed a mild increase in airways resistance in both groups. A greater and significant increase in tissue damping (from 603.5 ± 100.3 to 714.5 ± 81.9 cmH2O/L) and elastance (from 4160.2 ± 462.6 to 5018.2 ± 622.5 cmH2O/L) was found only in the treated group.
These results suggest that interstitial edema has a small but significant impact on the mechanical features of lung tissues and that these changes begin at very early stages, before the beginning of accumulation of extravascular fluid into the alveoli.
The functional organisation of the lung extracellular matrix comprises basically two large macromolecular families. The fibrillar components, including collagen I and III and elastic fibers, provide the elasticity of the lung tissue on stretching and de-stretching which is mechanically defined as lung compliance, that is the ratio between the change in lung volume and the corresponding change in transpulmonary pressure. Other macromolecular components, that include hyaluronan (HA) and proteoglycans (PGs), fill the voids among the fibrillar structures and the cells and keep the various structures assembled thanks to multiple linkages, allowing however reciprocal movements of the structures . HA and PGs also play an important role in the control of extravascular lung water. Being highly hydrophilic, they can bind water to form gel-like structures and, furthermore, a peculiar physical property of their hydrated state is the mechanical resistance to compressive forces [1, 2]. These molecules are therefore mostly involved in the tissue response to edema formation.
Previous studies evaluated the changes in mechanical properties of the respiratory system after inducing pulmonary congestion[3, 4] or severe lung edema [5–9], but no measurements were reported in the condition of interstitial lung edema in vivo implying an increase in extravascular lung water that does not exceed 5–10%.
We wondered whether the mechanical properties of lung tissue are significantly affected when the scaffold of the extravascular matrix is put under tension by interstitial edema. To this purpose, respiratory mechanics was monitored by low frequency forced oscillation technique (FOT) [10, 11] in closed-chest mechanically ventilated rats by inducing interstitial lung edema by slow rate saline overload.
We studied 17 Wistar male rats (weight 325–375 g). Animal use and care procedures were approved by the institutional animal care and use committee and complied with guidelines set by the American Physiological Society. The animals were randomly assigned to two groups, one group received saline infusion as described below ('treated', n = 8) and the other one was used as control group (n = 9). The experimental protocol was exactly the same for the two groups excluding the infusion of isotonic saline solution which was performed only in the treated group.
The animals were pre-anesthetized by ether vapours, then anesthetized with an intraperitoneal dose of 50% diluted urethane (1200 mg/kg) and placed in a supine position. A jugular vein was cannulated for drug delivery (both groups) and saline infusion (treated only). Tracheostomy was performed, and a 50-mm plastic cannula (2.5-mm inner diameter) was inserted into the distal trachea. Mechanical ventilation was provided by the same device used for the estimation of mechanical impedance as described below with a tidal volume (VT) of 2.5 ml and a frequency of 14 breaths/min. A positive end-expiratory pressure of 1 cmH2O was maintained for all the duration of the experiment. Before connecting the animal to the ventilator, paralysis was accomplished by pancuronium bromide (1 mg/kg body wt initial dose, supplemented by 0.33 mg/kg every 40 min).
Protocol and experimental set-up
Once connected to the mechanical ventilator, the animals were ventilated for four hours with the same ventilator settings. During all this period, only the treated group received infusion of saline solution at a constant rate of 0.75 ml/kg/min.
OVW waveforms characteristics.
Multiplicative factors (NSND3)
Waveform OVW 1
Waveform OVW 2
The assessment of mechanical impedance required one minute for each waveform with a two min ventilation period between the two OVW. The measurements were repeated every 15 min throughout all the experiment. Since the measurement procedure may induce alveolar derecruitment, after each measurement the lung was inflated to 20–25 cmH2O for a few seconds before resuming the baseline ventilation.
At the end of the experiment, the animals were sacrificed by an overdose of urethane. The trachea was tied in the neck, the chest was widely opened and the lungs were excised. One lung was removed after tying its main bronchus in order to prevent collapse, immersed in formaline and subsequently processed using routine histological techniques. The other lung was cut into three-four pieces that were used to compute the wet to dry weight ratio (W/D) as an index of lung water content.
Several slices of the processed lung of each animal were analysed by light microscopy at a magnification of 100×. A pathologist was asked to blindly give a score from 0 to 3 to the degree of perivascular edema, peribronchial edema and alveolar flooding of each animal.
Pressure and flow at the airway opening (Pao and ) were measured by a transducer (PXLA0025DN, Sensym, Milpitas, CA) connected to the tracheal tube and by a Fleisch-type pneumotachograph (model 0000 connected to a pressure transducer PXLA02X5DN, 0–2.5 cm H2O; Sensym, Milpitas, CA). All connections were made by thick silicon tubing kept as short as possible to provide a satisfying common mode rejection ratio and frequency response. All the signals were sampled at 200 Hz by the same A/D-D/A board used to control the ventilator and recorded by a personal computer. The flow signal was integrated to give lung volume (VL). The volume drift resulted from the integration of the flow signal was removed by estimating the linear trend on the integrated signal and removing it from the recording.
The frequency response of the measuring systems was assessed by using the same mechanical model used to optimize the controller parameters and it was flat up to 20 Hz.
Estimation of input impedance:
where f is frequency and j is the imaginary unit.
Pao and signals were low-pass filtered with a cut-off frequency of 20 Hz when the base frequency was 0.07825 Hz and of 40 Hz when it was 0.313 Hz. The signals were resampled at 40 Hz and 160 Hz respectively as in . Zrs was computed as the ratio between the auto-spectrum of the flow signal and the cross-spectrum between the pressure and flow signals, in accordance with the cross-spectrum method . The spectra were obtained by averaging ~5 periodograms computed by applying the Fast Fourier Transform to data segments of 512 samples each. The coherence functions of the impedances were computed as described in . In the present study we considered only impedance data with a coherence value greater than 0.95. Data at the oscillatory frequency of 4.61 Hz were discarded as in several measurements the coherence was lower than the threshold likely because of the effect of cardiac activity.
Raw was corrected for the contribution of the tracheal tube.
Tissue hysteresivity (η) was calculated as the ratio G/H.
All the data are expressed as mean ± SD. Significance of differences was tested by two-way ANOVA for repeated measurements using time and treatment group as factors. Multiple comparison after ANOVA was performed using Holm-Sidak test with control group and baseline as control conditions. Statistical tests were performed with a significance level of p < 0.05.
Evaluation of the degree of edema. Wet to dry (W/D) ratio and degree of perivascular edema, peribronchial edema and alveolar flooding scored by light microscopy averaged for control and treated groups. Data are reported as mean ± SD in adimensional unit.
4.30 ± 0.72
5.23 ± 0.59
Degree of perivascular edema
1.20 ± 0.45
Degree of peribronchial edema
Because of the wide inter-individual variability of baseline values, there are very slight differences in Zrs at the end of the experiment between control and treated group.
In the present study we investigated the impact of the development of interstitial lung edema on respiratory system mechanical properties. We used low-frequency FOT combined to the constant-phase modeling as this method permits to identify how an experimental intervention differentially affects airway caliber, de-recruitment of lung units and regional heterogeneity of function, and this information cannot be obtained by simpler measures of lung function.
The scorings of Table 2 essentially indicate that there is no water accumulation in the alveolar spaces, in agreement with the results that Conforti et al. obtained with a similar model of mild pulmonary edema. Therefore, this study provides the first data concerning modifications in the overall mechanical properties of the respiratory system in vivo during interstitial lung edema measured by low-frequency FOT in absence of airway and alveolar flooding.
Anaesthesia and paralysis, together with prolonged mechanical ventilation, can induce changes in lung mechanics per se [21, 22], which are likely to be the reason of the slight increase of Raw and G observed in the control group within the four hours. Derecruitment of lung units was prevented by regular positive pressure inflations (recruitment maneuvers) and the use of PEEP, and this was confirmed by the absence of elevation of H with time in the control group.
In our study we used a forcing waveform with different amplitudes for the different frequency components, as the lowest frequency was used to provide also ventilation to the animal. It has been shown that the impedance of the lung is amplitude-dependent . However, this phenomenon was small in the healthy lung and increased markedly after oleic acid injury, where the damage is much more severe than the one induced by our model . The use of two different waveforms allowed us to evaluate the impact of this effect, as the high amplitude frequency in the OVW2 signal used to ventilate the animal is very close to the third small amplitude frequency component of OVW1 (Table 1). As the values of Zrs at these two frequencies are in very good agreement (Figure 2), the use of a higher amplitude at low frequency did not significantly affect the assessment of Zrs in our experimental conditions.
Non-linearity of the respiratory system may also affect the measurements, since the OVW is a composite signal, but the use of appropriate frequency components  should have minimized this issue. Moreover, even if this may have affected the absolute values of the impedances reported in figure 2, the possible impact is similar for all the measurements.
Comparing our data with those from other studies, we found that our measurements of Zrs and constant phase model parameters at baseline are similar to those reported in the literature . However, we found slightly lower hysteresivity (η = 0.14) compared to  (η = 0.18) and to  (η = 0.2). Sakai et al. found that the mechanism that plays a major role in the differences in η is heterogeneous constriction of the airways. The lower values of η reported in the present paper may be due to the use of OVW, which applies physiological tidal volume amplitudes making the lung more homogeneous than during measurement performed with small amplitude signals .
Finally, in this study we are measuring the total respiratory system impedance, which is the result of both lung and chest wall mechanical properties. In particular, the latter constitutes a significant component of the total input impedance [24, 26]. However, it has been shown that even during severe edema induced by oleic acid infusion, the chest wall mechanical properties are not affected. These experimental data support the hypothesis that saline infusion does not induce significant changes in the configuration or mechanical properties of the chest wall because, although edema is a general phenomenon affecting all body tissues, the interstitial pressure increases much more in the lung compared to muscles  and, accordingly, one would expect the impact of interstitial edema to be much more relevant in the lung compared to the intercostal muscles.
In conclusion, the differences of Zrs between the treated and the control groups and between the baseline and end-experiment in the treated group, shown in figures 2 and 3, should not be affected by all these issues, but they likely reflect the changes in visco-elastic properties of the lung tissue in the specific mechanical conditions of interstitial edema.
Changes in respiratory mechanics
The accumulation of water in the interstitial space induces significant changes in both Rrs and Xrs and, consequently, also in Raw, G, and H.
Figure 2 shows that in the treated group Rrs increased at all frequencies, while Xrs becomes more negative with a greater decrease at low frequencies. We found that Xrs, which mainly reflects the elastic properties of the tissues or the presence of closure or choke points in peripheral airways [27, 28], is much more sensitive than Rrs to the development of interstitial lung edema, displaying significant differences between baseline and end-experiment and between the control and the infused group at more frequencies.
We did not observe an evident transient increase of resistance as reported by Ishii and coworkers  which was attributed to vagal reflex. These differences are likely due to the differences in edema model: Ishii et al. induced lung edema by a marked elevation of left atrial pressure, inducing alveolar flooding in less than 100 min. Conversely, our model of lung edema was not associated to significant changes of systemic and lung capillary pressures [12, 29].
The differences in Zrs relative to baseline reached statistical significance after 135 min at most frequencies, suggesting that the mechanical properties of the lung are affected during the development of interstitial edema and/or vascular congestion even when the increase of extravascular water is not exceeding 5–10% [12, 20].
In order to partition airways and tissue contribution to changes in respiratory mechanics, we fitted the constant phase model on Rrs and Xrs data (Figure 3). Raw slightly increase in a similar way in the control and in the treated group. An increase in Raw can be associated to a reduction of airway calibre but, since in interstitial lung edema we did not find histological evidence of geometrical remodelling of the airways, this small change is likely related to the effects of the prolonged sedation and mechanical ventilation.
In the treated group G and H increased with time. An increase in G and H may be due to increased tissue damping and elastance, reduced lung volume, airway closures or heterogeneous airway constriction. We can exclude heterogeneous airways constriction because it is associated to increased η , which was not observed in our data. A reduction in lung volume could produce an increase in both G and H, however, η is also affected by changes in lung volume . As η was constant throughout the experiment, we can conclude that the changes in G and H that we observed are not due to lung volume changes. Finally, also airway closures are very unlikely to occur in our experimental conditions, as suggested also by the histological analysis. For all these reasons the increase in G and H is likely related to changes in the tissue mechanical properties.
Comparison to previous studies on lung edema
Previous studies correlating the development of lung edema with changes of lung mechanics were focused mainly either on the effects of pulmonary congestion only or on the effects of later stages of acute lung edema.
When the perturbation was limited to pulmonary congestion, mechanical properties of the lung were found to be minimally affected in the range of lung volume considered in our study [3, 4, 9]. The slight changes previously observed (i.e. decrease in compliance and an increase in pulmonary resistance), are, however, in the same direction as the changes reported in the present study.
Conversely, a major impact on lung mechanics was found when later stages of lung edema were induced by either volume loading , increase in microvascular permeability [6, 31] increase in left atrial pressure [7, 8] or aortic occlusion . However, these models caused pulmonary congestion, alveolar fluid accumulation, small airways compression/occlusion, conditions that clearly impact on the overall mechanical properties of the respiratory system including, of course, airways resistance and lung compliance. A further complication in the evaluation of respiratory mechanics during edema development is the repeated use of BAL that may cause ventilation heterogeneities .
Although the lung seems well designed to resist to edema formation, a sustained condition of interstitial edema is a cause of progressive loss of integrity of the interstitial matrix . The obvious question then is: what is the mechanical resistance of the matrix macromolecules to continuously increased stress? Our experimental evidence is that the transition to severe edema is a matter of minutes , suggesting that severe edema acutely develops when the damage to the extracellular matrix overcomes a critical threshold. Therefore, the condition of increased interstitial parenchymal stresses represents an unstable equilibrium between tissue repair and severe tissue lesion. The present data suggest that the assessment of the time course of lung mechanics by forced oscillation technique may provide useful information on the accumulation of water into the lung and/or the development of severe pulmonary congestion. For this reason, the approach developed in this study could have a potential impact for the study of the early phase of the development of lung edema and for the evaluation of possible countermeasures in experimental studies. Moreover, further studies should be addressed to evaluate if the use of FOT can represent a non-invasive, potentially clinically useful tool to detect the earliest stages of congestive heart failure, and institute therapy long before the condition becomes life-threatening.
This work was partially supported by the Italian Space Agency (ASI), DCMC contract.
The writers gratefully acknowledge Fabio Ciancitto for his technical assistance in the development of the experimental set-up and experimental activity.
- JuulSE, TN W, VC H: Proteoglycans. In The Lung: scientific foundations. Edited by: Crystal RG and West JB. New York, Raven; 1991:413–420.Google Scholar
- Hascall VC Jr., Kimura JH: Biosynthesis, secretion, and aggregation of proteoglycans by rat chondrosarcoma chondrocytes. Ala J Med Sci 1981, 18:29–35.PubMedGoogle Scholar
- FRANK NR: Influence of acute pulmonary vascular congestion on recoiling force of excised cats' lung. J Appl Physiol 1959, 14:905–908.PubMedGoogle Scholar
- FRANK NR, RADFORD EP Jr., Whittenberger JL: Static volumepressure interrelations of the lungs and pulmonary blood vessels in excised cats' lungs. J Appl Physiol 1959, 14:167–173.PubMedGoogle Scholar
- Tang GJ, Freed AN: The role of submucosal oedema in increased peripheral airway resistance by intravenous volume loading in dogs. Eur Respir J 1994, 7:311–317.View ArticlePubMedGoogle Scholar
- Barnas GM, Stamenovic D, Lutchen KR: Lung and chest wall impedances in the dog in normal range of breathing: effects of pulmonary edema. J Appl Physiol 1992, 73:1040–1046.PubMedGoogle Scholar
- Ishii M, Matsumoto N, Fuyuki T, Hida W, Ichinose M, Inoue H, Takishima T: Effects of hemodynamic edema formation on peripheral vs. central airway mechanics. J Appl Physiol 1985, 59:1578–1584.PubMedGoogle Scholar
- Hogg JC, Agarawal JB, Gardiner AJ, Palmer WH, Macklem PT: Distribution of airway resistance with developing pulmonary edema in dogs. J Appl Physiol 1972, 32:20–24.PubMedGoogle Scholar
- COOK CD, Mead J, SCHREINER GL, FRANK NR, Craig JM: Pulmonary mechanics during induced pulmonary edema in anesthetized dogs. J Appl Physiol 1959, 14:177–186.PubMedGoogle Scholar
- Lutchen KR, Yang K, Kaczka DW, Suki B: Optimal ventilation waveforms for estimating low-frequency respiratory impedance. J Appl Physiol 1993, 75:478–488.PubMedGoogle Scholar
- Hantos Z, Daroczy B, Suki B, Galgoczy G, Csendes T: Forced oscillatory impedance of the respiratory system at low frequencies. J Appl Physiol 1986, 60:123–132.View ArticlePubMedGoogle Scholar
- G M, D N, M DF, D V: Pulmonary interstitial pressure in intact in situ lung: transition to interstitial edema. J Appl Physiol 1993,74(3):1171–1177.Google Scholar
- Suki B, Lutchen KR: Pseudorandom signals to estimate apparent transfer and coherence functions of nonlinear systems: applications to respiratory mechanics. IEEE Trans Biomed Eng 1992, 39:1142–1151.View ArticlePubMedGoogle Scholar
- Dellaca RL, Black LD, Atileh H, Pedotti A, Lutchen KR: Effects of posture and bronchoconstriction on low-frequency input and transfer impedances in humans. J Appl Physiol 2004, 97:109–18.View ArticlePubMedGoogle Scholar
- Michaelson ED, Grassman ED, Peters WR: Pulmonary mechanics by spectral analysis of forced random noise. J Clin Invest 1975, 56:1210–1230.View ArticlePubMedPubMed CentralGoogle Scholar
- Aliverti A, Dellaca RL, Pedotti A: Transfer impedance of the respiratory system by forced oscillation technique and optoelectronic plethysmography. Ann Biomed Eng 2001, 29:71–82.View ArticlePubMedGoogle Scholar
- Hantos Z, Daroczy B, Suki B, Nagy S, Fredberg JJ: Input impedance and peripheral inhomogeneity of dog lungs. J Appl Physiol 1992, 72:168–178.View ArticlePubMedGoogle Scholar
- Fredberg JJ, Stamenovic D: On the imperfect elasticity of lung tissue. J Appl Physiol 1989, 67:2408–2419.PubMedGoogle Scholar
- Bates JH: Point:Counterpoint: lung impedance measurements are/are not more useful than simpler measurements of lung function in animal models of pulmonary disease. J Appl Physiol 2007, 103:1900–1901.View ArticlePubMedGoogle Scholar
- Conforti E, Fenoglio C, Bernocchi G, Bruschi O, Miserocchi GA: Morpho-functional analysis of lung tissue in mild interstitial edema. Am J Physiol Lung Cell Mol Physiol 2002, 282:L766-L774.View ArticlePubMedGoogle Scholar
- Brismar B, Hedenstierna G, Lundquist H, Strandberg A, Svensson L, Tokics L: Pulmonary densities during anesthesia with muscular relaxation--a proposal of atelectasis. Anesthesiology 1985, 62:422–428.View ArticlePubMedGoogle Scholar
- Hedenstierna G, Edmark L: The effects of anesthesia and muscle paralysis on the respiratory system. Intensive Care Med 2005, 31:1327–1335.View ArticlePubMedGoogle Scholar
- Barnas GM, Stamenovic D, Lutchen KR, Mackenzie CF: Lung and chest wall impedances in the dog: effects of frequency and tidal volume. J Appl Physiol 1992, 72:87–93.PubMedGoogle Scholar
- Hirai T, McKeown KA, Gomes RF, Bates JH: Effects of lung volume on lung and chest wall mechanics in rats. J Appl Physiol 1999, 86:16–21.PubMedGoogle Scholar
- Sakai H, Ingenito EP, Mora R, Abbay S, Cavalcante FS, Lutchen KR, Suki B: Hysteresivity of the lung and tissue strip in the normal rat: effects of heterogeneities. J Appl Physiol 2001, 91:737–747.PubMedGoogle Scholar
- Hantos Z, Daroczy B, Suki B, Nagy S: Low-frequency respiratory mechanical impedance in the rat. J Appl Physiol 1987, 63:36–43.PubMedGoogle Scholar
- Dellaca RL, Santus P, Aliverti A, Stevenson N, Centanni S, Macklem PT, Pedotti A, Calverley PM: Detection of expiratory flow limitation in COPD using the forced oscillation technique. Eur Respir J 2004, 23:232–240.View ArticlePubMedGoogle Scholar
- Johnson MK, Birch M, Carter R, Kinsella J, Stevenson RD: Use of reactance to estimate transpulmonary resistance. Eur Respir J 2005, 25:1061–1069.View ArticlePubMedGoogle Scholar
- Negrini D: Pulmonary microvascular pressure profile during development of hydrostatic edema. Microcirculation 1995, 2:173–180.View ArticlePubMedGoogle Scholar
- Lutchen KR, Suki B, Zhang Q, Petbk F, Dar£czy B, Hantos Z: Airway and tissue mechanics during physiological breathing and bronchoconstriction in dogs. J Appl Physiol 1994, 77:373–385.PubMedGoogle Scholar
- Novak Z, Petak F, Banfi A, Toth-Szuki V, Barati L, Kosa L, Bari F, Szekely E: An improved technique for repeated bronchoalveolar lavage and lung mechanics measurements in individual rats. Respir Physiol Neurobiol 2006, 154:467–477.View ArticlePubMedGoogle Scholar
- Miserocchi G, Negrini D, Passi A, De LG: Development of lung edema: interstitial fluid dynamics and molecular structure. News Physiol Sci 2001, 16:66–71.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.