| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
POINT-COUNTERPOINT COMMENTS
Stony Brook University
School of Medicine
The following letters are in response to Point:Counterpoint: "Lung impedance measurements are/are not more useful than simpler measurements of lung function in animal models of pulmonary disease" that appears in this issue.
To the Editor: Mitzner argues that constant phase modeling is not more useful that simpler methods and no research has substantiated the importance of several variables determined by constant phase modeling (6). This is untrue and indicates only a cursory attempt at a literature review. We clearly established a developmental pattern in normal healthy newborn rat pups where PV curves, tissue damping (G), elastance, and hysteresivity (eta) all change from days 10 to 17 (2). In particular, eta was found to change in a PEEP-sensitive manner consistent with the development of mature secretory cell lung function. Furthermore, we showed an alteration in this developmental pattern, particularly with G, static compliance, elastance, and airway resistance with CFTR gene expression levels (3, 4). Combined PV and constant phase modeling were essential to substantiating these conclusions.
Hysteresivity, G, and other constant phase parameters serve as sensitive developmental markers for pathology before histologic effects (4, 5). Thus, while we are still relating these parameters to specific structural and cellular changes, they become invaluable tools for identification of subtle gene developmental effects and as confirmation of PV curve changes. This is particularly important when dealing with new findings in controversial topics such as the role of CFTR in lung development.
Ockham's Razor states "all things being equal the simplest explanation tends to be correct," not the simplest data tends to be the most useful. As we seek to understand disease processes and pathophysiology, the more detailed information of constant phase modeling is simply better.
REFERENCES
Harvard University
To the Editor: Positions of both esteemed colleagues are regrettable (1, 5). Both are wrong. Impedance does not provide more information; it provides different information. And to say that Penh is "discredited" (1) is to fail to understand that screening tools are at times indispensable; to say otherwise is a disservice to science. Worse, to say that impedance parameters, and especially the hysteresivity, eta, have failed to live up to billing (5) is tantamount to surrendering hope of elucidating molecular mechanism. Ugh!
The constant-phase (equivalently, structural-damping) model derives from earlier work of Hildebrandt (4) and my own (3). But to write the model as does Bates (1) is to fail to understand what "constant-phase" means; the tissue term (G-iH) should properly be written, and should only be written, –iH(1+i eta), thus reflecting constancy of phase (arctan eta) and obviating problems with G.
Heterogeneity aside, eta has been shown to unlock secrets of underlying molecular dynamics, being conserved across scales from whole lung, parenchyma, smooth muscle, isolated cells, down to molecular cytoskeletal (CSK) regions (2, 3, 6). Drs. Bates and Mitzner excepted, the rest of the world now understands that virtually all CSK dynamics revolve around eta, which sets the CSK rheology exponent alpha, the rate of CSK remodeling, the extent of CSK fluidization in response to stretch, and the rate of subsequent CSK resolidification (2, 6). Failure of the CSK to fluidize, in turns, explains the inability to dilate the asthmatic airway with a deep inspiration.
Eta has not lived up to billing? Hogwash!
REFERENCES
University of Szeged
To the Editor: After reading the Point:Counterpoint debate (1, 5) with great interest, I would like to comment on the question of the respiratory impedance parameters, rather than the Penh issue, where there is virtually full consensus between the debaters and presumably every physiologist.
Dr. Mitzner describes the amazingly simple structural background of the airway resistance (5), whereas in fact total respiratory (Rrs) or lung resistance (RL) is determined in animal experiments, usually together with the other conventional measure, elastance. I take strong objection to the tacit use of Rrs or RL as synonyms for airway resistance, as implied here (5). Substantial current evidence indicates that, at breathing frequencies, the tissue resistance contributes equally to Rrs in many species, including mice (e.g., Ref. 6). The best argument is nevertheless from Arthur B. DuBois (2): "...The pressures measured are in overcoming airway resistance and tissue viscance, including transverse movement between regions of unequal phase of motion..." Here we realize that the resistance of the respiratory tissues was considered a significant component of Rrs as early as 1953, although its non-Newtonian, frequency-dependent nature (3) was not as yet revealed. Perhaps equally important, there is a notion here that peripheral inhomogeneity adds another component to the viscous losses; when these facts are taken together, and relying on different validation studies of the constant-phase model (4, 6), just what the tissue damping coefficient G reflects and how its determination helps in understanding the changes in the peripheral lung appear quite intuitive.
REFERENCES
National Institute of Environmental Health Sciences
National Institutes of Health
To the Editor: The opinions expressed by Bates (1) and Mitzner (5) on the usefulness of lung impedance versus simpler measurements of lung function in animal models of pulmonary disease raise a number of interesting points and it is likely that one's opinion on this subject will depend largely on one's definition of "useful." It is well recognized that beneficial and restrictive qualities exist for each type of measurement discussed and that most computerized ventilators now offer software packages that enable both to be performed. As such, we suggest that data from lung impedance and simpler measurements of lung function in animal studies of lung disease be presented together whenever possible (2, 3, 4). In doing so, journal reviewers, editors, and the scientific community at large will be provided with all information pertinent to lung function alterations that are associated with molecular, biochemical, pathological, and structural indexes of pulmonary disease in a particular study. While some data may be difficult to interpret or fully understand for some individuals, presentation of all data should encourage collaboration and open discussion of the information. By not choosing one lung function model over another, all potentially relevant data can be collected and made available for assessment by anyone, at any time, regardless of their persuasion. This appears all the more to be a sensible approach considering that even the relevance to human lung disease of some of the most common animal models currently in use (e.g., ovalbumin-induced allergic inflammation) is the subject of intense debate (6).
REFERENCES
Department of Biomedical Engineering
Boston University
To the Editor: Bates (1) and Mitzner (6) argue whether to measure lung impedance over a range of frequencies. More generally, the question is should we probe a system at different scales? Here is a useful analogy. Looking at the parenchyma with light microscope at low magnification, we see large-scale structure and heterogeneity of alveolar dimensions can confirm, for example, the presence of emphysema. However, we do not see what's in the alveolar walls. Let's zoom into a single wall with electron microscope at much smaller scales. We find cells, collagen, etc., which might give us clues about tissue remodeling, but now we will not see enough alveoli to identify emphysema. Thus different mechanisms can manifest themselves at different scales, which is the basis of the successful first order airway-tissue partitioning by the CP tissue model (plus resistance; Ref. 2) that exploits the separation of tissue properties at long times scales (low frequency) and airway resistance at shorter time scales (higher frequency). I further argue that an additional incremental increase in complexity can lead to new science. When we combine alveolar heterogeneity with the CP model (one extra parameter; Ref. 3) and apply it to impedance data from pallid mice, the model predicts collagen-related molecular abnormalities at an early age of 7 weeks (4). This is notable since the pallid mice are thought to develop adult emphysema only around 1 year of age (6).
The lesson is this: If we limit ourselves to a territory too familiar, then we might ask, "are we really doing new science?"
REFERENCES
Boston University
To the Editor: "Impedance" measurements are simply an efficient technique to estimate total lung RL and EL, perhaps even at several different frequencies simultaneously. Explicitly and unequivocally, RL governs the amount of energy dissipated and EL the amount of energy stored during oscillations at a particular frequency. Now Dr. Mitzner then properly asks a far more challenging question: What are the structural origins that can lead to RL and EL at any particular frequency or over a frequency range? Using well established experimental and modeling research we now know much of the spectrum of potential answers all of which motivate the measurement of RL and EL (2) The exciting challenge of trying to distinguish which specific structural conditions are most relevant is not a reason to run for the hills.
I conjecture Dr. Mitzner's main bone to pick is use of the constant phase model to interpret impedance and he has my sympathy: here, the G wants to have its cake and eat it to. The "G" wants to be embraced as a universal property fundamentally linked to tissue viscoelasticity. But, during disease G loses both physical meaning and anatomic correlation leading to misuse (4). In contrast, RL and EL explicitly reflect the complexity of lung structure on the effective energy dissipation and storage, respectively, at a particular frequency. The mistake often made is overly interpreting these properties by explicitly linking them to a particular lung anatomy (e. g., RL with airways and EL with tissues). This is a mistake grounded in lack of understanding rather than application.
REFERENCES
The Johns Hopkins University
To the Editor: Drs. Mitzner (6) and Bates (1) are to be commended for their spirited discussion on the usefulness of lung impedance measurements as compared with "simpler" measurements of lung function, such as airway resistance and lung elastance. However, the debate seems to have been sidetracked in two respects. First, "simpler" appears to be in the eye of the beholder. Does it refer to simplicity in concept, measurement, or interpretation? Dr. Mitzner's idea of resistance as the ideal "simple" parameter neglects not only that it is relatively difficult to measure, but that its interpretation as having a "direct link to airway size" presumes a model of the lung that is simplified beyond usefulness. To which airways of an actual diseased lung does this parameter refer? Second, as this debate illustrates, the utility of the forced oscillation technique is frequently confused with the merits of the constant phase model. Since its initial use to characterize the oscillatory mechanics of healthy canine lungs (2), it has been applied across several species and pathologic conditions for which it has never been validated.
But more to the issue at hand, the impedance spectrum of the lung has been shown to be exquisitely sensitive to many different mechanical derangements of pulmonary function (3, 4), particularly with respect to the heterogeneity of distributed mechanical properties (5) and independent of any model that may be conceived to describe oscillatory behavior. That the impedance spectrum can be reduced to generate the "simpler" measures of resistance and elastance should be reassuring. As such, it complements these simpler measurements of lung function.
REFERENCES
University of Vermont
To the Editor: First of all I concur with Drs. Bates and Mitzner in a resounding rejection of the use of Penh (1–3). Just as an example; in a recent abstract, mice treated with alcohol had reduced airway hyperresponsiveness tested with Penh (4), but the authors completely missed that alcohol affects the CNS and the breathing pattern. While some techniques may serve as screening tools, it does not take away the responsibility of the scientist to understand the limitation of a technique.
The ongoing debate on whether measuring lung impedance is more useful than simpler measures of respiratory mechanics leaves me somewhat bemused (3). The relationship between impedance and the internal structures of the system is not immediately obvious. Our understanding of the respiratory system is based on our own conceptualization of the system in terms of a mathematical model of the system. If the model fits the measured parameters reasonably well, then we take the model as a useful reflection of the real system. We know that the models we use are simple. Of course we will be better off with a more comprehensive model of the lung because it will afford a more detailed understanding of how the lung works. Whether one likes it (1) or not (3), the scientific community will persist to understand physiological phenomena in ever finer details and while there is a risk that we at some point will not see the lung for the airway tree, it is the call of the scientist to use and interpret the appropriate model.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
