Airway and tissue constrictions are greater in closed than in open-chest conditions.

We measured lung impedance (ZL) before and after four doses of methacholine (Mch) infusion in five intact chest (with esophageal balloon) and six open-chest dogs from 0.2 to 8 Hz with an optimal ventilator waveform. From ZL, we estimated airway resistance (R(aw)) and inertance (Iaw) and tissue viscance (GL) and elastance (HL). Two-way analysis of variance revealed that: (1) Mch had a strong influence on all parameters (p < 0.001), but small effect on hysteresivity, nL = GL/HL; (2) closed-chest GL and HL were significantly higher and Iaw lower than their open-chest values (p < 0.002, p < 0.05 and p < 0.0001); and (3) at the highest Mch dose, the relative increase in R(aw) was six times higher in the closed-chest condition. The reduced impact of Mch on open-chest mechanics may be due to constrictions superimposed on grossly different lung configurations and/or some humoral effects initiated by the thoracotomy. We conclude that Mch doses that elicit mild constriction in open-chest condition can cause a severe constriction in intact animals.

Differential responses of global airway, terminal airway, and tissue impedances to histamine.

The forced oscillation and alveolar capsule techniques were applied to determine the input impedance of the lungs and the airway transfer impedances between 0.2 and 20 Hz in six open-chest dogs in the control state, during intravenous infusion of histamine at seven rates between 0.25 and 16 micrograms.kg-1.min-1, and after the infusion. In each condition, the input impedances seen from the alveolar capsules, i.e., terminal airway impedance (Zaw,ter), were measured by imposing 2- to 200-Hz oscillations from the capsules (B. L. K. Davey and J. H. T. Bates. Respir. Physiol. 91:165-182, 1993). Airway resistance (Raw) and inertance and tissue damping and elastance were derived from the lung impedance data. For all dogs, histamine progressively increased Raw and the real part of airway transfer impedance (airway transfer resistance), reaching, at 16 micrograms.kg-1.min-1, 241 +/- 109 (SD) and 370 +/- 186%, respectively, of the control value but caused greater, although locally highly variable, increases (769 +/- 716% of the control value) in the real part of Zaw,ter extrapolated to zero frequency (R0). With increasing doses of histamine, the changes in R0 always preceded those in Raw and airway transfer resistance implying that bronchoconstriction developed first in the lung periphery. It is therefore concluded that the measurement of Zaw,ter offers a sensitive method for the detection of early nonuniform responses to bronchoconstrictor stimuli that are not yet reflected by the values of the overall Raw. In one-half of the cases, significant increases in tissue damping and elastance occurred before any change in R0; this suggests that the mechanisms of airway and parenchymal constrictions may be unrelated.

Airway and tissue mechanics during physiological breathing and bronchoconstriction in dogs.

In five open-chest dogs and with four to five alveolar capsules we used an optimal ventilator waveform (OVW) to follow frequency and tidal volume (VT) dependence of lung, airway, and tissue resistance (R) and elastance (E) before and during constant infusion of histamine (16 micrograms.kg-1.min-1). OVW contains sufficient flow energy between 0.234 and 4.7 Hz, avoids nonlinear harmonic interactions, and simultaneously ventilates with physiological VT. Each OVW breath permits a smooth estimate of frequency dependence of R and E for the whole lung. A constant-phase model analysis provided estimates of purely viscous resistance (Rvis), which represents the sum of airway resistance (Raw) and any purely newtonian component of tissue resistance (Rti), and parameters G and H, which govern frequency dependence of Rti and tissue elastance (Eti), respectively. Tissue structural damping (eta) is calculated as G/H. This model was applied to the whole lung and tissue impedance as estimated from each capsule. We found a small but inconsequential purely newtonian component of Rti, even during constriction. Four dogs showed a peak response at approximately 4 min in lung Rvis coupled (in time) to initial increases in G, H, eta, and airway inhomogeneities. In two of these dogs the response was severe. Tissue properties estimated from whole lung impedance (G, H, and eta) were nearly identical to values estimated from unobstructed capsules throughout infusion. By using a technique independent of alveolar capsules, our results indicate that a major if not dominant response to a constrictive agonist occurs in lung tissues, resulting in a large increase in Rti and Eti. With severe constriction, significant increases occur in Raw and airway inhomogeneities as well. Finally, separation of airway and tissue properties using input impedance estimated from the frequency-rich OVW avoids use of alveolar capsules and may prove an effective tool for partitioning airway and tissue properties in humans.

The measurement of cardiac output in dogs by impedance cardiography with different electrode arrangements.

This study was performed to compare cardiac output (CO) values determined by means of impedance cardiography (ICG) with the conventional four-band electrode array and with different spot electrode arrays in anaesthetised dogs. CO values determined at end-expiratory apnoea with hand-calculation (ICG1) and during several respiratory cycles with a computer program (ICG2) were compared with values obtained via simultaneous thermodilution (TD) measurements. Changes in CO during isoproterenol infusion, bleeding and reinfusion were also studied by means of ICG1, using one of the spot electrode arrays and TD. Band voltage electrodes yielded a significantly lower CO, whereas spot electrodes on the left thorax gave a significantly higher CO than that measured with TD. In spite of the high correlation coefficients in the different electrodes arrays, the bias between ICG1 and TD, and that between ICG2 and TD CO in the SL1-SL8 and BN-BX electrode arrays showed differences statistically significant from zero. The percentage changes in CO measured with ICG1 in the SN-SX electrode array and TD during isoproterenol infusion, bleeding and reinfusion also showed a high correlation. These results indicate that band voltage electrodes can be replaced by the more convenient spot electrodes in certain arrays. Further CO may be measured during several respiratory cycles by using a computer program. Thus, ICG is a reproducible, non-invasive method for the measurement of CO in anaesthetised dogs.

Partitioning of pulmonary impedance: modeling vs. alveolar capsule approach.

Pulmonary input impedance (ZL), transfer tissue impedances (Ztti), and transfer airway impedances (Ztaw) were measured in open-chest dogs and isolated canine lungs by means of small-amplitude pseudorandom oscillations between 0.2 and 21.1 Hz. In the determination of Ztti and Ztaw, local alveolar pressures (PA) sensed in alveolar capsules were used. The global impedances of the airways (Zaw) and tissues (Zti) were estimated by fitting to the ZL data between 0.2 and 4.9 Hz (open-chest dogs) and between 0.2 and 5.9 Hz (isolated lungs) two models based on Hildebrandt's formulations (Bull. Math. Biophys. 31: 651-667, 1969), the parameters of which included airway resistance (Raw) and inertance (Iaw) and tissue damping (GL) and elastance (HL). The tissue parameters of Ztti (Gti and Hti) were also obtained from model fitting, whereas the Ztaw data were evaluated in terms of resistance (Rtaw) and inertance (Itaw). Excellent agreement was found between HL and Hti in both experimental groups and between GL and Gti in the isolated lungs (r > or = 0.999). The damping coefficients were also closely related in the open-chest dogs (r = 0.95), but Gti overestimated GL slightly (by 9%). Raw was underestimated by Rtaw (by 3-33%) and Iaw by Itaw (by 2-16%), depending on the model type and, in the excised lungs, the number of punctures in the capsules. In the case of the airway parameters, the systematic differences were accompanied by lower r values (0.535-0.935), which are explained primarily by the regional variations in PA.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanical impedances of lungs and chest wall in the cat.

In nine anesthetized and paralyzed cats, the mechanical impedances of the total respiratory system (Zrs) and the lungs (ZL) were measured with small-volume pseudorandom forced oscillations between 0.2 and 20 Hz. ZL was measured after thoracotomy, and chest wall impedance (Zw) was calculated as Zw = Zrs-ZL. All impedances were determined by using input airflow [input impedance (Zi)] and output flow measured with a body box [transfer impedance (Zt)]. The differences between Zi and Zt were small for Zrs and negligible for ZL. At 0.2 Hz, the real and imaginary parts of ZL amounted to 33 +/- 4 and 35 +/- 3% (SD), respectively, of Zrs. Up to 8 Hz, all impedances were consistent with a model containing a frequency-independent resistance and inertance and a constant-phase tissue part (G-jH)/omega alpha, where G and H are coefficients for damping and elastance, respectively, omega is angular frequency, and alpha determines the frequency dependence of the real and imaginary parts. G/H was higher for Zw than for ZL (0.29 +/- 0.05 vs. 0.22 +/- 0.04, P less than 0.01). In four cats, the amplitude dependence of impedances was studied: between oscillation volumes of 0.8 and 3 ml, GL, HL, Gw, and Hw decreased on average by 3, 9, 26, and 29%, respectively, whereas the change in G/H was small for both ZL (7%) and Zw (-4%). The values of H were two to three times higher than the quasistatic elastances estimated with greater volume changes (greater than 20 ml).

Input impedance and peripheral inhomogeneity of dog lungs.

Tracheal pressure, central airflow, and alveolar capsule pressures in cardiac lobes were measured in open-chest dogs during 0.1- to 20-Hz pseudorandom forced oscillations applied at the airway opening. In the interval 0.1-4.15 Hz, the input impedance data were fitted by four-parameter models including frequency-independent airway resistance and inertance and tissue parts featuring a marked negative frequency dependence of resistance and a slight elevation of elastance with frequency. The models gave good fits both in the control state and during histamine infusion. At the same time, the regional transfer impedances (alveolar pressure-to-central airflow ratios) showed intralobar and interlobar variabilities of similar degrees, which increased with frequency and were exaggerated during histamine infusion. Results of simulation studies based on a lung model consisting of a central airway and a number of peripheral units with airway and tissue parameters that were given independent wide distributions were in agreement with the experimental findings and showed that even an extremely inhomogeneous lung structure can produce virtually homogeneous mechanical behavior at the input.

A comparison of interrupter and forced oscillation measurements of respiratory resistance in the dog.

We compared the values of resistance produced by the forced oscillation technique (FOT) and the flow interruption technique (IT) when applied to six anesthetized paralyzed tracheostomized dogs. The FOT returned values of respiratory system resistance as a function of frequency [Re(f)] between 0.25 and 20 Hz. The IT returned a single value of resistance (Rinit) calculated by dividing the immediate change in tracheal pressure occurring upon interruption by the preinterruption flow. We found Rinit to coincide closely with Re(f) in the frequency range 5-20 Hz. Rinit has previously been interpreted as the high-frequency resistance of a resistance-elastance model of the respiratory system airways and tissues. It has also been shown previously, by direct measurement of alveolar pressure in dogs, that Rinit from the lungs alone is an accurate measure of airways resistance while Rinit obtained from the total respiratory system equals airways resistance plus a modest contribution from the chest wall. Re(f) at a frequency of approximately 10 Hz thus appears to be a useful quantity to measure as an index of airways resistance in the dog.

Nonlinearity and harmonic distortion of dog lungs measured by low-frequency forced oscillations.

The nonlinearity of lung tissues and airways was studied in six anesthetized and paralyzed open-chest dogs by means of 0.1-Hz sinusoidal volume forcing at mean transpulmonary pressures (Ptp) of 5 and 10 cmH2O. Lung resistance (RL) and elastance (EL) were determined in a 32-fold range (15-460 ml) of tidal volume (VT), both by means of spectrum analysis at the fundamental frequency and with conventional time-domain techniques. Alveolar capsules were used to separate the tissue and airway properties. A very small amplitude dependence was found: with increasing VT, the frequency-domain estimates of RL decreased by 5.3 and 14%, whereas EL decreased by 20 and 22% at Ptp = 5 and 10 cmH2O, respectively. The VT dependences of the time-domain estimates of RL were higher: 10.5 and 20% at Ptp = 5 and 10 cmH2O, respectively, whereas EL remained the same. The airway resistance increased moderately with flow amplitude and was smaller at the higher Ptp level. Analysis of the harmonic distortions of airway opening pressure and the alveolar pressures indicated that nonlinear harmonic production is moderate even at the highest VT and that VT dependence is homogeneous throughout the tissues. In three other dogs it was demonstrated that VT dependences of RL and EL were similar in situ and in isolated lungs at both Ptp levels.

Mechanical impedance of the canine diaphragm. Part 1. Experimental system and measurements.

A technique which does not require the measurement of strain has been developed for the investigation of the incremental dynamic properties of soft tissue sheets. Radially prestressed and circularly clamped canine diaphragm samples were exposed to small-amplitude pseudorandom pressure variations. From the measurement of these pressure variations and the volume flow caused by the vibration of the membrane the incremental mechanical impedance spectrum was computed in the 0.25-5 Hz frequency range at three different levels of initial stress. The diaphragm tissue was found to be basically elastic. However, the small viscous component showed a sharp negative frequency dependence between 0.25 and 2 Hz. The quasistatic elastances of the samples were in good agreement with the elastance values derived from the impedance data. The relationship between the elastance and the initial stress was close to linear. It was concluded that the method is applicable to the study of the incremental dynamic properties of planar soft tissue samples.

Mechanical impedance of the canine diaphragm. Part 2. Theoretical model and parameter estimation.

In the paper the equation of motion of the small amplitude transverse forced vibration of a radially prestressed and circularly clamped thin membrane has been developed. The material of the membrane is considered to be homogeneous, isotropic, incompressible and viscoelastic. From the analytical solution of this equation the incremental mechanical impedance of the membrane was derived as a function of frequency, geometrical parameters and incremental viscoelastic coefficients of the material. The parameters of the model were fitted to experimental impedance data using a global optimisation procedure to obtain the incremental viscoelastic moduli of the canine diaphragm. The estimated quasi-static behaviour of the model is shown to be consistent with the results of experimental quasi-static measurements. It is concluded that the incremental viscoelastic moduli of a soft tissue and the stress dependence of these material coefficients can be determined by fitting the parameters of the model to the impedance data of that particular tissue.

Modeling of low-frequency pulmonary impedance in dogs.

The mechanical impedance of the lungs (ZL) was measured in open-chest dogs with small-amplitude pseudorandom volume oscillations between 0.125 and 5 Hz, at mean transpulmonary pressures (Ptp) of 0.2, 0.4, and 0.8 kPa. At the lowest frequencies, the pulmonary resistance showed a marked negative frequency dependence and mirrored the changes in the reactance with altered Ptp. The ZL data were evaluated on the basis of two models, each containing the same airway compartment with a resistance and an inertance. The tissue impedance (Zti) in model 1 was represented with two compliances and a resistance (L. E. Mount. J. Physiol. Lond. 127: 157-167, 1955), whereas in model 2 a two-parameter formulation implying rate-independent dissipated work and frequency-dependent elastance (J. Hildebrandt. J. Appl. Physiol. 28: 365-372, 1970) was employed. The estimation of model parameters showed that model 2 was superior to model 1 in both fitting performance and parameter insensitivity to weighting in the fitting criterion. The model 2 coefficients of damping and elastance, characterizing the real and imaginary parts of Zti, respectively, depended on the lung distension and were closely correlated. Although ZL exhibited a slight dependence on the peak-to-peak volume excursion, at a given oscillatory volume no inconsistency with linear tissue viscoelasticity was detected.

Generation of optimum pseudorandom signals for respiratory impedance measurements.

Spontaneous breathing may impair the reliability of forced oscillatory impedance estimates at low frequencies, especially when the oscillatory power is distributed among many frequency values. Since the amplitude of the external forcing is limited to avoid non-linearities, it is suggested that the total energy of a composite electrical signal driving the loudspeaker be maximized at a given amplitude by finding the optimum phase relationships of the signal components, and that the low-frequency components increase in energy at the expense of the less disturbed high-frequency region. In healthy children and adults and in obstructed patients, the coherences and the coefficients of variation of the respiratory system impedance (Zrs) at 2 and 3 Hz were studied in the case of three test signals of 2-15 Hz bandwidth. Signals T1 and T2 had a flat power spectrum, whereas the components of T3 decreased sharply between 2 and 5 Hz; T1 was generated by simple random selection of phase angles, while optimization for maximum energy was done for T2 and T3. Optimization alone (T2) increased the reliability of the Zrs estimates at all frequencies, whereas enhancement of the low-frequency power (T3) resulted in a radical improvement of the estimates at 2 and 3 Hz, without loss in reliability at higher frequencies.

Low-frequency respiratory mechanical impedance in the rat.

A modified forced oscillatory technique was used to determine the respiratory mechanical impedances in anesthetized, paralyzed rats between 0.25 and 10 Hz. From the total respiratory (Zrs) and pulmonary impedance (ZL), measured with pseudorandom oscillations applied at the airway opening before and after thoracotomy, respectively, the chest wall impedance (ZW) was calculated as ZW = Zrs - ZL. The pulmonary (RL) and chest wall resistances were both markedly frequency dependent: between 0.25 and 2 Hz they contributed equally to the total resistance falling from 81.4 +/- 18.3 (SD) at 0.25 Hz to 27.1 +/- 1.7 kPa.l-1 X s at 2 Hz. The pulmonary compliance (CL) decreased mildly, from 2.78 +/- 0.44 at 0.25 Hz to 2.36 +/- 0.39 ml/kPa at 2 Hz, and then increased at higher frequencies, whereas the chest wall compliance declined monotonously from 4.19 +/- 0.88 at 0.25 Hz to 1.93 +/- 0.14 ml/kPa at 10 Hz. Although the frequency dependence of ZW can be interpreted on the basis of parallel inhomogeneities alone, the sharp fall in RL together with the relatively constant CL suggests that at low frequencies significant losses are imposed by the non-Newtonian resistive properties of the lung tissue.

Respiratory mechanical impedance in the rat.

The forced oscillatory impedance of the total respiratory system (Zrs) was measured in seven anaesthetized, paralysed rats weighing 351 +/- 55 g. Tracheotomy was performed, and the animals were placed in the supine position in a body box. Pseudo-random pressure variations between 0.5 and 10 Hz were applied around the chest. Central airflow was measured with a heated screen pneumotachograph. Total respiratory resistance (Rrs) and elastance (Ers), corrected for the impedance of the tracheal cannula, were markedly frequency-dependent: Rrs fell from 37.3 +/- 19.1 kPa.l-1.s at 0.5 Hz to 17.6 +/- 4.4 at 2 Hz and 10.3 +/- 3.3 at 10 Hz; the corresponding Ers values were 453 +/- 14, 594 +/- 90 and 713 +/- 104 kPa.l-1, respectively. This indicates that in the frequency range encompassing spontaneous breathing rates the classical resistance-intertance-compliance model provides an inadequate description of the respiratory mechanics in the rat.

Forced oscillatory impedance of the respiratory system at low frequencies.

Respiratory mechanical impedances were determined during voluntary apnea in five healthy subjects, by means of 0.25- to 5-Hz pseudo/random oscillations applied at the mouth. The total respiratory impedance was partitioned into pulmonary (ZL) and chest wall components with the esophageal balloon technique; corrections were made for the upper airway shunt impedance and the compressibility of alveolar gas. Neglect of these shunt effects did not qualitatively alter the frequency dependence of impedances but led to underestimations in impedance, especially in the chest wall resistance (Rw), which decreased by 20-30% at higher frequencies. The total resistance (Rrs) was markedly frequency dependent, falling from 0.47 +/- 0.06 (SD) at 0.25 Hz to 0.17 +/- 0.01 at 1 Hz and 0.15 +/- 0.01 kPa X l-1 X s at 5 Hz. The changes in Rrs were caused by the frequency dependence of Rw almost exclusively between 0.25 and 2 Hz and in most part between 2 and 5 Hz. The effective total respiratory (Crs,e) and pulmonary compliance were computed with corrections for pulmonary inertance derived from three- and five-parameter model fittings of ZL. Crs,e decreased from the static value (1.03 +/- 0.18 l X kPa-1) to a level of approximately 0.35 l X kPa-1 at 2-3 Hz; this change was primarily caused by the frequency-dependent behavior of chest wall compliance.

Total respiratory impedance in healthy children.

Impedance of the total respiratory system was measured in 121 healthy children aged 4-16 years during spontaneous breathing by pseudo-random forced oscillations between 3 and 10 Hz. Total respiratory resistance (Rrs), inertance (Irs) and compliance (Crs) were determined by least-mean-squares fitting. Estimates for inertance were reliable only for the larger children, where the values of Irs (0.0127 +/- 0.0034 SD) were similar to those reported for normal adults. Rrs correlated significantly (P less than 0.001) with height (r = -0.868), age (r = -0.865), and, in a subpopulation of the 6- to 16-year-old children, with forced vital capacity (r = -0.803). The corresponding correlation coefficients for Crs were 0.873, 0.844, and 0.853, respectively. Crs amounted to about a third of the static total compliance values of Sharp et al. (J Appl Physiol 1970; 29: 775-779) over the same interval of heights. In these relationships no significant difference was found between boys and girls.

An improved forced oscillatory estimation of respiratory impedance.

A new scheme of estimation is proposed for the determination of the total respiratory impedance (Zr) by forced oscillations. Instead of estimating Zr from the airflow at the mouth and the pressure drop across the respiratory system, Zr can be estimated from cross-power spectra between the electrical signal of the driving apparatus and the measured pressure and flow signals. since the electrical signal is free from any disturbance originating from spontaneous breathing, unlike the conventional techniques the proposed 'indirect' estimation of Zr is not subject to systematic errors caused by the respiratory signal components. This is confirmed both by theoretical analysis and simulation of the forced oscillatory measurement.

Parameter estimation of transpulmonary mechanics by a nonlinear inertive model.

Transpulmonary mechanics of anesthetized intubated dogs were studied during control breathing and hemorrhage-induced hyperventilation by least-mean-squares parameter estimation using several model versions. The classical elastance-resistance model was modified to include nonlinear elastic and viscous pressure terms with and without a linear inertive pressure component. Inclusion of the nonlinear terms decreased the root-mean-square error of fitting (q) of the classical model on the average to 67% in the control period and to 58% during hyperventilation. An additional decrease due to inertance was 4% (control) and 22% (hyperventilation) and was associated with acceptable estimates of inertance [0.056 +/- 0.02 (SD) and 0.063 +/- 0.008 cmH2O . l-1 . s2, respectively]. When inertance alone was added to the classical model, negligible improvement in q and unrealistic values of inertance were obtained. Conventional measures (Edyn and midvolume resistance) were close to the corresponding least-mean-squares estimates (E and R) of all model versions, except that in hyperventilation neglecting the inertance caused Edyn to markedly overestimate E of nonlinear inertive model.