| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Center for Biomedical Engineering, University of Kentucky, Lexington, KY and 2Department of Physiology, National Taiwan University College of Medicine, Taipei, Taiwan
Submitted 6 July 2007 ; accepted in final form 12 September 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
VL). Total lung volume (VL) described by equation: VL = VL + VFRC = KAL1.5 assumed uniform (isotropic) inflation. Total VFRC averaged 0.55 ml, consisting of 0.10 ml tissue, 0.21 ml blood and 0.24 ml air. K averaged 1.84. In conscious mice in a sealed box, we measured the peak-to-peak box pressure excursions (Pb) and x-rays during several cycles. K was used to convert measured AL1.5 to VL values. We calculated Ve and Vt from the plot of VL vs. cos(
– 
). Phase angle was the minimum point of the Pb cycle to the x-ray exposure. Phase difference between the Pb and VL cycles () was measured from Pb values using both room- and body-temperature humidified box air. A similar analysis was used after aerosol exposures to bronchoconstrictor methacholine (Mch), except that depended also on increased Raw. In conscious mice, Ve (0.24 ml) doubled after Mch (50–125 mg/ml) aerosol exposure with constant Vt, frequency (f), Pb, and Raw. In anesthetized mice, in addition to an increased Ve, repeated 100 mg/ml Mch exposures increased both Pb and Raw and decreased f to apnea in 10 min. Thus conscious mice adapted to Mch by limiting Raw, while anesthesia resulted in airway closure followed by diaphragm fatigue and failure. bronchoconstrictor; methacholine aerosol; body plethysmography
The measurement of Vt in conscious mice is particularly difficult with available computed tomography (CT) and micro-CT imaging technology (15, 28) because of the relatively low temporal resolution with x-ray exposure times, which are too long to obtain lung images at end-inspiration and end-expiration at breathing frequencies approaching 5 breaths/s.
Accordingly, in this study we used a single x-ray pulse of 10 ms to obtain single projection images of the thorax of conscious mice breathing spontaneously in a sealed box. The short exposure time minimized image blur due to ventilatory and cardiogenic motion. In anesthetized mice, we measured lung area from x-ray images obtained at Ve and after imposed increases in lung volume. With the assumption of uniform (isotropic) lung expansion, the mathematical relationship between lung area and lung volume was used to infer lung volume in conscious mice from x-ray images taken at different points during the box pressure cycle. The lung volume based on the lung area was corrected for tissue and blood mass measured in separate experiments. Under control conditions, the phase difference between the box pressure and lung volume cycle was measured from box pressure excursions with both room- and body-temperature humidified box air. A sinusoidal analysis was used to determine lung volume at associated points during the lung volume cycle and to determine Vt and Ve, both for control conditions and after exposure to the bronchoconstrictor methacholine (Mch) aerosol. The results showed that Ve increased in response to Mch aerosol in both conscious and spontaneously breathing anesthetized mice. However, Mch produced an increase in Raw in the anesthetized mice but not in the conscious mice.
Glossary
Ph/Pg), phase angle difference between temperature-humidity and gas compression parts of the box pressure curve
PbPb
T
VL
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
Calibration of Lung Area From X-Ray Images vs. Lung Volume Measured in Anesthetized Mice
BALB/c mice were anesthetized with 50 mg/kg sodium pentobarbital delivered ip. After a tracheostomy, the supine animal was placed on a digital x-ray sensor (4.2 cm x 3.0 cm, Lightyear Technology) and ventilated (5 breaths/s and 0.15 ml tidal volume; Hugo Sachs Elektronik, Harvard Apparatus MiniVent, type 845). Airway pressure (Paw) was measured via a side tube from the tracheal cannula connected to a pressure transducer (Cobe) and a chart recorder (Gould TA2000). With the trachea occluded at end-expiration, vertical x-ray images of the thorax (ventral-dorsal direction) were taken with an x-ray source (10 ms exposure) at functional residual capacity (FRC) at 0 cmH2O Paw and after lung air inflation from FRC with 0.25 ml. The animal was ventilated for 1 min to wash out the CO2 accumulated during the x-ray measurements. Subsequently x-ray images were repeated at FRC and after 0.5 ml lung inflation. This procedure with 0.25- and 0.5-ml inflations was repeated to verify reproducibility. The x-ray source and parameters were as follows: Bowie Manufacturing, model PRX-90T, focal spot 1.8 mm diameter; 80 kV, 15 mA. X-ray source-to-sensor distance was 24 cm and lung-to-sensor distance was 0.5–1 cm. Spatial resolution caused by penumbra blur was 80 µm. Maximum spatial resolution based on a pixel dimension on the computer screen was 20 µm. Correction for magnification was less than 5%.
The areas of the lung and heart outlined on the x-ray images were measured with commercial software (Photoshop and NIH image) on a PC. We used the following criteria. The apical (superior) regions of the lung and heart were bounded by the inferior border of the second rib. The lateral regions of the lungs were bounded by the inner curved outlines of the ribs. The caudal (inferior) regions of the lung were bounded by the superior curved outline of the diaphragm. A 5-mm metal marker was placed on the x-ray sensor to relate number of pixels to area. The resolution of the measurements based on repeated marker length measures was 100 µm (2%). Measurements of the lung and heart areas were repeatable within 4% error.
Conscious Mice Studies
We measured Raw in conscious mice by means of the procedures outlined in a previous study (25), but in addition we measured lung volume using x-ray imaging. To measure lung volume each animal was positioned prone, head first, in a plastic transparent tube (Corning, 50 ml, 2.8 cm diameter and 10 cm length) located on the x-ray sensor (Fig. 1). The head of the animal was located at the cone-shaped far end of the tube with its thorax confined above the x-ray sensor. The plastic tube was connected to a copper tube (4 cm diameter and 19 cm length) to approximate the box volume (240 ml) of the previous study (25). The plastic and copper tubes were interconnected via several holes drilled in the cone end. Airflow of 1–10 l/min through the chamber provided by an aerosolyzer (DeVilbiss, Fig. 1) prevented the accumulation of CO2. As needed, box air CO2 was measured by a CO2 analyzer (Puritan-Bennett) via a length of PE100 tubing with the open tip located 2 cm from the nose of the animal. Temperature and humidity were measured continuously via a probe (Fisher Scientific 11–661-7B) located 2 cm from the nose of the animal. Box air was maintained at 21–24°C and at 100% relative humidity by air flow from the aerosolyzer filled with saline prior to sealing. The box pressure decay time constant after sealing was 
1.5 s, measured by a rapid injection of 0.1 ml air into the box. Box pressure excursions were constant between 2 and 6 Hz with 0.015 ml volume amplitude. After the box was sealed, box pressure excursions were measured by a pressure transducer (Validyne DP 45) for several seconds. During this period, an x-ray image was taken at random with a single x-ray pulse (exposure time of 10 ms, see Fig. 2). Airflow through the box was then restored for 1 min before resealing the box and taking another x-ray. Several (10–20) x-rays were taken randomly at different times along the box pressure cycle. In a subsequent improvement of this technique, we used a specially designed electronic interface (Aircastle Custom Products, Lexington, KY) that triggered via a switch the shutter of the x-ray generator at either the maximum or minimum point of the pressure cycle. This required the manual adjustment of a built-in delay that allowed the activation of the trigger at any assigned point of the pressure cycle. This feature was also used to obtain x-ray images near the midpoints of the pressure excursions, which were required for the studies with body-temperature box air conditions (see Effect of Body Temperature Box Air on Lung Volumes and Box Pressure in Conscious Mice). The time at which the x-ray was taken relative to the box pressure cycle was measured by a signal from the x-ray shutter and recorded on the chart recorder simultaneously with the box pressure excursions (Fig. 2). The peak-to-peak box pressure excursion was used to calculate tidal volume via the method of Drorbaugh and Fenn (13). This calculation required the box air compliance that was determined by imposing a sinusoidal (frequency 5 Hz and amplitude 0.015 ml) change in box volume and measuring the resulting box pressure excursion. A small correction for gas compression effects was made to the calculated tidal volume (Eq. 15 of Ref. 25), as discussed below.
|
|
The correction for the gas compression effect to the calculated tidal volume required knowledge of the relative contributions of the gas compression effect and the temperature-humidity effect to the box pressure excursion. This correction was also needed to determine Vt from x-ray data in mice with increased Raw. The correction was determined by measuring the box pressure excursion first with room-temperature box air, then with body-temperature humidified box air. For the latter measurement, the box air temperature was maintained at body temperature (37–39°C) with a thermostatic controller (Physitemp, TCAT-2AC) that provided heat to the box from an external infrared lamp when the temperature measured by the box thermometer was below 38°C. Supplemental heat to the air within the copper tube was provided by a voltage source to a silicon rubber heater (100 W, c-03125–22, Cole-Parmer) bonded to the outside of the copper tube. Humidity (100%) was provided by airflow from the saline aerosolyzer. The box air was stabilized at body-temperature conditions for 1 h prior to placing the animal in the box. The box was then sealed, and several x-rays were taken over a period of 6 min near the midpoints of the pressure excursions that represented only gas compression effects and airway flow. Accordingly, the midpoints corresponded to maximum and minimum points on the volume cycle that differed in phase by 90° from the flow cycle. By an independent calibration with the CO2 probe, the 6-min period caused an increase in CO2 partial pressure to 16 mmHg (2%) with no consistent change in either the box pressure excursions or breathing frequencies.
Effect of Mch Aerosol on Lung Volume in Conscious Mice
The animal was placed in the box at 21–24°C and 100% relative humidity with an airflow supplied by the saline aerosolyzer. The box was sealed and the box pressure excursions were measured over a period of 6 min and used to calculate tidal volume (13, 25). Several x-ray images of the thorax were taken and used to calculate tidal volume (Vt) and mean lung volume (Vm), as described later. Box pressure and x-ray images were remeasured after the box air was increased to 37–39°C at 100% relative humidity. After a recovery period of several hours, box pressure and x-ray images were measured with room-temperature box air in response to a 1-min aerosolized Mch (50 mg/ml) exposure. An increase in Mch concentration to 125 mg/ml together with repeated exposures for 1–3 min had no consistent change on the box pressure excursions or the calculated Raw.
Effect of Mch Aerosol on Lung Volume in Anesthetized Mice
The foregoing procedures used in conscious mice to measure lung volume at both room- and body-temperature box air conditions and after Mch aerosol exposure at room temperature were repeated in anesthetized spontaneously breathing mice in the prone position. The mice were anesthetized with an ip injection of ketamine (100 mg/kg) and xylazine (8.5 mg/kg) in 0.125 ml saline; this procedure provided anesthesia for 1 h. Under anesthesia, box pressure and x-ray images were measured at room temperature and after box air conditions were increased to 37–39°C and 100% relative humidity. After allowing the animal to recover from the anesthesia overnight, box pressure and x-ray images were measured under anesthesia at room temperature with saline aerosol exposure and after Mch aerosol exposure (100 mg/ml) with a flow rate of 10 l/min. Preliminary studies showed no increase in the box pressure excursions after Mch aerosol exposure of 25–75 mg/ml. After 1 min of 100-mg/ml Mch exposure, the box was sealed, and box pressure was recorded for 1 min. The 1-min, 100-mg/ml Mch exposure was repeated until the box pressure excursions increased. Usually three Mch exposures were required to obtain an increased response in the box pressure. The increase in the box pressure excursions occurred simultaneously with a reduction in breathing frequency, terminating in apnea. During the increased box pressure response, x-ray images were recorded at several minimum and maximum points of sequential box pressure cycles. Post mortem, the lung and heart were removed, separated, and weighed. The lung was dried to a constant weight in an oven at 70°C, and the wet-to-dry weight ratio (W/D) calculated.
Blood and Tissue Mass in the Isolated Collapsed Lung
The equation for Raw (Ref. 25, see Eq. 3 below) requires air volumes at FRC, while the x-ray data provided the total lung volume, which included tissue and blood mass in addition to air volume. Thus, to correct for tissue and blood mass to obtain air volume, we measured tissue mass and blood mass at FRC via two separate experiments. In the first experiment, the blood mass was separated from the (blood-free) tissue mass in the isolated collapsed lung via the following procedure. In the anesthetized mouse, radioactive tracer 125I-albumin (2 x 104 counts/s in 0.05 ml Ringer solution; bovine serum albumin, Perkin Elmer, Boston, MA) was injected into the jugular vein. Prior to use, any unbound 125I was removed by passing tracer through a desalting column (PD-10 desalting column; Amersham Biosciences, Piscataway, NJ; see Ref. 24). One minute after injection, a 0.4-ml sample of blood was withdrawn, and its specific radioactivity (counts/s per g) was measured in a gamma counter (WIZARD 1470, Perkin-Elmer, Billerica, MA). Post mortem, the lung was isolated and weighed, and its total radioactivity was measured. The trapped blood mass was calculated by dividing the total radioactivity in the isolated lung by the specific radioactivity of the blood. Tissue mass was the difference between the mass of the isolated lung and the trapped blood mass. The density of both blood and tissue was assumed to be 1 g/ml.
Lung Blood Mass in Spontaneously Breathing Anesthetized Mice at FRC
In the second experiment, we measured the blood mass in anesthetized mice via a procedure similar to that used previously in unanesthetized mice (16). In the control anesthetized mouse, one minute after tracer 125I-albumin was injected into the jugular vein, the mouse was rapidly euthanized by immersion in liquid N2 and placed overnight in a freezer (–20°C). Then the frozen mouse was transected normal to the cranial-caudal axis at the level of the diaphragm, and the thorax was sliced into three transverse 5-mm-thick sections. Frozen samples of the lung were removed from the sections and weighed, and their radioactivity was measured. Frozen samples of whole blood were removed from the heart and weighed, and their specific radioactivity was measured. Blood mass in the samples of lung as a fraction of blood-free tissue mass was calculated. Blood mass of the entire lung was calculated with the mean tissue mass measured in the isolated collapsed lung. In a separate group, we repeated the foregoing experiments in anesthetized mice subjected to three 1-min exposures to the 100-mg/ml Mch aerosol needed to cause an increased Raw. The Mch aerosol-induced increase in box pressure response and decrease in f were measured prior to immersion in liquid N2.
Statistics
Data are reported as mean values ± SD. We used paired t- and unmatched t-tests where appropriate to evaluate significant differences between two groups of data. We used P < 0.05 to be significant.
|
RESULTS: THEORY AND ANALYSIS OF DATA |
|---|
|
TOPABSTRACTMETHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
Table 1 (mean ± SD, n = 5) summarizes results of (blood-free) tissue and trapped blood mass measured in the isolated collapsed lung. Tissue mass averaged 0.10 g, and the trapped blood averaged 0.025 g. Table 2 summarizes the blood mass measured in the lungs of anesthetized spontaneously breathing mice. Lung blood mass averaged 0.21 g for the control mice and was reduced by 43% to 0.12 g after bronchoconstriction with three 1-min exposures to 100 mg/ml Mch aerosol. The reduced blood volume in anesthetized mice was attributed to an increased Ve that dominated any Mch-induced vascular dilation (1). The values of tissue and blood mass were subtracted from the total lung volume measured via the x-ray measurements of lung area to obtain the lung air volume at FRC in conscious and anesthetized mice (see Calibration of X-ray Lung Area to Total Lung Volume With Anesthetized Mice).
|
|
We assumed that the lung within the thorax was uniformly inflated so that total lung volume (VL) was proportional to AL1.5, where AL is area of the lung outline that includes the heart minus the area of the heart outline measured from each x-ray image (Fig. 3):
 | (1) |
|
An estimate of the total lung volume at FRC (VFRC) was obtained by the linear regression of the imposed air volume increments from FRC (VL) vs. the measured AL1.5 values with the following equation:
 | (2) |
VL vs. AL1.5 for five anesthetized mice: VL = 1.84AL1.5 – 0.55, (R2 = 0.75, n = 15, P < 10–5). The pooled data produced a K of 1.84 and VFRC of 0.55 ml. Analysis of each mouse separately produced mean values of K and VFRC of 1.90 ± 0.56 (SD, n = 5) and 0.58 ± 0.20 ml, respectively. In the control anesthetized mice, air volume at FRC (Ve) was 0.24–0.27 ml, after correction for tissue and blood mass.
|
|
The method used to evaluate Raw by means of body plethysmography requires the box pressure excursion, lung tidal volume, and end-expiratory lung air volume under both room- and body-temperature box air conditions. A detailed description has been published (25). In brief, for an animal breathing spontaneously in a sealed box, Raw is given by the following equation:
 | (3) |
Pb in terms of amplitudes Pg and Ph:
 | (4) |
 | (5) |
 | (6) |
Pb was calculated as half of Pb (the peak-to-peak box pressure excursion), and Abt was the average of the areas under the inspiratory and expiratory parts of the Pb vs. t curve, given by Pb/(
f) for a sine wave with frequency f.
|
Drorbaugh and Fenn (13) showed that the inspired gas volume dVL is proportional to dPh, so that the phase relationships of Pb and Pg to Ph (Eqs. 5 and 6) apply to VL. In the experiments, VL was measured at different points along the Pb vs. t curve and corrected for the phase angle to determine the VL values at end-inspiration and end-expiration as well as Vt (Fig. 7). For control (unconstricted) conditions with subscript 1, 1 was calculated by means of Eq. 6 after determining Ph1/Pg1 from the following equation (Eq. 15 of Ref. 25):
 | (7) |
Pb1 and Pg1 were the box pressure amplitudes with room- and body-temperature humidified box air conditions, respectively. The calculation of 1 for control conditions assumed that for constant Raw, Pg1 and Ph1 (that is, Vt1) at room temperature did not change with body-temperature box air conditions. These assumptions were verified experimentally (see Effect of Body-Temperature Humidified Box Air Conditions on Pg1, Vt, Ve, and f.).
|
To evaluate the x-ray data, we assumed that the cyclic variations of box pressure Pb and inspired volume were sinusoidal, and the total lung volume (VL) analogous to alveolar air volume was represented by the following equation (Fig. 7, cf. Eq. A3 of Ref. 25):
 | (8) |
of 40°, Vt of 0.2 ml, and VLm of 0.4 ml (Eq. 8).
In the experiments we collected x-ray images at different time points along several box pressure cycles. An example of an x-ray pulse recorded during a pressure cycle is shown in Fig. 2. The time (tp) from the start (minimum point) of the cycle (phase angle 1 = 0°) to the time of x-ray exposure was measured for each x-ray image and converted to a phase angle (1 = 2ftp). 1 varied from 0° (cos 0° = 1), to 180° (cos 180° = –1) at the maximum box pressure, to 360° (cos 360° = 1) at the end of the cycle. Values of AL1.5 for the x-ray images were converted to VL values with K of 1.84 (Eq. 1), the value measured in anesthetized mice (Fig. 4). The linear regression of VL1 vs. cos (1 – 1) values provided the tidal volume Vt1 as twice the slope magnitude and VLm1 as the intercept (Eq. 8). In the event that VL1 was measured at the maximum and minimum box pressures (VL1max and VL1min), VLm1 and Vt1 applying Eq. 8 were given by:
 | (9) |
 | (10) |
1 affected only Vt1 and not VLm1.
Figure 8 is an example of VL1 vs. cos (1 – 1) values measured in one animal. The regression equation (Eq. 8) was: VL1 = 0.63 – 0.094 cos(1 – 1) (R2 = 0.90, n = 8, P < 0.005). Here the measured value of 1 was 21°, Vt1 was 0.19 ml, and VLm1 was 0.63 ml. Table 3 summarizes Vt1 values (0.19 ± 0.053 ml) calculated with the corrected box pressure excursions by means of the method of Drorbaugh and Fenn (13), and Vt1 values (0.21 ± 0.051 ml) from the x-ray measurements of lung volume from five conscious mice under control conditions. The Vt1 calculated with the measured box pressure excursion was multiplied by cos 1 to give the corrected Vt1 (see Eq. 6 and Fig. 6). In both conscious and anesthetized control mice, 1 was small (20–24o, Tables 4–6) with cos 1 of 0.93. Thus the uncorrected Vt1 values obtained by neglecting 1 were overestimated by 7% with the box pressure excursion and underestimated by 7% with the x-ray measurements. There was no significant difference between the two sets of corrected Vt1 values (P > 0.05). This justified the representation of the measured box pressure curves as sinusoids and the use of the K value measured in anesthetized mice. Total VLm1 from the x-ray data (0.59 ± 0.033 ml, Table 3) produced a Vm value of 0.28 ml and Ve of 0.18 ml after correction for tissue and blood mass.
|
|
|
|
|
Pg1, Vt, Ve, and f
For control conditions, the foregoing corrections for Vt1 based on 1 and Ph1/Pg1 used the assumption that both Raw and Pg1 did not change with body-temperature humidified box air conditions (Eq. 7, Ref. 25). The constant Raw with body-temperature box air conditions was substantiated previously in anesthetized mice (25). With body-temperature humidified box air, VLm and Vt were calculated from the measured x-ray areas taken at the midpoints of the box pressure excursions, representing Pg1 in phase with flow and thus end-inspiratory and end-expiratory volumes. The results are summarized in Table 4. Ratios of each parameter measured with body-temperature box air conditions to that measured with room-temperature box air were tested vs. 1 to evaluate any significant change. Note that body-temperature box air conditions had no significant effect on Vt, Ve, or f, indicating no change in the flow amplitude (Q = fVt); and with a constant Raw, Pg1 equal to RawQ did not change compared with the room-temperature box air conditions. By contrast, the evaluation of Vt from the x-ray data with bronchoconstriction depended on any change in Pg2 from Pg1, which is treated in the following section.
Determination of Vt and Ve After Bronchoconstriction Due to Mch Aerosol Exposure
The determination of tidal volume (Vt2) for the bronchoconstricted animals required the simultaneous solution of both Vt2 and Pg2, the amplitude of the gas compression part of the box pressure curve that determines the increased Raw. Subscripts 1 and 2 refer to control and constricted conditions, respectively. We used the following procedure, analogous to that used for the control animals with the appropriate modifications.
Unique solutions for Vt2/Vt1 (that is, Ph2/Ph1) and Pg2/Pg1 were obtained from the experimental data as follows. First, starting with a trial value of Vt2/Vt1 (e.g., 1) and Ph1/Pg1 known from control conditions, Pg2/Pg1 was calculated by means of the following equation (Eq. 19 of Ref. 25), obtained by applying Eq. 4 to control and constricted conditions:
 | (11) |
Pb2/Pb1 was the measured box pressure amplitude ratio. Second, Ph2/Pg2 was given by the product of the three ratios:
 | (12) |
2 between the lung volume and the box pressure curves for constricted conditions was computed by means of the equation analogous to Eq. 6:
 | (13) |
 | (14) |
 | (15) |
Pg2/Pg1, Vt ratio, and Vm ratio after correction for tissue and blood mass. Note that no correction for 2 was needed for VLm2. Effect of Mch Aerosol Exposure in Conscious Mice
The foregoing analysis was applied to measurements in conscious mice exposed to Mch aerosol. The Raw ratio with bronchoconstriction was obtained by means of Eq. 3 with Abt ratio equal to Pg2/Pg1 divided by frequency ratio (f2/f1) together with the Vt and Vm ratios. Table 5 summarizes values for Vt, Vm, Ve, Ph/Pg, Pb, , Pg2/Pg1, f, Abt ratio, and Raw ratio for control and Mch aerosol-induced conditions in conscious mice. These results were obtained after a 1-min exposure to 50 mg/ml Mch aerosol and were independent of dose (50–125 mg/ml), time of exposure (1–3 min), or number of exposures. Vm2 was obtained from VLm2 by correcting for tissue and blood mass measured in anesthetized control mice, since Raw did not change with the Mch aerosol exposure. Values of Pb and f were obtained from the box pressure cycles associated with the x-ray images that were used to calculate Vt and Vm. There was no consistent change in Vt, Pb, f, or Raw with the Mch aerosol exposure. The important effect of Mch was to increase Vm by 2-fold and Ve by 2.3-fold. Thus any bronchoconstriction due to Mch was eliminated by breathing at a higher lung volume.
Effect of Mch-Induced Bronchoconstriction in Anesthetized Animals
Table 6 summarizes values for Vt, Vm, Ve, Ph/Pg, Pb, , Pg2/Pg1, f, Abt ratio, and Raw ratio for control and Mch aerosol-induced conditions in anesthetized mice. Vm2 was obtained from VLm2 by correcting for tissue and blood mass measured in anesthetized mice after 3 repeated 1-min exposures to Mch aerosol (Table 2). Note that the effect of the Mch aerosol (100 mg/ml) was to decrease f by 72% and to increase Vm by 3-fold and Raw by 8-fold. The first two Mch exposures had no effect on Pb or f relative to control. In general, after the third Mch exposure there was a monotonic increase in Pb and a decrease in f that lasted for 10–20 min until apnea occurred. During this period prior to apnea, average values of Pb and f were obtained from the box pressure cycles associated with the x-ray images that were used to calculate Vt and Vm, and there was no consistent change in Vt and Vm detected from the x-ray data. The reasons for these differences between the conscious and anesthetized mice are discussed below (see DISCUSSION).
Breathing Frequency
Frequency of breathing averaged 5.3 ± 0.26 Hz (Tables 4 and 5) in the conscious mice breathing 100% humidified room air. Body-temperature 100% humidified box air had no significant effect on f. Anesthesia reduced f to 4.0 ± 0.64 Hz (Table 6), somewhat smaller than the reduction observed previously (25).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
) between the box pressure and lung volume cycles. In control mice, was measured via the box pressure excursions with both room- (control) and body-temperature box air conditions. Vt, Ve, and f remained constant under the two box air conditions, verifying that with invariant Raw, gas compression effects were constant. For control conditions, Vt values calculated via the measured x-ray areas were similar to those measured via the corrected box pressure excursions. In mice subjected to repeated 1-min exposures to aerosolized Mch (50–125 mg/ml), tidal volume remained constant while end-expiratory volume increased 2-fold from control conditions. This resulted in no consistent change in the calculated Raw. By contrast, in anesthetized spontaneously breathing mice, 3 repeated doses of 100 mg/ml aerosolized Mch increased Raw by 8-fold and Ve by 4-fold, resulting in apnea within 10–20 min. Methods
The use of the lung area from x-ray images of the thorax to obtain the lung air volume entailed several assumptions. First, lung volume was assumed to scale as AL1.5 for uniform lung inflation. This was found to be a good approximation for anesthetized mice. Second, to obtain the lung area, we subtracted the area of the heart, which was assumed to remain at a constant volume (21). Third, the lung tissue and blood volumes measured independently in anesthetized mice were assumed to remain constant and to be distributed uniformly throughout the lung with lung inflation, and to be similar in conscious mice. We assumed that blood volume was constant with the lung volume increments imposed in the anesthetized mice and with tidal volume in the conscious animals. This issue requires verification, since there is evidence that cardiac output is reduced with lung inflation (2, 3, 39).
The determination of tidal volume and end-expiratory volume from the x-rays taken during quiet breathing entailed several assumptions. First, the box pressure excursions were assumed to be sinusoidal. This was a reasonable assumption, since the sum of the areas measured under the inspiratory and expiratory parts of the box pressure curve was approximated well by 2P/(f), the value for a sine wave (25). Second, in the conscious mice we used x-rays taken during different respiratory cycles and assumed that the box pressure and lung volume excursions, Ve, and f remained constant. This was supported by the measurements that showed little variation in box pressure excursion and frequency during the 10- to 30-min period for the x-ray measurements. However, the lack of a variation in the box pressure excursion did not exclude changes in Ve that would contribute to the variation in VL observed at constant cos (1 – 1) value during repeated respiratory cycles (see Fig. 8). By contrast, in anesthetized mice when Raw increased after Mch aerosol exposure, x-ray images were collected while Pb was increasing and f was decreasing. Vt and Vm computed under these conditions were similar to those of conscious mice, while the changes in Pb and f represented average values over the response time to apnea. Third, the 10-ms x-ray exposure consisted of a single pulse that was short enough compared with the respiratory period (200 ms) to minimize image blur due to ventilatory motion on the x-ray image. The error in phase was 18°, amounting to 5% at the end-expiratory and end-inspiratory points of the box pressure cycle. However, the 10-ms x-ray exposure produced a greater blurring of the heart outline caused by a smaller cardiac period (100 ms). This effect sometimes produced an ill-defined heart outline on the x-ray images, which were subsequently discarded. Our approach using a 10-ms exposure contrasts with the 170-ms exposures that are needed when using available micro-CT scanners (15), which would not provide definitive images at end-expiration and end-inspiration, and thus Vt, in conscious mice breathing at 5 Hz. A similar limitation applies to clinical high-resolution CT scanners (4, 28). Fourth, the configuration of the thorax relative to the dorsal-ventral axis was assumed to be constant during the x-ray measurements. X-ray images that showed a gross asymmetry of the thorax, as observed by changes in curvature of the spine, were not reported. Last, the end-expiratory air volume from the x-ray data was obtained by subtracting the volume occupied by the tissue and blood that were measured in separate anesthetized mice by means of 125I-albumin injected iv prior to euthanasia. Tissue mass averaged 0.10 g in the collapsed isolated lung, which averaged 0.12 g (Table 1). The trapped blood was 0.02 g or 17% of the lung mass, somewhat smaller than the 25% measured in sheep (11). The lung blood mass averaged 0.21 g in the anesthetized BALB/c mice under control conditions and was reduced by 43% to 0.12 g after Mch aerosol exposure (Table 2). The control lung blood mass was greater than the value of 0.09 g measured in conscious male albino mice of the CF-1 strain (16). We used the blood mass measured in control anesthetized mice to determine the end-expiratory air volume in conscious mice when the total lung volume increased 40%. This most likely underestimated the increase in the end-expiratory air volume, since blood mass was reduced with the increased lung volume-induced reduction in cardiac output (2, 3, 39).
Error in Ph/Pg for Differences Between 37–39°C Box Air and Actual Body Temperature
We used a box air temperature of 37–39°C to eliminate the effect of temperature and humidity change on the box pressure excursion as the inspired box air changed to body temperature conditions. However, the body temperature of the anesthetized mouse decreased with time after anesthesia, and the difference between the box air temperature of 37–39°C and actual body temperature contributed to the gas compression effects measured by the box pressure excursion. The error in Ph/Pg due to this effect was 10% via a sensitivity analysis (25). This was verified by the following experiment. Fig. 9B shows the measured peak-to-peak box pressure excursion (Pb) of an anesthetized mouse breathing spontaneously in a sealed box vs. the body-to-box air temperature difference (T) as the box air was raised from 29 to 37°C over an hour period. Humidity was maintained at 100% by saline aerosol, and body temperature was measured by a rectal probe. Fig. 9A shows the associated box air temperature and body temperature vs. T. Note that T was 0°C when box air temperature equaled body temperature at 32°C, and body temperature increased from 32 to 34°C as box air temperature increased from 29 to 37°C. The linear regression of Pb vs. T was: Pb = –0.00071T + 0.024, R2 = 0.39, n = 23, P = 0.0013. The correct gas compression contribution (Pg) to Pb was the intercept 0.024 cmH2O at T of 0°C. The regression indicated that for negative values of T, Pb was greater than Pg, and for positive values of T, Pb was less than Pg. However, the values of Pb below Pg were not consistent with theory that showed Pb increasing above Pg at T of 0°C for both negative and positive values of T as Ph contributes to Pb (Eq. 4). The change in T from negative to positive values is reflected in changes in phase between Ph and Pb. For negative T with box air below body temperature, Ph lags (follows) Pg by 90°, and Ph lags Pb by phase angle (Eq. 6 and Fig. 6). For positive T as box air increases above body temperature, Ph is opposite in direction to and leads Pg by 90°, and results in Ph leading Pb. Evaluation of the Pb data for only positive values of T showed the regression to be insignificant (P > 0.14). For a 3°C change in T from –3 to 0°C, the change in Pb was 0.0021 cmH2O or 9% of the gas compression effect. Thus the gas compression pressure excursion was overestimated by 9%, and the measured Ph/Pg (1.9) was overestimated by 10%.
|
Lung volumes during spontaneous breathing. Vt measured in conscious mice via the box pressure excursion (0.26 ± 0.033 ml, Tables 3–5, n = 15) agreed with values measured via the lung area from the x-ray images (0.25 ± 0.050 ml). This agreement justified the procedures used for determining lung volume from single projection x-ray images. The measured Vt values were comparable to those (0.21 ± 0.070 ml, n = 30) reported in a previous study (25). In anesthetized mice, Vt was somewhat smaller than values measured previously (0.18 ± 0.047 ml, n = 5 vs. 0.29 ± 0.11 ml, n = 30; Ref. 25). Ve averaged 0.28 ± 0.054 ml (Tables 3–5, n = 15) in conscious mice and 0.21 ± 0.15 ml (n = 10) in anesthetized mice. The latter values were comparable to those measured previously (25) by Ne dilution (0.25 ± 0.10 ml, n = 30).
In the present study, the response to Mch aerosol was to increase Ve by 2-fold in conscious mice and by 4.1-fold in anesthetized mice, with no significant change in Vt. By contrast in the previous study, Ve remained constant, while Vt increased by 40% in anesthetized mice exposed to Mch (25). These differences in response to Mch are most likely due to differences in the experimental protocols between the two studies (see below).
Our measurements of Ve in conscious and anesthetized control mice span the range (0.14–0.40 ml) measured in mice with computerized tomography (CT) and micro-CT (28, 15). Our measurements of Vt (0.18–0.33 ml) in conscious and anesthetized mice were generally greater than the value (0.09 ml) measured in anesthetized mice with micro-CT (15). In the latter study the x-ray exposure time of 170 ms during inspiration with a breathing period of 400 s produced the mean lung volume during inspiration, not the end-inspiratory lung volume, and resulted in an underestimate of tidal volume by a factor of two.
The measured increase in Ve in response to aerosolized Mch exposure in conscious mice was consistent with the behavior often observed in humans with increased Raw caused by chronic obstructive pulmonary disease (COPD) or by an acute asthmatic attack (32, 40). In the conscious mice Ve doubled with Mch exposure to partly offset the increase in the gas compression effect to the calculated Raw. However, the increased Ve cannot alone explain why conscious mice showed no increase in Raw with Mch exposure, because Raw increased in anesthetized mice with a similar increase in Ve. Lung hyperinflation secondary to Mch-induced airway constriction resulted in an increased expansile force of the lung parenchyma on the airway that opposed the contractile force of the airway smooth muscle (5–10, 12, 20, 31, 32, 34–37). The response to Mch in conscious mice was largely independent of the Mch concentration and suggested a plateau in the bronchoconstriction that was always compensated by the lung expansion. A plateau in bronchoconstriction has been observed in some studies of normal humans (33, 38) and experimental animals (31, 34). However, a plateau was not observed in anesthetized dogs in one study that used CT (8) or in anesthetized mice in the present study. In humans the increase in Ve with Mch was not attributed to the degree of bronchoconstriction or type of bronchoconstrictor agent per se, but to flow limitation occurring during spontaneous breathing (32, 33). An important factor that would reduce Raw with Mch in the conscious mice is the bronchodilatory effect on the airway smooth muscle after a deep inspiration, as observed in humans (4, 23, 30) and anesthetized rabbits (20). These time-dependent effects might be related to the reduced contractility of airway smooth muscle observed with increases in cyclic changes in muscle length in vitro (18). The mechanisms responsible for these time-dependent effects have been reviewed (19).
By contrast to the results of the conscious mice in the present study, the Mch-induced increased lung volume in the anesthetized mice breathing room air showed no plateau in bronchoconstriction, but seemed to reach airway closure that resulted in gas trapping (5, 22, 32). This effect was verified in two mice whose post mortem lungs failed to collapse after the chest was opened and could not be deflated by suction with a syringe attached to the cannulated trachea. Further evidence for airway closure was provided in two anesthetized mice that were allowed to breathe 100% O2. Here exposure to Mch aerosol similar to that of the mice breathing room air did not prevent the increased Raw or apnea. However, rather than the gas trapping that was observed in mice breathing room air, breathing O2 caused the lung to collapse as O2 was absorbed with closed airways. Thus, a collapsed atelectatic lung was observed on the x-ray images prior to apnea and after opening the chest post mortem, and airway closure was verified by the inability to expand the isolated atelectatic lung with high positive pressure (40 cmH2O).
The observed response to aerosolized Mch in conscious mice was consistent with a plateau in the bronchoconstriction response to Mch concentration often observed in humans (4, 29, 38). The plateau in Raw (34) in anesthetized dogs has been demonstrated by means of CT with high doses of Mch aerosol and not with relatively low doses of Mch directly applied to the airway smooth muscle (8); the difference was attributed to reduced Mch concentration with the aerosol. The latter effect was probably not the reason why airway collapse was not observed in our conscious mice, since the same Mch aerosol concentration produced airway closure in anesthetized mice.
The reasons for the apnea in anesthetized mice after the Mch exposures are speculative and might involve many factors. First, the 8-fold increase in Raw and 4-fold increase in Ve with Mch might result in diaphragmatic fatigue due to the extremely high expansive force required to expand the lung with closed airways (17). This was evident from the Mch-induced 4-fold increase in box pressure excursion (Pb2/Pb1) relative to control (Table 6). The increase in box pressure was consistent with a 3-fold increase in the force generated by the diaphragm, as reflected in the alveolar pressure amplitude (Palv) that equaled the viscous pressure loss (RawQ, Ref. 25) due to airway flow (Q is flow amplitude). In the control anesthetized mice with an Raw of 4.2 cmH2O·ml–1·s (Table 6) and Q of 2.4 ml/s (fVt with Vt of 0.19 ml and f of 4 Hz), Palv was 10 cmH2O. With Mch aerosol exposure, Raw increased 8-fold to 34 cmH2O·ml–1·s, and with a flow amplitude of 0.93 ml/s (Vt of 0.27 ml and f of 1.1 Hz), Palv increased to 32 cmH2O. Such a sustained effort from the diaphragm might result in diaphragm fatigue and in turn a reduced f. Second, diaphragm fatigue was exacerbated by the lung inflation-induced low cardiac output (2, 3, 39) and O2 delivered to the overworked and overcontracted muscle, and resulted in diaphragm muscle failure (17). Similar effects on cardiac output most likely occurred in the mice breathing 100% O2 after airway closure and lung atelectasis. A low cardiac output was consistent with a reduced blood volume measured in the constricted lung (Table 2). Reduced cardiac output, increased Ve and Raw, and gas trapping produced by iv Mch has been reported in anesthetized dogs (2). Diaphragm muscle failure was secondary to the reduced blood flow and was not due to lack of O2 to the blood. The blood flowing through the lung was well oxygenated with no observable evidence of cyanosis or O2 desaturation caused by pulmonary edema, since the lung W/D (4.87 ± 0.23, n = 5) measured post mortem was normal. Airway secretions due to the Mch aerosol, as observed in rats (22), might result in airway closure and gas trapping. However, this effect should also occur in the conscious mice and thus cannot by itself explain the apnea in the anesthetized mice.
One factor that might have contributed to the reduced Mch-induced Raw in conscious mice compared with anesthetized mice was the reduced f with anesthesia (25). The reduced f with anesthesia was not attributed solely to the lung inflation-induced inhibition of active respiration (Hering-Breuer inflation reflex) because f did not decrease in conscious mice with a Mch-induced increased Ve. Airway and parenchymal tissue elastance and resistance, demonstrated in in vivo (31, 35) and in vitro studies (18), might change with frequency under control conditions and in response to Mch.
Airway resistance (Raw). Raw measured in control conscious mice in the present study (3.1 cmH2O·ml–1·s, Tables 4 and 5) was 60% greater than values (1.9 cmH2O·ml–1·s) estimated previously for conscious mice based on Ve values measured in anesthetized mice (25). In the conscious mice of the present study, Raw showed no increase in response to aerosolized Mch, in contrast to the 8-fold increase estimated previously for both anesthetized and conscious mice (25). These differences in the control Raw and Raw ratio with Mch between the two studies might be caused by the following differences in the experimental protocols. First, the need to take x-rays of the thorax in the present study required placing the conscious mouse within a tube to position its thorax over the imaging sensor, and the mouse might have reacted to the physical confinement by an increase in Raw. This confinement might produce an emotional stress that contributed to the response to Mch, similar to the effects observed in humans with asthma (26). Second, the measurements of Vt and Ve required x-rays to be taken after several exposures to aerosolized Mch. The steady-state box pressure excursions obtained over the 10- to 30-min experimental period were smaller than the transient values produced over the 30-s exposure to Mch in the previous study. Third, the present study showed an increased Ve (2.3-fold) in response to Mch, which acted to offset the contribution of the Abt ratio (1.6-fold) to the calculated Raw ratio, while in the previous study Ve was assumed to be unchanged with Mch. Thus any airway constriction induced by Mch in conscious mice was effectively reduced by breathing at a higher lung volume.
Raw in the anesthetized mice averaged 4.2 cmH2O·ml–1·s, comparable to the value measured previously with a similar technique (25), 2.5-fold greater than measurements (1.7 cmH2O·ml–1·s) obtained via end-inflation airway occlusion (14), but 8-fold greater than values (0.5 cmH2O·ml–1·s) measured via the forced oscillation technique (27). The much smaller Raw measured by forced oscillation was most likely due to the relatively small viscous pressure loss under the imposed laminar flow conditions. However, the present technique measured both laminar and turbulent flow contributions of spontaneous breathing to the total viscous pressure loss that determines Raw (25). The relatively high Raw values measured in the present study were most likely not due to a significant contribution of the upper airway, since similar high Raw values were measured previously in tracheostomized animals (25). By contrast to conscious mice in the present study that showed no change in Raw with Mch, Raw in the anesthetized mice with Mch increased 8-fold, comparable to the increase of the previous study (25). However, the Raw response in the present study occurred only after three repeated 1-min exposures to 100 mg/ml Mch aerosol, which resulted in apnea after several minutes. This is in contrast to the previous study, which used a lower dose of Mch (25 mg/ml) with a shorter exposure time of 30 s followed by box pressure measurements of 20–30 s in a sealed box. Here changes in pressure excursion with Mch increased on average by 50% relative to control with no change in f, Ve, or Vt.
Concluding Remarks
We developed a method using single projection x-ray images of the thorax to estimate Ve and Vt in spontaneously breathing mice. We used the technique to show that conscious mice responded to repeated exposures to aerosolized Mch by doubling Ve with little change in Raw. This behavior was largely independent of dose up to 125 mg/ml and the number of exposures, and was consistent with the establishment of a plateau in Raw. By contrast in anesthetized mice, three repeated 1-min exposures to 100 mg/ml Mch aerosol increased Ve by 4-fold and Raw by 8-fold, with breathing frequency decreasing to apnea within 10–20 min. This behavior was consistent with an increasing bronchoconstriction to airway closure and gas trapping, which resulted in diaphragm fatigue and failure exacerbated by the lung inflation-induced reduced blood flow and O2 supplied to the diaphragm muscle.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTS: THEORY AND ANALYSIS...DISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Agrawal, S. K. Singh, V. P. Singh, E. Murphy, and I. Parikh Partitioning of nasal and pulmonary resistance changes during noninvasive plethysmography in mice J Appl Physiol, December 1, 2008; 105(6): 1975 - 1979. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Mitzner Why can't mice just learn to pant? J Appl Physiol, August 1, 2008; 105(2): 402 - 402. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
