Ventilation-perfusion relationships blood-free perfused rabbit lungs
in isolated
D. WALMRATH, R. KijNIG, C. ERNST, H. BRUCKNER, F. GRIMMINGER, Department of lnternul Medicine, Justus-Liebig- University Giessen, D-6300 Giessen, Federal Republic of Germany WALMRATH, II., R. K~NIG, C. ERNST, H. BRUCKNER, F. GRIMMINGER, AND W. SEEGER. Ventilation-perfusion relationships in isoluted blood-free perfused rabbit lungs. J. Appl. Physiol. 72(l): 374-382, 1992.--The multiple inert gas elimination technique (MIGET) was applied to blood-free perfused isolated rabbit lungs.Commonly acceptedcriteria for reliability of the method were found to be fulfilled in this model. Ventilation-perfusion (VA/Q) distributions in isolated control lungs correspondedto those repeatedly detected under physiological conditions. In particular, a narrow unimodal dispersionof perfusate flow was observed:perfusion of low-VA/Q areasranged below 1% and shunt flow --2-3%; perfusion of high-VA/Q regionswas not detected. Gas flow was characterized by narrow dispersion in the midrange-VA/Q areas. Application of a low level of PEEP (1 cmH,O) reduced shunt flow to 4%, and lowTjA@ areas were no longer noted. By using this PEEP-level, stable gas exchange conditions were maintained for S h of extracorporeal perfusion. Graded embolization with small air bubblescauseda typical rightward shift (to higher iTA@ ratios) of mean ventilation, associatedwith the appearanceof highVA/Q regionsand an increasein dead spaceventilation. Mean perfusion was shifted leftward, and shunt flow was approximately doubled. Whole lung lavage with saline for washout of surfactant evoked a progressive manifold increase in shunt flow, accompaniedby a moderate rise of perfusate flow to lowVA/Q areas. We conclude that the MIGET can be applied to isolated blood-free perfused rabbit lungs for assessmentof gas exchange and that typical patterns of VA/Q mismatch are reproduced in this model.
AND W. SEEGER
trace concentrations. This technique has been used in several studies in humans and intact animals. Moreover, in a limited number of investigations, the MIGET was applied to blood-perfused dog lung lobes under openchest conditions (7,13,17). Isolated lung perfusion models allow various manipulations of experimental conditions and parallel assessment of biochemical events. Blood-free perfusate can be used and repeatedly exchanged; the continuous monitoring of perfusion pressures, ventilation pressure, and lung weight as well as sequential determinations of the capillary filtration coefficient, vascular compliance, microvascular resistance distribution, and pressure-volume characteristics are established (18-22). The use of synthetic perfusate is particularly suitable for biochemical analysis of a variety of mediators, assumed to be related to alterations in lung physiology. In the present communication we describe the adoption of the MIGET to buffer-perfused rabbit lungs removed from the thoracic cage. The possibility of detailed gas exchange analysis in an isolated lung model broadens the scope of this investigative tool. M%THODS
Materials. Hydroxyethylamylopectine (mol wt 2OO,OOO) was obtained from Fresenius (Oberursel, FRG). All other analytic grade biochemicals including diethyl ether and acetone were purchased from Merck (Darmstadt, FRG). Halothane was supplied by ICI-Pharma (Plankstadt, ventilation-perfusion ratio; ventilation-perfusion mismatch; FRG). Gas mixtures of sulfur hexafluoride (SF,), ethane, gasexchange; multiple inert gaselimination technique; shunt and cyclopropane (10, 20, and 70%; quantified by gas flow; embolization; bronchoalveolar lavage; positive end-expichromatography) were from Messer Griesheim (Frankratory pressure furt-Griesheim, FRG). Columns for gas chromatography were obtained already packed with Poropak Q mesh 80/ 100 and 5-A molecular sieve from Chrompack (FrankTHE UNDERSTANDING of pulmonary gas exchange has furt, FRG). undergone several major advances since the early 1900s holated lung model. The perfused lung model has been (8,26,31). One of the very important developments was previously described (19,22,23). Briefly, rabbits of either the multiple inert gas elimination technique (MIGET), sex weighing 2.4-3.0 kg were anesthetized with pentobarcommonly regarded as a powerful tool for assessment of bital sodium (60-90 mg/kg iv) and anticoagulated with gas exchange. It is based on the observation that the re- heparin ( 1,000 U/kg). Tracheostomy was performed, and tention of any gas is dependent on the blood-to-gas par- the animals were ventilated with room air with a Harvard tition coefficient of that gas and the distribution of ventirespirator (cat ventilator, Hugo Sachs Elektronik, lation and perfusion (5, 6). Wagner et al. (29, 30) rea- March Hugstetten, FRG). After midsternal thoracotsoned that it would be possible to obtain more omy, catheters were placed in the pulmonary artery and information. ab?ut the entire spectrum of ventilationthe left atrium, and perfusion with Krebs-Henseleit perfusion (VA/Q) distribution by measuring the ex- buffer, containing 4% (wt/vol) hydroxyethylamylopecchange of a variety of gases of different solubilities in tine, was started (BP 720 Fresenius roller pump, pulsa374 0161-7567/92 $2.00 Copyright 0 1992 the American Physiological Society Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
total
RATIO
IN ISOLATED
system heated at 37OC
LUNG MODEL
375
expiration gas mixing and sampling box iure I
pressure r-l
venous pressure regulation
1 flexible perfusate reservoir FIG.
1.
V
Schematic depiction of isolated lung model.
tile flow, Bad Homburg, FRG). For washout of blood, fluid was initially not recirculated, and flow was slowly increased to 100 ml/min. After lung passage of at least 2 liters buffer, a closed circuit (total volume 250 ml) for recirculation of Krebs-Henseleit buffer was used. Left atria1 pressure was set 1.5 mmHg in all experiments. In parallel with onset of artificial perfusion, room air supplemented with 4% CO, was used for ventilation. Tidal volume (9-13 ml/kg) and frequency (6-10 breaths/min) were adapted to maintain the pH of the recirculating buffer between 7.35 and 7.37. The PO, and PCO, values in the postlung buffer fluid were 90-110 Torr and 38 and 43 Torr. In standard protocol, no positive end-expiratory pressure (PEEP) was applied. After complete removal from the thorax, lungs were enclosed in a flexible gastight metalline polyester-aluminium foil bag (Hostaphan, Kalle Folienverarbeitung, Wiesbaden, FRG) fixed to a Plexiglas port for inlet and outlet of perfusion and TABLE
ventilation tubing (Fig. 1). They were freely suspended from a force transducer for monitoring of organ weight. Pressures in the pulmonary artery, left atrium, and trachea were registered with small-diameter tubing threaded into the perfusion catheters and the trachea and connected to pressure transducers. Perfusate samples were taken by use of these tubes. Gas samples were taken from the outlet of a copper expiration gas mixing box (Fig. 1). The whole system was heated to 37”C, ascertained by temperature sensoring in the pulmonary artery catheter and the expiration gas mixing box. Lungs included in the study were those that 1) had a homogeneous white appearance with no signs of hemostasis, edema, or atelectasis; 2) had a pulmonary artery and ventilation pressure in the normal range; and 3) were isogravimetric during an initial steady-state period of at least 45 min. After this initial baseline period was finished, time was set at zero.
1. Reproducibility of inert gas measurements SF,
Partition coefficient (37°C) Repeated measurements* of a pooled gas sample, % Repeated measurements* of a pooled buffer sample, %
Ethane
Cyclopropane
Halothane
Ether
Acetone
250
0.0024
0.0259
0.16
0.64
9.0
0.56
0.3
0.47
0.45
0.72
0.8
1.2
1.6
1.7
2.7
2.1
3.0
Coefficients of variation are given for gas chromatographic analysis of gases and for complete buffer equilibration gas chromatographic measurement of head space gas. * n = 10.
procedure with subsequent
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
376
VENTILATION-PERFUSION
RATIO
2. Ventilation-perfusion distribution in control lungs-absence of PEEP
TABLE
Shunt, %& Low
VA/$,
%&
Normal VA@, . %&
Qmean
Log SDQ Dead space, %V High VA/Q, %v vmean
Log SD. RSS
0 min
20 min
40 min
60 min
2.6t1.33 0.45t0.35
2.43t1.69 0.520.42
2.78t2 0.68kO.73
2.93t1.75 0.78t0.71
96.9t1.5 1.18tO.18 0.51~0.11 47.86t4.47 0 1.39t0.24 ;3,‘:;.;;7
97k1.57 1.22-t0.12 0.55kO.14 47.5t3.87 0 1.5t0.2 0.41kO.07 4.58t2.86
96.4k1.71 1.18+0.17 0.55-to.19 48.95k3.92 0 1.39t0.24 0.37+0.047 4.87t2.94
96.3t1.71 1.18t0.2 0.6ft0.2 48.06~4.05 0 1.44t0.22 0.42to. 1 4.51t3.32
l
-.
Values are means + SD for 6 control lungs at different times after termination of steady-state period. VA/Q, ventilation-perfusion ratio; Q mean,mean perfusion; log SD, and log SD,, log standard deviation of perfusion and ventilation, respectively; Vm,, mean ventilation; RSS, residual sum of squares. Distribution of perfusion and ventilation to areas with different iTA@ ratios is expressed as ptrcentage of total perfusion and tptal ventilation,,respectively: Shunt, VA/Q = -0; low VA/ Q, 0.005 < VA/Q c 0.1; normal VA/Q, 0.1~ VA/Q < 10; high VA/Q, 10 c VA@ < 100; dead space, iTA& > 100. PEEP, positive end-expiratory pressure. VA/Q
Determination
The MIGET
in Isolated Lungs by MIGET
was used as described by Wagner
et al.
(27, 29). Gus infusion and sampling. Six inert gases (SF,, ethane, cyclopropane, halothane, diethylether, and acetone) were dissolved in isotonic saline and continuously infused at a rate of 0.3 ml/min before perfusate passage through a bubble trap, which served as a mixing chamber. After an equilibration period of at least 30 min, lo-ml perfusate samples were simultaneously taken from the pulmonary artery and left atrium. A corresponding 30.ml gas sample was drawn from the expiration gas mixing chamber, in consideration with the time delay in gas flow due to the tubing system and the box volume. Gas-tight glass syringes (Becton Dickinson) were used throughout, and each sampling was performed in duplicate. Gas chromatographic analysis. Extraction of the gases dissolved in the buffer fluid was carried out by equilibration with nitrogen in a shaking water bath (37°C) as described (27). Preceding experiments addressing the time necessary for tonometry of all gases were performed. Incubation periods of 10, 20, 30, 40, 50, and 60 min were TABLE
3. Ventilation-perfusion
Shunt, %& Low VA/Q, %& Normal VA&,%(& .
Qmean
Log SDg Dead space, %v High v~/Q,%v vmean
Log SDf, RSS
distribution
IN ISOLATED
LUNG MODEL
investigated, and 40 min were found to be sufficient for complete equilibration. This time span was thus used for tonometry throughout the study. The gas phases after equilibration of the buffer samples as well as the exhaled gases were analyzed by gas chromatography. Separation and quantification of ethane, cyclopropane, halothane, diethyl ether, and acetone were performed with a gas chromatograph equipped with flame ionization detector (model 3300, Varian), with use of a commercially available Poropak-Q column. SF, was measured by a gas chromatograph fitted with an electron capture detector (Carlo Erba, model Fractovap, Milano, Italy), by use of a 5-A molecular sieve. In each experiment the Krebs-Henseleit buffer-to-gas partition coefficient was determined. Two modes of equilibration were used: tonometry of inert gas containing buffer fluid with nitrogen and tonometry of diluted inert gases with fresh buffer fluid. Conformity of the partition coefficients obtained by both modes was noted, and therefore only the first of the two techniques was used in the later experiments. The ratios of arterial to mixed venous partial pressures (retention) and of expired to mixed venous partial pressures (excretion) were calculated for each gas and were plotted against buffer-gas partition coefficients (retention-solubility and excretion-solubility curves). For each gas, the means of the two retentions and the two excretions from the duplicate samples were used for estimation of the VA/Q distribution by least squares analysis with enforced smoothing (5) with a computer program gratiously supplied by P. D. Wagner. Table 1 gives the reproducibility of buffer equilibration procedure and gas chromatographic analysis for the different gases. Experimental
Protocols
Twenty-seven isolated lung experiments were performed. Normal lungs. No interventions were undertaken. VA/ Q measurements were carried out at 0,20,40, and 60 min (n = 6). Normal lungs with PEEP. PEEP of 1 cmH,O was applied in seven lungs, and samples were taken at 0,20,40, and 60 min. In three of these lungs, the total observation period was prolonged to 4 h, and additional sampling was performed at 120, 180, and 240 min (n = 9).
in lungs ventilated
with 1 cmH,O PEEP
0 min
20 min
40 min
60 min
0.5t0.3 0 99.5t0.33 l.lt0.14 0.4t0.08 49.35t6.12 0 1.31t0.2 0.4kO.08 4.58t2.39
0.4t0.25 0 99.6k0.25 1.17-t0.16 0.37kO.08 46t6.31 0 1.37-t0.22 0.37~~0.08 5.94t2.14
0.51t0.3 0 99.520.3 l.lt0.13 0.38t0.06 47.5t7.67 0 1.38-t0.26 0.38t0.06 4.51t2.5
0.66-10.49 0 99.24t0.65 l.Zt0.18 0.38kO.08 47.5k7.51 0 1.31t0.19 0.38t0.08 5.27tl.78
Values are means t SD for 7 lungs ventilated with PEEP at different times after termination and explanation. In 3 lungs, observation period was prolonged to 4 h.
120
min
0.63t0.38 0 99.3t0.3 1.08t0.16 0.34t0.05 46.6k4.2 0 1.23t0.13 0.34t0.045 7.1k1.78
180 min
240 min
1.2t0,46 0 98.2kO.4 l.Ot0.18 0.381~0.035 50.41~3.6 0 1.17t0.25 0.39t0.46 4.41k2.5
1.36t0.42 0.033_to.o57 98.kkO.35 1.03+0.1 0.41t0.036 49.8~16.18 0 1.2bO.l 0.39t0.021 2.6kO.87
of steady-state period. See Table 2 for definitions
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
PEEP
RATIO
IN
ISOLATED
LUNG
377
MODEL
PEEP 1 cm $0 ......._.......... . . ..__._........‘.‘.~~ ... ..... ,,,.,_.............,..... . _’
1 cm $0
shunt(hg shu&Jng
log-Q L, =I14 - L, =1.5
\ w L, =a59 LpO.80
L, d.53 L, t1.64
2) 1)
b
low $&Iung
2)
L,=O.5 \o L, =0.75
low v;/&ng
1)
40 60 hid 20 0 2. Sequence of changes in ventilation-perfusion ratio (VA/Q) induced by alternate use of positive end-expiratory pressure (PEEP) in 2 isolated lungs with spontaneous moderate edema formation. Both lungs revealed slight visible signs of edema at end of steady-state period. At this time, total weight gain was 2.5 g in Lung I (L,) and 3 g in Lung 2 (L,). After baseline measurement, 1 cmH,O PEEP was applied for tyo 2P-min periods, interrupted by 20 min without PEEP. PEEP-induced reduction of shunt flow and perfusion to low-VA/Q areas, coincident with a decrease in dispersion of perfusion distribution (log SD*), is evident. At end of total 60-min period, lung weight in these lungs had increased to 3.2 (L,) and 3.6 (L2) g. FIG.
In two lungs that displayed signs of edema at the end of the steady-state period and thus missed the entry criteria for the study, the influence of alternating PEEP (0 and 1 cmH,O) on the VA/Q pattern was additionally investigated. Embolism. After an initial VA/Q measurement, air bubbles of l-l.5 mm diam were infused into the pulmonary artery until the pulmonary arterial pressure (PAP) doubled. VA/($ measurements were carried out 15 and 45 min after embolus application was stopped (n = 6). Lavczge. After an initial VA/Q measurement, 12 ml isotonic saline (37°C) were instilled into the trachea and withdrawn after 40-60 s. This lavage procedure was repeated four times at 5-min intervals. Ninety percent of the total fluid was recovered by aspiration. VA@ was determined 15 and 45 min after termination of lavage (n = 6). Data analysis. All values are given as percent means t SD or as coefficient of variation. RESULTS
Steady-state conditions. Mean perfusion flow measured at the roller pump was Ill.3 t 3.7 ml/min in all experiments. With use of the data of the inert gas analysis, a corresponding mean flow of 116 t 8.3 ml/min was calculated on the basis of the Fick principle. The residual sum
of squares was ~10 in all experimental groups; the average of all experiments was 4.05 t 1.98. The residual sum of squares is the result of testing the compatibility of the inert gas data to the derived VA/Q distribution by the least square method. Values ~10 represent data with small errors (14). In five control lungs, the loss of inert gases via the lung surface into the space of the surrounding foil bag was calculated. Repetitive samples were taken from the pulmonary artery, the left atrium, and the exhaled gas as well as from the gas in the foil bag space. Calculated in relation to the exhaled amounts of the different gases, the following losses into the bag space showed up: SF, 0.52 t 0.17%, ethane 1.09 t- O.l%, cyclopropane 1.75 + 0.45%, halothane 1.38 t O.Sl%, diethyl ether 1.23 t 0.9%, and acetone L34 t 0.45%. Baseline conditions. After termination of the steadystate period, all lungs had a mean PAP of 7-12 mmHg and a peak inflation pressure of 6-10 mmHg. These data correspond to previous investigations in the perfused rabbit lung (19-23). In the absence of embolization or these pressures did not change lavage challenge, throughout the observation period, and no weight gain occurred. Baseline measurements revealed normal VA/Q distribution in all lungs (Figs. 3 and 4, Tables 2-5). Shunt flow approximated 2-3%, and perfusion flow to poorly ventilated areas (low-VA/& ratios; see Table 2) was cl%; no perfusion to high-VA/Q regions was observed.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
378
VENTILATION-PERFUSION - 0’ before
Embolitation
Shunt 2.4 46
RATIO
IN ISOLATED
Dead
LUNG MODEL - 0’ before
Embolization
-
l.Oq
space 50.5 %
Retention 0.8 -
ventilation
0
,I b
”
0
1
I
0.01
0.1
Ventilation
- Perfusion
I
1
s z L B 0.6
)I L
\ .g
0.4-
5 d OP-
0.0 -
I
t
lo
100
Buffer
Ratio (log scale1
- 15’ after Embolization Shunt 7.3 96
- Gas Partition
Coefficient
- 15’ after Embolization
Dead
space
62.6
Excretion
-f
w
(log scale) -
%
Excrethn
;r ”
0.h
of1
Ventitation
- Perfusion
i Ratio
- 45’ after Embolization Shunt 4.6 %
lb
0
per fusion
l
t 0
1
,‘,
I oml
Buffer
(log scale)
Dead
ventilatiin
IdO
1 OM
I 0.1
I 1
- Gas Partition Coefficient - 45’ after Embolization
space
1 m
I loo
(log scale) -
61 96
Excretion
6
”
#‘I
Cr.‘01
Ventilation
o’.l
- Perfusion
i
lb
do
Ratio flog scale)
t
0
*’
I
oml
Buffer
1
Odl
I
0.1
1
1
- Gas Partition Coefficient
I
lo
I
loo
Uog scale)
FIG. 3. Left: sequence of iTA@ distribution changes induced by air embolization in an isolated perfused lung. Baseline conditions (preembolization) and iTA@ distributions 15 and 45 min after embolization are given. Right: measured retentions and excretions plotted against buffer-gas partition coefficients (dashed line) and predicted data for a homogeneous lung with corresponding alveolar ventilation and perfusion flow (solid line).
The dispersion of perfusion (log SDQ) varied between 0.51 and 0.61, thus reflecting narrow unimodal perfusion distribution. Ventilation approximated 50% to both midrange&/Q regions and nonperfused regions (dead space), and no distribution to high- or low-V~/Q areas was noted. Log standard deviation of ventilation (log SDv) ranged between OS37 and 0.47, indicating narrow dispersion of gas flow in the midrange-VA/Q
modes. Repetitive measurements in the control lungs did not reveal changes in i7A@ distribution within an observation period of 1 h (Table 2). influence OfpEEp. The application of PEEP induced a decrease in shunt flow to 5 h), only very low shunt flow
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
4. Ventilution-perfusion undergoing air embolization
TABLE
0 min
Shunt, %& Low
VA/$,
%Q
Normal VA/Q, %Q .
2.62t0.7 1.17t0.78 96.3~~0.6 1.18t0.22 0.68kO.21 46.82&9
distribution
RATIO
of lungs
15 min
45 min
5.65t1.3 0.62t1.25 92.5t0.58 0.64kO.16 0.9t0.2 61t5.67
5.3t1.44 1.22t1.89 92.8k2.88 0.77t0.23 0.9t0.34 59.2t7.7 5.25t5.59 2.1t0.65
IN
ISOLATED
LUNG
MODEL
379
4, Table 5). Moreover, a moderate increase in perfusion to low-v~/Q regions, a rise in ventilation to high-VA/Q areas, and an increase of dead space calculation were noted. DISCUSSION
The conditions of gas exchange in the presently used isolated lung model differ from those under physiological circumstances in several respects. First, studies were 0 10.9t5.75 performed in the complete absence of streaming corpusir mean 1.52t0.4 3.28+1.34 Log SDv 0.4t0.07 1.69t0.2 1.2rtrO.68 cular elements, which might influence the distribution of RSS 3.71rt1.73 1.3lkO.97 0.85*0.44 perfusate flow. Erythrocytes are pertinent for gas transport under physiological conditions, but they do not conValues are means + SD for 6 lungs before (0 min) as well as 15 and 45 min after graded air embolization. See Table 2 for definitions and tribute to a major extent to inert gas transport, and large explanation. heterogeneities of hematocrit have minimal influence on the derived VA/& distribution (2, 13, 34). Second, posi(~2%) and perfusion to low-v~/Q regions (~1%) was tive-pressure ventilation was performed. Similarly, other noted. Dead space did not increase in response to PEEP investigators (13, 27) used this type of artificial ventilaapplication, and no ventilation of high-VA/Q areas tion for analysis of gas exchange in blood-perfused isowas observed. In two lungs with spontaneous slight lated lung lobes in open-chest dogs. Third, lungs were edema formation, PEEP application resulted in a reprofreely suspended from the trachea. This erect position ducible reduction of shunt flow and perfusion of low-VA/ will influence the zones of flow distribution. Because of & areas (Fig. 2). the small size of the rabbit lung, however, the quantitaEmbolization. Graded pulmonary artery injection of tive impact of this factor might be small. Fourth, lungs small air bubbles caused a doubling in PAP and a slow were removed from the thoracic cage. This might give increase in lung weight of 1.5 t 0.5 g over 15 min. VA/Q rise to increased gas escape via the pleural surface. Direct determinations performed 15 min later (Fig. 3, Table 4) estimation of the gas loss into the surrounding bag did, showed an increase in dead spat? ventilation to an aver- however, demonstrate a negligible amount. Accordingly, age of 61%. Mean ventilation (V,,,) was shifted rightthe perfusion flow calculated from the gas analysis data ward (3.28 compared with 1.52 before embplization), as- by the Fick principle matched the directly measured flow sociated with the appearance of high-VA/Q regions well, thus excluding major gas losses. Fifth, the absence (average 11% of ventilation). Broader dispersion of ven- of peripheral deoxygenation might influence the tone of tilation was reflected by a fourfold rise in the log SDv lung vessels and thereby the distribution of perfusion. data, as calculated for the complete bimodal gas distribuDespite all these differences between in vivo conditions tion (Table 4), and a quite high SDv of its main mode and isolated lung perfusion, the present study suggests (0.65 t 0.13). Mean perfusion (&,,,) was shifted leftthat the application of the MIGET to this model is valid ward (0.64 compared with 1.18 before embolization), al- and that the conditions of gas exchange correspond to though there was no increase in perfusion of low-VA@ those in intact animals. The control lungs revealed a ratios. However, within the perfusion distribution to narrow and unimodal c$stribution of perfusion and,ven*midrange-VA/Q areas, 69.0 t 10.0% went to VA/Q units tilation, centered on a VA/Q-ratio of -1. No high-VA/Q between 0.1 and 1 (compared with 37.4 t 17.1% before areas were detected, and no gas flow and only a very embolization), and 23.4 t 10.5% was matched to units small fraction of perfusate (maximum 2%) were disbetween 1 and 10 (compared with 58.9 + 17.3% before tributed to low-VA/Q regions. Shunt flow averaged 2-3%, embolization). Shunt flow was approximately doubled in and dead space varied between 44 and 56%. These findembolized lungs. Forty-five minutes after injection of the ings correspond to numerous studies in normal dogs (3, air bubbles, PAP was reversed by -50-60%. Weight gain 15,16,25,27) and in sheep (10). Log SDQ and log SDv are had not further increased or declined at that time. A ten- commonly used as indicators of dispersion, independent of shunt flow and dead space ventilation, which are disredency for redistribution of gas flow to normal-VA/Q garded for calculation of this variable. For unimodal and areas was noted, as indicated by the decline in high-VA/Q areas from - 11 to -5% (Table 4), whereas the increased logarithmic normal distributions, the log SD values rise shunt flow remained virtually unchanged. with increasing degree of VA/Q mismatch (32). Normal Lauage. Repetitive bronchoalveolar lavage of the whole ranges for humans are 0.3-0.6 (31, 33) and for closedlung caused a near doubling in ventilation pressure in all chest dog experiments 0.4-0.65 (log SD,) and 0.66-1.66 experiments (7.5 t 3.8 before compared with 14.2 t 4.2 (log SDv) (2, 7, 15). In isolated dog lung lobe studies, mmHg after lavage), accompanied by a mean PAP rise of log SDg values of 0.75 and 0.93 and log SDv values of 0.82 were described (13, 17). The currently calculated mean 5.0 t 2.6 mmHg. After termination of lavage, the lungs log SDQ values of 0.51-0.61 and mean log SDV values of retained maximally 5 g weight (nonaspirated lavage 0.37-0.42 are thus in good accordance with measurefluid), but no further increase in organ weight occurred ments under physiologic+ conditions. until the end of the observation period. Lavage resulted The low degree of VA/Q heterogeneity was a remarkin a marked increase in shunt flow, amounting to an average of 6.4% after 15 min and 20.0% after 45 min (Fig. able finding, inasmuch as the lungs were ventilated with& mean
Log SDQ Dead spat?, %V High VA/B, %v
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
380
VENTILATION-PERFUSION - 0’ before
Lavage
Shunt 0.2 8
RATIO
IN ISOLATED
Dead
LUNG
MODEL
- 0’ before space
Lavage
-
1.0
54 %
0.8
0.6
0.4
02
0.0 I
I
1
a
1
0.01
0.1
1
10
.‘=
0
Ventilation
- Perfusion
Shunt 9.7 46
0
0
,‘fl
8 0.001
Buffer
I 0.01
1 0.1
- Gas Partition
m 1
space
59.5
8 lo
Coefficient
- 15’ after Lavage
Dead
perfusion
t 0
Ratio (log scale)
- 15’ after Lavage
ventilation
I
1 100
c
? loo
(log scale)
-
8
Excretion
,,
I
=,’
0
I
1
1
I
I
t
0.01
0.1
1
10
100
0
Ventilation
- Perfusion
-*
I
0.001
Buffer
Ratio (log scale)
- 45’ after Lavage Shunt 23.1 96
I
I
I
0.1
1
10
- Gas Partition
Coefficient
- 45’ after Lavage
Dead
I
0.01
space
60.6
8
1
loo
(log scale1
-
1.0q
%
0.8 -
ventilation
0
Derf usion
0
.-5 Z Ii8 0.81 Excretion
\ g
0.4s
5 CT: 029
0.0. r
0
,‘/
I
m
8
8
t
0.01
0.1
1
lo
loo
V&Nation
- Perfusion
Ratio flog scale)
f 0
, 0.001
Buffer
8 0.01
8 0.1
I 1
- Gas Partition Coefficient
8 lo
8
t
loo
(log scale)
FIG. 4. Left; sequence of VA@ changes induced by whole lung bronchoalveolar lavage. Baseline conditions (prelavage) and VA/Q distributions 15 and 45 min after lavage are given. Right: measured retentions and excretions plotted against buffer-gas partition coefficients (broken line) and predicted data for a homogeneous lung with corresponding alveolar ventilation and perfusion flow (solid line).
out PEEP, i.e., in the absence of any transpulmonary pressure gradient during expiration. Application of a small level of PEEP (1 cmH,O) further improved perfusion homogeneity. Shunt flow decreased to values 4 h in the isolated lung model. The impact of PEEP was particularly evident in two lungs with moderate spontaneous edema formation, in which reproducible reduction in shunt flow
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
RATIO
TABLE 5. Ventilation-perfusion
distribution in lungs undergoing whole lung bmnchoalueolar lavage 0 min
Shunt, %& Low VA/Q, %a Normal VA/$, %a . Q
L:zDg Dead spac,e, %v High VA/Q, %ir v
mean
Log SDv RSS
2.04H.47 0.6rtO.76
15 min
45 min
6.43t2.93 4.1t3.36
1.05t0.08 0.47t0.19
0.74t0.16
2Ozk7.8 4.51-t-2.66 70.7t4.18 0.6t0.2
1.19t0.31
1.42t0.22
49.726.7
56.1t4.72
58k4.5 4.3k4.64 2.7t0.74 1.05t0.36 0.69kO.45
97t1.44
0
89.1t4.75
0
1.25kO.19
1.62t0.27
0.37-tO.08 7*55-r- 1.83
0.74t0.25 3.43t2.36
Values are means t SD for 6 lungs before (0 min) as well as 15 and 45 min after repetitive bronchoalveolar lavage of the whole lung. See Table 2 for definitions and explanation.
and low-VA/Q perfusion were achieved. Reversible reopening of atelectatic areas might underlie this observation. Embolization with small air bubbles evoked a typical change in VA/Q pattern. Bimodal distribution was observed, characterized by areas with decreased blood flow, which is the mechanism for production of high-VA/Q and dead space units as well as rightward shift of Vm,,. This pattern is compatible with partial or complete vascular occlusion. Correspondingly, perfusion was shifted leftward (decrease in & mean), suggesting redistribution of flow to remaining nonoccluded areas. These observations are consistent with studies in glass bead, thrombo-, and gas embolization in dogs (3,4,9, 27). The presently noted substantial increase in dead space ventilation is in accordance with one study in thromboembolization (3), but similar observations were not made in the other cited embolization investigations. Two possible explanations may be offered. First, ventilation of unperfused areas might be decreased under physiological conditions because of hypocapnic bronchoconstriction, thus limiting the embolization-related increase in dead space ventilation (1, 12, 23, 24). Because of the artificial respiration with CO,-enriched air, this mechanism would not be operative in the isolated lung model. Accordingly, no increase in ventilation pressure in response to the embolization was presently noted. Second, there may be species differences in the extent of collateral perfusion and collateral ventilation, which may both influence dead space development with vascular occlusive events. Whole lung bronchoalveolar lavage is an established procedure for surfactant depletion and induction of progressive atelectasis formation (11). Thus marked ongoing increase in shunt flow in response to this interve@ioF was observed. In addition, some perfusion of HOW-VA/Q areas was noted, which suggested lung regions with markedly reduced but not completely absent ventilation. Surprisingly, an increase in dead space ventilation was calculated in lungs exposed to the lavage procedure. This was already obvious within 15 min and did not further increase until 45 min. There is no ready explanation for this finding; however, trapping of the highly soluble gases in the remaining (nonreaspirated) lavage fluid in the air spaces might underlie this observation. In conclusion, the present study demonstrates the fea-
IN
ISOLATED
LUNG
381
MODEL
sibility of adopting the MIGET to blood-free perfused isolated lungs. Commonly accepted criteria for reliability of the technique were fulfilled. A pattern of VA/Q distribution was observed, which corresponds to that noted under physiological conditions. Pathophysiol.ogic.al interventions evoked typical alterations in the VA/Q profile. The MIGET appears appropriate for assessment of gas exchange in isolated perfused lungs. The authors thank H. Michnacs for excellent technical assistance and Dr. C. Vogdt and E. Weil (Messer Griesheim, FRG) for help and expertise in establishing a gas chromatographic unit for gas measurement in our laboratory. They also thank Dr. P. D. Wagner for kind assistance with the setup of the multiple inert gas elimination technique, for supplying the computer program, and for proofreading the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft. This manuscript includes parts of the thesis of R. Kiinig. Address for reprint requests: W. Seeger, Zentrum fiir Innere Medizin, Justus-Liebig-Universitgt Giessen, Klinikstrasse 36, D-6300 Giessen, FRG. Received 17 January; accepted in final form 8 July 1991. REFERENCES
R. J., W. G. WOLFE, P. A. EBERT, AND D. C. SABISTON. Effects of carbon dioxide on bronchoconstriction after pulmonary artery occlusion. Am. J. Physiol. 214: 772-775, 1968. 2. BALGOS, A. A., D. C. WILLFORD, AND J. B. WEST. Ventilation-perfusion relationships during induced normovolemic polycythemia in dogs. J. Appl. Physid 65: 1686-1692, 1986. 3. DANTZKER, D. R., P. D. WAGNER, V. W. TORNABENE, N. P. ALAZRAKI, AND J. B. WEST. Gas exchange after pulmonary thrombembolization in dogs. Circ. Res. 42: 92-103, 1978. 4. DELCROIX, M., C. M~LOT, P. LEJEUNE, M. LEEMAN, AND R. NAEIJE. Effects of vasodilators on gas exchange in acute canine embolic pulmonary hypertension. Anesthesiology 72: 77-84, 1990. 1. ALLGOOD,
5. FARHI, L. E, Elimination of inert l-11,1967. 6. FARHI, L. E., AND T. YOKOYAMA.
inequality on elimination
gas by the lung. Respir.
Physiol.
3:
Effects of ventilation-perfusion of inert gases. Respir. Physiol. 3: 12-20,
1967. 7. HEDENSTIERNA, G., F. C. WHITE, R. MAZZONE, AND P. D. WAGNER. Redistribution of pulmonary blood flow in the dog with PEEP ventilation. J. Appl. Physiol. 46: 278-287, 1979. 8. HLASTALA, M. P. Multiple inert gas elimination technique. J. AppZ. l’hysiol. 56: l-7, 1984. M. P., H. T. ROBERTSON, AND B. K. Ross. Gas ex9. HLASTALA, change abnormalities produced by venous gas emboli. Respir. Physiol. 36: 1-17, 1979. 10. HUTTEMEIER, P. C., C. RINGSTED, K. ELIASEN, AND T. MOGENSEN. Ventilation-perfusion inequality during endotoxin-induced
pulmonary vasoconstriciton in conscious sheep: mechanisms of hypoxia. Clin. Physiol. Oxf. 8: 351-358, 1988. 11. LACHMANN, B., B. ROBERTSON, AND J. VOGEL. In vivo lung lavage as an experimental model of the respiratory distress syndrome. Acta Anuesthesiol. 12. LEVY, S. E., AND
Stand.
24: 231-246,
1980.
D. H. SIMMONS. Redistribution of alveolar ventilation following pulmonary thrombembolism in the dog. J. Appl.
Physiol. 36: 60-68, 1974. 13. OHLSSON, J., M. MIDDAUGH, AND M. P. HLASTALA. Reduction of lung perfusion increases irA/Q heterogeneity. J. Appl. Physiol. 66: 2423-2430,1989. 14. POWELL, F. L., AND P. D. WAGNER. Measurement of continuous distributions of ventilation-perfusion in non-alveolar lungs. Respir. Fhysiol. 48: 219-232, 1982. 15. ROMALDINI, H., R. RODRIGUEZ-ROISIN, P. D. WAGNER, AND J. B. WEST. Enhancement of hypoxic pulmonary vasoconstriction by almitrine in the dog. Am. Rev. Respir. Dis. 128: 288-293, 1983. 16. RIJBINFELD, A. R., P. D. WAGNER, AND J. B. WEST. Gas exchange during acute experimental canine asthma. Am. Rev. Respir. Dis. 118: 525-536,1978. P. T., J. SOLWAY, L. D. H. WOOD, AND J. I. 17. SCHUMACKER,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
382
18.
19.
20. 21.
22.
23.
24. 25.
VENTILATION-PERFUSION
RATIO
SZNAJDER. Lobar contribution to VA/Q inequality during constant-flow ventilation. J. Appl. Physiol. 65: 2132-2137, 1988. SEEGER, W., M. BAIJER, AND S. BHAKDI. Staphylococcal alphatoxin elicts hypertension in isolated rabbit lungs. Evidence for thromboxane formation and the role of extracellular calcium. J. Clin. Invest. 74: 849-858, 1984. SEEGER, W., M. MENGER, D. WALMRATH, G. BECKER, F. GRIMMINGER, AND H. NEUHOF. Arachidonic lipoxygenase pathways and increased vascular permeability in isolated rabbit lungs. Am. Rev. Respir. Dis. 136:964-972,1987. SEEGER, W., G. ST~HR, H. R. D. WOLF, AND H. NEUHOF. Alteration of surfactant function due to protein leakage: special interaction with fibrin monomer. J. Appl. Physiol. 58: 326-338, 1985. SEEGER, W., D. WALMRATH, M. MENGER, AND H. NEUHOF. Increased lung vascular permeability after arachidonic acid and hydrostatic challenge. J. Appl. Physiol. 61: 1781-1789, 1986. SEEGER, W., H. WALTER, N. SUTTORP, M. MUHLY, AND S. BHAKDI. Thromboxane-mediated hypertension and vascular leakage evoked by low doses of Escherichia coli hemolysin in rabbit lungs. J. Clin. Invest. 84: 220-227, 1989. SEVERINGHAUS, J. W., E. W. SWENSON, T. N. FINLEY, M. T. LATEGOLA, AND J. WILLIAMS. Unilateral hypoventilation produced in dogs by occluding one pulmonary artery. J. Appl. Physiol. 16: 5360, 1961. SWENSON, E. W., T. N. FINLEY, AND S. V. GUZMAN. Unilateral hypo-ventilation in man during temporary occlusion of one pulmonary artery. J. C&n. Invest. 40: 828-835, 1961. TORNABENE, V. W., J. B. FORTUNE, P. D. WAGNER, AND N. A.
IN ISOLATED
26. 27.
LUNG MODEL
HALASZ. Gas exchange after pulmonary fat embolism in dogs. J. Thorac. Cardiouasc. Surg. 78: 589-599, 1979. WAGNER, P. D. Recent advances in pulmonary gas exchange. Int. AnesthesioZ. Clin. 15: 81-111, 1977. WAGNER, P. D., H. A. SALTZMAN, AND J. B. WEST. Measurement
of continuous distributions 28. 29.
of ventilation-perfusion
ratios: theory.
J. Appl. Physiol. 36: 588-599, 1974. WAGNER, P. D., P. F. NAUMANN, AND R. B. LARAVUSO. Simultaneous measurement of eight foreign gases in blood by gas chromatography. J. Appl. Physiol. 36: 600-605, 1974. WAGNER, P, D., R. B. LARAVUSO, E. GOLDZIMMER, P. F. NAUMANN, AND J. B. WEST. Distributions of ventilation-perfusion ratios in dogs with normal and abnormal lungs. J. Appl. Physiol. 38:
1099-1109,1975. 30. WAGNER, P. D., R. B. LAwvuso, R. R. UHL, AND J. B, WEST. Continuous distributions of ventilation-perfusion ratios in normal subjects breathing air and 100% 0,. J. Clin. Invest. 54: 54-68,1974. 31. WAGNER, P, D., AIVD J. B. WEST. Ventilation perfusion relationships. In: Pulmonary Gus Exchange, edited by J. B. West. New York: Academic, 1980, vol. 1, chapt. 7, p. 219-259. 32. WEST, J. B. Ventilation-perfusion inequality and overall gas exchange in computer models of the lung. Respir. Physiol. 7: 88-100, 1969. 33. WEST, J. B. Ventilation-perfusion relationships. Am. Rev. Respir. Dis. 116: 919-943,1977. 34. YOUNG, I. H., AND P. D. WAGNER. Effect of intrapulmonary hematocrit maldistribution on 02, COa, and inert gas exchange. J. Appl. Physiol. 46: 240-248, 1979.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
in isolated
D. WALMRATH, R. KijNIG, C. ERNST, H. BRUCKNER, F. GRIMMINGER, Department of lnternul Medicine, Justus-Liebig- University Giessen, D-6300 Giessen, Federal Republic of Germany WALMRATH, II., R. K~NIG, C. ERNST, H. BRUCKNER, F. GRIMMINGER, AND W. SEEGER. Ventilation-perfusion relationships in isoluted blood-free perfused rabbit lungs. J. Appl. Physiol. 72(l): 374-382, 1992.--The multiple inert gas elimination technique (MIGET) was applied to blood-free perfused isolated rabbit lungs.Commonly acceptedcriteria for reliability of the method were found to be fulfilled in this model. Ventilation-perfusion (VA/Q) distributions in isolated control lungs correspondedto those repeatedly detected under physiological conditions. In particular, a narrow unimodal dispersionof perfusate flow was observed:perfusion of low-VA/Q areasranged below 1% and shunt flow --2-3%; perfusion of high-VA/Q regionswas not detected. Gas flow was characterized by narrow dispersion in the midrange-VA/Q areas. Application of a low level of PEEP (1 cmH,O) reduced shunt flow to 4%, and lowTjA@ areas were no longer noted. By using this PEEP-level, stable gas exchange conditions were maintained for S h of extracorporeal perfusion. Graded embolization with small air bubblescauseda typical rightward shift (to higher iTA@ ratios) of mean ventilation, associatedwith the appearanceof highVA/Q regionsand an increasein dead spaceventilation. Mean perfusion was shifted leftward, and shunt flow was approximately doubled. Whole lung lavage with saline for washout of surfactant evoked a progressive manifold increase in shunt flow, accompaniedby a moderate rise of perfusate flow to lowVA/Q areas. We conclude that the MIGET can be applied to isolated blood-free perfused rabbit lungs for assessmentof gas exchange and that typical patterns of VA/Q mismatch are reproduced in this model.
AND W. SEEGER
trace concentrations. This technique has been used in several studies in humans and intact animals. Moreover, in a limited number of investigations, the MIGET was applied to blood-perfused dog lung lobes under openchest conditions (7,13,17). Isolated lung perfusion models allow various manipulations of experimental conditions and parallel assessment of biochemical events. Blood-free perfusate can be used and repeatedly exchanged; the continuous monitoring of perfusion pressures, ventilation pressure, and lung weight as well as sequential determinations of the capillary filtration coefficient, vascular compliance, microvascular resistance distribution, and pressure-volume characteristics are established (18-22). The use of synthetic perfusate is particularly suitable for biochemical analysis of a variety of mediators, assumed to be related to alterations in lung physiology. In the present communication we describe the adoption of the MIGET to buffer-perfused rabbit lungs removed from the thoracic cage. The possibility of detailed gas exchange analysis in an isolated lung model broadens the scope of this investigative tool. M%THODS
Materials. Hydroxyethylamylopectine (mol wt 2OO,OOO) was obtained from Fresenius (Oberursel, FRG). All other analytic grade biochemicals including diethyl ether and acetone were purchased from Merck (Darmstadt, FRG). Halothane was supplied by ICI-Pharma (Plankstadt, ventilation-perfusion ratio; ventilation-perfusion mismatch; FRG). Gas mixtures of sulfur hexafluoride (SF,), ethane, gasexchange; multiple inert gaselimination technique; shunt and cyclopropane (10, 20, and 70%; quantified by gas flow; embolization; bronchoalveolar lavage; positive end-expichromatography) were from Messer Griesheim (Frankratory pressure furt-Griesheim, FRG). Columns for gas chromatography were obtained already packed with Poropak Q mesh 80/ 100 and 5-A molecular sieve from Chrompack (FrankTHE UNDERSTANDING of pulmonary gas exchange has furt, FRG). undergone several major advances since the early 1900s holated lung model. The perfused lung model has been (8,26,31). One of the very important developments was previously described (19,22,23). Briefly, rabbits of either the multiple inert gas elimination technique (MIGET), sex weighing 2.4-3.0 kg were anesthetized with pentobarcommonly regarded as a powerful tool for assessment of bital sodium (60-90 mg/kg iv) and anticoagulated with gas exchange. It is based on the observation that the re- heparin ( 1,000 U/kg). Tracheostomy was performed, and tention of any gas is dependent on the blood-to-gas par- the animals were ventilated with room air with a Harvard tition coefficient of that gas and the distribution of ventirespirator (cat ventilator, Hugo Sachs Elektronik, lation and perfusion (5, 6). Wagner et al. (29, 30) rea- March Hugstetten, FRG). After midsternal thoracotsoned that it would be possible to obtain more omy, catheters were placed in the pulmonary artery and information. ab?ut the entire spectrum of ventilationthe left atrium, and perfusion with Krebs-Henseleit perfusion (VA/Q) distribution by measuring the ex- buffer, containing 4% (wt/vol) hydroxyethylamylopecchange of a variety of gases of different solubilities in tine, was started (BP 720 Fresenius roller pump, pulsa374 0161-7567/92 $2.00 Copyright 0 1992 the American Physiological Society Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
total
RATIO
IN ISOLATED
system heated at 37OC
LUNG MODEL
375
expiration gas mixing and sampling box iure I
pressure r-l
venous pressure regulation
1 flexible perfusate reservoir FIG.
1.
V
Schematic depiction of isolated lung model.
tile flow, Bad Homburg, FRG). For washout of blood, fluid was initially not recirculated, and flow was slowly increased to 100 ml/min. After lung passage of at least 2 liters buffer, a closed circuit (total volume 250 ml) for recirculation of Krebs-Henseleit buffer was used. Left atria1 pressure was set 1.5 mmHg in all experiments. In parallel with onset of artificial perfusion, room air supplemented with 4% CO, was used for ventilation. Tidal volume (9-13 ml/kg) and frequency (6-10 breaths/min) were adapted to maintain the pH of the recirculating buffer between 7.35 and 7.37. The PO, and PCO, values in the postlung buffer fluid were 90-110 Torr and 38 and 43 Torr. In standard protocol, no positive end-expiratory pressure (PEEP) was applied. After complete removal from the thorax, lungs were enclosed in a flexible gastight metalline polyester-aluminium foil bag (Hostaphan, Kalle Folienverarbeitung, Wiesbaden, FRG) fixed to a Plexiglas port for inlet and outlet of perfusion and TABLE
ventilation tubing (Fig. 1). They were freely suspended from a force transducer for monitoring of organ weight. Pressures in the pulmonary artery, left atrium, and trachea were registered with small-diameter tubing threaded into the perfusion catheters and the trachea and connected to pressure transducers. Perfusate samples were taken by use of these tubes. Gas samples were taken from the outlet of a copper expiration gas mixing box (Fig. 1). The whole system was heated to 37”C, ascertained by temperature sensoring in the pulmonary artery catheter and the expiration gas mixing box. Lungs included in the study were those that 1) had a homogeneous white appearance with no signs of hemostasis, edema, or atelectasis; 2) had a pulmonary artery and ventilation pressure in the normal range; and 3) were isogravimetric during an initial steady-state period of at least 45 min. After this initial baseline period was finished, time was set at zero.
1. Reproducibility of inert gas measurements SF,
Partition coefficient (37°C) Repeated measurements* of a pooled gas sample, % Repeated measurements* of a pooled buffer sample, %
Ethane
Cyclopropane
Halothane
Ether
Acetone
250
0.0024
0.0259
0.16
0.64
9.0
0.56
0.3
0.47
0.45
0.72
0.8
1.2
1.6
1.7
2.7
2.1
3.0
Coefficients of variation are given for gas chromatographic analysis of gases and for complete buffer equilibration gas chromatographic measurement of head space gas. * n = 10.
procedure with subsequent
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
376
VENTILATION-PERFUSION
RATIO
2. Ventilation-perfusion distribution in control lungs-absence of PEEP
TABLE
Shunt, %& Low
VA/$,
%&
Normal VA@, . %&
Qmean
Log SDQ Dead space, %V High VA/Q, %v vmean
Log SD. RSS
0 min
20 min
40 min
60 min
2.6t1.33 0.45t0.35
2.43t1.69 0.520.42
2.78t2 0.68kO.73
2.93t1.75 0.78t0.71
96.9t1.5 1.18tO.18 0.51~0.11 47.86t4.47 0 1.39t0.24 ;3,‘:;.;;7
97k1.57 1.22-t0.12 0.55kO.14 47.5t3.87 0 1.5t0.2 0.41kO.07 4.58t2.86
96.4k1.71 1.18+0.17 0.55-to.19 48.95k3.92 0 1.39t0.24 0.37+0.047 4.87t2.94
96.3t1.71 1.18t0.2 0.6ft0.2 48.06~4.05 0 1.44t0.22 0.42to. 1 4.51t3.32
l
-.
Values are means + SD for 6 control lungs at different times after termination of steady-state period. VA/Q, ventilation-perfusion ratio; Q mean,mean perfusion; log SD, and log SD,, log standard deviation of perfusion and ventilation, respectively; Vm,, mean ventilation; RSS, residual sum of squares. Distribution of perfusion and ventilation to areas with different iTA@ ratios is expressed as ptrcentage of total perfusion and tptal ventilation,,respectively: Shunt, VA/Q = -0; low VA/ Q, 0.005 < VA/Q c 0.1; normal VA/Q, 0.1~ VA/Q < 10; high VA/Q, 10 c VA@ < 100; dead space, iTA& > 100. PEEP, positive end-expiratory pressure. VA/Q
Determination
The MIGET
in Isolated Lungs by MIGET
was used as described by Wagner
et al.
(27, 29). Gus infusion and sampling. Six inert gases (SF,, ethane, cyclopropane, halothane, diethylether, and acetone) were dissolved in isotonic saline and continuously infused at a rate of 0.3 ml/min before perfusate passage through a bubble trap, which served as a mixing chamber. After an equilibration period of at least 30 min, lo-ml perfusate samples were simultaneously taken from the pulmonary artery and left atrium. A corresponding 30.ml gas sample was drawn from the expiration gas mixing chamber, in consideration with the time delay in gas flow due to the tubing system and the box volume. Gas-tight glass syringes (Becton Dickinson) were used throughout, and each sampling was performed in duplicate. Gas chromatographic analysis. Extraction of the gases dissolved in the buffer fluid was carried out by equilibration with nitrogen in a shaking water bath (37°C) as described (27). Preceding experiments addressing the time necessary for tonometry of all gases were performed. Incubation periods of 10, 20, 30, 40, 50, and 60 min were TABLE
3. Ventilation-perfusion
Shunt, %& Low VA/Q, %& Normal VA&,%(& .
Qmean
Log SDg Dead space, %v High v~/Q,%v vmean
Log SDf, RSS
distribution
IN ISOLATED
LUNG MODEL
investigated, and 40 min were found to be sufficient for complete equilibration. This time span was thus used for tonometry throughout the study. The gas phases after equilibration of the buffer samples as well as the exhaled gases were analyzed by gas chromatography. Separation and quantification of ethane, cyclopropane, halothane, diethyl ether, and acetone were performed with a gas chromatograph equipped with flame ionization detector (model 3300, Varian), with use of a commercially available Poropak-Q column. SF, was measured by a gas chromatograph fitted with an electron capture detector (Carlo Erba, model Fractovap, Milano, Italy), by use of a 5-A molecular sieve. In each experiment the Krebs-Henseleit buffer-to-gas partition coefficient was determined. Two modes of equilibration were used: tonometry of inert gas containing buffer fluid with nitrogen and tonometry of diluted inert gases with fresh buffer fluid. Conformity of the partition coefficients obtained by both modes was noted, and therefore only the first of the two techniques was used in the later experiments. The ratios of arterial to mixed venous partial pressures (retention) and of expired to mixed venous partial pressures (excretion) were calculated for each gas and were plotted against buffer-gas partition coefficients (retention-solubility and excretion-solubility curves). For each gas, the means of the two retentions and the two excretions from the duplicate samples were used for estimation of the VA/Q distribution by least squares analysis with enforced smoothing (5) with a computer program gratiously supplied by P. D. Wagner. Table 1 gives the reproducibility of buffer equilibration procedure and gas chromatographic analysis for the different gases. Experimental
Protocols
Twenty-seven isolated lung experiments were performed. Normal lungs. No interventions were undertaken. VA/ Q measurements were carried out at 0,20,40, and 60 min (n = 6). Normal lungs with PEEP. PEEP of 1 cmH,O was applied in seven lungs, and samples were taken at 0,20,40, and 60 min. In three of these lungs, the total observation period was prolonged to 4 h, and additional sampling was performed at 120, 180, and 240 min (n = 9).
in lungs ventilated
with 1 cmH,O PEEP
0 min
20 min
40 min
60 min
0.5t0.3 0 99.5t0.33 l.lt0.14 0.4t0.08 49.35t6.12 0 1.31t0.2 0.4kO.08 4.58t2.39
0.4t0.25 0 99.6k0.25 1.17-t0.16 0.37kO.08 46t6.31 0 1.37-t0.22 0.37~~0.08 5.94t2.14
0.51t0.3 0 99.520.3 l.lt0.13 0.38t0.06 47.5t7.67 0 1.38-t0.26 0.38t0.06 4.51t2.5
0.66-10.49 0 99.24t0.65 l.Zt0.18 0.38kO.08 47.5k7.51 0 1.31t0.19 0.38t0.08 5.27tl.78
Values are means t SD for 7 lungs ventilated with PEEP at different times after termination and explanation. In 3 lungs, observation period was prolonged to 4 h.
120
min
0.63t0.38 0 99.3t0.3 1.08t0.16 0.34t0.05 46.6k4.2 0 1.23t0.13 0.34t0.045 7.1k1.78
180 min
240 min
1.2t0,46 0 98.2kO.4 l.Ot0.18 0.381~0.035 50.41~3.6 0 1.17t0.25 0.39t0.46 4.41k2.5
1.36t0.42 0.033_to.o57 98.kkO.35 1.03+0.1 0.41t0.036 49.8~16.18 0 1.2bO.l 0.39t0.021 2.6kO.87
of steady-state period. See Table 2 for definitions
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
PEEP
RATIO
IN
ISOLATED
LUNG
377
MODEL
PEEP 1 cm $0 ......._.......... . . ..__._........‘.‘.~~ ... ..... ,,,.,_.............,..... . _’
1 cm $0
shunt(hg shu&Jng
log-Q L, =I14 - L, =1.5
\ w L, =a59 LpO.80
L, d.53 L, t1.64
2) 1)
b
low $&Iung
2)
L,=O.5 \o L, =0.75
low v;/&ng
1)
40 60 hid 20 0 2. Sequence of changes in ventilation-perfusion ratio (VA/Q) induced by alternate use of positive end-expiratory pressure (PEEP) in 2 isolated lungs with spontaneous moderate edema formation. Both lungs revealed slight visible signs of edema at end of steady-state period. At this time, total weight gain was 2.5 g in Lung I (L,) and 3 g in Lung 2 (L,). After baseline measurement, 1 cmH,O PEEP was applied for tyo 2P-min periods, interrupted by 20 min without PEEP. PEEP-induced reduction of shunt flow and perfusion to low-VA/Q areas, coincident with a decrease in dispersion of perfusion distribution (log SD*), is evident. At end of total 60-min period, lung weight in these lungs had increased to 3.2 (L,) and 3.6 (L2) g. FIG.
In two lungs that displayed signs of edema at the end of the steady-state period and thus missed the entry criteria for the study, the influence of alternating PEEP (0 and 1 cmH,O) on the VA/Q pattern was additionally investigated. Embolism. After an initial VA/Q measurement, air bubbles of l-l.5 mm diam were infused into the pulmonary artery until the pulmonary arterial pressure (PAP) doubled. VA/($ measurements were carried out 15 and 45 min after embolus application was stopped (n = 6). Lavczge. After an initial VA/Q measurement, 12 ml isotonic saline (37°C) were instilled into the trachea and withdrawn after 40-60 s. This lavage procedure was repeated four times at 5-min intervals. Ninety percent of the total fluid was recovered by aspiration. VA@ was determined 15 and 45 min after termination of lavage (n = 6). Data analysis. All values are given as percent means t SD or as coefficient of variation. RESULTS
Steady-state conditions. Mean perfusion flow measured at the roller pump was Ill.3 t 3.7 ml/min in all experiments. With use of the data of the inert gas analysis, a corresponding mean flow of 116 t 8.3 ml/min was calculated on the basis of the Fick principle. The residual sum
of squares was ~10 in all experimental groups; the average of all experiments was 4.05 t 1.98. The residual sum of squares is the result of testing the compatibility of the inert gas data to the derived VA/Q distribution by the least square method. Values ~10 represent data with small errors (14). In five control lungs, the loss of inert gases via the lung surface into the space of the surrounding foil bag was calculated. Repetitive samples were taken from the pulmonary artery, the left atrium, and the exhaled gas as well as from the gas in the foil bag space. Calculated in relation to the exhaled amounts of the different gases, the following losses into the bag space showed up: SF, 0.52 t 0.17%, ethane 1.09 t- O.l%, cyclopropane 1.75 + 0.45%, halothane 1.38 t O.Sl%, diethyl ether 1.23 t 0.9%, and acetone L34 t 0.45%. Baseline conditions. After termination of the steadystate period, all lungs had a mean PAP of 7-12 mmHg and a peak inflation pressure of 6-10 mmHg. These data correspond to previous investigations in the perfused rabbit lung (19-23). In the absence of embolization or these pressures did not change lavage challenge, throughout the observation period, and no weight gain occurred. Baseline measurements revealed normal VA/Q distribution in all lungs (Figs. 3 and 4, Tables 2-5). Shunt flow approximated 2-3%, and perfusion flow to poorly ventilated areas (low-VA/& ratios; see Table 2) was cl%; no perfusion to high-VA/Q regions was observed.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
378
VENTILATION-PERFUSION - 0’ before
Embolitation
Shunt 2.4 46
RATIO
IN ISOLATED
Dead
LUNG MODEL - 0’ before
Embolization
-
l.Oq
space 50.5 %
Retention 0.8 -
ventilation
0
,I b
”
0
1
I
0.01
0.1
Ventilation
- Perfusion
I
1
s z L B 0.6
)I L
\ .g
0.4-
5 d OP-
0.0 -
I
t
lo
100
Buffer
Ratio (log scale1
- 15’ after Embolization Shunt 7.3 96
- Gas Partition
Coefficient
- 15’ after Embolization
Dead
space
62.6
Excretion
-f
w
(log scale) -
%
Excrethn
;r ”
0.h
of1
Ventitation
- Perfusion
i Ratio
- 45’ after Embolization Shunt 4.6 %
lb
0
per fusion
l
t 0
1
,‘,
I oml
Buffer
(log scale)
Dead
ventilatiin
IdO
1 OM
I 0.1
I 1
- Gas Partition Coefficient - 45’ after Embolization
space
1 m
I loo
(log scale) -
61 96
Excretion
6
”
#‘I
Cr.‘01
Ventilation
o’.l
- Perfusion
i
lb
do
Ratio flog scale)
t
0
*’
I
oml
Buffer
1
Odl
I
0.1
1
1
- Gas Partition Coefficient
I
lo
I
loo
Uog scale)
FIG. 3. Left: sequence of iTA@ distribution changes induced by air embolization in an isolated perfused lung. Baseline conditions (preembolization) and iTA@ distributions 15 and 45 min after embolization are given. Right: measured retentions and excretions plotted against buffer-gas partition coefficients (dashed line) and predicted data for a homogeneous lung with corresponding alveolar ventilation and perfusion flow (solid line).
The dispersion of perfusion (log SDQ) varied between 0.51 and 0.61, thus reflecting narrow unimodal perfusion distribution. Ventilation approximated 50% to both midrange&/Q regions and nonperfused regions (dead space), and no distribution to high- or low-V~/Q areas was noted. Log standard deviation of ventilation (log SDv) ranged between OS37 and 0.47, indicating narrow dispersion of gas flow in the midrange-VA/Q
modes. Repetitive measurements in the control lungs did not reveal changes in i7A@ distribution within an observation period of 1 h (Table 2). influence OfpEEp. The application of PEEP induced a decrease in shunt flow to 5 h), only very low shunt flow
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
4. Ventilution-perfusion undergoing air embolization
TABLE
0 min
Shunt, %& Low
VA/$,
%Q
Normal VA/Q, %Q .
2.62t0.7 1.17t0.78 96.3~~0.6 1.18t0.22 0.68kO.21 46.82&9
distribution
RATIO
of lungs
15 min
45 min
5.65t1.3 0.62t1.25 92.5t0.58 0.64kO.16 0.9t0.2 61t5.67
5.3t1.44 1.22t1.89 92.8k2.88 0.77t0.23 0.9t0.34 59.2t7.7 5.25t5.59 2.1t0.65
IN
ISOLATED
LUNG
MODEL
379
4, Table 5). Moreover, a moderate increase in perfusion to low-v~/Q regions, a rise in ventilation to high-VA/Q areas, and an increase of dead space calculation were noted. DISCUSSION
The conditions of gas exchange in the presently used isolated lung model differ from those under physiological circumstances in several respects. First, studies were 0 10.9t5.75 performed in the complete absence of streaming corpusir mean 1.52t0.4 3.28+1.34 Log SDv 0.4t0.07 1.69t0.2 1.2rtrO.68 cular elements, which might influence the distribution of RSS 3.71rt1.73 1.3lkO.97 0.85*0.44 perfusate flow. Erythrocytes are pertinent for gas transport under physiological conditions, but they do not conValues are means + SD for 6 lungs before (0 min) as well as 15 and 45 min after graded air embolization. See Table 2 for definitions and tribute to a major extent to inert gas transport, and large explanation. heterogeneities of hematocrit have minimal influence on the derived VA/& distribution (2, 13, 34). Second, posi(~2%) and perfusion to low-v~/Q regions (~1%) was tive-pressure ventilation was performed. Similarly, other noted. Dead space did not increase in response to PEEP investigators (13, 27) used this type of artificial ventilaapplication, and no ventilation of high-VA/Q areas tion for analysis of gas exchange in blood-perfused isowas observed. In two lungs with spontaneous slight lated lung lobes in open-chest dogs. Third, lungs were edema formation, PEEP application resulted in a reprofreely suspended from the trachea. This erect position ducible reduction of shunt flow and perfusion of low-VA/ will influence the zones of flow distribution. Because of & areas (Fig. 2). the small size of the rabbit lung, however, the quantitaEmbolization. Graded pulmonary artery injection of tive impact of this factor might be small. Fourth, lungs small air bubbles caused a doubling in PAP and a slow were removed from the thoracic cage. This might give increase in lung weight of 1.5 t 0.5 g over 15 min. VA/Q rise to increased gas escape via the pleural surface. Direct determinations performed 15 min later (Fig. 3, Table 4) estimation of the gas loss into the surrounding bag did, showed an increase in dead spat? ventilation to an aver- however, demonstrate a negligible amount. Accordingly, age of 61%. Mean ventilation (V,,,) was shifted rightthe perfusion flow calculated from the gas analysis data ward (3.28 compared with 1.52 before embplization), as- by the Fick principle matched the directly measured flow sociated with the appearance of high-VA/Q regions well, thus excluding major gas losses. Fifth, the absence (average 11% of ventilation). Broader dispersion of ven- of peripheral deoxygenation might influence the tone of tilation was reflected by a fourfold rise in the log SDv lung vessels and thereby the distribution of perfusion. data, as calculated for the complete bimodal gas distribuDespite all these differences between in vivo conditions tion (Table 4), and a quite high SDv of its main mode and isolated lung perfusion, the present study suggests (0.65 t 0.13). Mean perfusion (&,,,) was shifted leftthat the application of the MIGET to this model is valid ward (0.64 compared with 1.18 before embolization), al- and that the conditions of gas exchange correspond to though there was no increase in perfusion of low-VA@ those in intact animals. The control lungs revealed a ratios. However, within the perfusion distribution to narrow and unimodal c$stribution of perfusion and,ven*midrange-VA/Q areas, 69.0 t 10.0% went to VA/Q units tilation, centered on a VA/Q-ratio of -1. No high-VA/Q between 0.1 and 1 (compared with 37.4 t 17.1% before areas were detected, and no gas flow and only a very embolization), and 23.4 t 10.5% was matched to units small fraction of perfusate (maximum 2%) were disbetween 1 and 10 (compared with 58.9 + 17.3% before tributed to low-VA/Q regions. Shunt flow averaged 2-3%, embolization). Shunt flow was approximately doubled in and dead space varied between 44 and 56%. These findembolized lungs. Forty-five minutes after injection of the ings correspond to numerous studies in normal dogs (3, air bubbles, PAP was reversed by -50-60%. Weight gain 15,16,25,27) and in sheep (10). Log SDQ and log SDv are had not further increased or declined at that time. A ten- commonly used as indicators of dispersion, independent of shunt flow and dead space ventilation, which are disredency for redistribution of gas flow to normal-VA/Q garded for calculation of this variable. For unimodal and areas was noted, as indicated by the decline in high-VA/Q areas from - 11 to -5% (Table 4), whereas the increased logarithmic normal distributions, the log SD values rise shunt flow remained virtually unchanged. with increasing degree of VA/Q mismatch (32). Normal Lauage. Repetitive bronchoalveolar lavage of the whole ranges for humans are 0.3-0.6 (31, 33) and for closedlung caused a near doubling in ventilation pressure in all chest dog experiments 0.4-0.65 (log SD,) and 0.66-1.66 experiments (7.5 t 3.8 before compared with 14.2 t 4.2 (log SDv) (2, 7, 15). In isolated dog lung lobe studies, mmHg after lavage), accompanied by a mean PAP rise of log SDg values of 0.75 and 0.93 and log SDv values of 0.82 were described (13, 17). The currently calculated mean 5.0 t 2.6 mmHg. After termination of lavage, the lungs log SDQ values of 0.51-0.61 and mean log SDV values of retained maximally 5 g weight (nonaspirated lavage 0.37-0.42 are thus in good accordance with measurefluid), but no further increase in organ weight occurred ments under physiologic+ conditions. until the end of the observation period. Lavage resulted The low degree of VA/Q heterogeneity was a remarkin a marked increase in shunt flow, amounting to an average of 6.4% after 15 min and 20.0% after 45 min (Fig. able finding, inasmuch as the lungs were ventilated with& mean
Log SDQ Dead spat?, %V High VA/B, %v
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
380
VENTILATION-PERFUSION - 0’ before
Lavage
Shunt 0.2 8
RATIO
IN ISOLATED
Dead
LUNG
MODEL
- 0’ before space
Lavage
-
1.0
54 %
0.8
0.6
0.4
02
0.0 I
I
1
a
1
0.01
0.1
1
10
.‘=
0
Ventilation
- Perfusion
Shunt 9.7 46
0
0
,‘fl
8 0.001
Buffer
I 0.01
1 0.1
- Gas Partition
m 1
space
59.5
8 lo
Coefficient
- 15’ after Lavage
Dead
perfusion
t 0
Ratio (log scale)
- 15’ after Lavage
ventilation
I
1 100
c
? loo
(log scale)
-
8
Excretion
,,
I
=,’
0
I
1
1
I
I
t
0.01
0.1
1
10
100
0
Ventilation
- Perfusion
-*
I
0.001
Buffer
Ratio (log scale)
- 45’ after Lavage Shunt 23.1 96
I
I
I
0.1
1
10
- Gas Partition
Coefficient
- 45’ after Lavage
Dead
I
0.01
space
60.6
8
1
loo
(log scale1
-
1.0q
%
0.8 -
ventilation
0
Derf usion
0
.-5 Z Ii8 0.81 Excretion
\ g
0.4s
5 CT: 029
0.0. r
0
,‘/
I
m
8
8
t
0.01
0.1
1
lo
loo
V&Nation
- Perfusion
Ratio flog scale)
f 0
, 0.001
Buffer
8 0.01
8 0.1
I 1
- Gas Partition Coefficient
8 lo
8
t
loo
(log scale)
FIG. 4. Left; sequence of VA@ changes induced by whole lung bronchoalveolar lavage. Baseline conditions (prelavage) and VA/Q distributions 15 and 45 min after lavage are given. Right: measured retentions and excretions plotted against buffer-gas partition coefficients (broken line) and predicted data for a homogeneous lung with corresponding alveolar ventilation and perfusion flow (solid line).
out PEEP, i.e., in the absence of any transpulmonary pressure gradient during expiration. Application of a small level of PEEP (1 cmH,O) further improved perfusion homogeneity. Shunt flow decreased to values 4 h in the isolated lung model. The impact of PEEP was particularly evident in two lungs with moderate spontaneous edema formation, in which reproducible reduction in shunt flow
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
VENTILATION-PERFUSION
RATIO
TABLE 5. Ventilation-perfusion
distribution in lungs undergoing whole lung bmnchoalueolar lavage 0 min
Shunt, %& Low VA/Q, %a Normal VA/$, %a . Q
L:zDg Dead spac,e, %v High VA/Q, %ir v
mean
Log SDv RSS
2.04H.47 0.6rtO.76
15 min
45 min
6.43t2.93 4.1t3.36
1.05t0.08 0.47t0.19
0.74t0.16
2Ozk7.8 4.51-t-2.66 70.7t4.18 0.6t0.2
1.19t0.31
1.42t0.22
49.726.7
56.1t4.72
58k4.5 4.3k4.64 2.7t0.74 1.05t0.36 0.69kO.45
97t1.44
0
89.1t4.75
0
1.25kO.19
1.62t0.27
0.37-tO.08 7*55-r- 1.83
0.74t0.25 3.43t2.36
Values are means t SD for 6 lungs before (0 min) as well as 15 and 45 min after repetitive bronchoalveolar lavage of the whole lung. See Table 2 for definitions and explanation.
and low-VA/Q perfusion were achieved. Reversible reopening of atelectatic areas might underlie this observation. Embolization with small air bubbles evoked a typical change in VA/Q pattern. Bimodal distribution was observed, characterized by areas with decreased blood flow, which is the mechanism for production of high-VA/Q and dead space units as well as rightward shift of Vm,,. This pattern is compatible with partial or complete vascular occlusion. Correspondingly, perfusion was shifted leftward (decrease in & mean), suggesting redistribution of flow to remaining nonoccluded areas. These observations are consistent with studies in glass bead, thrombo-, and gas embolization in dogs (3,4,9, 27). The presently noted substantial increase in dead space ventilation is in accordance with one study in thromboembolization (3), but similar observations were not made in the other cited embolization investigations. Two possible explanations may be offered. First, ventilation of unperfused areas might be decreased under physiological conditions because of hypocapnic bronchoconstriction, thus limiting the embolization-related increase in dead space ventilation (1, 12, 23, 24). Because of the artificial respiration with CO,-enriched air, this mechanism would not be operative in the isolated lung model. Accordingly, no increase in ventilation pressure in response to the embolization was presently noted. Second, there may be species differences in the extent of collateral perfusion and collateral ventilation, which may both influence dead space development with vascular occlusive events. Whole lung bronchoalveolar lavage is an established procedure for surfactant depletion and induction of progressive atelectasis formation (11). Thus marked ongoing increase in shunt flow in response to this interve@ioF was observed. In addition, some perfusion of HOW-VA/Q areas was noted, which suggested lung regions with markedly reduced but not completely absent ventilation. Surprisingly, an increase in dead space ventilation was calculated in lungs exposed to the lavage procedure. This was already obvious within 15 min and did not further increase until 45 min. There is no ready explanation for this finding; however, trapping of the highly soluble gases in the remaining (nonreaspirated) lavage fluid in the air spaces might underlie this observation. In conclusion, the present study demonstrates the fea-
IN
ISOLATED
LUNG
381
MODEL
sibility of adopting the MIGET to blood-free perfused isolated lungs. Commonly accepted criteria for reliability of the technique were fulfilled. A pattern of VA/Q distribution was observed, which corresponds to that noted under physiological conditions. Pathophysiol.ogic.al interventions evoked typical alterations in the VA/Q profile. The MIGET appears appropriate for assessment of gas exchange in isolated perfused lungs. The authors thank H. Michnacs for excellent technical assistance and Dr. C. Vogdt and E. Weil (Messer Griesheim, FRG) for help and expertise in establishing a gas chromatographic unit for gas measurement in our laboratory. They also thank Dr. P. D. Wagner for kind assistance with the setup of the multiple inert gas elimination technique, for supplying the computer program, and for proofreading the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft. This manuscript includes parts of the thesis of R. Kiinig. Address for reprint requests: W. Seeger, Zentrum fiir Innere Medizin, Justus-Liebig-Universitgt Giessen, Klinikstrasse 36, D-6300 Giessen, FRG. Received 17 January; accepted in final form 8 July 1991. REFERENCES
R. J., W. G. WOLFE, P. A. EBERT, AND D. C. SABISTON. Effects of carbon dioxide on bronchoconstriction after pulmonary artery occlusion. Am. J. Physiol. 214: 772-775, 1968. 2. BALGOS, A. A., D. C. WILLFORD, AND J. B. WEST. Ventilation-perfusion relationships during induced normovolemic polycythemia in dogs. J. Appl. Physid 65: 1686-1692, 1986. 3. DANTZKER, D. R., P. D. WAGNER, V. W. TORNABENE, N. P. ALAZRAKI, AND J. B. WEST. Gas exchange after pulmonary thrombembolization in dogs. Circ. Res. 42: 92-103, 1978. 4. DELCROIX, M., C. M~LOT, P. LEJEUNE, M. LEEMAN, AND R. NAEIJE. Effects of vasodilators on gas exchange in acute canine embolic pulmonary hypertension. Anesthesiology 72: 77-84, 1990. 1. ALLGOOD,
5. FARHI, L. E, Elimination of inert l-11,1967. 6. FARHI, L. E., AND T. YOKOYAMA.
inequality on elimination
gas by the lung. Respir.
Physiol.
3:
Effects of ventilation-perfusion of inert gases. Respir. Physiol. 3: 12-20,
1967. 7. HEDENSTIERNA, G., F. C. WHITE, R. MAZZONE, AND P. D. WAGNER. Redistribution of pulmonary blood flow in the dog with PEEP ventilation. J. Appl. Physiol. 46: 278-287, 1979. 8. HLASTALA, M. P. Multiple inert gas elimination technique. J. AppZ. l’hysiol. 56: l-7, 1984. M. P., H. T. ROBERTSON, AND B. K. Ross. Gas ex9. HLASTALA, change abnormalities produced by venous gas emboli. Respir. Physiol. 36: 1-17, 1979. 10. HUTTEMEIER, P. C., C. RINGSTED, K. ELIASEN, AND T. MOGENSEN. Ventilation-perfusion inequality during endotoxin-induced
pulmonary vasoconstriciton in conscious sheep: mechanisms of hypoxia. Clin. Physiol. Oxf. 8: 351-358, 1988. 11. LACHMANN, B., B. ROBERTSON, AND J. VOGEL. In vivo lung lavage as an experimental model of the respiratory distress syndrome. Acta Anuesthesiol. 12. LEVY, S. E., AND
Stand.
24: 231-246,
1980.
D. H. SIMMONS. Redistribution of alveolar ventilation following pulmonary thrombembolism in the dog. J. Appl.
Physiol. 36: 60-68, 1974. 13. OHLSSON, J., M. MIDDAUGH, AND M. P. HLASTALA. Reduction of lung perfusion increases irA/Q heterogeneity. J. Appl. Physiol. 66: 2423-2430,1989. 14. POWELL, F. L., AND P. D. WAGNER. Measurement of continuous distributions of ventilation-perfusion in non-alveolar lungs. Respir. Fhysiol. 48: 219-232, 1982. 15. ROMALDINI, H., R. RODRIGUEZ-ROISIN, P. D. WAGNER, AND J. B. WEST. Enhancement of hypoxic pulmonary vasoconstriction by almitrine in the dog. Am. Rev. Respir. Dis. 128: 288-293, 1983. 16. RIJBINFELD, A. R., P. D. WAGNER, AND J. B. WEST. Gas exchange during acute experimental canine asthma. Am. Rev. Respir. Dis. 118: 525-536,1978. P. T., J. SOLWAY, L. D. H. WOOD, AND J. I. 17. SCHUMACKER,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
382
18.
19.
20. 21.
22.
23.
24. 25.
VENTILATION-PERFUSION
RATIO
SZNAJDER. Lobar contribution to VA/Q inequality during constant-flow ventilation. J. Appl. Physiol. 65: 2132-2137, 1988. SEEGER, W., M. BAIJER, AND S. BHAKDI. Staphylococcal alphatoxin elicts hypertension in isolated rabbit lungs. Evidence for thromboxane formation and the role of extracellular calcium. J. Clin. Invest. 74: 849-858, 1984. SEEGER, W., M. MENGER, D. WALMRATH, G. BECKER, F. GRIMMINGER, AND H. NEUHOF. Arachidonic lipoxygenase pathways and increased vascular permeability in isolated rabbit lungs. Am. Rev. Respir. Dis. 136:964-972,1987. SEEGER, W., G. ST~HR, H. R. D. WOLF, AND H. NEUHOF. Alteration of surfactant function due to protein leakage: special interaction with fibrin monomer. J. Appl. Physiol. 58: 326-338, 1985. SEEGER, W., D. WALMRATH, M. MENGER, AND H. NEUHOF. Increased lung vascular permeability after arachidonic acid and hydrostatic challenge. J. Appl. Physiol. 61: 1781-1789, 1986. SEEGER, W., H. WALTER, N. SUTTORP, M. MUHLY, AND S. BHAKDI. Thromboxane-mediated hypertension and vascular leakage evoked by low doses of Escherichia coli hemolysin in rabbit lungs. J. Clin. Invest. 84: 220-227, 1989. SEVERINGHAUS, J. W., E. W. SWENSON, T. N. FINLEY, M. T. LATEGOLA, AND J. WILLIAMS. Unilateral hypoventilation produced in dogs by occluding one pulmonary artery. J. Appl. Physiol. 16: 5360, 1961. SWENSON, E. W., T. N. FINLEY, AND S. V. GUZMAN. Unilateral hypo-ventilation in man during temporary occlusion of one pulmonary artery. J. C&n. Invest. 40: 828-835, 1961. TORNABENE, V. W., J. B. FORTUNE, P. D. WAGNER, AND N. A.
IN ISOLATED
26. 27.
LUNG MODEL
HALASZ. Gas exchange after pulmonary fat embolism in dogs. J. Thorac. Cardiouasc. Surg. 78: 589-599, 1979. WAGNER, P. D. Recent advances in pulmonary gas exchange. Int. AnesthesioZ. Clin. 15: 81-111, 1977. WAGNER, P. D., H. A. SALTZMAN, AND J. B. WEST. Measurement
of continuous distributions 28. 29.
of ventilation-perfusion
ratios: theory.
J. Appl. Physiol. 36: 588-599, 1974. WAGNER, P. D., P. F. NAUMANN, AND R. B. LARAVUSO. Simultaneous measurement of eight foreign gases in blood by gas chromatography. J. Appl. Physiol. 36: 600-605, 1974. WAGNER, P, D., R. B. LARAVUSO, E. GOLDZIMMER, P. F. NAUMANN, AND J. B. WEST. Distributions of ventilation-perfusion ratios in dogs with normal and abnormal lungs. J. Appl. Physiol. 38:
1099-1109,1975. 30. WAGNER, P. D., R. B. LAwvuso, R. R. UHL, AND J. B, WEST. Continuous distributions of ventilation-perfusion ratios in normal subjects breathing air and 100% 0,. J. Clin. Invest. 54: 54-68,1974. 31. WAGNER, P, D., AIVD J. B. WEST. Ventilation perfusion relationships. In: Pulmonary Gus Exchange, edited by J. B. West. New York: Academic, 1980, vol. 1, chapt. 7, p. 219-259. 32. WEST, J. B. Ventilation-perfusion inequality and overall gas exchange in computer models of the lung. Respir. Physiol. 7: 88-100, 1969. 33. WEST, J. B. Ventilation-perfusion relationships. Am. Rev. Respir. Dis. 116: 919-943,1977. 34. YOUNG, I. H., AND P. D. WAGNER. Effect of intrapulmonary hematocrit maldistribution on 02, COa, and inert gas exchange. J. Appl. Physiol. 46: 240-248, 1979.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 11, 2019.
