Society for Critical Care Anesthesiologists
Section Editor: Avery Tung
Inhibition of Constitutive Nitric Oxide Synthase Does
Not Influence Ventilation–Perfusion Matching in
Normal Prone Adult Sheep With Mechanical Ventilation
Mats J. Johansson, MD, PhD,*† John-Peder Escobar Kvitting, MD, PhD,†‡ Torun Flatebø,§
Anne Nicolaysen,§ Gunnar Nicolaysen, MD, PhD,§ and Sten M. Walther, MD, PhD*†
BACKGROUND: Local formation of nitric oxide in the lung induces vasodilation in proportion
to ventilation and is a putative mechanism behind ventilation–perfusion matching. We hypothesized that regional ventilation–perfusion matching occurs in part due to local constitutive nitric
oxide formation.
METHODS: Ventilation and perfusion were analyzed in lung regions (≈1.5 cm3) before and after inhibition of constitutive nitric oxide synthase with Nω-nitro-l-arginine methyl ester (L-NAME) (25 mg/
kg) in 7 prone sheep ventilated with 10 cm H2O positive end-expiratory pressure. Ventilation and
perfusion were measured by the use of aerosolized fluorescent and infused radiolabeled microspheres, respectively. The animals were exsanguinated while deeply anesthetized; then, lungs
were excised, dried at total lung capacity, and divided into cube units. The spatial location for each
cube was tracked and fluorescence and radioactivity per unit weight determined.
RESULTS: After administration of L-NAME, pulmonary artery pressure increased from a mean of
16.6–23.6 mm Hg, P = .007 but Pao2, Paco2, and SD log(V/Q) did not change. Distribution of
ventilation was not influenced by L-NAME, but a small redistribution of perfusion from ventral to
dorsal lung regions was observed. Perfusion to regions with the highest ventilation (fifth quintile
of the ventilation distribution) remained unchanged after L-NAME.
CONCLUSIONS: We found minimal or no influence of constitutive nitric oxide synthase inhibition
by L-NAME on the distributions of ventilation and perfusion, and ventilation–perfusion in prone,
anesthetized, ventilated, and healthy adult sheep with normal gas exchange. (Anesth Analg
2016;123:1492–9)
T
he matching of ventilation and perfusion is important for efficient gas exchange in the lung. In the
normal lung, the distribution of blood flow and
ventilation historically has been explained by an effect of
gravity on regional perfusion pressure and regional lung
compliance. The gravitational model, derived from clinical
and experimental observations, states that the distribution
of lung blood flow is determined by the interplay between
alveolar, arterial, and venous pressures resulting in a perfusion gradient with increased blood flow in basal areas of
the lung independent of body position.1 Additional mechanisms, however, were proposed to explain patterns of
perfusion inconsistent with the gravitational model, such
From the *Department of Cardiothoracic Anesthesia and Intensive Care;
†Division of Cardiovascular Medicine, Department of Medical and Health
Sciences; ‡Department of Cardiothoracic Surgery, Linköping University
Hospital, Linköping, Sweden; and §Department of Physiology, Institute
of Basic Medical Sciences, Faculty of Medicine, University of Oslo, Oslo,
Norway.
Accepted for publication July 11, 2016.
Funding: Institutional funds at Faculty of Medicine and Health Sciences,
Linköping University, Sweden; Faculty of Medicine, University of Oslo,
Norway; The Anders Jahres Foundation for Promotion of Sciences, Norway;
AGA Gas AB, Lidingö, Sweden.
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journal’s website (www.anesthesia-analgesia.org).
Reprints will not be available from the authors.
Address correspondence to John-Peder Escobar Kvitting, MD, PhD, Department of Cardiothoracic Surgery, Linköping University Hospital, SE-581 85
Linköping, Sweden. Address e-mail to [email protected].
Copyright © 2016 International Anesthesia Research Society
DOI: 10.1213/ANE.0000000000001556
1492www.anesthesia-analgesia.org
as relatively preserved dorsal lung blood flow in the prone
position.2–5
One possible mechanism for this nongravitational distribution of lung blood flow is local formation of nitric oxide
(NO) in the pulmonary vasculature and airways.6,7 NO is
found in exhaled air and plays an important role in lung disease.8,9 The role of NO in healthy lungs, however, is less well
understood. NO is formed throughout the lung by the vascular endothelium in response to tangential shear stresses
exerted by the viscous drag of blood flow.10 Nonuniform
NO formation initially was described by Pelletier et al,11
who found increased NO formation in dorsocaudal lung
regions in horse lungs. Evidence of increased NO formation
in dorsal compared with ventral lung regions also has been
reported in humans by Rimeika et al.12 In standing horses,
goats and sheep, prone dogs, and prone human subjects,
this increased local NO formation may then increase vascular conductance in dorsal lung regions, facilitating dorsal
blood flow and preventing a gravitational distribution of
lung blood flow.13–17 Although high-resolution imaging of
lung blood flow in prone position reveals perfusion gradients that favor dependent lung regions,18–20 these gradients
usually are considerably smaller than those in the supine
position.
A NO-dependent mechanism controlling ventilation–
perfusion (V/Q) matching in healthy lungs is supported by
studies indicating that NO is produced by the lung when the
lung is stretched, for example, by mechanical ventilation with
positive end-expiratory pressure (PEEP).21–23 This ventilationdependent formation of NO also may explain the very precise alignment of perfusion to ventilation through the effect
December 2016 • Volume 123 • Number 6
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
of NO on small distal pulmonary arteries. Existing data are
unclear regarding the role of NO in the local matching of perfusion to ventilation in the lung. Infusion of the potent pulmonary vasodilator prostacyclin in healthy primates did not
affect pulmonary perfusion distribution,24 suggesting a lack
of vasomotor tone. In addition, inhibition of endogenous NO
formation in awake, spontaneously breathing sheep did not
affect distribution of ventilation and perfusion.15 More recent
high-resolution magnetic resonance imaging data, however,
do indicate an effect of NO in hypoxic vasoconstriction.25
One potential explanation for the divergent effects of NO in
different studies is differences in transpulmonary pressures
between spontaneous and mechanical ventilation. Previous
work showed that the distribution of perfusion was significantly different between spontaneous and mechanical ventilation.26 The authors of the aforementioned sheep study15 did
not examine the potential impact of mechanical ventilation
on endogenous NO formation. Hence, we performed the current study in mechanically ventilated healthy sheep to clarify
the role of endogenous NO formation on blood flow distribution and V/Q matching. The animals were studied in the
prone position with 10 cm H2O PEEP.
Endogenous NO formation was blocked by intravenous administration of Nω-nitro-l-arginine methyl ester
(L-NAME), a potent inhibitor of nitric oxide synthase
(NOS). Melsom et al15 verified that NO could not be detected
in expired air when L-NAME was infused in doses greater
than 10 mg/kg, suggesting complete inhibition of endogenous NO production. Because the normal distribution of
lung blood flow in prone sheep demonstrates minimal or no
gravitational gradient, alterations leading to ventral redistribution and V/Q mismatch should be detected.
METHODS
Experimental Procedure
The experimental protocol was approved by the local
Animal Experimentation Committee at the University of
Oslo, Norway. Seven healthy sheep (30–40 kg) were studied. A foreleg vein was cannulated, and anesthesia induced
with pentobarbital sodium and maintained by infusion
4–7 mL/h (40 mg/mL) and a bolus of meperidine (8 mg/
mL). Animals were then tracheotomized (to avoid contamination of NO from the upper airways) and catheters introduced under sterile conditions into the carotid artery, the
external jugular vein, and the pulmonary artery via surgical cutdowns in the neck. Cardiac output was measured by
use of the thermodilution technique. Mechanical ventilation
was provided with a Servo 900C ventilator (Siemens Elema,
Solna, Sweden) in pressure-controlled mode with pressures
set to yield tidal volumes = 6–8 mL/kg with PEEP = 10 cm
H2O. The respiratory rate was adjusted to maintain normocapnia. The animals were ventilated with nitric oxide-free
air (AGA, Lidingö, Sweden). Blood gases were measured
with an automated blood-gas analyzer (model ABL 520,
Radiometer, Copenhagen, Denmark).
Inhibition of NOS
Endogenous NO generation was blocked by the inhibition
of NOS with L-NAME (Fluka Chemie, Buchs, Switzerland)
25 mg/kg, a competitive l-arginine analog.
December 2016 • Volume 123 • Number 6
Labeling of Perfusion and Ventilation
Radiolabeled microspheres were infused into the right atrium
as previously described.27 To summarize, approximately 106
microspheres (diameter 15.5 μm; NEN, Boston, MA) labeled
with different radioactive isotopes (106Ru or 95Nb) suspended
in saline were infused at 1 mL/min using a glass syringe
while being stirred continuously with a magnetic bar within
the suspension and an external magnetic stirrer placed
beneath the syringe. A fluorescent aerosol was made from an
aqueous solution of yellow–green, orange or red fluorescent
polystyrene microspheres (Fluo-Spheres, 0.2-µm diameter,
carboxylate-modified, Molecular Probes, Eugene, OR) and
delivered as previously described.28 To summarize, the solution was nebulized continuously with an Acorn Optimist II
nebulizer (MedicAid, Pagham, UK) that eliminated particles
>5 µm. The median mass aerodynamic diameter of the aerosol
was 1.1 µm (25th and 75th percentiles were 0.60 and 1.40 µm,
respectively) at a driving pressure of 500 kPa and a flow of
8 L/min. The fluorescent aerosol was fed continuously into
the inspiratory limb of the breathing circuit. The inspiratory
limb acted as spacer with a volume exceeding tidal volume.
A short connector allowed air to displace from the inspiratory
limb into the expiratory outlet. The direction of the airflow
during inspiration and the direction of the air expired by the
animal was controlled by 2 low resistance 1-way valves in a
Y-piece connected to the endotracheal tube.28 The radiolabeled
and fluorescent microspheres were delivered simultaneously
to the animal during a period of approximately 8 minutes.
Processing the Lungs
The animals were exsanguinated at the end of the experiment while still deeply anesthetized. The lungs and trachea
were excised en bloc and trimmed from adhering nonpulmonary tissue (eg, lymphatic glands, cardiac muscle). They
were then hung by the trachea and expanded with pressurized air (20–25 cm H2O). Numerous holes were made
through the lung pleura with a 22-gauge needle to facilitate
drying. After drying, the right lung was cut in 1-cm-thick
slices horizontal to the direction of gravity when the microspheres were given. Postmortem lung size and expansion
varied resulting in lungs being cut into median 18 planes,
range 17–20 planes. Therefore, to ease comparison between
animals, the vertical size of the lungs was normalized to 10
planes with use of the formulae Pij = 10 pij pmaxj , where Pij
was the normalized plane i (rounded to nearest integer), pij
was plane number i, and pmaxj was the highest plane number
(from 17 to 20) in the jth animal. The slices were divided into
cube units and the location of each unit was mapped. The
pieces were weighed and placed in labeled plastic tubes,
after which radioactive emission was counted in a gamma
counter (Cobra Autogamma 5002; Packard, Downers Grove,
IL). Fluorochromes were then eluted from the polystyrene
microspheres by soaking each region in 3 mL of 2-ethoxyethyl acetate (Aldrich-Chemie, Steinheim, Germany) for
approximately 24 hours with intermittent mixing. The
intensity of the fluorescent dye was measured in the supernatant of each sample using a LS50B spectrophotometer
(Perkin-Elmer, Buckinghamshire, UK). All intensities were
corrected for background contribution from 2-ethoxyethyl
acetate. There was no overlap between the signals from the
different fluorescent dyes.
(
)
www.anesthesia-analgesia.org 1493
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Effects of L-NAME on Gas Exchange
Experimental Protocol
All 7 animals were studied on a firm shallow v-shaped
surface in the prone sphinx posture with the legs flexed
beside the body and the head supported above the surface while mechanically ventilated as described previously
first at baseline and then with constitutive nitric oxide synthase (cNOS) inhibition. Repeated baseline measurements
were taken 1 hour apart in 5 of the 7 animals before cNOS
inhibition.
Calculations and Statistical Analysis
The number of animals needed in the experiment was calculated on the basis of previous studies of perfusion heterogeneity in healthy, mechanically ventilated sheep16 and
observations in dogs showing that spatial perfusion patterns were stable over time.29 We assumed that inhibition
of NO would increase gravitational heterogeneity from 0.25
(SD 0.08) at baseline16 to 0.35 (SD 0.10). Seven animals were
needed, given α = .05, β = .80, and correlation 0.7 for paired
measurements. To assess the influence of time, we used 5
animals as indicated by a previous study.29
Radioactivity and fluorescence per cubed lung piece was
normalized by dividing each radioactive or fluorescent signal
by the weight of the lung piece. Weight-normalized relative
radioactivity and fluorescence in each piece was calculated
by dividing the signal of each piece by the mean signal of
all lung pieces, yielding an overall normalized mean relative
perfusion and ventilation of 1.0 for each sheep and experimental condition. Normalized radioactivity and fluorescence
per unit weight in cubed lung pieces were then interpreted
as equivalent to regional perfusion and ventilation, respectively. To compare the effect of NO inhibition in regions with
high and low ventilation we grouped, for each animal, lung
regions in quintiles by their normalized ventilation.
For each animal and experimental condition, total variance of perfusion and ventilation were partitioned into
variance due to vertical height and residual variance. Total
variance was equal to the squared coefficient of variation
(CVtotal2) because mean regional flows were 1.0 as a result
of the normalization procedure described previously.
The total variance (total heterogeneity) was partitioned
into variance because of vertical height (gravitational
heterogeneity, CVgrav) and residual variance (isogravitational heterogeneity, CVisograv) with use of the formula
2
2
(CVtotal )2 = CVgrav + CVisograv .30 CVisograv was determined
within each isogravitational plane after calculating regional
perfusion and ventilation relative to perfusion and ventilation in the plane from which the region originated.
V/Q heterogeneity was evaluated with the SD of logV/Q
[SDlog(V/Q)].
We applied linear regression to describe relationships
between V and Q, changes in Q between experimental conditions, log (V/Q), and vertical height (horizontal plane
number) per animal. All data were tested for normality with
the Shapiro-Wilk W test and were found to be normally distributed (Supplemental Digital Content, Supplemental Table
E1, http://links.lww.com/AA/B495). Data are presented as
mean, SD, and 99% confidence interval (CI) and mean differences between experimental conditions were analyzed with
a paired 2-tailed t-test. A P < .01 was deemed significant to
(
) (
)
1494 www.anesthesia-analgesia.org
account for multiple comparisons, and STATA 14.1 (STATA
Corp LP, College Station, TX) was used for the analyses.
RESULTS
The median number of lung regions processed per animal
was 947 (range 825–1062). Repeated baseline measurements showed that perfusion distribution was stable and
physiologic variables were unchanged, except for a significant fall in cardiac output from 3.9 to 3.0 L/min output (Supplemental Figure E1, http://links.lww.com/AA/
B584; and Supplemental Table E2, http://links.lww.com/
AA/B495). The effects of cNOS inhibition on physiologic
variables are shown in Table 1. Pulmonary artery pressure
(PAP) increased by 7.0 mm Hg (99% CI 0.5–13.6 mm Hg)
and cardiac output decreased by 1.7 L (99% CI 0.8–2.5 L)
after cNOS inhibition. Airway pressures, tidal volumes, and
arterial carbon dioxide tensions remained unchanged during cNOS inhibition.
Effects of cNOS Inhibition on Global Indices of
Gas Exchange
Systemic artery oxygenation and alveolar-arterial oxygen
tension difference did not change with cNOS inhibition nor
did estimates of gas exchange based on regional V/Q ratios
(Table 2).
Effects of cNOS Inhibition on V, Q, and V/Q
Distributions
The topographic distribution of ventilation and measures of
ventilation heterogeneity did not change with cNOS inhibition (Figure 1 and Table 2). The vertical distribution of pulmonary perfusion did change, with redistribution from ventral
to dorsal lung regions during cNOS inhibition (Figure 2, left
panel). The vertical distribution of the change in perfusion
from baseline to cNOS inhibition ( Q cNOS inhibition − Q Baseline )
was significantly different from zero (−0.082, 99% CI −0.15
to −0.011, P = .005 compared with a zero slope). This change
indicates an effect of cNOS inhibition on ventral–dorsal distribution. Neither isogravitational nor gravitational perfusion heterogeneity differed between experimental conditions
(P = .045 and P = .07, respectively; Table 2). The vertical distribution of V/Q did not change significantly with L-NAME
(P = .12 for difference in vertical slope between baseline and
cNOS inhibition, Figure 2, right panel).
Effects of cNOS Inhibition on Perfusion of Lung
Regions With High Ventilation
Regional ventilation was grouped in quintiles for each animal and the mean proportion of each quintile per horizontal plane was visualized (Figure 3). The mean proportion of
perfusion to regions with high ventilation did not change
(from 27.6% [99% CI 24.4–30.8] at baseline to 26.9% [99% CI
24.2–29.7] with cNOS inhibition, P = .54; Figure 4).
DISCUSSION
We found that inhibition of cNOS with L-NAME did not
change regional V/Q matching and global gas exchange
in prone sheep ventilated with PEEP. We found minimal
redistribution of pulmonary perfusion from ventral to dorsal lung regions with L-NAME administration, implying
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Table 1. Hemodynamic and Ventilation Variables
Mean systemic arterial pressure (mm Hg)
Mean pulmonary arterial pressure (mm Hg)
Cardiac output (L/min)
Mean airway pressure (cm H2O)
Tidal volume (L)
Baseline,
Mean (SD)
137.0 (7.7)
16.6 (3.8)
3.9 (0.7)
15.6 (1.1)
0.289 (0.033)
cNOS-Inhibition,
Mean (SD)
130.6 (21.1)
23.6 (7.6)
2.2 (0.6)
15.4 (0.5)
0.287 (0.028)
Difference (Baseline–cNOS
Inhibition), Mean (99% CI)
6.4 (−16.0 to 28.8)
−7.0 (−13.6 to −0.5)a
1.7 (0.8 to 2.5)b
0.1 (−1.1 to 1.4)
0.003 (−0.021 to 0.026)
N = 7 for all analyses.
Abbreviation: cNOS, constitutive nitric oxide synthase.
a
P = .007.
b
P = .0003 paired t-test.
Table 2. Gas Exchange Parameters and Measures of Ventilation and Perfusion
Arterial oxygen tension, kPa
Arterial carbon dioxide tension, kPa
Alveolar-arterial oxygen tension difference, kPa
Mean log(V/Q)
Mean SDlog(V/Q)
QCV total
QCV gravitational
QCV isogravitational
VCV total
VCV gravitational
VCV isogravitational
Baseline, Mean (SD)
12.7 (1.5)
4.3 (0.5)
3.4 (1.0)
−0.02 (0.02)
0.29 (0.12)
0.42 (0.06)
0.26 (0.09)
0.31 (0.06)
0.57 (0.09)
0.29 (0.12)
0.48 (0.07)
cNOS-Inhibition, Mean (SD)
12.0 (1.3)
4.6 (0.6)
3.8 (0.8)
−0.03 (0.05)
0.30 (0.10)
0.39 (0.05)
0.17 (0.06)
0.35 (0.05)
0.59 (0.20)
0.25 (0.11)
0.51 (0.18)
Difference (Baseline–cNOS
Inhibition), Mean (99% CI)
0.7 (−1.6 to 3.0)
−0.3 (−1.1 to 0.5)
−0.4 (−2.0 to 1.2)
−0.002 (−0.06 to 0.05)
−0.001 (−0.13 to 0.13)
0.02 (−0.08 to 0.13)
0.09 (−0.06 to 0.24)
−0.04 (−0.09 to 0.02)a
−0.02 (−0.32 to 0.28)
0.04 (−0.17 to 0.25)
−0.04 (−0.35 to 0.27)
N = 7 for all analyses.
Abbreviations: CI, confidence interval; cNOS, constitutive nitric oxide synthase; CV, coefficient of variation; Q, perfusion; V, ventilation.
a
P = .045 paired t-test.
significance of these results, we should note some methodological issues.
Methodological Considerations
Figure 1. Mean (±1 SD) normalized ventilation per horizontal plane
at baseline (open circle) and with constitutive nitric oxide synthase
(cNOS) inhibition (closed circle). N = 7 per plane and experimental
stage.
that preferential NO formation in dorsal segments does
not occur, at least not in mechanically ventilated adult
sheep. Furthermore, we also found no apparent redistribution of pulmonary perfusion away from regions
with high ventilation after cNOS inhibition, arguing
against the hypothesis that NO is formed preferentially in
regions with high alveolar stretch. Before considering the
December 2016 • Volume 123 • Number 6
We used adult sheep which, like humans, respond to cNOS
inhibition with increased pulmonary vascular resistance, for
2 reasons.10 First, previous studies in prone sheep with PEEP
ventilation noted a remarkably uniform distribution of perfusion with no gravitational gradient.3,16 This finding suggests the presence of an active mechanism that mitigates the
expected gravity-driven distribution of lung blood flow and
leads to increased perfusion in dorsal lung regions. Second,
arterial oxygenation was well maintained in the prone position and 10 cm H2O PEEP had minimal impact in comparison with 0 PEEP.16,28 Thus, if local NO formation was to play
a role in intrinsic V/Q matching, the prone position should
be suitable for study. To manipulate intrinsic NO formation,
we used the potent specific inhibitor of cNOS, L-NAME,31
at a dose previously demonstrated to block cNOS.15 We did
not use inhaled aerosolized L-NAME because it may have
interfered with ventilation measurements. L-NAME is a
prodrug lacking NOS inhibitory activity unless it is hydrolyzed to Nω-nitro-l-arginine, which inhibits both neural and
endothelial NOS.32 Overall, pulmonary vascular resistance
increased with L-NAME administration reflecting a fall in
NO production.15,33–35 Because of the long duration of cNOS
inhibition, we could not randomize the sequence of experimental stages to exclude the possibility that the observed
effects resulted from the passage of time. Repeated baseline
measurements, however, indicated that the distribution of
perfusion was stable over time, in agreement with previous
research.29
www.anesthesia-analgesia.org 1495
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Effects of L-NAME on Gas Exchange
Figure 2. Left panel: mean (±1 SD) normalized
perfusion per plane at baseline (open circles)
and with constitutive nitric oxide synthase
(cNOS) inhibition (closed circles). Note that
perfusion in dorsal planes was greater with
cNOS inhibition. N = 7 per experimental stage.
Right panel: mean (±1 SD) log ventilation–perfusion ratios [log(V/Q)] per horizontal plane at
baseline (open circle) and with cNOS inhibition
(closed circle). N = 7 per stage.
Figure 3. Ventilation quintiles during constitutive nitric oxide synthase (cNOS) inhibition, partitioned per horizontal plane. For each
sheep, lung regions were grouped in quintiles by their normalized
ventilation. Data show mean proportions (N = 7) per plane. Note
that lung regions from the fifth compartment (High V) were not evenly
distributed in the lung; they were less frequent in dorsal horizontal
planes.
Regional perfusion was quantified with intravenous
administration of 15-µm radioactive microspheres, which
lodge in the pulmonary microcirculation in proportion to
local perfusion. This method has been validated previously
against a molecular tracer technique.36
Regional ventilation was determined with inhalation of
a wet aerosol of 0.2-µm fluorescent microspheres. Although
our measurements may have been distorted by deposition
of microspheres on conducting airways, previous studies with a similar aerosol found no evidence of airway (vs
alveolar) deposition.13,28,37 In addition, regional distribution
of this aerosol is similar to the distribution of Technegas,
a nanoparticle aerosol, which is mostly deposited by
diffusion-mediated dispersion in gas-exchanging parenchyma.28,30 Animals that inhaled 1-μm fluorescent microspheres aerosols had less than 1% of total deposition on
large conducting airways, and photomicrographs of aerosol
1496 www.anesthesia-analgesia.org
Figure 4. Proportion of total normalized perfusion per ventilation
compartment. For each sheep and experimental condition, lung
regions were grouped in quintiles by their normalized ventilation
at baseline (open bars) and with constitutive nitric oxide synthase
(cNOS) inhibition (closed bars). Bars show mean values (n = 7).
deposition on those lungs did not reveal deposition in the
1–3 mm airways.37 The lungs were inflated postmortem and
dried at total lung capacity to obtain uniform expansion of
alveoli. Absolute values for regional ventilation and pulmonary perfusion were not determined, because our focus was
to examine the influence of cNOS inhibition on the change
in regional ventilation and perfusion. Flow and ventilation
in each unit dissected were thus normalized to mean ventilation and perfusion of all units per animal and experimental condition.
Although earlier work using similar high-resolution
methods generated data that were analyzed using the multiple inert gas elimination technique,38 we could not use
the multiple inert gas elimination technique calculation
method because we measured V and Q in the right lung
only. Ventilation and perfusion distributions, however, are
similar in both lungs36,37 and the response to cNOS inhibition should thus not differ between lungs.
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Microsphere methods require that the lung be dissected
into regions of interest that traverse anatomic boundaries.
Because acinar compliance and conductance are not randomly scattered in the lung but clustered,39 we considered
a spatial resolution of 1–2 cm2 in the present study (equivalent to the size of an acinar cluster) sufficient to quantify
relevant changes of V and Q. This topographic resolution is
on the same order as that achieved with positron emission
tomography40 and magnetic resonance imaging.41 Positron
emission tomography is the only nondestructive method
that has been compared quantitatively with global gas
exchange in normal and diseased conditions.40 Improved
topographic resolution may provide insights into mechanisms that contribute to V/Q heterogeneity, but comparing
heterogeneity between methods is problematic because it
is scale-dependent (increases with increasing resolution42)
and it is influenced by different procedures for exclusion of
airways and large blood vessels. Finally, our study was limited by the low number of animals studied and our results
should thus be viewed as exploratory.
Role of NO on Global and Local V/Q Matching
We noted a small decrease in relative perfusion to ventral
lung regions and a corresponding increase in dorsal lung
perfusion with cNOS inhibition. This change led to a more
uniform distribution of pulmonary perfusion along the vertical (gravitational) axis. When we compared our results
with previous findings in spontaneously breathing awake
sheep,15 both studies found a dorsal redistribution of blood
flow which oppose other studies suggesting preferential
formation of NO in the dorsal lung regions.11,12,43 In the
present and previous sheep study15 V/Q heterogeneity was
unaffected overall and when analyzed per plane (isogravitational or height-corrected V/Q heterogeneity, compare
Figure 2B).
In the current study, the isogravitational effect on lung
perfusion remained unchanged with L-NAME, indicating
that cNOS inhibition did not play a role in perfusion distribution in regions with similar perfusion pressures. Our
finding that L-NAME caused some redistribution of blood
flow from the ventral to dorsal lung (in a direction opposite
to gravity) is unexplained. It may be due to cNOS induced
increases in PAP, which may have increased perfusion of
dorsal capillary beds. Similar ventral-to-dorsal redistribution of lung blood flow has been reported with increased
PAPs of the same magnitude as in our study.3,44–46 We cannot separate, however, in vivo effects of cNOS inhibition
on regional V/Q matching from concomitant effects of
increased PAP. V/Q heterogeneity in the most dorsal planes,
however was similar with cNOS inhibition compared with
baseline. It appears unlikely that a decrease in cardiac output and increase in PAP would result in similar V/Q matching as fine tuning of Q to V by local NO. We conclude from
the present and prior sheep study15 that in prone spontaneously breathing or mechanically ventilated sheep gas
exchange was undisturbed by blockage of endogenous NO
formation. The stable matching of V and Q despite mechanical ventilation, NO blockage and quite large changes in
cardiac output and PAP suggest that a robust, non–NOdependent mechanism matches V and Q in prone sheep
December 2016 • Volume 123 • Number 6
during normoxia. Additional studies are needed to clarify
to what extent the global hemodynamic changes contribute
to the redistribution of perfusion we observed.
Although our data do not support NO-mediated
increases in dorsal blood flow, other evidence suggests otherwise. In a 1986 study in dogs, conductance to blood flow
was greater in dorsal than in ventral lung segments, independent of position.47 It has been postulated that local formation of NO in dorsal lung segments could be a possible
explanation. In 1998, Pelletier et al11 also reported increased
vasodilation from dorsal regions of horse lungs consistent
with increased dorsal release of NO. In-vitro studies of
healthy human and pig lungs reporting greater cNOS activity in dorsal lung segments further support increased dorsal
NO formation.12,43
It is difficult to reconcile the aforementioned findings in
the literature with our results. One possible explanation is
that the in vitro studies may not reflect the in vivo state. This
explanation is backed by Glenny and coworkers, who found
no detectable vasomotor tone during baseline conditions in
anesthetized baboons.24 The absence of basal vasomotor
tone in their study provide indirect evidence for the lack
of NO-mediated vasodilation in normal lung, since without
vasoconstriction, NO cannot exert its vasodilatory effect.
Is There an Association Between Alveolar
Stretching and Increase in Perfusion Caused by
NO?
NO formation may be mediated by other factors besides
shear stresses on vascular endothelium. Neuronal NOS
expressed in nerve fibers in upper and lower airway smooth
muscle also add to NO formation48 as does endothelial NOS
in human bronchial epithelium.49 Such local NO formation
may also contribute to the matching of perfusion to ventilation. We found no decrease, however, in local perfusion in
response to cNOS inhibition when we examined perfusion
to lung regions with high ventilation. The theory that lung
formation of NO is induced by alveolar ventilation originates from observations that NO produced in the airways
improved oxygenation50 and by studies finding higher levels of expired NO with increasing PEEP levels.21,22 In our
study, however, we did not find that NO formation due to
tidal ventilation plays any role in the regulation of local perfusion and matching of perfusion to ventilation.
Clinical Inferences
In the normal lung, vascular tone is presumed to be minimal during normoxia and not responsible for matching of
ventilation and perfusion.24 Ventilation/perfusion matching
in healthy lungs was traditionally explained by a common
influence of gravity on ventilation and perfusion. Perfusion
increases in dependent lung due to gravity and ventilation increases in the same segments, because these poorly
inflated regions are more easily expanded than nondependent lung.1 However, many studies using high-resolution
imaging contradict the notion of a gravitational effect on
perfusion.2,5,16,17,28 In our study, we could not demonstrate a
NO-dependent mechanism for matching V and Q, at least not
in prone healthy sheep mechanically ventilated with PEEP.
Interestingly, recent data show that inhaled NO significantly
www.anesthesia-analgesia.org 1497
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Effects of L-NAME on Gas Exchange
altered the perfusion pattern in normoxic healthy humans
in supine position.25 Thus, some baseline vasomotor tone
exists during normoxia in humans, presumably mediated by
hypoxic pulmonary vasoconstriction in dorsal lung regions.25
This finding may not be relevant for the prone position
because turning prone is associated with better gas exchange.5
More studies are needed to clarify the role of active and passive mechanisms that regulate matching of ventilation and
perfusion in normal and diseased lungs.
CONCLUSIONS
The present study does not support a role of endogenous
NO in normal V/Q matching in prone sheep ventilated
with PEEP. Our data suggest that V/Q matching may be
regulated by other mechanisms besides local NO production. Further research is needed to better understand the
mechanisms underlying V/Q matching in the normal
mechanically ventilated lung.
DISCLOSURES
Name: Mats J. Johansson, MD, PhD.
Contribution: This author helped Conceive and design the study,
collect and interpret the data, and draft the manuscript.
Name: John-Peder Escobar Kvitting, MD, PhD.
Contribution: This author helped interpret, process, and clean the
data, and draft the manuscript.
Name: Torun Flatebø.
Contribution: This author helped design and collect the data.
Name: Anne Nicolaysen.
Contribution: This author helped design and collect the data.
Name: Gunnar Nicolaysen, MD, PhD.
Contribution: This author helped conceive and design the study,
interpret the data, and draft the manuscript.
Name: Sten M. Walther, MD, PhD.
Contribution: This author helped conceive and design the study,
collect, analyze and interpret the data, and draft the manuscript.
This manuscript was handled by: Avery Tung, MD, FCCM.
References
1. West JB. Regional differences in the lung. Chest 1978;74:426–437.
2. Reed JH Jr, Wood EH. Effect of body position on vertical distribution of pulmonary blood flow. J Appl Physiol. 1970;28:303–311.
3. Hakim TS, Dean GW, Lisbona R. Effect of body posture on spatial distribution of pulmonary blood flow. J Appl Physiol (1985)
1988;64:1160–1170.
4. Parker JC, Ardell JL, Hamm CR, Barman SA, Coker PJ. Regional
pulmonary blood flow during rest, tilt, and exercise in unanesthetized dogs. J Appl Physiol (1985) 1995;78:838–846.
5.Mure M, Domino KB, Lindahl SG, Hlastala MP, Altemeier
WA, Glenny RW. Regional ventilation-perfusion distribution
is more uniform in the prone position. J Appl Physiol (1985)
2000;88:1076–1083.
6. Griffith TM, Edwards DH, Davies RL, Harrison TJ, Evans KT.
EDRF coordinates the behaviour of vascular resistance vessels.
Nature 1987;329:442–445.
7.Tucker JF, Brave SR, Charalambous L, Hobbs AJ, Gibson A.
L-NG-nitro arginine inhibits non-adrenergic, non-cholinergic
relaxations of guinea-pig isolated tracheal smooth muscle. Br J
Pharmacol. 1990;100:663–664.
8. Brett SJ, Evans TW. Measurement of endogenous nitric oxide
in the lungs of patients with the acute respiratory distress syndrome. Am J Respir Crit Care Med. 1998;157:993–997.
9. LaForce C, Brooks E, Herje N, Dorinsky P, Rickard K. Impact
of exhaled nitric oxide measurements on treatment decisions
in an asthma specialty clinic. Ann Allergy Asthma Immunol.
2014;113:619–623.
10.Hampl V, Herget J. Role of nitric oxide in the pathogen
esis of chronic pulmonary hypertension. Physiol Rev.
2000;80:1337–1372.
1498 www.anesthesia-analgesia.org
11. Pelletier N, Robinson NE, Kaiser L, Derksen FJ. Regional differences in endothelial function in horse lungs: possible role in
blood flow distribution? J Appl Physiol (1985) 1998;85:537–542.
12. Rimeika D, Nyrén S, Wiklund NP, et al. Regulation of regional
lung perfusion by nitric oxide. Am J Respir Crit Care Med.
2004;170:450–455.
13.Melsom MN, Kramer-Johansen J, Flatebø T, Müller C,
Nicolaysen G. Distribution of pulmonary ventilation and perfusion measured simultaneously in awake goats. Acta Physiol
Scand. 1997;159:199–208.
14. Neumann PH, Kivlen CM, Johnson A, Minnear FL, Malik AB.
Effect of alveolar hypoxia on regional pulmonary perfusion.
J Appl Physiol Respir Environ Exerc Physiol. 1984;56:338–342.
15.Melsom MN, Flatebø T, Nicolaysen G. No apparent effect
of nitric oxide on the local matching of pulmonary perfusion and ventilation in awake sheep. Acta Physiol Scand.
2000;168:361–370.
16. Walther SM, Johansson MJ, Flatebø T, Nicolaysen A, Nicolaysen
G. Marked differences between prone and supine sheep in
effect of PEEP on perfusion distribution in zone II lung. J Appl
Physiol (1985) 2005;99:909–914.
17.Nyrén S, Mure M, Jacobsson H, Larsson SA, Lindahl SG.
Pulmonary perfusion is more uniform in the prone than in the
supine position: scintigraphy in healthy humans. J Appl Physiol
(1985) 1999;86:1135–1141.
18.Treppo S, Mijailovich SM, Venegas JG. Contributions of pulmonary perfusion and ventilation to heterogeneity in V(A)/Q
measured by PET. J Appl Physiol (1985) 1997;82:1163–1176.
19.Musch G, Layfield JD, Harris RS, et al. Topographical dis
tribution of pulmonary perfusion and ventilation, assessed
by PET in supine and prone humans. J Appl Physiol (1985)
2002;93:1841–1851.
20.Nyrén S, Radell P, Lindahl SG, et al. Lung ventilation and
perfusion in prone and supine postures with reference to
anesthetized and mechanically ventilated healthy volunteers.
Anesthesiology 2010;112:682–687.
21. Persson MG, Lönnqvist PA, Gustafsson LE. Positive end-expiratory pressure ventilation elicits increases in endogenously
formed nitric oxide as detected in air exhaled by rabbits.
Anesthesiology 1995;82:969–974.
22. Strömberg S, Lönnqvist PA, Persson MG, Gustafsson LE. Lung
distension and carbon dioxide affect pulmonary nitric oxide
formation in the anaesthetized rabbit. Acta Physiol Scand.
1997;159:59–67.
23.Kuebler WM, Uhlig U, Goldmann T, et al. Stretch activates
nitric oxide production in pulmonary vascular endothelial cells
in situ. Am J Respir Crit Care Med. 2003;168:1391–1398.
24.Glenny RW, Robertson HT, Hlastala MP. Vasomotor tone
does not affect perfusion heterogeneity and gas exchange in
normal primate lungs during normoxia. J Appl Physiol (1985)
2000;89:2263–2267.
25. Asadi AK, Sá RC, Kim NH, et al. Inhaled nitric oxide alters the
distribution of blood flow in the healthy human lung, suggesting active hypoxic pulmonary vasoconstriction in normoxia.
J Appl Physiol (1985) 2015;118:331–343.
26. Walther SM, Domino KB, Glenny RW, Hlastala MP. Pulmonary
blood flow distribution in sheep: effects of anesthesia,
mechanical ventilation, and change in posture. Anesthesiology
1997;87:335–342.
27. Iversen PO, Standa M, Nicolaysen G. Marked regional heterogeneity in blood flow within a single skeletal muscle at rest and
during exercise hyperaemia in the rabbit. Acta Physiol Scand.
1989;136:17–28.
28. Johansson MJ, Wiklund A, Flatebø T, Nicolaysen A, Nicolaysen
G, Walther SM. Positive end-expiratory pressure affects
regional redistribution of ventilation differently in prone and
supine sheep. Crit Care Med. 2004;32:2039–2044.
29.Glenny RW, McKinney S, Robertson HT. Spatial pattern of
pulmonary blood flow distribution is stable over days. J Appl
Physiol (1985) 1997;82:902–907.
30.Melsom MN, Flatebø T, Sjaastad OV, Aulie A, Nicolaysen
G. Minor redistribution of ventilation and perfusion within
the lung during exercise in sheep. Acta Physiol Scand.
1999;165:283–292.
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
31.Rees DD, Palmer RM, Schulz R, Hodson HF, Moncada S.
Characterization of three inhibitors of endothelial nitric oxide
synthase in vitro and in vivo. Br J Pharmacol. 1990;101:746–752.
32. Alderton WK, Cooper CE, Knowles RG. Nitric oxide synthases:
structure, function and inhibition. Biochem J. 2001;357:593–615.
33. Koizumi T, Gupta R, Banerjee M, Newman JH. Changes in pulmonary vascular tone during exercise. Effects of nitric oxide
(NO) synthase inhibition, L-arginine infusion, and NO inhalation. J Clin Invest. 1994;94:2275–2282.
34.Lorente JA, Landín L, Renes E, et al. Role of nitric oxide
in the hemodynamic changes of sepsis. Crit Care Med.
1993;21:759–767.
35. Meyer J, Lentz CW, Herndon DN, Nelson S, Traber LD, Traber
DL. Effects of halothane anesthesia on vasoconstrictor response
to NG-nitro-L-arginine methyl ester, an inhibitor of nitric oxide
synthesis, in sheep. Anesth Analg. 1993;77:1215–1221.
36.Melsom MN, Flatebø T, Kramer-Johansen J, et al. Both
gravity and non-gravity dependent factors determine
regional blood flow within the goat lung. Acta Physiol Scand.
1995;153:343–353.
37. Robertson HT, Glenny RW, Stanford D, McInnes LM, Luchtel
DL, Covert D. High-resolution maps of regional ventilation
utilizing inhaled fluorescent microspheres. J Appl Physiol (1985)
1997;82:943–953.
38.Altemeier WA, Robertson HT, Glenny RW. Pulmonary gasexchange analysis by using simultaneous deposition of
aerosolized and injected microspheres. J Appl Physiol (1985)
1998;85:2344–2351.
39. Glenny RW, Bauer C, Hofmanninger J, Lamm WJ, Krueger MA,
Beichel RR. Heterogeneity and matching of ventilation and
perfusion within anatomical lung units in rats. Respir Physiol
Neurobiol. 2013;189:594–606.
40.Vidal Melo MF, Layfield D, Harris RS, et al. Quantification
of regional ventilation-perfusion ratios with PET. J Nucl Med.
2003;44:1982–1991.
December 2016 • Volume 123 • Number 6
41.Henderson AC, Sá RC, Theilmann RJ, Buxton RB, Prisk GK,
Hopkins SR. The gravitational distribution of ventilation-perfusion ratio is more uniform in prone than supine posture in the
normal human lung. J Appl Physiol (1985) 2013;115:313–324.
42.Glenny RW, Robertson HT. Fractal properties of pulmonary
blood flow: characterization of spatial heterogeneity. J Appl
Physiol (1985) 1990;69:532–545.
4 3.Rimeika D, Wiklund NP, Lindahl SG, Wiklund CU.
Regional differences in nitric oxide-mediated vasorelaxation in porcine pulmonary arteries. Acta Anaesthesiol
Scand. 2006;50:947–953.
44.Lamm WJ, Starr IR, Neradilek B, Polissar NL, Glenny RW,
Hlastala MP. Hypoxic pulmonary vasoconstriction is heterogeneously distributed in the prone dog. Respir Physiol Neurobiol.
2004;144:281–294.
4 5.Hlastala MP, Lamm WJ, Karp A, Polissar NL, Starr IR,
Glenny RW. Spatial distribution of hypoxic pulmonary
vasoconstriction in the supine pig. J Appl Physiol (1985)
2004;96:1589–1599.
46.Gerbino AJ, Altemeier WA, Schimmel C, Glenny RW.
Endotoxemia increases relative perfusion to dorsal-caudal lung
regions. J Appl Physiol (1985) 2001;90:1508–1515.
47.Beck KC, Rehder K. Differences in regional vascular con
ductances in isolated dog lungs. J Appl Physiol (1985)
1986;61:530–538.
48.Li CG, Rand MJ. Evidence that part of the NANC relaxant
response of guinea-pig trachea to electrical field stimulation is
mediated by nitric oxide. Br J Pharmacol. 1991;102:91–94.
49. Shaul PW, North AJ, Wu LC, et al. Endothelial nitric oxide synthase is expressed in cultured human bronchiolar epithelium.
J Clin Invest. 1994;94:2231–2236.
50. Lundberg JO, Lundberg JM, Settergren G, Alving K, Weitzberg
E. Nitric oxide, produced in the upper airways, may act in an
‘aerocrine’ fashion to enhance pulmonary oxygen uptake in
humans. Acta Physiol Scand. 1995;155:467–468.
www.anesthesia-analgesia.org 1499
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.