Content uploaded by Michael K Stickland
Author content
All content in this area was uploaded by Michael K Stickland on Sep 03, 2017
Content may be subject to copyright.
Direct demonstration of 25- and 50-m arteriovenous pathways in healthy
human and baboon lungs
Andrew T. Lovering,
1,2
Michael K. Stickland,
1
Amy J. Kelso,
2
and Marlowe W. Eldridge
1,2,3
1
John Rankin Laboratory of Pulmonary Medicine; and
2
Departments of Pediatrics, and
3
Biomedical
Engineering, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin
Submitted 19 September 2006; accepted in final form 27 November 2006
Lovering AT, Stickland MK, Kelso AJ, Eldridge MW. Direct
demonstration of 25- and 50-m arteriovenous pathways in
healthy human and baboon lungs. Am J Physiol Heart Circ Physiol
292: H1777–H1781, 2007. First published December 1, 2006;
doi:10.1152/ajpheart.01024.2006.—Postmortem microsphere studies
in adult human lungs have demonstrated the existence of intrapulmo-
nary arteriovenous pathways using nonphysiological conditions. The
aim of the current study was to determine whether large diameter
(⬎25 and 50 m) intrapulmonary arteriovenous pathways are func-
tional in human and baboon lungs under physiological perfusion and
ventilation pressures. We used fresh healthy human donor lungs
obtained for transplantion and fresh lungs from baboons (Papio c.
anubis). Lungs were ventilated with room air by using a peak inflation
pressure of 15 cmH
2
O and a positive end-expiratory pressure of 5
cmH
2
O. Lungs were perfused between 10 and 20 cmH
2
O by using a
phosphate-buffered saline solution with 5% albumin. We infused a
mixture of 25- and 50-m microspheres (0.5 and 1 million total for
baboons and human studies, respectively) into the pulmonary artery
and collected the entire pulmonary venous outflow. Under these
conditions, evidence of intrapulmonary arteriovenous anastomoses
was found in baboon (n⫽3/4) and human (n⫽4/6) lungs. In those
lungs showing evidence of arteriovenous pathways, 50-m micro-
spheres were always able to traverse the pulmonary circulation, and
the fraction of transpulmonary passage ranged from 0.0003 to 0.42%.
These data show that intrapulmonary arteriovenous pathways ⬎50
m in diameter are functional under physiological ventilation and
perfusion pressures in the isolated lung. These pathways provide an
alternative conduit for pulmonary blood flow that likely bypasses the
areas of gas exchange at the capillary-alveolar interface that could
compromise both gas exchange and the ability of the lung to filter out
microemboli.
intrapulmonary arteriovenous anastomoses; shunt; baboon; pulmo-
nary circulation
RECENT WORK has demonstrated the transpulmonary passage of
saline contrast microbubbles during exercise, but not at rest, in
healthy adult human subjects without cardiac anomalies, sug-
gesting that intrapulmonary arteriovenous pathways are re-
cruited during hyperdynamic conditions (5, 14, 22). Although
the actual size of the saline contrast microbubbles is not
known, theoretical and experimental data have estimated that
the size distribution of microbubbles that survive to enter the
pulmonary microcirculation to be 60 to 90 m in diameter (8,
10, 16, 31, 32). These estimates prompted Eldridge et al. (5) to
suggest that these inducible intrapulmonary arteriovenous
pathways must be at least 60 m in diameter. Postmortem
studies in adult human lungs have documented the existence of
intrapulmonary arteriovenous pathways for more than 100
years (20). More recent studies using microspheres have dem-
onstrated that these vessels are functional under various con-
ditions. For example, Tobin and Zariquiey (25) demonstrated
that arteriovenous anastomoses were functional in atelectatic,
fresh human lungs with perfusion pressures ranging from 50 to
300 mmHg. The authors poured “several hundred” glass mi-
crospheres ranging from 10 to 750 m in diameter into the
perfusate and found that an average of 38 glass microspheres
with diameters up to 500 m traversed the pulmonary circu-
lation in 45% (9/20) of the lungs (25). In a subsequent study,
Tobin (24) studied 10 human lungs that were previously
frozen. These atelectatic lungs were perfused with 2,000 –
5,000 glass microspheres 200 ⫾25 m in diameter at an
unreported perfusion pressure. The microsphere injection was
followed by a perfusion of a radiopaque formalin with visual-
ization of microspheres in the pulmonary veins (24). Although
these studies demonstrate the existence of intrapulmonary
arteriovenous pathways, the experimental perfusion and infla-
tion pressures were either not reported or nonphysiological.
Thus it is not known if these anastomoses are functional under
physiological conditions. Furthermore, because the authors did
not report the amount of microspheres able to traverse the
pulmonary circulation relative to the total number of each size
infused, the relative number of the intrapulmonary arterio-
venous pathways cannot be determined. Accordingly, the aim
of the current study was to determine whether or not large-
diameter (⬎50 m) intrapulmonary arteriovenous pathways
are functional in isolated human and non-human primate lungs
under physiological ventilation and perfusion pressures and to
determine the percentage of microspheres able to traverse the
pulmonary circulation under these conditions.
METHODS
All procedures involving animals were approved by the University
of Wisconsin Animal Care and Use Committee. The human lungs
were originally procured for organ donation and key identifiers were
not obtained; therefore, the human lung studies received an exemption
from the University of Wisconsin Health Sciences Institutional Re-
view Board.
Animal lung preparation. Non-human primate lungs were har-
vested from baboons (Papio c. anubis)(n⫽4, 12.6 –15.6 kg body wt,
all male). All animals were anesthetized and subsequently heparinized
and euthanized by exsanguination. Lungs were gravity flushed in situ
via a peripheral vein with 2 liters of lactated Ringer. The trachea was
cannulated with a cuffed endotracheal tube, the trachea was clamped
Address for reprint requests and other correspondence: A. T. Lovering, John
Rankin Laboratory of Pulmonary Medicine, Dept. of Pediatrics and Population
Health Sciences, Univ. of Wisconsin School of Medicine and Public Health,
Rm. 4245 MSC, 1300 Univ. Ave., Madison, WI 53706-1532 (E-mail:
atlovering@wisc.edu).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Am J Physiol Heart Circ Physiol 292: H1777–H1781, 2007.
First published December 1, 2006; doi:10.1152/ajpheart.01024.2006.
0363-6135/07 $8.00 Copyright ©2007 the American Physiological Societyhttp://www.ajpheart.org H1777
by 10.220.33.4 on September 2, 2017http://ajpheart.physiology.org/Downloaded from
to maintain the lungs at functional residual capacity, and the lungs and
heart were removed intact from the animal and immediately prepared
for ventilation and perfusion studies.
The lungs were placed on a Styrofoam platform, and the pulmonary
artery and the left atrium were cannulated in two baboons (baboons 3
and 4). The left atrium was cannulated, and the right and left branches
of the pulmonary artery were cannulated in two baboons (baboons 1
and 2) such that the left and right lobes could be perfused indepen-
dently. After vascular cannulations, the lungs were ventilated via the
endotracheal tube to a peak inflation pressure of 15 cmH
2
O, with 5
cmH
2
O positive end-expiratory pressure at 20 –25 breaths/min with an
inspiratory duration-to-total breath duration ratio of 0.5. Airflow
pressures were measured using a Grass PT5 pressure transducer.
Lungs were perfused using a nonpulsatile, reservoir system similar
to that detailed by Conhaim et al. (2, 3). The reservoir had an overflow
outlet so that the perfusion pressure could be held constant using a
pump to deliver the perfusate to the reservoir. The pressure was
measured using a Gould-Statham P23 D6 pressure transducer, which
was set to level with the pulmonary artery. Pulmonary artery perfusion
pressure was manipulated by adjusting the height of the reservoir
above the heart. The venous outflow from the left atrial cannula was
level with the heart (or pulmonary veins) so that left atrial pressure
was atmospheric, or 0 cmH
2
O. Flow rate was determined using a
graduated cylinder. Before microsphere infusion began, the lungs
were perfused at 10 –15 cmH
2
O for 5–10 min until flows were stable.
All lungs were kept at room temperature to minimize edema forma-
tion.
Human lung procurement and preparation. Healthy human lungs
(n⫽6, one female) were procured by the University of Wisconsin
Organ Procurement Team. Lungs were procured from donors who
died from either a cerebrovascular accident, cerebral anoxia, or head
trauma. All of the human lungs used in this study were appropriate for
transplantation. Most often the recipient patient condition dictated
whether a bilateral or single lobe transplant could be done safely. If a
single lobe transplant was necessary, then the surgeon would choose
the lobe in the best condition. If both lobes were in optimal condition,
then the surgeon would typically prefer to transplant the larger (right)
lung. Alternately, the lungs were split at procurement and the patient
condition did not allow for transplant. Because of our IRB waiver, we
were only provided with the donor’s age, weight, and cause of death
and were not provided with the exact reasons for nontransplantation.
The average donor age was 29 ⫾12 y (range 13– 42 y). Lungs were
gravity flushed in situ with 4 liters of University of Wisconsin (UW)
solution (ViaSpanTM, DuPont Pharmaceuticals) with 10 mg of phen-
tolamine injected into the pulmonary artery before aortic cross clamp.
Lungs were inflated and removed, and the trachea was stapled closed.
This was followed by a 2-liter gravity flush using UW solution (5 mg
phentolamine/l). The lungs were then stored in UW solution refrig-
erated on ice, and all lungs were studied within 24 h of procurement.
We received either the right or left lung (see Table 2). As detailed
above, the lungs were placed on a Styrofoam platform, the airway,
pulmonary artery, and pulmonary veins were cannulated, and the
lungs were ventilated and perfused. Edema formation during experi-
mentation was determined by visual inspection, and lungs 4 and 5
appeared to become edematous during the experimentation.
Microsphere studies. The perfusate for all studies was phosphate-
buffered saline (PBS) with 5% albumin to prevent alveolar edema. All
lungs were ventilated with a peak inflation pressure of 15 cmH
2
O and
perfused at 20 cmH
2
O except for the left lungs of baboons 1 and 2,
which were perfused at 10 cmH
2
O. We used 25 ⫾2.5 and 50 ⫾4m
(means ⫾SD) fluorescent-labeled (red and green, respectively) mi-
crospheres (Duke Scientific), which were combined in solution and
injected together. The number of dry microspheres per gram for a
given vial of microspheres was provided by Duke scientific. A stock
solution of 1 million microspheres/ml (10 ml of PBS total volume, 1%
albumin) was prepared so that the required amount of microspheres
could then be withdrawn from the stock solution. Five hundred
thousand total microspheres were used for baboon studies (250 K of
each size), and one million microspheres (500 K of each size) were
used for human studies. The solution of microspheres (5 ml total
volume of PBS, 1% albumin to prevent microspheres from adhering
to each other) was infused at 1 ml/min into the pulmonary artery
cannula using a constant infusion pump via an infusion plug. The
microsphere solution was continuously agitated during the infusion to
maintain an even distribution within the solution. Venous outflow was
collected beginning at infusion and continuing for 1 min following the
termination of the infusion for 6 min total. Venous outflow rates (i.e.,
pulmonary blood flow) at a given perfusion pressure were obtained
pre- and postmicrosphere injection to estimate the degree of emboli-
zation (i.e., percent change on venous outflow) caused by the micro-
spheres. Paired t-test was used to determine whether changes in flow
rates were significant and significance was set at P⬍0.05.
Microspheres analysis. The vials containing the venous outflow
samples were gently agitated to ensure a uniform distribution of
microspheres and then were immediately vacuum filtered using a
42-mm total diameter, 36-mm functional diameter, 0.45-m pore
filter (Millipore). Filters were imaged using fluorescent microscopy
(Nikon Eclipse 50i, EXFO X-Cite 120 fluorescent illumination sys-
tem; Spot Camera v7.4 slider, Spot Advanced software v4.1) to
determine the number and sizes of microspheres not trapped in the
lung. Microspheres were counted manually on a given filter. If ⬎500
microspheres of a given size were found on a filter, then microspheres
from those filters were resuspended in PBS (1% albumin), and
refiltered in aliquots so that there would be ⬍500 microspheres on a
given filter. Resuspension was necessary for human lung 1.
RESULTS
Pressure-flow relationships pre- and postmicrosphere injec-
tion and the percentages of microspheres that traversed the
lungs are listed in Tables 1 (baboon) and 2 (human).
Baboon lungs. The left and right lung from baboons 1 and 2
were perfused individually. The percentages of 25- and 50-m
microspheres that traversed these lungs ranged from 0.003% to
0.2% (Table 1). Lungs from baboons 3 and 4were perfused as
a pair. The percentages of 25- and 50-m microspheres that
traversed these lungs ranged from 0 to 0.01% (Table 1). Pre-
and postmicrosphere injection venous outflow rates revealed
variable changes in individual and paired lobe perfusions, but
the changes in flow were not significant.
Human lungs. Left lung shunt fractions ranged from 0 to
0.29% for 25-m microspheres and 0 to 0.42% for 50-m
microspheres (Table 2). Right lung shunt fractions ranged from
0.0005 to 0.01% for 25-m microspheres and 0.0003 to 0.02%
for 50-m microspheres (Table 2). Venous outflow rates re-
vealed that flow was reduced on average after microsphere
injection (Table 2), but these reductions were not significant.
Lungs 4 and 5became slightly edematous during the experi-
ment, despite the use of identical conditions for previous lungs,
which did not become edematous.
Across species, there was no significant relationship be-
tween the percent change in venous outflow and the percentage
of microspheres that were able to traverse the pulmonary
circulation (r⫽⫺0.18 and ⫺0.16 for 25- and 50-m micro-
spheres, respectively). Thus lungs with greater reductions in
venous outflow (i.e., a greater degree of embolization) did not
consistently demonstrate more transpulmonary passage of mi-
crospheres.
H1778 INTRAPULMONARY ARTERIOVENOUS ANASTOMOSES IN HEALTHY LUNGS
AJP-Heart Circ Physiol •VOL 292 •APRIL 2007 •www.ajpheart.org
by 10.220.33.4 on September 2, 2017http://ajpheart.physiology.org/Downloaded from
DISCUSSION
Our findings directly demonstrated that intrapulmonary ar-
teriovenous pathways ⬎25 and 50 m are functional under
physiological perfusion and ventilation pressures in human and
baboon lungs.
Transpulmonary passage via arteriovenous anastomoses or
distended capillaries? Under physiological perfusion pres-
sures, microspheres ⬎7–10 m should be filtered by the
capillaries and other vessels and are therefore unable to tra-
verse the pulmonary circulation in the absence of arteriovenous
anastomoses. However, overembolization due to extensive
microsphere occlusion of the pulmonary vasculature may cause
excessive regional pulmonary pressures resulting in either the
opening of an arteriovenous anastomoses or microspheres
being forced through a grossly distended capillary. In the
current study we attempted to minimize the degree of overem-
bolization. Nevertheless, there were instances where pulmo-
nary venous outflow was reduced as much as 30% following
microspheres infusion. Importantly, our perfusion system
maintained a constant pressure so that when embolization
occurred, flow was reduced and an excessively elevated per-
fusion pressure was prevented. The extent of capillary disten-
tion is unknown in humans. Estimates from animal studies
have demonstrated in horses, dogs, and rabbits that averages of
pulmonary capillary radii at high pressures are between 3.2 and
3.6 m (1, 29), and therefore it is extremely unlikely that
capillaries would distend to 25–50 m. Furthermore, we found
that there was no correlation between the degree of emboliza-
tion, measured by change in pulmonary venous outflow, and
the percentage of microspheres able to traverse the pulmonary
circulation. Thus it seems highly unlikely that microspheres
were either traveling through arteriovenous anastomoses
opened by nonphysiological pressures or through abnormally
distended capillaries.
Intrapulmonary arteriovenous pathways in primates. We
chose to examine lungs from humans and baboons because of
the difficulty in obtaining fresh healthy human lungs. Both
humans and baboons have a common evolutionary ancestry
with the hominoid lineage (humans, gorillas, chimpanzees,
etc.) and the Old World monkeys (macaques, baboons, etc.)
diverging ⬃25 to 30 million years ago (7, 17, 21). Because of
this common ancestry, we hypothesized that lungs from both
species would have intrapulmonary arteriovenous anastomoses
⬎50 m. Our data clearly support the idea that other hominids
as well as other Old World monkeys could potentially have
intrapulmonary arteriovenous conduits.
Why is there a disparity between reports of intrapulmonary
arteriovenous anastomoses in humans? Despite our current
findings and the findings of others (24 –26, 28), there are
reports that suggest these vessels do not exist in humans (12,
27). Based on our findings in exercising humans, these path-
ways appear to be recruitable since they are not open at rest but
Table 1. Pressure-flow relationships before and after microsphere injection and percent
microsphere passage in baboon lungs
Body
Weight, kg
Premicrosphere Injection Postmicrosphere Injection
Microsphere
Passage, %
P, cmH
2
OQ, ml/min R (P/Q) P, cmH
2
OQ, ml/min R (P/Q) ⌬Q (%) 25 m50m
Baboon 1
Right lung 15.6 20 68 0.29 20 ND ND ND 0.02 0.02
Left lung 10 36 0.28 10 38 0.26 6 0.1 0.2
Baboon 2
Right lung 13.4 20 72 0.28 20 72 0.28 0 0 0
Left lung 10 41 0.24 10 28 0.36 ⫺32 0 0.003
Baboon 3 14.2 20 184 0.11 20 148 0.14 ⫺20 0 0
Baboon 4 12.6 20 156 0.13 20 ND ND ND 0.01 0.01
Ind avg @ 10 cmH
2
O14.5 10 39 0.26 10 33 0.31 ⫺13 0.05 0.10
Ind avg @ 20 cmH
2
O14.5 20 70 0.29 20 72 0.28 0 0.01 0.01
Pr avg @ 20 cmH
2
O13.4 20 170 0.12 20 148 0.14 ⫺20 0.01 0.01
P, pressure; Q, flow; R, ratio of P to Q; ⌬Q, change in flow; Ind Avg, Individual lobe perfusion average; Pr Avg, paired lobe perfusion average; ND, no data
obtained.
Table 2. Pressure-flow relationships before and after microsphere injection and percent
microsphere passage in human lungs
Premicrosphere Injection Postmicrosphere Injection
Microsphere
Passage, %
P, cmH
2
OQ, ml/min R (P/Q) P, cmH
2
OQ, ml/min R (P/Q) ⌬Q (%) 25 m50m
Human 1 Left lung 20 375 0.05 20 267 0.07 ⫺29 0.29 0.42
Human 2 Right lung 20 326 0.06 20 215 0.09 ⫺34 0.01 0.002
Human 3 Right lung 20 136 0.15 20 148 0.14 9 0.0005 0.02
Human 4 Left lung 20 126 0.16 20 92 0.22 ⫺27 0 0
Human 5 Left lung 20 90 0.22 20 104 0.19 16 * 0
Human 6 Right lung 20 116 0.17 20 116 0.17 0 0 0.0003
Average 20 195 0.14 20 157 0.15 ⫺11 0.06 0.07
Abbreviations same as in Table 1. *No 25-m microspheres used.
H1779INTRAPULMONARY ARTERIOVENOUS ANASTOMOSES IN HEALTHY LUNGS
AJP-Heart Circ Physiol •VOL 292 •APRIL 2007 •www.ajpheart.org
by 10.220.33.4 on September 2, 2017http://ajpheart.physiology.org/Downloaded from
become functional during exercise (5, 14, 22). Furthermore,
these pathways are likely small in number, as suggested by our
findings of small shunt fractions. Indeed, these inducible path-
ways may be very difficult to identify with histological meth-
ods, particularly when the conditions required to open them are
essentially unknown. Our findings provide strong, compelling
evidence that pulmonary arteriovenous pathways (at least 50
m in diameter) are functional in healthy human lungs under
physiological ventilation and perfusion pressures.
Significance of intrapulmonary arteriovenous pathways in
human and baboon lungs. We have previously shown that
microbubbles do not traverse the pulmonary circulation at rest
but are able to traverse the pulmonary circulation during
exercise in the majority of healthy humans (5, 22). Eldridge
and colleagues (5) hypothesized that these bubbles were trav-
eling through exercise-inducible arteriovenous anastomoses
that were between 60 and 90 m in diameter. Their estimate of
the diameter of these vessels was based on the size of micro-
bubble that would have a survival time sufficient to traverse the
pulmonary circulation under high pressure and high flow con-
ditions (15, 30 –32). The current study supports our previous
hypothesis by directly demonstrating that there are intrapul-
monary arteriovenous vessels with functional diameters ⬎50
m in both healthy human lungs and baboons. Most likely, our
estimate of the percentage of microspheres able to traverse the
pulmonary circulation represents a minimum shunt fraction.
Indeed, with the increased pulmonary pressures and flows
associated with exercise, it is likely that more, and potentially
larger, intrapulmonary arteriovenous pathways would be re-
cruited.
We have previously hypothesized that blood traveling
through intrapulmonary arteriovenous anastomoses during ex-
ercise would add to the venous admixture and negatively
impact gas exchange inefficiency (5, 13, 22). Stickland et al.
(22) demonstrated a good correlation between the alveolar-to-
arterial oxygen difference and the onset of intrapulmonary
shunt as detected by saline contrast echocardiography. Indeed,
during heavy exercise, a shunt percentage as small as 1–2% of
cardiac output would have a significant effect on gas exchange
(5, 6, 13). In the current study, the amount of microspheres
passing through the isolated lung was ⬍1%. Importantly, the
experimental conditions we used for our isolated ventilated and
perfused lung studies are significantly different from the con-
ditions the lung endures during exercise. Our data demonstrate
that pressures and flows obtained in the isolated, ventilated,
and perfused lung, which mimic zone I and II conditions at
peak inspiration, are sufficient to open some of these pathways.
It is likely that the greater pulmonary pressures and flows
achieved during whole body exercise would create conditions
that would recruit more of these pathways, because their
patency is likely flow and/or pressure dependent (22).
Non-gas exchange consequences of arteriovenous anasto-
moses. In addition to contributing negatively to pulmonary gas
exchange, these pathways would impact the lung’s role as a
biological filter. The importance of the lung’s role as a biolog-
ical filter becomes apparent when this filter is bypassed, as in
the case of an intracardiac shunt (patent foramen ovale, PFO),
or when the filter becomes compromised, as in the case of
intrapulmonary shunting caused by a disease process such as
hereditary hemorrhagic telangiectasia (HHT). For example, it
is well known that patients with either a PFO or HHT are more
susceptible to embolic stroke and neurological symptoms (4, 9,
11, 18). Furthermore, recent work has linked intrapulmonary
shunting and migraine headaches (23). Although the vessels
described in this report can only be estimated at slightly ⬎50
m in functional diameter and therefore may not be able to
pass millimeter-sized emboli, recent studies in the rat demon-
strate neurological impairment from microemboli less than 100
m (19).
To summarize, we have demonstrated here that intrapulmo-
nary arteriovenous pathways exist in healthy human and ba-
boon lungs under physiological conditions. These pathways
were found to have a functional diameter ⬎50 m in diameter,
indicating that these vessels are not capillaries. If blood flow-
ing through these conduits did not participate in gas exchange
and was 1–3% of cardiac output, the efficiency of pulmonary
gas exchange would be reduced, particularly under conditions
such as exercise. These vessels were functional under physio-
logical perfusion and ventilation pressures, but clearly, the
conditions required to open these pathways have yet to be fully
elucidated.
ACKNOWLEDGMENTS
We thank Dr. Rob Conhaim and Kal Watson for help setting up the
perfusion system in our lab. We also thank Dr. Rudolf Braun, Dr. Robert Love,
and Dr. Takushi Kohmoto for invaluable assistance.
GRANTS
Support for this project was provided by National Heart, Lung, and Blood
Institute (NHLBI) Grant HL-15469, a Grant-in-Aid from the American Heart
Association 0550176Z, and the Department of Pediatrics, University of Wis-
consin Medical School. A. T. Lovering was supported by a NHLBI Training
Grant T32 HL-07654.
REFERENCES
1. Birks EK, Mathieu-Costello O, Fu Z, Tyler WS, West JB. Comparative
aspects of the strength of pulmonary capillaries in rabbit, dog, and horse.
Respir Physiol 97: 235–246, 1994.
2. Conhaim RL, Watson KE, Heisey DM, Leverson GE, Harms BA.
Perfusion heterogeneity in rat lungs assessed from the distribution of 4-m
diameter latex particles. J Appl Physiol 94: 420 – 428, 2003.
3. Conhaim RL, Watson KE, Heisey DM, Leverson GE, Harms BA.
Thromboxane receptor analog, U-46619, redistributes pulmonary micro-
vascular perfusion in isolated rat lungs. J Appl Physiol 96: 245–252, 2004.
4. Di Tullio M, Sacco RL, Gopal A, Mohr JP, Homma S. Patent foramen
ovale as a risk factor for cryptogenic stroke. Ann Intern Med 117:
461– 465, 1992.
5. Eldridge MW, Dempsey JA, Haverkamp HC, Lovering AT, Hokan-
son JS. Exercise-induced intrapulmonary arteriovenous shunting in
healthy humans. J Appl Physiol 97: 797– 805, 2004.
6. Gledhill N, Frosese AB, Dempsey JA. Ventilation to perfusion distribu-
tion during exercise in health. In: Muscular Exercise and the Lung, edited
by Dempsey JA and Reed CE. Madison, WI: University of Wisconsin
Press, 1977, p. 325–344.
7. Goodman M, Porter CA, Czelusniak J, Page SL, Schneider H, Sho-
shani J, Gunnell G, Groves CP. Toward a phylogenetic classification of
Primates based on DNA evidence complemented by fossil evidence. Mol
Phylogenet Evol 9: 585–598, 1998.
8. Hagan AD, DeMaria AN. Clinical Applications of Two-Dimensional
Echocardiography and Cardiac Doppler. Boston, MA: Little Brown 1989.
9. Kjeldsen AD, Oxhoj H, Andersen PE, Green A, Vase P. Prevalence of
pulmonary arteriovenous malformations (PAVMs) and occurrence of
neurological symptoms in patients with hereditary haemorrhagic telangi-
ectasia (HHT). J Intern Med 248: 255–262, 2000.
10. Krowka MJ, Cortese DA. Pulmonary aspects of liver disease and liver
transplantation. Clin Chest Med 10: 593– 616, 1989.
11. Lechat P, Mas JL, Lascault G, Loron P, Theard M, Klimczac M,
Drobinski G, Thomas D, Grosgogeat Y. Prevalence of patent foramen
ovale in patients with stroke. N Engl J Med 318: 1148 –1152, 1988.
H1780 INTRAPULMONARY ARTERIOVENOUS ANASTOMOSES IN HEALTHY LUNGS
AJP-Heart Circ Physiol •VOL 292 •APRIL 2007 •www.ajpheart.org
by 10.220.33.4 on September 2, 2017http://ajpheart.physiology.org/Downloaded from
12. Liebow DD. Recent advances in pulmonary anatomy. In: Pulmonary
Structure and Function, edited by DeRuich AVS and O’Connor M.
Boston, MA: Little Brown, 1962, p. 17–19.
13. Lovering AT, Haverkamp HC, Eldridge MW. Responses and limita-
tions of the respiratory system to exercise. Clin Chest Med 26: 439 –57,
2005.
14. Lovering AT, Stickland MK, Eldridge MW. Intrapulmonary shunt
during normoxic and hypoxic exercise in healthy humans. Adv Exp Med
Biol 588: 31– 46, 2006.
15. Meerbaum S. Principles of echo contrast. In: Advances in Echo Imaging
Using Contrast Enhancement, edited by Nanda N and Schlief R. Nether-
lands: Kluwer, 1993, p. 9 –42.
16. Meltzer RS, Tickner EG, Popp RL. Why do the lungs clear ultrasonic
contrast? Ultrasound Med Biol 6: 263–269, 1980.
17. Purvis A. A composite estimate of primate phylogeny. Philos Trans R Soc
Lond B Biol Sci 348: 405– 421, 1995.
18. Ranoux D, Cohen A, Cabanes L, Amarenco P, Bousser MG, Mas JL.
Patent foramen ovale: is stroke due to paradoxical embolism? Stroke 24:
31–34, 1993.
19. Rapp JH, Pan XM, Yu B, Swanson RA, Higashida RT, Simpson P,
Saloner D. Cerebral ischemia and infarction from atheroemboli ⬍100
micron in size. Stroke 34: 1976 –1980, 2003.
20. Sappey PC. Traite´da´natomie descriptive. Paris: Delahaye, 1879.
21. Stewart CB, Disotell TR. Primate evolution- in and out of Africa. Curr
Biol 8: R582–R588, 1998.
22. Stickland MK, Welsh RC, Haykowsky MJ, Petersen SR, Anderson
WD, Taylor DA, Bouffard M, Jones RL. Intra-pulmonary shunt and
pulmonary gas exchange during exercise in humans. J Physiol 561:
321–329, 2004.
23. Thenganatt J, Schneiderman J, Hyland RH, Edmeads J, Mandzia JL,
Faughnan ME. Migraines linked to intrapulmonary right-to-left shunt.
Headache 46: 439 – 443, 2006.
24. Tobin CE. Arteriovenous shunts in the peripheral pulmonary circulation
in the human lung. Thorax 21: 197–204, 1966.
25. Tobin CE, Zariquiey MO. Arteriovenous shunts in the human lung. Proc
Soc Exp Biol Med 75: 827– 829, 1950.
26. von Hayek H. The Human Lung. Translated from Die Menschliche
Lunge, by Krahl VE. New York and London: Hafner, 1960.
27. Wagenvoort CA, Heath DE, Edwards JE. The Pathology of the Pul-
monary Vasculature. Springfield, IL: Thomas, 1964, p. 25, 108.
28. Weibel E. Blood vessel anastomoses in the human lungs. Z Zellforsch
Mikrosk Anat 50: 653– 692, 1959.
29. West JB, Mathieu-Costello O, Jones JH, Birks EK, Logemann RB,
Pascoe JR, Tyler WS. Stress failure of pulmonary capillaries in race-
horses with exercise-induced pulmonary hemorrhage. J Appl Physiol 75:
1097–1109, 1993.
30. Woods TD, Patel A. A critical review of patent foramen ovale detection
using saline contrast echocardiography: when bubbles lie. JAmSoc
Echocardiogr 19: 215–222, 2006.
31. Yang WJ. Dynamics of gas bubbles in whole blood and plasma. J Bio-
mech 4: 119 –125, 1971.
32. Yang WJ, Echigo R, Wotton DR, Hwang JB. Experimental studies of
the dissolution of gas bubbles in whole blood and plasma. I. Stationary
bubbles. J Biomech 4: 275–281, 1971.
H1781INTRAPULMONARY ARTERIOVENOUS ANASTOMOSES IN HEALTHY LUNGS
AJP-Heart Circ Physiol •VOL 292 •APRIL 2007 •www.ajpheart.org
by 10.220.33.4 on September 2, 2017http://ajpheart.physiology.org/Downloaded from
