JOURNALOF
Vol.
APPLIED
38, No. 6, June
PHYSIOLOGY Printed
1975.
Strength
in U.S.A.
of pulmonary
to regional
alveolar
vascular
response
hypoxia
C. A. HALES, B. AHLUWALIA, AND H. KAZEMI Medical Services (Pulmonary Unit), Massachusetts General Hospital and Department of Medicine, Harvard Medical School, Boston, Massachusetts 02114
HALES, C. A., B. AHLUWALIA, AND H. KAZEMI. Strength of pulmonary vascular response to regional alveolar hypoxia. J. Appl. Physiol. 38(6): 1083-1087. 1975 .-Regional alveolar hypoxia in regional pulmonary vasoconstriction which the lung induces diverts blood flow from the hypoxic area. However, the predominant determinant of the distribution of perfusion in the normal so that more perfusion occurs at the base than erect lung is gravity at the apex. To de termine the strength of the regional alveolar hypoxic response in diverting flow with or against the gravity gradient a divided tracheal cannula was placed in anesthetized dogs and unilateral alveolar hypoxia created by ventilating one lung with nitrogen while ventilating the other lung with oxygen to preserve normal systemic oxygenation. Scintigrams of the distribution of perfusion obtained with intravenous 13N and the MGH positron camera revealed a 34 and 32% decrease in perfusion to the hypoxic lung in the supine and erect positions and a 26y0 decrease in the decubitus position with the hypoxic lung dependent (P = 0.94 from supine shift), indicating nearly equal vasoconstriction with shift of perfusion away from the hypoxic lung in all positions. Analysis of regional shifts in perfusion revealed an equal vasoconstrictor response from apex to base in the supine position but a greater response in the lower lung zones in the erect position where perfusion was also greatest.
pulmonary
perfusion
; ni trogen
13; vasoconstriction
ALTHOUGH MULTIPLE FACTORS affect the distribution of ventilation and perfusion in the lung, a major role has been attributed to the force of gravity (12). The pulmonary vasculature, however, is also known to respond to regional alveolar hypoxia with regional vasoconstriction and diversion of blood flow away from areas of hypoxia. The strength of the vasoconstriction induced by regional alveolar hypoxia is unknown, and whether it can effectively divert perfusion from a hypoxic dependent lung zone against gravity to a better aerated upper lung zone is unclear. The alveolar hypoxic vasoconstrictor response is, for example, rendered ineffective by increases in overall Zpulmonary resistance (8). Indeed, Arborelius et al. (1) were unable to demonstrate a shift in perfusion against gravity when, in the decubitus position, the dependent lung was made hypoxic. On the other hand, Dugard and Naimark (3) and Fowler and Read (6) h ave demonstrated that generahzed hypoxemia shifts distribution of pulmonary perfusion upward, perhaps indicating that the force of the constriction in the lower lobe vessels is capable of diverting blood flow up against gravity. In these studies, however, the hypoxemia
may have created a systemic response such as increased cardiac output, which could by itself cause upward shift of perfusion without invoking a regional vasoconstriction in the lower lobes. To test the strength of the vasoconstrictor response to alveolar hypoxia in the absence of systemic hypoxemia, a dog model was used in which one lung was ventilated with oxygen to maintain normal arterial oxygenation while the other lung was ventilated with nitrogen to produce unilateral alveolar hypoxia. This lung was then studied in the supine, erect, and lateral decubitus positions to determine the strength of the vasoconstrictor response to alveolar hypoxia with and against gravity. In addition, the relative magnitude of the vasoconstrictor response was assessed in basal as well as apical lung zones to determine if the vasoconstrictor response to hypoxia is of varying magnitude in different regions of the lung as has been suggested (6). METHODS
Mongrel dogs weighing 18-20 kg were anesthetized with 30 mg/kg pentobarbital intravenously and then intubated with a double-lumen Riisch endotracheal tube. Each arm of the endotracheal tube was attached to one side of a double-volume-cycled respiratory pump (Harvard Apparatus Company, lMillis, Mass.) that had been set to deliver a tidal volume of 7.5 ml/kg to each lung at a rate of 14-16 times/min. Adequacy of tracheal division was tested by ventilating one lung with 100 % 02 and the other with 100 % Nz. Division was judged adequate if the end tidal Paz was 30 mmHg or less on the nitrogen-ventilated lung after breathing that gas for 7 min as has been described elsewhere (8, 9). Tracheal division was determined prior to each run and only those animals with adequate tracheal division throughout the experiment are reported. Arterial blood samples were obtained in hzparinized syringes and end-tidal alveolar gas was sampled simultaneously in dry glass syringes from three-way stopcocks attached to each arm of the endotracheal cannula. Paz PCO~in blood and gas samples, and blood pH were determined at 38°C by appropriate electrodes (Instrumentation Laboratory, Lexington, Mass.). Each dog was placed in the supine position on a wooden V-shaped cradle which was constructed to prevent any side-to-side rolling of the animal. The dog was attached to the cradle with multiple straps which allowed the animal
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
1084
HALES,
and frame to be moved as one unit from the supine to a new stable upright or lateral decubitus position without movement of the animal on the frame. The relative aerating volume and perfusion to each lung was determined with the radionuclide 13N and the positron camera. The camera which has previously been described in detail (2) consists of two opposing sets of 127 NaI (TI) crystals, each 2.0 by 3.8 cm, which operate in coincidence to cover an area 27 x 30 cm. The data are stored and processed by a PDP-9 computer (Digital Equipment Corp., Maynard, Mass.). With this camera both lungs are imaged but, in addition, activity can be quantitated for the entire lung or in any area of the field viewed. Measurements of perfusion were made with the intravenous injection of i3N, a cyclotron-produced positronemitting isotope with a half-life of 10 min (7). One millicurie of i3N in 20 ml of saline was injected intravenously during a 10-s breath hold at end-tidal expiration. The activity recorded by the camera depends on the evolution of the relatively insoluble radioactive isotope into alveolar air as the bolus of injected 13N is distributed to the lung by blood flow. The relative distribution of perfusion to each lung was determined by dividing the activity in each lung by the total activity in the two lungs. The distribution of perfusion was determined before and after addition of unilateral alveolar hypoxia in all three positions. In addition, a regional analysis of the distribution of perfusion and perfusion per unit volume within each lung was done in five dogs. Each lung was divided into four equal zones from apex to base which covered all of the aerating volume of that lung as viewed by the camera during steady state rebreathing of 13N. The volume image was obtained by allowing the animal to rebreathe from a spirometer containing 2 mCi of r3N in 4 liters of air. A COZ absorber was in line and 02 was added at a rate of flow adjusted to maintain a constant volume in the system. Rebreathing was continued until equilibration as determined by total counts detected by the camera had stabilized over a 60-s period. This usually required 2-3 min. To avoid atelectasis, the lungs were periodically hyperinflated to five times the tidal volume by depressing the bell on the spirometer. A 10-s breath hold at FRC was then instituted and an image obtained by the positron camera which was taken to represent the distribution of lung volume. After this, unilateral alveolar hypoxia was induced by ventilating one lung with 100 % oxygen and the other lung with 100 % Nz for 7-10 min and repeating the perfusion studies. After the measurements in the supine position the animals were then placed at random in erect and latera decubitus positions and all the studies repeated. Blood and alveolar gas tensions were measured in all experiments at the end of the period of unilateral alveolar hypoxia, just prior to determination of regional blood flow with 13N. The animals were ventilated with room air between runs and were hyperinflated to 30 cmHzO by occluding the expiratory port every 10 min to prevent atelectasis. Previous studies from this laboratory have shown that there is no difference in the vascular hypoxic response between the two lungs of a given animal and for ease of description, the
TABLE
response
Qpoxic pulmonary in six dogs
AND
Control Air Breathing)
nilateral Alveola Hypoxia
%
Position
Erect
50.2 f1.9 51.5 f2.1
Left side down Values
Chaqge in Q
%Q Right lung
Supine
KAZEMI
uasoconstrictor
1.
(Room
AHLUWALIA,
are means
49.8 f1.5 48.5 lt1.8 56.3 zk7.9 zk SEM.
49 zt4.2 53.6 f3.4
*P
’
51 k4.2 46.3 k3.4
67 rt2.0 67 Itl. 5 59 f2.7
Left lung 33 f2.4 33 Al.9 41 fl.9
-34%6* -32%‘,* -26%:
< 0.01.
FIG. 1. An example of positron Camera image of the distribution of perfusion in the lung using intravenous injections of i3N at end-tidal expiration in the anesthetized dog in the supine position. In upper half of picture is the perfusion image and in lower half is the summation of all counts in the image for each lung. Left picture represents control perfusion on room air and the right picture represents the perfusion after 10 min of ventilation of the left lung with 1OO70 nitrogen while the right lung was on 100% 02. There was a 547, reduction in perfusion to the hypoxic lung in this animal. nitrogen-ventilated lung will be referred lung (9). Statistical analysis was done by comparsion data by the t-test.
to as the
left
of the paired
RESULTS
Room air breathing. In six dogs in the supine position while breathing room air the mean arterial Paz was 77 mmHg (SEM f 5.5), PcoZ 40 mmHg (SEM St 1.7), and pH 7.34 (SEM f 0.01). The right lung received 50.2% (SEM f 1.9) of the cardiac output and the left 49.8 % (SEM f 1.5). Distribution of the relative aerating lung volume to the two lungs, as determined by rebreathing of 13N, was 49 % (SEM f 4.2) to the right lung and 51 % (SEM =t 4.2) to the left (Table 1).
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
STRENGTH
OF
HYPOXIC
VASCULAR
1085
RESPONSE
In the erect position, the right lung received 51.5 % (SEM & 2.1) of the total perfusion and 53.6 cio (SEM & 3.4) of the aerating lung volume, while the left lung received 48.5 % (SEM =t 1.76) of the perfusion and had 46.3 % (SEM f 3.4) of the volume. In the decubitus position with the left side dependent, the left or the down lung received 55.5 % (SEM =t 3.7) of the total perfusion and had 43 % (SEM f 7.9) of the lung volume. The nondependent lung received 44.5 % (SEM k 3.2) of the perfusion but had 56.3 % (SEM f 7.9) of the lung volume. The animals were resting on their left sides but were partially suspended in the decubitus position by the straps and the frame which may have prevented some of the compression of the lung by the chest wall in the dependent lung. Alveolar fwpoxia. In the supine position in six dogs, 10 min of unilateral nitrogen ventilation or alveolar hypoxia of the left lung reduced the mean perfusion to 33 % of total cardiac output (p = 0.008 from control) representing a 34 % reduction in blood flow to the left lung (Table 1). An example of a perfusion scintigram is given in Fig. 1. The mean arterial Pea was 126 mmHg (SEM rt 6.9), Pcoz 43 mmHg (SEM rt 3.6), and pH 7.31 (SEM =t 0.03). Average end-tidal Poq was 17 mmHg (SEM =t 2.4) in the left lung. In the erect position, the mean perfusion to the alveolar hypoxic left lung was 33 % of total perfusion (P = 0.006 from room air erect control) or a 32 ‘/;I reduction in blood flow, about the same order of magnitude as that in the supine position (Table 1). An example of a perfusion scintigram is given in Fig. 2. The mean arterial PoZ was 121 mmHg (SEM f 4.8), Pcos 41 mmHg (SEM f 2), and pH 7.39 (SEM f 0.02). End-tidal POT was 18 mmHg (SEM f 2.8) in the left lung. In the left side down position, the mean perfusion to the alveolar hypoxic left lung was 41 c/;, of total perfusion (I-’ = 0.007 from rcom air decubitus control) which represents a 26 % overall reduction in perfusion to the dependent
of FIO. 2. .in example of positron camera image of the distribution perfusion in the lung using intravenous injections of W at end-tidal expiration in an anesthetised dog in the erect position. Control perfusion on room air is on left and on right is perfusion during ventilation of the left lung with 100$/o NP . Hvpovic lung in this animal decreased its perfusion bv 49oj,.
FIG. 3. An example of positron Camera image of the distribution of perfusion in the left side dependent position in the anesthetized dog. Upper picture represents control perfusion on room air. Lower picture is perfusion during ventilation of the dependent lung with loo%, N2 and demonstrates in this dog a 56y0 shift of perfusion away from the dependent lung and, therefore, directly against gravity. hypoxic lung (P = 0.94 from supine shift). An example is illustrated in Fig. 3. In association with the reduction in perfusion to the hypoxic lung the mean arterial Paz was 74 mmHg (SEM f 3.2) (P = 0.002 from supine unilateral nitrogen, the Pcoa 45 mmHg (SEM f 1.7), and pH of 7.30 (SEM f 0.02). Mean end-tidal Paz was 18 mmHg (SEM f 1.3) in the left lung. EJect of alveolar h@oxia on apical versus basal lung vessels. The distribution of perfusion per unit lung volume was analyzed in four equal zones from apex to base of each lung in five dogs in the supine and erect positions before and after 10 min of unilateral alveolar hypoxia. In the supine position, perfusion per unit volume decreased uniformly from apex to base in response to unilateral alveolar hypoxia (Fig. 4). In the erect position, a gravity-dependent distribution of perfusion per unit volume in the hypoxic lung persisted in response to unilateral alveolar hypoxia. All lung zones decreased perfusion per unit volume although the greatest percentage decrease of 36 % occurred in the third zone from the apex followed by a 30 % reduction at the base (Fig. 5). The apical zone reduced its perfusion by 22 % which was significantly less than that occurring in zone 3 or 4 (P < 0.01). In the erect position, an increase in perfusion to the other
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
1086
HALES,
ROOM
r ’ lu k+ c Zb4 2; o
AIR
1.2
1.0
ALVEOLAR HYPOXIA
0.8
a
u
L
h Q
s 3
1
0.6
J
@
I UPPER ZONE
0.4’
-Lo-
cIIIII--o--
0--
I U.MID ZONE
-L-C
0 -----
“‘v-0
I L.MID ZONE
I LOWER ZONE
FIG. 4. Perfusion per unit volume to the left lung in the supine position in 5 anesthetized dogs. Control room air values are solid circles and hypoxic values are closed circles. Reduction in perfusion per unit volume is essentially the same from apex to base.
ROOM
1 .6 -
AIR
1.41,2
-
I .o -
0.8
-
0.6
-
0.4
-
_1
021-L .
UPPER ZONE
FIG. 5. Perfusion tion in 5 anesthetized and hypoxic values volume occurs from perfusion is greatest.
per
U.MID ZONE
unit volume dogs. Control are closed circles. apex to base but
L.MID ZONE
LOWER ZONE
to the left lung in the erect posiroom air values are solid circles Reduction in perfusion per unit is most marked at the base where
lung (ventilated with oxygen) occurred in all four zones but predominantly and significantly to the lowest zone which increased its perfusion per unit volume by 40 % (P = 0.000 1 compared to the apical zone). DISCUSSION
The pulmonary vasoconstrictor response to regional or, as in these experiments unilateral, alveolar hypoxia is potent, able to effectively function whether the hypoxic lung is dependent or not. In the supine and erect positions 34 %
AHLUWALIA,
AND
KAZEMI
and 32 70 of perfusion was diverted from the hypoxic lung to the oxygenated lung while 26 % of the perfusion was diverted from the hypoxic lung in the dependent position (P = 0.94 from the supine shift). Thus, the relative magnitude of the response was not affected by the position of the hypoxic lung. The hypoxic lung acts as an augmented physiologic shunt since any mixed venous blood going to it does not pick up oxygen during passage through the pulmonary capillary bed. Indeed, the mixed venous blood loses some oxygen to the alveoli of the nitrogen-ventilated lung resulting in an end capillary blood Pa,, from the nitrogen lung which is lower than mixed venous PoZ. In the supine and erect positions the alveolar hypoxic response reduced the shunt to 34 and 32 % of cardiac output allowing maintenance of a Pa,, of 126 and 12 1 mmHg, respectively. In the decubitus position the dependent lung received relatively more perfusion, 56 %, than in the other two positions. Therefore, although the magnitude of the perfusion shift by the alveolar hypoxic response was the same, the fraction of cardiac output, 41 %, continuing to perfuse the hypoxic lung was greater, producing a lower mean Pa,, of 74 mmHg in these animals. There was no significant difference in the regional pulmonary vasoconstrictor response to alveolar hypoxia in the supine position from apex to base (Fig. 4). In the erect position, however, there was a statistically significant greater vasoconstrictor response at the base than at the apex (Fig. 5), although some response continued to be present at the apex. The greater response at the base is therefore probably related to the increased perfusion in the basal vessels in the erect position. If the actual decrease in diameter which brings about the vasoconstritor responses to hypoxia is small, tending only to affect the most distended vessels, then the predominant effect of vasoconstriction would be seen in the basal vessels and less effect would be seen in the less distended apical vessels, a hypothesis supported by the findings of the present experiments. In summary, the pulmonary vessels respond equally to alveolar hypoxia from apex to base in the supine position. However, in the erect position, the vasoconstrictor response is somewhat greater at the base. In addition, the pulmonary vasoconstrictor response to alveolar hypoxia is a potent mechanism of control of the distribution of perfusion, capable of shifting perfusion away from hypoxic alveoli even when the hypoxic lung is dependent and the perfusion shift must be directly against gravity. ‘This HL-06664 Pulmonary Received
study
was supported by National Institutes of Health Grants and HL-05767. C. A. Hales is the recipient of a Young Investigator Award (HL- 17200).
for
publication
2 August
1974.
REFERENCES 1. ARBORELIUS, M.. G. LUNDIN, L. SVANBERG, AND J. G. DEFARES. Influence of unilateral hypoxia on blood flow through the lungs in man in lateral position. J. /@fll. physiol. 15 : 595-597, 1960. 2. BROWNELL, G., C. A. BURNHAM, B. HOOP, AND H. KAZEMI. Positron scintigraphy with short lived cyclotron-produced radiopharmaceuticals and a multicrystal positron camera. In: Medical Radioisotopes Scintigraphy 1972. Vienna: International Atomic Energy Agency, 1973, vol. 1, p. 3 13-330.
3. DUGARD, of pulmonary 4. FISHMAN, circulation. 5. FISHMAN,
A.,
AND A. NAIMARK. blood
flow.
J. .&pl.
Effect physiol.
of hypoxia
on
23 : 663-671,
A. P. Respiratory gases in the regulation Physiol. Rev. 41 : 234-235, 196 1. A. I?, A. HIMMELSTEIN, H.W. FRITTS,AND
Blood flow through each J. CZin. Invest. 34 : 637-646,
lung in 1955.
man
during
unilateral
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
distribution 1967.
of pulmonary A. COURNAND. hypoxia.
STRENGTH
OF
HYPOXIC
VASCULAR
1087
RESPONSE
6.
FOWLER, K. T., AND J. READ. Effect of alveolar hypoxia on zonal distribution of pulmonary blood flow. J. AppZ. Physiol. 18 : 244250, 1963. 7. GREENE, R., B. HOOP, AND H. KAZEMI. Use of 13N in studies of airway closure and regional ventilation. J. Nucl. Med. 12 : 7 19-723, 1971. 8. HALES, C. A., AND H. KAZEMT. Hypoxic vascular response of the lung: effect of aminophylline and epinephrine. Am. Rev. Respirat. Diseases 110 : 126-l 32, 1974. 9. KAZEMI, H., P. E. BRUECKE, AND E. F. PARSONS. Role of the
10. 11.
12.
autonomic nervous system in the hypoxic response of the pulmonary vascular bed. Respiration Physiol. 15 : 245-257, 1972. LLOYD, JR., T. C. Effects of alveolar hypoxia on pulmonary vascular resistance. J. k&$. Physiol. 19 : 1086-1094, 1964. STRIEDER, D. J., B. A. BARNES, S. ARONOW, P. S. RUSSELL, AND H. KAZEMI. Xenon 133 study of ventilation and perfusion in normal and transplanted dog lungs. J. k&Z. Physiol. 23 : 359-366, 1967. WEST, J. B. Regional differences in gas exchange in the lung of erect man. J. AppZ. Physiol. 17 : 893-898, 1962.
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
Vol.
APPLIED
38, No. 6, June
PHYSIOLOGY Printed
1975.
Strength
in U.S.A.
of pulmonary
to regional
alveolar
vascular
response
hypoxia
C. A. HALES, B. AHLUWALIA, AND H. KAZEMI Medical Services (Pulmonary Unit), Massachusetts General Hospital and Department of Medicine, Harvard Medical School, Boston, Massachusetts 02114
HALES, C. A., B. AHLUWALIA, AND H. KAZEMI. Strength of pulmonary vascular response to regional alveolar hypoxia. J. Appl. Physiol. 38(6): 1083-1087. 1975 .-Regional alveolar hypoxia in regional pulmonary vasoconstriction which the lung induces diverts blood flow from the hypoxic area. However, the predominant determinant of the distribution of perfusion in the normal so that more perfusion occurs at the base than erect lung is gravity at the apex. To de termine the strength of the regional alveolar hypoxic response in diverting flow with or against the gravity gradient a divided tracheal cannula was placed in anesthetized dogs and unilateral alveolar hypoxia created by ventilating one lung with nitrogen while ventilating the other lung with oxygen to preserve normal systemic oxygenation. Scintigrams of the distribution of perfusion obtained with intravenous 13N and the MGH positron camera revealed a 34 and 32% decrease in perfusion to the hypoxic lung in the supine and erect positions and a 26y0 decrease in the decubitus position with the hypoxic lung dependent (P = 0.94 from supine shift), indicating nearly equal vasoconstriction with shift of perfusion away from the hypoxic lung in all positions. Analysis of regional shifts in perfusion revealed an equal vasoconstrictor response from apex to base in the supine position but a greater response in the lower lung zones in the erect position where perfusion was also greatest.
pulmonary
perfusion
; ni trogen
13; vasoconstriction
ALTHOUGH MULTIPLE FACTORS affect the distribution of ventilation and perfusion in the lung, a major role has been attributed to the force of gravity (12). The pulmonary vasculature, however, is also known to respond to regional alveolar hypoxia with regional vasoconstriction and diversion of blood flow away from areas of hypoxia. The strength of the vasoconstriction induced by regional alveolar hypoxia is unknown, and whether it can effectively divert perfusion from a hypoxic dependent lung zone against gravity to a better aerated upper lung zone is unclear. The alveolar hypoxic vasoconstrictor response is, for example, rendered ineffective by increases in overall Zpulmonary resistance (8). Indeed, Arborelius et al. (1) were unable to demonstrate a shift in perfusion against gravity when, in the decubitus position, the dependent lung was made hypoxic. On the other hand, Dugard and Naimark (3) and Fowler and Read (6) h ave demonstrated that generahzed hypoxemia shifts distribution of pulmonary perfusion upward, perhaps indicating that the force of the constriction in the lower lobe vessels is capable of diverting blood flow up against gravity. In these studies, however, the hypoxemia
may have created a systemic response such as increased cardiac output, which could by itself cause upward shift of perfusion without invoking a regional vasoconstriction in the lower lobes. To test the strength of the vasoconstrictor response to alveolar hypoxia in the absence of systemic hypoxemia, a dog model was used in which one lung was ventilated with oxygen to maintain normal arterial oxygenation while the other lung was ventilated with nitrogen to produce unilateral alveolar hypoxia. This lung was then studied in the supine, erect, and lateral decubitus positions to determine the strength of the vasoconstrictor response to alveolar hypoxia with and against gravity. In addition, the relative magnitude of the vasoconstrictor response was assessed in basal as well as apical lung zones to determine if the vasoconstrictor response to hypoxia is of varying magnitude in different regions of the lung as has been suggested (6). METHODS
Mongrel dogs weighing 18-20 kg were anesthetized with 30 mg/kg pentobarbital intravenously and then intubated with a double-lumen Riisch endotracheal tube. Each arm of the endotracheal tube was attached to one side of a double-volume-cycled respiratory pump (Harvard Apparatus Company, lMillis, Mass.) that had been set to deliver a tidal volume of 7.5 ml/kg to each lung at a rate of 14-16 times/min. Adequacy of tracheal division was tested by ventilating one lung with 100 % 02 and the other with 100 % Nz. Division was judged adequate if the end tidal Paz was 30 mmHg or less on the nitrogen-ventilated lung after breathing that gas for 7 min as has been described elsewhere (8, 9). Tracheal division was determined prior to each run and only those animals with adequate tracheal division throughout the experiment are reported. Arterial blood samples were obtained in hzparinized syringes and end-tidal alveolar gas was sampled simultaneously in dry glass syringes from three-way stopcocks attached to each arm of the endotracheal cannula. Paz PCO~in blood and gas samples, and blood pH were determined at 38°C by appropriate electrodes (Instrumentation Laboratory, Lexington, Mass.). Each dog was placed in the supine position on a wooden V-shaped cradle which was constructed to prevent any side-to-side rolling of the animal. The dog was attached to the cradle with multiple straps which allowed the animal
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
1084
HALES,
and frame to be moved as one unit from the supine to a new stable upright or lateral decubitus position without movement of the animal on the frame. The relative aerating volume and perfusion to each lung was determined with the radionuclide 13N and the positron camera. The camera which has previously been described in detail (2) consists of two opposing sets of 127 NaI (TI) crystals, each 2.0 by 3.8 cm, which operate in coincidence to cover an area 27 x 30 cm. The data are stored and processed by a PDP-9 computer (Digital Equipment Corp., Maynard, Mass.). With this camera both lungs are imaged but, in addition, activity can be quantitated for the entire lung or in any area of the field viewed. Measurements of perfusion were made with the intravenous injection of i3N, a cyclotron-produced positronemitting isotope with a half-life of 10 min (7). One millicurie of i3N in 20 ml of saline was injected intravenously during a 10-s breath hold at end-tidal expiration. The activity recorded by the camera depends on the evolution of the relatively insoluble radioactive isotope into alveolar air as the bolus of injected 13N is distributed to the lung by blood flow. The relative distribution of perfusion to each lung was determined by dividing the activity in each lung by the total activity in the two lungs. The distribution of perfusion was determined before and after addition of unilateral alveolar hypoxia in all three positions. In addition, a regional analysis of the distribution of perfusion and perfusion per unit volume within each lung was done in five dogs. Each lung was divided into four equal zones from apex to base which covered all of the aerating volume of that lung as viewed by the camera during steady state rebreathing of 13N. The volume image was obtained by allowing the animal to rebreathe from a spirometer containing 2 mCi of r3N in 4 liters of air. A COZ absorber was in line and 02 was added at a rate of flow adjusted to maintain a constant volume in the system. Rebreathing was continued until equilibration as determined by total counts detected by the camera had stabilized over a 60-s period. This usually required 2-3 min. To avoid atelectasis, the lungs were periodically hyperinflated to five times the tidal volume by depressing the bell on the spirometer. A 10-s breath hold at FRC was then instituted and an image obtained by the positron camera which was taken to represent the distribution of lung volume. After this, unilateral alveolar hypoxia was induced by ventilating one lung with 100 % oxygen and the other lung with 100 % Nz for 7-10 min and repeating the perfusion studies. After the measurements in the supine position the animals were then placed at random in erect and latera decubitus positions and all the studies repeated. Blood and alveolar gas tensions were measured in all experiments at the end of the period of unilateral alveolar hypoxia, just prior to determination of regional blood flow with 13N. The animals were ventilated with room air between runs and were hyperinflated to 30 cmHzO by occluding the expiratory port every 10 min to prevent atelectasis. Previous studies from this laboratory have shown that there is no difference in the vascular hypoxic response between the two lungs of a given animal and for ease of description, the
TABLE
response
Qpoxic pulmonary in six dogs
AND
Control Air Breathing)
nilateral Alveola Hypoxia
%
Position
Erect
50.2 f1.9 51.5 f2.1
Left side down Values
Chaqge in Q
%Q Right lung
Supine
KAZEMI
uasoconstrictor
1.
(Room
AHLUWALIA,
are means
49.8 f1.5 48.5 lt1.8 56.3 zk7.9 zk SEM.
49 zt4.2 53.6 f3.4
*P
’
51 k4.2 46.3 k3.4
67 rt2.0 67 Itl. 5 59 f2.7
Left lung 33 f2.4 33 Al.9 41 fl.9
-34%6* -32%‘,* -26%:
< 0.01.
FIG. 1. An example of positron Camera image of the distribution of perfusion in the lung using intravenous injections of i3N at end-tidal expiration in the anesthetized dog in the supine position. In upper half of picture is the perfusion image and in lower half is the summation of all counts in the image for each lung. Left picture represents control perfusion on room air and the right picture represents the perfusion after 10 min of ventilation of the left lung with 1OO70 nitrogen while the right lung was on 100% 02. There was a 547, reduction in perfusion to the hypoxic lung in this animal. nitrogen-ventilated lung will be referred lung (9). Statistical analysis was done by comparsion data by the t-test.
to as the
left
of the paired
RESULTS
Room air breathing. In six dogs in the supine position while breathing room air the mean arterial Paz was 77 mmHg (SEM f 5.5), PcoZ 40 mmHg (SEM St 1.7), and pH 7.34 (SEM f 0.01). The right lung received 50.2% (SEM f 1.9) of the cardiac output and the left 49.8 % (SEM f 1.5). Distribution of the relative aerating lung volume to the two lungs, as determined by rebreathing of 13N, was 49 % (SEM f 4.2) to the right lung and 51 % (SEM =t 4.2) to the left (Table 1).
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
STRENGTH
OF
HYPOXIC
VASCULAR
1085
RESPONSE
In the erect position, the right lung received 51.5 % (SEM & 2.1) of the total perfusion and 53.6 cio (SEM & 3.4) of the aerating lung volume, while the left lung received 48.5 % (SEM =t 1.76) of the perfusion and had 46.3 % (SEM f 3.4) of the volume. In the decubitus position with the left side dependent, the left or the down lung received 55.5 % (SEM =t 3.7) of the total perfusion and had 43 % (SEM f 7.9) of the lung volume. The nondependent lung received 44.5 % (SEM k 3.2) of the perfusion but had 56.3 % (SEM f 7.9) of the lung volume. The animals were resting on their left sides but were partially suspended in the decubitus position by the straps and the frame which may have prevented some of the compression of the lung by the chest wall in the dependent lung. Alveolar fwpoxia. In the supine position in six dogs, 10 min of unilateral nitrogen ventilation or alveolar hypoxia of the left lung reduced the mean perfusion to 33 % of total cardiac output (p = 0.008 from control) representing a 34 % reduction in blood flow to the left lung (Table 1). An example of a perfusion scintigram is given in Fig. 1. The mean arterial Pea was 126 mmHg (SEM rt 6.9), Pcoz 43 mmHg (SEM rt 3.6), and pH 7.31 (SEM =t 0.03). Average end-tidal Poq was 17 mmHg (SEM =t 2.4) in the left lung. In the erect position, the mean perfusion to the alveolar hypoxic left lung was 33 % of total perfusion (P = 0.006 from room air erect control) or a 32 ‘/;I reduction in blood flow, about the same order of magnitude as that in the supine position (Table 1). An example of a perfusion scintigram is given in Fig. 2. The mean arterial PoZ was 121 mmHg (SEM f 4.8), Pcos 41 mmHg (SEM f 2), and pH 7.39 (SEM f 0.02). End-tidal POT was 18 mmHg (SEM f 2.8) in the left lung. In the left side down position, the mean perfusion to the alveolar hypoxic left lung was 41 c/;, of total perfusion (I-’ = 0.007 from rcom air decubitus control) which represents a 26 % overall reduction in perfusion to the dependent
of FIO. 2. .in example of positron camera image of the distribution perfusion in the lung using intravenous injections of W at end-tidal expiration in an anesthetised dog in the erect position. Control perfusion on room air is on left and on right is perfusion during ventilation of the left lung with 100$/o NP . Hvpovic lung in this animal decreased its perfusion bv 49oj,.
FIG. 3. An example of positron Camera image of the distribution of perfusion in the left side dependent position in the anesthetized dog. Upper picture represents control perfusion on room air. Lower picture is perfusion during ventilation of the dependent lung with loo%, N2 and demonstrates in this dog a 56y0 shift of perfusion away from the dependent lung and, therefore, directly against gravity. hypoxic lung (P = 0.94 from supine shift). An example is illustrated in Fig. 3. In association with the reduction in perfusion to the hypoxic lung the mean arterial Paz was 74 mmHg (SEM f 3.2) (P = 0.002 from supine unilateral nitrogen, the Pcoa 45 mmHg (SEM f 1.7), and pH of 7.30 (SEM f 0.02). Mean end-tidal Paz was 18 mmHg (SEM f 1.3) in the left lung. EJect of alveolar h@oxia on apical versus basal lung vessels. The distribution of perfusion per unit lung volume was analyzed in four equal zones from apex to base of each lung in five dogs in the supine and erect positions before and after 10 min of unilateral alveolar hypoxia. In the supine position, perfusion per unit volume decreased uniformly from apex to base in response to unilateral alveolar hypoxia (Fig. 4). In the erect position, a gravity-dependent distribution of perfusion per unit volume in the hypoxic lung persisted in response to unilateral alveolar hypoxia. All lung zones decreased perfusion per unit volume although the greatest percentage decrease of 36 % occurred in the third zone from the apex followed by a 30 % reduction at the base (Fig. 5). The apical zone reduced its perfusion by 22 % which was significantly less than that occurring in zone 3 or 4 (P < 0.01). In the erect position, an increase in perfusion to the other
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
1086
HALES,
ROOM
r ’ lu k+ c Zb4 2; o
AIR
1.2
1.0
ALVEOLAR HYPOXIA
0.8
a
u
L
h Q
s 3
1
0.6
J
@
I UPPER ZONE
0.4’
-Lo-
cIIIII--o--
0--
I U.MID ZONE
-L-C
0 -----
“‘v-0
I L.MID ZONE
I LOWER ZONE
FIG. 4. Perfusion per unit volume to the left lung in the supine position in 5 anesthetized dogs. Control room air values are solid circles and hypoxic values are closed circles. Reduction in perfusion per unit volume is essentially the same from apex to base.
ROOM
1 .6 -
AIR
1.41,2
-
I .o -
0.8
-
0.6
-
0.4
-
_1
021-L .
UPPER ZONE
FIG. 5. Perfusion tion in 5 anesthetized and hypoxic values volume occurs from perfusion is greatest.
per
U.MID ZONE
unit volume dogs. Control are closed circles. apex to base but
L.MID ZONE
LOWER ZONE
to the left lung in the erect posiroom air values are solid circles Reduction in perfusion per unit is most marked at the base where
lung (ventilated with oxygen) occurred in all four zones but predominantly and significantly to the lowest zone which increased its perfusion per unit volume by 40 % (P = 0.000 1 compared to the apical zone). DISCUSSION
The pulmonary vasoconstrictor response to regional or, as in these experiments unilateral, alveolar hypoxia is potent, able to effectively function whether the hypoxic lung is dependent or not. In the supine and erect positions 34 %
AHLUWALIA,
AND
KAZEMI
and 32 70 of perfusion was diverted from the hypoxic lung to the oxygenated lung while 26 % of the perfusion was diverted from the hypoxic lung in the dependent position (P = 0.94 from the supine shift). Thus, the relative magnitude of the response was not affected by the position of the hypoxic lung. The hypoxic lung acts as an augmented physiologic shunt since any mixed venous blood going to it does not pick up oxygen during passage through the pulmonary capillary bed. Indeed, the mixed venous blood loses some oxygen to the alveoli of the nitrogen-ventilated lung resulting in an end capillary blood Pa,, from the nitrogen lung which is lower than mixed venous PoZ. In the supine and erect positions the alveolar hypoxic response reduced the shunt to 34 and 32 % of cardiac output allowing maintenance of a Pa,, of 126 and 12 1 mmHg, respectively. In the decubitus position the dependent lung received relatively more perfusion, 56 %, than in the other two positions. Therefore, although the magnitude of the perfusion shift by the alveolar hypoxic response was the same, the fraction of cardiac output, 41 %, continuing to perfuse the hypoxic lung was greater, producing a lower mean Pa,, of 74 mmHg in these animals. There was no significant difference in the regional pulmonary vasoconstrictor response to alveolar hypoxia in the supine position from apex to base (Fig. 4). In the erect position, however, there was a statistically significant greater vasoconstrictor response at the base than at the apex (Fig. 5), although some response continued to be present at the apex. The greater response at the base is therefore probably related to the increased perfusion in the basal vessels in the erect position. If the actual decrease in diameter which brings about the vasoconstritor responses to hypoxia is small, tending only to affect the most distended vessels, then the predominant effect of vasoconstriction would be seen in the basal vessels and less effect would be seen in the less distended apical vessels, a hypothesis supported by the findings of the present experiments. In summary, the pulmonary vessels respond equally to alveolar hypoxia from apex to base in the supine position. However, in the erect position, the vasoconstrictor response is somewhat greater at the base. In addition, the pulmonary vasoconstrictor response to alveolar hypoxia is a potent mechanism of control of the distribution of perfusion, capable of shifting perfusion away from hypoxic alveoli even when the hypoxic lung is dependent and the perfusion shift must be directly against gravity. ‘This HL-06664 Pulmonary Received
study
was supported by National Institutes of Health Grants and HL-05767. C. A. Hales is the recipient of a Young Investigator Award (HL- 17200).
for
publication
2 August
1974.
REFERENCES 1. ARBORELIUS, M.. G. LUNDIN, L. SVANBERG, AND J. G. DEFARES. Influence of unilateral hypoxia on blood flow through the lungs in man in lateral position. J. /@fll. physiol. 15 : 595-597, 1960. 2. BROWNELL, G., C. A. BURNHAM, B. HOOP, AND H. KAZEMI. Positron scintigraphy with short lived cyclotron-produced radiopharmaceuticals and a multicrystal positron camera. In: Medical Radioisotopes Scintigraphy 1972. Vienna: International Atomic Energy Agency, 1973, vol. 1, p. 3 13-330.
3. DUGARD, of pulmonary 4. FISHMAN, circulation. 5. FISHMAN,
A.,
AND A. NAIMARK. blood
flow.
J. .&pl.
Effect physiol.
of hypoxia
on
23 : 663-671,
A. P. Respiratory gases in the regulation Physiol. Rev. 41 : 234-235, 196 1. A. I?, A. HIMMELSTEIN, H.W. FRITTS,AND
Blood flow through each J. CZin. Invest. 34 : 637-646,
lung in 1955.
man
during
unilateral
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
distribution 1967.
of pulmonary A. COURNAND. hypoxia.
STRENGTH
OF
HYPOXIC
VASCULAR
1087
RESPONSE
6.
FOWLER, K. T., AND J. READ. Effect of alveolar hypoxia on zonal distribution of pulmonary blood flow. J. AppZ. Physiol. 18 : 244250, 1963. 7. GREENE, R., B. HOOP, AND H. KAZEMI. Use of 13N in studies of airway closure and regional ventilation. J. Nucl. Med. 12 : 7 19-723, 1971. 8. HALES, C. A., AND H. KAZEMT. Hypoxic vascular response of the lung: effect of aminophylline and epinephrine. Am. Rev. Respirat. Diseases 110 : 126-l 32, 1974. 9. KAZEMI, H., P. E. BRUECKE, AND E. F. PARSONS. Role of the
10. 11.
12.
autonomic nervous system in the hypoxic response of the pulmonary vascular bed. Respiration Physiol. 15 : 245-257, 1972. LLOYD, JR., T. C. Effects of alveolar hypoxia on pulmonary vascular resistance. J. k&$. Physiol. 19 : 1086-1094, 1964. STRIEDER, D. J., B. A. BARNES, S. ARONOW, P. S. RUSSELL, AND H. KAZEMI. Xenon 133 study of ventilation and perfusion in normal and transplanted dog lungs. J. k&Z. Physiol. 23 : 359-366, 1967. WEST, J. B. Regional differences in gas exchange in the lung of erect man. J. AppZ. Physiol. 17 : 893-898, 1962.
Downloaded from www.physiology.org/journal/jappl at Midwestern Univ Lib (132.174.254.157) on February 14, 2019.
