Effects of hypoxic pulmonary vasoconstriction on pulmonary gas exchange

Brimioulle, Serge, Philippe Lejeune, and Robert Naeije.Effects of hypoxic pulmonary vasoconstriction on pulmonary gasexchange. J. Appl. Physiol. 81(4):1535-1543, 1996.Several reports have suggested that hypoxicpulmonary vasoconstriction (HPV) might result in deterioration ofpulmonary gas exchange in severe hypoxia. We therefore investigated theeffects of HPV on gas exchange in normal and diseased lungs. Weincorporated a biphasic HPV stimulus-response curve observed in intactdogs (S. Brimioulle, P. Lejeune, J. L. Vachièry, M. Delcroix, R. Hallemans, and R. Naeije, J. Appl.Physiol. 77: 476-480, 1994) into a 50-compartment lung model (J. B. West, Respir.Physiol. 7: 88-110, 1969) to control the amount ofblood flow directed to each lung compartment according to the localhypoxic stimulus. The resulting model accurately reproduced the bloodgas modifications caused by HPV changes in dogs with acute lung injury.In single lung units, HPV had a moderate protective effect on alveolaroxygenation, which was maximal at near-normal alveolarPO₂ (75-80 Torr), mixed venousPO₂ (35 Torr), andPO₂ at which hemoglobin is 50%saturated (24 Torr). In simulated diseased lungs associated with40-60 Torr arterial PO₂,however, HPV increased arterial PO₂ by 15-20 Torr. We conclude that HPV can improve arterialoxygenation substantially in respiratory failure.

     

Effects of altitude acclimatization on pulmonary gas exchange during exercise

Pulmonary gas exchange was studied in eight normal subjects both before and after 2 wk of altitude acclimatization at 3,800 m (12,470 ft, barometric pressure = 484 Torr). Respiratory and multiple inert gas tensions, ventilation, cardiac output (Q), and hemoglobin concentration were measured at rest and during three levels of constant-load cycle exercise during both normoxia [inspired PO2 (PIO2) = 148 Torr] and normobaric hypoxia (PIO2 = 91 Torr). After acclimatization, the measured alveolar-arterial PO2 difference (A-aPO2) for any given work rate decreased (P less than 0.02). The largest reductions were observed during the highest work rates and were 24.8 +/- 1.4 to 19.7 +/- 0.8 Torr (normoxia) and 22.0 +/- 1.1 to 19.4 +/- 0.7 Torr (hypoxia). This could not be explained by changes in ventilation-perfusion inequality or estimated O2 diffusing capacity, which were unaffected by acclimatization. However, Q for any given work rate was significantly decreased (P less than 0.001) after acclimatization. We suggest that the reduction in A-aPO2 after acclimatization is a result of more nearly complete alveolar/end-capillary diffusion equilibration on the basis of a longer pulmonary capillary transit time.     

Intrapulmonary shunting and pulmonary gas exchange during normoxic and hypoxic exercise in healthy humans.

Exercise-induced intrapulmonary arteriovenous shunting, as detected by saline contrast echocardiography, has been demonstrated in healthy humans. We have previously suggested that increases in both pulmonary pressures and blood flow associated with exercise are responsible for opening these intrapulmonary arteriovenous pathways. In the present study, we hypothesized that, although cardiac output and pulmonary pressures would be higher in hypoxia, the potent pulmonary vasoconstrictor effect of hypoxia would actually attenuate exercise-induced intrapulmonary shunting. Using saline contrast echocardiography, we examined nine healthy men during incremental (65 W + 30 W/2 min) cycle exercise to exhaustion in normoxia and hypoxia (fraction of inspired O(2) = 0.12). Contrast injections were made into a peripheral vein at rest and during exercise and recovery (3-5 min postexercise) with pulmonary gas exchange measured simultaneously. At rest, no subject demonstrated intrapulmonary shunting in normoxia [arterial Po(2) (Pa(O(2))) = 98 +/- 10 Torr], whereas in hypoxia (Pa(O(2)) = 47 +/- 5 Torr), intrapulmonary shunting developed in 3/9 subjects. During exercise, approximately 90% (8/9) of the subjects shunted during normoxia, whereas all subjects shunted during hypoxia. Four of the nine subjects shunted at a lower workload in hypoxia. Furthermore, all subjects continued to shunt at 3 min, and five subjects shunted at 5 min postexercise in hypoxia. Hypoxia has acute effects by inducing intrapulmonary arteriovenous shunt pathways at rest and during exercise and has long-term effects by maintaining patency of these vessels during recovery. Whether oxygen tension specifically regulates these novel pathways or opens them indirectly via effects on the conventional pulmonary vasculature remains unclear.     

Pulmonary gas exchange in humans during normobaric hypoxic exercise

Previous studies (J. Appl. Physiol. 58: 978-988 and 989-995, 1985) have shown both worsening ventilation-perfusion (VA/Q) relationships and the development of diffusion limitation during heavy exercise at sea level and during hypobaric hypoxia in a chamber [fractional inspired O2 concentration (FIO2) = 0.21, minimum barometric pressure (PB) = 429 Torr, inspired O2 partial pressure (PIO2) = 80 Torr]. We used the multiple inert gas elimination technique to compare gas exchange during exercise under normobaric hypoxia (FIO2 = 0.11, PB = 760 Torr, PIO2 = 80 Torr) with earlier hypobaric measurements. Mixed expired and arterial respiratory and inert gas tensions, cardiac output, heart rate (HR), minute ventilation, respiratory rate (RR), and blood temperature were recorded at rest and during steady-state exercise in 10 normal subjects in the following order: rest, air; rest, 11% O2; light exercise (75 W), 11% O2; intermediate exercise (150 W), 11% O2; heavy exercise (greater than 200 W), 11% O2; heavy exercise, 100% O2 and then air; and rest 20 minutes postexercise, air. VA/Q inequality increased significantly during hypoxic exercise [mean log standard deviation of perfusion (logSDQ) = 0.42 +/- 0.03 (rest) and 0.67 +/- 0.09 (at 2.3 l/min O2 consumption), P less than 0.01]. VA/Q inequality was improved by relief of hypoxia (logSDQ = 0.51 +/- 0.04 and 0.48 +/- 0.02 for 100% O2 and air breathing, respectively). Diffusion limitation for O2 was evident at all exercise levels while breathing 11% O2.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effects of acid-base status on acute hypoxic pulmonary vasoconstriction and gas exchange.

To investigate the relationship between hypoxic pulmonary vasoconstriction and respiratory and metabolic acidosis and respiratory alkalosis, the pulmonary gas exchange and pulmonary hemodynamic responses were measured in anesthetized, paralyzed, and mechanically ventilated dogs in two sets of experiments (series A, n = 6; series B, n = 10). The animals were treated with acute hypoxia, CO2 inhalation, hyperventilation, and dinitrophenol in various combinations. Multiple regression analysis indicated that mean pulmonary arterial pressure (Ppa) was significantly correlated with end-tidal PO2, mixed venous PO2, and the mean pulmonary capillary pH (average of arterial and mixed venous pH) as independent variables [series A: r = +0.999, standard error of estimate (SEE) = 0.4 mmHg; series B: r = +0.98, SEE = 1.4 mmHg]. Similar analyses of mean values published by other authors from an acute study on humans with exercise at sea level and simulated altitudes of 10,000 and 15,000 ft also indicated a good relationship (n = 14, r = +0.98, SEE = 2.1 mmHg). The mean data (n = 19) obtained in Operation Everest II at various exercise loads and simulated altitudes gave a correlation of r = +0.87, SEE = 6.1 mmHg. These empirical analyses suggest that variations in the rise of Ppa with hypoxia can be accounted for in vivo by the superimposed acid-base status. Furthermore, ventilation-perfusion inhomogeneity, as estimated in the dogs from end-tidal and arterial O2 and CO2 differences and assuming no true shunt or diffusion impairment, was highly correlated with Ppa and mean pulmonary capillary pH (r = +0.999 in series A, r = +0.77 in series B). The human data from the above studies also showed significant correlations between Ppa and directly measured ventilation-perfusion (standard deviation of perfusion obtained from inert gas measurements). These observations indicate that the beneficial effects of hyperventilation during hypoxia may be related to the marked alkalosis that serves to reduce Ppa and improve pulmonary gas exchange efficiency.     

Effects of exhaustive endurance exercise on pulmonary gas exchange and airway function in women.

Seventeen fit women ran to exhaustion (14 +/- 4 min) at a constant speed and grade, reaching 95 +/- 3% of maximal O(2) consumption. Pre- and postexercise lung function, including airway resistance [total respiratory resistance (Rrs)] across a range of oscillation frequencies, was measured, and, on a separate day, airway reactivity was assessed via methacholine challenge. Arterial O(2) saturation decreased from 97.6 +/- 0.5% at rest to 95.1 +/- 1.9% at 1 min and to 92.5 +/- 2.6% at exhaustion. Alveolar-arterial O(2) difference (A-aDO(2)) widened to 27 +/- 7 Torr after 1 min and was maintained at this level until exhaustion. Arterial PO(2) (Pa(O(2))) fell to 80 +/- 8 Torr at 1 min and then increased to 86 +/- 9 Torr at exhaustion. This increase in Pa(O(2)) over the exercise duration occurred due to a hyperventilation-induced increase in alveolar PO(2) in the presence of a constant A-aDO(2). Arterial O(2) saturation fell with time because of increasing temperature (+2.6 +/- 0.5 degrees C) and progressive metabolic acidosis (arterial pH: 7.39 +/- 0.04 at 1 min to 7.26 +/- 0.07 at exhaustion). Plasma histamine increased throughout exercise but was inversely correlated with the fall in Pa(O(2)) at end exercise. Neither pre- nor postexercise Rrs, frequency dependence of Rrs, nor diffusing capacity for CO correlated with the exercise A-aDO(2) or Pa(O(2)). Although several subjects had a positive or borderline hyperresponsiveness to methacholine, this reactivity did not correlate with exercise-induced changes in Rrs or exercise-induced arterial hypoxemia. In conclusion, regardless of the degree of exercise-induced arterial hypoxemia at the onset of high-intensity exercise, prolonging exercise to exhaustion had no further deleterious effects on A-aDO(2), and the degree of gas exchange impairment was not related to individual differences in small or large airway function or reactivity.     

Influence of inhaled nitric oxide on gas exchange during normoxic and hypoxic exercise in highly trained cyclists.

This study tested the effects of inhaled nitric oxide [NO; 20 parts per million (ppm)] during normoxic and hypoxic (fraction of inspired O(2) = 14%) exercise on gas exchange in athletes with exercise-induced hypoxemia. Trained male cyclists (n = 7) performed two cycle tests to exhaustion to determine maximal O(2) consumption (VO(2 max)) and arterial oxyhemoglobin saturation (Sa(O(2)), Ohmeda Biox ear oximeter) under normoxic (VO(2 max) = 4.88 +/- 0.43 l/min and Sa(O(2)) = 90.2 +/- 0.9, means +/- SD) and hypoxic (VO(2 max) = 4.24 +/- 0.49 l/min and Sa(O(2)) = 75.5 +/- 4.5) conditions. On a third occasion, subjects performed four 5-min cycle tests, each separated by 1 h at their respective VO(2 max), under randomly assigned conditions: normoxia (N), normoxia + NO (N/NO), hypoxia (H), and hypoxia + NO (H/NO). Gas exchange, heart rate, and metabolic parameters were determined during each condition. Arterial blood was drawn at rest and at each minute of the 5-min test. Arterial PO(2) (Pa(O(2))), arterial PCO(2), and Sa(O(2)) were determined, and the alveolar-arterial difference for PO(2) (A-aDO(2)) was calculated. Measurements of Pa(O(2)) and Sa(O(2)) were significantly lower and A-aDO(2) was widened during exercise compared with rest for all conditions (P < 0.05). No significant differences were detected between N and N/NO or between H and H/NO for Pa(O(2)), Sa(O(2)) and A-aDO(2) (P > 0.05). We conclude that inhalation of 20 ppm NO during normoxic and hypoxic exercise has no effect on gas exchange in highly trained cyclists.     

Effect of acute increases in pulmonary vascular pressures on exercise pulmonary gas exchange.

The purpose of this study was to determine the effect of acute increases in pulmonary vascular pressures, caused by the application of lower-body positive pressure (LBPP), on exercise alveolar-to-arterial PO2 difference (A-aDO2), anatomical intrapulmonary (IP) shunt recruitment, and ventilation. Eight healthy men performed graded upright cycling to 90% maximal oxygen uptake under normal conditions and with 52 Torr (1 psi) of LBPP. Pulmonary arterial (PAP) and pulmonary artery wedge pressures (PAWP) were measured with a Swan-Ganz catheter. Arterial blood samples were obtained from a radial artery catheter, cardiac output was calculated by the direct Fick method, and anatomical IP shunt was determined by administering agitated saline during continuous two-dimensional echocardiography. LBPP increased both PAP and PAWP while upright at rest, and at all points during exercise (mean increase in PAP and PAWP 3.7 and 4.0 mmHg, respectively, P<0.05). There were no differences in exercise oxygen uptake or cardiac output between control and LBPP. Despite the increased PAP and PAWP with LBPP, A-aDO2 was not affected. In the upright resting position, there was no evidence of shunt in the control condition, whereas LBPP caused shunt in one subject. At the lowest exercise workload (75 W), shunt occurred in three subjects during control and in four subjects with LBPP. LBPP did not affect IP shunt recruitment during subsequent higher workloads. Minute ventilation and arterial PcO2 were not consistently affected by LBPP. Therefore, small acute increases in pulmonary vascular pressures do not widen exercise A-aDO2 or consistently affect IP shunt recruitment or ventilation.     

Muscle oxygenation and pulmonary gas exchange kinetics during cycling exercise on-transitions in humans.

Near-infrared spectroscopy (NIRS) was utilized to gain insights into the kinetics of oxidative metabolism during exercise transitions. Ten untrained young men were tested on a cycle ergometer during transitions from unloaded pedaling to 5 min of constant-load exercise below (VT) the ventilatory threshold. Vastus lateralis oxygenation was determined by NIRS, and pulmonary O2 uptake (Vo --> Vo2) was determined breath-by-breath. Changes in deoxygenated hemoglobin + myoglobin concentration Delta[deoxy(Hb + Mb)] were taken as a muscle oxygenation index. At the transition, [Delta[deoxy(Hb + Mb)]] was unmodified [time delay (TD)] for 8.9 +/- 0.5 s at VT (both significantly different from 0) and then increased, following a monoexponential function [time constant (tau) = 8.5 +/- 0.9 s for VT]. For >VT a slow component of Delta[deoxy(Hb + Mb)] on-kinetics was observed in 9 of 10 subjects after 75.0 +/- 14.0 s of exercise. A significant correlation was described between the mean response time (MRT = TD + tau) of the primary component of Delta[deoxy(Hb + Mb)] on-kinetics and the tau of the primary component of the pulmonary Vo2 on-kinetics. The constant muscle oxygenation during the initial phase of the on-transition indicates a tight coupling between increases in O2 delivery and O2 utilization. The lack of a drop in muscle oxygenation at the transition suggests adequacy of O2 availability in relation to needs.     

Frequency domain analysis of ventilation and gas exchange kinetics in hypoxic exercise.

The kinetics of O2 up-take (VO2), CO2 output (VCO2), ventilation (VE), and heart rate (HR) were studied during exercise in normoxia and hypoxia [inspired O2 fraction (FIO2) 0.14]. Eight male subjects each completed 6 on- and off-step transitions in work rate (WR) from low (25 W) to moderate (100-125 W) levels and a pseudorandom binary sequence (PRBS) exercise test in which WR was varied between the same WRs. Breath-by-breath data were linearly interpolated to yield 1-s values. After the first PRBS cycle had been omitted as a warm-up, five cycles were ensemble-averaged before frequency domain analysis by standard Fourier methods. The step data were fit by a two-component (three for HR) exponential model to estimate kinetic parameters. In the steady state of low and moderate WRs, each value of VO2, VCO2, VE, and HR was significantly greater during hypoxic than normoxic exercise (P less than 0.05) with the exception of VCO2 (low WR). Hypoxia slowed the kinetics of VO2 and HR in on- and off-step transitions and speeded up the kinetics of VCO2 and VE in the on-transition and of VE in the off-transition. Frequency domain analysis confined to the range of 0.003-0.019 Hz for the PRBS tests indicated reductions in amplitude and greater phase shifts in the hypoxic tests for VO2 and HR at specific frequencies, whereas amplitude tended to be greater with little change in phase shift for VCO2 and VE during hypoxic tests.(ABSTRACT TRUNCATED AT 250 WORDS)     

Early onset of pulmonary gas exchange disturbance during progressive exercise in healthy active men.

Some recent studies of competitive athletes have shown exercise-induced hypoxemia to begin in submaximal exercise. We examined the role of ventilatory factors in the submaximal exercise gas exchange disturbance (GED) of healthy men involved in regular work-related exercise but not in competitive activities. From the 38 national mountain rescue workers evaluated (36 +/- 1 yr), 14 were classified as GED and were compared with 14 subjects matched for age, height, weight, and maximal oxygen uptake (VO2 max; 3.61 +/- 0.12 l/min) and showing a normal response (N). Mean arterial PO2 was already lower than N (P = 0.05) at 40% VO2 max and continued to fall until VO2 max (GED: 80.2 +/- 1.6 vs. N: 91.7 +/- 1.3 Torr). A parallel upward shift in the alveolar-arterial oxygen difference vs. %VO2 max relationship was observed in GED compared with N from the onset throughout the incremental protocol. At submaximal intensities, ideal alveolar PO2, tidal volume, respiratory frequency, and dead space-to-tidal volume ratio were identical between groups. As per the higher arterial PCO2 of GED at VO2 max, subjects with an exaggerated submaximal alveolar-arterial oxygen difference also showed a relative maximal hypoventilation. Results thus suggest the existence of a common denominator that contributes to the GED of submaximal exercise and affects the maximal ventilatory response.     

Pulmonary gas exchange during exercise in pigs

Increased ventilation-perfusion(A/)inequality is observed in ~50% of humans during heavy exercise andcontributes to the widening of the alveolar-arterialO₂ difference(A-aD_O2). Despite extensive investigation, the cause remains unknown. As a firststep to more direct examination of this problem, we developed an animalmodel. Eight Yucatan miniswine were studied at rest and duringtreadmill exercise at ~30, 50, and 85% of maximalO₂ consumption (O_{2 max}). Multipleinert-gas, blood-gas, and metabolic data were obtained. TheA-aD_O2increased from 0 ± 3 (SE) Torr at rest to 14 ± 2 Torr duringthe heaviest exercise level, but arterialPO₂(Pa_O2) remained at resting levels during exercise. There was normalA/inequality [log SD of the perfusion distribution(log) = 0.42 ± 0.04] at rest, and moderate increases(log = 0.68 ± 0.04, P < 0.0001) wereobserved with exercise. This result was reproducible on a separate day.TheA/inequality changes are similar to those reported in highly trainedhumans. However, in swine, unlike in humans, there was no inert gasevidence for pulmonary end-capillary diffusion limitation during heavyexercise; there was no systematic difference in the measuredPa_O2 and the Pa_O2 as predicted from the inertgases. These data suggest that the pig animal model iswell suited for studying the mechanism of exercise-inducedA/ inequality.

     

Effect of prolonged heavy exercise on pulmonary gas exchange in horses

During short-term maximal exercise,horses have impaired pulmonary gas exchange, manifested by diffusionlimitation and arterial hypoxemia, without marked ventilation-perfusion(A/)inequality. Whether gas exchange deteriorates progressively duringprolonged submaximal exercise has not been investigated. Sixthoroughbred horses performed treadmill exercise at ~60% of maximaloxygen uptake until exhaustion (28-39 min). Multipleinert gas, blood-gas, hemodynamic, metabolic rate, and ventilatory datawere obtained at rest and 5-min intervals during exercise. Oxygenuptake, cardiac output, and alveolar-arterialPO₂ gradient were unchanged after thefirst 5 min of exercise. Alveolar ventilation increased progressivelyduring exercise, from increased tidal volume and respiratory frequency,resulting in an increase in arterialPO₂ and decrease in arterialPCO₂. At rest there was minimal A/inequality, log SD of the perfusion distribution (logSD) = 0.20. This doubled by 5 min of exercise (logSD = 0.40) butdid not increase further. There was no evidence of alveolar-end-capillary diffusion limitation during exercise. However, there was evidence for gas-phase diffusion limitation at all time points, and enflurane was preferentially overretained. Horses maintainexcellent pulmonary gas exchange during exhaustive, submaximal exercise. AlthoughA/inequality is greater than at rest, it is less than observed in mostmammals and the effect on gas exchange is minimal.

     

Effects of human pregnancy and aerobic conditioning on alveolar gas exchange during exercise

This study examined the effects of aerobic conditioning during the second and third trimesters of human pregnancy on ventilatory responses to graded cycling. Previously sedentary pregnant women were assigned randomly to an exercise group (n = 14) or a nonexercising control group (n = 14). Data were collected at 15-17 weeks, 25-27 weeks and 34-36 weeks of pregnancy. Testing involved 20 W.min-1 increases in work rate to a heart rate of 170 beats.min-1 and (or) volitional fatigue. Breath-by-breath ventilatory and alveolar gas exchange measurements were compared at rest, a standard submaximal .VO2 and peak exercise. Within both groups, resting .V(E), .V(A), and V(T)/T(I) increased significantly with advancing gestation. Peak work rate, O2 pulse (.VO2/HR), .V(E), .V(A) respiratory rate, V(T)/T(I), .VO2, .VCO2, and the ventilatory threshold (T(vent)) were increased after physical conditioning. Chronic maternal exercise has no significant effect on pregnancy-induced changes in ventilation and (or) alveolar gas exchange at rest or during standard submaximal exercise. Training-induced increases in T(vent) and peak oxygen pulse support the efficacy of prenatal fitness programs to improve maternal work capacity.     

Effect of beta-adrenergic blockade during exercise on ventilation and gas exchange.

The ventilatory effects of beta-adrenergic blockade during steady-state exercise were studied in eight normal subjects using intravenous propranolol hydrochloride (0.2 mg/kg). Heart rate decreased in all subjects by an average of 17%. Coincident with the phase of decreasing heart rate was a significant decrease in both minute ventilation (VE) and CO2 output (VCO2), averaging 9.6 and 9.2%, respectively. Both functions returned to prepropranolol levels after heart rate had reached its reduced steady-state value. The change in VE was significantly correlated with the change in VCO2 (r = 0.85, P less than 0.005), and was associated with negligible changes in endtidal CO2 tensions and ventilatory equivalents for CO2. We interpret these studies as showing that the transient isocapnic hypopnea concomitant with an acute reduction in cardiac output was secondary to a transient decrease in CO2 flux (cardiac output x mixed venous CO2 content). This decrease in VE appears to be induced by the acute decrease in cardiac output ("cardiodynamic hypopnea"), in fashion similar to the previously described cardiodynamic hyperpnea.     

Effect of prolonged, heavy exercise on pulmonary gas exchange in athletes

During maximalexercise, ventilation-perfusion inequality increases, especially inathletes. The mechanism remains speculative. Wehypothesized that, if interstitial pulmonary edema is involved, prolonged exercise would result in increasing ventilation-perfusion inequality over time by exposing the pulmonary vascular bed to highpressures for a long duration. The response to short-term exercise wasfirst characterized in six male athletes [maximal O₂ uptake(O_{2 max}) = 63 ml · kg¹ · min¹] by using 5 minof cycling exercise at 30, 65, and 90%O_{2 max}. Multiple inert-gas, blood-gas, hemodynamic, metabolic rate, and ventilatory data were obtained. Resting log SD of the perfusion distribution (logSD) was normal [0.50 ± 0.03 (SE)] and increased with exercise (logSD = 0.65 ± 0.04, P < 0.005), alveolar-arterialO₂ difference increased (to 24 ± 3 Torr), and end-capillary pulmonary diffusion limitation occurred at 90%O_{2 max}. The subjectsrecovered for 30 min, then, after resting measurements were taken,exercised for 60 min at ~65%O_{2 max}.O₂ uptake, ventilation, cardiacoutput, and alveolar-arterial O₂difference were unchanged after the first 5 min of this test, but logSD increased from0.59 ± 0.03 at 5 min to 0.66 ± 0.05 at 60 min(P < 0.05), without pulmonary diffusion limitation. LogSD was negativelyrelated to total lung capacity normalized for body surface area(r = 0.97,P < 0.005 at 60 min). These data are compatible with interstitial edema as a mechanism and suggest that lungsize is an important determinant of the efficiency of gas exchangeduring exercise.

     

Mechanisms of improvement in pulmonary gas exchange during isovolemic hemodilution.

Severe anemia is associated with remarkable stability of pulmonary gas exchange (S. Deem, M. K. Alberts, M. J. Bishop, A. Bidani, and E. R. Swenson. J. Appl. Physiol. 83: 240-246, 1997), although the factors that contribute to this stability have not been studied in detail. In the present study, 10 Flemish Giant rabbits were anesthetized, paralyzed, and mechanically ventilated at a fixed minute ventilation. Serial hemodilution was performed in five rabbits by simultaneous withdrawal of blood and infusion of an equal volume of 6% hetastarch; five rabbits were followed over a comparable time. Ventilation-perfusion (VA/Q) relationships were studied by using the multiple inert-gas-elimination technique, and pulmonary blood flow distribution was assessed by using fluorescent microspheres. Expired nitric oxide (NO) was measured by chemiluminescence. Hemodilution resulted in a linear fall in hematocrit over time, from 30 +/- 1.6 to 11 +/- 1%. Anemia was associated with an increase in arterial PO(2) in comparison with controls (P < 0.01 between groups). The improvement in O(2) exchange was associated with reduced VA/Q heterogeneity, a reduction in the fractal dimension of pulmonary blood flow (P = 0.04), and a relative increase in the spatial correlation of pulmonary blood flow (P = 0. 04). Expired NO increased with anemia, whereas it remained stable in control animals (P < 0.0001 between groups). Anemia results in improved gas exchange in the normal lung as a result of an improvement in overall VA/Q matching. In turn, this may be a result of favorable changes in pulmonary blood flow distribution, as assessed by the fractal dimension and spatial correlation of blood flow and as a result of increased NO availability.