Anesth Analg 2004;99:547-551
© 2004 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000125111.56859.7D
CRITICAL CARE AND TRAUMA
Prevention of Hemodilution-Induced Inhibition of Hypoxic Pulmonary Vasoconstriction by N-Acetylcysteine in Dogs
François Kerbaul, MD*,
Philippe Van der Linden, MD PhD
,
Sébastien Pierre, MD
,
Benoît Rondelet, MD
,
Christian Melot, MD PhD
,
Serge Brimioulle, MD PhD
, and
Robert Naeije, MD PhD
*Department of Anesthesia and Intensive Care, Timone Hospital, Marseille, France;
Department of Anesthesia, Centre Hospitalo-Universitaire, Charleroi, Belgium;
Department of Physiology, Faculty of Medicine, Free University of Brussels, Brussels, Belgium; and
Department of Intensive Care, Erasme Hospital, Free University of Brussels, Belgium
Address correspondence and reprint requests to Robert Naeije, MD, Department of Physiology, Erasme Campus CP 604, 808, Lennik Rd., B-1070 Brussels, Belgium. Address e-mail to rnaeije{at}ulb.ac.be
 |
Abstract
|
|---|
We investigated the possible contributions of reactive oxygen species and of viscosity changes to hemodilution-induced inhibition of hypoxic pulmonary vasoconstriction (HPV) in dogs. Fourteen isoflurane-anesthetized dogs were randomly assigned to receive N-acetylcysteine (NAC) 200 mg/kg IV (n = 7) or placebo (n = 7). Mean pulmonary artery pressure (Ppa) was measured with cardiac output maintained constant by a manipulation of venous return in hyperoxia (fraction of inspired oxygen, 0.4) and in hypoxia (fraction of inspired oxygen, 0.1) at baseline and after stepwise reductions in hematocrit from 40% to 20%. Measured Ppa was compared with predicted Ppa by using a viscoelastic model. HPV was expressed as hypoxic Ppa minus hyperoxic Ppa. Hemodilution was associated with a decrease in HPV from 7 ± 1 mm Hg to 3 ± 1 mm Hg (P < 0.01), and this was completely prevented by NAC (HPV was unchanged, from 8 ± 1 to 8 ± 1 mm Hg; not significant). Hemodilution in the model decreased HPV from 8 ± 1 mm Hg to 6 ± 1 mm Hg (P < 0.05). We conclude that hemodilution-induced inhibition of HPV is in part explained by viscosity changes and can be prevented by the administration of NAC, which is possibly explained by the scavenging of reactive oxygen species.
IMPLICATIONS: Hemodilution inhibits hypoxic pulmonary vasoconstriction. This study showed that this potential cause of altered gas exchange is partly explained by viscosity changes and is completely preventable by the administration of the reactive oxygen species scavenger N-acetylcysteine.
 |
Introduction
|
|---|
Isovolemic hemodilution is currently used to reduce blood transfusion requirements during surgery (1). However, a decrease in hematocrit inhibits hypoxic pulmonary vasoconstriction (HPV) (26). Therefore, hemodilution could alter the matching of pulmonary perfusion (
) to alveolar ventilation in hypoxic lung regions. Accordingly, hemodilution has been reported to increase pulmonary shunt in rabbits with left lung atelectasis (7). Thus, hemodilution could contribute to the alteration of pulmonary gas exchange and resulting hypoxemia often observed with general anesthesia.
The mechanisms of hemodilution-induced inhibition of HPV remain incompletely understood. In isolated perfused rabbit lungs, hemodilution decreased exhaled nitric oxide (NO), and hemodilution-induced inhibition of HPV can be prevented by an L-arginine analog, suggesting that scavenging of NO by a normal number of red blood cells is important to normal pulmonary vascular reactivity (6). However, red blood cells also act as reactive oxygen species (ROS) scavengers. It has been shown that ROS scavengers attenuate endotoxin-induced impairment of HPV in intact mice (8), in contrast with several in vitro studies on isolated arteries or perfused lungs that suggest that ROS mediates HPV (9). However, because changes in blood viscosity affect pulmonary vascular resistance (PVR), one could wonder whether this effect might be proportionally more important in hypoxia-induced vasoconstriction, thus accounting for the observed changes in HPV after hemodilution.
We therefore investigated the effects of antioxidant treatment by N-acetylcysteine (NAC) on hemodilution-induced changes in hyperoxic and hypoxic PVR in intact anesthetized dogs. Although most previous studies were performed on isolated perfused lungs, we reasoned that an intact animal model would allow us to observe the integrated effects of all possible adaptive reflexes and thus produce results more directly transposable to clinical anesthesia practice. Because anemia is associated with an increase in cardiac output, we defined PVR by measurements of mean pulmonary artery pressure (Ppa) at cardiac output (
) maintained constant to avoid the confounding effects of passive, flow-associated changes in pulmonary vascular pressures (10). To sort out the effects of viscosity changes on PVR, we compared the measured Ppa with that predicted by a pulmonary circulation model that explains Ppa by a distribution of compliances, resistances, and viscosities at a given level of flow (11).
 |
Methods
|
|---|
The study design was reviewed and approved by the animal ethics committee of the Free University of Brussels. All procedures complied with the Guiding Principles in the Care and Use of Animals of the American Physiological Society.
Fourteen mongrel dogs (2530 kg) were premedicated with ketamine (20 mg/kg IM) and placed supine. Anesthesia was induced with sodium pentobarbital (10 mg/kg IV) and maintained with 1.41% end-tidal isoflurane (1 minimum alveolar anesthetic concentration [MAC]) throughout the experiment. Larger concentrations of isoflurane (2.1% end-tidal; 1.5 MAC) were required only during splenectomy. Muscle paralysis was obtained with pancuronium bromide (0.2 mg/kg IV) and maintained with a continuous infusion of 0.2 mg · kg1 · h1 IV after tracheal intubation was performed. The dogs were mechanically ventilated and equipped with pulmonary and systemic catheters for hemodynamic measurements and blood sampling, as previously described (11,12). A balloon catheter (Percor Stat-DL 10.5F; Datascope, Paramus, NJ) was advanced into the inferior vena cava through a right femoral venotomy to produce a titratable decrease in cardiac output (
) (11,12). Inspired and expired fractions of oxygen, CO2, and isoflurane were measured by using an ULTIMA II infrared spectrophotometer (Datex, Helsinki, Finland). Hemodynamics and arterial blood gases were measured as previously reported (11,12). Hemoglobin concentration and saturation were measured by using a cooximeter (OSM 3; Radiometer, Copenhagen, Denmark). Body-surface area (m2) was calculated as 0.112 x weight (kg2/3). A splenectomy was performed via a midline laparotomy to prevent autotransfusion.
In each experimental condition (variation of fraction of inspired oxygen [FIO2] and/or hematocrit),
, heart rate (HR), systemic arterial pressure (Psa), Ppa, occluded Ppa (Ppao), right atrial pressure (Pra), and arterial blood gases were measured after ensuring steady-state conditions (stable Psa, Ppa, end-tidal PCO2, and HR) for 30 min. These measurements were performed at uncontrolled
and were then followed by a measurement of Ppa and Ppao at
transiently constrained to 3.5 L · min1 · m2 by stepwise inflations of the inferior vena cava balloon to control venous return.
Each animal was slowly hemodiluted by repeated withdrawals of 510 mL/kg of arterial blood and replacements by the same volume of 6% hydroxyethylstarch 200/0.5 (Haes-Stéril 6%; Fresenius AG, Bad Hombourg, Germany) from the baseline hematocrit of 40% ± 1% down to a hematocrit of 20%. When the investigated hematocrit was reached, a 30-min period of stabilization was observed before the next blood gas and hemodynamic measurements.
The animals were randomized to pretreatments with NAC or placebo. NAC (Lysomucil 10%®; Zambon Group SpA, Vicenza, Italy) was given as an IV bolus of 150 mg/kg over 15 min, followed by a continuous infusion of 50 mg/kg over 4 h. These doses have been recommended to obtain a maximal antioxidant activity in paracetamol intoxications (13). The placebo consisted of the same amount of IV vehicle physiological saline.
After the NAC or placebo infusions had been started, hemodynamic and blood gas measurements were obtained at an FIO2 of 0.4 while steady-state conditions (stable, continuously monitored Ppa, Psa, end-tidal CO2, and HR) had been ensured for at least 30 min. All the measurements were then repeated at the fifth minute of ventilation with an FIO2 of 0.1. We previously showed that a maximal HPV in intact anesthetized dogs is obtained at an FIO2 of 0.1 (12). The same sequence of hyperoxic and hypoxic measurements was then repeated after hemodilution, still under the continuous infusion of either NAC or placebo, and a stabilization period of at least 30 min was reobtained.
The Ppa at different levels of flow was calculated by using a previously reported viscoelastic pulmonary circulation model based on a distribution of compliances, resistances, and viscosities derived from cat lung morphometric studies (14) adapted to dog lungs after a scaling procedure (11). The smallest-order arteriole diameters were decreased in hypoxia (15), at unchanged initial hematocrit, to fit as closely as possible the measured Ppa at the
of 3 L · min1 · m2 in hypoxia. Thereafter, Ppa was recalculated with everything kept unchanged excepted for the hematocrit, which was decreased to 20%.
Results were expressed as means ± SEM. The hemodynamic and blood gas measurements were submitted to a two-way repeated-measures analysis of variance. When the F ratio of the analysis of variance reached a critical P < 0.05 value, the Scheffé test was used to compare specific situations (16).
 |
Results
|
|---|
Regarding HPV and hemodilution in the control group (Table 1, Fig. 1), progressive hemodilution in hyperoxia increased
and decreased Psa without changes in HR, Ppa, Ppao, and Pra. Arterial blood gases at baseline showed a pH of 7.37 ± 0.01, a PaO2 of 209 ± 8 mm Hg, and a PaCO2 of 36 ± 1 mm Hg, and these did not significantly change with hemodilution. Hypoxia increased Ppa and
without other hemodynamic changes, did not affect arterial pH, and decreased PaO2 to 37 ± 2 mm Hg and PaCO2 to 35 ± 1 mm Hg. As shown in Figure 1, hemodilution significantly attenuated HPV, and this was actually partly predicted by the viscoelastic model.
View this table:
[in this window]
[in a new window]
|
Table 1. Hemodynamics and Blood Gases at Uncontrolled Cardiac Output During Progressive Hemodilution in the Control Group (n = 7)
|
|

View larger version (24K):
[in this window]
[in a new window]
|
Figure 1. Hypoxic pulmonary vasoconstriction expressed as hypoxic transpulmonary pressure (mean pulmonary artery pressure minus occluded pulmonary artery pressure) at cardiac output maintained constant at 3.5 L · min1 · m2 at the hematocrits (Ht) of 40% (empty columns) and 20% (shaded columns) in dogs pretreated with a placebo (control) or with N-acetylcysteine (NAC). The responses to hemodilution predicted in a viscoelastic model of the pulmonary circulation are also shown. Hemodilution inhibited the hypoxic response. This was partly explained by a decrease in viscosity and was prevented by NAC. Vertical bars indicate the SEM. *P < 0.05, low hematocrit (Ht 20%) versus normal hematocrit (Ht 40%).
|
|
Regarding HPV and hemodilution in the NAC-treated group (Table 2, Fig. 1), at the initial baseline FIO2 measurement, the NAC-treated dogs tended to present with higher cardiac output and HR and slightly lower Psa, Pra, Ppao, and hemoglobin, but none of these differences reached a P < 0.05 significance. The effects of hemodilution and hypoxia on hemodynamics and blood gases were similar to those observed in the control group, except that the hemodilution-induced decrease in Psa did not reach a P < 0.05 level of significance, and hemodilution did not increase
in hypoxia. As shown in Figure 1, NAC prevented the inhibition of HPV induced by hemodilution.
View this table:
[in this window]
[in a new window]
|
Table 2. Hemodynamics and Blood Gases at Uncontrolled Cardiac Output During Progressive Hemodilution in the Group Pretreated with N-Acetylcysteine (n = 7)
|
|
 |
Discussion
|
|---|
These results show that hemodilution-induced inhibition of HPV is in part explained by decreased viscosity and can be completely prevented by NAC. In this study, vasomotor responses were defined by the measurement of the gradient between Ppa and Ppao at a cardiac output kept constant by a manipulation of venous return. Indeed, it has repeatedly been shown that pulmonary vascular pressures and PVR are flow-dependent variables due to the distribution of compliances and resistances within the pulmonary circulation and possibly also to flow-induced release of endothelium-derived vasodilating mediators (10,11). In this study, both hypoxia and hemodilution were associated, as expected, with important increases in cardiac output, obscuring the hemodilution-induced inhibition of HPV, as can be seen in Table 1. However, normalization of flow disclosed the importance of hemodilution-induced inhibition of HPV (Fig. 1).
The hypoxic pressor response varies greatly from one species to another and from one individual to another (17). Variation of HPV has been attributed to an imbalance of endothelium-derived vasodilators and vasoconstrictors, which are believed to normally modulate the response (18). We wondered whether anesthesia could have affected HPV in this study. Barbiturates may slightly decrease HPV (19), but the very short duration of action of a single bolus of thiopental (approximately seven minutes) used at the beginning of the experiments in this study makes such an effect unlikely. Isoflurane anesthesia has been reported to attenuate HPV (20). However, in these experiments, the concentrations of isoflurane did not differ from one experimental condition to another and did not exceed 1 MAC, thus allowing the preservation of a sufficient hypoxic response to be investigated.
The fact that normal pulmonary vasoreactivity requires the presence of red blood cells has intrigued researchers for years (26). Although anemia-associated accumulations of adenosine or vasodilating prostaglandins have been excluded (3,4), evidence has been gathered that a normal amount of red blood cells is necessary for the scavenging of tonic endothelium-derived endogenous NO production (6). In isolated perfused rabbit lungs, anemia is associated with an increased exhaled NO, and hemodilution-induced inhibition of HPV can be prevented by the administration of an L-arginine analog (6).
Red blood cells also act as ROS scavengers (21), and this could modulate HPV as well (22). There are several, albeit contradictory, lines of evidence that ROS are involved in pulmonary artery smooth muscle cell oxygen sensing (9). These include a reduced production of H2O2 from a microsomal nicotinamide-adenine dinucleotide (NADH) oxidase leading to decreased guanylate cyclase activation and increased vasorelaxation (23), an activation of sarcolemmal nicotinamide-adenine dinucleotide phosphate (NADPH) oxidase leading to increased superoxide anion production signaling vasoconstriction (24,25), a decrease in mitochondrial electron transport leading to decreased ROS generation, a shift of ratios of redox couples to a more reduced state leading to vasoconstriction through inhibition of Kv (voltage gated potassium) channels (2628), and activation of proximal electron chain transport generation of ROS mediating vasoconstriction (29). Although most recently the notion of a hypoxia-induced decrease in ROS was supported by observations on endothelium-denuded resistive pulmonary arteries by using three different sensitive detection methods (30), observations on pulmonary artery smooth muscle cells overexpressing catalase rather suggested H2O2 to be the signaling molecule for HPV (28). As recently emphasized, understanding the ups and downs of ROS in HPV will require further experimental work, including verification that intracellular calcium movements induced by hypoxia and H2O2 are exactly comparable (9). In the meantime, results of the administration of antioxidants in various experimental preparations will not help to solve the controversy, with reported inhibition (25,29) as well as enhancement (8,22,26,31,32) of HPV. Our results confirm previous observations on endotoxic mice (8) that antioxidant therapy in intact animals restores HPV.
NAC has been reported to enhance NO-mediated systemic vasodilation in a variety of experimental models, including the spontaneously hypertensive rat (33). NAC may also decrease adrenergic systemic vasoconstriction (33). These effects may explain the tendency to increased HR and
with decreased Psa, Pra, and Ppao in our NAC-treated dogs at baseline. However, there was no evidence of NAC-induced pulmonary vasodilation in these experiments.
Hematocrit varies along the pulmonary arterial tree (34), so the hemodilution-induced changes in hematocrit and associated changes in viscosity at the site of resistive vessels are impossible to measure directly. We therefore compared the pulmonary hemodynamic effects of hemodilution with those predicted by a viscoelastic model based on a realistic distribution of compliances, resistances, and viscosities (11). The results show that approximately one-third of the hemodilution-induced decrease in PVR in hypoxia is explained by associated changes in viscosity.
A possible clinical implication of these results is that a simple, nontoxic intervention such as the administration of NAC may prove to be a useful strategy for preserving HPV and arterial oxygenation in patients who are undergoing hemodilution and are at risk of altered ventilation/perfusion matching.
 |
Acknowledgments
|
|---|
Supported by ADEREM (Association pour le développement de la Recherche Biologique et MédicaleCHU de Marseillegrant 57083/04.2001), Marseilles, France; Foundation for Cardiac Surgery, Belgium; and Fonds de la Recherche Scientifique Médicale Grant 3.4516.02, Belgium. BR was a fellow of the Erasmus Foundation, Brussels, Belgium.
 |
References
|
|---|
- Practical guidelines for blood component therapy: a report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology 1996; 84: 73247.[Web of Science][Medline]
- Hauge A. Conditions governing the pressor response to ventilation hypoxia in isolated perfused rat lungs. Acta Physiol Scand 1968; 72: 3344.[Web of Science][Medline]
- McMurtry IF, Hookway BW, Roos SD. Red blood cells but not platelets prolong vascular reactivity of isolated rat lungs. Am J Physiol 1978; 234: H18691.
- Hakim TS, Malik AB. Hypoxic pulmonary vasoconstriction in blood and plasma perfused lungs. Respir Physiol 1988; 72: 10921.[Web of Science][Medline]
- Weissman N, Grimminger F, Walmrath D, Seeger W. Hypoxic vasoconstriction in buffer-perfused rabbit lungs. Respir Physiol 1995; 100: 15969.[Web of Science][Medline]
- Deem S, Swenson ER, Alberts MK, et al. Red-blood-cell augmentation of hypoxic pulmonary vasoconstriction: hematocrit dependence and the importance of nitric oxide. Am J Respir Crit Care Med 1998; 157: 11816.[Abstract/Free Full Text]
- Deem S, Bishop MJ, Alberts MK. Effect of anemia on intrapulmonary shunt during atelectasis in rabbits. J Appl Physiol 1995; 79: 19517.[Abstract/Free Full Text]
- Baboolal HA, Ichinose F, Ullrich R, et al. Reactive oxygen species scavengers attenuate endotoxin-induced impairment of hypoxic pulmonary vasoconstriction in mice. Anesthesiology 2002; 97: 122733.[Web of Science][Medline]
- Sham JSK. Hypoxic pulmonary vasoconstriction: ups and downs of reactive oxygen species. Circ Res 2002; 91: 64951.[Free Full Text]
- Naeije R. Pulmonary vascular resistance: a meaningless variable? Intensive Care Med 2003; 29: 5269.[Web of Science][Medline]
- Mélot C, Delcroix M, Closset J, et al. Starling resistor vs. distensible vessel models for embolic pulmonary hypertension. Am J Physiol 1995; 267: 81727.
- Brimioulle S, Lejeune P, Vachiéry JL, et al. The stimulus-response curve of hypoxic pulmonary vasoconstriction in intact dogs: effects of ASA. J Appl Physiol 1994; 77: 47680.[Abstract/Free Full Text]
- Prescott LF, Illingworth RN, Crichley JA, et al. Intravenous N-acetylcysteine: the treatment of choice for paracetamol poisoning. BMJ 1979; 2: 1097110.
- Zhuang FY, Fung YC, Yen RT. Analysis of blood flow in cats lung with detailed anatomical and elasticity data. J Appl Physiol 1983; 55: 13418.[Abstract/Free Full Text]
- Marshall BE, Marshall C. A model for hypoxic vasoconstriction of the pulmonary circulation. J Appl Physiol 1988; 64: 6877.[Abstract/Free Full Text]
- Winer BJ. Statistical principles in experimental design. New York: McGraw-Hill, 1971.
- Grover RF, Vogel JHK, Averill KH, Blount SG Jr. Pulmonary hypertension: individual and species variability relative to vascular reactivity. Am Heart J 1966; 66: 13.
- Leeman M, Zegers de Beyl V, Biarent D, et al. Inhibition of cyclooxygenase and of nitric oxide synthase in hypoxic vasoconstriction and in oleic acid lung injury. Am J Respir Crit Care Med 1999; 159: 138390.[Abstract/Free Full Text]
- Wetzel RC, Martin LD. Pentobarbital attenuates pulmonary vasoconstriction in isolated sheep lungs. Am J Physiol 1989; 257: H898903.
- Lennon PF, Murray PA. Attenuated hypoxic pulmonary vasoconstriction during isoflurane anesthesia is abolished by cyclooxygenase inhibition in chronically instrumented dogs. Anesthesiology 1996; 84: 40414.[Web of Science][Medline]
- Winterbourn CC, Stern A. Human red cells scavenge extracellular hydrogen peroxide and inhibit formation of hypochlorous acid and hydroxyl radical. J Clin Invest 1987; 80: 148691.
- Yamaguchi K, Asano K, Takasugi T, et al. Modulation of hypoxic pulmonary vasoconstriction by antioxidant enzymes in red blood cells. Am J Respir Crit Care Med 1996; 153: 2117.[Abstract]
- Mohazzab KM, Fayngersch RP, Kaminsky PM, et al. Potential role of the NADH oxidoreductase-derived reactive O2 species in calf pulmonary arterial PO2-elicited responses. Am J Physiol 1995; 269: 163744.
- Marshall C, Mamary AJ, Verhoeven AJ, et al. Pulmonary artery NADPH-oxidase is activated in hypoxic pulmonary vasoconstriction. Am J Respir Cell Mol Biol 1996; 15: 63344.[Abstract]
- Grimminger F, Weissmann N, Spriestersbach R, et al. Effects of NADPH oxidase inhibitors on hypoxic vasoconstriction in buffer-perfused rabbit lungs. Am J Physiol 1995; 268: L74752.
- Archer SL, Huang J, Henry T, et al. A redox-based O2 sensor in rat pulmonary vasculature. Circ Res 1993; 73: 110012.[Abstract/Free Full Text]
- Reeve HL, Weir EK, Nelson DP, et al. Opposing effects of oxidants and antioxidants on K+ channel activity and tone in rat vascular tissue. Exp Physiol 1995; 80: 82534.[Abstract]
- Michelakis ED, Hampl V, Nsair A, et al. Diversity in mitochondrial function explains differences in vascular oxygen sensing. Circ Res 2002; 90: 130715.[Abstract/Free Full Text]
- Waypa GB, Chandel NS, Schumacker PT. Model for hypoxic pulmonary vasoconstriction involving mitochondrial oxygen sensing. Circ Res 2001; 88: 125966.[Abstract/Free Full Text]
- Waypa GB, Marks JD, Mack MM, et al. Mitochondrial reactive oxygen species trigger calcium increases during hypoxia in pulmonary artery myocytes. Circ Res 2002; 91: 71926.[Abstract/Free Full Text]
- Archer SL, Peterson D, Nelson DP, et al. Oxygen radicals and antioxidant enzymes alter pulmonary reactivity in the rat lung. J Appl Physiol 1989; 66: 10211.[Abstract/Free Full Text]
- Abdalla S, Will JA. Potentiation of the hypoxic contraction of guinea-pig isolated pulmonary arteries by two inhibitors of superoxide dismutase. Gen Pharmacol 1995; 26: 78592.[Web of Science][Medline]
- Girouard H, Chulak C, Wu L, et al. N-acetylcysteine improves nitric oxide and alpha-adrenergic pathways in mesenteric beds of spontaneously hypertensive rats. Am J Hypertens 2003; 16: 57784.[Web of Science][Medline]
- Fahraeus R, Lindquist T. The viscosity of the blood in narrow capillary tubes. Am J Physiol 1949; 159: 4018.[Free Full Text]
Accepted for publication February 10, 2004.