| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Anesthesiology, Westfälische-Wilhelms-Universität, Münster, Germany 48149; and 2 Departments of Anesthesiology, Physiology, and Biophysics, The University of Texas Medical Branch, and The Shriners Burns Institute, Galveston, Texas 77555-0833
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
We investigated the effects of modified
hemoglobin on regional blood flow and function of different organs
during hyperdynamic sepsis. Fourteen sheep were surgically prepared for
the study. After a 5-day recovery period, a continuous infusion of live
Pseudomonas aeruginosa bacteria was
begun and maintained for 48 h. At 24 h, after a hyperdynamic
circulation had developed, the animals were randomly assigned to two
groups: 1) a treatment group
(n = 7) that received an infusion with
100 mg/kg pyridoxalated hemoglobin polyoxyethylene conjugate (PHP) over
30 min and 2) a control group (n = 7) that received only the
vehicle. PHP infusion increased mean arterial pressure from 86 ± 2.8 to 101.8 ± 3.5 mmHg (P < 0.05) and systemic vascular resistance index from 769 ± 42.1 to 1,087 ± 56.8 dyn · s · m2 · cm
5
(P < 0.05). PHP
infusion did not decrease regional blood flow, measured with
fluorescent microspheres, below the baseline values in any of the
analyzed tissues. None of the investigated blood chemistry variables
showed any changes indicative of impaired organ function after PHP
infusion. In our model of ovine sepsis we found no side effects after
PHP infusion that would limit the use of PHP as a nitric oxide
scavenger in sepsis.
sepsis; modified hemoglobin; hemoglobin; nitric oxide; regional blood flow; kidney function
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
BECAUSE OF THE HIGH INCIDENCE and mortality of sepsis and septic shock and associated high health care costs, sepsis and its treatment are prominent public health and scientific issues (19). However, recent intervention trials have found no benefit from anti-inflammatory therapies in sepsis. During the last few years it has become evident that the biogenic gas nitric oxide mediates the loss of vascular smooth muscle cell contractility in septic shock, either by directly activating guanylate cyclase or by indirectly causing a contractile and energetic failure of this muscle cell (10, 30). This vasodilatation is mainly responsible for disturbances in the macro- and microcirculation of the septic patient. Many mediators and toxins such as cytokines or endotoxins can promote the expression of the inducible form of nitric oxide synthase, which is capable of producing large amounts of nitric oxide over a prolonged period of time (28). Several investigators have shown that inhibition of nitric oxide synthase in animal models of hyperdynamic sepsis normalizes mean arterial pressure and systemic vascular resistance or even improves survival (5, 29).
Nitric oxide also mediates the hypertensive response to modified hemoglobin (26). In contrast to hemoglobin in red blood cells, extracellular hemoglobin rapidly penetrates the endothelial barrier of the vascular wall and reacts with nitric oxide (1). The extremely rapid reaction between oxyhemoglobin and nitric oxide results in the formation of methemoglobin and nitrate (1). When modified hemoglobin is used as a red blood cell substitute, vasoconstriction resulting from nitric oxide scavenging would be an unwelcome side effect. However, in hyperdynamic sepsis with increased nitric oxide production, the nitric oxide-scavenging effects of modified hemoglobin could be advantageous.
Recently, it was demonstrated that cell-free hemoglobin can reverse the
endotoxin-mediated hyporesponsivity of rat aortic rings to
-adrenergic agents (16) and that modified hemoglobin ameliorates the hypotensive response to endotoxin in a porcine model of
endotoxemia (2). In a pilot experiment, we investigated the hemodynamic
effects of different doses of pyridoxalated hemoglobin polyoxyethylene
conjugate (PHP) in an ovine model of chronic hyperdynamic sepsis (3).
In that experiment we noted that doses of 50, 100, and 200 mg/kg of PHP
normalized mean arterial pressure and systemic vascular resistance in
septic sheep. The purpose of the present study was to investigate
whether the vasoconstiction due to PHP treatment during sepsis affects
regional blood flow to different organs and whether PHP treatment has
detrimental effects on different organ functions, with a special focus
on kidney, liver, and pancreas function.
| |
MATERIALS AND METHODS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
The procedures and experimental design described in this study were approved by the Animal Care and Use Committee of The University of Texas Medical Branch and were conducted in compliance with the guidelines for the care and use of animals established by the American Physiological Society as well as those of the National Institutes of Health.
Animal Preparation
Twenty female range ewes of the merino breed (body weight 32.2 ± 1.9 kg, range 25-42 kg) were surgically prepared for chronic study. After a 12-h fasting period, the sheep were anesthetized with 1.5-2.5 vol% halothane in oxygen, which was delivered through an animal face mask until anesthesia was adequate for endotracheal intubation (ID 10 mm; Mallinckrodt, Glen Falls, NY). After intubation, the animals were mechanically ventilated with tidal volumes of 15 ml/kg and frequencies of 12-15 breaths/min. Anesthesia was maintained with halothane in oxygen (1.0-1.5 vol%). Right femoral arterial and venous catheters were placed through a femoral incision in the abdominal aorta and in the abdominal vena cava. A Swan-Ganz thermal dilution catheter (model 93A-131-7F, American Edwards Laboratories, Irvine, CA) was positioned through the jugular vein into the pulmonary artery. After a left lateral thoracotomy at the fifth intercostal space, a Silastic catheter (0.062 in. ID, 0.125 in. OD; Dow Corning, Midland, MI) was inserted into the left atrium. After wound closure, the animals were weaned from ventilation and allowed to recover for at least 5 days. During this time, the sheep were monitored three times a day for appearance, adequacy of pain control, temperature, oral intake, and fecal and urinary output. If their body temperature exceeded 39.6°C, intravenous antibiotic treatment was begun and maintained until the body temperature was normal for >24 h. All antibiotics were stopped the day before the experiment. During the recovery and study periods, the animals were held in metabolic cages with free access to food and water. The day before the experiment the animals were anesthetized with ketamine, and a urethral Foley urinary retention catheter was placed. Thereafter, all sheep were connected to continuously flushing pressure transducers (Baxter, Irvine, CA), which were attached to hemodynamic monitors (model 78304A, Hewlett-Packard, Santa Clara, CA). The animals received a continuous infusion of Ringer lactate (2 ml · kg1 · h1),
and urine was collected until the experiment was started.
The animals were excluded from the study if, on the day of the experiment, blood temperature was above 39.6°C, white blood cell count was above 10,000 cells/µl, cardiac index was above 6.5 l/min, heart rate was above 100 beats/min, or mean pulmonary arterial pressure was above 25 mmHg.
Experimental Protocol
The experiments were performed in awake sheep. After baseline measurements, a continuous infusion of live Pseudomonas aeruginosa bacteria at a dose of 1 ×105 colony-forming units (CFU) · min1 · kg1
was started and maintained throughout the 48-h experimental period. The
infusion rate of Ringer lactate was adjusted to keep left atrial
pressure at baseline levels ± 3 mmHg. After 24 h of sepsis, all
animals (n = 14) were randomly and
equally assigned to one of the following two groups.
Treatment group (n = 7). After 24 h of P. aeruginosa infusion, an infusion of 100 mg/kg PHP (Apex Bioscience, Research Triangle Park, NC) was given to the animals in this group. PHP was dissolved in Plasmalyte infusion solution (Baxter Healthcare, Deerfield, IL) to a total infusion volume of 200 ml. The hemoglobin solution was given intravenously through a 0.2-µm infusion filter (Gelman Sciences, Ann Arbor, MI) over a 30-min period.
Control group (n = 7). The animals in this group received an infusion of 200 ml of Plasmalyte infusion solution over a 30-min period.
All animals were observed during the next 24 h. Cardiac outputs were determined by using the thermal dilution technique (cardiac output computer 9520, American Edwards Laboratory, Santa Ana, CA). Ten milliliters of cold (1-4°C) dextrose were used as the indicator. The average of three injections was computed for cardiac output. Hemodynamic and oxygenation variables were measured and calculated according to standard formulas. At every time point, arterial and mixed venous blood gases, oxyhemoglobin and methemoglobin concentration were determined (1302 pH/ blood-gas analyzer and CO-oximeter 282, Instrumentation Laboratory, Lexington, MA). The blood-gas results were not corrected for the body temperature of the sheep.Bacterial Preparation and Quantitative Cultures
In this study, live P. aeruginosa was used for induction of sepsis (strain ISR 12-4-4, isolated from a thermally injured patient at Brooke Army Hospital, San Antonio, TX). These bacteria were prepared from a frozen stock culture. After being thawed, 0.8 ml of the stock culture solution was inoculated into a trypticase soy broth (Difco Laboratories, Detroit, MI) and incubated for 24 h at 37°C. After being harvested by centrifugation, the bacterial concentration was counted in a Petroff-Hausser count chamber (Hausser Scientific, Blue Bell, PA). The bacteria were then resuspended in 250 ml of 0.9% saline to a final concentration so that with an infusion rate of 2 ml · kg1 · h1
the sheep received 1 × 105
CFU · kg1 · min1.
Samples of the suspension were taken from every infusion bag, and
pulmonary and aortic blood samples were taken every 4 h for the
duration of the experiment. From every sample, 200 µl were transferred into agar plates. These plates were incubated for 24 h at
37°C. After incubation, bacterial colony-forming units were counted
with a Darkfield Quebec colony counter (model 3330, American Optical,
Buffalo, NY), and the number of colony-forming units per milliliter was
calculated for arterial (CFUa)
and mixed venous blood
(CFU
).
Because the lung is the only important organ that clears bacteria
between the mixed venous and the arterial blood, the bacterial
clearance by the lungs can be calculated. The bacterial clearance of
the lungs can give a rough estimation of the phagocytotic activity of
the reticuloendothelial system of the lung. By using the paired amounts
of CFUa and
CFU,
the pulmonary bacterial clearance was calculated by using the following formula: pulmonary bacterial clearance = (CFU 
CFUa)/CFU · 100.
Hematology and Blood and Urine Chemistry
Hematology, blood, and urine chemistry parameters were analyzed to investigate the function of different organs after PHP treatment. Disturbances in renal function were assessed by measurement of plasma nitrogen, renal clearance of creatinine, and urine output. As parameters of liver damage and cholestasis
-glutamyl transferase (-GT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase were analyzed. Pancreatic damage was
estimated by using plasma lipase levels. To evaluate the effects of
hemoglobin infusion on plasma iron levels, plasma iron and iron-binding
capacity were determined. Plasma conjugated dienes were measured as an
estimation of tissue damage due to oxygen free radicals (32).
Hematologic variables (red and white blood cell counts) were determined
immediately after blood withdrawal by using a Coulter counter (S880,
Coulter Electronics, Hialeah, FL). The white blood cells were
differentiated into neutrophils and lymphocytes by using blood smears.
For the determination of the other serum and plasma parameters, blood
samples were immediately centrifuged and frozen at 80°C.
After these samples were thawed, plasma osmolality, plasma colloid
oncotic pressure, serum electrolytes, -GT, AST, ALT, alkaline
phosphatase, lipase, creatinine, urea nitrogen, bilirubin, iron,
iron-binding capacity, lactate, and urine creatinine were determined by
using standard laboratory tests used in a large university clinical
laboratory. The serum amylase levels of sheep in a pretest were below
the range of the used clinical assay (<20 U/l). Therefore, no serum
amylase levels of the investigated sheep are presented in the study.
All other parameters showed reliable results. In vitro interference of
low concentrations of PHP with laboratory tests was investigated by the
manufacturer of PHP (Apex Bioscience; unpublished observations), and
only interference with the lactate dehydrogenase determination and the
bilirubin assays was noticed. Plasma conjugated dienes were measured in
thawed EDTA-plasma samples, by using the method described by Ward et al. (32). Creatinine clearance was calculated from the serum and urine
creatinine levels and the urine volume excreted during a 12-h period
before the experiment started and during the experimental period every
12 h.
Regional Blood Flow Measurements
Regional blood flow in various organs was determined by using the fluorescent-microsphere technique. At the 0-, 24-, 25-, 28-, and 48-h time points of the experiment ~5 million fluorescent microspheres (15 ± 0.1 µm, E-Z TRAC, Interactive Medical Technology, Los Angeles, CA) were injected into the left atrium. Simultaneously, reference blood was continuously withdrawn from the abdominal aorta, with a withdrawal rate of 10 ml/min. The withdrawal was started 30 s before the microsphere injection and stopped 120 s after the injection. The different colors of the microspheres were administered in random order.Euthanasia and Necropsy
After 48 h, all animals were anesthetized with 500 mg ketamine and euthanized with the venous injection of 50 ml of saturated KCl solution. At necropsy the left and right lungs were removed for the determination of lung wet-to-dry weight ratios (24). Tissue samples were obtained from the myocardium of the right and left ventricle, liver, spleen, pancreatic head, distal ileum, colon, left and right kidney, and skeletal nonrespiratory muscle. These tissue samples were used together with the reference blood samples for the determination of regional blood flow to these tissues. This analysis was performed at Interactive Medical Technology. Each sample was weighed and then digested in NaOH. After digestion the number of different fluorescent microspheres was determined by flow cytometry. Regional blood flow in milliliters per gram per minute was calculated by using the following formula: blood flow = (number of counted spheres/tissue weight in grams) · (1/number of spheres in the reference blood in milliliters per minute).Statistical Analysis
All data are given as means ± SE. For statistical analysis of hemodynamic variables, the 0-, 24-, 25-, 28-, and 48-h time points were used. Statistical analysis was performed by using analysis of variance. If significant differences were detected, Fisher's least significant differences procedure was used, followed by a Bonferroni correction for the number of comparisons. Comparisons between the wet-to-dry weight ratios of the treatment and the control group were made by using Student's t-test.| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
During the first 24 h of sepsis, six sheep died because of pulmonary hypertension or hypodynamic septic shock. All animals that survived the first 24 h of bacteremia (n = 14) were included in the analysis of this study. These animals had a hyperdynamic circulation with an increased cardiac index, an increased heart rate, a decreased systemic vascular resistance index, and a decreased mean arterial pressure (Figs. 1 and 2).
|
|
After 5 min, when one-sixth of the PHP infusion was given, a
significant increase in mean arterial pressure was observed. Mean
arterial pressure increased from 86 ± 2.8 mmHg at 24 h to 96.7 ± 2.7 mmHg at 24 h and 5 min (P < 0.05 vs. 24 h), to 97.8 ± 2.9 at 24 h and 15 min
(P < 0.05 vs. 24 h), and to 101.8 ± 3.5 mmHg at 24 h and 30 min
(P < 0.05 vs. 24 h). Mean arterial
pressure stayed above the 24-h value during the next 4-6 h. This
increase in mean arterial pressure during the first hour after PHP
infusion was caused by an increase in the systemic vascular resistance index (Fig. 1), which rose in the PHP group from 769 ± 42.1 to 1,087 ± 56.8 dyn · s · m2 · cm5
(P < 0.05). At 28 h,
the increase in mean arterial pressure after PHP infusion, based on a
nonsignificant increase in systemic vascular resistance index and a
cardiac output, tended to be increased above 25-h values.
After PHP infusion, systemic vascular resistance index did not exceed
baseline values. In most of the PHP-treated animals, a decreased
cardiac index and heart rate were observed (Fig. 2). These changes did
not reach statistical significance. In both groups, pulmonary arterial
pressure increased 6-10 mmHg during the first hour after bacterial
infusion was started and remained elevated for the rest of the
experiment. PHP infusion caused a further increase in pulmonary
arterial pressure from 30 ± 1.9 mmHg at 24 h to 33 ± 1.7 mmHg
at 25 h. At 25 and 28 h, the pulmonary arterial pressure was
significant higher in the PHP group than in the control group
(P < 0.05) (Fig.
3). This increase in pulmonary arterial
pressure had no effect on arterial oxygenation. The arterial
PO2 was 80 ± 5.1 Torr at 24 h and
81 ± 3.5 Torr at 24.5 h. Mixed venous
PO2 was also unchanged, with values
of 39 ± 3 Torr at 24 h and 39 ± 3 Torr at 24.5 h.
The lungs of the treatment animals at the end of the experiment did not
show macroscopic signs of pulmonary edema at necropsy. Wet-to-dry
weight ratios as a measure of pulmonary edema were 5.0 ± 0.2 in the
left lungs and 5.0 ± 0.1 in the right lungs of the treatment group
animals and were 5.2 ± 0.1 in the left lungs and 5.2 ± 0.1 in
the right lungs of the control group animals. No significant
differences between the treatment and the control group were detected.
|
Eight hours after PHP was given, all macrohemodynamic parameters in the treated animals had returned to levels that existed before PHP administration and were similar to those of the control group throughout the remaining experimental period.
The results of the microsphere studies are shown in Table
1. After 24, 25, and 28 h, none of the
examined organ tissues in the control or in the treatment group showed
a significant increase in blood flow. A marked but not significant drop
in regional blood flow to the pancreatic head occurred in the treatment
group (2.9 ± 0.3 vs. 1.9 ml · g1 · min1).
This drop was not associated with an increase in serum lipase levels
(150 U/l at 24 h vs. 122 ± 29 at 25 h; Table
2). After 48 h of sepsis, blood flow was
significantly elevated in comparison to baseline in the liver of the
treatment and the control group, the spleen of both groups, the left
and the right ventricle of both groups, and the left and right kidney
cortex of the treatment group. No difference in microvascular blood
flow was seen between the right and left kidney cortex in either group,
thus confirming that adequate mixing of the microspheres had taken
place at the time of microsphere injection.
|
|
Blood chemistry and hematology parameters are presented in Tables 2 and 3. Only bilirubin and iron levels differed significantly between control and treatment group. Both increased after PHP infusion in the treatment group. Iron levels decreased during the first 24 h of sepsis in the control group from 102 ± 14 µg/dl at 0 h to 33 ± 13 µg/dl at 24 h (P < 0.05) and in the PHP group from 91 ± 9.6 µg/dl at 0 h to 25 ± 2.7 µg/dl at 24 h (P < 0.05). After PHP infusion, iron levels increased back to 60 ± 4.9 µg/dl at 25 h (P < 0.05) but never exceeded iron-binding capacity. During the time course of sepsis, no significant differences in urine output and creatinine clearance were found between the two animal groups (Table 4).
|
|
Pulmonary bacterial clearance, a measurement that provides information about the effectiveness of the reticuloendothelial system, decreased in the first 24 h of sepsis to 0.56 ± 0.09 in the control group (P < 0.05 vs. 0 h) and 0.52 ± 0.08 in the PHP group (P < 0.05 vs. 0 h) (Fig. 4). Pulmonary bacterial clearance was not different between both groups throughout the experiment.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
In the present study we investigated the effects of nitric oxide scavenging with PHP in ovine sepsis, with a special focus on microvascular blood flow, organ function, and organ damage. We conducted our experiments in an established sepsis model with chronically instrumented awake sheep. In this model, sepsis was induced and maintained by a continuous infusion of live P. aeruginosa bacteria. Advantages of this animal model of sepsis are as follows: 1) it mimics the hemodynamic situation of septic patients (5); 2) therapeutic interventions were made after 24 h of sepsis, a time when the inducible form of the nitric oxide synthase is stimulated and produces large amounts of nitric oxide (28); and 3) all observations were made in awake sheep; therefore, no side effects of anesthetics or interactions with anesthetics confounded the results. The animals included in this study developed a hyperdynamic sepsis, characterized by decreased mean arterial pressure, decreased systemic vascular resistance, and increased heart rate and cardiac output. The infusion of 100 mg/kg PHP in the hyperdynamic sheep normalized mean arterial pressure and systemic vascular resistance for 4-6 h. The vasoconstrictive effects of PHP were not associated with any significant decrease in microvascular blood flow to different organs. Furthermore, after PHP infusion, no changes in blood chemistry or hematology parameters suggestive of organ dysfunction or damage were observed.
In a previous study we showed that different doses of PHP restored mean arterial pressure in hyperdynamic ovine sepsis by increasing systemic vascular resistance (3). Of three effective doses of PHP, we used the intermediate dose for the present study to ensure adequate effects without toxicity from overdosage. Infusion of 100 mg/kg PHP caused a rapid increase in mean arterial pressure and systemic vascular resistance. The speed of hemodynamic changes during PHP infusion was impressive. Five minutes after the PHP infusion was started, after only 16.6 mg/kg PHP in 33.3 ml fluid had been given, mean arterial pressure increased significantly. The hemodynamic changes after such small amounts of hemoglobin cannot be explained only by volume effects of the hemoglobin infusion. Also, changes in plasma osmolality or plasma oncotic pressure with consequent increases in intravascular volume are unlikely (Table 2).
The observed increases in mean arterial pressure and systemic vascular resistance are most likely due to nitric oxide-scavenging effects of hemoglobin and to elevation of the vasoconstrictive peptide endothelin-1 after hemoglobin infusion (1, 12). We observed a short insignificant decrease in cardiac index and heart rate during and after PHP infusion. Significant decreases in cardiac output were seen after inhibition of nitric oxide synthase in our ovine sepsis model (4). Similar to our observations and contrary to the observations with nitric oxide synthase inhibition, other investigators also saw no change in cardiac output after infusion of modified hemoglobin (18, 27).
One major concern about the use of modified hemoglobins, especially the use of hemoglobins as a treatment of hyperdynamic sepsis, is that hemoglobin can cause a severe vasoconstriction and therefore reduce nutritive blood flow to different organs. This reduction in nutritive blood flow could impair organ function in, for example, the pancreas, kidney, and liver. Therefore, we investigated regional blood flow after PHP infusion in hyperdynamic sepsis. In addition, we investigated a couple of blood chemistry parameters that could indicate impaired organ function after vasoconstriction due to PHP. For the measurement of regional blood flow the fluorescent-microsphere technique was used. The advantage of the microsphere technique is that blood flow in a large number of different tissues can be measured. By this technique, capillary blood flow is measured, in contrast to the flow-probe technique in which only blood flow in larger vessels can be determined. A disadvantage of the use of the microsphere technique in awake large animals is that movements of the animals during the microsphere injection can disturb measurements and increase the variation of blood flow results. In our study, baseline blood flows to different organs (Table 1) were in the same range as those described by others (4, 5, 31). During the first 24 h of sepsis, cardiac output increased significantly, but none of the analyzed organs showed a significant increase in organ blood flow, with the exception of the liver in the treatment animals. Booke et al. (4) made similar observations by using the same animal model and colored microsspheres. The increase in cardiac output without a significant increase in capillary blood flow could be explained by an increased amount of precapillary shunting during sepsis, which cannot be determined with the microsphere technique.
After 48 h of sepsis, organ blood flows in the control group showed redistribution of blood flow away from pancreas, ileum, and colon toward the heart. The reduction in pancreatic blood flow (25) as well as the increase in myocardial blood flow (25) during sepsis have been described in previous studies. PHP infusion during sepsis did not decrease organ blood flow significantly. Also, no change in plasma lactate levels, a rough indicator of the adequacy of nutritive blood flow to the organs, was seen after PHP infusion.
The only significant change in regional blood flow after PHP infusion was an increase in blood flow to the spleen at 48 h and an increase in ileal blood flow at 28 and 48 h. However, studies investigating intestinal blood flow after hemoglobin infusion report varying results. In contrast to our findings, Aranow et al. (2) noted a reduced ileal mucosal perfusion after cross-linked hemoglobin in porcine endotoxemia but no change in mucosal pH. Compared with a control group there was no decrease in mesenteric artery blood flow after hemoglobin infusion. Gulati et al. (12) even observed a significant increase in regional blood flow to the gastrointestinal tract after infusion of diaspirin-cross-linked hemoglobin. Ulatowski et al. (31), on the other hand, saw a significant decrease in intestinal blood flow after infusion of bovine cross-linked hemoglobin in cats compared with a control group. They, however, used a much higher dose of hemoglobin and conducted their experiments in an animal hemorrhage model, which may explain their contrasting results. No significant change in myocardial perfusion in the right and left ventricles was seen after PHP infusion in our study, although cardiac output tended to decrease below the 24-h values for the first 20 h after PHP was given. Whereas in ex vivo studies coronary vasoconstrictive properties of modified hemoglobin were seen (20), in recent in vivo studies no coronary vasoconstriction or reduction in myocardial blood flow was detected after hemoglobin application (12, 31). Gulati et al. (12) showed that the increase in myocardial blood flow after application of cross-linked hemoglobin can be attenuated by an endothelin-A-receptor antagonist, indicating the important role of endothelin release and endothelin effects after hemoglobin infusion. In the same study, a decreased renal vascular resistance and an increased renal blood flow after cross-linked hemoglobin administration were seen. This vasodilatatory effect of hemoglobin also disappeared after infusion of an endothelin-A-receptor antagonist.
In our study we also observed no significant effects of PHP on regional blood flow to the kidney cortex, and neither plasma creatinine, plasma urea nitrogen, creatinine clearance, nor urine output as markers of kidney function were affected by PHP infusion during hyperdynamic sepsis. Renal impairment after hemoglobin infusion is more frequent if the infused hemoglobin is not purified and contains some stroma or if the hemoglobin is not modified (13, 17). Nitric oxide plays a crucial role in the modulation of renal blood flow and glomerular filtration (7). Therefore, nitric oxide-scavenging properties of hemoglobin could aggravate kidney dysfunction. Additionally, direct toxic effects of hemoglobin on the tubular kidney function were described. We did not observe any impairment of kidney function after PHP infusion during ovine sepsis. First, tubular damage due to the highly purified and chemically modified PHP is unlikely. Second, in contrast to normal kidneys, in septic patients, a loss of renal vascular autoregulation occurs, and nitric oxide synthase inhibition or nitric oxide scavenging can improve glomerular filtration pressure by a more pronounced effect on the efferent than on the afferent renal arteriole (14, 15).
Nitric oxide seems to play an important role in the organ function of
the normal liver and in preventing microcirculatory dysfunction in the
liver during sepsis (21, 23). However, we did not observe any
detrimental effects of the nitric oxide scavenger PHP on the liver
during ovine sepsis. Serum levels of AST, ALT, and
-GT as markers of liver damage were analyzed. The serum levels of AST and -GT were unchanged, and ALT even decreased after PHP infusion. Our observations are in accordance with those of
Eldridge et al. (9), who saw no change in liver function tests and
liver histology after PHP infusion in comparison to a control group,
and with those of Gulati et al. (12), who described an unchanged
regional liver blood flow after infusion of cross-linked hemoglobin.
However, it must be emphasized that the measurement of liver blood flow
with microspheres, which were used in the study of Gulati et al. and in
our study, is an insufficient method to estimate nutritive blood flow
to the liver, because only the arterial part of the liver circulation
is measured and the regional blood flow that was contributed by the
portal circulation is not measured. Therefore, we measured only
microvascular blood flow to the liver to compare our results with
results from other groups.
Increased iron levels in the blood could be harmful during infection, because the virulence of P. aeruginosa and other bacteria can be increased by increased iron levels (6). We saw a significant increase in iron levels after infusion of PHP, but these iron levels were still below baseline and never exceeded the iron-binding capacity, so that no free iron could increase bacterial growth. Unfortunately, no organ bacterial counts were made during our study; therefore, no final comments can be made on the effects of PHP on bacterial growth in different organs after PHP infusion in sepsis. Another concern about elevated iron levels and the application of free hemoglobin is related to the oxidative damage to tissues by generating oxygen free radicals in the presence of iron and hemoglobin. In sepsis, when levels of antioxidants are reduced, that could be a detrimental side effect of PHP therapy. In our experiments, we saw no change in plasma conjugated diene levels, which indicates that no relevant lipid oxidation occurred, although free hemoglobin was given in sepsis.
After PHP infusion we observed a significant increase in total bilirubin levels in the treatment group. Free hemoglobin can induce the enzyme heme oxygenase and therefore increase the production of bilirubin (22). Also, an interference of the bilirubin test used with PHP could explain the observed elevation of bilirubin (9).
In contrast to other authors (18), we saw no change in methemoglobin concentrations after infusion of PHP.
This lack of change is most likely related to the very low dose of modified hemoglobin we gave and not related to specific advantages of PHP in comparison with other hemoglobins. Modified hemoglobin is in part cleared from the circulation by the reticuloendothelial system; therefore, the question arises as to whether the use of modified hemoglobin as a treatment of hyperdynamic septic shock impairs the host defense to infections.
We investigated the activity of the reticuloendothelial system in our ovine sepsis model by calculating the pulmonary bacterial clearance (Fig. 4) of live P. aeruginosa bacteria. No differences were seen between groups, in pulmonary bacterial clearance, or in white blood cells counts and white blood cell differentiation (Table 3). These data are in accord with observations from Crowley et al. (8), who saw no changes in white blood cell count and in Escherichia coli clearance after infusion of stroma-free cross-linked hemoglobin in dogs. Our results are in contrast to those of Griffiths et al. (11), who found that PHP infusion cause death in experimental E. coli peritonitis in mice. They conclude that the promotion of fulminant sepsis is a potential danger when using modified hemoglobins. Although our ovine sepsis model is not suitable for the investigation of sepsis lethality, a promotion of sepsis owing to PHP infusion would have been noted during the 24-h observation period after PHP infusion. The difference between our results and those of Griffiths et al. may be related to the different animal models, the different times when PHP was given or the different bacteria that were used. Further investigation regarding bacterial growth, bacterial clearance, and outcome should be performed to evaluate the use of PHP as a possible treatment for septic shock.
In conclusion, infusion of a low dose of PHP normalized mean arterial pressure and systemic vascular resistance in an ovine model of hyperdynamic sepsis during the first 4 h after infusion. The application of modified hemoglobin was not associated with significant decreases in organ blood flows. Investigation of clinically relevant blood chemistry parameters gave no indication of toxicity. Also, no detrimental effects on kidney function were observed. No interactions between PHP infusion and the host response to severe infection were seen. The only observed side effect of this new therapeutic approach for hyperdynamic septic shock was an increase in pulmonary arterial pressure and pulmonary vascular resistance. The change in pulmonary vascular tone was not associated with changes in oxygenation, and no pulmonary edema was found in the lungs of the treated animals. Our study showed that a short-duration, low-dose infusion of modified hemoglobin was a safe and effective treatment of hypotension in an ovine model of hyperdynamic sepsis.
| |
ACKNOWLEDGEMENTS |
|---|
The study was supported in part by Apex Bioscience (Research Triangle Park, NC).
| |
FOOTNOTES |
|---|
Address for reprint requests: D. L. Traber, Dept. of Anesthesiology, The Univ. of Texas Medical Branch, Galveston, TX 77555-0833 (E-mail: traber{at}utmb.edu).
Received 23 September 1996; accepted in final form 30 January 1998.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
1.
Alayash, A. I.,
and
R. E. Cashon.
Reactions of nitric oxide and hydrogen peroxide with hemoglobin-based blood substitutes.
Ann. NY Acad. Sci.
738:
378-381,
1994[Medline].
2.
Aranow, J. S.,
H. Wang,
J. Zhuang,
and
M. P. Fink.
Effect of human hemoglobin on systemic and regional hemodynamics in a porcine model of endotoxemic shock.
Crit. Care Med.
24:
807-814,
1996[Medline].
3.
Bone, H. G., M. Booke, P. J. Schenarts, M. O'Connor, L. D. Traber, and D. L. Traber.
Comparison of different doses of pyridoxalated hemoglobin
polyoxyethylene conjugate in the treatment of hyperdynamic sepsis.
Crit. Care Med. 24, Suppl.: A27, 1996.
4.
Booke, M.,
F. Hinder,
R. McGuire,
L. D. Traber,
and
D. L. Traber.
Nitric oxide synthase inhibition versus norepinephrine in ovine sepsis: effects on regional blood flow.
Shock
5:
362-370,
1996[Medline].
5.
Booke, M.,
J. Meyer,
W. Lingnau,
F. Hinder,
L. D. Traber,
and
D. L. Traber.
Use of nitric oxide inhibitors in animal models of sepsis.
New Horizons
3:
123-138,
1995[Medline].
6.
Bullen, J. J.,
C. G. Ward,
and
S. N. Wallis.
Virulence and the role of iron in Pseudomonas aeruginosa infections.
Infect. Immun.
10:
443-450,
1974
7.
Chintala, M. S.,
P. J. Chiu,
S. Vemulapalli,
R. W. Watkins,
and
E. J. Sybertz.
Inhibition of endothelial derived relaxing factor (EDRF) aggravates ischemic acute renal failure in anesthetized rats.
Arch. Pharm.
348:
305-310,
1993.
8.
Crowley, J. P.,
J. Metzger,
A. Gray,
L. E. Pivacek,
G. Cassidy,
and
C. R. Valeri.
Infusion of stroma-free cross-linked hemoglobin during acute gram-negative bacteremia.
Circ. Shock
41:
144-149,
1993[Medline].
9.
Eldridge, J.,
R. Russell,
R. Christenson,
R. Sakamoto,
J. Williams,
M. Parr,
B. Trump,
P. Delaney,
and
C. Mackenzie.
Liver function and morphology after resuscitation from severe hemorrhagic shock with hemoglobin solutions or autologous blood.
Crit. Care Med.
24:
663-671,
1996[Medline].
10.
Gomez-Jimenez, J.,
A. Salgado,
M. Mourelle,
M. C. Martín,
R. M. Segura,
R. Peracaula,
and
S. Moncada.
L-Arginine: nitric oxide pathway in endotoxemia and human septic shock.
Crit. Care Med.
23:
253-258,
1995[Medline].
11.
Griffiths, E.,
A. Cortes,
N. Gilbert,
P. Stevenson,
S. Macdonald,
and
D. Pepper.
Haemoglobin-based blood substitutes and sepsis.
Lancet
345:
158-160,
1995[Medline].
12.
Gulati, A.,
A. C. Sharma,
and
G. Singh.
Role of endothelin in the cardiovascular effects of diaspirin crosslinked and stroma reduced hemoglobin.
Crit. Care Med.
24:
137-147,
1996[Medline].
13.
Hess, J. R.,
V. W. Macdonald,
and
W. W. Brinkley.
Systemic and pulmonary hypertension after resuscitation with cell-free hemoglobin.
J. Appl. Physiol.
74:
1769-1778,
1993
14.
Hinder, F.,
M. Booke,
L. D. Traber,
N. Matsumoto,
K. Nishida,
S. Rogers,
and
D. L. Traber.
Nitric oxide synthase inhibition during experimental sepsis improves renal excretory function in the presence of chronically increased atrial natriuretic peptide.
Crit. Care Med.
24:
131-136,
1996[Medline].
15.
Hoffend, J.,
A. Cavarape,
K. Endlich,
and
M. Steinhausen.
Influence of endothelium-derived relaxing factor on renal microvessels and pressure-dependent vasodilation.
Am. J. Physiol.
265 (Renal Fluid Electrolyte Physiol. 34):
F285-F292,
1993
16.
Kilbourn, R. G.,
G. Joly,
B. Cashon,
J. Deangelo,
and
J. Bonaventura.
Cell-free hemoglobin reverses the endotoxin-mediated hyporesponsivity of rat aortic rings to -adrenergic agents.
Biochem. Biophys. Res. Commun.
260:
10395-4608,
1994.
17.
Lee, R.,
N. Atsumi,
E. E. Jacobs, Jr.,
W. G. Austen,
and
G. J. Vlahakes.
Ultrapure, stroma-free, polymerized bovine hemoglobin solution: evaluation of renal toxicity.
J. Surg. Res.
47:
407-411,
1989[Medline].
18.
Lee, R.,
K. Neya,
T. A. Svizzero,
and
G. J. Vlahakes.
Limitations of the efficacy of hemoglobin-based oxygen-carrying solutions.
J. Appl. Physiol.
79:
236-242,
1995
19.
Lowry, S. F.
Sepsis and its complications: clinical definitions and therapeutic prospects.
Crit. Care Med.
22:
S1-S2,
1994[Medline].
20.
Macdonald, V. W.,
R. M. Winslow,
M. A. Marini,
and
M. T. Klinker.
Coronary vasoconstrictor activity of purified and modified human hemoglobin.
Biomaterials
18:
263-282,
1990.
21.
Nishida, J.,
R. S. McCuskey,
D. McDonnell,
and
E. S. Fox.
Protective role of NO in hepatic microcirculatory dysfunction during endotoxemia.
Am. J. Physiol.
267 (Gastrointest. Liver Physiol. 30):
G1135-G1141,
1994
22.
Otterbein, L.,
S. L. Sylvester,
and
A. M. K. Choi.
Hemoglobin provides protection against lethal endotoxemia in rats: the role of heme oxygenase-1.
Am. J. Respir. Cell Mol. Biol.
13:
595-601,
1995[Abstract].
23.
Pastor, C. M.,
and
T. R. Billiar.
Regulation and functions of nitric oxide in the liver in sepsis and inflammation.
New Horizons
3:
65-72,
1995[Medline].
24.
Pearce, M. L.,
J. Yamashita,
and
J. Beazell.
Measurement of pulmonary edema.
Circ. Res.
16:
482-488,
1965
25.
Raper, R. F.,
W. J. Sibbald,
J. Hobson,
and
F. S. Rutledge.
Effect of PGE1 on altered distribution of regional blood flows in hyperdynamic sepsis.
Chest
100:
1703-1711,
1991
26.
Schultz, S. C.,
B. Grady,
F. Cole,
I. Hamilton,
K. Burhop,
and
D. S. Malcolm.
A role for endothelin and nitric oxide in the pressor response to diaspirin cross-linked hemoglobin.
J. Lab. Clin. Med.
122:
301-308,
1993[Medline].
27.
Sharma, A. C.,
and
A. Gulati.
Yohimbine modulates diaspirin crosslinked hemoglobin-induced systemic hemodynamics and regional circulatory effects.
Crit. Care Med.
23:
874-884,
1995[Medline].
28.
Szabo, C.
Alterations in nitric oxide production in various forms of circulatory shock.
New Horizons
3:
2-32,
1995[Medline].
29.
Szabo, C.,
G. J. Southan,
and
C. Thiemermann.
Beneficial effects and improved survival in rodent models of septic shock with S-methylisothiourea sulfate, a potent and selective inhibitor of inducible nitric oxide synthase.
Proc. Natl. Acad. Sci. USA
91:
12472-12476,
1994
30.
Szabo, C.,
B. Zingarelli,
and
A. L. Salzman.
Role of poly-ADP ribosyltransferase activation in the vascular contractile and energetic failure elicited by exogenous and endogenous nitric oxide and peroxynitrite.
Circ. Res.
78:
1051-1063,
1996
31.
Ulatowski, J. A.,
T. Nishikawa,
B. Matheson-Urbaitis,
E. Bucci,
R. J. Traystman,
and
R. C. Koehler.
Regional blood flow alterations after bovine fumaryl 
-crosslinked hemoglobin transfusion and nitric oxide synthase inhibition.
Crit. Care Med.
24:
558-565,
1996[Medline].
32.
Ward, P. A.,
G. O. Till,
J. R. Hatherill,
T. M. Annesley,
and
R. G. Kunkel.
Systemic complement activation, lung injury, and products of lipid peroxidation.
J. Clin. Invest.
76:
517-527,
1985.
This article has been cited by other articles:
|
|
 |
|
  M. Lange, K. Broking, C. Hucklenbruch, C. Ertmer, H. Van Aken, M. Lucke, H.-G. Bone, and M. Westphal Hemodynamic effects of titrated norepinephrine in healthy versus endotoxemic sheep Innate Immunity, February 1, 2007; 13(1): 53 - 57. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Langenberg, L. Wan, S. M. Bagshaw, M. Egi, C. N. May, and R. Bellomo Urinary biochemistry in experimental septic acute renal failure Nephrol. Dial. Transplant., December 1, 2006; 21(12): 3389 - 3397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mullon, G. Giacoppe, C. Clagett, D. McCune, and T. Dillard Transfusions of Polymerized Bovine Hemoglobin in a Patient with Severe Autoimmune Hemolytic Anemia N. Engl. J. Med., June 1, 2000; 342(22): 1638 - 1643. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Caron, P. Menu, B. Faivre-Fiorina, P. Labrude, A. Alayash, and C. Vigneron Systemic and renal hemodynamics after moderate hemodilution with HbOCs in anesthetized rabbits Am J Physiol Heart Circ Physiol, June 1, 2000; 278(6): H1974 - H1983. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
0.05 vs. 0 h.
P 
P 
