INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

DESPITE ADVANCES in the pathophysiological mechanisms of sepsis and refinements in the management of septic patients, the exceedingly high mortality and morbidity rates due to sepsis, septic shock, and multiple organ failure in surgical intensive care units have not been significantly reduced during the past two decades (2, 3). Because mediators or factors responsible for the transition from the early, hyperdynamic to the late, hypodynamic phase of sepsis have not been fully identified and consequently not prevented, progressive cell and organ dysfunction and failure occur. Studies have shown that the cardiovascular and hemodynamic response to sepsis includes an early, hyperdynamic phase characterized by increased cardiac output, increased tissue perfusion, and decreased peripheral resistance (38). This is followed by a late, hypodynamic phase characterized by reduced microvascular blood flow in various tissues (38, 40). Although the mechanisms responsible for the transition from hyperdynamic circulation during early sepsis to hypodynamic circulation during late sepsis are not completely understood, it has been postulated that pharmacological modulation of the cardiovascular system during the progression of sepsis may delay or even prevent the occurrence of the hypodynamic and hypocardiovascular response (30, 35).

A number of studies have indicated that administration of the phosphodiesterase inhibitor pentoxifylline (PTX, a xanthine derivative) during sepsis or bacteremia produced beneficial effects on cell and organ function (4, 14, 17, 25, 28) without producing significant adverse hemodynamic side effects (1). Studies by Zhang et al. (43) have indicated that administration of PTX improved tissue oxygen extraction capacity during endotoxic shock. In addition, administration of PTX following murine endotoxic shock increased the survival rate (24). Our previous studies have shown that administration of PTX early after the onset of sepsis maintained hepatocellular function and improved cardiac performance during the early stage of sepsis (34). In addition, PTX protected vascular endothelial cells during both hyperdynamic and hypodynamic sepsis (39). Moreover, PTX has been shown to reduce the production of tumor necrosis factor- (TNF-), interleukin-6 (IL-6), and interleukin-1 (IL-1) during intra-abdominal sepsis (17). It remains unknown, however, whether this agent also has any salutary effects on cardiovascular responses and tissue oxygen utilization during the late stage of polymicrobial sepsis. Because polymicrobial sepsis is characterized by an early, hyperdynamic phase followed by a late, hypodynamic phase and because lethality does not usually occur during the hyperdynamic phase in the model of cecal ligation and puncture (CLP) (38), we hypothesized that prevention of the occurrence of the late, hypodynamic phase of sepsis reduces the mortality rate. The primary aim of this study therefore was to determine whether administration of PTX early after the onset of sepsis prevents the occurrence of the hypodynamic response during the progression of polymicrobial sepsis. The effects of PTX on systemic and regional perfusion, oxygen delivery (DO₂), oxygen consumption (O₂), morphological changes, and survival rate were therefore investigated in a rat model of polymicrobial sepsis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal model of sepsis. Polymicrobial sepsis was induced by CLP in male Sprague-Dawley rats (275-345 g) as described previously (5). In brief, animals were fasted overnight before CLP was performed but were allowed water ad libitum. At the time of the experiment, the rats were anesthetized with methoxyflurane inhalation, and a 2-cm ventral midline incision was performed. The cecum was exposed, ligated just distally to the ileocecal valve to avoid intestinal obstruction, punctured twice with an 18-gauge needle, squeezed gently to force out a small amount of feces, and then returned to the abdominal cavity. The abdominal incision was closed in layers, and the animals received 3 ml/100 g body wt normal saline solution subcutaneously immediately after CLP was performed (i.e., fluid resuscitation). Rats in the sham-operated group (control) underwent the same surgical procedure except that the cecum was neither ligated nor punctured. Various parameters were then measured at 20 h after the onset of sepsis or sham operation. Twenty hours after CLP represents the late, hypodynamic phase and 2-10 h after CLP represents the early, hyperdynamic phase of polymicrobial sepsis (38). The animals were divided into three groups: 1) sham operation, 2) CLP with vehicle (normal saline solution) administration, and 3) CLP with PTX treatment (refer to the next section for details). There were six animals in each group. It should be noted that these groups of animals were used for determining systemic and regional hemodynamics, DO₂, O₂, and plasma levels of alanine aminotransferase (ALT) and lactate. The mortality study was performed in a different set of animals for obvious reasons. Similarly, the morphological study was conducted in additional animals to avoid any possible effects of blood sampling on tissue structure. The experimental procedures and care of the animals were performed in adherence with the Animal Welfare Act and National Institutes of Health guidelines for the use of experimental animals. This project was approved by the Institutional Animal Care and Use Committee of Rhode Island Hospital.

Administration of PTX. At 1 h after the onset of sepsis, the left femoral vein was cannulated with a PE-50 tubing under methoxyflurane anesthesia. PTX (50 mg/kg body wt in normal saline solution at 0.67 ml/100 g body wt; Sigma, St. Louis, MO) was infused through the venous catheter over a period of 30 min at a constant infusion rate. The femoral artery was also cannulated with PE-50 tubing and connected to a blood pressure analyzer (Micro-Med, Louisville, KY) to monitor blood pressure changes during PTX infusion. Our preliminary results indicate that an infusion rate of PTX at 1.67 mg · kg¹ · min¹ does not produce significant hypotension. After the completion of PTX infusion, the femoral vein and femoral artery were decannulated. Vehicle-treated septic animals and sham-operated animals received vehicle (normal saline solution) intravenously over 30 min. In additional sham-operated animals (n = 4), PTX (50 mg/kg body wt) was infused at 1 h after surgery over a period of 30 min.

Determination of cardiac output and regional blood flow. At 20 h after CLP or sham operation, the animals were anesthetized again with methoxyflurane. Both the right femoral artery and vein were cannulated with PE-50 tubing. An additional PE-50 catheter was inserted into the left ventricle via the right carotid artery. The position of the PE-50 catheter tip in the left ventricle was confirmed by the left ventricular pulse pressure. After blood sampling (~1.0 ml) from the femoral arterial catheter for the measurement of ALT, lactate, and systemic hematocrit, a 3.5-F catheter (Sherwood, St. Louis, MO) was placed in the hepatic vein through the jugular vein to collect a hepatic venous sample for determining its oxygen content as described under Determination of organ DO₂ and O₂. Strontium-85-labeled microspheres (DuPont-NEN, Boston, MA) were suspended in 10% dextran containing 0.05% Tween 80 surfactant to prevent aggregation. The microspheres were dispersed with a vortex shaker for 3 min and a 0.20- to 0.25-ml suspension of microspheres with an activity of ~4 µCi/rat was infused into the left ventricle for 20 s at a constant rate. An estimated 150,000 microspheres were injected into each rat. The reference blood sample was withdrawn from the femoral arterial catheter beginning 20 s before the microsphere infusion and continued for 80 s at a rate of 0.7 ml/min. Normal saline (0.8 ml) was infused via the left ventricular catheter immediately after microsphere infusion for 40 s. At the end of each experiment, the rat was euthanized by an overdose of methoxyflurane. Various organs (i.e., the liver, spleen, pancreas, stomach, small intestine, and cecum only in sham-operated animals, large intestine, mesentery, and kidneys) were harvested, washed with normal saline, and gently blotted on filter paper. The organs were weighed and placed in one or more counting tubes, and the radioactivity was counted on a Wallac automatic gamma-counter (1470 Wizard; Wallac, Gaithersburg, MD). The reference blood sample was transferred into a counting tube and counted. Cardiac output, stroke volume, and total peripheral resistance (TPR), blood flow in the liver (including hepatic arterial blood flow and portal venous blood flow), small intestine, and kidneys were calculated according to the method described in our previous publication (31, 32).

Determination of organ DO₂ and O₂. At 20 h after CLP or sham operation, blood samples (~0.3 ml each) were drawn from the femoral artery (representing the systemic arterial blood), superior vena cava close to the right atrium (representing the systemic venous blood), renal vein, hepatic vein, and portal vein, respectively. Blood oxygen content was determined using an OSM3 hemoximeter (Radiometer, Copenhagen, Denmark). It should be noted that blood samples for oxygen content measurement were collected following microsphere administration. Systemic or regional DO₂ was then calculated by multiplying arterial oxygen content with cardiac output or organ blood flow, respectively. Systemic or regional O₂ was determined by multiplying the difference between arterial oxygen content and venous oxygen content with cardiac output or organ blood flow, respectively. Small intestinal O₂ was calculated by the difference in oxygen content between the systemic arterial blood and portal venous blood, and O₂ in the kidneys was calculated by the difference between the systemic arterial blood and renal venous blood. Hepatic O₂ was determined by calculating the difference in oxygen content between the hepatic inflow blood (i.e., hepatic arterial blood and portal venous blood) and hepatic outflow blood (hepatic venous blood). Moreover, oxygen extraction ratio (ERO₂, %) was determined by O₂/DO₂ × 100.

Measurement of plasma ALT and lactate. Whole blood was collected in EDTA-coated test tubes, and the plasma was separated by centrifugation immediately after blood sampling. The plasma samples were stored at 70°C until assayed for ALT and lactate using Sigma assay kits according to the manufacturer's instruction.

Histological examination. The morphological alterations in the liver, small intestine, and kidneys were examined at 20 h after the onset of sepsis by light microscopy. Tissue samples (n = 2/group) obtained from both sham-operated animals and septic animals treated with PTX or vehicle were submerged in 10% Formalin in neutral buffered solution (Sigma) for immediate fixation and later embedded in paraffin. The tissue blocks were then sectioned at a thickness of 5 µm, floated on warm water, and transferred to glass slides, where they were stained with hematoxylin and eosin, dehydrated, and coverslipped. Histological examinations were performed using a light microscope and documented by photographs.

Mortality study. In additional groups of rats, CLP was performed in two groups of 10 animals each. At 1 h after CLP, either PTX or vehicle (normal saline solution) was infused intravenously as previously mentioned. At 20 h after the onset of sepsis, the necrotic cecum was excised and the abdominal cavity was washed twice by using 40 ml of warm, sterile normal saline solution. The abdominal incision was then closed in layers. The procedure of cecal excision in CLP animals was performed to mimic the clinical situation in which septic focus is routinely removed whenever possible. The experimental animals were then allowed food ad libitum and monitored for 10 days to record the time of death in the nonsurvivors.

Statistical analysis. Results are presented as means ± SE. One-way analysis of variance (ANOVA) and Tukey's test were employed for the comparison among different groups of animals. In addition, Fisher's exact test was used to compare the difference in mortality rates. Differences were considered significant at P  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of PTX on systemic hemodynamic parameters DO₂ and O₂. As shown in Table 1, cardiac output decreased by 41.6% (P < 0.05) at 20 h after CLP; however, treatment with PTX prevented the reduction of cardiac output. Similarly, systemic DO₂ decreased by 35.5% (P < 0.05) at 20 h after the onset of sepsis, and administration of PTX maintained DO₂ during late sepsis. Although systemic O₂ was similar in both sham and septic animals at 20 h after the surgical procedure, PTX treatment after CLP significantly increased systemic O₂ (Table 1). In addition, systemic ERO₂ increased by 34.2-53.3% (P < 0.05) at 20 h after CLP, irrespective of PTX administration (Table 1). It appears that the significant increase in systemic ERO₂ at 20 h after the onset of sepsis plays a major role in the maintenance of systemic O₂ under such conditions. Mean arterial pressure was maintained above 120 mmHg at 20 h after CLP, irrespective of PTX administration. Although stroke volume decreased by 35.4% (P < 0.05) at 20 h after CLP, administration of PTX maintained stroke volume at the sham level in late sepsis (Table 1). In contrast, TPR increased by 72.2% (P < 0.05) at 20 h after CLP, and PTX treatment early after the onset of sepsis prevented such an increase in TPR (Table 1). The septic animals were hemoconcentrated at 20 h after CLP as evidenced by significantly increased systemic hematocrit (from an average of 44.8 to 53.8%). Administration of PTX, however, decreased systemic hematocrit to an average of 50.5% (Table 1).

View this table: [in this window] [in a new window]   Table 1.   Effects of PTX infusion on CO, systemic DO2, O2, ERO2, and other hemodynamic parameters at 20 h after CLP

Effects of PTX on hepatic blood flow, DO₂, and O₂. As shown in Table 2, total hepatic blood flow (i.e., the sum of portal blood flow and hepatic arterial blood flow) decreased by 51.6% (P < 0.05) at 20 h after the onset of sepsis. Administration of PTX early after CLP, however, maintained hepatic perfusion. Similarly, hepatic DO₂ decreased by 43.5% (P < 0.05) at 20 h after CLP, whereas administration of PTX early after the onset of sepsis prevented the decrease in hepatic DO₂. DO₂ in the liver increased by 135.8% (P < 0.05) in PTX-treated vs. vehicle-treated animals at 20 h after CLP (Table 2). In contrast to DO₂, hepatic O₂ did not decrease significantly during late sepsis. Administration of PTX, however, increased hepatic O₂ by 155.2% (P < 0.05) at 20 h after the onset of sepsis. Hepatic ERO₂ was not altered with either vehicle or PTX treatment during late sepsis (Table 2).

View this table: [in this window] [in a new window]   Table 2.   Effects of PTX infusion on BF, regional DO2, O2, and ERO2 at 20 h after CLP

Effects of PTX on intestinal blood flow, DO₂, and O₂. Regional perfusion in the small intestine decreased by 64.7% (P < 0.05) at 20 h after CLP; however, treatment with PTX prevented the decrease in intestinal blood flow (Table 2). Similarly, intestinal DO₂ and O₂ decreased by 60.5 and 44.4% (P < 0.05), respectively, at 20 h after the onset of sepsis. Administration of PTX, however, increased intestinal DO₂ and O₂ by 263.7 and 253.3% (P < 0.05), respectively, compared with vehicle-treated animals. Intestinal ERO₂ increased significantly at 20 h after CLP, irrespective of PTX administration (Table 2).

Effects of PTX on renal blood flow, DO₂, and O₂. Renal blood flow decreased by 42.3% (P < 0.05) at 20 h after CLP, and treatment with PTX prevented the decrease of intestinal blood flow (Table 2). Similar to the tissue perfusion, renal DO₂ decreased by 36.4% (P < 0.05) at 20 h after the onset of sepsis. Administration of PTX, however, increased renal DO₂ by 85.8% (P < 0.05) compared with vehicle-treated animals (Table 2). Although renal O₂ did not change at 20 h after the onset of sepsis, infusion of PTX early after the onset of sepsis significantly increased renal O₂ even compared with sham-operated animals. Similar to intestinal ERO₂, renal ERO₂ increased significantly at 20 h after CLP irrespective of PTX administration (Table 2).

Effects of PTX on plasma levels of ALT and lactate. Plasma levels of ALT increased by 718.2% (P < 0.05) at 20 h after CLP. PTX treatment, however, significantly attenuated the increase in plasma levels of ALT (P < 0.05) (Table 3). Similarly, whereas plasma levels of lactate increased by 173.4% (P < 0.05) at 20 h after CLP, administration of PTX attenuated the increase in plasma levels of lactate (P < 0.05, Table 3).

Effects of PTX on morphological alterations. In contrast to sham-operated animals (Fig. 1A), patches of hepatocytes were necrotic with eosinophilic cytoplasm and condensed nuclei at 20 h after CLP with vehicle treatment (Fig. 1B). The compromised hepatocytes, however, were significantly attenuated with PTX treatment (Fig. 1C). In the small intestine, epithelial cells along the villi were damaged or detached at 20 h after CLP (Fig. 2B) compared with the section of a sham animal. However, PTX treatment markedly improved intestinal structure and reduced villi damage (Fig. 2C). At 20 h after CLP, renal histology revealed faintly stained tubules due to tubular edema (Fig. 3B). Moreover, epithelial cells showed degeneration, and Bowman's space was narrowed (Fig. 3B). Although slightly swollen tubules could still be seen in the PTX-treated septic animals compared with shams (Fig. 3A), the glomerular capsule appeared normal (Fig. 3C).

View larger version (87K): [in this window] [in a new window]   Fig. 1.   A: photomicrograph of hepatic section from a sham-operated rat. Normal liver structures are demonstrated. B: photomicrograph of hepatic section from a septic rat with vehicle (normal saline) treatment. Patches of hepatocytes show necrosis with eosinophilic cytoplasm nuclei that are condensed and intensely stained with hematoxylin. C: photomicrograph of section from a septic rat with pentoxifylline (PTX) treatment. Liver structures appear normal. Bar = 100 µm.

View larger version (74K): [in this window] [in a new window]   Fig. 2.   A: photomicrograph of small intestinal section from a sham-operated rat. Villus and lamina propria structures appear normal. B: photomicrograph of small intestinal section from a septic rat with vehicle (normal saline) treatment. Tissue section shows neutrophil accumulation, as well as swollen and partially damaged villi. C: photomicrograph of small intestinal section from a septic rat with PTX treatment. Villus and lamina propria were significantly improved. Bar = 200 µm.

View larger version (93K): [in this window] [in a new window]   Fig. 3.   A: photomicrograph of renal section from a sham-operated rat. This section demonstrates normal structures of glomeruli, Bowman's space, and tubule lumina. B: photomicrograph of renal section from a septic rat with vehicle (normal saline) treatment. Limited number of tubular epithelial cells underwent degeneration and Bowman's space was narrowed. C: photomicrograph of renal section from a septic rat with PTX treatment. Although some slightly swollen tubules can be seen, glomerular capsule appears normal. Bar = 100 µm.

Effects of PTX on mortality rate. The mortality rate after CLP and cecal excision with vehicle treatment was 30 and 40% at day 1 and day 2, respectively (Table 4). It increased to 50% at days 3-10 after the completion of CLP and cecal excision. However, treatment with PTX prevented lethality throughout the 10-day observation period (P < 0.05 on days 2-10) (Table 4).

Effects of PTX on hemodynamic parameters and oxygen utilization in sham-operated animals. As indicated in Table 5, administration of PTX did not significantly alter cardiac output, regional blood flow, or systemic and regional DO₂ or O₂ in sham-operated animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Sepsis produces an aberrant and excessive host response causing a diffuse inflammatory insult that can lead to cell and organ dysfunction and failure associated with a high mortality rate (2, 7). Although cardiovascular and hemodynamic responses are enhanced during the early phase of polymicrobial sepsis, the late phase of sepsis is characterized by decreased cardiac output and tissue perfusion as well as increased vascular resistance (21, 38). In late sepsis, the inadequate tissue perfusion cannot meet both systemic and regional metabolic and oxygen demands. Because microcirculatory injury is considered a fundamental mechanism in the development of organ dysfunction and multiple organ failure during sepsis (9, 16), the primary objective in the management of sepsis is to maintain hemodynamic and cardiovascular stability and adequate tissue perfusion, and to meet the increasing metabolic and oxygen demands. Although studies have demonstrated that administration of PTX produces beneficial effects following sepsis (17, 25, 26, 34), endotoxic shock (43), or trauma and hemorrhage (36, 41), it remains unknown whether infusion of this agent early after the onset of sepsis delays or prevents the occurrence of hypodynamic responses during the progression of polymicrobial sepsis. The current study was therefore conducted to determine the effects of administration of PTX on various hemodynamic parameters, oxygen utilization, tissue histology, and mortality rate during sepsis.

The results indicate that cardiac output, stroke volume, blood flow in the liver, small intestine and kidneys, and systemic and regional DO₂ (i.e., hepatic, intestinal, and renal DO₂) decreased significantly, and TPR increased significantly at 20 h after CLP. Although we did not measure hemodynamic parameters and oxygen utilization during the early stage of sepsis in this study, previous work from our laboratory has demonstrated that cardiac output and hepatic perfusion increased, whereas TPR decreased significantly at 2-10 h after CLP (33). Moreover, perfusion in the small intestine, kidneys, and spleen increased at 5 h after CLP (40). The increased hemodynamic response during the early stage of sepsis was associated with an enhanced heart performance as evidenced by the increased maximal rate of the left ventricular pressure rise and fall (±dP/dt_max) (44). Furthermore, our preliminary results have shown that systemic and regional DO₂ increased significantly at 5 h after CLP. These findings, taken together, clearly indicate that cardiovascular and hemodynamic response during polymicrobial sepsis are biphasic, i.e., an early, hyperdynamic and hypercardiovascular response (2-10 h after CLP) followed by a late, hypodynamic response (20 h after CLP or later). In the septic patient, however, cardiac output is usually well maintained even at the late stage of sepsis (22). The discrepancy in cardiovascular responses during sepsis between clinical settings and animal studies appears to be due to the fact that the septic patient receives continuous fluid administration. In this regard, studies have shown that chronic resuscitation after CLP in the rat significantly reduces the mortality rate at 5 days after the onset of sepsis (42). Because continuous administration of large volumes of fluid in clinical settings produces the "hyperdynamic state" even in the patient with severe septic shock, the limitation of this study is that the septic animals only received fluid resuscitation (3 ml/100 g body wt normal saline) at the time of CLP procedure. Although early fluid resuscitation is needed to produce the hyperdynamic response during the early stage of sepsis (32), the lack of continuous fluid resuscitation may in part contribute to the reduced cardiac output observed at 20 h after CLP.

Our previous studies have indicated that cardiac contractility and ultrastructure did not appear to be compromised at 20 h after CLP (44). In the present study, however, cardiac output decreased significantly at the above time point. It is likely that the reduced circulating blood volume observed under such conditions (37) may be responsible for the low cardiac output at 20 h after the onset of sepsis. It should be noted that our previous studies (31, 33) have shown that the decrease in cardiac output at 20 h after CLP was not statistically different from sham-operated animals. Such a discrepancy in the changes of cardiac output during the late stage of sepsis between the current result and previous studies could be because cardiac output was measured by radioactive microspheres in the present study and by dye-dilution technique in the previous studies (31, 33). In addition, by using the radioactive microsphere technique, we previously reported that cardiac output was 17.39 ± 3.47 ml · min¹ · 100 g body wt¹ at 20 h after CLP (32), which is similar to the value indicated in the present study (16.3 ± 2.3 ml · min¹ · 100 g body wt¹, Table 1). Despite the fact that cardiac output decreased by more than 20% at 20 h after CLP in that study, such a decrease was not statistically different from sham-operated animals and could be due to the relatively large standard errors (32).

Under conditions of the reduced DO₂, such as hypoxia, anemia, and low cardiac output, tissue oxygen demands are inadequately met. If DO₂ decreases sufficiently, the compensatory mechanism (i.e., an increase in tissue oxygen extraction) can be exhausted, and the low DO₂ eventually limits O₂ and lactic acidosis occurs (6). Thus maintenance of DO₂ would be expected to be helpful in sepsis (12). In this regard, the present study clearly shows that administration of PTX at 1 h after the onset of sepsis markedly increased systemic and regional (i.e., hepatic, intestinal, and renal) DO₂ and O₂. Treatment with PTX also prevented the decrease in cardiac output, stroke volume, and blood flow in the liver, small intestine, and kidneys. In the vehicle-treated septic animals, plasma levels of ALT and lactate were markedly elevated, indicating hepatic damage and tissue hypoxia. Moderate morphological alterations in the liver, small intestine, and kidneys were observed at 20 h after the onset of sepsis. It is not surprising that the histological changes occurred under such conditions inasmuch as our previous studies have demonstrated that the microvascular perfusion decreased significantly at 20 h after CLP (40). The reduction in tissue perfusion during the late stage of sepsis was also confirmed by the present study. In addition, circulating levels of liver enzymes (e.g., ALT and aspartate aminotransferase) increased at 10-20 h after CLP (37), indicating hepatocellular injuries. Similarly, plasma levels of ALT were found to be elevated by more than sevenfold (Table 3) in this study. The elevated levels of lactate and ALT as well as the morphological alterations in the examined organs, however, were significantly attenuated by PTX treatment. In addition, PTX treatment decreased the mortality rate from 50 to 0% at days 3-10 after CLP and cecal excision. Thus the beneficial effects of PTX on cardiac output, tissue perfusion, and systemic and regional DO₂ and O₂ allow the host to meet the increased metabolic and oxygen demands during the late stage of sepsis. Such an increase in tissue perfusion and oxygen utilization therefore prevented the transition from the hyperdynamic phase to the hypodynamic phase, relieved tissue hypoxia, and attenuated organ damage during the late stage of sepsis. In view of this, PTX appears to be a useful adjunct for maintaining hemodynamic stability and oxygen utilization and preventing lethality from sepsis. In this regard, it is encouraging that randomized, double-blind clinical studies have been performed to determine the effects of PTX in septic patients. Staubach et al. (25) recently reported that continuous intravenous administration of PTX beneficially influenced cardiopulmonary functions in patients with sepsis without adverse side effects. Although larger trials are required to evaluate the efficacy of PTX in improving organ function in relation to the outcome of patients with severe sepsis, that study has clearly shown that the multiple organ dysfunction score decreased and the fraction of inspired oxygen ratios was improved in septic patients receiving PTX treatment (25).

Since Chalkiadakis et al. (4) reported the beneficial effects of PTX during sepsis in 1985, various studies have demonstrated that administration of PTX during sepsis or endotoxic shock preserves cardiac output and small intestine microvascular blood flow (26), reduces multiple organ damage (11), prevents TNF--induced lung injury (15), improves the tissue oxygen extraction capabilities (43) and oxygen utilization (1), and increases survival of septic animals (24). Although the precise mechanisms responsible for the beneficial effects of PTX in late sepsis remain unknown, studies have indicated that the phosphodiesterase inhibitor PTX increases intracellular cAMP levels, which increase the activity of protein kinase A (cAMP-dependent kinase) and upregulate cAMP-dependent response element binding protein (a nuclear factor) to downregulate the expression of TNF- messenger RNA (8, 18). It has been proposed that the downregulation of the proinflammatory cytokine TNF- during adverse circulatory conditions may be the mechanism by which PTX produces various beneficial effects (8, 27, 36). In addition to the downregulation of TNF- by PTX, Voisin et al. (29) reported that pretreatment with this agent in a rat model of Escherichia coli bacteremia prevented the increase in IL-1 and IL-6. Similarly, Lundblad et al. (17) reported that the above-mentioned proinflammatory cytokines were reduced by PTX treatment during fulminant intra-abdominal sepsis. Moreover, we have previously demonstrated that administration of PTX following hemorrhagic shock significantly decreased circulating levels of TNF and IL-6 (36). Furthermore, Nelson et al. (19) have recently shown that PTX reduced lipopolysaccharide binding protein mRNA in the liver and small intestine of septic animals. Thus downregulation of proinflammatory cytokines and other mediators may be responsible for the beneficial effects of PTX observed in the present study. PTX also improves erythrocyte deformability, decreases neutrophil oxidative burst, adherence, and lysozyme degranulation, and inhibits the increase in free intracellular calcium (13, 20). Such non-TNF- effects of this agent may also play an important role in the beneficial effect PTX on the survival of septic animals (10). Our previous studies have shown that PTX modulates cardiovascular responses during polymicrobial sepsis (34, 39). Administration of PTX early after the onset of sepsis increases left ventricular performance parameters such as maximal rates of pressure rise and fall (±dP/dt_max) and ventricular peak systemic pressure and reduces ventricular diastolic pressure at 5 h after CLP (34). In addition, endothelium-dependent (acetylcholine-induced) vascular relaxation decreased significantly at 10-20 h after CLP. Administration of PTX, however, maintained acetylcholine-induced vascular relaxation at both time points, suggesting that this agent improved the release of nitric oxide produced by constitutive nitric oxide synthase (i.e., endothelial cell function) (39). More recently, our preliminary results have indicated that vascular responsiveness to a novel vasodilator peptide adrenomedullin decreased significantly at 20 h after CLP in large blood vessels (the aorta) and in the resistance blood vessels (small arteries and arterioles) in the small intestine. Administration of PTX, however, attenuated the reduced adrenomedullin-induced vascular relaxation, which was associated with downregulation of proinflammatory cytokines such as TNF-, IL-1, and IL-6. It appears that the maintenance of vascular adrenomedullin responsiveness (presumably through downregulation of proinflammatory cytokines) may play an important role in producing the beneficial effects observed in the present study. Nonetheless, further studies are required to provide further insight into the mechanism by which PTX produces its beneficial effects.

Because PTX is a potent vasodilatory agent, it could be argued that the beneficial effects of PTX on various hemodynamic parameters and oxygen utilization are the direct result of its effect on vasculature. This, however, was not the case in the present study, since infusion of PTX at 1 h after sham operation did not affect cardiac output, organ blood flow, and tissue oxygen utilization at 20 h after surgery. It is likely, however, that the maintenance of systemic and regional DO₂ and O₂ during late sepsis following PTX administration is the direct result of better cardiac output and tissue perfusion. Further studies are required to determine whether the effect of PTX on tissue perfusion are caused by the modulation of vascular responsiveness to endogenous vasoactive agents such as adrenomedullin during polymicrobial sepsis. Whereas administration of PTX resulted in various beneficial effects when it was given early after the onset of sepsis, the beneficial cardiopulmonary effects of this agent were lost once septic shock was established (23). Thus although PTX prevents or delays the transition from the early, hyperdynamic phase to the late, hypodynamic phase of sepsis, it remains unknown whether this agent produces any salutary effects if administered during the late stage of sepsis or septic shock (i.e., the terminal stage of sepsis).

In summary, our results indicate that administration of PTX early after the onset of sepsis maintained cardiac output and regional blood flow, increased systemic and regional DO₂ and O₂, reduced multiple organ damage, and attenuated lactic acidosis. Moreover, PTX treatment significantly decreased the mortality rate of septic animals. Because PTX prevents the occurrence of hypodynamic sepsis, this agent appears to be a useful adjunct for maintaining hemodynamic stability and preventing lethality from sepsis.