AJP - Regu AJP: Renal Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Regul Integr Comp Physiol 281: R1177-R1185, 2001;
0363-6119/01 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Quezado, Z.
Right arrow Articles by Eichacker, P. Q.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Quezado, Z.
Right arrow Articles by Eichacker, P. Q.
Vol. 281, Issue 4, R1177-R1185, October 2001

Acute G-CSF therapy is not protective during lethal E. coli sepsis

Zenaide Quezado, Chantal Parent, Waheedullah Karzai, Michael Depietro, Charles Natanson, William Hammond, Robert L. Danner, Xizhong Cui, Yvonne Fitz, Steven M. Banks, Eric Gerstenberger, and Peter Q. Eichacker

Critical Care Medicine Department, Clinical Center, National Institutes of Health, Bethesda, Maryland 20892


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

We investigated whether decreases in circulating polymorphonuclear neutrophils (PMN) during lethal Escherichia coli (E. coli) sepsis in canines are related to insufficient host granulocyte colony-stimulating factor (G-CSF). Two-year-old purpose-bred beagles had intraperitoneal E. coli-infected or -noninfected fibrin clots surgically placed. By 10 to 12 h following clot, both infected survivors and nonsurvivors had marked increases (P = 0.001) in serum G-CSF levels (mean peak G-CSF ng/ml ± SE, 1,931 ± 364 and 2,779 ± 681, respectively) compared with noninfected controls (134 ± 79), which decreased at 24 to 48 h. Despite increases in G-CSF, infected clot placement caused delayed (P = 0.06) increases in PMN (mean ± SE change from baseline in cells × 103/mm3 at 24 and 48 h) in survivors (+3.9 ± 3.9 and +13.8 ± 3.6) compared with noninfected controls (+13.1 ± 2.8 and +9.1 ± 2.5). Furthermore, infected nonsurvivors had decreases in PMN (-1.4 ± 1.0 and -1.1 ± 2.3, P = 0.006 compared with the other groups). We next investigated whether administration of G-CSF immediately after clot placement and continued for 96 h to produce more rapid and prolonged high levels of G-CSF after infection would alter PMN levels. Although G-CSF caused large increases in PMN compared with control protein from 2 to 48 h following clot in noninfected controls, it caused much smaller increases in infected survivors and decreases in infected nonsurvivors (P = 0.03 for the ordered effect of G-CSF comparing the three groups). Thus insufficient host G-CSF is unlikely the cause of decreased circulating PMN in this canine model of sepsis. Other factors associated with sepsis either alone or in combination with G-CSF itself may reduce increases or cause decreases in circulating PMN.

granulocyte colony-stimulating factor; infection


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

IN THE INTACT HOST, the polymorphonuclear neutrophil (PMN) is a key component in host defense during acute bacterial infection (20). In most patients, despite increased extravascular recruitment of PMN during infection, accelerated release of cells from the bone marrow increases circulating PMN numbers. However, in some patients, severe infection accompanied by sepsis and septic shock can be associated with reductions in circulating PMNs and a worsened outcome (1, 13). Understanding the causes of such reductions may provide measures to improve treatment and outcome in these patients.

Granulocyte colony-stimulating factor (G-CSF) regulates PMN production and function (21, 24). In the normal host, increased production of G-CSF appears central in the recruitment and stimulation of circulating PMNs at sites of infection. Levels of both G-CSF and circulating PMN have been shown to increase acutely during bacterial infection. Reductions or inhibition of G-CSF decrease neutrophil release from the bone marrow and worsen host defense and outcome. Whether reductions in circulating PMN in patients with severe infection and sepsis are related to insufficient release of host G-CSF is not clear.

In this investigation, we studied the relationship between decreased circulating PMNs and serum G-CSF levels as well as the effects of supplementing the host G-CSF response with recombinant G-CSF at the onset of lethal Escherichia coli (E. coli) sepsis. We used an extensively validated antibiotic-treated canine model of lethal intraperitoneal E. coli sepsis. In this model, infected animals present a pattern and time course of bacterial and host mediator release as well as cardiovascular and pulmonary dysfunction similar to that observed in humans with sepsis (5, 17). After the onset of sepsis in this model, circulating PMNs are decreased in nonsurvivors compared with survivors (17). In addition, the therapeutic use of antibiotics and fluid, a routine practice in intensive care units, has been shown to improve the survival of infected animals (18). With the use of this same model, we showed previously that administration of recombinant G-CSF before the onset of sepsis caused marked increases in PMN during sepsis and increased hemodynamic function and survival (6).


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

To evaluate the effect of supplementing host G-CSF on circulating PMN during lethal E. coli sepsis, we conducted two experiments. In the first experiment, we studied endogenous G-CSF and PMN levels, and in the second, we administered recombinant G-CSF in canines challenged with E. coli-impregnated thrombin-fibrin clots placed surgically in the peritoneum.

Animal Care

The protocol used in this study was approved by the Animal Care and Use Committee of the Clinical Center of the National Institutes of Health.

Experiment 1

In canines receiving either sterile (n = 10) or E. coli-infected (n = 17) clots, G-CSF and circulating and lung lavage PMN levels and hemodynamics were determined. In all animals, 7 days before and 1, 2, 4, and 24 days after clot placement, complete and differential blood counts (CBC; ZB1, Coulter Electronics, Hialeah, FL, and Wright stain) were measured, and blood cultures were obtained (5, 17). In addition, on these same days as well as 1 h before and at 2- to 4-h intervals up to 12 h after sterile or E. coli-infected clot placement, plasma endotoxin [detection sensitivity 0.1 endotoxin U/ml (EU/ml)] and serum tumor necrosis factor (TNF; detection sensitivity 0.5 pg/ml) and G-CSF levels were determined (5, 6). G-CSF levels were measured using a modification of the NFS-60 bioassay (2). Aliquots of serum were taken from dogs at various time points after inoculation and were mixed with NSF-60 cells in thymidine-free medium, either with or without polyclonal rabbit antibody to recombinant G-CSF, and incubated at 37°C in 5% CO2. Then 1 µCi of [3H]thymidine was added for 4-6 h to the wells; cells were harvested, and counts were determined by liquid scintillation. Antibody-inhibitable counts were determined from a standard recombinant canine G-CSF dose-response curve, and the concentration of canine G-CSF was interpolated for each sample.

Experiment 2

Canines receiving either sterile (n = 8) or E. coli-infected clot (n = 24) were randomized to receive control protein or low- or high-dose canine recombinant G-CSF. All animals had circulating and lung lavage PMN levels and hemodynamics determined. Blood samples were drawn for cultures 7 days before and 1, 2, 4, and 21 days after clot placement. In addition, in all animals on these same days as well as 1 h before and at 2- to 4-h intervals up to 12 h after clot placement, blood was drawn for CBCs and plasma endotoxin and serum TNF levels.

Immediately after sterile or E. coli-infected clot implantation, animals were randomized to begin treatment with either canine recombinant G-CSF or control protein. In the low-dose regimen protocol, animals were treated with G-CSF or control protein 40 µg/kg body wt sc every 12 h for 2 days and then every 24 h for 2 days. In the high-dose regimen protocol, animals were treated with G-CSF or control protein 40 µg/kg body wt via both subcutaneous and intravenous routes (total G-CSF or control protein 80 µg · kg-1 · dose-1) at the same time points as the low-dose regimen. These doses were chosen, based on pharmacokinetic data, to simulate the high levels of host G-CSF noted in septic animals in experiment 1 (14). In addition, daily prophylactic administration of G-CSF over 7 days resulting in cumulative doses comparable to the individual ones used in this study had been shown to increase neutrophil numbers during sepsis and to be protective in this canine model (6). The canine recombinant G-CSF used in this study was supplied by Amgen (Thousand Oaks, CA) and was prepared as previously described (6). Control animals received canine serum protein.

Infected and Sterile Clot Implantation

In both experiments 1 and 2, 2-yr-old purpose-bred beagles (8-12 kg) had surgical intraperitoneal placement of either a thrombin-fibrin clot infected with E. coli 0111 (1.5 × 1010 colony-forming U/kg body wt) or a sterile thrombin-fibrin clot (5, 17). Clot implantation was performed under general anesthesia and required 30 to 60 min. All further studies were performed in awake animals with local anesthesia for procedures. All animals were treated with ceftriaxone (100 mg/kg every 24 h iv) starting 6 h after sterile or infected clot placement for a total of 5 days.

Catheter Placement and Hemodynamic and Laboratory Measures

In both experiments, 7 days before and 1, 2, 4, and 21 days after clot placement, awake animals had femoral and pulmonary arterial catheters placed, using local anesthesia (1% lidocaine) (5, 17). After this, intravascular catheter hemodynamics, arterial and mixed venous blood gas determinations, and radionuclide cineangiographic left ventricular ejection fractions (LVEFs) were performed before and immediately after each of two sequential volume challenges (15 and 45 ml/kg body wt of Ringer solution warmed to 37°C and infused over 15 and 45 min, respectively). Hemodynamic measures included intravascular systemic and pulmonary arterial pressure determinations and thermodilution cardiac outputs. Body temperature was measured using the pulmonary artery catheter, and hemodynamic data were indexed by body weight. LVEF measurements were performed before the initial volume infusion and after the final one. Arterial and mixed venous blood samples were analyzed with a Radiometer ABL-300 blood gas analyzer (Copenhagen, Denmark). Alveolar-to-arterial oxygen gradients were calculated. In addition, routine chemistries, lactate, and liver function tests were measured. Catheters were removed at the end of each study day. Additional volume infusions (Ringer, 50 ml/kg body wt) were administered to all animals 2 and 6 h after clot placement to maintain hydration.

Tracheostomy and Bronchoalveolar Lavage

In all studies, 14 days before clot placement, animals had a permanent tracheostomy surgically constructed under general anesthesia as previously described (4). The tracheal stoma was allowed to mature for 7 days before initiation of studies. Seven days before and 1, 2, 4, and 21 days after clot placement, animals underwent bronchoalveolar lavage (BAL). This was performed via the tracheal stoma in awake animals with topical anesthesia (1% lidocaine). BAL cell counts and differentials and protein determinations were obtained as previously described (3).

Statistics

Survival data were analyzed using a two-sample Wilcoxon test (26). Cardiopulmonary, BAL, and laboratory data were analyzed using ANOVA (25). In experiment 1, baseline data were analyzed using a one-way ANOVA controlling for study group. To examine changes over time, two additional effects were added to the ANOVA model, dog and time, as well as all interaction terms including time. All interaction terms involving dog were pooled to form the error term in the ANOVA. In experiment 2, baseline data were analyzed using an ANOVA with two main effects, infection (septic and sterile clot) and dose of G-CSF (control, low, and high dose), and the interaction between these two main effects. As in experiment 1, the effect of time is examined by adding both dog and time main effects to the model and all interactions involving time. When no significant differences were observed between the low- and high-dose groups of G-CSF, these groups were combined to increase the power of the experiment to find G-CSF effects.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Experiment 1: G-CSF and PMN Levels During Lethal E. coli Sepsis

Clinical findings. Seven of 17 infected animals and no noninfected controls died after clot placement (P < 0.01) (Fig. 1). Infected animals demonstrated signs of sepsis, i.e., weakness and lethargy. Noninfected controls appeared well throughout.


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1.   Proportion of animals surviving following noninfected or Escherichia coli (E. coli)-infected clot implantation.

Serum G-CSF. At baseline before clot placement, circulating G-CSF levels did not differ significantly (P = NS) among study groups. Within 10 to 12 h following clot placement, both infected survivors and nonsurvivors had marked increases (P < 0.001) in serum G-CSF levels compared with noninfected controls (Table 1). At recovery, G-CSF levels in infected surviving animals were similar to noninfected controls and baseline (both P = NS; Table 1).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Serial mean blood G-CSF levels before and after clot placement

Circulating PMNs. At baseline before clot placement, circulating PMN numbers did not differ significantly (P = NS) among study groups. Noninfected control animals had increases in PMN that were maximal at 24 h and then decreased at 48 h after clot placement (Table 2, Fig. 2). Infected survivors, in a pattern that was different (P = 0.06) compared with noninfected controls, had increases in PMNs that were not maximal until 48 h after clot placement. Infected nonsurvivors had PMN levels that decreased and were lower (P = 0.006) compared with noninfected controls and infected survivors at both 24 and 48 h after clot placement. At recovery, circulating PMNs in infected surviving animals were similar to noninfected controls and baseline (both P = NS). Other circulating cells did not differ (P = NS) comparing infected and control animals at any time during the study.

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Serial mean circulating PMN before and after clot placement



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 2.   Mean (±SE) serial changes in circulating polymorphonuclear neutrophil (PMN) following clot placement in noninfected controls and in E. coli-infected survivors and nonsurvivors.

Cardiopulmonary changes. At baseline before clot placement, circulating cardiopulmonary parameters did not differ significantly (P = NS) among study groups. Twenty-four to 48 h following clot placement, infected animals compared with controls had significant decreases in mean arterial pressure (MAP; P < 0.01) (Table 3), LVEF (P < 0.01; Table 3), cardiac index (CI; P < 0.02) (Table 3), and arterial oxygen pressure (PaO2; P < 0.05) (Table 4) and increases in BAL PMNs (P < 0.05) (Table 4). At recovery, all parameters in infected surviving animals were similar (P = NS) to noninfected controls.

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Mean MAP, LVEF, and CI


                              
View this table:
[in this window]
[in a new window]
  		Table 4.   Mean PaO2 and BAL PMN counts

Other laboratory data. At baseline before clot placement, laboratory data did not differ significantly (P = NS) among study groups. After clot placement, blood cultures were positive for E. coli in infected animals but not uninfected controls. Plasma endotoxin (EU/ml) and serum TNF (pg/ml) levels (mean peak levels ± SE, calculated based on the maximal measure noted in each animal), respectively, were increased in infected survivors (41 ± 1 and 191 ± 75) and infected nonsurvivors (1,018 ± 303 and 228 ± 107) compared with noninfected controls (1 ± 1 and 2 ± 1) up to 48 h after clot placement (P < 0.001 for both). Compared with infected survivors, infected nonsurvivors had significant (P < 0.05; data not shown) increases in plasma endotoxin levels. All other laboratory data did not differ significantly (P = NS) between groups at any time during study.

Experiment 2: Recombinant G-CSF Administration During Lethal E. coli Sepsis

Clinical findings. All animals receiving sterile noninfected clot (n = 8) survived. Among the low-dose regimen (40 µg · kg-1 · dose-1) animals challenged with E. coli-infected clot, mortality rates were similar in both control protein and G-CSF groups (2 of 6 animals in each; P = NS) (Fig. 3). In the high-dose regimen (80 µg · kg-1 · dose-1) animals, G-CSF resulted in a trend toward increased mortality rates compared with the control protein group (4 of 6 vs. 1 of 6, respectively; P = 0.1 Wilcoxon for harm).


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 3.   Proportion of animals surviving following E. coli-infected clot implantation in animals treated with control protein or granulocyte colony-stimulating factor (G-CSF) in low (40 µg/kg body wt; A) or high (80 µg/kg body wt; B) doses.

Circulating PMNs. At baseline before clot placement, circulating cell numbers did not differ significantly (P = NS) among study groups. In noninfected animals receiving control protein, clot placement was associated with increases in circulating PMN that were maximal at 24 h (Table 5). With control protein, PMN 2 to 24 h after clot placement in infected survivors and nonsurvivors was decreased compared with noninfected controls (P < 0.01; Table 5). Both regimens of G-CSF produced similar (P = NS; Table 5) changes in PMN within either noninfected controls, infected survivors or nonsurvivors. Therefore, results with the two G-CSF regimens in each of these groups were combined to increase the likelihood of detecting significant effects with G-CSF. From 2 to 48 h after clot placement, G-CSF produced large increases in PMN in noninfected controls, smaller increases in infected survivors, and decreases in infected nonsurvivors (Table 5, Fig. 4). The effects of G-CSF on changes in PMN number over this time period comparing these groups were ordered (noninfected controls > infected survivors > infected nonsurvivors, P = 0.03). Circulating PMN in infected survivors was lower (P = 0.01) than control protein-treated noninfected controls from 2 to 48 h following clot placement (Table 5, Fig. 4).

                              
View this table:
[in this window]
[in a new window]
  		Table 5.   Serial mean circulating PMN before and after placement of clot in animals receiving G-CSF or control protein



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4.   Mean (±SE) serial changes in circulating PMN following clot placement in noninfected controls and in E. coli-infected survivors and nonsurvivors treated with G-CSF or control protein.

Cardiopulmonary changes. At baseline before clot placement, cardiopulmonary parameters did not differ significantly (P = NS) among study groups. After clot placement at 24 and 48 h, compared with noninfected controls, all animals receiving an E. coli-infected clot exhibited decreases in MAP (P = 0.03; Table 6), LVEF (P = 0.0001; Table 6), CI (P = 0.004; Table 6), and PaO2 (P = 0.002; Table 7) and increases in BAL PMN concentrations (P = 0.05; Table 7). Compared with control protein, G-CSF had no different effect (P = NS) on cardiopulmonary and BAL parameters in either noninfected- or infected-clot animals at any time during the study.

                              
View this table:
[in this window]
[in a new window]
  		Table 6.   Serial mean MAP, LVEF, and CI before and after placement of clot in animals receiving G-CSF or control protein


                              
View this table:
[in this window]
[in a new window]
  		Table 7.   Serial mean PaO2 and BAL PMN before and after placement of clot in animals receiving G-CSF or control protein

Laboratory analysis. At baseline before clot placement, all laboratory parameters as outlined in METHODS did not differ significantly (P = NS) among study groups. After E. coli-infected clot placement in animals, there were significant increases in blood bacteria counts and plasma endotoxin and serum TNF levels compared with sterile clot animals (all P = 0.001, data not shown). G-CSF did not produce significant (P = NS) changes in any laboratory parameter besides circulating PMN.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The findings from this study suggest that insufficient host G-CSF is not the reason why the number of circulating PMN is reduced during E. coli sepsis in this canine model. We found that infected clot placement was associated with marked increases in serum G-CSF levels by 10 to 12 h. Despite these early increases in serum G-CSF, infection was associated with delayed increases in PMN in survivors and decreases in PMN in nonsurvivors. We then investigated whether increasing serum G-CSF levels earlier or for a longer duration would increase circulating PMN in the model and change outcome. High doses (40 or 80 µg/kg) of recombinant G-CSF were administered at the onset of sepsis and for up to 96 h during the time of maximum sepsis. Treatment with G-CSF did produce a small increase in PMN in infected survivors. However, these increases were not as great as changes with G-CSF in noninfected controls nor were they as high as increases noted in control protein-treated noninfected animals. Furthermore, G-CSF treatment resulted in paradoxical reductions in PMN in infected nonsurvivors, and in the highest dose, it was associated with a trend toward worsened mortality rate. Thus factors other than insufficient G-CSF appear to result in the reduced circulating PMN noted in this model.

One possibility is that sepsis suppresses bone marrow release of PMN. However, kinetic studies have shown that the acute stages of lethal bacterial infection in canines were associated with marked increases in bone marrow release of neutrophils (15). Furthermore, in infected animals in the present study, treatment with G-CSF in survivors was associated with additional small increases in PMN, suggesting the presence of a marrow still responsive to stimulatory agents.

Alternatively in this model, one could postulate that despite additional exogenous G-CSF, extravascular uptake of PMN might have surpassed the marrow's ability to maintain circulating PMN levels. This uptake may have occurred at the site of infection. The E. coli bacteria employed in this model has itself been shown to provide a strong stimulus for extravascular PMN recruitment (8). Also, challenge in rats with E. coli-impregnated agar pellets intraperitoneally causes increased peritoneal neutrophil numbers that are augmented by pretreatments with G-CSF (3). Endotoxin and TNF, two mediators recognized to stimulate the activation and adhesion of PMN to vascular endothelium at sites of infection, were greatly increased in infected animals, more so in nonsurvivors than survivors (23). The possibility that a decrease in circulating PMN is associated with increased intraperitoneal recruitment of PMN has been shown in other peritonitis models (27). However, unlike the present investigation, in that study, G-CSF was given prophylactically and increased survival. Nevertheless, it is possible that decreases in circulating PMN in the setting of additional exogenous G-CSF reflect increased intraperitoneal recruitment of PMN in our model.

Extravascular uptake of PMN may have occurred in noninfected tissues as well. BAL PMN numbers were increased and arterial oxygenation decreased in septic animals, suggesting that mediators associated with sepsis had caused PMN recruitment and inflammatory injury at sites distant from the original site of infection. Previous studies have documented that peritonitis in this model causes PMN recruitment to other organs as well (19). Nevertheless, BAL PMN numbers did not differ in infected survivors and nonsurvivors nor were they altered by G-CSF treatment, suggesting that recruitment of PMN to noninfected organs does not contribute significantly to the reductions in circulating PMN number.

Despite higher although not significantly different serum G-CSF levels, infected nonsurvivors had greater decreases in PMN levels than survivors. These differences may relate to the counterregulatory effects increased circulating PMNs have been shown to have on serum G-CSF levels in normal animals (13). However, treatment with G-CSF in infected nonsurvivors was associated with further reductions in PMN. In combination, these findings raise the possibility that G-CSF may itself contribute to the reductions noted in circulating PMN when infection is most severe in this model. Both endotoxin and TNF, which were increased following the onset of sepsis and appeared higher in nonsurvivors than in survivors, have been shown to interact with G-CSF to increase PMN activation and adhesion (12, 16, 28-31). We previously showed in rats with E. coli pneumonia associated with endotoxemia and high TNF levels that G-CSF treatment reduced circulating PMN number and worsened tissue injury and survival rates (8, 11). In contrast, in rats with S. aureus pneumonia and significantly lower serum TNF levels than with E. coli, G-CSF increased PMNs and survival. Of note in the present study was the trend toward increased mortality rate in E. coli-infected animals receiving the highest dose of G-CSF. Increased clearance of neutrophils related to mediators produced during E. coli infection in combination with G-CSF treatment may have worsened inflammatory tissue injury or further reduced host defense.

Results of some clinical studies of G-CSF raise the possibility that severity of infection may influence the effect of G-CSF on PMN. One group of investigators has shown that in septic patients, the response to G-CSF may vary according to the patient's Acute Physiology and Chronic Health Evaluation (APACHE II) score (6). Patients with lower APACHE II scores had a good response to G-CSF, whereas those with higher APACHE II scores had a poor response. In addition, "good responders" to G-CSF had a 10% mortality rate, whereas "poor responders" had a 100% mortality rate. In a study by Gross-Weege et al. (10), G-CSF administration produced greater increases in circulating PMN counts in 10 patients with the systemic inflammatory response without infection, all of whom survived, than in 10 patients with sepsis and infection, only six of whom survived. Nelson et al. (22) showed that in patients with community-acquired pneumonia resulting in a low mortality rate (9%), G-CSF caused significant increases in circulating PMN number. In our study, in infected survivors, G-CSF very modestly increased, but in infected nonsurvivors, it decreased circulating PMN in the setting of lethal E. coli sepsis. Taken together, these results suggest that the response to G-CSF during sepsis may relate, in part, to severity of illness and the predicted mortality of a given population.

In contrast to the present study, we previously showed that initiation of G-CSF treatment several days before infected clot placement in this canine model resulted in five- to sixfold increases in circulating PMN counts that were maintained with additional doses of G-CSF treatment during sepsis (6). These increases in PMN were associated with reductions in bacteremia, endotoxemia, and TNF levels and improvements in cardiopulmonary function and survival in septic animals. Thus increasing bone marrow reserve with G-CSF in this model before the initiation of infection may have overcome reductions in circulating PMN related to the tissue uptake of cells and improved host defense and survival. A similar regimen of prophylactic G-CSF also increased circulating PMN and survival in canines challenged with intrabronchial E. coli (9). Therefore, the timing of therapy with G-CSF in relationship to the onset of infection may impact the host's response to G-CSF.

In conclusion, insufficient levels of G-CSF do not appear to cause the decreases in circulating PMN in this canine model of lethal E. coli sepsis. The administration of G-CSF early during infection and for longer periods (96 h), although causing small increases in circulating PMN in survivors, did not improve overall survival rates and was associated with paradoxical reductions in PMN in nonsurvivors. Thus other factors associated with sepsis in this model, either alone or in combination with G-CSF itself, may reduce increases or cause decreases in circulating PMN. These findings suggest that in patients, augmenting a normal host G-CSF response following the onset of severe sepsis associated with reductions in circulating PMN may do little to improve overall outcome.


    ACKNOWLEDGEMENTS

The writers thank S. Richmond, L. Wilson, D. Dolan, and A. Hilton for technical support during this study; Dr. J. Bacher for veterinary care and surgical procedures; and J. Candotti for preparation of the manuscript.


    FOOTNOTES

Address for reprint requests and other correspondence: P. Q. Eichacker, Critical Care Medicine Dept., National Institutes of Health, Bldg. 10, Rm. 7D43, Bethesda, MD 20892.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 6 April 2001; accepted in final form 1 June 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Bone, RC, Sprung CL, and Sibbald WJ. Definitions for sepsis and organ failure. Crit Care Med 20: 724-726, 1992[ISI][Medline].

2.   Dale, DC, Lau S, Nash R, Boone T, and Osborne W. Effect of endotoxin on serum granulocyte and granulocyte-macrophage colony-stimulating factor levels in dogs. J Infect Dis 165: 689-694, 1992[ISI][Medline].

3.   Dunne, JR, Dunkin BJ, Nelson S, and White JC. Effects of granulocyte colony stimulating factor in a nonneutropenic rodent model of Escherichia coli peritonitis. J Surg Res 61: 348-354, 1996[ISI][Medline].

4.   Eichacker, PQ, Hoffman WD, Farese A, Banks SM, Kuo GC, MacVittie TJ, and Natanson C. TNF but not IL-1 in dogs causes lethal lung injury and multiple organ dysfunction similar to human sepsis. J Appl Physiol 71: 1979-1989, 1991[Abstract/Free Full Text].

5.   Eichacker, PQ, Hoffman WD, Farese A, Danner RL, Suffredini AF, Waisman Y, Banks SM, Mouginis T, Wilson L, Rothlein R, Elin RJ, Hosseini JM, MacVittie TJ, and Natanson C. Leukocyte CD18 monoclonal antibody worsens endotoxemia and cardiovascular injury in canines with septic shock. J Appl Physiol 74: 1885-1892, 1993[Abstract/Free Full Text].

6.   Eichacker, PQ, Waisman Y, Natanson C, Farese A, Hoffman WD, Banks SM, and MacVittie TJ. Cardiopulmonary effects of granulocyte colony-stimulating factor in a canine model of bacterial sepsis. J Appl Physiol 77: 2366-2373, 1994[Abstract/Free Full Text].

7.   Endo, S, Inada K, Inoue Y, Yamada Y, Takakuwa T, Kazai T, Nakae H, Kuwata Y, Hoshi S, and Yashida M. Evaluation of recombinant granulocyte colony stimulating factor (rhG-CSF) therapy in granulopoietic patients with sepsis. Curr Med Res Opin 13: 233-241, 1991.

8.   Freeman, BD, Correa R, Karzai W, Natanson C, Patterson M, Banks S, Fitz Y, Danner RL, Wilson L, and Eichacker PQ. Controlled trials of rG-CSF and CD11b-directed MAb during hyperoxia and E. coli pneumonia in rats. J Appl Physiol 80: 2066-2076, 1996[Abstract/Free Full Text].

9.   Freeman, BD, Quezado Z, Zeni F, Natanson C, Danner RL, Banks S, Quezado M, Fitz Y, Bacher J, and Eichacker PQ. rG-CSF reduces endotoxemia and improves survival during E. coli pneumonia. J Appl Physiol 83: 1467-1475, 1997[Abstract/Free Full Text].

10.   Gross-Weege, W, Weiss M, Schneider M, Wenning M, Harms B, Dumon K, Ohmann C, and Roher HD. Safety of a low-dosage Filgrastim (rhG-CSF) treatment in non-neutropenic surgical intensive care patients with an inflammatory process. Intensive Care Med 23: 16-22, 1997[ISI][Medline].

11.   Karzai, W, von Specht BU, Parent C, Haberstroh J, Wollersen K, Natanson C, Banks SM, and Eichacker PQ. G-CSF during Escherichia coli versus Staphylococcus aureus pneumonia in rats has fundamentally different and opposite effects. Am J Respir Crit Care Med 159: 1377-1382, 1999[Abstract/Free Full Text].

12.   Khwaja, A, Carver JE, and Linch DC. Interactions of granulocyte-macrophage colony-stimulating factor (CSF), granulocyte CSF, and tumor necrosis factor alpha  in the priming of the neutrophil respiratory burst. Blood 79: 745-753, 1992[Abstract/Free Full Text].

13.   Knaus, WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, Sirio CA, Murphy DJ, Lotring T, Damiano A, and Harrell FE, Jr. The APACHE III prognostic system risk prediction of hospital mortality for critically ill hospitalized adults. Chest 100: 1619-1636, 1991[Abstract/Free Full Text].

14.   Kuwabara, T, Kobayashi S, and Sugiyama Y. Pharmacokinetics and pharmacodynamics of a recombinant human granulocyte colony-stimulating factor. Drug Metab Rev 28: 625-658, 1996[ISI][Medline].

15.   Marsh, JC, Boggs DR, Cartwright GE, and Wintrobe MM. Neutrophil kinetics in acute infection. J Clin Invest 46: 1943-1953, 1967.

16.   Mur, E, Zabernigg A, Hilbe W, Eisterer W, Halder W, and Thaler J. Oxidative burst of neutrophils in patients with rheumatoid arthritis: influence of various cytokines and medication. Clin Exp Rheumatol 15: 233-237, 1997[ISI][Medline].

17.   Natanson, C, Danner RL, Fink MP, MacVittie TJ, Walker RI, Conklin JJ, and Parrillo JE. Cardiovascular performance with E. coli challenges in a canine model of human sepsis. Am J Physiol Heart Circ Physiol 254: H558-H569, 1988[Abstract/Free Full Text].

18.   Natanson, C, Danner RL, Reilly JM, Doerfler ML, Hoffman WD, Akin GL, Hosseini JM, Banks SM, Elin RJ, MacVittie TJ, and Parrillo J. Antibiotics versus cardiovascular support in a canine model of human septic shock. Am J Physiol Heart Circ Physiol 259: H1440-H1447, 1990[Abstract/Free Full Text].

19.   Natanson, C, Fink MP, Ballantyne HK, MacVittie TJ, Conklin JJ, and Parrillo JE. Gram-negative bacteremia produces both severe systolic and diastolic cardiac dysfunction in a canine model that simulates human septic shock. J Clin Invest 78: 259-270, 1986.

20.   Natanson, C, Hoffman WD, Suffredini AF, Eichacker PQ, and Danner RL. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann Intern Med 120: 771-783, 1994[Abstract/Free Full Text].

21.   Nelson, S, and Bagby GJ. Granulocyte colony-stimulating factor and modulation of inflammatory cells in sepsis. Clin Chest Med 17: 319-332, 1996[ISI][Medline].

22.   Nelson, S, Summer W, Bagby G, Nakamura C, Stewart L, Lipscomb G, and Andresen J. Granulocyte colony-stimulating factor enhances pulmonary host defenses in normal and ethanol-treated rats. J Infect Dis 164: 901-906, 1991[ISI][Medline].

23.   Parent, C, and Eichacker PQ. Neutrophil and endothelial cell interactions in sepsis. The role of adhesion molecules. Infect Dis Clin North Am 13: 427-447, 1999[ISI][Medline].

24.   Root, RK, and Dale DC. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor: comparisons and potential for use in the treatment of infections in nonneutropenic patients. J Infect Dis 179, Suppl2: S342-S352, 1999.

25.   Scheffé, H. The Analysis of Variance. New York: Wiley, 1959.

26.   Siegel, S. Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill, 1956.

27.   Villa, P, Shaklee CL, Meazza C, Agnello D, Ghezzi P, and Senaldi G. Granulocyte colony-stimulating factor and antibiotics in the prophylaxis of a murine model of polymicrobial peritonitis and sepsis. J Infect Dis 178: 471-477, 1998[ISI][Medline].

28.   Wiltschke, C, Krainer M, Budinsky AC, Berger A, Muller C, Zeillinger R, Speiser P, Kubista E, Eibl M, and Zielinski CC. Reduced mitogenic stimulation of peripheral blood mononuclear cells as a prognostic parameter for the course of breast cancer: a prospective longitudinal study. Br J Cancer 71: 1292-1296, 1995[ISI][Medline].

29.   Wright, SD, Ramos RA, Hermanowski-Vosatka A, Rockwell P, and Detmers PA. Activation of the adhesive capacity of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J Exp Med 173: 1281-1286, 1991[Abstract/Free Full Text].

30.   Yagisawa, M, Yuo A, Kitagawa S, Yazaki Y, Togawa A, and Takaku F. Stimulation and priming of human neutrophils by IL-1 alpha  and IL-1 beta : complete inhibition by IL-1 receptor antagonist and no interaction with other cytokines. Exp Hematol 23: 603-608, 1995[ISI][Medline].

31.   Yuo, A, Kitagawa S, Kasahara T, Matsushima K, Saito M, and Takaku F. Stimulation and priming of human neutrophils by interleukin-8: cooperation with tumor necrosis factor and colony-stimulating factors. Blood 78: 2708-2714, 1991[Abstract/Free Full Text].

Am J Physiol Regul Integr Comp Physiol 281(4):R1177-R1185


This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
P. C. Minneci, K. J. Deans, B. Hansen, C. Parent, C. Romines, D. A. Gonzales, S.-X. Ying, P. Munson, A. F. Suffredini, J. Feng, et al.
A canine model of septic shock: balancing animal welfare and scientific relevance
Am J Physiol Heart Circ Physiol, October 1, 2007; 293(4): H2487 - H2500.
[Abstract] [Full Text] [PDF]

Home page
 Antimicrob. Agents Chemother.Home page
L. E. T. Stearne, A. G. Vonk, B. J. Kullberg, and I. C. Gyssens
Effect of Recombinant Murine Granulocyte Colony-Stimulating Factor with or without Fluoroquinolone Therapy on Mixed-Infection Abscesses in Mice
Antimicrob. Agents Chemother., September 1, 2005; 49(9): 3668 - 3675.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Quezado, Z.
Right arrow Articles by Eichacker, P. Q.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Quezado, Z.
Right arrow Articles by Eichacker, P. Q.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online