Lipopolyamines as a therapeutic strategy in experimental Gram-negative bacterial sepsis.

Lipopolyamines are a class of polycationic amphiphilic compounds that have been shown to bind with high affinity to polyanionic macromolecules, including both DNA and bacterial lipopoly-saccharide (LPS). One of these compounds, termed DOSPER (1,3-di-oleoyloxy-2-(6-carboxyl-spermyl)- propylamide), is non-cytotoxic and has been shown to inhibit LPS-mediated cytokine release and lethality in endotoxin challenge models. In the study reported here, the activity of DOSPER was tested in neutropenic rats with invasive Gram-negative bacteremia caused by Pseudomonas aeruginosa. DOSPER alone was ineffective (0/8) at influencing mortality, but provided a significant survival advantage if administered in combination with a bactericidal antibiotic, ceftazidime (10/12; P<0.05). Ceftazidime alone was partially protective (6/12) while the control group had no survivors (0/8). DOSPER administration markedly reduced circulating endotoxin levels (P<0.01) and interleukin-6 levels (P<0.05) but had no significant effect on bacteremia and bacterial concentrations of P. aeruginosa in liver or spleen tissue. Lipopolyamines may be potentially valuable as a therapeutic adjunct in treatment of Gram-negative bacterial sepsis.

Evaluation of the safety of recombinant P-selectin glycoprotein ligand-immunoglobulin G fusion protein in experimental models of localized and systemic infection.

P-selectin is a major component in the early interaction between platelets, endothelial cells, and inflammatory cells in the initial phases of the innate immune response. The major ligand for P-selectin is P-selectin glycoprotein ligand-1 (PSGL-1) and this ligand is expressed on the surface of monocyte, lymphocyte, and neutrophil membranes. A truncated form of recombinant human P-selectin glycoprotein ligand-1 has been covalently linked to immunoglobulin G (rPSGL-Ig) and this fusion peptide functions as a competitive inhibitor of PSGL-1. As an inhibitor of neutrophil-endothelial cell adherence, rPSGL-Ig is in early clinical development for the treatment of ischemia reperfusion injury. To determine the potential for deleterious effects from inhibition in P-selectin-mediated neutrophil attachment in the presence of bacterial infection, the effects of therapeutic doses of rPSGL-Ig were tested in three standard laboratory sepsis models. The experimental models included: the murine systemic Listeria monocytogenes infection model, the Pseudomonas aeruginosa bacteremia model in neutropenic rats, and the cecal ligation and puncture (CLP)-induced peritonitis model in rats. Recombinant human PSGL-Ig had no adverse effects on mortality or immune clearance in systemic bacterial infection in any of the three infection models. The PSGL-1 inhibitor did significantly decrease local neutrophil infiltration and bacterial clearance in the peritoneum following CLP, but this did not increase the systemic levels of proinflammatory cytokines, the quantitative levels of bacteremia, or the overall mortality rate following CLP. The results indicate that rPSGL-Ig did not exacerbate infection in these experimental sepsis models.

Active immunization with a detoxified Escherichia coli J5 lipopolysaccharide group B meningococcal outer membrane protein complex vaccine protects animals from experimental sepsis.

The passive infusion of antibodies elicited in rabbits with a detoxified J5 lipopolysaccharide (LPS)/group B meningococcal outer membrane protein complex vaccine protected neutropenic rats from heterologous lethal gram-negative bacterial infection. In this study, active immunization was studied in neutropenic rats infected with Pseudomonas aeruginosa, in the presence or absence of ceftazidime therapy, and with Klebsiella pneumoniae. This vaccine elicited a > 200-fold increase in anti-J5 LPS antibody, which remained elevated throughout the duration of cyclophosphamide-induced neutropenia and for < or = 3 months. There was improved survival among immunized versus control animals: 48% (13/28) versus 7% (2/29) in Pseudomonas-challenged rats; 61% (11/18) versus 0% (0/10) in Pseudomonas- and ceftazidime-treated rats; and 64% (9/14) versus 13% (2/15) in Klebsiella-challenged rats (P < 0.01 for each comparison). Immunized animals had lower levels of bacteria in organs and lower levels of circulating endotoxin at the onset of fever. In conclusion, active immunization with an anti-endotoxin vaccine improved survival after infection with > or = 2 heterologous, clinically relevant bacterial species in immunocompromised animals. Active immunization with this vaccine merits further investigation.

The activity of tissue factor pathway inhibitor in experimental models of superantigen-induced shock and polymicrobial intra-abdominal sepsis.

OBJECTIVES: To study recombinant human tissue factor pathway inhibitor (rhTFPI) in a superantigen-induced shock model and in a cecal ligation and puncture (CLP) model of peritonitis in mice. DESIGN: Prospective, randomized, experimental study. SETTING: An experimental animal research laboratory. SUBJECTS: Eighty BALB/c mice for the superantigen model, and 56 BALB/c mice for the CLP model. INTERVENTIONS: In the superantigen-induced shock model, animals received rhTFPI (350 mg/kg) subcutaneously every 12 hrs (n = 30) or saline control (n = 30) for 60 hrs after staphylococcal enterotoxin B (SEB; 10 microg iv) and a sublethal dose of E. coli 0111:B4 lipopolysaccharide (LPS; 75 microg ip). Control groups received SEB alone (n = 10) and LPS alone (n = 10). In the CLP model, rhTFPI or saline was given every 8 hrs for 48 hrs by using a 21-gauge needle (n = 9) or 23-gauge needle (n = 14) for CLP. A sham surgery control group (n = 10) was also included. MEASUREMENTS AND MAIN RESULTS: There was 0% mortality in the SEB and LPS control groups. The mortality rate was 64% in the saline control group that received both SEB and LPS (19 of 30), whereas the rhTFPI- treated animals had a mortality rate of 20% (6 of 30; p < .01). The rhTFPI-treated group had significantly lower interleukin-6 levels (61.8 +/- 41 pg/mL vs. 285 +/- 63 pg/mL; p < .05) than the control group but no differences in tumor necrosis factor-alpha or interferon-gamma levels. In the CLP experiment, rhTFPI-treated animals did not have any survival advantage over the control group after the large-bore (21-gauge) needle puncture. The rhTFPI group had significantly improved 7-day mortality rate after CLP with the small-bore needle (23-gauge; 21.4% [rhTFPI] vs. 71.4% [control], p < .01). Plasma LPS, interleukin-6, interferon-gamma, and tumor necrosis factor-alpha levels were unchanged by rhTFPI treatment, but significantly reduced LPS (p = .006) and IFNgamma (p = .001) levels were found in the peritoneal fluid. CONCLUSIONS: Tissue factor pathway inhibitor significantly improves the mortality rate in models of superantigen-induced shock and polymicrobial intra-abdominal infection, supporting its potential use in clinical trials for septic shock.

Differential antibiotic-induced endotoxin release in severe melioidosis.

Severe melioidosis is a life-threatening, systemic bacterial infection caused by Burkholderia pseudomallei. A prospective, randomized treatment trial was conducted in northeast Thailand to compare ceftazidime (a penicillin-binding protein [PBP]-3-specific agent that causes release of large amounts of endotoxin in vitro) and imipenem (a PBP-2-specific agent that kills B. pseudomallei more rapidly but releases low amounts of endotoxin) in severe melioidosis over a 6-h time course after the first dose of antibiotic. Despite similar clinical, microbiological, endotoxin, and cytokine measures at study entry, ceftazidime-treated patients (n=34) had significantly greater systemic endotoxin (P<.001) than patients treated with imipenem (n=34) after the first dose of antibiotic. No overall difference in mortality was observed (35% in both groups [95% confidence interval, 20%-50%]). Differential antibiotic-induced endotoxin release is demonstrable in severe melioidosis. These differences in endotoxin release did not appear to have a significant impact on survival in this group of patients.

Recombinant human interleukin-11 has anti-inflammatory actions yet does not exacerbate systemic Listeria infection.

To determine whether recombinant human (rh) interleukin (IL)-11 disrupts the clearance of microbial pathogens, mice were challenged with Listeria monocytogenes after receiving high-dose rhIL-11, anti-tumor necrosis factor (TNF) monoclonal antibody (MAb), anti-IL-11 MAb, or saline control. The LD50 was not affected by rhIL-11 but was 10-fold lower in the anti-TNF MAb group (P<.001). Plasma IL-6, IL-1beta, and TNF-alpha levels were not different between rhIL-11-treated animals and the control group; however, interferon-gamma levels were significantly reduced by IL-11 treatment (2477 vs. 0 pg/mL, P<.01). Compared with the control group, the quantitative level of L. monocytogenes in hepatic and splenic tissue was unchanged by rhIL-11 but was significantly increased by TNF or IL-11 inhibition. The results indicate that IL-11 down-regulates cytokine production but does not exacerbate systemic infection in the murine Listeria infection model.

Relationship between plasma levels of lipopolysaccharide (LPS) and LPS-binding protein in patients with severe sepsis and septic shock.

Plasma endotoxin and lipopolysaccharide-binding protein (LBP) levels were measured in a group of 253 patients at the onset of severe sepsis and/or septic shock. Endotoxin levels were significantly greater than control levels (n=33; mean +/- SD, 5.1+/-7.3 pg/mL) in 78.3% of patients. Median endotoxin levels in patients with sepsis were 300 pg/mL (25%-75% interquartile range, 110-726 pg/mL). LBP levels were elevated in 97% of patients compared with normal control values of 4.1+/-1.65 microgram/mL. Median LBP levels in patients with sepsis were 31.2 microgram/mL (interquartile range, 22.5-47.7 microgram/mL). Median endotoxin levels at study entry were more highly elevated (515 vs. 230 pg/mL; P<.01), and LBP levels were less highly elevated (28.0 vs. 33.2 microgram/mL; P<.05) in nonsurvivors than survivors over the 28-day study period. No correlation was found between endotoxin and LBP levels. The quantitative level of both endotoxin and LBP may have prognostic significance in patients with severe sepsis.

Additive effects of human recombinant interleukin-11 and granulocyte colony-stimulating factor in experimental gram-negative sepsis.

Recombinant human granulocyte colony-stimulating factor (rhG-CSF) is widely used to promote granulocyte recovery from a variety of pathologic states. Recombinant human interleukin-11 (rhIL-11) has recently become available clinically as a platelet restorative agent after myelosuppressive chemotherapy. Preclinical data has shown that rhIL-11 limits mucosal injury after chemotherapy and attenuates the proinflammatory cytokine response. The potential efficacy of combination therapy with recombinant human forms of rhIL-11 and rhG-CSF was studied in a neutropenic rat model of Pseudomonas aeruginosa sepsis. At the onset of neutropenia, animals were randomly assigned to receive either rhG-CSF at a dose of 200 micrograms/kg subcutaneously every 24 hours for 7 days; rhIL-11 at 200 micrograms/kg subcutaneously every 24 hours for 7 days; the combination of both rhG-CSF and rhIL-11; or saline control. Animals were orally colonized with Pseudomonas aeruginosa 12.4.4 and then given a myelosuppressive dose of cyclophosphamide. rhG-CSF resulted in a slight increase in absolute neutrophil counts (ANC), but did not provide a survival advantage (0 of 12, 0% survival) compared with the placebo group (1 of 12, 8% survival). rhIL-11 was partially protective (4 of 10, 40% survival); the combination of rhG-CSF and rhIL-11 resulted in a survival rate of 80% (16 of 20; P <.001). rhIL-11 alone or in combination with rhG-CSF resulted in preservation of gastrointestinal mucosal integrity (P <.001), lower circulating endotoxin levels (P <.01), and reduced quantitative levels of P. aeruginosa in quantitative organ cultures. These results indicate that the combination of rhIL-11 and rhG-CSF is additive as a treatment strategy in the prevention and treatment of experimental Gram-negative sepsis in immunocompromised animals. This combination may prove to be efficacious in the prevention of severe sepsis in neutropenic patients.

Recombinant human interleukin-11 in experimental Pseudomonas aeruginosa sepsis in immunocompromised animals.

The therapeutic potential of recombinant human interleukin-11 (rhIL-11) was tested in a neutropenic rat model that mimics the clinical consequences of myelosuppressive chemotherapy complicated by Pseudomonas aeruginosa sepsis. rhIL-11-treated animals (150 micrograms/kg intravenously every 24 h for 3 days) had reduced endotoxin levels (P < .05) and less pulmonary edema fluid (P < .001) and were protected (P < .01) against thinning and necrosis of the intestinal mucosa compared with the control group. The survival rate in rhIL-11-treated animals was 40% (19/47), whereas it was 0 (0 of 19) in the control group (P < .01). The addition of ciprofloxacin (10 mg/kg every 12 h) resulted in a survival rate of 9 (60%) of 15, while the combination of rhIL-11 and ciprofloxacin resulted in 100% survival (15/15; P < .05). These results indicate that rhIL-11 supports mucous membrane integrity of the alimentary tract and decreases the systemic inflammatory response to experimental gram-negative infection in immunocompromised animals.

Potential hazards of combination immunotherapy in the treatment of experimental septic shock.

Using an actual infection model of Pseudomonas aeruginosa sepsis in neutropenic rats, the potential utility of a combination anticytokine approach for the treatment of sepsis was tested. A dimeric tumor necrosis factor binding protein (TNF-BP) consisting of two soluble recombinant human TNF type 1 receptors linked with polyethylene glycol was used with recombinant human interleukin-1 receptor antagonist (IL-1ra). Despite having levels of bacteremia and endotoxemia similar to the control group (survivors, 0/18), 30% of IL-1ra-treated animals survived (P < .05); 31% of TNF-BP-treated animals survived (P < .01). Unexpectedly, the combination of IL-1ra plus TNF-BP proved to be uniformly fatal (survivors, 0/20). Endotoxin (P < .0001) and bacteremia (P < .01) levels were >10-fold higher than levels in animals treated with IL-1ra alone, TNF-BP alone, or placebo. Disseminated microabscesses in major organs were found in animals treated with combination immunotherapy. Combination anticytokine therapy may exacerbate systemic infection and worsen outcome in experimental sepsis.

Activity of lipopolysaccharide-binding protein-bactericidal/permeability-increasing protein fusion peptide in an experimental model of Pseudomonas sepsis.

A chimeric protein consisting of the N-terminal domain of lipopolysaccharide-binding protein and the C-terminal domain of bactericidal/permeability-increasing protein demonstrated a dose-dependent survival benefit (P = 0.001) and reduced endotoxin levels (P < 0.01) in neutropenic rats with Pseudomonas aeruginosa sepsis. This lipopolysaccharide-binding protein-bactericidal/ permeability-increasing peptide has favorable pharmacokinetics and antiendotoxin properties which may be of value for human sepsis.

Affinity-purified Escherichia coli J5 lipopolysaccharide-specific IgG protects neutropenic rats against gram-negative bacterial sepsis.

Antibodies were raised in rabbits by immunization with the heat-killed J5 mutant of Escherichia coli O111 (Rc chemotype). Serum antibodies were separated into purified IgG and IgM by sequential affinity chromatography on protein G-Sepharose and anti-rabbit IgG-Sepharose columns. J5 lipopolysaccharide (LPS)-specific IgG was prepared by affinity chromatography of purified IgG on a J5 LPS-EAH Sepharose 4B affinity column. Purified IgM, IgG, and J5 LPS-specific IgG protected neutropenic rats against lethal challenge with Pseudomonas aeruginosa 12:4:4 (Fisher Devlin immunotype 6). Nine of 16 rats treated with the IgM fraction were protected (P < .001). Thirteen of 20 rats treated with the purified IgG and 6 of 8 treated with J5 LPS-specific IgG were protected compared with none of 25 treated with IgG made from the preimmune serum of the same rabbit (P < .001). These results demonstrate that purified J5 LPS-specific IgG protects against the lethal consequences of gram-negative bacteremia.

Relative concentrations of endotoxin-binding proteins in body fluids during infection.

Endotoxin initiates the systemic inflammatory response, haemodynamic changes, and multi-organ failure that may occur as a consequence of systemic gram-negative bacterial infection. The serum protein lipopolysaccharide-binding protein (LBP) binds to the lipid A component of bacterial endotoxin and facilitates its delivery to the CD14 antigen on the macrophage, where inflammatory cytokines are released and a cascade of host mediators is initiated. The neutrophil granular protein bactericidal/permeability-increasing protein (BPI) competes with LBP for endotoxin binding and functions as a molecular antagonist of LBP-endotoxin interactions. We have measured concentrations of both proteins in body fluids from 49 consecutive patients. In 16 of 17 samples of fluid from closed-space infections, BPI was present in greater concentration than LBP (median BPI/LBP ratio 7.6 [95% CI 2.32-22.1]). The ratio of BPI and LBP was not significantly different from 1.0 in abdominal fluid from 10 patients with peritonitis (ratio 0.235 [0.18-0.47]), whereas the BPI/LBP ratio was low in 22 non-infected body fluids (0.01 [0.001-0.04]) and concentrations of both proteins approached those in normal human plasma. BPI concentrations were directly correlated with the quantity of neutrophils within clinical samples (rs = 0.81, p < 0.0001). Thus, within abscess cavities BPI is available in sufficient quantities for effective competition with LBP for endotoxin. BPI may attenuate the local inflammatory response and the systemic toxicity of endotoxin release during gram-negative infections.

Human neutrophil bactericidal/permeability-increasing protein reduces mortality rate from endotoxin challenge: a placebo-controlled study.

OBJECTIVES: To study the toxicology and pharmacology of the endotoxin-neutralizing agent, bactericidal/permeability-increasing protein. DESIGN: Prospective, randomized, placebo-controlled laboratory study. SETTING: Academic research laboratory. SUBJECTS: CD-1 mice (n = 259); Sprague Dawley rats (n = 26); New Zealand White rabbits (n = 19). INTERVENTIONS: Pharmacokinetics of intravenously injected bactericidal/permeability-increasing protein was assessed in mice. Toxicology was tested in mice and rats. Efficacy of intravenously administered bactericidal/permeability-increasing protein as an endotoxin-neutralizing agent was tested in mice, rats, and rabbits. MEASUREMENTS AND MAIN RESULTS: Administration of a single 10-mg/kg bolus injection of bactericidal/permeability-increasing protein resulted in no alterations in hematologic, renal, or hepatic function, activity level, or weight gain in animals observed over a 7-day study period. A single bolus injection (10 mg/kg) of bactericidal/permeability-increasing protein protected 15 of 16 mice from a lethal endotoxin challenge (mortality rate 1/16 [6.25%]) compared with a 100% (16/16) mortality rate in the saline-treated controls (p < .001). Bactericidal/permeability-increasing protein administered up to 1 hr after endotoxin provided significant protection against lethal endotoxin challenge. Furthermore, bactericidal/permeability-increasing protein reduced the induration and dermal necrosis observed in the localized dermal Shwartzman reaction. CONCLUSIONS: Bactericidal/permeability-increasing protein is a potent antiendotoxin that neutralizes endotoxin in vivo and prevents mortality in animal models of lethal endotoxemia.

Differential effect of isotype on efficacy of anti-tumor necrosis factor alpha chimeric antibodies in experimental septic shock.

Immune complexes containing human gamma (g)1 or murine g2a antibodies generate secondary effector mechanisms via Fc receptor binding or complement activation, whereas those containing human g4 or murine g1 antibodies generally do not. Therefore, isotype selection of therapeutic antibodies may have important clinical consequences. In a rabbit model of human tumor necrosis factor (rhuTNF)-induced pyrexia, a murine/human chimeric g4 anti-human TNF-alpha monoclonal antibody (mAb) (cCB0011) showed a dose-dependent inhibition of pyrexia, whereas a g1 isotype variant of the same mAb gave a marked pyrexia that was seen at all doses indicative of an immune complex-mediated response. To investigate whether isotype difference could influence mAb efficacy in pathological disease states, hamster/murine chimeric g1 and g2a anti-murine TNF-alpha mAbs (TN3g1, TN3g2a) were studied in experimental shock in mice and rats. In lipopolysaccharide-induced shock in mice, treatment with TN3g1 mAb at 30 and 3 mg/kg resulted in 90% survival by 72 h (p < or = 0.004), and prolonged survival to 45 h (p < or = 0.05), respectively, compared with 100% mortality by 27 h in controls. In contrast, a g2a isotype variant of the same mAb (30 mg/kg) resulted in only 10% survival by 72 h (p < or = 0.05). In a neutropenic sepsis model in rats there was greater survival in animals receiving the g1 isotype of TN3 compared with g2a isotype variant (70 vs. 27%; p < or = 0.005) with 100% mortality in the controls. These differences were not due to the pharmacokinetic profiles of the mAbs. In models of experimental shock antibody isotype can affect outcome with inactive isotypes (human g4 and murine g1) being more efficacious than active isotypes (human g1 and murine g2a).

The efficacy of combination immunotherapy in experimental Pseudomonas sepsis.

Immunotherapy has been shown to be an effective adjuvant in the management of septic shock. A neutropenic rat model of septic shock induced by infection with Pseudomonas aeruginosa 12.4.4 (Fisher immunotype 6) was used to determine the relative efficacy of single, double, and triple combination immunotherapy. A Pseudomonas O serotype-specific, opsonophagocytic monoclonal antibody (MAb), polyclonal J5 antiserum, and a MAb directed against tumor necrosis factor-alpha (TNF) were studied as single therapy and in combination. The combination of all three immunotherapeutic agents resulted in a 77% survival rate (33/43 animals). This level of protection was superior to that achieved with any combination of two antibody treatments (50%-60% survival; P = .029) or single antibody therapy (25%-43% survival; P < .001) or compared with a control group (0/25 survivors; P < .0001). Immunotherapy directed against multiple steps of the septic process is more active than single or double antibody regimens and may offer an improved approach to the adjunctive treatment of septic shock.

Efficacy of anti-endotoxin monoclonal antibody E5 alone or in combination with ciprofloxacin in neutropenic rats with Pseudomonas sepsis.

Pathogen-free rats were rendered neutropenic, given oral feedings of Pseudomonas aeruginosa 12.4.4, then monitored for fever. At the onset of fever, rats were given intravenous treatment with either anti-endotoxin monoclonal antibody (MAb) E5 or control MAb B55. Survival was significantly greater in E5- than in B55-treated animals (P < .01). Serum levels of both lipopolysaccharide and tumor necrosis factor-alpha were significantly reduced in E5- versus B55-treated rats 24 h after treatment (P < .01 and < .05, respectively). Rats were also treated with E5 or B55 in combination with a suboptimal dose of ciprofloxacin at fever onset and again 24 h later. Survival was significantly greater in ciprofloxacin-treated animals given E5 than in animals given B55 (P < .005). Posttreatment endotoxin levels were decreased in animals receiving E5 in combination with ciprofloxacin (P < .001) compared with B55-treated animals. These results indicate that therapy with anti-endotoxin MAb E5 alone or in combination with antimicrobial therapy improves survival in this bacteremic infection model of Pseudomonas sepsis.

Efficacy of antilipopolysaccharide and anti-tumor necrosis factor monoclonal antibodies in a neutropenic rat model of Pseudomonas sepsis.

Monoclonal antibodies (MAb) directed against bacterial lipopolysaccharide (LPS) and tumor necrosis factor-alpha (TNF) provide partial protection in experimental models of septic shock. To determine if additional benefit accrues from a combination of anti-TNF and anti-LPS MAb in the treatment of septic shock, a neutropenic rat model was developed to study active infection with Pseudomonas aeruginosa 12.4.4. Animals were treated intravenously with an irrelevant MAb (group 1); anti-TNF MAb (group 2); MAb directed against P. aeruginosa 12.4.4 LPS (group 3); or a combination of anti-TNF and anti-LPS MAb (group 4). None of the control animals in group 1 survived the 7-d period of neutropenia (0/16). In contrast, the survival rate was 44% in group 2 (P less than 0.02); 37% in group 3 (P less than 0.05); and 75% in group 4 (P less than 0.0002). The combination of monoclonal antibodies provided greater protection than either MAb given alone (P less than 0.05). Serum TNF levels during infection were significantly greater in groups 1 and 3 (20.1 +/- 3.3 U, mean +/- SE) than in groups 2 and 4 (0.9 +/- 0.8 U, P less than 0.0001). These results indicate that a combination of monoclonal antibodies to LPS and TNF have additive benefit in experimental Pseudomonas aeruginosa sepsis. This immunotherapeutic approach may be of potential utility in the management of serious, gram-negative bacterial infection in neutropenic patients.

Efficacy of a monoclonal antibody directed against tumor necrosis factor in protecting neutropenic rats from lethal infection with Pseudomonas aeruginosa.

A monoclonal antibody directed against murine tumor necrosis factor-alpha (TNF) was studied in a neutropenic rat model to determine its efficacy in protecting animals from lethal infection with Pseudomonas aeruginosa. Anti-TNF monoclonal antibody at a dose of 20 mg/kg given intravenously at 0 and 120 h resulted in a 53% survival rate (8/15) compared with no survival in control animals (0/15) (P less than .005). The combination of anti-TNF monoclonal antibody and oral ciprofloxacin at a suboptimal dose of 2.5 mg/kg/day resulted in a 100% survival rate in neutropenic animals (16/16), while ciprofloxacin alone produced only a 67% survival rate (10/15) during the 7-day period of neutropenia (P less than .05). Thus anti-TNF monoclonal antibody alone or in addition to antimicrobial agents improved survival in neutropenic animals after infection with P. aeruginosa.