The effect of emergency department delay on outcome in critically ill medical patients: evaluation using hospital mortality and quality of life at 6 months.

OBJECTIVE: To assess the impact of delay in emergency department (ED) on outcome of critically ill patients admitted to the medical intensive care unit (MICU). Outcome was defined as hospital mortality and as health-related quality of life (HRQoL) at 6 months after intensive care assessed by the 15D measure. The 15D is a generic, 15-dimensional, standardized measure of HRQoL. We hypothesized that prolonged stay in the ED is related to worse outcome. DESIGN AND SETTING: A prospective follow-up cohort study in university hospital. SUBJECTS: All consecutive 1675 patients admitted to the MICU between July 2002 and June 2004. RESULTS: The 15D questionnaire was mailed to all patients alive at 6 months after admission. Of all MICU patients, 64% were admitted from ED. The mean length of stay in the ED was 6.2 h (95%CI 5.9-6.5 h). The hospital mortality rate was 24.4% (20.0% in the ED vs. 33.0% in the non-ED cohort, P < 0.001) and it was associated with higher age and degree of physiological derangement at admission. Neither the length of ED stay was associated with hospital mortality (P = 0.82) nor with HRQoL at 6 months after MICU admission (P = 0.34). Altogether, HRQoL at 6 months was significantly lower compared with the age- and sex-matched general population (P < 0.001). CONCLUSIONS: In a university hospital, the length of ED stay was not associated with the outcome of critically ill medical patients. However, we feel that the effect of ED treatment and delay on outcome and outcome prediction in the critically ill patients deserves further evaluation.

Expression of cyclooxygenase-2 in dysplasia of the stomach and in intestinal-type gastric adenocarcinoma.

PURPOSE: Cyclooxygenase (Cox) is the key enzyme in conversion of arachidonic acid to prostanoids. Two Cox genes have been cloned, and expression of Cox-2 mRNA and protein has been shown to be elevated in several human malignancies and in animal models of carcinogenesis. The purpose of this study was to investigate Cox-2 protein expression in human gastric dysplasias and adenocarcinomas. EXPERIMENTAL DESIGN: Performance of several Cox-2 antibodies was evaluated, after which Cox-2 protein expression was studied in 67 gastric cancer specimens and in eight definitive dysplasias by using immunohistochemistry. RESULTS: Cox-2 positivity was detected in 58% (25/43) of the intestinal-type (well-differentiated) tumors and 6% (1/18) of diffuse-type (poorly differentiated) tumors. Consistent with these data, we detected higher expression of Cox-2 mRNA, protein, and enzymatic activity in well-differentiated gastric cancer cell lines (MKN-28 and MKN-74) when compared with poorly differentiated cell lines (HSC-39 and KATO III). Cox-2 immunoreactivity was localized to the carcinoma cells, but the stroma of the tumors was negative. However, strong Cox-2 positivity was consistently detected in stromal cells at sites of erosions and ulcerations. Furthermore, four of nine (44%) definitive dysplasias of the stomach that showed no evidence of invasion were positive for Cox-2. CONCLUSIONS: Cox-2 is expressed by the neoplastic cells in the intestinal-type gastric adenocarcinoma and by precarcinogenic (dysplastic) lesions leading to this disease.

Regulated expression of prostaglandin E(2) receptors EP2 and EP4 in human ovarian granulosa-luteal cells.

Prostaglandins (PGs) have been implicated in regulation of ovarian function. We have previously shown that the expression of cyclooxygenase-2 and the receptor for PGF(2 alpha) are expressed in periovulatory human granulosa cells and upregulated by gonadotropins and cytokines in cultured human ovarian granulosa-luteal (GL) cells. We now show that transcripts for PGE(2) receptor subtypes EP2 and EP4 are expressed in freshly isolated human granulosa cells and in mouse ovaries as detected by Northern blot analysis. However, EP2 and EP4 receptor mRNA levels were low or nondetectable in cultured human GL cells suggesting that these transcripts may be under hormonal and/or cytokine regulation in the ovaries in vivo. Indeed, the proinflammatory cytokine interleukin-1 beta (IL-1 beta) stimulated expression of EP2 and EP4 transcripts in concentration- and time-dependent manner in the GL cells. Furthermore, the transcript for EP2 receptor was localized in the corpus luteum of the mouse ovary by in situ hybridization, and EP2 protein was expressed in human corpus luteum as detected by immunohistochemistry. Our data suggest that IL-1 beta induces expression of EP2 and EP4 receptors in human GL cells, and that EP2 receptor is expressed in both human and murine luteal glands.

Expression of cyclooxygenase-2 in human transitional cell carcinoma of the urinary bladder.

Recent studies suggest that expression of cyclooxygenase-2 (Cox-2) is elevated in transitional cell carcinoma (TCC) of the urinary bladder and that inhibition of Cox-2 activity suppresses bladder cancer in experimental animal models. We have investigated the expression of Cox-2 protein in human TCCs (n = 85), in in situ carcinomas (Tis) of the urinary bladder (n = 17), and in nonneoplastic urinary bladder samples (n = 16) using immunohistochemistry. Cox-2 immunoreactivity was detected in 66% (67 of 102) of the carcinomas, whereas only 25% (4 of 16) of the nonneoplastic samples were positive (P: < 0.005). Cox-2 immunoreactivity localized to neoplastic cells in the carcinoma samples. The rate of positivity was the same in invasive (T1-3; 70%, n = 40) and in noninvasive (Tis and Ta; 65%, n = 62) carcinomas, but noninvasive tumors had a higher frequency (32%) of homogenous pattern of staining (>90% of the tumor cells positive) than the invasive carcinomas (10%) (P: < 0.05). However, several invasive TCCs exhibited the strongest intensity of Cox-2 staining in the invading cells, whereas other parts of the tumor were virtually negative. Finally, strong Cox-2 positivity was also found in nonneoplastic ulcerations (2 of 2) and in inflammatory pseudotumors (2 of 2), in which the immunoreactivity localized to the nonepithelial cells. Taken together, our data suggest that Cox-2 is highly expressed in noninvasive bladder carcinomas, whereas the highest expression of invasive tumors associated with the invading cells, and that Cox-2 may also have a pathophysiological role in nonneoplastic conditions of the urinary bladder, such as ulcerations and inflammatory pseudotumors.

Expression of cyclooxygenase-2 and prostanoid receptors by human myometrium.

Prostanoids play an important role in the regulation of parturition. All reproductive tissues, including fetal membranes, decidua, and myometrium, have the capacity to synthesize prostanoids, and fetal membranes have been shown to express elevated levels of cyclooxygenase-2 (Cox-2) at the onset of labor. We have now investigated the expression of Cox-2 in human myometrium. Myometrial samples collected from women in labor during lower segment cesarean section expressed 15-fold higher levels of Cox-2 messenger ribonucleic acid (mRNA) compared to myometrial specimens collected from women not in labor, as detected by Northern blot analysis. Immunohistochemical detection of Cox-2 protein showed cytoplasmic staining in the smooth muscle cells of the myometrium. Cultured myometrial cells expressed low levels of Cox-2 mRNA under baseline conditions, but interleukin-1beta (IL-1beta) caused a 17-fold induction of expression of the Cox-2 transcript after incubation for 6 h. IL-1beta also induced expression of biologically active Cox-2 protein, as detected by immunofluorescence, Western blot analysis, and measuring the conversion of arachidonic acid to prostanoids in the presence and absence of a Cox-2-selective inhibitor, NS-398. PGE2 receptor subtype EP2 mRNA was expressed in cultured myometrial smooth muscle cells, whereas transcripts for EP1, EP3, EP4, FP, and IP were low or below the detection limit as measured by Northern blot analysis. However, IL-1beta stimulated expression of EP4 receptor mRNA. Our data suggest that expression of Cox-2 transcript is elevated at the onset of labor in myometrial smooth muscle cells, which may depend on induction by cytokines. As, in addition to Cox-2, the expression of prostanoid receptors is regulated, not only the production of prostanoids, but also responsiveness to them, may be modulated.

Expression of cyclooxygenase-2 in human lung carcinoma.

Epidemiological studies indicate that the use of aspirin decreases incidence of and mortality from gastrointestinal cancers. A major target of aspirin and other nonsteroid anti-inflammatory drugs is cyclooxygenase (Cox), the rate-limiting enzyme in the conversion of arachidonic acid to prostanoids. Two Cox genes have been cloned (Cox-1 and Cox-2), of which Cox-2 has recently been found to be expressed in several human carcinomas. We have now studied the expression of Cox-2 mRNA and protein in human lung adenocarcinoma, squamous cell carcinoma, and small cell lung cancer. Cox-2 mRNA steady-state levels were high in well-differentiated adenocarcinoma samples, but low in poorly differentiated adenocarcinoma, squamous cell carcinoma, and small cell lung cancer, as detected by Northern blot analysis. Immunohistochemistry showed Cox-2 staining in 19 of 21 adenocarcinomas. However, well-differentiated adenocarcinomas contained more Cox-2 staining than the poorly differentiated ones. Expression of the Cox-2 protein was also seen in all 11 squamous cell carcinomas studied, although the level of staining seemed to be less than that in the adenocarcinomas. Small cell lung cancer specimens (n = 4) stained with a relatively weak intensity. Interestingly, atypical alveolar epithelium, which associates with asbestosis and idiopathic fibrosing alveolitis and is considered to be a precursor lesion for lung cancer, expressed the Cox-2 protein. Our data, thus, suggest that Cox-2 is expressed in human lung carcinomas and in precursor lesions leading to this malignancy.

Aerosolized pentamidine as alternative primary prophylaxis against Pneumocystis carinii pneumonia in adult hepatic and renal transplant recipients.

STUDY OBJECTIVE: To examine the safety and efficacy of aerosolized pentamidine (AP) as alternative primary prophylaxis against Pneumocystis carinii pneumonia (PCP) in adult liver and kidney transplant recipients. DESIGN: Retrospective review of medical records. SETTING: Tertiary care urban teaching hospital with active liver and kidney transplant programs. PATIENTS: Adult liver and kidney transplant recipients intolerant of trimethoprim-sulfamethoxazole (TMP-SMX) therapy and referred to the AP clinic between June 1991 and December 1994. INTERVENTIONS: Each patient received monthly AP, 300 mg, delivered by a nebulizer (Respirgard-II), preceded by inhaled albuterol, 180 micrograms. During the period of follow-up, information related to side effects of AP and incidence of PCP was recorded. RESULTS: A total of 35 patients were identified, 18 liver and 17 kidney transplant recipients. Fourteen patients received AP as initial prophylaxis because of prior sensitivity to TMP-SMX. In another 19 patients, initial TMP-SMX therapy was discontinued for leukopenia (5), elevated liver function test values (4), rash (3), nausea (2), renal failure (2), seizure (2), and thrombocytopenia (1). In addition, two patients received AP in the setting of organ rejection. Liver transplant recipients received AP for an average of 4.28 +/- 1.6 months, and renal transplant recipients received AP for an average of 5.71 +/- 4.3 months. Adverse effects of AP included bronchospasm (two), dyspnea (one), cough (one), and nausea (one). AP therapy was discontinued in only one patient due to severe bronchospasm. There were no cases of PCP in the 35 patients receiving AP. CONCLUSIONS: These observations suggest that AP is well tolerated and may be an effective alternative for PCP prophylaxis in adult liver and kidney transplant recipients intolerant to TMP-SMX therapy.

Pertussis toxin has eukaryotic-like carbohydrate recognition domains.

Bordetella pertussis is bound to glycoconjugates on human cilia and macrophages by multiple adhesins, including pertussis toxin. The cellular recognition properties of the B oligomer of pertussis toxin were characterized and the location and structural requirements of the recognition domains were identified by site-directed mutagenesis of recombinant pertussis toxin subunits. Differential recognition of cilia and macrophages, respectively, was localized to subunits S2 and S3 of the B oligomer. Despite greater than 80% sequence homology between these subunits, ciliary lactosylceramide exclusively recognized S2 and leukocytic gangliosides bound only S3. Substitution at residue 44, 45, 50, or 51 in S2 resulted in a shift of carbohydrate recognition from lactosylceramide to gangliosides. Mutational exchange of amino acid residues 37-52 between S2 and S3 interchanged their carbohydrate and target cell specificity. Comparison of these carbohydrate recognition sequences to those of plant and animal lectins revealed that regions essential for function of the prokaryotic lectins were strongly related to a subset of eukaryotic carbohydrate recognition domains of the C type.

Integrin-mediated localization of Bordetella pertussis within macrophages: role in pulmonary colonization.

The adherence of Bordetella pertussis to human respiratory cilia is critical to the pathogenesis of whooping cough but the significance of bacterial attachment to macrophages has not been determined. Adherence to cilia and macrophages is mediated by two large, nonfimbrial bacterial proteins, filamentous hemagglutinin (FHA), and pertussis toxin (PT). PT and FHA both recognize carbohydrates on cilia and macrophages; FHA also contains an Arg-Gly-Asp (RGD) sequence which promotes bacterial association with the macrophage integrin complement receptor 3 (CR3). We determined that virulent B. pertussis enter and survive in mammalian macrophages in vitro and that CR3 is important for this uptake process. We then determined the relative contribution of CR3 versus carbohydrate-dependent interactions to in vivo pulmonary colonization using a rabbit model. B. pertussis colonized the lung as two approximately equal populations, one extracellular population attached to ciliary and macrophage surface glycoconjugates and another population within pulmonary macrophages. Loss of the CR3 interaction, either by mutation of FHA or treatment with antibody to CR3, disrupted accumulation of viable intracellular bacteria but did not prevent lung pathology. In contrast, elimination of carbohydrate-bound bacteria, either by a competitive receptor analogue or an anti-receptor antibody, was sufficient to prevent pulmonary edema. We propose that CR3-dependent localization of B. pertussis within macrophages promotes persistence of bacteria in the lung without pulmonary injury. On the other hand, the presence of extracellular bacteria adherent to cilia and macrophages in carbohydrate-dependent interactions is associated with pulmonary pathology.

Recognition of a bacterial adhesion by an integrin: macrophage CR3 (alpha M beta 2, CD11b/CD18) binds filamentous hemagglutinin of Bordetella pertussis.

During the course of whooping cough, Bordetella pertussis interacts with alveolar macrophages and other leukocytes on the respiratory epithelium. We report here mechanisms by which these bacteria adhere to human macrophages in vitro. Whole bacteria adhere by means of two proteins, filamentous hemagglutinin (FHA) and pertussis toxin, either of which is sufficient to mediate adherence. FHA interacts with two classes of molecules on macrophages, galactose-containing glycoconjugates and the integrin CR3 (alpha M beta 2, CD11b/CD18). The interaction between CR3 and FHA involves recognition of the Arg-Gly-Asp (RGD) sequence at positions 1097-1099 in FHA. This study demonstrates that bacterial adherence can be based on the interaction of a bacterial adhesin RGD sequence with an integrin and that bacterial adhesins can have multiple binding sites characteristic of eukaryotic extracellular matrix proteins.

The role of cytokines in the generation of inflammation and tissue damage in experimental gram-positive meningitis.

Cytokines mediate many host responses to bacterial infections. We determined the inflammatory activities of five cytokines in the central nervous system: TNF-alpha, IL-1 alpha, IL-1 beta, macrophage inflammatory protein 1 (MIP-1), and macrophage inflammatory protein 2 (MIP-2). Using a rabbit model of meningeal inflammation, each cytokine (except IL-1 beta) induced enhanced blood brain barrier permeability, leukocytosis in cerebrospinal fluid, and brain edema. Homologous antibodies to each mediator inhibited leukocytosis and brain edema, and moderately decreased blood brain barrier permeability. In rabbits treated with anti-CD-18 antibody to render neutrophils dysfunctional for adhesion, each cytokine studied lost the ability to cause leukocytosis and brain edema. After intracisternal challenge with pneumococci, antibodies to TNF or IL-1 prevented inflammation, while anti-MIP-1 or anti-MIP-2 caused only a 2-h delay in the onset of inflammation. We suggest these cytokines have multiple inflammatory activities in the central nervous system and contribute to tissue damage during pneumococcal meningitis.

Reduction of inflammation, tissue damage, and mortality in bacterial meningitis in rabbits treated with monoclonal antibodies against adhesion-promoting receptors of leukocytes.

We tested if specific inhibition of recruitment of leukocytes across the blood brain barrier from the vascular compartment to the cerebrospinal fluid (CSF) space reduced tissue damage and improved the outcome of infection in a rabbit model of experimental meningitis. The CD11/CD18 complex of receptors on leukocytes promotes adhesion of these cells to endothelia, a process required for egress of cells into the extravascular space. Intravenous injection of the anti-CD18 mAb IB4 effectively blocked the development of leukocytosis in the CSF of animals challenged intracisternally with living bacteria, bacterial endotoxin, or bacterial cell wall. This effect was associated with protection from blood brain barrier injury as measured by exclusion of serum proteins from CSF in mAb-treated animals. The densities of bacteria in CSF and the degrees of bacterial killing due to ampicillin were not affected by the antibody. Animals receiving the antibody experienced a delay in the development of bacteremia and a significantly reduced inflammatory response during ampicillin-induced bacterial killing. Therapy with mAb IB4 prevented development of brain edema and death in animals challenged with lethal doses of Streptococcus pneumoniae. These studies indicate that the major mechanism of leukocyte migration across the blood brain barrier involves the CD11/CD18 receptors and that inflammatory leukocytes recruited by this mechanism are a major cause of blood brain barrier injury and cerebral edema during meningitis.

Comparative evaluation of potential components for group B meningococcal vaccine by passive protection in the infant rat and in vitro bactericidal assay.

Seventeen monoclonal antibodies to one of three main cell surface antigens of Neisseria meningitidis group B were tested for protective efficacy in the infant rat using as challenge seven strains of different class 2/3 protein serotypes, class 1 protein (P1) subtypes and LPS immunotypes. Type-specific protection indicated both by a reduction of bacteraemia and meningitis and survival of the animals was regularly obtained with antibodies to the P1 protein and to LPS. By contrast, only one of seven antibodies to the serotype-specific class 2/3 protein was protective, even though four of them were highly bactericidal. The animal protection test and in vitro bactericidal assay were otherwise concordant. These data form important guidelines for the design of vaccines to prevent group B meningococcal infections.

Beneficial effects of a specific leukotriene receptor antagonist in splanchnic artery occlusion shock.

The purpose of this study was to examine the effects of a new potent peptidoleukotriene receptor antagonist, SK&F 104353, in splanchnic artery occlusion shock. SK&F 104353 was administered as a 1 mg/kg initial bolus followed by an infusion of 3 mg/kg per h for the entire 2 h post-reperfusion observation period. In a group of conscious rats, this dose of SK&F 104353 shifted the LTD4 dose response curve rightward 10-fold, indicating effective antagonism of peptidoleukotriene responses in the rat. Anesthetized rats subjected to splanchnic artery occlusion shock survived an average of only 98 +/- 8 min whereas all animals receiving SK&F 104353 survived the 2 h reperfusion period (P less than 0.02 from vehicle). Therefore, the survival rate of the splanchnic artery occlusion shock group of rats receiving SK&F 104353 was improved to 100% compared with 50% survival for the vehicle-treated splanchnic artery occlusion shock group (P less than 0.025). In the splanchnic artery occlusion shock + SK&F 104353 group the increase in the plasma activities of the lysosomal hydrolase, cathepsin D, and the cardiotoxic peptide, myocardial depressant factor, were significantly attenuated in comparison to the splanchnic artery occlusion shock + vehicle group (P less than 0.025). These data indicate that the peptidoleukotriene receptor antagonist, SK&F 104353 is beneficial in splanchnic artery occlusion shock, and furthermore suggests that it may be a therapeutically useful agent in bowel ischemic shock.

Role of type 1 and S fimbriae in the pathogenesis of Escherichia coli O18:K1 bacteremia and meningitis in the infant rat.

The role of fimbriae in the pathogenesis of Escherichia coli infection was studied in the infant rat model. Rat pups were challenged intraperitoneally at the age of 5 days with E. coli K1 (strain IH3080, O18:K1:H7) and three different subpopulations (type 1, type S, or nonfimbriated) of it. All bacterial subpopulations were able to produce peritonitis, bacteremia, and meningitis. However, the type 1 fraction was the least virulent and the type S fraction was the most virulent, as judged by the bacterial counts in body fluids and by the mortality rates of the pups. Fimbrial phase variation to mainly the type-S-fimbriated forms was observed in all body fluids. An initially type-S-fimbriated inoculum remained predominantly type S fimbriated in the peritoneal fluid and blood. In the cerebrospinal fluid, however, about 50% of the bacteria were type S fimbriated and 50% were nonfimbriated 1 h after challenge with the type-S-fimbriated subpopulation; at later times the share of type-S-fimbriated bacteria also increased in the cerebrospinal fluid.

Experimental meningococcal meningitis in the infant rat.

Experimental infection of the infant rat could be established with intraperitoneal challenge with strains of all three meningococcal groups A, B, and C tested. 5-day-old rats were challenged with 3 different doses (10(2), 10(4), and 10(6) bacteria/animal) and the development of peritonitis, bacteraemia and meningitis detected as a function of time at 3, 6, and 9 h. Mortality was followed up to 24 h. Group B strains caused a rapidly developing and sustained high level bacteraemia and meningitis with all challenge doses. Bacteraemia and meningitis following challenge with MenA and MenC were of somewhat shorter duration and reached lower peak levels. Repeated 'rat passages' of meningococcal strains that had been kept stored for long periods markedly increased their virulence. This study shows that the infant rat model can be applied for studying pathogenesis of meningococcal bacteraemia and meningitis. Previous work from this laboratory has shown it to be suited also for studying antibody-mediated protection from this disease.

Protective efficacy of monoclonal antibodies to class 1 and class 3 outer membrane proteins of Neisseria meningitidis B:15:P1.16 in infant rat infection model: new prospects for vaccine development.

The protective efficacy of monoclonal antibodies to class 1 and class 3 outer membrane proteins of Neisseria meningitidis B:15:P1.16 was tested in an infant rat infection model. Four monoclonal antibodies to class 1 protein had bactericidal titres exceeding 20,000 and they protected infant rats completely against bacterial challenge with meningococci carrying the same class 1 protein, P1.16. One monoclonal antibody to class 3 protein was highly bactericidal (titer greater than 20,000), whereas two others had no bactericidal activity. All these antibodies gave some protection from infection, resulting in mortalities varying from 66 to 83% as compared to 100% in control rats who had received either unrelated monoclonal antibody or saline. These results strongly speak for class 1 outer membrane protein as a vaccine candidate for meningococcus group B.