Risk factors for early hospital readmission after cardiac operations.

OBJECTIVE: Early hospital readmissions after cardiac procedures are both costly and harmful to patients. We investigated the factors that predispose to readmission to develop strategies to minimize this problem. METHODS: As part of a prospective data collection, patients having cardiac procedures at our institution are routinely tracked for 30 days after their discharge from the hospital. We reviewed 2650 patients in our cardiac database who underwent operations over the past 5 years. We used univariate and multivariate statistical techniques to identify risks for readmission. RESULTS: Of 2574 discharged patients, 252 (9.8%) required readmission. The most common causes of readmission are cardiac (42%), pulmonary (19%), gastrointestinal (10%), extremity complications (6.7%; deep vein thrombophlebitis, peripheral arterial vascular disease, and saphenous vein harvest site problems), sternal wound problems (7.5%), and metabolic problems (4%). Of more than 70 variables studied, only 6 are significant multivariate predictors of readmission: female sex (P =.002); diabetes (P =.001); chronic lung problems (P =.011); increased distance between home and hospital (P >.001); preoperative atrial fibrillation (P =.002); and preoperative chronic renal insufficiency (P =.002). Type of operation, redo procedures, and other intraoperative and postoperative variables are not important multivariate predictors of readmission. Prolonged hospital length of stay for the initial procedure did not cause more frequent readmission. The costs of initial hospitalization (operating room costs combined with postoperative in-hospital costs) were not significantly increased in those patients who required readmission. CONCLUSIONS: The high-risk patient for readmission is a woman with diabetes, chronic lung disease, renal insufficiency, and preoperative atrial fibrillation who lives at a distance from the hospital. Readmission does not depend on periprocedural variables (eg, cardiopulmonary bypass time) or on postoperative complications. High procedural costs from the initial hospitalization do not predispose to readmission. These results suggest interventions that may reduce readmission.

2-Amino-5-chlorophenol toxicity in renal cortical slices from Fischer 344 rats: effect of antioxidants and sulfhydryl agents.

2-Amino-5-chlorophenol is nephrotoxic through an unidentified mechanism. This study examined the in vitro toxicity of 2-amino-5-chlorophenol in renal cortical slices from Fischer 344 rats and specifically assessed induction of lipid peroxidation and depletion of renal glutathione. Renal cortical slices exposed to 0, 0.25, 0.5, and 1 mM 2-amino-5-chlorophenol exhibited a concentration- and time-dependent increase in lactate dehydrogenase (LDH) leakage. Pyruvate-directed gluconeogenesis was diminished in a concentration-dependent manner following a 90-min incubation with 0, 0.25, 0.5, and 1 mM 2-amino-5-chlorophenol. Lipid peroxidation was induced within 60 min by 1 mM 2-amino-5-chlorophenol in renal slices relative to control tissue. Total glutathione (GSH) levels were decreased below control values within 30 min of exposure to 0.5 and 1 mM 2-amino-5-chlorophenol. These results indicated that GSH levels were decreased prior to the appearance of increased LDH leakage and diminished membrane integrity. 2-Amino-5-chlorophenol toxicity was increased in renal slices isolated from animals pretreated with buthionine sulfoximine (BSO, 890 mg/kg ip). Pretreatment of renal slices with the phenolic antioxidant N,N'-diphenyl-1, 4-phenylenediamine (DPPD, 50 microM) or the iron chelator deferoxamine did not reduce 2-amino-5-chlorophenol cytotoxicity. These results suggest that 2-amino-5-chlorophenol toxicity was not mediated through an iron-dependent mechanism. 2-Amino-5-chlorophenol cytotoxicity was reduced by a 15-min pre-incubation with 2 mM ascorbate or a 30-min preincubation with the thiol-containing agents GSH (1 mM) or dithiothreitol (1 mM, DTT). Pretreatment with GSH, DTT, or ascorbate reduced LDH leakage and lipid peroxide generation induced by 2-amino-5-chlorophenol. These results suggest that 2-amino-5-chlorophenol cytotoxicity involved free radical generation through an iron-independent mechanism. Toxicity was reduced by the presence of the antioxidant ascorbate or by addition of glutathione.

Cephaloridine in vitro toxicity and accumulation in renal slices from normoglycemic and diabetic rats.

Previous work has shown a reduction in cephaloridine nephrotoxicity in a diabetic rat model. The following studies examined in vitro cephaloridine toxicity in renal slices from normoglycemic and diabetic Fischer 344 rats. Diabetes was induced by acute intraperitoneal injection of 35 mg/kg streptozotocin. Renal cortical slices were isolated from normoglycemic and diabetic animals. Tissues were exposed to 0-5 mM cephaloridine for 15-120 min. Pyruvate-directed gluconeogenesis was diminished in all groups exposed to 2-5 mM cephaloridine for 60-120 min. Leakage of lactate dehydrogenase (LDH) was apparent only in the normoglycemic group in the presence of 4-5 mM cephaloridine for 120 min. LDH leakage was not increased at any cephaloridine concentration in the diabetic tissue. Total glutathione levels were compared in renal cortical slices exposed to cephaloridine for 30-120 min. Baseline values for glutathione were comparable between normoglycemic and diabetic tissue suggesting that the mechanism for reduced toxicity was not due to higher glutathione levels in diabetic tissue. Total glutathione levels were diminished more rapidly in normoglycemic than diabetic tissue by incubation with 5 mM cephaloridine. Comparison of cephaloridine accumulation indicated that diabetic tissue accumulated less cephaloridine than the normoglycemic group when tissues were incubated with 0-2 mM cephaloridine. However, renal slice accumulation was similar between normoglycemic and diabetic groups following in vitro incubation with 4-5 mM cephaloridine. These results suggest that the mechanism for reduced in vitro cephaloridine toxicity in diabetic tissue cannot be limited to differences in accumulation and must include an unidentified cellular component.

In Vitro toxicity of 2- and 4-chloroaniline: comparisons with 4-amino-3-chlorophenol, 2-amino-5-chlorophenol and aminophenols.

Chloroanilines have been associated with renal and hepatic toxicity. This study (a) examined the in vitro hepatic and renal toxicity of 2-chloroaniline and 4-chloroaniline, (b) further examined whether aromatic ring hydroxylation would increase toxicity of the parent compound and (c) compared toxicity between respective aminochlorophenol and aminophenol. Renal and hepatic slices were exposed to varying concentrations of 2-chloroaniline, 4-chloroaniline. 4-amino-3-chlorophenol, 2-amino-5-chlorophenol, 2-aminophenol or 4-aminophenol. Toxicity was monitored by measurement of pyruvate-directed gluconeogenesis and leakage of lactate dehydrogenase (LDH). Hepatic tissue was less susceptible to toxicity than kidney tissue for all compounds since LDH leakage was elevated only in renal tissue. Gluconeogenesis was reduced in renal cortical slices exposed to 0.1 mum aminochlorophenols or 4-aminophenol, whereas a concentration of 0.5 mum was necessary for the chloroanilines and 2-aminophenol. LDH release was increased in renal slices by aminochlorophenols and aminophenols but not by the chloroanilines. The nephrotoxic potential in renal cortical slices was 4-aminophenol > 2-amino-5-chlorophenol > 4-amino-3-chlorophenol > 2-aminophenol > 2-chloroaniline = 4-chloroaniline. These results suggest that aromatic ring hydroxylation increased in vitro toxicity of the chloroanilines. Comparison of aminophenols with aminochlorophenols indicated that addition of a halogen can have variable effects on toxicity.

A mutant epidermal growth factor receptor common in human glioma confers enhanced tumorigenicity.

The development and neoplastic progression of human astrocytic tumors appears to result through an accumulation of genetic alterations occurring in a relatively defined order. One such alteration is amplification of the epidermal growth factor receptor (EGFR) gene. This episomal amplification occurs in 40-50% of glioblastomas, which also normally express endogenous receptors. Moreover, a significant fraction of amplified genes are rearranged to specifically eliminate a DNA fragment containing exons 2-7 of the gene, resulting in an in-frame deletion of 801 bp of the coding sequence of the extracellular domain. Here we used retroviral transfer of such a mutant receptor (de 2-7 EGFR) into glioblastoma cells expressing normal endogenous receptors to test whether the mutant receptor was able to augment their growth and malignancy. Western blotting analysis showed that these cells expressed endogenous EGFR of 170 kDa as well as the exogenous de 2-7 EGFR of 140-155 kDa. Although holo-EGFRs were phosphorylated on tyrosine residues only after exposure of the cells to ligand, de 2-7 EGFRs were constitutively phosphorylated. In tissue culture neither addition of EGF nor expression of the mutant EGFR affected the rate of cell growth. However, when cells expressing mutant EGFR were implanted into nude mice subcutaneously or intracerebrally, tumorigenic capacity was greatly enhanced. These results suggest that a tumor-specific alteration of the EGFR plays a significant role in tumor progression perhaps by influencing interactions of tumor cells with their microenvironment in ways not easily assayed in vitro.

Anti-CD31 delays platelet adhesion/aggregation at sites of endothelial injury in mouse cerebral arterioles.

The arterioles on the surface of the mouse brain (pial arterioles) were observed by in vivo microscopy. A focus of minor endothelial damage was produced in a single pial arteriole in each mouse by briefly exposing the site to a helium neon laser after an intravenous injection of Evans blue. Mice were injected 10 minutes before injury with a monoclonal antibody (MAb) to mouse CD31, also known as platelet endothelial cell adhesion molecule. This treatment doubled (P < .01) the time required for the laser to produce a recognizable platelet aggregate. In additional experiments, an MAb to mouse CD61 and an MAb to mouse intercellular adhesion molecule 1 had no effect. The data support previous observations indicating that platelet adhesion/aggregation in this model is induced by endothelial injury without exposure of basal lamina. The data are consistent with the hypothesis that the endothelial injury exposes or activates a platelet endothelial cell adhesion molecule on the endothelium which is blocked by the MAb directed against CD31. This may be the first demonstration of an effect of an anti-platelet endothelial cell adhesion molecule on platelet endothelial cell adhesion molecule on platelet adhesion/aggregation in vivo.

Monoclonal antibody-mediated protection and neutralization of motility in experimental Proteus mirabilis infection.

A panel of monoclonal antibodies with specificity for a wound isolate of Proteus mirabilis was established. Of nine antibodies studied in detail, three were broadly reactive with various Proteus isolates, while six reacted in a serotype-specific fashion with the strain used for immunization. Five of the six serotype-specific antibodies were reactive with lipopolysaccharide. The sixth serotype-specific antibody, 4-F (immunoglobulin G1 [IgG1]), was potently protective in a burn wound sepsis model and recognized a protein antigen. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot (immunoblot) analysis were used to determine that 4-F was reactive with flagellar protein. Approximately 1.3 micrograms of the antibody was sufficient to provide protection against 8 50% lethal doses of wound isolate, and approximately 26 micrograms provided full protection against challenge with 333 50% lethal doses. In vitro test results indicated that 4-F inhibited the motility of the wound isolate, and in vivo testing showed that it inhibited dissemination of the inoculum from the burn site to the liver and spleen. Whereas the antibody was highly effective in preventing the death of mice subsequent to challenge at a burn site, no protection was seen following an intraperitoneal challenge. These results may therefore indicate that the protection observed in the burn model is solely a reflection of the capacity of 4-F to neutralize bacterial motility.

Detection of IgG-associated determinants in reduced and alkylated preparations of human IgG3 by monoclonal antibodies.

Using classical typing antisera, previous experiments have failed to demonstrate IgG3 in partially reduced and alkylated preparations of human IgG intended for intravenous application (IGIV). To establish that IgG3 is actually present in such preparations, we designed an enzyme-linked immunosorbent assay (ELISA) using monoclonal antibodies as solid-phase reagents and protein A-purified IgG3 as antigen. Three different samples of reduced and alkylated antigen were used: (1) IgG3 isolated from a ready-for-infusion IGIV; (2) IgG3 which was purified from an intramuscular (Cohn fraction II) IgG solution before being subjected to a mild reduction and alkylation procedure, and (3) completely reduced and alkylated IgG3. The reduction and alkylation procedure did not affect the solubility of IgG3, indicating that IGIV prepared in this manner should contain normal quantities of IgG3. In the ELISA, solid-phase monoclonals which were cross-reactive with multiple IgG subclasses clearly reacted with reduced and alkylated IgG3. Furthermore, there was no substantial difference between the quantities of modified and native antigen required for 50% maximal ELISA signal. In contrast, solid-phase monoclonals with IgG3-restricted specificity did not recognize reduced and alkylated material. These results indicate that IGIV prepared by reduction and alkylation has a normal IgG3 content and confirm that some IgG3-specific determinants are altered by the modification procedure.

Antigenic relationships of murine coronaviruses: analysis using monoclonal antibodies to JHM (MHV-4) virus.

Monoclonal antibodies were produced to JHMV-DL, a neurotropic member of the mouse hepatitis virus (MHV) or murine coronavirus group. Of 23 antibodies isolated, 10 were specific for the major envelope glycoprotein, gp180/90, 10 for the nucleocapsid protein, pp60, and 3 for the minor envelope glycoprotein, gp25. Eleven different MHV isolates were used in antibody binding assays to study antigenic relationships among the viruses. Each MHV isolate tested had a unique pattern of antibody binding, indicating that each is a distinct strain. Conservation of JHMV-DL antigenic determinants varied among the three proteins, with pp60 showing intermediate conservation, gp180/90 little conservation, and gp25 marked conservation in the different MHV strains. Monoclonal antibodies to pp60 proved most useful in delineating antigenic relationships among MHV strains. These antigenic groups correlated with pathogenic types, indicating that pp60 may be one of the gene products which mediates the distinct disease patterns manifested by different murine coronaviruses.

Natural killer cell activity during mouse hepatitis virus infection: response in the absence of interferon.

The ability of the JHM3 strain of mouse hepatitis virus (MHV) to induce natural killer (NK) cells was examined. Infection of C57BL/6 (B6) mice with this virus resulted in the augmentation of natural cytotoxicity against YAC-I target cells in the absence of a detectable interferon response. The cells responsible for this increased cytotoxicity were sensitive to complement-mediated lysis with an anti-Q-5 reagent but not with a Thy 1.2 antiserum, indicating that they possess an NK-like surface phenotype. Although variation in the NK response of individual B6 mice following JHM virus infection was found, even the animal with the most responsive NK cell population had no detectable interferon in the spleen. This finding contrasted with observations with an unrelated virus (lymphocytic choriomeningitis virus) and a serologically related strain of MHV. Infection with both of these viruses induced augmented NK cell activity and interferon responses. In addition, we found that neither the ability to mount an augmented NK cell response nor preferential lysis of virus-infected targets correlated with resistance or susceptibility to JHM virus infection.

Specific recognition of the product of a transferred major histocompatibility complex gene by cytotoxic T lymphocytes.

Mouse L cells transfected with a genomic clone containing the H-2Ld gene (8-5 cells) were shown to function as targets for H-2Ld-specific cytotoxic T lymphocytes (CTL). The CTL-mediated lysis of 8-5 cells was shown to be H-2Ld specific by the use of (i) CTL with restricted reactivity, (ii) unlabeled target inhibiton, and (iii) monoclonal antibody inhibiton. We also demonstrated that 8-5 cells could function as targets for antibody-plus-complement-mediated cell lysis. Specificity was confirmed by using H-2Ld-specific monoclonal antibodies. These experiments demonstrate that the gene products of a major histocompatibility complex genomic clone can be functionally expressed in a foreign cell and can mediate immunologically specific cellular interactions.

Immunity to MCA-induced rat sarcomas: analysis of in vivo and in vitro results.

Three in vivo techniques were used to establish the specificity of tumor immunity induced after sensitization of F344 rats to syngeneic MCA-induced sarcomas: (1) post-excision resistance to tumor challenge, (2) passive tumor neutralization (the Winn test), and (3) concomitant immunity. In general, these assays revealed unique non-cross-reactive antigens associated with each of three sarcomas, FMF1, FMM2, and FMM3. However, spleen cells from tumor-sensitized rats did not demonstrate cell-mediated cytotoxicity in vitro corresponding to the specificity of tumor resistance in vivo. In the (3H)-proline cytotoxicity assay, spleen cells from FMM3 tumor-bearing rats or from FMM3 tumor-immune rats were not selectively cytotoxic for cultured FMM3 target cells. Parallel analysis of spleen cells from normal or FMM3-sensitized rats using the Winn assay and the (3H)-proline assay revealed that (1) spleen cell cytotoxicity in vitro did not correlate with effective tumor protection in vivo; and (2) normal spleen cells were cytotoxic against cultured sarcoma target cells in vitro and inhibited tumor growth in vivo. Thus, passive tumor protection by normal spleen cells in vivo corresponded with the demonstration of natural cytotoxicity in vitro, but induced specific anti-tumor reactivity was observed only in vivo.