Bacterial infections are an important cause of mortality in cirrhosis but there is a paucity of multi-center studies. The aim was to define factors predisposing to infection-related mortality in hospitalized cirrhotic patients. Methods A prospective, cohort study of cirrhotic patients with infections was performed at eight North American tertiary-care hepatology centers. Data were collected on admission vitals, disease severity [MELD and sequential organ failure (SOFA)] scores], first infection site, type [community-acquired, health care-associated (HCA) or nosocomial], and second infection occurrence during hospitalization. The outcome was mortality within 30 days. A multi-variable logistic regression model predicting mortality was created. Results 207 patients (55 years, 60% men, MELD 20) were included. Most first infections were HCA (71%), then nosocomial (15%) and community-acquired (14%). Urinary tract infections (52%), spontaneous bacterial peritonitis (SBP, 23%) and spontaneous bacteremia (21%) formed the majority of the first infections. Second infections were seen in 50 (24%) patients and were largely preventable: respiratory, including aspiration (28%), urinary, including catheter-related (26%), fungal (14%) and C. difficile (12%) infections. Forty-nine patients (23.6%) who died within 30 days had higher admission MELD (25 vs 18, p<0.0001), lower serum albumin (2.4g.dL vs. 2.8g/dL, p=0.002), and second infections (49% vs. 16%, p<0.0001) but equivalent SOFA scores (9.2 vs. 9.9, p=0.86). Case fatality rate was highest for C. difficile (40%), respiratory (37.5%) and spontaneous bacteremia (37%), and lowest for SBP (17%) and urinary infections (15%). The model for mortality included admission MELD (OR: 1.12), heart rate (OR:1.03) albumin (OR:0.5) and second infection (OR:4.42) as significant variables. Conclusions Potentially preventable second infections are predictors of mortality independent of liver disease severity in this multi-center cirrhosis cohort.

We evaluated the clinical characteristics of patients with Francisella philomiragia (formerly Yersinia philomiragia) isolated from normally sterile sites. Isolates from 14 patients were received by the Centers for Disease Control between 1975 and 1987: 9 were from blood; 2 from lung biopsies; and 1 each from pleural, peritoneal, and cerebrospinal fluid. Underlying problems included chronic granulomatous disease in 5 patients, near-drowning in 5, and a myeloproliferative disease in 2. All 13 patients for whom records were available had a febrile syndrome compatible with bacterial infection. Pneumonia and fever-bacteremia were the commonest clinical syndromes reported. In 7 cases, F. philomiragia was the only sterile-site isolate, and the clinical syndrome did not resolve without appropriate antibiotics. Familiarity with this organism is important because of its ability to cause serious disease in chronic granulomatous disease and near-drowning patients. Further study may yield new insights into pathogenic and host defense mechanisms.

Barrett's esophagus develops when refluxed gastric juice injures the esophageal squamous lining and the injury heals through a metaplastic process in which intestinal-type columnar cells replace squamous ones. The progenitor cell that gives rise to Barrett's metaplasia is not known, nor is it known why the condition is predisposed to malignancy. We studied the contribution of bone marrow stem cells to the development of Barrett's esophagus in an animal model. Twenty female rats were given a lethal dose of irradiation followed by tail vein injection of bone marrow cells from male rats. Ten days later, the female rats were randomly assigned to undergo either esophagojejunostomy, a procedure that causes reflux esophagitis with intestinal metaplasia, or a sham operation. The rats were killed at 8 weeks and serial sections of the snap-frozen esophagi were cut and mounted on slides. The first and last sections were used for histological evaluation and the intervening sections were immunostained for cytokeratin to identify epithelial cells and analyzed for Y chromosome by fluorescence in situ hybridization (FISH). Histological evaluation of the esophagi from rats that had esophagojejunostomy revealed ulcerative esophagitis and multiple areas of intestinal metaplasia. FISH analyses showed that some of the squamous epithelial cells and some of the columnar epithelial cells lining the glands of the intestinal metaplasia were positive for Y chromosome. These observations suggest that multi-potential progenitor cells of bone marrow origin contribute to esophageal regeneration and metaplasia in this rat model of Barrett's esophagus.

Cell-mediated cytotoxicity plays an important role in the clearance of noncytopathic viruses from infected tissues. Perforin-dependent cytotoxic mechanisms have been noted to play an important role in the clearance of infections from multiple extrahepatic organs. In contrast, mice with defects in the Fas/Fas ligand (FasL)-mediated cytotoxicity pathway exhibit delayed clearance of adenovirus from the liver without apparent delay in the clearance of viral infections from extrahepatic organs. The present studies examined the role of cytotoxic effector mechanisms in intrahepatic immune responses to a replication-defective, recombinant β-galactosidase-encoding adenovirus (AdCMV-lacZ). Delayed clearance of AdCMV-lacZ from the livers of FasL-defective B6.gld mice, but not perforin-deficient B6.pfp−/− mice, was noted despite no significant differences in initial hepatic CD8+ T cell IFN-γ or TNF responses or in activation of intrahepatic cytotoxic lymphocytes cells capable of killing AdCMV-lacZ-infected fibroblast targets. In contrast, AdCMV-lacZ-infected hepatocyte targets were far more sensitive to killing by intrahepatic cytotoxic lymphocytes from B6.pfp−/− than from B6.gld mice, and residual levels of virus-specific killing of hepatocyte targets by FasL-defective B6.gld CTL were blocked by TNF inhibition. These results suggest that inherent resistance of hepatocytes to cytotoxicity mediated by perforin-dependent mechanisms leaves Fas/FasL-dependent, cell-mediated cytotoxicity as the major pathway for CTL-mediated killing of virally infected hepatocytes and accounts for the more prominent role of perforin-independent anti-viral mechanisms in immune responses in the liver.

No abstract

TNF-TNFR2 interactions promote MHC class II-stimulated alloresponses while TNF-TNFR1 interactions promote MHC class I-stimulated alloresponses. The present studies were designed to evaluate whether TNF-TNFR2 interactions were involved in the in vivo generation of CD4+ T cell-mediated intestinal graft-versus-host disease (GVHD) in the (C57BL/6J (hereafter called B6)→B6 × B6.C-H-2bm12 (bm12))F1 GVHD model. Briefly, 5 × 106 splenic CD4+ T lymphocytes from B6.TNFR2−/− or control B6 mice were transferred with 1–2 × 106 T cell-depleted B6 bone marrow cells (BMC) to irradiated MHC class II-disparate (bm12 × B6)F1 mice. Weight loss, intestinal inflammation, and the surface expression of CD45RB (memory marker) on intestinal and splenic lymphocytes were assessed. IL-2 and IFN-α mRNA levels in intestinal lymphocytes were assessed by nuclease protection assays. A significant reduction in weight loss and intestinal inflammation was observed in recipients of the TNFR2−/−CD4+ SpC. Similarly, a significant decrease was noted in T cell numbers and in CD45RBlow (activated/memory) expression on intestinal but not CD4+ T cells in recipients of TNFR2−/−CD4+ spleen cells. IL-2 and IFN-α mRNA levels were reduced in the intestine in the recipients of TNFR2−/− splenic CD4+ T cells. These results indicate that TNF-TNFR2 interactions are important for the development of intestinal inflammation and activation/differentiation of Th1 cytokine responses by intestinal lymphocytes in MHC class II-disparate GVHD while playing an insignificant role in donor T cell activation in the spleen.

Deficiency or inhibition of tumor necrosis factor (TNF) significantly prolongs hepatic expression of recombinant adenoviral vectors. To explore mechanisms responsible for this observation, the present studies examined the effects of TNF versus TNF receptor 1 (TNFR1) or TNFR2 deficiency on the course of antiviral-immune responses to a replication-deficient, beta-galactosidase-encoding recombinant adenovirus (AdCMV-lacZ). Clearance of AdCMV-lacZ was significantly delayed in TNF-deficient mice. Less pronounced but significant delays in AdCMV-lacZ clearance were observed in TNFR2-deficient but not TNFR1-deficient mice. Numbers of interferon-gamma expressing intrahepatic lymphocytes (IHL) were similar in AdCMV-lacZ-infected, TNF-deficient, TNFR1-deficient, TNFR2-deficient, and control mice. However, IHL isolated from AdCMV-lacZ-infected, TNF-deficient or AdCMV-lacZ-infected, TNFR2-deficient mice exhibited decreased levels of FasL expression and adenovirus-specific cytolytic T lymphocyte (CTL) activity. Similar defects in allo-specific killing of Fas-sensitive hepatocyte targets by TNF-deficient or TNFR2-deficient but not TNFR1-deficient CTL were also noted. No defects in generation of allo-specific cytotoxicity directed against perforin-sensitive target cells were noted in TNF-, TNFR1-, or TNFR2-deficient lymphocytes. These findings indicate that TNF/TNFR2 interactions facilitate generation of FasL-dependent CTL effector pathways that play an important role in in vivo antiviral-immune responses in the liver.

Osteoporosis is increasingly reported in the aging HIV-positive population, and co-infection with hepatitis C virus (HCV) may further increase the risk of osteoporosis. However, it remains unclear whether HCV-related increased fracture risk is a function of the severity of liver disease. We calculated the time-updated alanine aminotransferase to platelet ratio index (APRI) score (an indirect marker of hepatic fibrosis) in all HIV-infected patients enrolled in the Veterans Affairs' Clinical Case Registry between 1984 and 2009. The association between HCV co-infection and incident osteoporotic fracture (defined as closed wrist, vertebral, or hip fracture) was assessed in univariate and multivariate Cox survival models adjusting for traditional risk factors for osteoporosis and APRI score or the presence of cirrhosis. A total of 772 osteoporotic fractures were identified among 56,660 HIV-infected patients (98.1% male; 31.3% HCV co-infected; median age 44.0 years) contributing 305,237 patient-years of follow-up. Fracture rates were significantly higher among HIV/HCV patients than HIV-only patients (2.57 versus 2.07/1000 patient-years, relative risk = 1.24, p < 0.0001). In a Cox multivariable model including age, race, smoking, drug use, body mass index, and antiretroviral therapy, HCV co-infection remained an independent predictor of osteoporotic fractures after controlling for presence of cirrhosis (hazard ratio [HR] = 1.32; p <0.001) or APRI score (HR = 1.30; p = 0.003). Among HIV/HCV co-infected patients, cirrhosis strongly predicted osteoporotic fractures (HR = 1.65; 95% confidence interval [CI] 1.11-2.44; p = 0.012), but APRI score was a weaker predictor (HR = 1.008; 95% CI 1.002-1.014; p = 0.015). In conclusion, among HIV-infected patients, severity of liver disease partly explains the HCV-associated increased risk of osteoporotic fractures. Other determinants of this increased risk remain to be defined.