Acute Kidney Injury in Cirrhosis: Baseline Serum Creatinine Predicts Patient Outcomes.

OBJECTIVES: The International Ascites Club (IAC) recently defined Stage 1 acute kidney injury (AKI) for cirrhosis as an acute increase in serum creatinine (SCr) by ≥0.3 mg/dl or by ≥50% in <48 h from a stable value within 3 months. The baseline SCr may influence AKI risk and patient outcomes. The objective of this study is to determine in cirrhosis whether the baseline SCr has any effect on the in-hospital AKI course and patient survival. METHODS: North American Consortium for the Study of End-Stage Liver Disease is a consortium of tertiary-care hepatology centers prospectively enroling non-elective cirrhotic inpatients. Patients with different baseline SCr levels (≤0.5, 0.51-1.0, 1.01-1.5, >1.5 mg/dl) were evaluated for the development of AKI, and compared for AKI outcomes and 30-day survival. RESULTS: 653 hospitalized cirrhotics (56.7±10years, 64% men, 30% with infection) were included. The incidence of AKI was 47% of enrolled patients. Patients with higher baseline SCr were more likely to develop AKI, with significantly higher delta and peak SCr (P<0.001) than the other groups, more likely to have a progressive AKI course (P<0.0001), associated with a significantly reduced 30-day survival (P<0.0001). Multivariate logistic regression showed that the delta SCr during an AKI episode to be the strongest factor impacting AKI outcomes and survival (P<0.001), with a delta SCr of 0.70 mg/dl having a 68% sensitivity and 80% specificity for predicting 30-day mortality. CONCLUSIONS: Admitted cirrhotic patients with higher baseline SCr are at higher risk for in-hospital development of AKI, and more likely to have AKI progression with reduced survival. Therefore, such patients should be closely monitored and treated promptly for their AKI.

Gender-Specific Differences in Baseline, Peak, and Delta Serum Creatinine: The NACSELD Experience.

BACKGROUND: Women have lower serum creatinine values than men for similar renal function. AIMS: We aimed to determine the differential effect of baseline, peak, and delta creatinine between genders on outcomes in infected hospitalized cirrhotic patients. METHODS: North American Consortium for the Study of End-Stage Liver Disease is a 15-center consortium of tertiary care hepatology centers prospectively enrolling infected cirrhotic inpatients. Baseline, peak, and delta creatinine during hospitalization were compared between genders, and their impact on overall survival, transplant-free survival, probability of transplantation, and need for renal replacement therapy (RRT) was analyzed. RESULTS: In total, 532 patients with cirrhosis (males = 59% median admission MELD = 20) had significantly lower median baseline (1.07 vs. 1.30 mg/dL, p < 0.0001) and peak creatinine (1.47 vs. 1.59 mg/dL, p = 0.024) in women than men during hospitalization for an infection, but both genders had similar delta creatinine levels (0.30 vs. 0.30 mg/dL, p = 0.957). Thirty-day mortality was similar between genders. RRT was not used more often in women (19 vs. 12%, p = 0.050), and women were 1.8 times more likely than men to receive RRT at the same creatinine (p = 0.028). Both peak and delta creatinine significantly predicted 6-month overall and transplant-free survival (p < 0.0001), but the probability of liver transplantation was affected by the interaction between gender and both peak and delta creatinine. CONCLUSIONS: Infected hospitalized cirrhotic women are significantly more likely than men to receive RRT at similar creatinine levels. Gender-specific differences in baseline, peak, and delta creatinine need further investigation to determine whether women need acute kidney injury treatment at lower creatinine thresholds than men.

Prospective Multicenter Observational Study of Overt Hepatic Encephalopathy.

BACKGROUND: Overt hepatic encephalopathy (OHE) is a frequent complication of decompensated cirrhosis. AIMS: A multicenter prospective observational study was performed to assess the most commonly recorded presenting manifestations of OHE and its associated health-care burden. METHODS: Qualifying patients must have experienced ≥1 OHE episode within 30 days of enrollment (qualifying OHE) and were followed for recurrence (on-study OHE). RESULTS: Two hundred and sixty-five patients were enrolled at 30 sites and followed for up to 9 months (mean 72 days). Seventy-two patients experienced 122 on-study episodes; with 72, 23, and 13 having ≥1, ≥2, or ≥3 on-study episodes with median days to occurrence of the 1st, 2nd, and 3rd episode of 34, 19, and 11, respectively. The most frequently recorded OHE manifestations included confusion (78 %), change in mental status (57 %), disorientation (48 %), lethargy (46 %), and asterixis (45 %). West Haven grade was used inconsistently and recorded for only 28 % of episodes. Most qualifying and on-study episodes occurred on rifaximin (60 and 82 %, respectively) and were associated with hospitalization (68 and 85 %, respectively). Twenty-three patients experienced ≥2 on-study episodes within 2 months of enrollment on average (median 45 days) and accounted for 60 % of on-study episodes. CONCLUSIONS: In this prospective study, OHE's most commonly recorded presenting manifestations included confusion, altered mental status, disorientation, lethargy, and asterixis. As reflected by frequent recurrence and hospitalizations, OHE, particularly the approximately 10 % of "high-resource-utilizing" patients with frequent recurrence, continues to pose a major unmet medical need and health-care burden despite the use of rifaximin.

Risk of cardiovascular disease in HIV, hepatitis C, or HIV/hepatitis C patients compared to the general population.

BACKGROUND: As a result of effective antiretroviral therapy HIV patients are living longer, and their risk of cardiovascular disease (CVD) is a growing concern. It remains unknown whether coinfection with hepatitis C (HCV) changes an HIV person's CVD risk, and how the risks compare to the general population. The objective of this study was to compare the Framingham Risk Score (FRS) and vascular age differences in persons with HIV, HCV or HIV/HCV disease to the general population. METHODS: HIV, HCV, and HIV/HCV patients with clinic visits between 2004 and 2009 were sampled from medical clinics in Rochester, NY. Uninfected persons were randomly selected from the National Health and Nutrition Examination Survey (NHANES), and individually matched on gender, race, and age. We stratified by infection group and conducted separate multivariable linear regression analyses between each infection group and the gender, race, and age matched participants from NHANES. RESULTS: Rochester patients (HIV = 239, HCV = 167, HIV/HCV = 182) were compared 3 : 1 with the NHANES participants. After controlling for weight, marital status, current pharmacotherapies and the matching variables of gender, race, and age, HIV/HCV patients had a 2% higher general FRS compared with the general population (p = 0.03), and vascular age differences that were 4.1 years greater (p = .01). HCV patients had a 2.4% higher general FRS than the general population (p < .001), and vascular age differences that were 4.4 years greater (p < .001). CVD risk was elevated but not significantly different between HIV patients and the general population. CONCLUSION: Cardiovascular disease risk is elevated among HIV/HCV and HCV infected persons compared with the general population.

Zollinger-Ellison syndrome, acromegaly, and colorectal neoplasia.

Zollinger-Ellison syndrome (ZES) and acromegaly are two hypersecretory states in which colorectal neoplasia has been described, but the incidence in the former condition may not be increased. We describe four patients with colorectal neoplasia associated with the ZES and review other published cases. Tissue ELISA with Adnab-9 antibody, a putative colorectal cancer risk marker, from a patient with ZES and from seven patients with acromegaly was compared to 13 controls at average risk for colorectal neoplasia. The patient with ZES without detectable colonic neoplasia and seven patients with acromegaly had increased binding of Adnab-9 in the colonic mucosa by ELISA. The difference was significant for the acromegaly patients compared to the controls (p < 0.05). The accumulated 34 instances of colorectal neoplasia in ZES patients suggests that this association may not be rare. Adnab-9 expression, detectable in both ZES and acromegaly, may reflect predisposition to colorectal neoplasia in both hyper-secretory states. Therefore, while a basis for association of colorectal neoplasia and hypergastrinemia exists, the clinical data are not compelling enough to warrant surveillance of patients with ZES. To resolve this problem, more definitive case control studies should be conducted.

Fish oil dietary supplementation for prevention of indomethacin induced gastric and small bowel toxicity in healthy volunteers.

OBJECTIVE: To determine if N-3 fatty acid (fish oil) dietary supplements could favorably alter indomethacin induced gastric and small bowel toxicity related to use of nonsteroidal antiinflammatory drugs (NSAID). METHODS: Healthy volunteers consumed 8 g of N-3 fatty acids for 16 weeks, while controls consumed corn oil. Subjects ingested indomethacin 50 mg tid between Weeks 12 and 16. Upper gastrointestinal (GI) endoscopy with biopsies and Cr-EDTA swallows were performed at Week 12 and again at Week 16. Biopsy specimens were graded for inflammation and endoscopic scores were recorded. RESULTS: No significant differences were seen between groups in any study variable, although the direction of change favored the fish oil subjects for inflammation scores in both the stomach and duodenum (fish oil subjects -0.50 +/- 1.2 stomach, -0.28 +/- 0.97 duodenum; and corn oil subjects +0.10 +/- 0.84 stomach, +0.20 +/- 0.79 duodenum; p = 0.086). Direction of change in stomach inflammation showed a trend favoring fish oil (p = 0.056 by chi square). CONCLUSION: Although no significant differences were seen between groups, the changes observed in inflammation suggest a possible benefit of fish oil in the amelioration of NSAID induced GI inflammation.

Omeprazole inhibits growth of cancer cell line of colonic origin.

The direct effects of omeprazole on colonic cells has not been evaluated. Controversy exists regarding the potential adverse effects of omeprazole on cell proliferation. In order to mimic the in vivo situation in the patient treated with omeprazole, proliferation cell culture experiments were performed, monitoring directly the effects of gastrin and omeprazole both alone and in combination. Three colonic cancer cell lines were used, two with neuroendocrine features (NCI-H716, LCC-18) and one (DLD-1) not known to have these features. In these in vitro proliferation experiments, only the NCI-H716 colorectal cancer cell line responded to omeprazole by decreased proliferation (P < 0.05). The effect was concentration dependent shown for all doses of omeprazole used. Gastrin had a statistically significant effect on increasing proliferation in the NCS-H716 cell line alone but only at the highest concentration (10(-6) M). Omeprazole has a cytostatic effect on one of three colorectal cancer cell lines but the mechanism for this effect of omeprazole and its potential role in treatment awaits elucidation.

Urinary organ specific neoantigen. A potentially diagnostic test for colorectal cancer.

Urinary organ-specific neoantigen from colorectal cancer patients has been used to make a monoclonal antibody, BAC 18.1. In this study we assessed the potential of this antibody for the diagnosis of colorectal cancer. We evaluated binding in both urine and effluent samples and compared it with effluent carcinoembryonic antigen standardized for both volume (nanograms per milliliter) and protein. Urinary organ-specific antigen as detected by BAC 18.1 was significantly greater in 29 cancer patients (A405: 0.717 +/- 0.500) vs 27 controls [0.121 +/- 0.273 (P < 0.05)]. Considerable overlap of binding of BAC 18.1 was observed in the colonic effluent of patients with CRC (N = 13), adenomas (N = 26), inflammatory bowel disease (N = 8), or having a normal colonoscopic examination (N = 24). CEA levels (nanograms per milliliter) were significantly elevated in the effluent samples of patients with a past history of colorectal cancer, as compared to that of normal individuals (P < 0.05). The presence of the M(r) 30,000 organ-specific neoantigen in colonic effluent was also demonstrated by western blot. Organ-specific neoantigen originates in the colon and is excreted into the urine, so the BAC 18.1 binding levels in the urine may be a diagnostic aid for CRC.

Adenoma-derived antibody, Adnab-9 recognizes a membrane-bound glycoprotein in colonic tissue and effluent material from patients with colorectal neoplasia.

The binding in pre-colonoscopic effluent of Adnab-9, a monoclonal antibody raised against colonic adenomas, was evaluated for specificity in the diagnosis of colorectal cancer. A heterogeneous group of 58 patients was evaluated by ELISA. Effluent samples and tissue extracts were subjected to Western blotting or ELISA to confirm specificity. Immunohistochemistry was performed on the cancer tissue sections. The proportion of positive effluent binding was higher in the cancer when compared to the normal group (P = 0.036). A dominant 87 M(r) band was found in adenoma extracts and some effluent samples. Adnab-9 binding in effluent samples predominated in membrane-bound fractions. Immunohistochemistry showed no specific staining in the cancer cells. The antigen recognised is a glycoprotein shown by effects of N-glycanase digestion and not cross-reactive with carcinoembryonic antigen. Non-gastro-intestinal tissue extracts did not bind Adnab-9. The major 87 M(r) adenoma-derived antigen may be found in effluent material, particularly in the membrane-bound fraction.

Cellular distribution of a colonic adenoma-associated antigen as defined by monoclonal antibody Adnab-9.

Adenomatous colonic polyps constitute a precursor for colorectal cancer. Antibodies to these precancerous lesions might identify specific early tumor antigens. Adnab-9 is a murine monoclonal antibody raised against membranes of colonic adenomas. Adnab-9 binding in colonic washings (effluent) correlates with the presence of colorectal cancer. Immunohistochemical staining with Adnab-9 shows cytoplasmic reactivity in scattered cells in 4 of 31 adenomatous tissue sections, 0 of 14 sections of colorectal cancer cells, and 1 of 8 normal-appearing colonic mucosa specimens examined. Adnab-9 recognized a dominant M(r) 87,000 protein species in tissue extracts in the membrane-bound fraction of effluent by Western blotting. Adnab-9 binding by enzyme-linked immunosorbent assay in adenomatous extracts is higher than cancer or normal tissue, is membrane-bound, and is absent from established colorectal cancer cell lines. This distribution and nature of immunostaining suggest that Adnab-9 recognizes a determinant associated with the membrane component of a subpopulation of adenoma cells which may have a role in early colorectal neoplasia.

Oral colon lavage solutions containing polyethylene glycol may interfere with ELISA detection of tumor-associated antigens in colonic effluent.

Immunologic methods for detection of colorectal neoplasia based on examination of stool or colonic effluent are being developed. Most current oral lavage preparations contain polyethylene glycol (PEG), and if PEG adversely interferes with immunologic testing these tests may become less useful. We describe a decrease in sensitivity of ELISA for tumor-associated antigens (TAA) when effluent samples are diluted in PEG-electrolyte lavage solution, equivalent to a commonly used oral lavage solution based on PEG. Radioisotope-labeled antigen binding to plastic plates was decreased by dilution in the PEG lavage solution. Antigen binding, present in colonic effluent collected by the laxative purge method, was absent in effluent collected by PEG oral lavage from the same patient. We conclude that PEG and PEG-containing lavage solutions interfere with ELISA detection of TAA in colonic effluents. We speculate that the in vitro, and possibly the in vivo, effect occurs at the level of antigen binding to the plate either by a steric effect or alteration of charge by the nonpolar properties of PEG.