948 Refeeding After Total Parenteral Nutrition (TPN) Leads to a PRO-Inflammatory State Within the Intestinal Mucosa

948 Refeeding After Total Parenteral Nutrition (TPN) Leads to a PRO-Inflammatory State Within the Intestinal Mucosa

AGA Abstracts Hospitalier de l'Université de Montréal). Liver tests were obtained in parallel. Twenty-three patients were classified into F0 to F4 ac...

379KB Sizes 0 Downloads 9 Views

AGA Abstracts

Hospitalier de l'Université de Montréal). Liver tests were obtained in parallel. Twenty-three patients were classified into F0 to F4 according to their liver stiffness (LS) results, based on Corpechot et al. 1. A fibrosis score of ≥F2 was considered clinically significant. Cholestasis was defined by the presence of alkaline phosphatase levels superior to the upper limit of normal. Results: A total of 23 patients underwent transient elastography measurements. Reliable readings were obtained in 22 patients. The average LS value was 8,21 +/- 3,49 kPa (3,4-34,8 kPa). Patients were classifies into: F0-F1 for 14, F2 for 4, F3 for 1 and F4 for 3 patients respectively. Therefore, significant fibrosis (≥F2) was present in 8 patients (36%). LS results were higher in patients with biochemical cholestasis: 12,5 +/-6,9 kPa in cholestatic versus 4,63 +/- 0,49 kPa (p=0,05) in non-cholestatic subjects. Conversely, patients with significant fibrosis had higher alkaline phosphatase levels (169 +/- 58 units per liter versus 97 +/- 14 units per liter, p=0,045). Furthermore, there was a correlation (r=0,36) between alkaline phosphatase values and the degree of fibrosis as evaluated by transient elastography (p<0.0001). Conclusion: Measuring transient elastography is feasible in patients on TPN. There is a significant correlation between liver stiffness and the presence of biochemical cholestasis. To determine whether increased liver stiffness is consecutive to the presence of liver fibrosis, a liver biopsy will need to be performed. When TPN-induced cholestasis is accompanied with increased liver stiffness, our results suggest that the phenomenon is significant. We need further follow-up to determine the prognostic value of elevated liver stiffness in predicting the occurrence of liver-related complications in TPN patients. 1Corpechot C, Naggar AL, Poujol-Robert A, et al. Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC. Hepatology 2006; 43:1118-24.

via a single taste stimulant, glutamate (GLM), via (T1R1/T1R3) and GLM receptors (GLMRs), would prevent TPN-associated atrophy and loss of epithelial barrier function (EBF). Methods: C57BL/6J male mice were randomized to SHAM, TPN, or TPN+GLM (oral glutamate water 75mM) and treated for 7 days. Expression of TRs, GLMRs and transporters was investigated with real-time PCR array. Intestinal morphology was assessed by measuring villus height and crypt depth. EBF was evaluated by transepithelial resistance(TER) and immunofluorescence (IF) staining of tight junction (TJ) proteins ZO-1 and occludin. Epithelial cell (EC) proliferation was investigated with proliferating cell nuclear antigen (PCNA) and BrdU staining. Results: TPN (vs. SHAM) led to significant changes in the abundances of several T1R members, as well as some GLMRs and transporters (Table), with most actually increasing. Oral GLM supplementation significantly reversed such changes, especially in GLMRs. We next investigated the effect of GLM on TPN mice. TPN+GLM mice (vs. TPN) showed significant increases in intestinal villus height and crypt depth (Table). Correspondingly, PCNA and BrdU staining both displayed 2-fold increases in TPN+GLM vs. TPN mice, which was further confirmed with 4-fold upregulation of PCNA expression at RNA level in TPN+GLM group (Table). SGLT1 also returned to SHAM values with GLM. TER significantly decline with TPN, and this rose moderately in the TPN+GLM group. IF staining of ZO-1 and occludin displayed a recovery of the continuity of epithelial TJ network, which was disrupted by TPN (Figure). Conclusion: TPN led to significant alterations in TRs and GLMRs in the small intestine. Loss of gut luminal sensing to nutrients may have contributed to TPN-associated intestinal atrophy. GLM significantly prevented atrophy and the impaired TJ functions of the intestine in TPN mice. This suggests that stimulating intestinal TRs and GLMRs through GLM supplementation could be a potential strategy for protecting TPN patients against associated severe complications.

1-Corpechot C, Naggar AL, Poujol-Robert A, et al. Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC. Hepatology 2006; 43:1118-24. 945 Cell Death Results From Digested Formula but Not Digested Fresh Human Breast: A Possible Mechanism for Necrotizing Enterocolitis Alexander H. Penn, Angelina E. Altshuler, James W. Small, Sharon F. Taylor, Karen R. Dobkins, Geert W. Schmid-Schönbein

a: p<0.05 vs SHAM; b: p<0.05 vs TPN

Background: Necrotizing enterocolitis (NEC) is a leading cause of death in premature infants and is characterized by hemorrhagic necrosis in the small intestine. Premature infants fed formula have more permeable intestines and are more likely to develop NEC than premature infants fed breast milk, though mechanisms are unclear. In ischemia, which increases intestinal permeability, our data show free fatty acids (FFAs) in the intestinal lumen can cause hemorrhagic necrosis. Therefore, milk or formula exposed to digestive enzymes resulting in formation of FFAs may also impact intestinal cells. Objective: To determine if digestion by enzymes present in the small intestine would result in cytotoxic levels of unbound FFAs in either formula or breast milk. Design/Methods: We digested infant formula or human breast milk (fresh or stored at -80 °C; N=12) with pancreatic lipase, proteases (trypsin and chymotrypsin), lipase+proteases, or fluid from the lumen of rat small intestine, measured FFA concentrations, and tested for cytotoxicity on cells (neutrophils, endothelial cells, and small intestinal epithelial cells) present in intestinal tissue. To determine their role in formula cytotoxicity, we reduced unbound FFAs (by glass fiber filtration or addition of bovine serum albumin), or added a lipase inhibitor (0.25 mg/ml orlistat) before or after digestion. Functional FFA binding capacity of control and protease digested breast milk and formula was determined by measuring cytotoxicity after addition of varying concentrations of the exogenous FFA, oleic acid. Results: Lipase digestion of formula, but not breast milk, caused significant cell death in neutrophils (60.0±15.6% vs. 6.2±7.4%, P<6x10-8) as well as endothelial (P<8x10-5) and epithelial cells (P<6x10-6 with lipase; P<0.0004 with luminal fluid). FFAs were significantly elevated in lipase-digested formula compared to breast milk (P<0.0003) and cell death was decreased with lipase inhibitor pretreatment (P<0.0006), addition of albumin (binds FFAs; P<8x10-5), or glass fiber filtration (also binds FFAs; P<8x10-5), suggesting FFAs cause the cytotoxicity. Protease digestion increased FFA binding capacity of formula and milk (P<0.05) but only enough to decrease cytotoxicity in milk (P<0.01). Conclusion: We suggest that FFA induced cytotoxicity is a possible mechanism for the pathogenesis of NEC.

The effect of oral GLM on TJ network of jejunum epithelium under TPN administration for 7 days. (Merge images of ZO-1 (Red) and occludin (Green) staining, 40×). Note that GLM restored the continuity of TJ of the jejunum villi (arrow). 948 Refeeding After Total Parenteral Nutrition (TPN) Leads to a PROInflammatory State Within the Intestinal Mucosa Yongjia Feng, Eiichi A. Miyasaka, Daniel H. Teitelbaum Background: TPN leads to a decline in epithelial barrier function(EBF) and profound increases in pro-inflammatory cytokines. TPN also leads to changes in the mucosally-associated intestinal microflora. Clinically when patients move from TPN to enteral feeding complications arise (intolerance and infections). The aim of this study was to investigate the effect of refeeding in a TPN mice model. Methods: C57BL/6J male mice received enteral nutrition (Control) or TPN for 7 days. A third group was given TPN for 7 days followed by cessation of TPN and given chow for 24 hours (n=6-8/group), small intestine was then harvested. Chemokine, cytokine and related genes were measured with real-time PCR , and EC apoptosis via TUNEL staining. Transepithelial resistance (TER) and FITC-Dextan concentration was measured to assess EBF, intestinal microflora was determined by 454 pyrosequencing in small ileum and colon. The paneth cell marker lysozyme was measured with IF staining, and paneth-derived cryptins were measured with real-time PCR. Results: TPN significantly increased TUNEL+ ECs in TPN mice (1% vs 10% control vs TPN, Table). This high apoptosis index did not reverse in the refeeding group. The pro-inflammatory cytokines TNF-α and IL-6 increased 2-fold with TPN mice vs. controls. TNF-α expression in refeeding mice rose even higher (4-fold) vs. controls, and IL-6 expression remained similar to the TPN group. Toll-like receptor 4 (TLR4) expression increased with TPN, and rose much more (4-fold) with refeeding. These increases were also seen with chemokine monocyte chemoattractant protein-1 (MCP1). The Treg-derived cytokines IL-10 and TGFβ were down-regulated with

947 Glutamate Supplementation Prevents Loss of TPN-Associated Intestinal Atrophy and Epithelial Barrier Function Weidong Xiao, Yongjia Feng, Daniel H. Teitelbaum Background: The underlying mechanisms contributing to total parenteral nutrition (TPN)associated intestinal atrophy are currently unknown. The intestine has recently been shown to contain a number of taste receptors (TRs) along the gastrointestinal tract, which may regulate energy supply via intestinal sensing and taste perception. We hypothesized that the loss of luminal nutrients with TPN would lead to a decline in TRs, and that this may be a contributing factor to intestinal atrophy. Additionally, we tested whether selective stimulation

AGA Abstracts


1 month at a midwestern academic medical center from 2003 to 2011. Medical records were analyzed for patient age, race, gender, indication for PN, duration of therapy, and serum 25-hydroxy vitamin D levels. Patients with end stage renal disease, hypercalcemia, and parathyroid disease were excluded. All patients received daily PN containing one vial of multivitamin with 200 IU of vitamin D3. Vitamin D deficiency was defined as a serum blood level <10 ng/ml, vitamin D insufficiency 10-30 ng/ml, and vitamin D sufficiency >30 ng/ml. Results: A total of 48 patients were identified, 36 female, 12 male. The median age was 51 years (range 19-85 years). Of the 48 patients, 28 (58.3%) were Caucasian, 18 (37.5%) were African-American, and 2 (4.2%) other. The most common indications for home PN were Crohn's Disease, complications of bariatric surgery, short bowel syndrome, scleroderma, ischemic bowel, and radiation enteritis. The average 25-hydroxy vitamin D level was 23.9 + 3.5 ng/ml at a mean duration of home PN for 7.9 + 4.3 months. Of the 48 patients, 32 (66.6%) had 25-hydroxy vitamin D levels less than 30 ng/ml. Women on home PN had a lower 25-hydroxy vitamin D level than men (18.8 + 4.0 ng/ml, 41.4 +7.5 ng/ml respectively). The lowest 25-hydroxy vitamin D levels were found in African-Americans (11.8 + 6.0 ng/ml). Conclusions: Low vitamin D levels are common in home PN patients. Standard one vial multivitamin preparations are not adequate for vitamin D sufficiency defined as >30 ng/ml, particularly in women and African-Americans. 950 Evaluation of Bone Mineral Density Changes in Patients Under Home Parenteral Nutrition Nicolas Benoît, Meriem Boufassa, Michel Lemoyne, Mickael Bouin, Louise D'Aoust INTRODUCTION: Osteoporosis is a very frequent condition in patients with intestinal failure or those under home parenteral nutrition (HPN). The exact role of HPN remains a controversial subject in the literature. Our hypothesis was that an adequate follow-up of patients under HPN lead to a stabilisation or even an improvement of the bone mineral density on the long range. METHODS: Retrospective study of patients under HPN followed by the Gastroenterology Service of St-Luc Hospital from 2004 to 2011. Exclusion criteria were the lack of data on bone mineral density or the absence of an osteodensitometry at the beginning of the HPN. Follow-up included regular visits with an endocrinologist, an initial dual energy X-ray absorptiometry at the start of HPN and every two years thereafter. Bone mineral densities at the hip (g/cm2) were collected and their evolution was measured. A reduction or an increase in bone mineral density was considered significant for a drop or a rise of 0,04 g/cm2 respectively. RESULTS: 44 patients under HPN were identified. Data on bone mineral density were collected for 30 individuals. A mean of 4 dual energy X-ray absorptiometry were performed for each patient. On average, we observed a drop of 0,01 g/cm2 (95 % confidence interval (CI); drop of 0,03 - rise of 0,01 g/cm2) in bone mineral density for the entire follow-up. Globally, more than 70 % of individuals showed a stabilisation or an increase in their bone mineral density on the long range. It looks like there was no difference in the evolution of the bone status between people with osteoporosis initially and those without osteoporosis at the beginning of HPN. In addition, a small waiting time between the diagnosis of intestinal insufficiency and the beginning of HPN seems to ensure a better constancy of the bone mineral density (odd ratio: 5,8 ; P value = 0,19). This trend seems similar when the duration of HPN was short (odd ratio: 4,5 ; P value = 0,16). CONCLUSION: An adequate follow-up of patients under HPN ensures a stabilisation of bone mineral density on the long range. 951

compared to TPN group, *p<0.05,**p<0.01, ***p<0.001

Combined Effects of Angiotensin II Receptor Blocker and Metformin on Suppression of Intestinal Polyp Formation in MIN Mice Akinori Yanaka, Yoshie Hifumi, Gen Fujii, Masafumi Yamamoto, Katsuya Nakano, Misato Shimura, Michihiro Mutoh Background and Aims: Obesity causes Type-2 diabetes and hypertension, and it has been demonstrated that obesity has a positive association with colorectal cancer risk. Recently, epidemiological studies and basic animal studies revealed that obesity-associated cancers involve dysregulated expression of hormones and cytokines, such as adiponectin, insulin, PAI-1, IL-6 and so on. On the other hand, metformin, an anti-diabetes drug, and angiotensin II receptor blockers (ARBs), such as losaltan, used as anti-hypertension drugs, have shown suppressive effects on cancer risk in clinical experiments. However, the effects of ARB plus metformin on intestinal tumorigenesis have not been demonstrated yet. Thus, we investigated the combination effects on development of intestinal polyp in Apc-deficient mice. Methods: Male Min mice, Apc-deficient mice, were administrated with 100 ppm losartan (LS) in drinking water, 1000 ppm metformin (MT) in diet (AIN-76A), or LS+MT at doses of 100 and 1000 ppm each for 9 weeks. At the age of 15 weeks, the number and size of intestinal polyps in the Min mice with each treatment were examined. Levels of adipocytokines, especially IL-6, and phosphorylation of STAT3 were also evaluated to clarify the effect of IL-6 on intestinal tumor development. Results: Treatment with LS, MT, or LS+MT changed the total number of polyps to 99.5%, 102.9% and 63.2% (p < 0.01) of the untreated value, respectively. Serum levels of IL-6 in the untreated-control mice were 20.6±7.0 pg/ml, and the levels in LS, MT, and LS+MT treated mice were 15.4±7.1 (mean±SD), 24.1±8.1 and 12.6±2.0 pg/ml (p < 0.05 vs untreated), respectively. Other cytokines were not clearly changed. IL-6 mRNA expression levels in liver tissue were significantly suppressed by LS or MT treatment, and were additionally suppressed lower than 25% (p < 0.01) of the untreated value by LS+MT treatment. IL-6 mRNA expression levels in non-tumorous parts and tumorous parts of small intestine were not suppressed by LS, MT or LS+MT treatment. Thus, we confirmed the expression of IL-6 receptor and STAT3 activation (Y705 phosphorylation) by immunohistochemistry and find that STAT3 was clearly activated only in tumorous parts of the small intestine. Conclusion: We clearly demonstrated the suppressive effect of LS and MT on intestinal polyp development in Min mice. The mechanism involved in the suppressive effect of combination treatment on intestinal polyp formation in Min mice was partly explained by synergistic suppression of IL-6 and resultant inactivation of STAT3. These data suggest that ARB and MT could be useful as chemopreventive agents against obesity-associated colorectal cancer.

Phylum level analysis after ribosomal database project (RDP) classification of 454-pyrosequencing of ileal and colon mucosa-associated bacteria samples. Percentages of most common phyla are shown in the graph. There were significantly fewer Firmicutes ,more proteoobacteria ,Verrucomicrobia and Bacteroidetes in the TPN samples compared to the control samples. Refeeding 24 hours increased Firmicutes and decreased Proteobacteria and Bacteroidetes percentages compared to TPN samples, but also led to expansion of an undescribed portion of the microbiome 949 Is Standard One Vial Multivitamin Adequate for Vitamin D Sufficiency in Home Parenteral Nutrition Patients? Edwin K. McDonald, Carol E. Semrad, Elizabeth A. Wall Background: Vitamin D is important in bone homeostasis and may play a role in cancer, cardiovascular disease, and autoimmune disease including diabetes mellitus. Recent studies suggest that 25-hydroxy vitamin D levels greater than 30 ng/ml offer protection from these conditions. U.S. adult standard one vial multivitamin preparations typically contain 200 IU of vitamin D3. Patients on home parenteral nutrition (PN) are dependent on multivitamin additives to provide vitamin D supplementation. It is not known whether standard one vial multivitamin preparations are adequate to achieve sufficient vitamin D levels. Aim: To evaluate vitamin D status in a diverse population of patients receiving home PN. Methods: A retrospective chart review was performed on patients receiving home PN for greater than


AGA Abstracts

AGA Abstracts

TPN, only TGFβ expression recovered with refeeding. Pyrosequencing data showed that TPN led to significant changes in mucosally-associated intestinal microflora; refeeding partially recovered this TPN-associated microflora shift. However, refeeding led to increased numbers of unclassified strains. Lysozyme expression was down-regulated in TPN group vs controls, along with a down-regulation of cryptins. Refeeding did not affect either change. TER decreased 50% in TPN mice and refeeding led to a partial recovery. Passage of FITC-Dextran was increased with TPN; and refeeding partially reversed this change. Refeeding led to a striking loss of tight junction proteins ZO-1 and Occludin vs. TPN administration, and a marked increase in the pore-forming Claudin-2. Conclusions: Refeeding after TPN administration partially recovered both TPN-associated loss of EBF and changes in the microbiome. However, refeeding led to an increase of several pro-inflammatory factors, and did not reverse the TPN-associated loss of paneth cell function. These refeeding effects may contribute to feeding intolerance and infectious complications. The process of refeeding patients may be more complex than previously perceived. summary data