Tube Feed Necrosis after Major Gastrointestinal Oncologic Surgery: Institutional Lessons and a Review of the Literature.

BACKGROUND: Small bowel necrosis after enteral feeding through a jejunostomy tube (tube feed necrosis, TFN) is a rare, serious complication of major abdominal surgery. However, strategies to reduce the incidence and morbidity of TFN are not well established. Here, in the largest series of TFN presented to date, we report our institutional experience and a comprehensive review of the literature. METHODS: Eight patients who experienced TFN from 2000 to 2014 after major abdominal surgery for oncologic indications at the University of Cincinnati were reviewed. Characteristics of post-operative courses and outcomes were reviewed prior to and after a change in tube-feeding protocol. The existing literature addressing TFN over the last three decades was also reviewed. RESULTS: Patients with TFN ranged from 50 to 74 years old and presented with upper gastrointestinal tract malignancies amenable to surgical resection. Six and two cases of TFN occurred following pancreatectomy and esophagectomy, respectively. Prior to TF protocol changes, which included initiation at a low rate, titrating up more slowly and starting at one-half strength TF, three of six cases of TFN (50%) resulted in mortality. With the new TF protocol, there were no deaths, goal TF rate was achieved 3 days later, symptoms of TFN were recognized 3 days earlier, and re-operation was conducted 1 day earlier. CONCLUSION: This case series describes a change in clinical practice that is associated with decreased morbidity and mortality of TFN. Wider implementation and further refinement of this tube-feeding protocol may reduce TFN incidence at other institutions and in patients with other conditions requiring enteral nutrition.

Surgery induces human mononuclear cell arginase I expression.

BACKGROUND: Arginase is a metabolic enzyme for the amino acid arginine that participates in the immune response to trauma. We hypothesize that surgical trauma induces arginase expression and activity in the human immune system. METHODS: Peripheral mononuclear cell (MNC) arginase activity and expression and plasma nitric oxide metabolites and interleukin (IL)-10 were measured in patients undergoing elective general surgery. Twenty-two healthy volunteers served as a comparison population. RESULTS: MNC arginase activity increased within 6 hours of surgery (p < 0.05) and coincided with increased arginase I protein expression. Plasma nitric oxide metabolites decreased significantly postoperatively (p < 0.05). Patients lacking an elevation in IL-10 failed to demonstrate increased MNC arginase activity. CONCLUSION: Increased MNC arginase expression may contribute to postsurgical immune dysfunction by affecting arginine use and availability and nitric oxide metabolism in the immune system. Plasma IL-10 may play a role in regulating MNC arginase activity.

Alterations in arginine metabolic enzymes in trauma.

Arginine is the sole substrate for nitric oxide (NO) synthesis by NO synthases (NOS) and promotes the proliferation and maturation of human T-cells. Arginine is also metabolized by the enzyme arginase, producing urea and ornithine, the precursor for polyamine production. We sought to determine the molecular mechanisms regulating arginase and NOS in splenic immune cells after trauma. C3H/HeN mice underwent laparotomy as simulated moderate trauma or anesthesia alone (n = 24 per group). Six, 12, 24, or 48 h later, 6 animals from each group were sacrificed, and splenectomy was performed and plasma collected. Six separate animals had neither surgery nor anesthesia and were sacrificed to provide resting values (t = 0 h). Spleen arginase I and II and iNOS mRNA abundance, arginase I protein expression, and arginase activity were determined. Plasma NO metabolites (nitrite + nitrate) were also measured. Trauma increased spleen arginase I protein expression and activity (P = 0.01) within 12 and for at least 48 h after injury and coincided with up-regulated arginase I mRNA abundance at 24 h. Neither arginase II nor iNOS mRNA abundance in the spleen was significantly increased by trauma at 24 h. Plasma nitrite + nitrate was decreased in animals 48 h post-injury compared to anesthesia controls (P < 0.05). Trauma induces up-regulation of arginase I gene expression in splenic immune cells within 24 h of injury. Arginase II is not significantly up-regulated at that time point. Arginase I, rather than iNOS appears to be the dominant route for arginine metabolism in splenic immune cells 24 h after trauma.

Arginase I expression and activity in human mononuclear cells after injury.

OBJECTIVE: To determine the effect of trauma on arginase, an arginine-metabolizing enzyme, in cells of the immune system in humans. SUMMARY BACKGROUND DATA: Arginase, classically considered an enzyme exclusive to the liver, is now known to exist in cells of the immune system. Arginase expression is induced in these cells by cytokines interleukin (IL) 4, IL-10, and transforming growth factor beta, corresponding to a T-helper 2 cytokine profile. In contrast, nitric oxide synthase expression is induced by IL-1, tumor necrosis factor, and gamma interferon, a T-helper 1 cytokine profile. Trauma is associated with a decrease in the production of nitric oxide metabolites and a state of immunosuppression characterized by an increase in the production of IL-4, IL-10, and transforming growth factor beta. This study tests the hypothesis that trauma increases arginase activity and expression in cells of the immune system. METHODS: Seventeen severely traumatized patients were prospectively followed up in the intensive care unit for 7 days. Twenty volunteers served as controls. Peripheral mononuclear cells were isolated and assayed for arginase activity and expression, and plasma was collected for evaluation of levels of arginine, citrulline, ornithine, nitrogen oxides, and IL-10. RESULTS: Markedly increased mononuclear cell arginase activity was observed early after trauma and persisted throughout the intensive care unit stay. Increased arginase activity corresponded with increased arginase I expression. Increased arginase activity coincided with decreased plasma arginine concentration. Plasma arginine and citrulline levels were decreased throughout the study period. Ornithine levels decreased early after injury but recovered by postinjury day 3. Increased arginase activity correlated with the severity of trauma, early alterations in lactate level, and increased levels of circulating IL-10. Increased arginase activity was associated with an increase in length of stay. Plasma nitric oxide metabolites were decreased during this same period. CONCLUSIONS: Markedly altered arginase expression and activity in cells of the human immune system after trauma have not been reported previously. Increased mononuclear cell arginase may partially explain the benefit of arginine supplementation for trauma patients. Arginase, rather than nitric oxide synthase, appears to be the dominant route for arginine metabolism in immune cells after trauma.

Beta adrenoceptor regulation of macrophage arginase activity.

BACKGROUND: Arginase, which metabolizes L-arginine within the urea cycle, is essential for production of polyamines and affects production of nitric oxide by depletion of L-arginine, the common substrate for both arginase and nitric oxide synthase. Having shown that trauma increases splenic macrophage arginase activity, we seek to define the mechanisms for this. RAW macrophage arginase activity and expression are increased by 8-bromo-cAMP in vitro. We hypothesize that since catecholamines increase cAMP, trauma-induced splenic arginase activity may be mediated by post-injury catecholamine release. METHODS: RAW 264.7 macrophage arginase activity was measured in vitro in response to 4 catecholamines with or without propranolol or lipopolysaccharide (LPS). C57BL/6 mice underwent laparotomy as a model of moderate trauma after propranolol treatment, with and without intraperitoneal Escherichia coli LPS administration as a simulated pro-inflammatory stimulus. RESULTS: Macrophage arginase activity increased in vitro in response to catecholamines or LPS (P < .05). Propranolol pretreatment blocked macrophage arginase activity induced by epinephrine (10 mumol/L) in vitro (P < .05). Trauma or LPS alone increased splenic arginase activity in vivo (P < .05). Propranolol did not alter LPS-induced splenic arginase activity but did significantly reduce trauma-induced splenic arginase activity (P < .05). CONCLUSIONS: Catecholamines alone increase macrophage arginase activity through beta-adrenoceptor activation. Increased splenic arginase activity induced by moderate trauma is decreased by beta-adrenoceptor blockade, suggesting that trauma-induced arginase activity is partly mediated by endogenous catecholamines.

Trauma increases extrahepatic arginase activity.

BACKGROUND: Although expressed primarily in the liver, arginase activity also is present in extrahepatic tissues and specifically in macrophages, where it may play diverse physiologic roles in wound healing, cellular proliferation, and the regulation of nitric oxide production. Arginase activity in immune cells is upregulated by certain cytokines such as IL-4, IL-10, and TGF-beta and by catecholamines. Since the release of these substances is increased after trauma, we hypothesized that arginase activity would also be increased in immune cells after trauma. The current work tests this hypothesis. METHODS: A model of surgical trauma was created in C3H/HeN mice by performing an exploratory laparotomy. Tissue arginase activity and arginase I protein expression were determined. As a control, arginase activity and expression were also stimulated with the use of endotoxin. In addition, we evaluated the expression of inducible nitric oxide synthase and the accumulation of nitric oxide metabolites in plasma. RESULTS: Surgical trauma was associated with a significant increase in arginase activity in splenic and renal tissues (P < .05). Splenic macrophages from trauma animals exhibited arginase activity levels approximately 10 times those of controls (P < .05). Endotoxin alone increased arginase activity in the spleen, but this increase was less than that of trauma alone (P < .05). Arginase activity remained elevated after trauma for up to 4 days and normalized by day 7. Arginase I expression was upregulated by trauma in both splenic and renal tissue and by endotoxin in the spleen only. Despite upregulation of inducible nitric oxide synthase in trauma animals, circulating nitric oxide metabolites were decreased 2 days after trauma compared with controls (P < .05). Endotoxin-induced nitric oxide metabolites were also reduced in trauma animals compared with endotoxin treatment alone (P < .05), but this normalized by day 4. CONCLUSIONS: Extrahepatic arginase expression and activity is increased after trauma and may provide the necessary precursors for cellular proliferation and repair or may play a regulatory role in the production of nitric oxide.

Autoradiographic localization of cholecystokinin (CCK) receptor expression during the development of azaserine-induced rat pancreatic carcinoma.

The peptide hormone cholecystokinin (CCK) has been shown to stimulate the growth of azaserine-induced preneoplastic nodules in the rat pancreas. Previously, our labortory demonstrated by classical binding studies that CCK receptors are overexpressed in azaserine-induced rat pancreatic neoplasms. In the present study, we utilized autoradiography to determine the temporal course of this increased receptor binding. Male Lewis rats were given azaserine or saline injections and sacrificed at 2, 4, 8, 12, and 18 months of age. Pancreatic tissue was harvested and autoradiography using 125l-labeled. CCK-8 was performed. Densitometry measurements of azaserine-induced pancreatic nodules, internodular pancreas, and normal pancreatic tissue (from saline-treated controls) of each age group were taken with an image analyzer. There was no statistically significant difference in CCK binding to internodular pancreas and normal pancreas at any age. At 2 months of age, there was no significant increase in CCK binding to azaserine-induced pancreatic nodules. However, at 4, 8, 12, and 18 months of age there was significantly greater CCK binding to azaserine-induced pancreatic nodules than to both internodular pancreas and normal pancreas (p < 0.001 for all groups). At 18 months of age, one azaserine-treated animal developed a pancreatic acinar cell carcinoma, which likewise exhibited significantly greater CCK binding than internodular pancreas or normal pancreas (p < 0.001 for both). These findings demonstrate increased CCK binding in azaserine-induced preneoplastic pancreatic nodules and pancreatic acinar cell carcinoma, compatible with our previous demonstration of receptor overexpression in these tissues. Increased CCK binding first becomes apparent by 4 months following exposure to azaserine. These result suggest that overexpression of CCK receptors, located specifically on preneoplastic and neoplastic pancreatic lesions, results in increased CCK binding and is involved in the mediation of CCK-stimulated growth during azaserine-induced pancreatic carcinogenesis.

Gastrin receptor expression during azaserine-induced rat pancreatic carcinogenesis.

The hormone gastrin is thought to stimulate the growth of certain pancreatic carcinoma cell lines. We have previously detected the presence of the gastrin receptor in rat pancreatic carcinoma cell lines but not in normal rat pancreas. We had not, however, previously demonstrated that gastrin receptor is expressed in pancreatic carcinomas developing in the rat in vivo. Therefore, in the present study, we examined rat pancreatic tissue at various stages in azaserine-induced pancreatic carcinogenesis for gastrin binding and for the presence of gastrin receptor mRNA to determine the temporal expression pattern of the gastrin receptor during the in vivo development of pancreatic cancer. Autoradiography of pancreatic tissue using (125)I-gastrin-17-I from all azaserine-treated and control animals at 2, 4, 8, and 12 months of age demonstrated no specific gastrin binding. At 18 months of age, normal pancreas, azaserine-induced premalignant pancreatic nodules, and internodular pancreas demonstrated no specific gastrin binding. One of three azaserine-treated animals developed an area of pancreatic acinar cell carcinoma at 18 months of age which exhibited significant specific gastrin binding of 141.8 - 32.8 fmole/gm of tissue. Southern blot analysis of pancreatic RNA isolated from animals at 12 months of age revealed no gastrin receptor mRNA; however, by 18 months of age, gastrin receptor mRNA was present in all azaserine-treated animals but absent in control animals. In summary, specific gastrin binding is present in in vivo azaserine-induced pancreatic acinar cell carcinoma but absent in normal pancreas and azaserine-induced premalignant pancreatic nodules. Gastrin receptor mRNA is first expressed in azaserine-treated rat pancreas at some point between 12 and 18 months of age. These results demonstrate that expression of gastrin receptor is altered in azaserine-treated rat pancreas and may play a role in the development of pancreatic cancer.