Nutrisi in Gastric Ulcer

Nutrition in Clinical Practicehttp://ncp.sagepub.com/ When to Feed the Patient With Gastrointestinal Bleeding Stephen A. McClave and Wei-Kuo Chang Nutr Clin Pract 2005 20: 544 DOI: 10.1177/0115426505020005544 The online version of this article can be found at: http://ncp.sagepub.com/content/20/5/544 Published by: http://www.sagepublications.com On behalf of: The American Society for Parenteral & Enteral Nutrition Additional services and information forNutrition in Clinical Practice can be found at: Email Alerts: http://ncp.sagepub.com/cgi/alerts Subscriptions: http://ncp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav >> Version of Record - Oct 1, 2005 What is This? Downloaded from by guest on January 26, 2013 ncp.sagepub.com McClave.edu. Taiwan ABSTRACT: Whether to provide artificial enteral nutrition therapy to a patient with evidence of gastrointestinal bleeding (GIB) creates a difficult clinical dilemma. evidence of GIB is not an automatic contraindication to further enteral feeding. Kentucky 40202. The clinical significance of GIB is variable. 0884-5336/05/2005-0544$03. the etiology of Correspondence: Stephen A. National Defense Medical Center.00/0 Nutrition in Clinical Practice 20:544-550. Depending on the etiology of the GIB. Concern that enteral feeding may contribute to the morbidity associated with GIB leads to delays in initiating enteral therapy or to cessation of feeding in the patient in whom artificial nutrition support has already been started. aspirate to clinically significant GIB with hemodynamic compromise. In other patients.. Taipei. Louisville. MD*. Surprisingly. actually increase risk for rebleeding in other patients. or serve as a moot point with no relation to further bleeding or morbidity in still other patients. the etiology of the GIB is such that enteral feeding may protect the patient and reduce the likelihood for continued bleeding. Jackson St. MD† *Departments of Medicine. 550 S. and †Tri-Service General Hospital. ranging from innocuous guaiac-positive coffee-ground gastric . Louisville. Kentucky. enteral nutrition may protect the gut mucosa and reduce further bleeding in some patients. The nutrition support specialist needs a full understanding of the physiology behind the varying diagnoses for GIB to know whether feedings can be initiated or continued or whether enteral feedings need to be withheld for 48-72 hours until risk for rebleeding and further morbidity is minimized. Professor of Medicine. Division of Gastroenterology/Hepatology. University of Louisville School of Medicine. In some patients.Clinical Dilem m a When to Feed the Patient With Gastrointestinal Bleeding Stephen A. The decision to start enteral feeding or continue feeding once evidence of GIB develops is based on the etiology of the bleed. McClave. an endoscopic evaluation is needed to distinguish the differential etiology of the GIB. Electronic mail may be sent to samcclave@louisville. In many cases. causing them to stop enteral feeding or withhold initiation of feedings in the first place. October 2005 Copyright © 2005 American Society for Parenteral and Enteral Nutrition 5 4 4 Evidence of gastrointestinal bleeding (GIB) is a common fear factor for nutrition support specialists. MD. and Wei-Kuo Chang. and patients are more likely to require blood transfusion. The pathophysiologic mechanism of acute variceal bleeding involves Downloaded from ncp. Over 20-30 erosions may be present. The etiology of the GIB in this setting invariably involves stress gastropathy. undergoes endoscopic intervention. involving peptic ulcer disease. the rest in the body and antrum of the stomach). The lesions in stress gastropathy are multiple diffuse erosions or very shallow ulcers. These lesions tend to be distal (the majority occurring in the duodenal bulb. Because the lesions are so numerous and so superficial throughout the gastric fundus. a different scenario involves the patient who presents to the emergency room with a massive acute upper GIB. The volume of bleeding is usually greater in this scenario. The pathophysiologic mechanism in peptic ulcer disease involves Helicobacter pylori infection or use of nonsteroidal antiinflammatory drugs. the differential diagnosis is wider. hypotension and hemodynamic instability are common. gastritis. the most common is either a gastric or duodenal ulcer. tend to be solitary and deep. septic patient receiving mechanical ventilation who develops GIB several days after admission to the ICU. esophagitis. performing endoscopy and using hemostasis techniques to stop the bleeding are not needed. In this latter patient. most commonly in the fundus of the stomach. Mallory-Weiss tear. On the other hand. the volume of bleeding may be greater. 2013 . There are usually 2 distinct scenarios for GIB in the intensive care unit (ICU) setting. and is then admitted directly to the ICU.sagepub. The pathophysiologic mechanism of these lesions is mucosal ischemia. Of the different etiologies.com by guest on January 26. these lesions are otherwise asymptomatic and are painless clinically. and usually results in a fairly low volume of bleeding. and clinically are more likely to be associated with abdominal pain. or acute variceal bleeding. tending to occur proximally in the gastrointestinal tract. One scenario involves the critically ill. In fact.the GIB is different. usually the clinical conditions and presentation are so typical that endoscopy is not required to confirm the diagnosis (the diagnosis may be assumed and supportive therapy may be initiated). Except for the bleeding. and providing enteral nutrition may be a liability jeopardizing the chances for hemostasis and increasing risk for rebleeding. For patients in the second scenario involving massive upper GIB. The enteral feeding can be started later only when risk from rebleeding has diminished. injection therapy. visible signs of blood. gastrointestinal bleeding. raising mortality dramatically. Understanding the difference between these 2 scenarios is imperative for the clinician to know whether or not it is safe to provide nutrition support by the enteral route (Table 1). visible vessel. cirrhosis and portal hypertension.October 2005 FEEDING THE GI BLEEDER 545 Table 1 Differentiating those patients who may be fed immediately from those for whom feedings should be delayed after presentation of GIB May usually be fed immediately Stress gastropathy Peptic ulcer disease with clean ulcer base or flat spot Mallory-Weiss tear Angiodysplasia Lower GI bleeds Feeds should be delayed for 48 h Peptic ulcer disease with adherent clot.sagepub. enteral feeding is safe. active ooze/spurter. The patient has to be stabilized first with regard to their bleed. enteral feeding may be temporarily contraindicated. 2. with no Downloaded from ncp. this same body of literature has shown that one of the best strategies for preventing GIB in the critically ill patient is to provide enteral feeding.3 “Occult” GIB refers to guaiac-positive stool or gastric aspirate. “Overt” bleeding is defined by retrospective.2 Amazingly. its use serves to protect the patient. More recent prospective studies have focused on the differentiation between any evidence of GIB and “clinically important hemorrhage” and have thus shown that the morbidity and mortality associated with GIB in the ICU is much less than previously thought. suggested that the development of any GIB in the ICU had a profound adverse effect on outcome.1. mechanical hemoclips.1 Concern over this issue led to excessive use of acidreducing agents. or rubber band ligation. Specific definitions of bleeding in the ICU have become very important. Endoscopy is required to establish the diagnosis (from among the numerous possibilities) and to achieve hemostasis through such techniques as thermocoagulation. Continuing to provide feedings in these patients may actually increase the likelihood for rebleeding. providing stress prophylaxis to almost any patient admitted to the ICU. For patients in the first scenario involving stress gastropathy.com Patients Who Develop GIB After Admission to the ICU The literature in the past. or lesion requiring electrothermal coagulation therapy Esophageal or gastric varices GIB. which was mainly . and continuing to provide feedings may reduce the likelihood for further bleeding. 1%. tachycardic. whereas the mortality in those without clinically important hemorrhage was only 9.4 Surprisingly.3 According to these definitions. it is not felt to be costby guest on January 26. 2.0%. superoxide radicals. the incidence of clinically important hemorrhage ranges only from 3. Increased permeability between the gastric epithelial cells. Reduction in blood flow to the gastric mucosa sets up a vicious cycle.3 When blood flow is restored to the gastric mucosa. 2 In a large study from the United States.3 As recent studies have shown that the incidence of clinically important hemorrhage is much less than previously thought. These factors lead to further reduction in gastric blood flow. and decreases in prostaglandin synthesis now afford the opportunity for further injury to the mucosal defect from acid present within the gastric lumen. the deaths in these patients with GIB were felt to be related to their underlying disease process and not to the GIB itself. Hemodynamic changes mean that the patient is hypotensive.5%.3 One report showed that full and aggressive volume resuscitation alone decreased the incidence of GIB to 0.7% to 6.4 The main factor in the pathophysiology of the GIB in these patients is stress gastropathy and mucosal ischemia. Reduced perfusion to the gastric mucosa results in a buildup of lactic acidosis and a drop in the intramural pH of the gastric wall. reduction in thickness and bicarbonate content of the mucous layer. The ability to stimulate blood flow to the gut explains why ultimately enteral feeding is such a good stress prophylactic agent against GIB.” however. whereas the incidence of overt GIB in the ICU ranges from 5% to 25%. The initial insult of hypoperfusion leads to the increased production of nitric oxide.2. the mortality in those patients with a clinically important hemorrhage was 48. The end result is a defect in the mucosal barrier. and a reduction in the release of cytoprotective prostaglandins. The need for transfusion usually is defined by a patient requiring 2 units of blood transfused.6%. more recent studies have shown that.3 In a Canadian multicenter trial (where patients in the ICU tend to have greater severity of critical illness than those patients in the United States). which increase the overall inflammatory response. 3 Reduction of blood flow to the gastric mucosa sets up the injury. is defined as overt bleeding plus either hemodynamic changes or need for blood transfusion. or orthostatic.the presence of hematemesis (bright-red blood or coffee grounds per nasogastric tube). mortality in those patients that demonstrated evidence of clinically important hemorrhage was 31%. 2013 . The restoration of blood flow flushes out the vascular bed. “Clinically important hemorrhage. the resultant reperfusion hyperemia may facilitate even greater injury. releasing a number of superoxide radicals and inflammatory cytokines. hematochezia (bright-red blood per rectum). or melena (black stool per rectum). com . infusion of enteral formula into the stomach is more effective than infusion distal to the pylorus.001). In one study. according to the alkalinity of the formulas. the practice of stress prophylaxis was one of the biggest financial drains on the hospital pharmacy.10 What may be much more important in preventing stress-induced gastropathy is the effect of enteral nutrition on intestinal blood flow. In the past. 20. the alkaline formula may buffer gastric acid. 51% of respondents indicated that they discontinued primary acid-reducing agents on initiation of enteral nutrition. Most of the studies would suggest that.8 and 1 study actually showed a decrease in the pH in response to enteral nutrition (or at least the pH was more difficult to control once enteral nutrition was started). In one study. and increased risk of aspiration pneumonia. Adverse events are associated with the use of acid-reducing agents. however.sagepub. most of the literature involving enteral nutrition and stress gastropathy focused on control of intragastric pH. likelihood for bacterial colonization of the stomach. Enteral nutrition formulas are alkaline.5 Early on.10 In contrast. respectively. their actual effect on intragastric pH is variable.1 studies showed that gastric intraluminal pH increased in response to infusion of enteral nutrition. infusion of formula into the duodenum kept the pH 4. Few clinicians. 6 Two studies.546 MCCLAVE AND CHANG effective for all patients in the ICU to receive acidreducing medications. In a recent survey. such as allergic reactions. p . whereas on the other hand. at which time an oxygen probe was placed in the cecal wall. enteral nutrition was more effective in raising the pH than histamine-2 (H-2) blockers.5 in 32% of 143 aspirates.5 in 88% of 196 aspirates (p .9 This variable effect from infusion of enteral nutrition may be related to the fact that Vol.001). 0. In a second Downloaded from ncp. Braga et al11 showed that mean intestinal tissue oxygen tension was significantly higher in the group randomized to enteral nutrition compared with those randomized to parenteral nutrition (43 mm Hg vs 31 mm Hg.5-7. 5 on one hand. showed no change in pH in response to infusion of enteral nutrition. However. 1 In a recent review by Maclaren et al. What is clear is that if the management for stress prophylaxis was solely focused on raising the pH. In a study of 257 patients operated on for cancer.10 However.0. 7. the luminal nutrients may stimulate further production of gastric acid. with a pH in the range of 5. proton pump inhibitors. the intraluminal pH should rise as a response to infusion of enteral formula. infusion into the stomach held the pH4.9% of respondents indicated that they used enteral nutrition as primary agents for stress prophylaxis. or antacid therapy. regard use of enteral nutrition as an effective stress prophylactic agent. No. However. though. the difference in the effect of postpyloric enteral nutrition to intragastric infusion was clinically negligible with regard to preventing clinically important hemorrhage. 17 Subsequently.73%. no acid-reducing therapy was used. 1 In a large prospective study of 1200 ICU patients. hemodynamic changes were monitored as patients began receiving enteral feeding.6 days was associated with a decrease in the incidence of GIB from 7. over the next 2 years. With sophisticated monitoring systems. 1 5 prospective randomized trials that compared acid-reducing agents to placebo showed no clinical benefit. increased significantly. those receiving enteral nutrition. and those receiving both.15. the timing of initiation of enteral nutrition affected the incidence of GIB. Cardiac output was essentially unchanged.8%) who were randomized to receive placebo in fact received enteral nutrition. Interestingly.5 to 4. post hoc analysis showed that continuous enteral nutrition was associated with a 70% reduction in the rate of GIB. Revelly et al12 studied postoperative cardiovascular patients who received a balloon pump for congestive heart failure after surgery. Reducing the time to initiation of enteral nutrition from 15. however. In the 1 prospective randomized trial that did show a benefit of acid-reducing agents to placebo. 17 The most impressive study was a by guest on January 26.8%) received enteral nutrition.3% for the subsequent 2 years with conversion from the acid-reducing therapy to enteral feeding (p .0% (p .1 In the recent review by Maclaren et al.05). Protection from the enteral feeding in controls may explain why there was no difference in incidence of GIB compared with study patients who received drug. Mean systemic blood pressure decreased from 75 mm to 70 mm Hg with the infusion of enteral nutrients.8%-72. this effect of enteral nutrients on increasing splanchnic blood flow may be a much bigger factor than its effect on pH in explaining its ability to prevent bleeding in the critically ill patient.16 In a large retrospective study over a 2-year period in burn patients.98% to 0. and the incidence of serious upper GIB was decreased to one-third (from 1. there was no significant difference in the incidence of GIB between 3 groups of patients. patients received an IV H-2 receptor blocker agent for stress prophylaxis while in the ICU. In a total of 562 patients. the incidence of upper GIB was reduced from 8.13 In a prospective study of 166 patients with spinal cord injury initially receiving a H-2 receptor blocking agent. those receiving acidreducing therapy. A surprisingly large volume of studies in the literature would suggest that enteral nutrition is a very effective stress prophylactic agent. 2013 . p NS) from the first to the second time period as well.study.6% to 2.3% during the first 2 years to 3.14 In a prospective randomized controlled trial of 181 burn patients. Splanchnic blood flow.12 With mucosal ischemia as the main pathophysiologic mechanism for stress-induced gastropathy. the fewest number of controls (3. and enteral nutrition was used instead as the primary agent for stress prophylaxis for patients while hospitalized in the unit. a large percentage of the controls in these studies (4.05). confidence intervals.October 2005 FEEDING THE GI BLEEDER prospective. multicenter trial by the Canadian Critical Care Trials Group which looked at 1770 critically ill patients in the ICU in an effort to identify those factors that increased the risk of GIB vs those factors that decreased risk. clinical shock.20 As use of enteral nutrition was substituted for the decreasing use of acid-reducing agents. In the 1405 patients with no risk factors.1%.001).3% to 19. if not better. the incidence of stress ulceration remained unchanged.com . In these patients. use of EN alone should provide adequate protection or prophylaxis against GIB.001). 19 In a prospective analysis in trauma patients after the introduction of an ICU protocol. risk of GIB is significantly greater according to the presence of well-defined risk factors. Cook et al 2 identified 2 major risk factors that 547 increase risk of bleeding in the ICU.18 It is not surprising then that in 2 recent surveys.2 Thus. In a certain subset of patients in the ICU. In a landmark study. Devlin et al 19 showed that the use of pharmacologic acidreducing agents was reduced significantly from 71% to 21% ( p .3%(although the difference did not meet statistical significance).7%. Critically ill Downloaded from ncp.08). the same investigator showed that the use of pharmacologic agents decreased from 70% to 26% (p . and consideration should be made to adding an acid-reducing medication. interesting trends have been demonstrated over the past few years between use of acid-reducing therapy and use of enteral nutrition for stress prophylaxis. According to the relative risk. In a retrospective analysis over a 4year period in a medical ICU involving almost 3000 patients. Interestingly. enteral feeding was at least as good.18 Two factors were found to decrease risk of bleeding: use of IV ranitidine and enteral tube feeding. A third risk factor. a general recommendation based on this study would be that stress prophylaxis provided by EN should be supplemented with an acid-reducing agent (preferably a proton-pump inhibitor) in the critically ill patient who has both evidence of coagulopathy and is receiving mechanical ventilation. at reducing risk of bleeding as the IV ranitidine. the incidence of clinically important hemorrhage was 3. Mechanical ventilation for 48 hours is associated with a 16fold increase in bleeding (p .001). the rate of GIB was shown to decrease from 25.2 In 847 patients with at least 1 positive risk factor. Use of enteral nutrition during this period increased from 51. the incidence of bleeding was only 0. One other group of patients that may require the combination of EN and acid-reducing agents is those patients with burns or head injury.20 In the majority of critically ill patients. provision of EN alone may not give sufficient protection against stress gastropathy and GIB. and p values. or the patient who shows signs of overt GIB.1% to 79.1%.001). just missed statistical significant as a factor associated with increased risk of bleeding (p .sagepub. and coagulopathy is associated with a fourfold increase in bleeding (p . and the risk of rebleeding is related to the presence or absence of a visible vessel. whereas a visible vessel that involves an actual knee or elbow of the vessel sticking up through the bed is associated with a 43% rebleeding rate.patients who have sustained burns or a head injury may have the additional component of a hypersecretory state and excessive acid production.21 Often when these patients are placed in the ICU. Patients Who Present With Acute GIB and Then Are Admitted to the ICU In the second scenario where a patient presents the emergency room with an acute upper GIB. the pathophysiologic mechanism most often is H pylori infection in 85% of cases. which provides injection of vasoconstrictive agents and electrocautery thermocoagulation. The Curling’s ulcers that develop in burn patients and the Cushing’s ulcers that develop in head-injury patients tend to be deep solitary lesions (as opposed to the multiple. 2013 . the potential etiologic factors are more variable. and an acid-reducing agent may be required in addition to provide adequate protection. whereas an ulcer with a flat spot (whether it be red. An active ooze is associated with a by guest on January 26.21 The other 25% of patients usually present with bleeding esophageal varices. factors that complicate their course are present.22 Less than 1% of these patients with upper GIB and ulcer disease will involve other factors such as Zollinger-Ellison syndrome. With bleeding peptic ulcer disease. such as coagulopathy and splanchnic ischemia. or blue) is associated with a 10% risk of rebleeding.22 An adherent clot attached to the ulcer bed is associated with a 22% risk of bleeding. In these patients. gastric ulcer. is evaluated by endoscopy. Endoscopic surveys in the past have shown that 75% of these patients have lesions that are acid related. and then admitted to the ICU. or viral infection. or esophagitis. with use of nonsteroidal antiinflammatory agents accounting for the remaining 10%-15% of cases. superficial erosions that occur in critically ill patients without burns and head injuries). The concept of a visible vessel refers to the situation where the ulcer has eroded down to an actual blood vessel (usually an artery) and that the presence of this vessel protruding up through the bed of the ulcer is associated with a high likelihood for rebleeding. These lesions tend to be solitary. black. Crohn’s disease. achieves hemostasis in 90% of these cases. Endoscopic therapy.22 It is important for nutritionists to understand the concept of visible vessel or bleeding stigmata. and to know how to differentiate those patients who can be fed immediately from those for whom feeding should be delayed. to appreciate the risk of further bleeding. Mallory-Weiss tear. stress prophylaxis with enteral feeding alone may be insufficient. systemic lupus erythematous. with equal distribution between gastritis. An ulcer with a clean base is associated with a 5% risk of rebleeding. and duodenal ulcer. has a tendency to lyse the clots. No. or active bleeding.25 It is important for the nutritionist to understand these concepts in order to know when it is safe to refeed the patient with a bleeding peptic ulcer. The period immediately after endoscopic therapy is somewhat tenuous and risky. 23 In a large prospective multicenter trial. and then 2. In contrast. These patients can be fed immediately after the endoscopic procedure.001).com .0 days.5 to 2.0-7. In general.7 units (p . and thus feedings may need to be withheld for 48 hours. 5 (omeprazole) keeps the pH 6.sagepub. whereas an active arterial spurter is associated with 65%-85% chance of continued bleeding. a significant decrease in the hospital length of stay from 5.0 on day 1. the impact of proton-pump inhibitors on reducing rate of rebleeding was significant in a different study by Lau et al. The remaining patients will have an adherent clot. and maintaining the pH 5 neutralizes almost all of the acid present in the stomach.0 for a full 3 days after initiation of the drug. it decreases to 3. and a significant reduction in the mean units of blood transfused from 3.7% ( p . 25 when given aggressively in high enough doses. use of this aggressive dosing of a proton-pump inhibitor was associated with a reduction in the rebleeding rate at 7 days from 5.548 MCCLAVE AND CHANG 55% rebleeding rate. Clotting is adversely affected by low pH and the presence of acid.7 by day 3.25 Compared with placebo.8% down to 1. aggressive therapy with an IV proton-pump inhibitor Vol. have a much higher rate of rebleeding. Pepsin is inactivated by a pH of 4.0. The proton-pump inhibitor omeprazole was given as an 80-mg IV bolus dose. followed by a continuous infusion at 8 mg/h for 72 hours (after which 20 mg a day were given for the next 18 days). very early development of tachyphylaxis to H-2 receptor blocking agents. secreted by the stomach.22 These lesions have such a low risk of rebleeding that they do not require electrocoagulation therapy.0 on day 2.0 to 4. Netzer et al 23 showed that there is a rapid.5. 20. Although the pH in response to IV H-2 blockers is 5. 62% of cases will present with either a clean ulcer base or a flat spot.3 The ability of various acid-reducing medications to maintain a high pH is variable. Raising the pH 6 is required to assure maintenance of the clot over a lesion once the bleeding has stopped. visible vessel. The risk of rebleeding after endoscopic therapy is about 20%. and thus require endoscopic therapy. and increasing the pH to 6 becomes important in stabilizing the clot over the visible vessel and reducing risk of further bleeding. 3 Clotting is optimized if the pH can be increased to a range of 5.04). Providing enteral nutrition during the first 4872 hours after endoscopic therapy may interfere with the ability of these acid-reducing medications to Downloaded from ncp. the likelihood for rebleeding is still significant.21 Control of gastric acid. Pepsin. there was no effect from H-2 blockers on reducing risk of morbidity from rebleeding.24 In contrast. In one study. and endoscopic therapy is rarely required. Varices bleed because of a buildup of pressure in the splanchnic by guest on January 26.25 These patients also may resume enteral feeding as soon as tolerated. and there is no need to delay refeeding in these patients. p NS). and these lesions do not always require endoscopic therapy.8 days vs 9. intragastric feedings might have less effect on decreasing pH through stimulation of gastric acid secretion than intrajejunal feedings (because of a buffering effect from the alkaline formulas).6 vs 3. In a small study of 26 patients with upper GIB who underwent injection therapy for peptic ulcer disease. Although theoretically. and hospital length of stay was reduced significantly in the early refeeding group (6. it is prudent to wait 48 hours before restarting feeding. respectively. conservative recommendations would be to start enteral feedings immediately in the patient with peptic ulcer disease who is found on endoscopy to have a clean ulcer base or flat spot. 2013 . In those patients with ulcer disease and a visible vessel who have undergone endoscopic hemostasis therapy. Surprisingly. de Ledinghen et al 26 randomized patients to early refeeding the day after endoscopic therapy vs delayed feeding to begin 3 days after injection therapy. develops into a higher-pressure system with dilated. cirrhosis. In contrast. Although this study would suggest that it is not only safe but actually beneficial for patients to receive enteral feeding soon after endoscopic therapy.3 units. Esophageal varices develop because of difficulty in venous drainage through a scarred cirrhotic liver. which is normally a low-pressure compliant system.01) compared.0 days. Thus. the number of units of blood transfused was less in the early refeeding group (2. specifically in patients with GIB due to peptic ulcer disease. most experts recommend waiting at least 48 hours after endoscopic therapy before initiating enteral feeding. and portal hypertension tend to have large-volume bleeds with an associated high morbidity and mortality. Other etiologies that account for patients with acute upper GIB include angiodysplasia and Mallory-Weiss tear. with a tear of the mucosa usually in the area of the gastroesophageal junction.control intragastric pH and continue to stabilize the clot over the visible vessel. patients who present with upper GIB from esophageal varices. the study is small and experts are reluctant to make widespread recommendations based on its results. tortuous veins. Risk of rebleeding after electrocautery therapy is fairly low. the effect by either route has not been studied well. Up to 95% of these lesions stop bleeding spontaneously. with the group receiving delayed refeeding. p . Thus. The portal venous system. Mallory-Weiss tears involve retching and vomiting. Angiodysplasias tend to be lowervolume venous bleeds (as compared with largervolume arterial bleeds associated with peptic ulcer disease).26 Only 1 patient in the delayed feeding group developed recurrent GIB (vs none in the early refeeding group). 27 Of concern was the fact that the rebleeding in the early-fed group occurred earlier.sagepub. NG tubes have to be removed to perform banding. Although numbers were too small to reach statistical significance. one would probably wait at least 48 hours in these high-risk patients (although there are no data in this specific situation). as the largevolume bleed that ensues decompresses the system. Above this pressure. who underwent sclerotherapy or band ligation (followed by infusion of octreotide). feeding should be delayed for at least 48 hours after endoscopic therapy for bleeding esophageal varices. The varices tend to bleed from 1 site on 1 varix. thought to Downloaded from ncp.27 Early refeeding may cause a shift in blood flow to the splanchnic circulation. again by de Ledinghen et al. the compression of the rubber band causes fibrosis and scarring at the site of the varix. If a TIPS (transjugular intrahepatic portosystemic shunt) were placed. Endoscopic management involves band ligation of these varices. the nutritionist needs to be familiar with these issues to know whether it is safe to feed the patient with acute variceal bleeding. Also. melena. There is reluctance to replace tubes right away for fear of knocking off the bands prematurely (but nasogastric tubes nonetheless may be placed gingerly at approximately 48 hours). If bleeding recurs and platelets are 25. platelets are not routinely given before endoscopic intervention. A banding device placed on the end of an endoscope flips a rubber band over the varix. hospital length of stay was delayed 2 days (14. Withholding enteral nutrition is rarely an issue for patients who present with lower GIB. and although risk for rebleeding is increased. Eventually. scar tissue is left behind.5 vs 12. Patients were randomized after endoscopic therapy to either early refeeding 1 day after banding or late refeeding in which feedings were not started until 3 days after banding. One small study. p NS). and mortality was slightly higher in the early group than the late group (25% vs 20%. the banded varix falls off. the rate of rebleeding was shown to be higher in the early group compared with the late group (33% vs 10%. and the varices are obliterated. then platelet transfusion may be used. The clinical presentation of lower GIB invariably involves hematochezia. p NS). mor- 549 tality in the early-fed patients was related to rebleeding (and aspiration and encephalopathy).October 2005 FEEDING THE GI BLEEDER circulation above 12 mm Hg. which could lead to increases in pressure and increased risk of rebleeding from the varices. Over a matter of days. Again. In a prospective study of 235 patients who presented with severe hematochezia.000. closer to the time of banding than the rebleeding that occurred in the late-refeeding group.com . and only in rare circumstances. 27 studied 22 patients over a year period that presented with acute variceal bleeding. Cirrhotic patients are usually thrombocytopenic due to hypersplenism. p NS). the varices will literally burst at a single point and lead to significant high-volume bleeding.9 days. For these reasons. With respect to the lower GIB itself. Patients with GIB for whom enteral nutrition should be held for at least 48 hours include those patients who present with an upper GIB and are found to have a peptic ulcer disease with visible vessel or those patients who present with esophageal varices. Rigaud D. Am J Respir Crit Care Med. polyps. 2. 22 The etiologies for the colonic bleeds included diverticulosis. Guyatt GH. Bonfils S. a definitive colonic source for the bleed was found in 75%. Liu C. controlled. Ann Intern Med. N Engl J Med. it is appropriate to continue enteral nutrition as long as it does not interfere with the patient preparation for colonoscopy. 2013 . 8. 2002. References 1. Bonten MJM. The evidence of a clinically important hemorrhage may warrant endoscopic evaluation. 4. Gaillard CA. Pathophysiology of the upper gastrointestinal tract in the critically ill patient: rationale for the therapeutic benefits of acid suppression. Enteral nutrition in patients receiving mechanical ventilation. ischemia. Comparison of omeprazole and ranitidine for stress ulcer prophylaxis. Hance AJ. Gaillard CA. 1986. Hinthorn DR. 1994. especially if the clinician is prone to withhold enteral nutrition in response to signs of any GIB. Am J Med. cancer. The role of intragastric acidity and stress ulcer prophylaxis on colonization and infection in mechanically ventilated ICU patients. et al. Fennerty MB. et al. Van Tiel FH.90:58-63. 2001. Fish DN.have a lower GIB. by guest on January 26. Smeets HGW. Prophylaxis for stress-related gastric hemorrhage in the medical intensive care unit: a randomized. Stobberingh EE. single-blind study. Van Der Geest S. For most lower GI bleeds. 1986. Use of enteral nutrition for stress ulcer prophylaxis. Ben-Menachem T.30(6 suppl):S351-S355. 9. Crit Care Med. and the decision to continue or withhold enteral nutrition usually depends on issues related to the patient preparation required before colonoscopy. Beysens AJ. 6.22 Providing enteral nutrition to these patients rarely affects the risk for rebleeding. Fogel R. Van der Geest S. 1994. Seelig CB.121:568-575. or withhold enteral nutrition is dependent on the specific etiology of the GIB. Chest. Effect of ranitidine and enteral feeding. Ann Pharmacother. Gilbert C.42:1255-1259. hemorrhoids. Dig Dis Sci. Chastre J. Crit Care Med. The decision to initiate. and angiomas. Continuous enteral feeding counteracts preventative measures for gastric colonization in intensive care unit patients. Risk factors for gastrointestinal bleeding in critically ill patients: Canadian Critical Care Trials Group. Levy MJ. Intragastric pH profile during acute respiratory failure in patients with chronic obstructive pulmonary disease.152:1825-1834. continue. MacLaren R.22:934-944. Accary JP. Van Tiel FH. 1994. Conclusions Evidence of GIB in the intensive care setting has variable clinical significance. Ranney JE. 3. Patel RV. Jarvis CL. et al.80:827-832. It is appropriate to continue enteral feeding in the patient suspected to have stress gastropathy or in that patient found to have peptic ulcer disease with a clean ulcer base or flat spot. enteral feeding is usually a moot point.330:377-381.35:1614-1623. 1997. rectal ulcer. 5. Cook DJ. Bonten MJM. Fuller HD. Robinson NJ. 1995. 7. Pingelton SK. Lucas CE. Gersbach P. Weigelt JA. 5 10. Solem LD. 1990. Braga M.27:540-547. Gianotti L. 12. Cook DJ. et al. Zuidema GD. Di Carlo V. Guyatt G.171:366-372. N Engl J Med. 11. 1997. 20. Cook R. Risk factors for clinically important upper gastrointestinal bleeding in patients requiring mechanical ventilation. Revelly JP. Kitler ME.27:2812-2817. 2001. Antacid therapy and nutritional supplementation in the prevention of Curling’s ulcer. Surg Gynecol Obstet. The value of early enteral nutrition in the prophylaxis of stress ulceration in the severely burned patient. Strate RG. Pagliarello J. Preventing post-operative acute bleeding of the upper part of the gastrointestinal tract. Downloaded from ncp. Berger MM. Early postoperative enteral nutrition improves gut oxygenation and reduces costs compared with total parenteral nutrition. 17. Hay A. Early metabolic and splanchnic responses to enteral nutrition in postoperative cardiac surgery patients with circulatory compromise. Cayeux C. Tappy L. Germann G.com . Surg Gynecol Obstet. Raff T. Enterline JP. Gentilini O. Valentine RJ.550 MCCLAVE AND CHANG Vol. Allo M. Intensive Care Med. Kuric J. Salis C. Does nasoenteral feeding afford adequate gastroduodenal stress prophylaxis? Crit Care Med. 13.23:313-318. 1999. Bounan KR. Turner WW. Ledgerwood AM. Cook DJ. Marshall JC.148:367-369. Crit Care Med.sagepub. 1979.27:140-145. 16. 14. Chiolero R. Hartmann B. Griffith L.14:599-601. Nutritional support: prophylaxis against stress bleeding after spinal cord injury.338:791-797. 1989. 1986. Burns.29:242-248. Fischer RP. A comparison of sucralfate and ranitidine for the prevention of upper gastrointestinal bleeding patients requiring mechanical ventilation. No. 18. Marshall J. Crit Care Med. 1998. Heyland DK. 2001. Parisis V. 15. et al. Paraplegia. 26. Jensen DM.94:351-357. Early feeding or enteral nutrition in patients with cirrhosis after bleeding from esophageal varices? A randomized controlled study. Beau P. by guest on January 26. 23. 1999. Cothell J. Devlin JW. Recent advances in the endoscopic diagnosis and therapy of upper gastrointestinal. Med Clin North Am.19:452-460. Zarowitz BJ. Beau P. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers.86:1319-1356. Acute upper gastrointestinal bleeding in critically ill patients: causes and treatment modalities. Langman MJ. Effect of repeated injection and continuous infusion of omeprazole and ranitidine on intragastric pH over 72 hours. Devlin JW. Ann Pharmacother. Ben-Menachem T. 30(6 suppl):S365-S368.343:310-316. Mannant PR. Kovacs TO. 2002. Am J Gastroenterol.32:869-874. 2002. Gaia C. 24. Stress ulcer prophylaxis in medical ICU patients: annual utilization in relation to the incidence of endoscopically proven stress ulceration. Sung JJ. Walt RP. Dig Dis Sci. 25. 27. et al. 1998. et al. 22. small intestinal. de Ledinghen V. and colonic bleeding.19. When should patients with bleeding peptic ulcer resume oral intake? A randomized controlled study. Lau JY. Peters MJ. Continuous intravenous famotidine for haemorrhage from peptic ulcer. de Ledinghen V. 1992. Sandoz M. Fogel RP. 1997. 1998. Netzer P. N Engl J Med.42:536-541. Mannant PR. 21. Freemantle NP.340:1058-1062. Crit Care Med. 2000. Lancet.22:282285. Pharmacotherapy. Ulep SK. Gastroenterol Clin Biol. et al. 2013 . Dulchavsky SA. 1999. Tyburski JG. Mann SG. Conrad SA. Claire KS. et al. Impact of stress ulcer prophylaxis guidelines on drug costs and frequency of major gastrointestinal bleeding. Lee KK. 20.