Journal of Pediatric Critical Care

P - ISSN : 2349-6592    |    E - ISSN : 2455-7099

Year : 2016 | Volume : 3 | Issue : 4 | Page : 52-62

Intensive Care Management of Upper Gastrointestinal Bleeding In Children:A Review

Veena Raghunathan*, Maninder Singh Dhaliwal**, Sakshi Karkra***, Abdul Elkadri****

* Consultant, Pediatric Intensive Care Unit, Medanta, The Medicity, Gurgaon
** Senior Consultant, Pediatric Intensive Care Unit, Medanta, The Medicity, Gurgaon
***Consultant, Pediatric Gastroenterology and Hepatology, Medanta, The Medicity, Gurgaon
**** Assistant Professor of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin

Correspondence Address:

Veena Raghunathan
Consultant, Pediatric Intensive Care unit, ICU 5, Medanta, the Medicity, Gurgaon – 122001, Phone number +91-9871221112, email id:
Received: 11-Oct-16/Accepted: 25-Oct-16/Published online: 22-Nov-16

Source of Funding:None Conflict of Interest:None


Abstract: Managing children having an upper gastrointestinal bleed (UGIB) can be anxiety-provoking for an intensivist, as the differential diagnosis can vary from a benign disorder to a life-threatening condition, with the potential for high fatality if the management is delayed. Common causes in children of UGIB as a presenting complaint in the emergency department include mucosal lesions and variceal hemorrhage. While in intensive care settings, UGIB is usually secondary and is common in critically sick children with risk factors like respiratory failure, shock, organ failure and trauma. Despite the varied presentation, the primary focus in a child with UGIB is resuscitation and stabilization followed by a diagnostic evaluation. Pharmacological methods in management of UGIB include vitamin K, acid suppression agents while patients with portal hypertension warrant special consideration where splanchnic vasoconstrictors may have a role. Emergency endoscopy in acute UGIB in children can be technically difficult and risky to the patient, and it must be performed only once the patient is adequately stabilized. In this review, an attempt has been made to discuss an intensivist approach to UGIB in children.

Key words : bleeding,Gastrointestinal,Pediatrics ,acute,portal hypertension

Acute upper GI bleeding (UGIB) is defined as gross bleeding from a source in the esophagus, stomach or duodenum, proximal to the ligament of Treitz (located at the duodeno-jejunal flexure). Gastrointestinal (GI) bleeding may be overt, occult or obscure. Overt UGIB usually presents as hematemesis or melena with or without hemodynamic compromise depending on the amount of bleeding. Hematochezia may occur rarely as the presenting sign in patients with extremely brisk UGIB. Occult bleeding is when there is no visible blood present. Obscure GI bleed is defined as bleeding from the GI tract that persists or recurs without any obvious etiology after initial diagnostic endoscopy and small bowel radiographs. It may be overt or occult. Children presenting to the emergency department with acute UGIB usually lead to a high level of panic and anxiety in the child, family and the treating physician. The causes of UGIB in children are varied and range from benign to life-threatening. Prompt recognition and treatment is essential to prevent morbidity and mortality in these cases.

Existing data indicates the incidence of UGIB in the emergency department to be < 1 %1,2 . It is more common within the PICU setting with the reported occurrence being 6-25% in critically ill children3,4. There is limited pediatric data related to mortality in UGIB in the emergency room setting, however, it appears to herald a bad prognosis in critically ill children who develop it. In a large prospective observational study, Cochran et al demonstrated that critically ill children with UGIB have significantly higher mortality (16% vs 1.3%) than those without UGIB3.

It is important to distinguish ‗true‘ gastrointestinal bleeding from other non-gastrointestinal causes of bleeding. Two important causes of non-gastrointestinal bleeding include: swallowed blood and food mimics. Swallowed blood from a maternal source, either at delivery or during breastfeeding, is commonly seen in neonates and small infants. This can usually be distinguished by taking a careful history, or by using the Apt-Downey test to distinguish neonatal from adult blood. In older children, bleeding from the nose, oral mucosal ulcers, palatal injury or sinusitis may also present as hematemesis. Detailed history and physical examination with a high index of suspicion can distinguish a gastrointestinal from a non-gastrointestinal source. Foods and drinks may mimic blood in vomitus/stool. These include licorice, fruit drinks, beets, cherries and drugs (phenytoin, rifampicin, iron, cefdinir) and bismuth containing products. Guiac/benzadine tests are helpful in the differentiation, but it must be noted that a false positive Guiac test can occur with the ingestion of tomatoes, red cherries, meat and turnips. The causes of UGIB in pediatric patients vary according to age (Table 1). In neonates, swallowed maternal blood or hemorrhagic disease of newborn should be considered. Reflux esophagitis is more common in infants and toddlers, whereas gastritis and ulcers are more common in older children and adolescents5,6. Broadly speaking, UGIB can be classified into variceal and non variceal bleeding, of which variceal bleeding predominates (40-95%) in India. In contrast, Western countries more commonly have UGIB caused by peptic ulcer disease, with H. pylori gastritis being found in up to 49% of children presenting with UGIB1,5,7. In a 10 year retrospective study of 447 children in Iran, the most common causes of UGIB were found to be erosive esophagitis, gastritis/erosions, esophagal varices and peptic ulcer disease6. Some important causes of UGIB are discussed briefly below:
Hemorrhagic and erosive gastropathy: Erosions and subepitheleal haemorrhages in gastric mucosa lead to UGIB in around 2-10% of patients. The most common precipitating factors are NSAID usage and stress8.
Peptic ulcer: Bleeding peptic ulcers may be primary (related to H. pylori, bile reflux) or secondary (related to NSAID usage, stress, drugs, corrosives etc). The incidence of bleeding ulcers appears to be decreasing over the last few years, and is probably due to decreasing H. pylori prevalence and increased use of acid suppressive medications. Peptic ulcers are most commonly found in the duodenum in children less than 10 years old. Hematemesis and perforation are usually found in secondary peptic ulcer disease9.
Variceal bleed and portal hypertension: Portal hypertension is defined as a Hepatic-Venous Pressure gradient > 5 mm Hg, and is the predominant cause of UGIB in India. In adults, the most common cause of portal hypertension is cirrhosis, whereas in children it is extrahepatic portal vein obstruction (EHPVO). In Indian children presenting with variceal bleeding, EHPVO has been noted in 87% of cases1. The mortality after index hematemesis in variceal bleeding is 30% and after recurrent variceal bleeding is as high as 70%.
Extra-hepatic Portal Vein Obstruction: The mean age of presentation is 5 to 6 years10. Usually, UGIB is massive and recurrent; however the risk of rebleeding after a major episode is lower than in cirrhosis 11. Despite massive bleeding, these patients do not develop encephalopathy. Other features include splenomegaly, absence of jaundice and normal liver function. It can be diagnosed by ultrasonography. Ascites may develop transiently following a bleed, however, persistent and massive ascites should prompt a search for co-existent cirrhosis. Also, when portal vein thrombosis occurs acutely, it is associated with the development of a progressive ascites along with acute abdominal pain (due to small intestinal ischemia or infarction) and melena if the thrombus extends into the superior mesenteric vein.
Chronic liver disease(CLD): In children, the major causes of chronic liver disease are biliary atresia, viral hepatitis and metabolic liver disease. Presentation is usually dominated by manifestations of the primary disease. While nearly 80 % of children with EHPVO experience at least one episode of UGIB, only 30-40% of children with CLD have upper gastrointestinal bleeds12.
Mallory –Weiss Tears : These occur at the gastroesophageal junction, primarily on the gastric side. Vomiting, retching or coughing is classically reported prior to hematemesis.
Foreign body ingestion: Button battery ingestion may rarely cause UGIB, but when found is associated with a high morbidity and mortality13.

Table 1: Etiology of UGIB according to age

Clinical Assessment:
The assessment in a patient presenting with UGIB, like in any acutely presenting child, must begin with the evaluation and stabilisation of the airway, breathing, and circulation (―The ABCs‖). Airway control is especially important, and intubation by rapid sequence intubation may be indicated in profuse vomiting or altered mentation to avoid the aspiration of blood which is associated with significant morbidity and mortality. Breathing should be assessed by respiratory rate, respiratory effort, chest rise and SPO2. Oxygen should be administered to a child with respiratory distress or palor. Hemodynamic instability may result in signs of shock. Tachycardia is the most sensitive indicator of blood loss in children14, with hypotension being a late finding. Other signs of shock include hypotension, delayed capillary refill, cold extremities and signs of end organ hypoperfusion such as change in mental status, oliguria, increased lactate. In such cases, volume resuscitation should be immediate. Once stabilised, detailed history and examination must be taken to evaluate the etiology of UGIB.

History and Examination:
Along with the ongoing hemodynamic stabilisation, specific details of the type and amount of bleeding must be obtained. Hematemesis is defined as the vomiting of blood. The color may vary from bright or dark red to brown or ―coffee-ground‖ appearance. This depends upon the amount of contact time between the hemoglobin and the acidic gastric environment, leading to oxidation of the iron molecule within the hemoglobin molecule. Hematemesis is the most common clinical presentation of UGIB in children15. Melena is the passage of dark tarry stool with a characteristic ‗sickeningly sweet‘ aroma. A history of melena is usually associated with more significant bleeding than in hematemesis15. The jet black color of melanotic stool is due to the action of bacteria on hemoglobin that has been converted to hematin or other hemochromes. However when an upper GI source bleeds very briskly, blood does not remain in the intestine long enough for melena to occur. For the presentation of melena to occur, it has been determined that there should be at least 50 ml of blood in volume which remains within the gastrointestinal tract for more than 6 hours16. Hematochezia is the passage of bright-red/ maroon-colored stools per rectum. It usually indicates a lower GI hemorrhage, though it can also be found in cases of profuse blood loss and rapid GI transit. The history of the amount of blood loss at the time of presentation is useful; however it must be kept in mind that the estimate of visualised blood loss by both lay people and healthcare professionals may be highly inaccurate. It has been found that parents overestimate blood loss while physicians and nurses tend to underestimate blood loss17. Chronicity of symptoms also determines the clinical presentation. A child with acute severe bleed will be more symptomatic whereas chronic slow GI bleed may be better tolerated. In the case of slow bleeding, loss of 13% of the total blood volume may occur with no compromise of hemodynamics18. The history also provides clues to the etiology of bleeding. A history of recent nasal and/or oral trauma points to a need for thorough nasopharygeal examination for source of bleeding. History suggestive of bleeding diathesis, GI disorders (such as Inflammatory Bowel Disease) or liver disease when present facilitate early initiation of disease- specific measures to help control bleeding. Usage of NSAIDS has been associated with a higher incidence of erosive gastritis and peptic ulcers in children5,19. Apart from NSAIDS, anti-platelet & anti-coagulation drugs also may contribute to increased risk of upper GI bleeding. Other clinical clues to diagnosis include:
1. Jaundice, ascites – Liver disease
2. H/o NSAID, steroids, tetracyclines – Medication induced gastritis
3. Pain – Peptic ulcer disease, Mallory Weiss
4. Painless profuse bleed, splenomegaly- Variceal bleed, Portal hypertension
5. Severe retching, emesis before hematemesis- Mallory Weiss
6. Mechanical ventilation, sick child- Stress gastritis
7. H/O aortic aneurysm/ surgery- Aortoenteric fistula

Physical Examination:
 General signs: Palor may indicate the severity of blood loss but may be absent in acute cases. The presence of ecchymosis may indicate an underlying bleeding disorder. The patient should be examined for icterus, spider nevi and other signs of liver disease. Thorough examination of ear/nose/ throat should be done to look for erosions/polyp and any other source of bleeding.
 Abdominal examination: Presence of hepatomegaly and/or spenomegaly may suggest liver disease and/or portal hypertension. Abdominal tenderness with or without mass raises suspicion of intussusception/ intestinal obstruction. Bowel peristalsis is best-assessed using auscultation. Usually in UGIB, bowel sounds are hyperactive. A rectal examination may help to identify any rectal varices or fissures that may indicate a lower GI source of bleeding.

Risk Stratification
Various risk stratification scores have been developed and validated in adults to classify patients being evaluated for suspected upper GI bleeding, but no similar tool exists for children. The existing adult clinical tools include Clinical Rockall Score (based on BUN, Hb, systolic BP, and co-morbities etc) and the Glasgow Blatchford Score (based on the severity of shock, diagnosis, co-morbities, and major stigmata of recent bleeding). Both can predict high and low risk for rebleeding and mortality20,21. Freedman et al attempted to identify predictors of significant UGIB (i.e. Hb drop > 2 g/dL within 48 hours, blood transfusion within 48 hours of presentation and need for endoscopy/surgical procedure to stop bleeding) in a cohort of 613 children. The investigators found that presence of any 1 of the following identified all children with clinically significant hemorrhage: melena, hematochezia, unwell appearance, or a moderate to large volume of fresh blood in vomitus. Other risk factors were older age (9.7 vs 2.9 years), tachycardia, and history suggestive of GI disease22.In the PICU setting, the risk factors identified to be associated with significant UGIB include: Respiratory failure requiring mechanical ventilation (especially high pressure settings), coagulopathy, organ failure, PRISM score ≥ 10, shock, long operative procedures >3 hours and trauma3,23.

Nasogastric Aspiration:
Nasogastric aspiration is the traditional tool used for years to stratify risk in GI bleeding. Presence of a positive nasogastric aspirate for blood is generally considered to be a good indicator of acute UGIB. Clear aspirate and/or lavage makes bleeding proximal to ligament of Treitz unlikely, though it does not rule it out. A large bleeding duodenal ulcer may not be detected by nasogastric aspiration, as it may be clear. Nasogastric tube insertion is usually the first procedure that is done in a patient of UGIB after initial hemodynamic and respiratory stabilization. Contrary to popular belief, the placement of a nasogastric tube does not appear to lead to bleeding by esophageal varices24. However, it is poorly tolerated and disliked by patients25. Nasogastric lavage can be performed by instilling 3- 5ml/kg of NS and aspiration of the effluent content. If the effluent cannot be aspirated, the nasogastric tube and/or the patient position should be adjusted or nasogastric tube checked for blockage due to large clots. Lavage is recommended with saline at room temperature, as iced or cold saline has been associated with the risk of iatrogenic hypothermia and a higher risk of rebleeding when stopped. In a retrospective review of 1190 patients, Luk et al found that positive nasogastric-tube aspirate findings were 93% predictive of an upper GI source of bleeding26. Thus nasogastric aspiration of blood can be used to guide the diagnostic approach: it predicts the presence of a high risk lesion and the need for urgent endoscopy, but it however has not been shown to have an effect on the length of stay, surgery or transfusion requirements21. Red blood in nasogastric tube (as compared to coffee-ground vomitus) and bright red stools appear to have worse prognosis27. Thus, nasogastric tube insertion and aspiration is a useful guide to confirm UGIB in most cases. It also helps to decompress stomach contents prior to endoscopy. However it is better to avoid nasogastric tube insertion after endoscopy and ligation of esophageal varices.
Laboratory Studies
Blood investigations:
In sick children, initial blood investigations should include blood gas, CBC, PT/PTT, renal and liver function tests, blood group & cross match. Patients with chronic slow bleeding may present with more severe anemia than patients with an acute bleed unless bleeding is massive. Thrombocytopenia may be associated with hypersplenism in cases of portal hypertension, or may indicate sepsis. Liver dysfunction leads to an abnormal coagulation profile (PT/INR) and in chronic cases, hypoalbuminemia. Liver enzymes may be elevated or normal in end stage liver disease, along with increased bilirubin levels. Prolonged PT/PTT may also indicate severe acute illness such as in sepsis or DIC. Patients with UGIB may have an elevated BUN/creatinine ratio as compared to lower GI bleeding. This is because RBC proteins are digested and re-absorbed by the small intestine. BUN/creatinine ratio ≥ 30 has been suggested to be useful to detect UGIB: previous studies have demonstrated specificity as high as 98% and sensitivity of 68.8% for detection of UGIB28. It has also been suggested as a predictor of blood transfusion requirement. However, the BUN/creatinine ratio may be elevated due to other factors such as pre-renal azotemia and other comorbities. In more recent studies which have included children, BUN/creatinine ratio has not been shown to be a good marker for the differentiation between an upper or lower GI source of bleeding29.

Stool / Vomitus guiac test:To confirm UGIB, stool can be tested for the presence of heme to identify a true GI source of bleeding, though the site of bleed requires further evaluation.

Conventional Angiography – This is performed by an intervention radiologist and is useful to diagnose difficult cases of obscure UGIB. It is ideally performed at time of active profuse bleed. As a bleeding rate of around 0.5 ml/min can be detected, it is more useful for an acute rather than a chronic UGIB. It additionally can detect vascular malformations even when no active bleeding is present30.

CT Angiography- CT has a reasonably good accuracy for the diagnosis of the source of UGIB. It‘s advantages are easy availability and is the least invasive technique. However, although it is good for diagnostic purposes, therapeutic intervention is not possible.

Meckel’s scan- In children, especially those presenting with painless hematochezia and normal endoscopic findings, scintigraphy and/or surgical exploration should be considered urgently to look for Meckel‘s diverticulum31.

Initial Management
As described earlier, assessment and stabilisation of the ABCs is the first step in the management of UGIB. Intubation and mechanical ventilation may be necessary cases where there is a risk of aspiration. Oxygen should be administered in cases of respiratory distress and/or pallor. In the case of shock, 2 large-bore intravenous lines should be placed. Isotonic fluid boluses of 20 ml/kg should be administered. If there is continued profuse bleeding, packed red blood cell transfusion may be required. Foley‘s catheterisation and hourly urine output monitoring should be done. Pending endoscopy, nasogastric tube insertion is helpful to guide management as is discussed above. The initial management steps are summarised in figure 1.

                                     Figure 1: Management: Ongoing UGIB

Transfusion guidelines
Acute profuse blood loss leads to compromised perfusion and poor oxygen delivery to tissues. Blood transfusion in such cases is essential. In UGIB, a restrictive transfusion strategy is recommended, with the recommended threshold of Hb requiring transfusion 7 g/dl. It has been demonstrated that this threshold significantly improves patient outcomes32.This is especially relevant in UGIB related to portal hypertension, where the Hb target should be 7-8 g/dl: more liberal blood transfusion has been shown to increase portal pressure and increase risk of rebleeding as well as increase the risk of mortality. Fresh frozen plasma (FFP) is generally transfused as part of massive transfusion protocol in any patient with exsanguination and severe hemorrhagic shock. However, there is much controversy regarding its role in UGIB in cirrhotic patients. This is mainly due to concerns for an excessive increase in plasma volume and consequently increased portal hypertension. FFP should probably not be administered to such patients solely with the aim of correction of coagulopathy, as PT/INR are not a good indicators of coagulability in cirrhotic patients32,33. Platelet transfusions should be considered in patients with active UGIB with thrombocytopenia, especially if the platelet count is <30 X 109/L. However, this should not delay endoscopy34. There is no role for Factor VIIa in variceal bleeding associated with cirrhosis.

Acid suppression therapy
In all children with UGIB, it is beneficial to start empirical treatment with proton pump inhibitors (PPI). PPI have been demonstrated to be more effective at lowering pH than H2 receptor blockers in adults. This has led to the preferred use of PPI over H2 receptor blockade in children. A recent meta-analysis demonstrated that PPI administration before endoscopy helped to reduce stigmata of recent/active bleeding and also requirement for endoscopy35. Apart from stress-ulcer bleeding, PPI use has been found to play a definite beneficial role in variceal bleeding. Gastritis is very often associated with portal hypertension in children, and thus PPI use has a role36. It also helps to decrease the size of post ligation ulceration37. Dosage recommendations for above acid suppression agents is given in Table 238. A recently concluded meta analysis stated that there is no clear benefit in use of continous PPI over intermittent PPI dosing39. The role of locally cytoprotective drugs like sucralfate, magnesium hydroxide in children is controversial. They may have complementary role and hence could be considered in ulcer related bleeding; however as they make endoscopic visualisation of lesions difficult, it may be worthwhile to wait until after endoscopic evaluation. 
Vasoactive drugs:
In UGIB due to portal hypertension, vasoactive drug therapy is recommended to decrease and/or control active bleeding. Vasoactive drugs used are mainly of two types: Vasopressin derivatives (vasopressin, terlipressin) and somatostatin derivatives (octreotide). These drugs are able to stop bleeding in up to 75-80% of cases40.

Octreotide is preferred in children due to a higher efficacy and fewer side effects. It is a synthetic somatostatin derivative with a long half life (1 ½ hours). It is believed to cause splanchnic vasoconstriction and inhibition of gastric secretion. Octreotide is administered as an infusion: loading 1 mcg/kg followed by 1-4 mcg/kg/h (maximum rate 50 mcg/h) for up to 48 hours.

In a retrospective study by Eroglu et al in patients with UGIB with and without portal hypertension, it stopped bleeding in 71% of patients with portal hypertension and 50% patients without portal hypertension41. Thus, even in cases with an unclear etiology and in cases of severe UGIB, octreotide could be considered along with acid suppression.

Vasopressin is a very potent vasoconstrictor, and hence carries a concern for peripheral, bowel and cardiac ischemia with its usage. The dosage is 0.002 to 0.005 units/kg/min. Due to the concern for ischemia, octreotide is preferred over vasopressin. However, it does play a role in severe ongoing UGIB which is unresponsive to other therapies. Terlipressin is another vasopressive derivative which has mainly been used in adults, with very few case reports of use in children. In the case of a patient presenting with shock on norepinephrine, specific splanchnic vasoconstrictor therapy with any of the above drugs should be considered in cases of suspected portal hypertension related UGIB. In case of variceal UGIB, vasoactive therapy when combined with endoscopic treatment is more efficiacious at stopping bleeding than endoscopic treatment alone. It can be continued for 3—5 days after endoscopic intervention to decrease chances of rebleeding42.

Beta blockers:In the case of variceal bleeding, Beta blocker therapy should be started for secondary prophylaxis. When a Beta blocker is introduced early after ligation & endoscopic treatment, it helps to prevent rebound portal hypertension and rebleeding. A recent meta-analysis of 23 studies found that combining endoscopic therapy and Beta blockers decreased rebleeding more than either intervention used alone43. The dose is 0.5–2 mg/kg/day in 2–4 divided doses, with the goal of 75% reduction in the heart rate.

Vitamin K
Vitamin K should be administered in the presence of coagulopathy and should be considered in all children with chronic liver, pancreatic/ intestinal disorders. The dose is 1-2 mg for infants and 5-10 mg for older children.

Patients with chronic liver disease who develop UGIB have high incidence of infections (up to 40%) which increases the risk of rebleeding and mortality44. Prophylactic antibiotic (3rd generation cephalosporin) administration in such cases is recommended for 5- 7 days.

Tranexamic Acid
Tranexamic Acid is an antifibrinolytic agent which has been used to enhance clot formation and thus reduce severity of UGIB. A recent Cochrane systemic review suggested that tranexamic acid may have a role in reducing mortality in UGIB, however its beneficial role is unclear when used along with anti-ulcer drugs and endoscopic methods45. Hence, additional randomised controlled trials are required to determine its role in serious or uncontrolled UGIB.
Upper GI endoscopy
Upper GI endoscopy has two important roles: 1. It is useful in establishing the etiological diagnosis and 2. plays a therapeutic purpose as well. The need and exact timing of endoscopy must be made in consultation with the pediatric gastroenterologist. Endoscopy is generally recommended to be performed within 24 hours of admission of a patient with suspected UGIB. This has been shown to have the highest yield and reduces blood product requirement, need for second look endoscopy and surgical interventions34,46. When variceal bleeding is suspected, endoscopy is recommended within 12 hours38. In cases of active bleeding with hemodynamic instability and severe anemia, urgent endoscopy is advantageous: however it should be done only once patient is adequately resuscitated. Prior to endoscopy, to facilitate emptying of stomach contents, nasogastric lavage or administration of prokinetic agents (erythromycin/ metoclopramide) is helpful.

Non variceal bleed
In the presence of Forest type Ia, Ib, IIa (high risk of rebleeding), endoscopic treatment should be performed (Table 3). Combined treatment (epinephrine + clips or thermal devices) is more efficacious47. In the case of IIB (adherent clot), dislodgement by irrigation or endoscopic treatment for a small sized clot can be performed. For endoscopic stigmata with low risk of bleeding, there is no need for any endoscopic treatment. Second look endoscopy (16 to 24 h after 1st endoscopy) is not routinely required, but may be performed when high-risk stigmata have been observed. However, no benefits in mortality have been demonstrated48.

Variceal bleeding
Endoscopic therapy of bleeding esophageal varices should be done by band ligation or sclerotherapy
1. EVL(Endoscopic Variceal Ligation): is done using multiband ligator
2. EST(Endoscopic Sclerotherapy): is done by the injection into (intravariceal) or around (perivariceal) varices of a sclerosant agent. Various sclerosants used are:
 Absolute Alcohol
 Polydoconol
 Sodium morrhuate
 Sodium tetradecylsulfate

Unsuccessful endoscopic treatment: Options:
Balloon tamponade:In ongoing bleeding, esophageal/gastric balloon tamponade can be considered when emergency endoscopy is not possible. It can also be considered after failure of endoscopic therapy pending radical treatment of portal hypertension. The success of the Sengstaken-Blakemore tube (with 2 balloons - esophagal + gastric) and Linton- Nachlas tube (one balloon for gastric varices) is variable- 40 to 90%49. Various problems with use include inhalational lung disease, esophagal ulceration and rupture. It should not be used for >24 hours. Another recent issue is that many providers have not used these interventions due to the rarity of bleeds requiring this level of intervention.

Angiography and selective arterial embolisation by a pediatric interventional radiologist can and should be considered for severe UGIB.

Consultation with a surgeonis also essential in cases of failed endoscopy. For variceal bleed, TIPS (transjugular intrahepatic portosystemic shunting) can be considered , this results in decompression of intrahepatic portal vein into hepatic veins . Portosystemic shunts can be considered in severe refractory cases50.

Key messages:
 UGIB is an alarming presentation in the pediatric population; timely recognition of severity and appropriate management are important to reduce morbidity and mortality.
 Pharmacological measures including PPI and vasoactive medications must be started promptly- these play a role in both non variceal and variceal bleeding.
 Ongoing bleeding must be replaced with appropriate blood products. Restrictive transfusion strategy is advocated.
 Endoscopy is important for both diagnosis and therapeutic intervention and is advantageous when done early (within 24 hours). In ongoing active bleed, it must be performed urgently as soon as patient is adequately stabilised hemodynamically.
 A multidisciplinary approach is needed to manage UGIB. The pediatric intensivist along with gastroenterologist, interventional radiologist and surgeon all play a vital role in management of life threatening bleeding.

1. Arora NK, Ganguly S, Mathur P, et al. Upper gastrointestinal bleeding: etiology and management. Indian J Pediatr 2002; 69:155-168
2. Chawla S, Seth D, Mahajan P, et al. Upper gastrointestinal bleeding in children. Clin Pediatr (Phila) 2007; 46:16-21
3.Cochran EB, Phelps SJ, Tolley EA, et al. Prevalence of, and risk factors for, upper gastrointestinal tract bleeding in critically ill pediatric patients. Crit Car Med 1992;20:1519- 1523
4. Lacroix J, Nadeau D, Laberge S, et al. Frequency of upper gastrointestinal bleeding in a pediatric intensive care unit. Crit Care Med1992; 20:35-42
5. Grimaldi-Bensouda L, Abenhaim L, Michaud L, et al. Clinical features and risk factors for upper gastrointestinal bleeding in children: a case-crossover study. Eur J Clin Pharmacol 2010; 66:831-837
6. Rafeey M, Shoaran M, Majidy H. Diagnostic endoscopy and clinical characteristics of gastrointestinal bleeding in children: a 10-year retrospective study. Iran Red Crescent Med J 2013;15:794-797
7. Boukthir S, Mazigh SM, Kalach N, et al. The effect of non-steroidal anti-inflammatory drugs and Helicobacter pylori infection on the gastric mucosa in children with upper gastrointestinal bleeding. Pediatr Surg Int 2010;26:227-230
8. Nithiwathanapong C, Reungrongrat S, Ukarapol N . Prevalence and risk factors of stress-induced gastrointestinal bleeding in critically ill children. World J Gastroenterol 2005;21:6839-6842
9. Laine L. Paterson W et al. Bleeding peptic ulcers. N. Engl J Med 1994; 331:717-727
10. Webb LJ, Sherlock S. The etiology, presentation and natural history of Extrahepatic portal venous obstruction.Q J Med 1979; 192:627-639
11. Martel C, Bologuesi M ,Bellen S,et al. Long Term follow-up study of adult patients with non-cirrhotic obstruction of portal system;comparision with cirrhosis patients. J Hepatol 1992;15:299-303
12. Bernard O, Alvarez F, Brinelle F,et al. Portal hypertension. Clin Gastroenterol 1985; 14:35-55
13. Brumbaugh DE, Colson SB, Sandoval JA, et al. Management of button battery-induced hemorrhage in children. J Pediatr Gastroenterol Nutr 2011;52:585-589
14. Colle I, Wilmer A, Le Moine O, et al. Upper gastrointestinal tract bleeding management: Belgian guidelines for adults and children. Acta Gastroenterol Belg 2011;74:45-66
15. Cleveland K, Ahmad N, Bishop P, et al. Upper gastrointestinal bleeding in children: an 11-year retrospective endoscopic investigation. World J Pediatr 2012; 8:123-128
16. Schiff L, Stevens RJ, Shapiron N, et al .Observations on the oral administration of citrated blood in man II. The effect on the stools. Am J Med Sci 1942;203:409-412
17. Tebruegge M, Misra I, Pantazidou A, et al. Estimating blood loss: comparative study of the accuracy of parents and health care professionals. Pediatrics 2009;124:e729-e736
18. Bhatia V, Lodha R. Upper gastrointestinal bleeding. Indian J Pediatr 2011;78:227-233
19. Kalyoncu D, Urganci N, Cetinkaya F. Etiology of upper gastrointestinal bleeding in young children. Indian J Pediatr 2009;76:899-901
20. Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper gastrointestinal haemorrhage. Lancet 2000;356:1318-1321
21. Rockall T, Logan R, Devlin H, et al. Risk assessment after acute upper gastrointestinal haemorrhage. Gut 1996;38:316-321
22. Freedman SB, Stewart C, Rumantir M, et al. Predictors of clinically significant upper gastrointestinal hemorrhage among children with hematemesis. J Pediatr Gastroenterol Nutr 2012; 54:737-743 23. Deerojanawong J, Peongsujarit D, Vivatvakin B, et al. Incidence and risk factors of upper gastrointestinal bleeding in mechanically ventilated children. Pediatr Crit Care Med 2009; 10:91-95
24. Huang ES, Karsan S, Kanwal F, et al. Impact of nasogastric lavage on outcomes in acute GI bleeding. Gastrointest Endosc 2011;74:971-980
25. Aljebreen AM, Fallone CA, Barkun AN. Nasogastric aspirate predicts high-risk endoscopic lesions in patients with acute upper-GI bleeding. Gastrointest Endosc 2004; 59:172–178
26. Sarin N, Monga N, Adams PC. Time to endoscopy and outcomes in upper gastrointestinal bleeding. Can J Gastroenterol 2009; 23:489-493
27. Frattaroli FM, Casciani E, Spoletini D, et al. Prospective study comparing multi-detector row CT and endoscopy in acute gastrointestinal bleeding. World J Surg 2009; 33:2209-2217
28. Urashima M, Toyoda S, Nakano T, et al. BUN/Cr ratio as an index of gastrointestinal bleeding mass in children. J Pediatr Gastroenterol Nutr 1991; 15:89-92
29. Kim KS, Kang CH, Kim JY. Availability of Blood Urea Nitrogen/Creatinine Ratio in Gastrointestinal Bleeding with Melena in Children. Pediatr Gastroenterol Hepatol Nutr 2015; 18(1): 30-38
30. Afshani E, Berger PE. Gastrointestinal tract angiography in infants and children. J Pediatr Gastroenterol Nutr 1986; 5:173-86.
31. Yau KK, Siu WT, Law BK, et al. Laparoscopy-assisted surgical management of obscure gastrointestinal bleeding secondary to Meckel's diverticulum in a pediatric patient: case report and review of literature. Surg Laparosc Endosc Percutan Tech 2005;15:374–377
32. Villanueva C, Colomo A, Bosch A , et al. Transfusion Strategies for Acute Upper Gastrointestinal Bleeding. N Engl J Med 2013;368:11-21
33. de Franchis R. Revising consensus in portal hypertension: report of the Baveno V consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol. 2010;53:762-768
34. Colle I, Wilmer A, Le Moine O, et al. Upper gastrointestinal tract bleeding management: Belgian guidelines for adults and children. Acta Gastroenterol Belg 2011;74:45-66
35. Sreedharan A, Martin J, Leontiadis GI, et al. Proton pump inhibitor treatment initiated prior to endoscopic diagnosis in upper gastrointestinal bleeding. Cochrane Database Syst Rev. 2010; 7:CD005415
36. El-Rifai N, Mention K, Guimber D, et al. Gastropathy and gastritis in children with portal hypertension. J Pediatr Gastroenterol Nutr 2007;45:137-140
37. Shaheen NJ, Stuart E, Schmitz SM, et al. Pantoprazole reduces the size of postbanding ulcers after variceal band ligation: a randomized, controlled trial. Hepatology 2005; 41:588-594
38. Boyle JT. Gastrointestinal bleeding in infants and children. Pediatr Rev 2008; 29:39-52
39. Sachar H, Vaidya K, Laine L. Intermittent vs continuous proton pump inhibitor therapy for high-risk bleeding ulcers: a systematic review and meta-analysis. JAMA Intern Med 2014; 174:1755-1762
40. Garcia-Tsao G, Bosch J. Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med 2010; 362:823-832
41. Eroglu Y, Emerick KM, Whitington PF, et al. Octreotide therapy for control of acute gastrointestinal bleeding in children. J Pediatr Gastroenterol Nutr 2004; 38:41-47
42. Besson I, Ingrand P, Person B, et al. Sclerotherapy with or without octreotide for acute variceal bleeding. N Engl J Med 1995; 333:555-560
43. Gonzalez R, Zamora J, Gomez-Camarero J, et al. Meta-analysis: combination endoscopic and drug therapy to prevent variceal rebleeding in cirrhosis. Ann Intern Med 2008; 149:109-122
44. Bernard B, Cadranel JF, Valla D, et al. Prognostic significance of bacterial infection in bleeding cirrhotic patients: a prospective study. Gastroenterology 1995; 108:1828–1834
45. Bennett C, Klingenberg SL, Langholz E, et al. Tranexamic acid for upper gastrointestinal bleeding. Cochrane Database of Systematic Reviews. 2014; (11): CD006640
46. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101–113
47. Laine L, McQuaid KR. Endoscopic therapy for bleeding ulcers: an evidence-based approach based on meta-analyses of randomized controlled trials. Clin Gastroenterol Hepatol. 2009;7:33-47
48. Tsoi KK, Chan HC, Chiu PW, et al. Second-look endoscopy with thermal coagulation or injections for peptic ulcer bleeding: a meta-analysis. J Gastroenterol Hepatol 2010; 25:8-13
49. Panes J, Teres J, Bosch J, et al. Efficacy of balloon tamponade in treatment of bleeding gastric and esophageal varices. Results in 151 consecutive episodes. Dig Dis Sci 1988; 33:454-459
50. Fox VL. Gastrointestinal bleeding in infancy and childhood. Gastroenterol Clin North Am 2000; 29:37–66