Pediatric Gastrointestinal Bleeding Workup

For upper GI bleeding, a nasogastric tube can be placed to confirm the presence of fresh blood and to evaluate the degree of active bleeding. If fresh or active bleeding is confirmed, esophagogastroduodenoscopy (EGD) can determine the source of upper GI bleeding in 90% of children when performed in the first 24 hours. Alternatively, colonoscopy identifies the cause of bleeding in 80% of children with lower GI bleeding.

A cohort study by Bose et al included 56 hospitalized infants with GI bleeding who underwent EGD and/or colonoscopy/flexible sigmoidoscopy. Seven of the endoscopies detected sources of bleeding (eg, gastric ulcers, duodenal ulcer, gastric angiodysplasia, esophageal varices, and an anastomotic ulcer); however, three patients in the study presented with a GI perforation soon after the procedure. ^[8]

In general, trace or small amounts of blood that are a one- or first-time occurrence are not of emergent concern. Children rarely require an extensive laboratory workup or invasive procedures, and parents can be advised to observe the child at home to see if these situations arise again.

Patients with substantial upper or lower GI bleeding, as determined from their history or examination, should receive a complete blood count (CBC), coagulation studies, and a chemistry panel. The CBC reveals anemia and thrombocytopenia.

A normal hematocrit may provide false reassurance regarding some children with hypovolemia and hemoconcentration.

Leukocytosis with increased bands may indicate an infectious etiology or complication responsible for the bleeding.

Elevated, abnormal prothrombin time indicates coagulopathy (ie, disseminated intravascular coagulation) or profound impairment of liver synthetic function.

A prolonged activated partial thromboplastin time indicates a hemophiliac patient or coagulopathy.

A chemistry panel may reveal a high blood urea nitrogen (BUN) level, suggesting an upper GI source that has had time to allow the body to reabsorb blood leading to a higher BUN level compared with a lower GI source.

For children who have tenderness in the right upper quadrant or a history suggestive of liver disease, aspartate aminotransferase and alanine aminotransferase enzyme levels may indicate hepatitis and increased risk of portal hypertension.

The history should also be used as a guide with regard to when fecal leukocytes, parasites, or cultures should be ordered, if an infectious etiology is suspected.

In cases of episodic or obscure bleeding, nuclear medicine radionucleotide studies, arteriography, and wireless video capsule endoscopy are used to assist in identifying the site of blood loss.

Radionuclear imaging with technetium-labeled red blood cells can be used to detect bleeding at a rate as low as 0.1 mL per minute. This technique is somewhat imprecise; however, it may direct localization for either selective angiography, suggest a need for video capsule endoscopy, or provide some direction for laparotomy search and resection, a notoriously difficult process for the control of GI bleeding.

Arteriography can be used to detect bleeding at a rate of 0.5 mL per minute and offers the advantage of providing treatment and diagnosis. The treatment consists of embolization and intra-arterial administration of vasoconstrictors.

When arteriography and nuclear scanning fail to diagnose or localize the cause of bleeding, further options remain, including repeat endoscopy and push enteroscopy (often also referred to as double balloon endoscopy or enteroscopy). ^[9]

In many cases, wireless video capsule endoscopy reveals the cause noninvasively, but its main disadvantage is the inability to collect tissue samples for biopsy examination. ^[10]

If all else fails, diagnostic laparoscopy and intraoperative endoscopy can be performed as means of last resort.

An age-specific discussion of diagnostic workups for GI bleeding follows.

Go to Imaging of Upper Gastrointestinal Bleeding and Imaging of Esophageal Varices for complete information on these topics.

Neonates

In most neonates with stress gastritis, the diagnosis is presumptive. If necessary, definitive diagnosis is made with upper endoscopy, demonstrating erythema, diffuse bleeding, erosions, or ulcerations of the gastric mucosa.

Upper gastrointestinal bleeding in children aged 1 month to 1 year

Diagnostic workup for GER often begins with a barium swallow. Other diagnostic modalities include pH probes, esophagoscopy, esophageal manometry, and nuclear medicine studies.

Gastritis is primary or secondary in etiology. Primary gastritis is associated with Helicobacter pylori infection and is the most common cause of gastritis in children. H pylori is detected using serum immunoglobulin G (IgG) levels, rapid urease testing (CLOtest), or mucosal biopsy.

Lower gastrointestinal tract bleeding in children aged 1 month to 1 year

In cases of intussusception, ultrasonography may be used as the initial diagnostic study to avoid a more invasive barium or pneumatic enema. The ultrasonographic finding of a "target" sign or a "pseudokidney" sign is highly suggestive for intussusception. Because ultrasonographic studies are only diagnostic, many clinicians choose to proceed directly to barium, saline, or pneumatic enema, which are both diagnostic and potentially therapeutic.

In patients with gangrenous bowel, upper and lower contrast studies aid in diagnosis.

Upper gastrointestinal tract bleeding in children aged 1-2 years

When a peptic ulcer that is not associated with H pylori infection is diagnosed, a fasting plasma gastrin level is measured to exclude Zollinger-Ellison syndrome.

Significant upper GI bleeding in patients with ulcer is evaluated and treated with immediate endoscopy. Biopsy samples are taken, if warranted.

Lower gastrointestinal tract bleeding in children aged 1-2 years

Most polyps in persons of this age group are the juvenile type and are located throughout the colon. These are benign hamartomas and usually require no treatment, because they autoamputate. In cases of polyps, colonoscopy is the diagnostic evaluation of choice, because it allows examination of the entire colon and the potential excision of bleeding polyps when they are identified. Colonoscopy is helpful in diagnosing other polyposis syndromes, such as familial polyposis syndromes and adenomatous polyps.

In patients with Meckel diverticulum, technetium-99m (^99m Tc) pertechnetate scanning is used to identify this congenital anomaly, with 90% accuracy. This isotope has a high affinity for parietal cells of gastric mucosa and allows identification of normal and ectopic gastric mucosa. The use of H2 blockers and proton pump inhibitors (PPIs), pentagastrin, and glucagon may enhance the accuracy of this test, since H2 blockers inhibit excretion of the isotope, pentagastrin enhances gastric mucosal uptake, and glucagon inhibits peristalsis.

Upper gastrointestinal tract bleeding in children older than age 2 years

After the initial stabilization of patients with upper GI bleeding, upper endoscopy is the preferred diagnostic and therapeutic tool. The esophagus and stomach are assessed for the presence of varices and in order to exclude gastritis or ulcer disease as the source of bleeding. Gastric varices are most commonly found in the fundus.

Lower gastrointestinal tract bleeding in children older than age 2 years

Vascular lesions include a wide variety of malformations, including hemangiomas, arteriovenous malformations, and vasculitis. If these lesions are located in the colon, colonoscopy may be diagnostic and therapeutic. However, brisk bleeding may obscure the visual field, making localization the bleeding impossible. Arteriography assists in localizing the source and embolizing the feeding vessel.

In all children, the presence of blood can be confirmed by the use of peroxide-based tests, such as the Hemoccult or Hematest for lower GI bleeding and Gastroccult for upper GI bleeding.

Certain ingestions, such as red meat, iron, and peroxidase-containing vegetables (eg, turnips, horseradish, broccoli, cauliflower, and cantaloupe), can give false-positive results.

Patients with suspected obstruction should undergo plain abdominal radiography. Abdominal radiography may also be helpful in neonates in whom NEC is a possibility; the images may show free air, pneumatosis intestinalis (bubbles in the bowel wall), or portal air.

Imaging for nonemergency pediatric GI bleeding may begin with barium contrast studies (barium swallows, upper GI series, small bowel follow-throughs, or barium enemas) to point to foreign bodies, esophagitis, IBD, or polyps. For neonates with malrotation with midgut volvulus, oral contrast may reveal a "corkscrew" of small bowel or a "bird's beak".

When intestinal malrotation is suspected, an immediate upper GI contrast study should be performed to confirm the diagnosis of malrotation with midgut volvulus. Free air in the perineum suggests bowel perforation.

For suspected intussusception, color Doppler ultrasonography can be used. Its sensitivity is 98-100%, and its specificity is 89-100%; these rates are operator-dependent.

Enema studies can successfully reduce intussusception in most patients.

Barium enema studies have traditionally been used with success rates of 50-90%. Rates improve when symptoms are present for less than 24 hours. Barium study is contraindicated if perforation is suspected.

Air and water-soluble contrast enemas have also been used with similar success rates. Air enemas require less radiographic exposure but have slightly higher perforation rates, while enemas with a water-soluble (or saline) contrast agent require experienced sonographers.

If an enema successfully reduces intussusception, in-hospital observation is warranted because reoccurrence is not uncommon.

A Meckel scan uses^99m Tc pertechnetate to highlight the ectopic gastric mucosa.

Arteriography is used to localize lesions when endoscopy has failed or when the patient cannot cooperate. The modality can be helpful for bleeding that is distal to the ligament of Treitz.

Children whose history and physical findings suggest significant bleeding from an upper GI source receive a nasogastric tube for diagnostic purposes.

Return of coffee-ground-like material or Gastroccult-positive material confirms an upper GI bleed.

Still, the false-negative rate is 16% if duodenopyloric regurgitation is absent. Therefore, a clear nasogastric aspirate alone cannot be used to rule out of a GI bleed.

Patients with severe upper GI bleeding should receive endoscopy within the first 12 hours of the hemorrhagic episode if they are sufficiently stable, because early endoscopy improves the diagnostic index.

The site of upper GI bleeding can be identified in 90% of cases when endoscopy is performed within 24 hours. This modality is also beneficial in predicting the likelihood of continued bleeding. ^[11]

The Forress classification divides endoscopic findings into the following 3 categories:

• I - Active hemorrhage (Ia = bright-red bleeding, Ib = slow bleeding)

• II - Recent hemorrhage (IIa = nonbleeding visible vessel, IIb = adherent clot on base of lesion, IIc = flat pigmented spot)

• III - No evidence of bleeding.

The incidence of rebleeding decreases dramatically, because less evidence of ulceration or bleeding is seen.

Push enteroscopy in some studies has been shown to have a higher diagnostic yield than standard esophagogastroduodenoscopy (EGD). Essentially, this modality uses a long endoscope that is placed through the mouth into the jejunum and can reach about 160 cm beyond the ligament of Treitz.

In one study, push enteroscopy identified a large number of mucosal lesions that could not be identified by a standard endoscope. ^[12]

Double-balloon enteroscopy is another technology that has shown high therapeutic and diagnostic yield. This modality employs a high-resolution video endoscope with latex balloons attached at the tip, with an overtube that can be inflated and deflated. It can be inserted either orally or anally.

Wireless capsule endoscopy is under investigation for children, although it can be used only diagnostically, not therapeutically.

For lower GI bleeds, colonoscopy can reveal the source of bleeding more effectively than barium enema can, and it has 80% sensitivity.

Colonoscopy should be performed only when the patient is stable and when blood and feces will not conceal proper visualization.

Sigmoidoscopy alone has also been used in children who have had symptoms of chronic lower GI bleeding for 1 year or longer. The study reveals the most common etiologies, such as juvenile colorectal polyps and nonspecific proctitis.