CLINICAL SCENARIO
A 55-year-old gentleman presented to the emergency department with an episode of massive hematemesis. He was diabetic for 10 years and was taking oral hypoglycemic agent for the past 10 years. He has suffered acute anterior wall myocardial infarction last year. He has undergone primary coronary intervention including coronary stenting. Following that he was maintained on aspirin and beta blockers. On examination, his pulse rate was 126 beats/min and blood pressure in the left arm in the supine posture was 94/60 mm Hg. He appeared pale and very anxious. His abdominal examination was unremarkable and there was no splenomegaly.
MAIN CLINICAL QUESTIONS
There are three main issues to be considered at this point: Assessment of volume of blood loss and appropriate resuscitation, identification of the source of bleeding, and to determine if bleeding still active or it has ceased.
Assessment of Severity of Blood Loss and Resuscitation
Assessment of severity of bleeding should be done as soon as the patient is received in emergency. Heart rate and blood pressure are good indicators of amount of blood loss. As reflected by the presence of tachycardia and supine hypotension, it is clear that the index patient is hemodynamically unstable and lost approximately 30% of blood volume already (Table 1). Hence, the first objective is resuscitation of the patient. Two large bore intravenous (IV) cannula or a central venous line should be inserted and crystalloids should be infused at a rate of 25–30 mL/kg. A nasogastric tube should be placed to look for ongoing bleeding and also to lavage the stomach for decreasing the risk of aspiration and also prior to endoscopic examination. Blood should be given to maintain hemoglobin at least 7–8 g/dL. Higher threshold is needed in hemodynamically unstable patient and patients with cardiac illness.
The initial hemoglobin level in patients with acute gastrointestinal (GI) bleeding may not reflect the amount of blood loss as whole blood is lost (both plasma and RBC). The loss of volume is compensated by activation of aldosterone–renin–angiotensin cascade which conserves the salt and water loss and help in increasing intravascular volume. At this point of time (say 24–48 hours), the hemoglobin level may decline because of the effect of hemodilution. The hemodilution also occurs because of infusion of crystalloids as part of initial resuscitation. Hence, one should interpret the hemoglobin report appropriately. For example, the hemoglobin level may remain normal even with the loss of a liter of blood while the patient shows the signs of tachycardia and hypotension.
Goals of resuscitation should be to achieve systolic blood pressure at least >100 mm Hg and heart rate <100 beats/min.
Is the bleeding active?
Although most bleeding episodes stops spontaneously, it is important to know if the bleeding is still ongoing. This can be assessed by persistence of tachycardia and low blood pressure. The activity of upper gastrointestinal (UGI) bleeding is also assessed on the return of gastric contents through the nasogastric tube. While fresh blood through nasogastric tube suggests active bleeding, altered blood or coffee ground blood suggests that the bleeding might have stopped. We should also remember that bleeding can occur intermittently and bleeding from duodenum and below may not regurgitate into the stomach and hence cannot be assessed through nasogastric tube aspirate. A nasogastric tube was placed in the index patient and it showed that the bleeding was still active.
What is the cause of bleeding?
Bleeding from esophagus to the ligament of Treitz at the duodenojejunal flexure is defined as UGI bleeding. The bleeding may manifest as hematemesis (vomiting of blood, bright red blood suggestive of recent or ongoing bleeding whereas coffee-ground emesis suggests old bleeding) or melena (black tarry stool). If the rate of bleeding is high, large amount of fresh blood can traverse through the GI tract and passed as red blood/altered blood in the stool (hematochezia). Hematochezia refers to passage of bright red- or maroon-colored blood per rectum and generally suggests lower GI as the source of bleeding.
A quick history and clinical examination may point towards a cause of bleeding lesion (Table 2). The usual causes of upper and lower GI bleeding are summarized in Table 3.
To assess: who are the patients at high risk of morbidity?
All patients with GI bleeding should be assessed for the risk which guides triaging for requirement of admission and the risk of re-bleeding after initial control. Pre-endoscopy risk scoring system includes the Glasgow–Blatchford Score (GBS), the Clinical Rockall Score, an artificial neural network score, and the AIMS65 score, of which GBS is commonly used. If the GBS is ≥6, >50% of patients will need endoscopic therapy (Table 4). Patients with UGI bleeding and having a GBS 0–1 can be managed on an outpatient basis.
|
|
When to consider for endoscopy?
After initial resuscitation and initiation of intravenous proton pump inhibitors and/or, UGI endoscopy should be planned based on risk assessment. Patients with hemodynamic instability should be shifted to intensive care unit and endoscopy should be done at the earliest after initial resuscitation. In all other patients, endoscopy examination should be done as soon as possible, preferably within 24 hours.
COMING BACK TO THE INDEX PATIENT
As discussed earlier, that the index patient was hypotensive, suggesting that he had a massive UGI bleeding and the estimated loss of blood volume was >30% of total blood volume. The aspirin which he was taking was withheld. He was immediately resuscitated with intravenous crystalloids. The hemoglobin at the time of presentation was 14.5 g/dL (same as his baseline value) in spite of massive bleeding. Three packed cell RBC were arranged and transfused rapidly. Simultaneously, a nasogastric tube was placed and it showed active bleeding. On quick clinical evaluation, he did not have clinical pointer toward any specific disease (Table 2). He was also given 80 mg of pantoprazole intravenously and then pantoprazole was continued at a rate of 8 mg/h infusion using infusion pump. With crystalloids and blood transfusion, his blood pressure rose to 124/74 mm Hg.
|
His blood was also sent for complete blood count, liver function test, INR, and renal function test. He was then wheeled into the endoscopy theater. Before shifting to endoscopy theater, a good gastric lavage was given to clean the stomach which would allow clear visualization of the mucosa and also decrease the risk of aspiration.
On endoscopic examination, there was some amount of liquid blood in the stomach, which was sucked using the endoscopic suction. He was found to have a gastric ulcer, approximately 1.5 cm with a nonbleeding visible vessel at the base of the ulcer. There was no active bleeding at this point of time. Based on the type and stigmata of peptic ulcers, the peptic ulcers are classified as per Forrest classification. Types I and II have high risk of bleeding whereas type III has low risk of rebleeding (Table 5). Various forms of endoscopic therapeutic strategies are now available and they are effective in controlling not only the active bleeding but also the recurrence of bleeding.
Since he had a gastric ulcer with visible vessel at the base of the ulcer, he had high chances of recurrence of bleeding, hence he needed an endotherapy even though the lesion was not bleeding actively. The lesion appeared most amenable to endoscopic application of hemoclips and hence two hemoclips were applied over the visible vessel successfully (Figs. 1A and B). There are multiple endoscopic methods for control of bleeding from peptic ulcer, which are summarized in Table 6.
|
After observing for a few minutes for any active bleeding after application of hemoclips, four pieces of biopsies were obtained from the antrum of the stomach.
Figs. 1A and B: Gastric ulcer with nonbleeding visible vessel (arrow) and endoscopic hemoclip application.
|
One piece was used to urea breath test, which was observed to be positive and three pieces in 10% formalin was sent to pathology laboratory for histopathological examination, which later confirmed presence of Helicobacter pylori (H. pylori) and chronic active gastritis.
Over next 72 hours, he was monitored closely using a monitoring chart including pulse, blood pressure, activity of melena, hemoglobin, and hematocrit. He remained stable and there was no recurrence of bleeding. At this time intravenous pantoprazole was stopped and he was put on oral pantoprazole. Since his rapid urease test was positive, suggesting infection by Helicobacter pylori. H. pylori should be tested in all patients with peptic ulcer and if found to be positive, they must be eradicated. Therefore, he was prescribed anti-H. pylori therapy. He was prescribed triple therapy including clarithromycin 500 mg twice a day, capsule amoxycillin 1 g twice a day and capsule pantoprazole 40 mg twice a day, all for total of 14 days. The pantoprazole in the dose of 40 mg was used further for 2 weeks to complete 4 weeks of therapy.
He was subsequently discharged on day 5. Since he had gastric ulcer, he underwent a repeat endoscopic examination for confirmation of healing of ulcer and indeed the ulcer had healed completely. Malignant gastric ulcers do not heal with proton pump inhibitor (PPI) and hence demonstration of healing is essential after 4–8 weeks. The biopsies were also taken from antrum again to look for eradication of H. pylori, both the rapid urease test and the histopathological examination confirmed eradication of H. pylori.
Summary of management of a patient with bleeding peptic ulcer has been described in Box 1.
CONCLUSION
- Hemoglobin may remain normal (as baseline) in patients with active GI bleeding.
- The three principles of treating patients with UGI bleeding include initial appropriate resuscitation and stabilization, assessment for the activity of bleeding, identification of bleeding lesions and then institution of appropriate therapeutic strategies to handle the bleeding lesion.
SUGGESTED READINGS
- Laine L, Barkun AN, Saltzman JR, Martel M, Leontiadis GI. ACG Clinical Guideline: Upper Gastrointestinal and Ulcer Bleeding. Am J Gastroenterol. 2021;116:899–917.
- Barkun AN, Almadi M, Kuipers EJ, Laine L, Sung J, Tse F, et al. Management of Nonvariceal Upper Gastrointestinal Bleeding: Guideline Recommendations From the International Consensus Group. Ann Intern Med. 2019;171:805–22.