Octreotide is a synthetic somatostatin analog that shares the four amino acid sequence of the natural hormone necessary for biologic effects. It has a longer half-life than somatostatin and thus is preferentially used in clinical settings. Octreotide has been reported to inhibit gastric secretion in humans and to reduce splanchnic and azygous blood flow in patients with cirrhosis and portal hypertension (1–3). It has been widely used to control acute variceal bleeding and to prevent variceal rebleeding. It is less frequently used in the management of gastrointestinal bleeding of other etiologies.
Most clinical trials involving octreotide have been performed in adults. Very few reports exist regarding its efficacy in children. In this retrospective study, we present the results of octreotide used as treatment for gastrointestinal bleeding in children with and without portal hypertension (PH). This report represents the largest pediatric series described.
MATERIALS AND METHODS
The pharmacy records of Children's Memorial Hospital were reviewed to identify the patients treated with octreotide between January 1995 and December 2000. This study includes children who received continuous intravenous (IV) octreotide infusion for control of acute upper or lower gastrointestinal bleeding, including bleeding from gastrointestinal stoma sites. Data were collected by medical chart review and included age, gender, primary disease, coagulation status, previous history of gastrointestinal bleeding, sclerotherapy or variceal banding, esophagogastroduodenoscopy (EGD) findings, hemoglobin drop secondary to bleeding, transfusion requirements, and duration of octreotide therapy. Cessation of bleeding during treatment and post-treatment rebleeding were the primary outcomes assessed. A complete response to octreotide was defined as cessation of bleeding, with stable hemodynamic status and hemoglobin concentration for the duration of the infusion. Post-treatment rebleeding was defined as any new episodes of bleeding from any site during the interval between discontinuation of octreotide and surgery, death, or loss to follow-up. Ongoing bleeding and recurrent bleeding during octreotide therapy were considered treatment failures. Data from patients with or without PH and with or without intrinsic liver disease were evaluated separately. Fisher exact tests were used for statistical analysis.
Criteria for initiation of octreotide and the duration of treatment were decided by the individual treating physician and, accordingly, were not standardized. A drop in hemoglobin and changes in vital signs caused by bleeding at initial presentation were the commonly stated indications for initiating treatment. Patients received a continuous IV infusion of octreotide at the rate of 1 to 2 μg/kg/h. Most patients also received an IV bolus dose of 1 to 2 μg/kg immediately before continuous infusion. Octreotide therapy typically was discontinued 24 hours after bleeding ceased, or if there was no clinical response, after 12 hours. Blood products, volume expanders, and vitamin K were used when indicated. Most patients received IV ranitidine therapy during the octreotide infusion.
All patients underwent endoscopy for evaluation of acute gastrointestinal bleeding when hemodynamically stable and within 24 hours of their first bleeding episode. For subsequent bleeding episodes, patients routinely underwent endoscopy if a recent endoscopic evaluation had not been performed. Sclerotherapy was performed on patients with Grades III to IV esophageal varices and varices with signs of active or recent bleeding.
IV octreotide therapy was used in 33 patients experiencing 49 acute gastrointestinal bleeding episodes.
Gastrointestinal Bleeding Associated with Portal Hypertension
Twenty-one patients with PH had 35 episodes of gastrointestinal bleeding treated with octreotide. The patient characteristics are shown in Table 1. Intrinsic liver disease was present in 15 patients, whereas 6 had portal vein thrombosis. Eleven patients had previous gastrointestinal bleeding, and 5 patients had undergone variceal sclerotherapy or banding. The median hemoglobin drop from the most recent stable hemoglobin concentration was 2.5 g/dL (range, 0.4–5.5 g/dL). The median total volume of packed red blood cells transfused from onset of bleeding until discontinuation of octreotide therapy was 21.7 mL/kg (range, 0–278 mL/kg). Thirteen patients, 10 with intrinsic liver disease, had coagulopathy at the time of 18 bleeding episodes, and 6 had thrombocytopenia.
Twenty-six upper endoscopic examinations were performed in the 21 patients, 21 while the patient was receiving octreotide infusion and 5 before the acute bleeding episode. Active bleeding was identified during endoscopy in eight patients and was attributed to esophageal varices (n = 4), gastric or gastrostomy varices (n = 2), portal hypertensive gastropathy (n = 1), and gastric ulcer (n = 1). Overall, endoscopy of the 21 patients revealed esophageal varices alone (n = 6) or in combination with gastric varices (n = 3), or in combination with portal hypertensive gastropathy (n = 6). Two patients had gastric varices alone. Two patients had all three: esophageal and gastric varices and portal hypertensive gastropathy. Thus, 90% of patients had varices visible in one or more areas of the gastrointestinal tract as judged by endoscopic examinations performed during or just before the bleeding episode (see Table 1). Findings also included portal hypertensive gastropathy in 8 patients and gastric or esophageal ulceration (4 were postsclerotherapy ulcers) in 7. One of the gastric ulcers was not visible during endoscopic examination because it was obscured by a gastrostomy tube. This ulcer was later identified at surgery and was ultimately treated by resection. The bleeding source could not be identified in one patient.
Outcomes of Octreotide Treatment
The duration of octreotide infusion ranged from 19 hours to 7 days. Sclerotherapy of esophageal varices also was performed in five episodes. The results of octreotide treatment are given in Table 2. Complete response was achieved in 25 episodes, including 4 in which sclerotherapy was also performed. There were 10 treatment failures. Post-treatment rebleeding was observed in 13 episodes at a median interval of 1 month (range, 1 day–8 months). Six post-treatment rebleeding episodes occurred within 2 weeks of the cessation of infusion. Two of the rebleeding episodes occurred in patients who also had sclerotherapy. The results of octreotide therapy in patients with or without liver disease are shown in Table 3. The responses to octreotide, post-treatment rebleeding, and mortality rates were not significantly different in these two groups. Four patients with intrinsic liver disease and one without liver disease had sclerotherapy concurrent with octreotide (P = NS, Fisher exact test). Patients with intrinsic liver disease had coagulopathy in 14 episodes and thrombocytopenia in 5 episodes. Those without intrinsic liver disease had coagulopathy in 4 episodes, and thrombocytopenia in 1 episode.
Eleven patients underwent liver transplantation. Six were treated with surgical shunts (5 mesentericoportal shunts and 1 distal splenorenal shunt). Two underwent transjugular intrahepatic portosystemic shunt. One required gastric resection. Definitive treatment was undertaken at a median of 9 days (range, 2–60 days) after the last episode of bleeding. There were four deaths in this population, occurring 3.5 to 35 days after the cessation of octreotide. One patient died of multiple organ failure, 1 died of post-transplant bacterial sepsis, and 2 died in the operating room during liver transplantation (1 with multiple organ failure and pulmonary hemorrhage, and 1 of uncontrollable intra-abdominal bleeding).
Gastrointestinal Bleeding not Associated with Portal Hypertension
Twelve patients without PH received octreotide during 14 episodes of gastrointestinal bleeding. Table 4 details the patient characteristics. This patient group included children with a variety of diagnoses that experienced life-threatening gastrointestinal bleeding requiring intensive care support. Two patients were liver transplant recipients (one also had a bone marrow transplant), and two had lymphoma. Leukemia, human immunodeficiency virus infection, systemic lupus erythematosus, total colonic Hirschsprung disease, neurodegenerative disease, necrotizing enterocolitis, Peutz-Jeghers syndrome, and subglottic stenosis were the primary diagnoses in the others. Co-morbidities included renal failure (4 patients), liver injury related to parenteral nutrition (2), sepsis (1), hemorrhagic pancreatitis (1), and gallbladder perforation (1). Two patients experienced bleeding after invasive procedures, one after polyp removal and one after liver biopsy. The patient with subglottic stenosis had received high-dose steroid therapy for several days before bleeding. The median hemoglobin drop from the most recent stable hemoglobin concentration was 3.5 g/dL (range, 1.2–7 g/dL). The median volume of packed red blood cells transfused from the onset of bleeding until termination of octreotide infusion was 47.4 mL/kg (range, 8–326 mL/kg). Five patients had coagulopathy, and three had thrombocytopenia.
EGD was performed in 7 of the 12 patients during the bleeding episode. None of the patients were actively bleeding during endoscopy. Gastric ulcers were identified in 2 patients, and 1 patient had a duodenal ulcer with erosive gastritis. A gastric ulcer was subsequently identified in 1 additional patient at the time of exploratory laparotomy. In the remaining 8 patients, including 3 who underwent EGD, the source of bleeding was not identified. Of the 5 patients who did not undergo EGD, 2 were post-procedure patients and 3 were deemed medically unstable for EGD.
Outcomes of Octreotide Treatment
The duration of octreotide infusion ranged from 3 hours to 36 days. Complete response was achieved in seven episodes, including both post-procedure bleeding episodes. Two patients, one with a duodenal ulcer and the other with bleeding of an unknown source, experienced post-treatment rebleeding within 4 days of the infusion. Treatment failed in seven episodes. Of the four patients who had a discrete ulcer as the bleeding source, two experienced a complete response.
Of 12 patients, 3 experienced response to medical therapy alone. Four patients were referred for surgery to control the initial bleeding episode or post-treatment rebleeding. There were six deaths in this group, including one patient who died despite operative management. Three patients died of complications of gastrointestinal bleeding while receiving octreotide. A liver transplant recipient experienced a sudden cardiac event during an acute gastrointestinal bleeding episode. Another patient with systemic lupus erythematosus (SLE) experienced recurrent major life-threatening bleeding while on octreotide therapy, and died during surgery of uncontrollable bleeding from an antral ulcer communicating with a pancreatic pseudocyst. One patient with short bowel syndrome died of bacterial sepsis after receiving octreotide for 36 days to treat presumed small bowel ulceration. The cause of death in the remaining patients included one patient with lymphoma who died of multiple organ failure 5 days after completion of the infusion, and two who died of infectious causes 10 days and 6 weeks, respectively, after completion of the infusion.
One patient with previous cardiac surgery experienced asymptomatic bradycardia (60 beat/min) while receiving octreotide. Another patient experienced agitation, which led to discontinuation of octreotide, but this patient tolerated octreotide well later in longer durations. One patient, a liver transplant recipient, had experienced intermittent upper and lower gastrointestinal bleeding in the prior week, the source of which could not be identified during endoscopy. Octreotide infusion was started to control recurrent bleeding, which had resulted in a drop in hemoglobin to 2.9 g/dL. The patient experienced a sudden cardiac event 5 hours after the octreotide infusion was started. In the hours before this event, the bleeding appeared to stop, and a normal hemoglobin concentration had been achieved with transfusion. A postmortem examination was not performed, but the clinical impression was that death was caused by massive recurrent gastrointestinal bleeding. The contribution of octreotide to the cardiac event remains unknown.
Gastrointestinal Bleeding Associated with Portal Hypertension
Bleeding from esophageal varices is the most serious and potentially life-threatening symptom of PH in children and adults (4). It leads to death in 5% to 9% of children with extrahepatic portal vein thrombosis and in 37% of those with intrinsic hepatic disease untreated by liver transplantation (5,6). Endoscopic sclerotherapy, portosystemic shunts, and liver transplantation are commonly used in the care of these patients. At centers where facilities and expertise to perform these treatments are not available, urgent control of active bleeding and prevention of early rebleeding medically is essential. Variceal pressure has been shown to be an independent predictive factor for variceal bleeding in adults, so agents that reduce variceal pressure have been used in the management of variceal bleeding (7). Octreotide has been shown to reduce splanchnic and azygous blood flow, a reliable index of flow through esophageal varices, and collateral blood flow in patients with cirrhosis and PH (2,3,8–11). Octreotide has been demonstrated to reduce intravariceal pressure by as much as 40% (12,13). However, some studies have shown more modest effects on portal pressure (2,3,8,10,14).
Studies of clinical outcomes in adult patients with variceal bleeding treated with octreotide are heterogenous with regard to patient characteristics, outcome measures studied, method of administration, and dose and duration of octreotide. In a recently published meta-analysis (15), octreotide was shown to improve the control of variceal bleeding compared with alternative medical therapies and placebo/no intervention. A 13% absolute reduction in bleeding rate was reported with octreotide. It was shown to have comparable efficacy to immediate sclerotherapy for control of bleeding.
Very few reports exist on the use of octreotide in children with gastrointestinal bleeding. Our group has previously reported arrest of acute variceal bleeding within 2 hours of infusion in four children with PH when octreotide was used in conjunction with sclerotherapy (EM Alonso, personal communication). Post-treatment rebleeding was controlled with reinstitution of octreotide in all cases. Siafakas et al. (16) successfully used octreotide in three patients with esophageal/gastric varices who had no response to other treatments. Zellos and Schwartz (17) used chronic subcutaneous octreotide therapy for persistent bleeding from esophageal varices. During 24 months of use, no further bleeding was observed.
We report results of octreotide therapy during 35 gastrointestinal bleeding episodes in 21 patients with PH, 90% of whom had varices. Although active variceal bleeding was not visualized in most cases, it was the most likely bleeding source. In the current study, the gastrointestinal bleeding ceased in 71% of 35 bleeding episodes during octreotide treatment. Although the response rate to octreotide (80%) appears to be higher in patients who had sclerotherapy in conjunction with octreotide, this response remains unclear because of the relatively small number of patients studied. In adults with PH, spontaneous cessation of variceal bleeding has been reported to occur in 40% to 50% (18). Bleeding in our series ceased at a higher rate than would be expected spontaneously in adults. Comparison data for children are not available.
Our data suggest that patients without intrinsic liver disease may have a higher response rate to octreotide and a lower mortality rate; however, these results did not reach statistical significance in this small cohort. Studies on larger groups of patients may be more revealing.
Portal hypertensive gastropathy (PHG) and gastric ulcers are other potential causes for gastrointestinal bleeding in patients with PH (19–21).
Kouroumalis et al. (22) showed that octreotide is effective in arresting acute bleeding from PHG within 48 hours. In our study, 39% of patients had PHG in conjunction with varices. The response rate to octreotide in this group was 50%. The only patient actively bleeding at EGD from PHG alone did not experience response to octreotide. Two of three patients with gastric ulcers and four with postsclerotherapy ulcers stopped bleeding during octreotide infusion. PH-induced impairment of the gastric mucosal defenses has been suggested in the pathogenesis of ulcers in patients with cirrhosis, but whether bleeding can be controlled with reduced portal pressure is not yet well established (23).
In all our patients with PH who experienced post-treatment rebleeding episodes, half recurred within 2 weeks after the octreotide infusion was discontinued. Since the risk of rebleeding after octreotide therapy is high, more definitive treatments should be explored while patients remain on the infusion.
In adult studies, mortality rates of patients receiving octreotide alone or in combination with sclerotherapy range from 15% to 38% (23–27). In the meta-analysis it was concluded that octreotide did not significantly decrease mortality in comparison to alternative pharmacologic and mechanical interventions (15). The mortality rate in our patients with PH was 19%. The lower mortality rate in our study may have been related to the generally better prognosis of PH in children than adults.
Gastrointestinal Bleeding not Associated with Portal Hypertension
Published experience using octreotide in patients without PH has been primarily in the treatment of bleeding caused by peptic ulceration. Octreotide has been reported to inhibit gastric acid secretion and to reduce gastroduodenal blood flow (1,28–30). Although Christiansen et al. (31) reported that octreotide was not significantly different from placebo in controlling peptic ulcer bleeding, Jenkins et al. (32) reported successful management of massive upper gastrointestinal bleeding from multiple peptic ulceration with octreotide. Lin et al. (33) compared octreotide with ranitidine in patients with peptic ulcer bleeding and showed a significantly higher homeostasis rate with octreotide.
In our study, the response rate of gastrointestinal bleeding in patients without PH was 50%. Most of our patients had significant co-morbid conditions, which may have reduced the likelihood of response to medical therapy. However, octreotide appeared to be effective in some of these patients, especially when the bleeding source was a discrete ulcer or a procedure site.
Octreotide occasionally can cause nausea, abdominal cramps, diarrhea, fat malabsorption, bradycardia, and hyperglycemia, which usually resolve spontaneously (34,35). Paralytic ileus, thrombocytopenia, and hepatic decompensation also have been rarely documented (26,36). Jenkins et al. (26) reported a significantly higher rate of multiple organ failure in patients receiving octreotide treatment and postulated that octreotide may exacerbate diminished renal and hepatic perfusion during hypotension related to bleeding episodes. Other studies have not supported this hypothesis. In fact, in clinical trials, octreotide has been shown to have an excellent safety profile. In our experience, octreotide was well tolerated. We observed one episode of asymptomatic bradycardia. Three of our patients experienced multiple organ failure. Because these three patients presented with multiple medical problems, it is impossible to determine whether octreotide caused systemic hemodynamic effects that diminished perfusion of vital organs. It is our impression that the benefit of expeditious control of serious gastrointestinal bleeding outweighs the potential risks of altered organ perfusion.
In this series of pediatric patients, octreotide appeared to be effective in achieving high rates of immediate bleeding control in patients with PH, rates comparable to those reported in adults. The role of octreotide therapy as an alternative to emergency endoscopic therapy in pediatric patients requires additional investigation. In addition, more definitive treatments, such as sclerotherapy, banding, shunt procedures, and liver transplantation, should be considered based on the underlying disease process. The effects of octreotide on PHG, ulcer-related bleeding, and gastrointestinal bleeding not associated with PH are not well established and also require additional study. In general, octreotide is well tolerated in children, with minimal adverse events.
The hypothetical contribution of octreotide to the development of multiple organ failure requires additional study but should lead to careful monitoring when the drug is used in critically ill patients.
1. Jenkins SA, Poulianos G, Coraggio F, et al. Somatostatin in the treatment of non-variceal upper gastrointestinal bleeding. Dig Dis 1998; 16:214–24.
2. Pringle SD, McKee RF, Garden OJ, et al. The effect of a long-acting somatostatin analogue on portal and systemic haemodynamics in cirrhosis. Aliment Pharmacol Ther 1988; 2:451–9.
3. McCormick PA, Dick R, Siringo S, et al. Octreotide reduces azygous blood flow in cirrhotic patients with portal hypertension. Eur J Gastroenterol Hepatol 1990; 2:489–92.
4. Karrer FM, Narkewicz MR. Esophageal varices: current management in children. Semin Pediatr Surg 1999; 8( 4):193–201.
5. Mowat AP. Prevention of variceal bleeding. J Pediatr Gastroenterol Nutr 1986; 5:679–87.
6. Fonkalsrud EW. Treatment of variceal hemorrhage in children. Surg Clin North Am 1990; 70:475–87.
7. El Atti EA, Nevens F, Bogaerts K, et al. Variceal pressure is a strong predictor of variceal hemorrhage in patients with cirrhosis as well as in patients with non-cirrhotic portal hypertension. Gut 1999; 45:618–21.
8. Lin HC, Tsai YT, Lee FY, et al. Hemodynamic evaluation of octreotide in patients with hepatitis B-related cirrhosis. Gastroenterology 1992; 103:229–34.
9. Navasa M, Bosch J, Chesta J, et al. Haemodynamic effects of subcutaneous administration of SMS 201–995, a long-acting somatostatin analogue in patients with cirrhosis and portal hypertension (abstract). J Hepatol 1988; 7(Suppl):S64.
10. Eriksson LS, Brundin T, Soderlund C, et al. Haemodynamic effects of a long acting somatostatin analogue in patients with liver cirrhosis. Scand J Gastroenterol 1987; 22:919–25.
11. Escorsell A, Feu F, Garcia-Pagan JC, et al. Hemodynamic effects of octreotide in cirrhotic patients with portal hypertension. A double-blind study (abstract). J Hepatol 1995; 23(Suppl 1):85.
12. Jenkins SA, Baxter JN, Corbett WA, et al. Effects of a somatostatin analogue SMS 201–995 on hepatic hemodynamic in the pig and on variceal pressure in man. Br J Surg 1985; 72:1009–12.
13. Cello JP, Chan MF. Octreotide therapy for variceal hemorrhage. Digestion 1993; 54(Suppl 1):20–6.
14. McKee R. A study of octreotide in esophageal varices. Digestion 1990; 45(Suppl):60–4.
15. Corley DA, Cello JP, Adkisson W, et al. Octreotide for acute esophageal variceal bleeding: a meta-analysis. Gastroenterology 2001; 120:946–54.
16. Siafakas C, Fox VL, Nurko S. Use of octreotide for the treatment of severe gastrointestinal bleeding in children. J Pediatr Gastroenterol Nutr 1998; 26:356–9.
17. Zellos A, Schwartz KB. Efficacy of octreotide in children with chronic gastrointestinal bleeding. J Pediatr Gastroenterol Nutr 2000; 30:442–6.
18. D'Amico G, Pagliaro L, Bosch J. The treatment of portal hypertension. A meta-analytic review. Hepatology 1995; 22:332–54.
19. D'Amico G, Montalbano L, Traina M, et al. Natural history of congestive gastropathy in cirrhosis. Gastroenterology 1990; 99; 1558–64.
20. Gostout CJ, Viggiano TR, Balm RK. Acute gastrointestinal bleeding from portal hypertensive gastropathy: prevalence and clinical features. Am J Gastroenterol 1993; 88:2030–3.
21. Kitano S, Dolgor B. Does portal hypertension contribute to the pathogenesis of gastric ulcer associated with liver cirrhosis? J Gastroenterol 2000; 35:79–86.
22. Kouroumalis EA, Koutroubakis IE, Manousos ON. Somatostatin for acute severe bleeding from portal hypertensive gastropathy. Eur J Gastroenterol Hepatol 1998; 10:509–12.
23. Freitas DS, Sofia C, Pontes JM, et al. Octreotide in acute bleeding esophageal varices: a prospective randomized study. Hepatogastroenterology 2000; 47:1310–4.
24. Besson I, Ingrand P, Person B, et al. Sclerotherapy with or without octreotide for acute variceal bleeding. N Engl J Med 1995; 333:555–60.
25. Sung JJY, Chung SCS, Lai CW, et al. Octreotide infusion or emergency sclerotherapy for variceal haemorrhage. Lancet 1993; 342:637–41.
26. Jenkins SA, Shields R, Davies M, et al. A multicentre randomised trial comparing octreotide and injection sclerotherapy in the management and outcome of acute variceal haemorrhage. Gut 1997; 41:526–33
27. Bildozola M, Kravetz D, Argonz J, et al. Efficacy of octreotide and sclerotherapy in the treatment of acute variceal bleeding in cirrhotic patients. Scand J Gastroenterol 2000; 35:419–25.
28. Whitehouse I, Beglinger C, Fried M, et al. The effect of an octapeptide somatostatin analogue (SMS 201–995) and somatostatin-14 on pentagastrin-stimulated gastric acid secretion: a comparative study in man. Hepatogastroenterologica 1984; 31:227–9.
29. Whitehouse I, Beglinger C, Ruettimann C, et al. Inhibition of pentagastrin-stimulated acid secretion after subcutaneous administration of a new somatostatin analogue. Gut 1986; 27:141–6.
30. Olson JA, Loud FB, Christianssen J. Inhibition of meal stimulated gastric acid secretion by the octapeptide somatostatin analogue SMS 201–995. Gut 1987; 28:464–7.
31. Christiansen J, Ottenjann R, Von Arx F. Placebo-controlled trial with the somatostatin analogue SMS 201–995 in peptic ulcer bleeding. Gastroenterology 1989; 97:468–574.
32. Jenkins SA, Taylor BA, Nott DM, et al. Management of massive upper gastrointestinal haemorrhage from multiple sites of peptic ulceration with somatostatin and octreotide: a report of five cases. Gut 1992; 33:404–7.
33. Lin HJ, Perng CL, Wang K, et al. Octreotide for arrest of peptic ulcer haemorrhage: A prospective, randomized controlled trial. Hepatogastroenterology 1995; 42:856–60.
34. Lamberts SWJ, van der Lely AJ, de Herder WW, et al. Octreotide. N Engl J Med 1996; 334:246–54.
35. McCormick PA, Chin J, Greenslade L, et al. Cardiovascular effects of octreotide in patients with hepatic cirrhosis. Hepatology 1995; 21:1255–60.
36. Drouin E, Martin S, Alvarez F. Hepatic decompensation in a child with biliary atresia on octreotide perfusion (abstract). J Pediatr Gastroenterol Nutr 1999; 29:512.