Lack of difference among terlipressin, somatostatin, and octreotide in the control of acute gastroesophageal variceal hemorrhage
Running head: Effects of three vasoactive drugs in variceal bleeding
Yeon Seok Seo,1 Soo Young Park,2 Moon Young Kim,3 Ju Hyun Kim,4 Jun Yong Park,5 Hyung Joon Yim,1 Byoung Kuk Jang,6 Hong Soo Kim,7 Taeho Hahn,8 Byung Ik Kim,9 Jeong Heo,10 Hyonggin An,11 Won Young Tak,2 Soon Koo Baik,3,12 Kwang Hyub Han,5,12 Jae Seok Hwang,6 Sang Hoon Park,8 Mong Cho,10 Soon Ho Um1,12
Department of Internal Medicine, Korea University College of Medicine, Seoul1; Kyungpook National University School of Medicine, Daegu2; Yonsei University Wonju College of Medicine, Wonju3; Gachon Medical School, Gil Medical Center, Incheon4; Yonsei University College of Medicine, Seoul5; Keimyung University College of Medicine, Daegu6; Soonchunhyang University College of Medicine, Cheonan7; Hallym University College of Medicine, Anyang8; Sungkyunkwan University College of Medicine, Kangbuk Samsung Hospital, Seoul9; Pusan National University College of Medicine, Busan10; Department of Biostatistics, Korea University College of Medicine, Seoul11; Liver Cirrhosis Clinical Research Center, Seoul, Korea12
Correspondence: Soon Ho Um, MD, PhD Division of Gastroenterology and Hepatology, Department of Internal Medicine, Korea University College of Medicine, 126-1, 5-Ga, Anam-Dong, Seongbuk-Gu, Seoul 136-705, Korea Tel: 82-2-920-6608, FAX: 82-2-953-1943, E-mail: [email protected]
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/hep.27006
Hepatology
Abbreviations: BP: blood pressure CI: confidence interval EIS: endoscopic injection sclerotherapy EVs: esophageal varices EVL: endoscopic variceal ligation EVO: endoscopic variceal obturation GOVs: gastroesophageal varices GVs: gastric varices Hb: hemoglobin HCC: hepatocellular carcinoma IGVs: isolated gastric varices MELD: Model for End-stage Liver Disease OR: odds ratio TIPS: transjugular intrahepatic portosystemic shunt
Hepatology
Page 2 of 33
Page 3 of 33
Hepatology
ABSTRACT Background & Aims: Vasoactive drugs are recommended to be started as soon as possible in suspected variceal bleeding, even before diagnostic endoscopy. However, it is still unclear whether the therapeutic efficacies of the various vasoactive drugs used are comparable. The aim of this prospective multicenter randomized non-inferiority trial was to characterize the effects of terlipressin, somatostatin, and octreotide, when they are initiated before endoscopic treatment in patients with acute variceal bleeding. Methods: Patients with liver cirrhosis and significant upper gastrointestinal bleeding were randomly assigned to receive early administration of terlipressin, somatostatin, or octreotide, followed by endoscopic treatment. Patients with non-variceal bleeding were excluded after endoscopy. The primary endpoint was 5-day treatment success, defined as control of bleeding without rescue treatment, rebleeding or mortality, with a non-inferiority margin of 0.1. Results: In total, 780 patients with variceal bleeding were enrolled: 261 in terlipressin group, 259 in somatostatin group, and 260 in octreotide group. At the time of initial endoscopy, active bleeding was noted in 43.7%, 44.4%, and 43.5% of these patients, respectively (P=0.748), and treatment success was achieved by day 5 in 86.2%, 83.4%, and 83.8% (P=0.636) with similar rates of control of bleeding without rescue treatment (89.7%, 87.6%, and 88.1%, P=0.752), or rebleeding (3.4%, 4.8%, and 4.4%, P=0.739), mortality (8.0%, 8.9%, and 8.8%, P=0.929). The absolute values of lower bound of confidence interval for terlipressin vs somatostatin, terlilpressin vs octreotide, and octreotide vs somatostatin were 0.095, 0.090, and 0.065, respectively. Conclusion: The hemostatic effects and safety did not differ significantly between terlipressin, somatostatin, and octreotide as adjuvants to endoscopic treatment in patients with acute gastroesophageal variceal bleeding. Keywords: variceal bleeding, hemostasis, safety, vasoactive drug, cirrhosis
Hepatology
Hepatology
INTRODUCTION Acute variceal bleeding is one of the critical complications in patients with liver cirrhosis.1 Although the mortality rate has decreased significantly during the past several decades due to remarkable improvements in diagnostic and therapeutic modalities for its management, acute variceal bleeding remains a leading cause of death in cirrhotic patients.1, 2 Therefore, appropriate and effective treatments should be performed promptly in these patients. Three vasoactive drugs—terlipressin, somatostatin, and octreotide—play a role in the control of variceal bleeding by reducing portal blood flow and portal pressure.3, 4 Previous studies showed that the use of vasoactive drugs was associated with a significantly lower risk of mortality and transfusion requirements, and an improved control of bleeding.5, 6 In particular, the treatment response in patients with variceal bleeding was significantly better when combining vasoactive drugs with endoscopic therapy than with pharmacological or endoscopic monotherapy.7-11 Therefore, current practice guidelines recommend the combination of pharmacological therapy and endoscopic therapy as the standard treatment for acute variceal bleeding, particularly the early administration of vasoactive drugs even before endoscopy in suspected variceal bleeding.12-14 Regarding the methods of endoscopic therapy, several studies suggested that the probability of rebleeding from esophageal varices (EVs) is significantly lower for endoscopic variceal ligation (EVL) than for endoscopic injection sclerotherapy (EIS).15, 16 There is no preference when selecting one of these three vasoactive drugs because their efficacies seem to be comparable, although some authors recommend terlipressin as the first choice17 because it is the only drug that was found to improve the survival in placebo-controlled trials18, 19 and a meta-analysis.5 However, it is unclear whether the earlier results could be directly applied to the current clinical situations, since most patients in the prior studies were treated with vasoactive drugs alone or with vasoactive drugs plus EIS while EVL is now primarily recommended for the endoscopic control of esophageal variceal bleeding in combination with vasoactive drugs. In addition, there are some concerns regarding the hemostatic efficacy of octreotide because it was ineffective in the setting of monotherapy in a randomized controlled trial20 and desensitization of octreotide was reported in cirrhotic patients with portal hypertension.21 The present multicenter, prospective, open-label randomized trial was conducted to compare the efficacies of the vasoactive drugs terlipressin, somatostatin, and octreotide in patients with cirrhosis and acute variceal bleeding when they were combined with endoscopic therapy. Because we hypothesized that these drugs have the same treatment efficacies, the present study was designed as a noninferiority test. The 5-day treatment
Hepatology
Page 4 of 33
Page 5 of 33
Hepatology
success rate was assessed, as determined by the control of bleeding free of rescue treatment, rebleeding and mortality.
PATIENTS AND METHODS Patients Eleven medical centers distributed throughout Korea participated in this study. Patients with liver cirrhosis presenting with either hematemesis or melena between October 1, 2006 and May 31, 2010 were assessed for inclusion in this trial after obtaining informed written consent from all potential patients or from their next of kin when the patients were unable to give this consent. The inclusion criteria were as follows: (i) liver cirrhosis diagnosed by previous liver biopsy or by compatible clinical, laboratory, and radiologic findings, (ii) age between 16 and 75 years, (iii) arrival at the hospital within 24 h after the occurrence of hematemesis and/or melena, (iv) clinically significant bleeding, defined as systolic blood pressure (BP) 20 mmHg, or pulse rate >100 beats/min,22 (v) patients agreeing to participate in this trial with informed consent. Patients were excluded if they presented any of the following: (i) non-cirrhotic portal hypertension, (ii) a history of endoscopic variceal therapy within 2 weeks before the episode, (iii) enrollment during the 6 week period before the index bleed, (iv) a history of severe cardiovascular disease, including acute myocardial infarction, atrioventricular block, congestive heart failure, ischemic heart disease, or severe hypertension (systolic BP >170 mmHg and/or diastolic BP >100 mmHg), (v) chronic renal failure, (vi) advanced hepatocellular carcinoma (HCC) invading portal vein, (vii) other malignancy, (viii) pregnancy, (ix) positive result in a human immunodeficiency virus test, (x) known hypersensitivity to any of the study drugs, or (xi) refusal to participate in the study. This study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the ethical committees for human investigations at all of the enrolled hospitals. The study protocol was registered at www.clinicaltrials.gov (ClinicalTrials.gov identifier: NCT00966355). Definitions Time zero (T0) was defined as the time of admission to the first hospital that the patient as taken to.23 In patients who presented a clinical sign of bleeding (e.g., hematemesis or melena) during hospitalization for other reasons, the time when the patient noticed the sign was considered as T0. The index bleeding was defined as controlled when there was no hematemesis, the hemoglobin (Hb) level was stable without requiring blood transfusions, and the vital signs were stable (systolic BP >100 mmHg and pulse rate 38°C that lasted more than 24 h and leukocytosis with a shift to the left, but without any other evidence of infection, were considered as having possible infections.24 Hepatic dysfunction was evaluated according to the Child-Pugh classification25 and the Model for End-stage Liver Disease (MELD) score employed by the United Network of Organ Sharing (http://www.unos.org). The stage of HCC was classified according to the Modified UICC classification.26 Definitions for the classifications according to the type of bleeding and to the source of bleeding are described in the Supporting Methods. Pharmacologic treatment Once enrolled, patients were randomly allocated to the following three groups using sealed opaque envelopes numbered according to a table of random numbers: terlipressin, somatostatin, and octreotide groups. Randomization was performed using computer-generated random numbers to allocate patients to the terlipressin, somatostatin, and octreotide groups in a 1:1:1 ratio. The study drugs were administered immediately after randomization as follows: terlipressin (2 mg) by i.v. bolus followed by 1 mg i.v. every 6 h for 5 days,
Hepatology
Page 6 of 33
Page 7 of 33
Hepatology
somatostatin (250 µg) by i.v. bolus followed by 250 µg/h continuous infusion for 5 days, and octreotide (50 µg) by i.v. bolus followed by 25 µg/h continuous infusion for 5 days. The vasoactive drug was stopped after endoscopy in cases where the bleeding was determined to be unrelated to gastroesophageal varices (GOVs). Endoscopic treatments After the initiation of pharmacologic treatment, endoscopic examination was performed as soon as patients had been resuscitated. Endoscopic treatment was applied immediately in patients diagnosed with bleeding from GOVs on endoscopy. EVL and endoscopic variceal obturation (EVO) with cyanoacrylate were considered as the primary therapies for bleeding from esophageal or gastric cardial varices and for bleeding from gastric fundal varices, respectively.14 If the primary therapy was not technically feasible, EIS and EVL/EIS were performed for esophageal/cardial varices and for fundal varices, respectively. In the present study, bleeding from EVs was treated with EVL in 92.0% of patients except for minor cases managed by EIS in (0.9%), bleeding from GOV1 were also mainly treated with EVL (71.0%) followed by EVO (22.6%), bleeding from GOV2 was principally treated with EVO (72.3%) followed by EVL (20.0%), and bleeding from isolated gastric varices type 1 (IGV1) was principally treated with EVO (79.3%) followed by EVL (10.3%). Detailed information about endoscopic treatments and other treatments are described in the Supporting Methods. Statistics Statistical analyses were performed with the Statistical Package for Social Sciences version 13.0 (SPSS, Chicago, IL, USA). Statistical significance was established at P
Running head: Effects of three vasoactive drugs in variceal bleeding
Yeon Seok Seo,1 Soo Young Park,2 Moon Young Kim,3 Ju Hyun Kim,4 Jun Yong Park,5 Hyung Joon Yim,1 Byoung Kuk Jang,6 Hong Soo Kim,7 Taeho Hahn,8 Byung Ik Kim,9 Jeong Heo,10 Hyonggin An,11 Won Young Tak,2 Soon Koo Baik,3,12 Kwang Hyub Han,5,12 Jae Seok Hwang,6 Sang Hoon Park,8 Mong Cho,10 Soon Ho Um1,12
Department of Internal Medicine, Korea University College of Medicine, Seoul1; Kyungpook National University School of Medicine, Daegu2; Yonsei University Wonju College of Medicine, Wonju3; Gachon Medical School, Gil Medical Center, Incheon4; Yonsei University College of Medicine, Seoul5; Keimyung University College of Medicine, Daegu6; Soonchunhyang University College of Medicine, Cheonan7; Hallym University College of Medicine, Anyang8; Sungkyunkwan University College of Medicine, Kangbuk Samsung Hospital, Seoul9; Pusan National University College of Medicine, Busan10; Department of Biostatistics, Korea University College of Medicine, Seoul11; Liver Cirrhosis Clinical Research Center, Seoul, Korea12
Correspondence: Soon Ho Um, MD, PhD Division of Gastroenterology and Hepatology, Department of Internal Medicine, Korea University College of Medicine, 126-1, 5-Ga, Anam-Dong, Seongbuk-Gu, Seoul 136-705, Korea Tel: 82-2-920-6608, FAX: 82-2-953-1943, E-mail: [email protected]
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/hep.27006
Hepatology
Abbreviations: BP: blood pressure CI: confidence interval EIS: endoscopic injection sclerotherapy EVs: esophageal varices EVL: endoscopic variceal ligation EVO: endoscopic variceal obturation GOVs: gastroesophageal varices GVs: gastric varices Hb: hemoglobin HCC: hepatocellular carcinoma IGVs: isolated gastric varices MELD: Model for End-stage Liver Disease OR: odds ratio TIPS: transjugular intrahepatic portosystemic shunt
Hepatology
Page 2 of 33
Page 3 of 33
Hepatology
ABSTRACT Background & Aims: Vasoactive drugs are recommended to be started as soon as possible in suspected variceal bleeding, even before diagnostic endoscopy. However, it is still unclear whether the therapeutic efficacies of the various vasoactive drugs used are comparable. The aim of this prospective multicenter randomized non-inferiority trial was to characterize the effects of terlipressin, somatostatin, and octreotide, when they are initiated before endoscopic treatment in patients with acute variceal bleeding. Methods: Patients with liver cirrhosis and significant upper gastrointestinal bleeding were randomly assigned to receive early administration of terlipressin, somatostatin, or octreotide, followed by endoscopic treatment. Patients with non-variceal bleeding were excluded after endoscopy. The primary endpoint was 5-day treatment success, defined as control of bleeding without rescue treatment, rebleeding or mortality, with a non-inferiority margin of 0.1. Results: In total, 780 patients with variceal bleeding were enrolled: 261 in terlipressin group, 259 in somatostatin group, and 260 in octreotide group. At the time of initial endoscopy, active bleeding was noted in 43.7%, 44.4%, and 43.5% of these patients, respectively (P=0.748), and treatment success was achieved by day 5 in 86.2%, 83.4%, and 83.8% (P=0.636) with similar rates of control of bleeding without rescue treatment (89.7%, 87.6%, and 88.1%, P=0.752), or rebleeding (3.4%, 4.8%, and 4.4%, P=0.739), mortality (8.0%, 8.9%, and 8.8%, P=0.929). The absolute values of lower bound of confidence interval for terlipressin vs somatostatin, terlilpressin vs octreotide, and octreotide vs somatostatin were 0.095, 0.090, and 0.065, respectively. Conclusion: The hemostatic effects and safety did not differ significantly between terlipressin, somatostatin, and octreotide as adjuvants to endoscopic treatment in patients with acute gastroesophageal variceal bleeding. Keywords: variceal bleeding, hemostasis, safety, vasoactive drug, cirrhosis
Hepatology
Hepatology
INTRODUCTION Acute variceal bleeding is one of the critical complications in patients with liver cirrhosis.1 Although the mortality rate has decreased significantly during the past several decades due to remarkable improvements in diagnostic and therapeutic modalities for its management, acute variceal bleeding remains a leading cause of death in cirrhotic patients.1, 2 Therefore, appropriate and effective treatments should be performed promptly in these patients. Three vasoactive drugs—terlipressin, somatostatin, and octreotide—play a role in the control of variceal bleeding by reducing portal blood flow and portal pressure.3, 4 Previous studies showed that the use of vasoactive drugs was associated with a significantly lower risk of mortality and transfusion requirements, and an improved control of bleeding.5, 6 In particular, the treatment response in patients with variceal bleeding was significantly better when combining vasoactive drugs with endoscopic therapy than with pharmacological or endoscopic monotherapy.7-11 Therefore, current practice guidelines recommend the combination of pharmacological therapy and endoscopic therapy as the standard treatment for acute variceal bleeding, particularly the early administration of vasoactive drugs even before endoscopy in suspected variceal bleeding.12-14 Regarding the methods of endoscopic therapy, several studies suggested that the probability of rebleeding from esophageal varices (EVs) is significantly lower for endoscopic variceal ligation (EVL) than for endoscopic injection sclerotherapy (EIS).15, 16 There is no preference when selecting one of these three vasoactive drugs because their efficacies seem to be comparable, although some authors recommend terlipressin as the first choice17 because it is the only drug that was found to improve the survival in placebo-controlled trials18, 19 and a meta-analysis.5 However, it is unclear whether the earlier results could be directly applied to the current clinical situations, since most patients in the prior studies were treated with vasoactive drugs alone or with vasoactive drugs plus EIS while EVL is now primarily recommended for the endoscopic control of esophageal variceal bleeding in combination with vasoactive drugs. In addition, there are some concerns regarding the hemostatic efficacy of octreotide because it was ineffective in the setting of monotherapy in a randomized controlled trial20 and desensitization of octreotide was reported in cirrhotic patients with portal hypertension.21 The present multicenter, prospective, open-label randomized trial was conducted to compare the efficacies of the vasoactive drugs terlipressin, somatostatin, and octreotide in patients with cirrhosis and acute variceal bleeding when they were combined with endoscopic therapy. Because we hypothesized that these drugs have the same treatment efficacies, the present study was designed as a noninferiority test. The 5-day treatment
Hepatology
Page 4 of 33
Page 5 of 33
Hepatology
success rate was assessed, as determined by the control of bleeding free of rescue treatment, rebleeding and mortality.
PATIENTS AND METHODS Patients Eleven medical centers distributed throughout Korea participated in this study. Patients with liver cirrhosis presenting with either hematemesis or melena between October 1, 2006 and May 31, 2010 were assessed for inclusion in this trial after obtaining informed written consent from all potential patients or from their next of kin when the patients were unable to give this consent. The inclusion criteria were as follows: (i) liver cirrhosis diagnosed by previous liver biopsy or by compatible clinical, laboratory, and radiologic findings, (ii) age between 16 and 75 years, (iii) arrival at the hospital within 24 h after the occurrence of hematemesis and/or melena, (iv) clinically significant bleeding, defined as systolic blood pressure (BP) 20 mmHg, or pulse rate >100 beats/min,22 (v) patients agreeing to participate in this trial with informed consent. Patients were excluded if they presented any of the following: (i) non-cirrhotic portal hypertension, (ii) a history of endoscopic variceal therapy within 2 weeks before the episode, (iii) enrollment during the 6 week period before the index bleed, (iv) a history of severe cardiovascular disease, including acute myocardial infarction, atrioventricular block, congestive heart failure, ischemic heart disease, or severe hypertension (systolic BP >170 mmHg and/or diastolic BP >100 mmHg), (v) chronic renal failure, (vi) advanced hepatocellular carcinoma (HCC) invading portal vein, (vii) other malignancy, (viii) pregnancy, (ix) positive result in a human immunodeficiency virus test, (x) known hypersensitivity to any of the study drugs, or (xi) refusal to participate in the study. This study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the ethical committees for human investigations at all of the enrolled hospitals. The study protocol was registered at www.clinicaltrials.gov (ClinicalTrials.gov identifier: NCT00966355). Definitions Time zero (T0) was defined as the time of admission to the first hospital that the patient as taken to.23 In patients who presented a clinical sign of bleeding (e.g., hematemesis or melena) during hospitalization for other reasons, the time when the patient noticed the sign was considered as T0. The index bleeding was defined as controlled when there was no hematemesis, the hemoglobin (Hb) level was stable without requiring blood transfusions, and the vital signs were stable (systolic BP >100 mmHg and pulse rate 38°C that lasted more than 24 h and leukocytosis with a shift to the left, but without any other evidence of infection, were considered as having possible infections.24 Hepatic dysfunction was evaluated according to the Child-Pugh classification25 and the Model for End-stage Liver Disease (MELD) score employed by the United Network of Organ Sharing (http://www.unos.org). The stage of HCC was classified according to the Modified UICC classification.26 Definitions for the classifications according to the type of bleeding and to the source of bleeding are described in the Supporting Methods. Pharmacologic treatment Once enrolled, patients were randomly allocated to the following three groups using sealed opaque envelopes numbered according to a table of random numbers: terlipressin, somatostatin, and octreotide groups. Randomization was performed using computer-generated random numbers to allocate patients to the terlipressin, somatostatin, and octreotide groups in a 1:1:1 ratio. The study drugs were administered immediately after randomization as follows: terlipressin (2 mg) by i.v. bolus followed by 1 mg i.v. every 6 h for 5 days,
Hepatology
Page 6 of 33
Page 7 of 33
Hepatology
somatostatin (250 µg) by i.v. bolus followed by 250 µg/h continuous infusion for 5 days, and octreotide (50 µg) by i.v. bolus followed by 25 µg/h continuous infusion for 5 days. The vasoactive drug was stopped after endoscopy in cases where the bleeding was determined to be unrelated to gastroesophageal varices (GOVs). Endoscopic treatments After the initiation of pharmacologic treatment, endoscopic examination was performed as soon as patients had been resuscitated. Endoscopic treatment was applied immediately in patients diagnosed with bleeding from GOVs on endoscopy. EVL and endoscopic variceal obturation (EVO) with cyanoacrylate were considered as the primary therapies for bleeding from esophageal or gastric cardial varices and for bleeding from gastric fundal varices, respectively.14 If the primary therapy was not technically feasible, EIS and EVL/EIS were performed for esophageal/cardial varices and for fundal varices, respectively. In the present study, bleeding from EVs was treated with EVL in 92.0% of patients except for minor cases managed by EIS in (0.9%), bleeding from GOV1 were also mainly treated with EVL (71.0%) followed by EVO (22.6%), bleeding from GOV2 was principally treated with EVO (72.3%) followed by EVL (20.0%), and bleeding from isolated gastric varices type 1 (IGV1) was principally treated with EVO (79.3%) followed by EVL (10.3%). Detailed information about endoscopic treatments and other treatments are described in the Supporting Methods. Statistics Statistical analyses were performed with the Statistical Package for Social Sciences version 13.0 (SPSS, Chicago, IL, USA). Statistical significance was established at P
