- Email: [email protected]
Intravenous vasopressin and gastrointestinal hemorrhage in children
Intravenous Vasopressin and Gastrointestinal Hemorrhage in Children By David W. Tuggle, Keith G. Bennett, Joni Scott, and William P. Tunell O k l a h o m a City, O k l a h o m a 9 Intravenous (IV) vasopressin has been used to control human upper gastrointestinal (GI) hemorrhage for over 30 years. Although the use of vasopressin has been studied extensively in adults, no study has evaluated its use in children. Vasopressin was used therapeutically in 15 episodes of esophageal variceal hemorrhage and two episodes of bleeding peptic ulcer. Nine of 17 episodes w e r e controlled with vasopressin alone (53%). Balloon tamponade and variceal sclerosis w e r e required for control in t h e remainder. Blood requirements averaged 53 m L / k g prior to control of hemorrhage. Metabolic complications occurred in 65% of the episodes. There w e r e two groups of patients identified: those receiving greater or those receiving less than .01 u n i t s / k g / m i n of IV vasopressin. All of the complications identified occurred when >.01 U / k g / m i n of vasopressin were used (P < .05). Control of bleeding was not improved with higher doses of vasopressin. These data suggest that the use of IV vasopressin at doses >.01 U / k g / m i n to control GI bleeding will increase the incidence of complications without improving control of hemorrhage. 9 1988
by Grune
& Stratton. Inc.
INDEX WORDS: Esophageal varices; vasopressin; gastrointestinal hemorrhage.
N T R A V E N O U S (IV) vasopressin has been used extensively in adults and children to control gastrointestinal (GI) bleeding) 4 Although several prospective double-blind, randomized studies have evaluated the effectiveness of vasopressin on controlling GI bleeding in adults, ~there have been no studies evaluating its effectiveness or the appropriate dosage for children. 3 The recommended dose of IV vasopressin for children varies significantly and is based on adult studies, z4 We evaluated our own experience using IV vasopressin to control GI hemorrhage in children in an attempt to standardize therapy.
I
MATERIALS AND METHODS Patients with upper GI bleeding episodes were included in this study if they met the following criteria: 1. Intravenous vasopressin must have been the initial therapeutic intervention if gastric lavage was unsuccessful. 2. Transfusion requirements were >20 m L / k g of packed RBC prior to instituting treatment. 3. If a previous bleeding episode during the same hospitalization had occurred, more than 24 hours should have elapsed since the cessation of hemorrhage and treatment, and the patient's coagulation profile was normal. 4. The patients were on no therapy of any kind to prevent hemorrhage. Between July 1, 1979, and February 1, 1987, 14 patients with 30 episodes of GI bleeding required vasopressin. Eight patients with 17 bleeding episodes met the criteria for inclusion in this study. Journal of Pediatric Surgery, Vo123, No 7 (July), 1988: pp 627-629
Intravenous vasopressin was used in each instance. Success was defined as cessation of hemorrhage within 24 hours as manifested by a clearing nasogastric lavage, stable pulse, BP, and hematocrit with the use of only IV vasopressin for control of hemorrhage. If control of bleeding was not established within 24 hours, or if excessive hemorrhage indicated the urgent need for additional therapeutic interventions, vasopressin therapy was thought to have failed and control was established by other methods. An unpaired Student's t test or Fisher's exact test was used for statistical comparison when appropriate. Values are expressed as mean and SE of the mean.
RESULTS
The mean patient age was 7.2 years, the mean weight was 22.4 kg. Vasopressin was delivered IV by peripheral vein or central venous catheter. Nine bleeding episodes were controlled with vasopressin alone (53%). Bleeding stopped in these patients between two and 24 hours (9.8 _+ 2.4 hours) after starting therapy. Blood requirements averaged 53 _+ 8 m L / k g prior to control of hemorrhage. There was no significant difference in transfusion requirements between those patients in whom control was established and those patients who failed control by vasopressin alone. Fifteen bleeding episodes were due to esophageal varices. Two episodes of hemorrhage were due to peptic ulcer. Eight of 15 episodes of esophageal variceal bleeding and one of two episodes of bleeding peptic ulcer were controlled with vasopressin alone. Bradycardia occurred in 14 of 17 episodes (82%). This was asymptomatic in all but one patient. Other complications occurred in 65% of the bleeding episodes (Table 1). Patients were begun on doses of vasopressin (0.1 to 0.2 U/min) as described in the literature. 2 The dose of vasopressin was increased if control of bleeding was not achieved expeditiously (one to two hours). When comparing patients with successful and failed therapy, there was no significant difference in the length of time vasopressin was used (39 _+ 8 hours v 49 _+ 9 hours). The maximum dosage received during each bleeding From the Section of Pediatric Surgery, Department of Surgery, The University of Oklahoma College of Medicine and the Oklahoma Children's Memorial Hospital, Oklahoma City. Presented at the 36th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, New Orleans, October 31 to November 2, 1987. Address reprint requests to David IV. Tuggle, MD, Pediatric Surgery Section, PO Box 26307, Oklahoma City, OK 73126. 9 1988 by Grune & Stratton, Inc. 0022-3468/88/2307-0006503.00/0 627
628
TUGGLE ET AL
Table 1. Type of Complication and Frequency of Occurrence in Patients Receiving Vasopressin Complication Electrolyte abnormalities (Na, K, CI, or Ca) Fluid overload Hypertension Cardiac dysrhythmias
Frequency 10 4 4 2
NOTE. Some patients had more than one type of complication.
episode was used to calculate the dose of vasopressin delivered per kilogram per minute. The maximum delivered dosage ranged from 0.004 U/kg/min to 0.04 U/kg/min. Patients were divided into two groups: those receiving >_0.01 U/kg/min (high dose) or <0.01 U/kg/min (low dose). These two groups were compared with respect to the success of therapy with vasopressin alone and the incidence of complications (Table 2). Control of bleeding did not improve with high dose vasopressin and there was a significantly greater incidence of complications in those patients receiving >_0.01 U/kg/min (P < .05). There was no significant difference in the length of time vasopressin was used in the high dose and low dose groups (49 _+ 8 hours v 39 _+ 7.6 hours). DISCUSSION
These data suggest that vasopressin alone may be useful in controlling upper GI bleeding in some children and that there is a rate of administration that will allow control of bleeding while minimizing complications in these difficult to manage patients. Apparently, if <0.01 U/kg/min of IV vasopressin is given to a child with GI bleeding, control can be frequently established without metabolic complications. This conclusion is supported extensively in the adult experimental and clinical literature. 1 We recommend starting vasopressin at 0.002 to 0.005 U/kg/min, which would allow increasing the dose of vasopressin before the risk of complications increases. Recommendations for the dose of vasopressin have been controversial in adult studies ~ and have been derived from adult data in pediatric reports. 2-4 In adult dogs, low dose vasopressin (0.003 U / k g / m i n ) decreases superior mesenteric artery flow by 51%, Table 2. Bleeding Episodes Distributed by Maximum Dosage of Vasopressin No. of Patients Receiving -->0.01 U/kg/min* No. of episodes No. of successes Complications
11 4 (36%) 11
No, of Patients Receiving <0.01 U/kg/min 6 5 (83%) 0 (P < .05)
*Note low success rate and high number of complications in patients receiving >_0.01 U/kg/min.
portal venous flow by 56%, and portal venous pressure by 73%.5 These effects are 85% of the decrease achieved with high dose vasopressin (0.014 U/kg/ min). However, cardiac output remained 40% higher in the dogs given low dose vasopressin. No difference in the response of portal pressure was noted between dogs with and without acute portal hypertension. High dose vasopressin decreased coronary blood flow, cardiac output, and myocardia contractility to a greater degree than did a lower dose. ~ This experimental data supports the concept of using lower doses of vasopressin to achieve control of bleeding while reducing the risk of complications. Other drugs can be used with or instead of vasopressin to control GI hemorrhage. A recent randomized study evaluated the use of vasopressin plus nitroglycerin in the control of variceal bleeding.6 These investigators found that the addition of IV nitroglycerin to vasopressin significantly improved control of bleeding while decreasing the incidence of complications. Although we have not used this combination of drugs, we feel it may be useful in patients with difficult-to-control bleeding. Other methods of pharmacologic treatment on the horizon include somatostatin and triglycyl vasopressinY '7 These new drugs may also improve control of bleeding while decreasing complications. When pharmacologic therapy fails, direct methods of control must be utilized. A pediatric-size Sengstaken-Blakemore tube can be used to arrest bleeding from esophageal varices and was used in three of our patients. Little has been written concerning the use of this tube in children.24 Our protocol using this adjunct involves placing the tube through the mouth into the stomach, inflating the gastric balloon with 30 mL air, and confirming the tube position with an x-ray. This step may decrease the risk of intraesophageal balloon inflation and esophageal rupture. If the balloon is in the proper position, a total of 150 to 175 mL of air is injected (depending on the patient's size) and tension is placed on the tube to control bleeding. A padded helmet is placed on the patient's head and the tube is taped to the helmet. We intubate our patients who require a Blakemore tube. This reduces the risk of respiratory arrest or aspiration while using this tube in small children. Although vasopressin and balloon tamponade will frequently control bleeding temporarily, definitive therapy is generally required in these patients. We prefer elective esophageal variceal sclerotherapy as has been described by Lilly.8 However, four of our patients required emergency sclerotherapy to halt bleeding. We use a flexible endoscope to perform this procedure and sodium morrhuate as a sclerosant.
VASOPRESSIN USE IN CHILDREN
629
This experience suggests a starting point for vasopressin (0.002 U / k g / m i n ) therapy in children and, perhaps more importantly, a point at which to stop and alter treatment (>0.01 U / k g / m i n ) . REFERENCES 1. Hussey KP: Vasopressin therapy for upper gastrointestinal tract hemorrhage--Has its efficacy been proven? Arch Intern Med 145:1263-1267, 1985 2. O'Neill JA Jr: Portal hypertension in children, in Dean RH, O'Neill JA (eds): Vascular Disorders of Childhood. Philadelphia, Lea & Febiger, 1983, p 148 3. Hyams JA, Leichtner AM, Schwartz AN: Recent advances in
diagnosis and treatment of gastrointestinal hemorrhage in infants and children. J Pediatr 106:1-9, 1985 4. Altman RP: Portal hypertension, in Welch K J, Randolph JG, Ravitch MM, et al (eds): Pediatric Surgery (ed 4). Chicago, Year Book, 1986, pp 1075-1086 5. Baur J, Laken R, Rosch J: Similarity of arterial and intravenous vasopressin on portal and systemic hemodynamics. Gastroenterology 69:13-19, 1975 6. Gimson AE, Westaby D, Hegarty J, et al: A randomized trial of vasopressin and vasopressin plus nitroglycerin in the control of acute variceal hemorrhage. Hepatology 6:410-413, 1986 7. Rector WG: Drug therapy for portal hypertension. Ann lnt Med 105:96-107, 1986 8. Lilly JR: Endoscopic sclerosis of esophageal varices in children. Surg Gynecol Obstet 152:513-514, 1981
Discussion G.W. Holcomb, III (Philadelphia): It would seem that your complications were manageable, yet if I interpreted your slide correctly, 36% of your patients on high dose vasopressin therapy were successfully managed. Therefore, it seemed that if one third of the patients were successfully managed with high dose vasopressin, would not this be a reasonable route to try initially if low dose vasopressin fails, and then go on to something else? David W. Tuggle (closing): We could not identify any patient who would seem to have benefited from more vasopressin in this study. It is a small group of patients. A prospective study needs to be done to
evaluate this mode of therapy. I think it is reasonable to use high dose vasopressin if you feel that you need to and cannot control bleeding with, or if you would have a difficult time placing, a Sengstaken-Blakemore tube, or difficulty in performing variceal sclerosis. I think it is not unreasonable to go on to high dose vasopressin recognizing that some of these complications may start to occur. This method has been used to control bleeding in three more patients since submission of this abstract. This difference in high dose and low dose vasopressin has persisted, and we believe that it is going to be a reasonably valid way to use vasopressin in these children.
by Grune
& Stratton. Inc.
INDEX WORDS: Esophageal varices; vasopressin; gastrointestinal hemorrhage.
N T R A V E N O U S (IV) vasopressin has been used extensively in adults and children to control gastrointestinal (GI) bleeding) 4 Although several prospective double-blind, randomized studies have evaluated the effectiveness of vasopressin on controlling GI bleeding in adults, ~there have been no studies evaluating its effectiveness or the appropriate dosage for children. 3 The recommended dose of IV vasopressin for children varies significantly and is based on adult studies, z4 We evaluated our own experience using IV vasopressin to control GI hemorrhage in children in an attempt to standardize therapy.
I
MATERIALS AND METHODS Patients with upper GI bleeding episodes were included in this study if they met the following criteria: 1. Intravenous vasopressin must have been the initial therapeutic intervention if gastric lavage was unsuccessful. 2. Transfusion requirements were >20 m L / k g of packed RBC prior to instituting treatment. 3. If a previous bleeding episode during the same hospitalization had occurred, more than 24 hours should have elapsed since the cessation of hemorrhage and treatment, and the patient's coagulation profile was normal. 4. The patients were on no therapy of any kind to prevent hemorrhage. Between July 1, 1979, and February 1, 1987, 14 patients with 30 episodes of GI bleeding required vasopressin. Eight patients with 17 bleeding episodes met the criteria for inclusion in this study. Journal of Pediatric Surgery, Vo123, No 7 (July), 1988: pp 627-629
Intravenous vasopressin was used in each instance. Success was defined as cessation of hemorrhage within 24 hours as manifested by a clearing nasogastric lavage, stable pulse, BP, and hematocrit with the use of only IV vasopressin for control of hemorrhage. If control of bleeding was not established within 24 hours, or if excessive hemorrhage indicated the urgent need for additional therapeutic interventions, vasopressin therapy was thought to have failed and control was established by other methods. An unpaired Student's t test or Fisher's exact test was used for statistical comparison when appropriate. Values are expressed as mean and SE of the mean.
RESULTS
The mean patient age was 7.2 years, the mean weight was 22.4 kg. Vasopressin was delivered IV by peripheral vein or central venous catheter. Nine bleeding episodes were controlled with vasopressin alone (53%). Bleeding stopped in these patients between two and 24 hours (9.8 _+ 2.4 hours) after starting therapy. Blood requirements averaged 53 _+ 8 m L / k g prior to control of hemorrhage. There was no significant difference in transfusion requirements between those patients in whom control was established and those patients who failed control by vasopressin alone. Fifteen bleeding episodes were due to esophageal varices. Two episodes of hemorrhage were due to peptic ulcer. Eight of 15 episodes of esophageal variceal bleeding and one of two episodes of bleeding peptic ulcer were controlled with vasopressin alone. Bradycardia occurred in 14 of 17 episodes (82%). This was asymptomatic in all but one patient. Other complications occurred in 65% of the bleeding episodes (Table 1). Patients were begun on doses of vasopressin (0.1 to 0.2 U/min) as described in the literature. 2 The dose of vasopressin was increased if control of bleeding was not achieved expeditiously (one to two hours). When comparing patients with successful and failed therapy, there was no significant difference in the length of time vasopressin was used (39 _+ 8 hours v 49 _+ 9 hours). The maximum dosage received during each bleeding From the Section of Pediatric Surgery, Department of Surgery, The University of Oklahoma College of Medicine and the Oklahoma Children's Memorial Hospital, Oklahoma City. Presented at the 36th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, New Orleans, October 31 to November 2, 1987. Address reprint requests to David IV. Tuggle, MD, Pediatric Surgery Section, PO Box 26307, Oklahoma City, OK 73126. 9 1988 by Grune & Stratton, Inc. 0022-3468/88/2307-0006503.00/0 627
628
TUGGLE ET AL
Table 1. Type of Complication and Frequency of Occurrence in Patients Receiving Vasopressin Complication Electrolyte abnormalities (Na, K, CI, or Ca) Fluid overload Hypertension Cardiac dysrhythmias
Frequency 10 4 4 2
NOTE. Some patients had more than one type of complication.
episode was used to calculate the dose of vasopressin delivered per kilogram per minute. The maximum delivered dosage ranged from 0.004 U/kg/min to 0.04 U/kg/min. Patients were divided into two groups: those receiving >_0.01 U/kg/min (high dose) or <0.01 U/kg/min (low dose). These two groups were compared with respect to the success of therapy with vasopressin alone and the incidence of complications (Table 2). Control of bleeding did not improve with high dose vasopressin and there was a significantly greater incidence of complications in those patients receiving >_0.01 U/kg/min (P < .05). There was no significant difference in the length of time vasopressin was used in the high dose and low dose groups (49 _+ 8 hours v 39 _+ 7.6 hours). DISCUSSION
These data suggest that vasopressin alone may be useful in controlling upper GI bleeding in some children and that there is a rate of administration that will allow control of bleeding while minimizing complications in these difficult to manage patients. Apparently, if <0.01 U/kg/min of IV vasopressin is given to a child with GI bleeding, control can be frequently established without metabolic complications. This conclusion is supported extensively in the adult experimental and clinical literature. 1 We recommend starting vasopressin at 0.002 to 0.005 U/kg/min, which would allow increasing the dose of vasopressin before the risk of complications increases. Recommendations for the dose of vasopressin have been controversial in adult studies ~ and have been derived from adult data in pediatric reports. 2-4 In adult dogs, low dose vasopressin (0.003 U / k g / m i n ) decreases superior mesenteric artery flow by 51%, Table 2. Bleeding Episodes Distributed by Maximum Dosage of Vasopressin No. of Patients Receiving -->0.01 U/kg/min* No. of episodes No. of successes Complications
11 4 (36%) 11
No, of Patients Receiving <0.01 U/kg/min 6 5 (83%) 0 (P < .05)
*Note low success rate and high number of complications in patients receiving >_0.01 U/kg/min.
portal venous flow by 56%, and portal venous pressure by 73%.5 These effects are 85% of the decrease achieved with high dose vasopressin (0.014 U/kg/ min). However, cardiac output remained 40% higher in the dogs given low dose vasopressin. No difference in the response of portal pressure was noted between dogs with and without acute portal hypertension. High dose vasopressin decreased coronary blood flow, cardiac output, and myocardia contractility to a greater degree than did a lower dose. ~ This experimental data supports the concept of using lower doses of vasopressin to achieve control of bleeding while reducing the risk of complications. Other drugs can be used with or instead of vasopressin to control GI hemorrhage. A recent randomized study evaluated the use of vasopressin plus nitroglycerin in the control of variceal bleeding.6 These investigators found that the addition of IV nitroglycerin to vasopressin significantly improved control of bleeding while decreasing the incidence of complications. Although we have not used this combination of drugs, we feel it may be useful in patients with difficult-to-control bleeding. Other methods of pharmacologic treatment on the horizon include somatostatin and triglycyl vasopressinY '7 These new drugs may also improve control of bleeding while decreasing complications. When pharmacologic therapy fails, direct methods of control must be utilized. A pediatric-size Sengstaken-Blakemore tube can be used to arrest bleeding from esophageal varices and was used in three of our patients. Little has been written concerning the use of this tube in children.24 Our protocol using this adjunct involves placing the tube through the mouth into the stomach, inflating the gastric balloon with 30 mL air, and confirming the tube position with an x-ray. This step may decrease the risk of intraesophageal balloon inflation and esophageal rupture. If the balloon is in the proper position, a total of 150 to 175 mL of air is injected (depending on the patient's size) and tension is placed on the tube to control bleeding. A padded helmet is placed on the patient's head and the tube is taped to the helmet. We intubate our patients who require a Blakemore tube. This reduces the risk of respiratory arrest or aspiration while using this tube in small children. Although vasopressin and balloon tamponade will frequently control bleeding temporarily, definitive therapy is generally required in these patients. We prefer elective esophageal variceal sclerotherapy as has been described by Lilly.8 However, four of our patients required emergency sclerotherapy to halt bleeding. We use a flexible endoscope to perform this procedure and sodium morrhuate as a sclerosant.
VASOPRESSIN USE IN CHILDREN
629
This experience suggests a starting point for vasopressin (0.002 U / k g / m i n ) therapy in children and, perhaps more importantly, a point at which to stop and alter treatment (>0.01 U / k g / m i n ) . REFERENCES 1. Hussey KP: Vasopressin therapy for upper gastrointestinal tract hemorrhage--Has its efficacy been proven? Arch Intern Med 145:1263-1267, 1985 2. O'Neill JA Jr: Portal hypertension in children, in Dean RH, O'Neill JA (eds): Vascular Disorders of Childhood. Philadelphia, Lea & Febiger, 1983, p 148 3. Hyams JA, Leichtner AM, Schwartz AN: Recent advances in
diagnosis and treatment of gastrointestinal hemorrhage in infants and children. J Pediatr 106:1-9, 1985 4. Altman RP: Portal hypertension, in Welch K J, Randolph JG, Ravitch MM, et al (eds): Pediatric Surgery (ed 4). Chicago, Year Book, 1986, pp 1075-1086 5. Baur J, Laken R, Rosch J: Similarity of arterial and intravenous vasopressin on portal and systemic hemodynamics. Gastroenterology 69:13-19, 1975 6. Gimson AE, Westaby D, Hegarty J, et al: A randomized trial of vasopressin and vasopressin plus nitroglycerin in the control of acute variceal hemorrhage. Hepatology 6:410-413, 1986 7. Rector WG: Drug therapy for portal hypertension. Ann lnt Med 105:96-107, 1986 8. Lilly JR: Endoscopic sclerosis of esophageal varices in children. Surg Gynecol Obstet 152:513-514, 1981
Discussion G.W. Holcomb, III (Philadelphia): It would seem that your complications were manageable, yet if I interpreted your slide correctly, 36% of your patients on high dose vasopressin therapy were successfully managed. Therefore, it seemed that if one third of the patients were successfully managed with high dose vasopressin, would not this be a reasonable route to try initially if low dose vasopressin fails, and then go on to something else? David W. Tuggle (closing): We could not identify any patient who would seem to have benefited from more vasopressin in this study. It is a small group of patients. A prospective study needs to be done to
evaluate this mode of therapy. I think it is reasonable to use high dose vasopressin if you feel that you need to and cannot control bleeding with, or if you would have a difficult time placing, a Sengstaken-Blakemore tube, or difficulty in performing variceal sclerosis. I think it is not unreasonable to go on to high dose vasopressin recognizing that some of these complications may start to occur. This method has been used to control bleeding in three more patients since submission of this abstract. This difference in high dose and low dose vasopressin has persisted, and we believe that it is going to be a reasonably valid way to use vasopressin in these children.