GASTROENTEROLOGY
1990;94:1303-1306
Multipolar Electrocoagulation Versus Injection Therapy in the Treatment of Bleeding Peptic Ulcers A Prospective, LOREN
Randomized
LAINE
Gastroenteroloav Section. DeDartment of Medicine, Medicine, Loskgeles, California
This study prospectively compares multipolar electrocoagulation and injection therapy in high-risk patients with bleeding ulcers. Patients were considered for entry if they had a bloody nasogastric aspirate, melena, or hematochezia and unstable vital signs, transfusion of r2 U of blood in 12 hours, or a decrease in hematocrit of ~6% in 12 hours. Sixty patients with endoscopic evidence of an ulcer with active bleeding (n = 26) or a nonbleeding visible vessel (n = 34) were randomly assigned to receive multipolar electrocoagulation or injection with absolute ethanol. Hemostasis was achieved in 14 of 14 actively bleeding patients with multipolar electrocoagulation vs. 10 of 12 (63%) treated with injection. No significant differences were observed between electrocoagulation and injection therapy in any parameter assessed during the hospitalization: incidence of further bleeding (6% vs. lo%), units of blood transfused after treatment (1.6r 0.6 vs. 1.3k 0.4), incidence of surgery for bleeding (6% vs. 7%), length of hospital stay in days (5.8 & 0.9 vs. 7.2 f 2.5), cost of hospitalization ($7160 k $1630 vs. $8520 + $2960), or mortality rate (3 % vs. 3 %). Treatment induced bleeding in nonbleeding visible vessels in 35% of subjects in each group, but this was controlled with continued treatment in all patients. One delayed perforation occurred 6 days after multipolar electrocoagulation. Multipolar electrocoagulation and injection therapy are of comparable efficacy in the treatment of patients with clinical evidence of a major upper gastrointestinal bleed and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel. leeding peptic ulcers remain an important health problem despite improvements in diagnosis and medical therapy. Neither the rate of emergency sur-
B
Trial
University
of Southern California School of
gery for bleeding ulcers nor the rate of mortality has declined in the past 30 years (l-3). Recently, however, endoscopic therapy has provided a new method of treatment for patients with bleeding ulcers. A 1989 Consensus Conference at the National Institutes of Health (NIH) concluded that patients with clinical evidence of a major bleed and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel are at high risk for persistent or recurrent bleeding and should receive endoscopic hemostatic therapy (3). The present study is designed to answer the next important question: which endoscopic therapy should be used? Multipolar electrocoagulation has been shown to be effective in the treatment of patients with endoscopic evidence of active upper gastrointestinal hemorrhage or a nonbleeding visible vessel in an ulcer crater (4,5), and the NIH Consensus Conference determined that multipolar electrocoagulation was one of the two “most promising techniques” for the treatment of bleeding ulcers (the other most promising technique was the use of the heater probe] (3). Prospective, controlled trials also have shown that injection therapy is effective in the treatment of bleeding ulcers (6,7), and the NIH Consensus Conference suggested that injection therapy warranted further study (3). This prospective, randomized trial compared multipolar electrocoagulation with injection o,f absolute ethanol for the treatment of patients with major upper gastrointestinal hemorrhage due to an ulcer with active bleeding or a nonbleeding vessel.
Abbreviation used in this paper: NIH, National Institutes of Health. o 1996 by the American Gastraenterological Association 0016-5065/96/$3.00
1304
LOREN LAINE
Patients and Methods Patients were considered for entry if they had a bloody nasogastric aspirate, melena, or hematochezia, and systolic blood pressure 590 mm Hg, heart rate 2110 beats/ min, or orthostatic changes in systolic blood pressure of 220 mm Hg or in heart rate of 220 beats/min; transfusion of ~2 U of blood in 12 hours; or a decrease in hematocrit of ~6% in 12 hours. Patients meeting these clinical criteria underwent upper gastrointestinal tract endoscopy. Patients with active bleeding from an ulcer [defined as a continuous flow of blood (pumping or oozing] for at least 5 minutes of endoscopic observation] or with a nonbleeding visible vessel (defined as a raised, rounded, relatively smooth-surfaced plug) in an ulcer crater were randomly assigned to receive either multipolar electrocoagulation or injection with absolute ethanol. Separate computer-generated randomization sequences were used for active bleeding and nonbleeding visible vessels. Patients were excluded if they were under 18 years of age, were unable or unwilling to give consent, were pregnant, were receiving anticoagulant therapy, had a prothrombin time of less than 30% or a platelet count of less than 50,000/mm3; if endoscopy could not be performed; if no bleeding site could be visualized; if there were two or more bleeding lesions; or if it was likely that performance of endoscopy might jeopardize the patient’s survival. Multipolar electrocoagulation was performed using a 50-W BICAP unit (Circon-ACMI, Stamford, CT) at a setting of 5-6. Two- to lo-second bursts of electrocoagulation were administered to the bleeding point or the nonbleeding visible vessel and the immediately surrounding area with application of a 3%mm BICAP probe as forcefully as possible The mean number of seconds of electrocoagulation per patient was 67 (range, 34-110). Injection therapy was performed using 0.1-0.2 mL of absolute ethanol per injection; a maximum of 2.0 mL was used. At least two injections per quadrant were attempted circumferentially 1-3 mm from the vessel or bleeding point, and a final injection was placed directly into the vessel. Visible vessels were seen to retract and frequently darken; on some occasions, blanching of the vessel and surrounding area was observed. An average of 1.5 mL of ethanol per patient was delivered (range, 1.0-2.0 mL). Endoscopic treatments were performed by the author in all cases, and the patients received only one course of endoscopic therapy. The patients and those caring for them were blinded to the nature of the treatment. Only the endoscopist was aware of a patient’s endoscopic therapy, and he did not participate in the further care of the patient. Physicians were informed of the initial endoscopic findings and of the patient’s bleeding status at the conclusion of treatment. Patients were followed up until discharge or death. An episode of rebleeding was defined as a recurrence of bloody emesis or nasogastric aspirate, melena, or hematochezia, coupled with a recurrence of unstable vital signs (as defined above] or a decrease in hematocrit after stabilization of ~4% in 24 hours. Repeat endoscopies were not mandated by the protocol. Quantitative data (expressed as mean + SE] were com-
GASTROENTEROLOGY Vol.
99, No. 5
pared by a Wilcoxon two-tailed, nonpaired, rank-sum test. Integral data were analyzed with Fisher’s Exact Test. A difference was regarded as significant if P < 0.05. This study was approved by the Research Committee of the Los Angeles County-USC Medical Center, and all patients signed a written consent form.
Results Six hundred five patients meeting the clinical entry criteria were screened. Sixty of these patients, who had an ulcer with active bleeding (n = 26) or a nonbleeding visible vessel (n = 34). were randomly assigned to receive either multipolar electrocoagulation or injection therapy. The other 545 were excluded for one of the following reasons: 169 because they had an ulcer without active bleeding or a nonbleeding visible vessel; 28 because they were entered into a prior trial of endoscopic treatment for nonbleeding visible vessels (5); 248 because they bled from another lesion (varices, 144; Mallory-Weiss tear, 59; hemorrhagic or erosive gastritis, 5; gastric cancer, 8; vascular malformations, 4; other lesions, 28); 30 because no bleeding site was identified: 15 because they had two or more bleeding lesions; 17 because they were unable to give consent; 12 because they refused consent; 1 because their physician refused consent: 12 because they had a severe coagulopathy or thrombocytopenia; and 13 because endoscopy could not be performed (hemodynamic instability, 7; agitation or lack of cooperation, 5; upper esophageal stricture, 1). The two treatment groups did not differ significantly in any of 21 characteristics evaluated at the time of presentation (selected characteristics are listed in Table 1). These characteristics included age, vital signs, transfusion requirements, number of gastric versus duodenal ulcers, use of alcohol or nonsteroidal antiinflammatory drugs, hours from presentation to entry,
Table 1. Selected Characteristics Before Randomization
Characteristics
of the Treatment Multipolar electrocoagulation [n = 311
Age ( vl Heart rate (beats/min] Systolic blood pressure 599 mm Hg (%I Hematocrit (%] Blood transfusion (U) Alcoholics [ %) Duodenal ulcers (%) Gastric ulcers (%) Active bleeding (%) Nonbleeding visible vessels [ %) NOTE. Data are given as mean + SE
Groups
Injection therapy (n = 29)
47 + 3
47 + 2
114 * 4
111 + 3
48
28 k
45 2
1.6 + 0.3
28 k 1 1.8 k 0.3
29
31
71
72
29
28
45
41
55
59
November
COAGULATION
1990
and laboratory values. Patients were relatively young with a mean age of 47 years. At presentation the mean heart rate was greater than 110 beats/min; 95% had “unstable” vital signs as defined earlier, and approximately half had a systolic blood pressure 590 mm Hg. At the time of entry, the patients’ mean transfusion requirement was just less than 2 U of blood. The results after randomization and treatment are shown in Table 2. Hemostasis was achieved in actively bleeding ulcers in all 14 patients treated with multipolar electrocoagulation and in 10 of 12 patients (83%) treated with injection (difference not significant). In the patients with nonbleeding visible vessels, rebleeding was seen in 2 of the 17 patients (12%) treated with multipolar electrocoagulation and in 1 of the 17 (6%) receiving injection therapy. Thus, further bleeding was seen in 2 of the 31 patients (6%) treated with multipolar electrocoagulation and in 3 of the 29 (lo%] treated with injection therapy. No significant differences were found in any of the parameters evaluated after treatment: units of blood transfused, number of patients requiring surgery for bleeding, days in the hospital, days in an intensive care unit, cost of hospitalization, or mortality rate. Patients with active bleeding and those with nonbleeding visible vessels were also analyzed separately. Again, no significant differences were noted between the multipolar electrocoagulation and injection therapy groups in any of the parameters assessed. Bleeding was induced by treatment of a nonbleeding visible vessel in 6 of 17 patients (35%) treated with multipolar electrocoagulation and in 6 of 17 patients (35%) treated with injection therapy. In all cases, bleeding stopped with continued endoscopic treatment. One patient in the multipolar electrocoagulation group had a perforation of a duodenal ulcer 9 days after treatment. This patient was a 64-year-old woman with decompensated cirrhosis and respiratory insufficiency requiring intubation who developed clinical evidence of a perforation 9 days after electrocoagula-
Table 2. Results of Treatment Multipolar electrocoagulation (n = 31) Further bleeding (%) Blood transfusions (U ] Surgery required for bleeding (“I] Hospital stay [days] Intensive care unit stay [days] Hospital cost ($1 Mortality rate (%)
6 1.8 k 0.6 6" 5.8 k 0.9 2.3 f 0.7 7160 + 1630 3
Injection therapy [n = 29) 10 1.3 f 0.4 7 7.2 SC2.5 2.4 + 0.6 8520 f 2960 3
NOTE. Data are given as mean * SE. “One additional patient underwent emergency surgery for a perforated duodenal ulcer 9 days after treatment.
VS. INJECTION
FOR BLEEDING
ULCERS
1305
tion treatment. Emergency surgery was performed, but the patient had postoperative complications including intraabdominal abscesses and died during repeat surgery on her 21st hospital day.
Discussion Endoscopic hemostatic therapy has been used increasingly in recent years as a nonsurgical alternative for the treatment of patients with bleeding ulcers. However, because clinical studies have been limited and sometimes conflicting (3), the use of endoscopic hemostasis has only recently gained formal acceptance (3). Now that the benefit of endoscopic therapy in patients at high risk for persistent or recurrent bleeding has been established (3-7), the next step in the evaluation of endoscopic therapy is to determine which method of hemostasis is preferable in terms of efficacy, safety, and cost. Multipolar electrocoagulation and the heater probe are thermal-contact devices with several theoretical advantages for use in the treatment of bleeding ulcers: (a) they coagulate 100% of canine arteries of less than 2 mm in diameter (8) [Swain et al. report that arteries in 27 human gastric ulcers with visible vessels that rebled averaged 0.7 mm in diameter (range, 0.1-1.8 mm] (9)]; (b) they cause limited tissue injury (lo), thereby decreasing the risk of perforation; (c) they are relatively inexpensive; and (d) they are portable. Multipolar electrocoagulation was used in the present study because its clinical efficacy has been documented at this institution (4,5), making it our standard form of therapy in patients with bleeding peptic ulcers. More recently, injection therapy has been evaluated for use in the treatment of nonvariceal upper gastrointestinal tract hemorrhage. Agents injected include absolute ethanol, polidocanol, and epinephrine. The attraction of injection therapy is its simplicity and low cost-only a sclerotherapy needle and the solution to be injected are necessary. Potential drawbacks to injection therapy are the significant tissue injury caused by sclerosant agents such as absolute ethanol or polidocanol (1~12) and the lack of an irrigation system in treating patients with active bleeding. Both multipolar electrocoagulation (4,5) and injection therapy (6,7) have been shown to be more effective than no endoscopic treatment in high-risk patients with bleeding ulcers. The present study, a prospective, randomized comparison of multipolar electrocoagulation and injection therapy in the treatment of patients with clinical and endoscopic features suggesting a high risk of further bleeding, shows that patients treated with either form of endoscopic therapy do extremely well, with no significant differences be-
GASTROENTEROLOGY
1306 LOREN LAINE
tween the two treatment groups. Less than 10% of the patients required surgery for bleeding, and the mortality rate was 3%. Bleeding was induced by both multipolar electrocoagulation and injection therapy in approximately one third of the nonbleeding visible vessels, but continued treatment stopped the bleeding in all cases. A delayed duodenal ulcer perforation was observed in one patient (hospitalized with hepatic and pulmonary failure] 9 days after multipolar electrocoagulation treatment. Previous reports of perforation after multipolar electrocoagulation treatment for upper gastrointestinal bleeding are extremely rare: Brearley et al. (13) report a patient who developed a duodenal ulcer perforation immediately after treatment [the patient did well without surgical treatment], and Rutgeerts et al. (14) report one patient with a perforated ulcer occurring 4 days after treatment. No perforations were noted in the patients receiving absolute ethanol. Because of the significant tissue injury induced by absolute alcohol (ll,l2), the maximum volume injected must be restricted; a maximum of 2 ml/patient was used in this study. In uncontrolled studies of absolute ethanol injection for hemostasis, the rate of perforation may approach 1% (15). Longterm sequelae of absolute ethanol injection in the gastroduodenal mucosa have not been well evaluated. In summary, both multipolar electrocoagulation and injection therapy seem to be safe and effective forms of treatment for selected high-risk patients with bleeding ulcers; results for the two treatments were comparable for all parameters assessed during hospitalization. Potential advantages of multipolar electrocoagulation over injection of absolute ethanol include induction of less tissue injury and availability of an irrigation system to wash bleeding lesions. The primary advantage of injection over electrocoagulation is that no initial capital outlay is required with injection therapy (the electrocoagulation unit costs $5995). Both multipolar electrocoagulation and injection therapy can be recommended for patients with clinical evidence of major bleeding and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel. References 1. Gustavsson S, Kelly KA, Melton LJ III, Zinsmeister AR. Trends in peptic ulcer surgery: a population-based study in Rochester, Minnesota, 1956-1985. Gastroenterology 1988:94:688-694.
Vol. 99, No. 5
2, Christensen
A, Bousfield R, Christiansen J. Incidence of perforated and bleeding peptic ulcers before and after the introduction of Hz-receptor antagonists. Ann Surg 1988;207:4-6. 3, Therapeutic endoscopy and bleeding ulcers-NIH Consensus Conference. JAMA 1989;262:1369-1372, 4. Laine L. Multipolar electrocoagulation in the treatment of active upper gastrointestinal tract hemorrhage: a prospective controlled trial. N Engl J Med 1987;316:1613-1617. 5. Laine L. Multipolar electrocoagulation in the treatment of peptic ulcers with nonbleeding visible vessels: a prospective controlled trial. Ann Intern Med 1989;110:510-514. 6. Panes J, Viver J, Forne M. Garcia-Olivares E, Marco C, Garau J. Controlled trial of endoscopic sclerosis in bleeding peptic ulcers. Lancet 1987;2:1292-1294. 7. Chung SCS. Leung JWC, Steele RJC, Crofts TJ, Li AKC. Endoscopic injection of adrenaline for actively bleeding ulcers: a randomised trial. Br Med J 1988;296:1631-1633. 8. Johnston JH, Jensen DM, Auth D. Experimental comparison of endoscopic yttrium-aluminum-garnet laser, electrosurgery, and heater probe for canine gut arterial coagulation: importance of compression and avoidance of erosion. Gastroenterology 1987: 92:1101-1108. 9. Swain CP, Storey DW, Bown SG, Heath J, Mills TN, Salmon PR, Northfield TC, Kirkham JS, O’Sullivan JP. Nature of the bleeding vessel in recurrently bleeding gastric ulcers. Gastroenterology 1986;90:595-608. 10.Johnston JH, Jensen DM. Mautner W. Comparison of endoscopic electrocoagulation and laser photocoagulation of bleeding canine gastric ulcers. Gastroenterology 1982;82:904-910. 11.Randall GM, Jensen DM, Hirabayashi K, Machicado GA. Controlled study of different sclerosing agents for coagulation of canine gut arteries. Gastroenterology 1989;96:1274-1281, 12.Rutgeerts P, Geboes K, Vantrappen G. Experimental studies of injection therapy for severe nonvariceal bleeding in dogs. Gastroenterology 1989;97:810-621, 13.Brearley S, Hawker PC, Dykes PW, Keighley MRB. Perendoscopic bipolar diathermy coagulation of visible vessels using a 3.2mm probe-a randomised clinical trial. Endoscopy 1987;19: 160-163. 14.Rutgeerts P, Vantrappen G, Van Hootegem P, Broeckaert L, Janssens J. Coremans G, Geboes K. Neodymium-YAG laser photocoagulation versus multipolar electrocoagulation for the treatment of severely bleeding ulcers: a randomized comparison. Gastrointest Endosc 1987;33:199-202. 15.Asaki S. Endoscopic hemostasis by local absolute ethanol injection for upper G-I tract bleeding-a multi center study. In: Okabe H, Honda T. Ohshiba S. eds. Endoscopic Surgery. New York: Elsevier, 1984:105-116.
Received December 4,1989. Accepted May 10,199O. Address requests for reprints to: Loren Laine, M.D., Gastroenterology Section, Department of Medicine, University of Southern California School of Medicine, 2025Zonal Avenue, Los Angeles, California 90033. The author thanks the nursing and medical staffs of the Medical Admitting Area and the Gastrointestinal Unit of the Los Angeles CountyU.S.C. Medical Center for their assistance and Dr. Hartley Cohen for his thoughtful comments.
1990;94:1303-1306
Multipolar Electrocoagulation Versus Injection Therapy in the Treatment of Bleeding Peptic Ulcers A Prospective, LOREN
Randomized
LAINE
Gastroenteroloav Section. DeDartment of Medicine, Medicine, Loskgeles, California
This study prospectively compares multipolar electrocoagulation and injection therapy in high-risk patients with bleeding ulcers. Patients were considered for entry if they had a bloody nasogastric aspirate, melena, or hematochezia and unstable vital signs, transfusion of r2 U of blood in 12 hours, or a decrease in hematocrit of ~6% in 12 hours. Sixty patients with endoscopic evidence of an ulcer with active bleeding (n = 26) or a nonbleeding visible vessel (n = 34) were randomly assigned to receive multipolar electrocoagulation or injection with absolute ethanol. Hemostasis was achieved in 14 of 14 actively bleeding patients with multipolar electrocoagulation vs. 10 of 12 (63%) treated with injection. No significant differences were observed between electrocoagulation and injection therapy in any parameter assessed during the hospitalization: incidence of further bleeding (6% vs. lo%), units of blood transfused after treatment (1.6r 0.6 vs. 1.3k 0.4), incidence of surgery for bleeding (6% vs. 7%), length of hospital stay in days (5.8 & 0.9 vs. 7.2 f 2.5), cost of hospitalization ($7160 k $1630 vs. $8520 + $2960), or mortality rate (3 % vs. 3 %). Treatment induced bleeding in nonbleeding visible vessels in 35% of subjects in each group, but this was controlled with continued treatment in all patients. One delayed perforation occurred 6 days after multipolar electrocoagulation. Multipolar electrocoagulation and injection therapy are of comparable efficacy in the treatment of patients with clinical evidence of a major upper gastrointestinal bleed and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel. leeding peptic ulcers remain an important health problem despite improvements in diagnosis and medical therapy. Neither the rate of emergency sur-
B
Trial
University
of Southern California School of
gery for bleeding ulcers nor the rate of mortality has declined in the past 30 years (l-3). Recently, however, endoscopic therapy has provided a new method of treatment for patients with bleeding ulcers. A 1989 Consensus Conference at the National Institutes of Health (NIH) concluded that patients with clinical evidence of a major bleed and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel are at high risk for persistent or recurrent bleeding and should receive endoscopic hemostatic therapy (3). The present study is designed to answer the next important question: which endoscopic therapy should be used? Multipolar electrocoagulation has been shown to be effective in the treatment of patients with endoscopic evidence of active upper gastrointestinal hemorrhage or a nonbleeding visible vessel in an ulcer crater (4,5), and the NIH Consensus Conference determined that multipolar electrocoagulation was one of the two “most promising techniques” for the treatment of bleeding ulcers (the other most promising technique was the use of the heater probe] (3). Prospective, controlled trials also have shown that injection therapy is effective in the treatment of bleeding ulcers (6,7), and the NIH Consensus Conference suggested that injection therapy warranted further study (3). This prospective, randomized trial compared multipolar electrocoagulation with injection o,f absolute ethanol for the treatment of patients with major upper gastrointestinal hemorrhage due to an ulcer with active bleeding or a nonbleeding vessel.
Abbreviation used in this paper: NIH, National Institutes of Health. o 1996 by the American Gastraenterological Association 0016-5065/96/$3.00
1304
LOREN LAINE
Patients and Methods Patients were considered for entry if they had a bloody nasogastric aspirate, melena, or hematochezia, and systolic blood pressure 590 mm Hg, heart rate 2110 beats/ min, or orthostatic changes in systolic blood pressure of 220 mm Hg or in heart rate of 220 beats/min; transfusion of ~2 U of blood in 12 hours; or a decrease in hematocrit of ~6% in 12 hours. Patients meeting these clinical criteria underwent upper gastrointestinal tract endoscopy. Patients with active bleeding from an ulcer [defined as a continuous flow of blood (pumping or oozing] for at least 5 minutes of endoscopic observation] or with a nonbleeding visible vessel (defined as a raised, rounded, relatively smooth-surfaced plug) in an ulcer crater were randomly assigned to receive either multipolar electrocoagulation or injection with absolute ethanol. Separate computer-generated randomization sequences were used for active bleeding and nonbleeding visible vessels. Patients were excluded if they were under 18 years of age, were unable or unwilling to give consent, were pregnant, were receiving anticoagulant therapy, had a prothrombin time of less than 30% or a platelet count of less than 50,000/mm3; if endoscopy could not be performed; if no bleeding site could be visualized; if there were two or more bleeding lesions; or if it was likely that performance of endoscopy might jeopardize the patient’s survival. Multipolar electrocoagulation was performed using a 50-W BICAP unit (Circon-ACMI, Stamford, CT) at a setting of 5-6. Two- to lo-second bursts of electrocoagulation were administered to the bleeding point or the nonbleeding visible vessel and the immediately surrounding area with application of a 3%mm BICAP probe as forcefully as possible The mean number of seconds of electrocoagulation per patient was 67 (range, 34-110). Injection therapy was performed using 0.1-0.2 mL of absolute ethanol per injection; a maximum of 2.0 mL was used. At least two injections per quadrant were attempted circumferentially 1-3 mm from the vessel or bleeding point, and a final injection was placed directly into the vessel. Visible vessels were seen to retract and frequently darken; on some occasions, blanching of the vessel and surrounding area was observed. An average of 1.5 mL of ethanol per patient was delivered (range, 1.0-2.0 mL). Endoscopic treatments were performed by the author in all cases, and the patients received only one course of endoscopic therapy. The patients and those caring for them were blinded to the nature of the treatment. Only the endoscopist was aware of a patient’s endoscopic therapy, and he did not participate in the further care of the patient. Physicians were informed of the initial endoscopic findings and of the patient’s bleeding status at the conclusion of treatment. Patients were followed up until discharge or death. An episode of rebleeding was defined as a recurrence of bloody emesis or nasogastric aspirate, melena, or hematochezia, coupled with a recurrence of unstable vital signs (as defined above] or a decrease in hematocrit after stabilization of ~4% in 24 hours. Repeat endoscopies were not mandated by the protocol. Quantitative data (expressed as mean + SE] were com-
GASTROENTEROLOGY Vol.
99, No. 5
pared by a Wilcoxon two-tailed, nonpaired, rank-sum test. Integral data were analyzed with Fisher’s Exact Test. A difference was regarded as significant if P < 0.05. This study was approved by the Research Committee of the Los Angeles County-USC Medical Center, and all patients signed a written consent form.
Results Six hundred five patients meeting the clinical entry criteria were screened. Sixty of these patients, who had an ulcer with active bleeding (n = 26) or a nonbleeding visible vessel (n = 34). were randomly assigned to receive either multipolar electrocoagulation or injection therapy. The other 545 were excluded for one of the following reasons: 169 because they had an ulcer without active bleeding or a nonbleeding visible vessel; 28 because they were entered into a prior trial of endoscopic treatment for nonbleeding visible vessels (5); 248 because they bled from another lesion (varices, 144; Mallory-Weiss tear, 59; hemorrhagic or erosive gastritis, 5; gastric cancer, 8; vascular malformations, 4; other lesions, 28); 30 because no bleeding site was identified: 15 because they had two or more bleeding lesions; 17 because they were unable to give consent; 12 because they refused consent; 1 because their physician refused consent: 12 because they had a severe coagulopathy or thrombocytopenia; and 13 because endoscopy could not be performed (hemodynamic instability, 7; agitation or lack of cooperation, 5; upper esophageal stricture, 1). The two treatment groups did not differ significantly in any of 21 characteristics evaluated at the time of presentation (selected characteristics are listed in Table 1). These characteristics included age, vital signs, transfusion requirements, number of gastric versus duodenal ulcers, use of alcohol or nonsteroidal antiinflammatory drugs, hours from presentation to entry,
Table 1. Selected Characteristics Before Randomization
Characteristics
of the Treatment Multipolar electrocoagulation [n = 311
Age ( vl Heart rate (beats/min] Systolic blood pressure 599 mm Hg (%I Hematocrit (%] Blood transfusion (U) Alcoholics [ %) Duodenal ulcers (%) Gastric ulcers (%) Active bleeding (%) Nonbleeding visible vessels [ %) NOTE. Data are given as mean + SE
Groups
Injection therapy (n = 29)
47 + 3
47 + 2
114 * 4
111 + 3
48
28 k
45 2
1.6 + 0.3
28 k 1 1.8 k 0.3
29
31
71
72
29
28
45
41
55
59
November
COAGULATION
1990
and laboratory values. Patients were relatively young with a mean age of 47 years. At presentation the mean heart rate was greater than 110 beats/min; 95% had “unstable” vital signs as defined earlier, and approximately half had a systolic blood pressure 590 mm Hg. At the time of entry, the patients’ mean transfusion requirement was just less than 2 U of blood. The results after randomization and treatment are shown in Table 2. Hemostasis was achieved in actively bleeding ulcers in all 14 patients treated with multipolar electrocoagulation and in 10 of 12 patients (83%) treated with injection (difference not significant). In the patients with nonbleeding visible vessels, rebleeding was seen in 2 of the 17 patients (12%) treated with multipolar electrocoagulation and in 1 of the 17 (6%) receiving injection therapy. Thus, further bleeding was seen in 2 of the 31 patients (6%) treated with multipolar electrocoagulation and in 3 of the 29 (lo%] treated with injection therapy. No significant differences were found in any of the parameters evaluated after treatment: units of blood transfused, number of patients requiring surgery for bleeding, days in the hospital, days in an intensive care unit, cost of hospitalization, or mortality rate. Patients with active bleeding and those with nonbleeding visible vessels were also analyzed separately. Again, no significant differences were noted between the multipolar electrocoagulation and injection therapy groups in any of the parameters assessed. Bleeding was induced by treatment of a nonbleeding visible vessel in 6 of 17 patients (35%) treated with multipolar electrocoagulation and in 6 of 17 patients (35%) treated with injection therapy. In all cases, bleeding stopped with continued endoscopic treatment. One patient in the multipolar electrocoagulation group had a perforation of a duodenal ulcer 9 days after treatment. This patient was a 64-year-old woman with decompensated cirrhosis and respiratory insufficiency requiring intubation who developed clinical evidence of a perforation 9 days after electrocoagula-
Table 2. Results of Treatment Multipolar electrocoagulation (n = 31) Further bleeding (%) Blood transfusions (U ] Surgery required for bleeding (“I] Hospital stay [days] Intensive care unit stay [days] Hospital cost ($1 Mortality rate (%)
6 1.8 k 0.6 6" 5.8 k 0.9 2.3 f 0.7 7160 + 1630 3
Injection therapy [n = 29) 10 1.3 f 0.4 7 7.2 SC2.5 2.4 + 0.6 8520 f 2960 3
NOTE. Data are given as mean * SE. “One additional patient underwent emergency surgery for a perforated duodenal ulcer 9 days after treatment.
VS. INJECTION
FOR BLEEDING
ULCERS
1305
tion treatment. Emergency surgery was performed, but the patient had postoperative complications including intraabdominal abscesses and died during repeat surgery on her 21st hospital day.
Discussion Endoscopic hemostatic therapy has been used increasingly in recent years as a nonsurgical alternative for the treatment of patients with bleeding ulcers. However, because clinical studies have been limited and sometimes conflicting (3), the use of endoscopic hemostasis has only recently gained formal acceptance (3). Now that the benefit of endoscopic therapy in patients at high risk for persistent or recurrent bleeding has been established (3-7), the next step in the evaluation of endoscopic therapy is to determine which method of hemostasis is preferable in terms of efficacy, safety, and cost. Multipolar electrocoagulation and the heater probe are thermal-contact devices with several theoretical advantages for use in the treatment of bleeding ulcers: (a) they coagulate 100% of canine arteries of less than 2 mm in diameter (8) [Swain et al. report that arteries in 27 human gastric ulcers with visible vessels that rebled averaged 0.7 mm in diameter (range, 0.1-1.8 mm] (9)]; (b) they cause limited tissue injury (lo), thereby decreasing the risk of perforation; (c) they are relatively inexpensive; and (d) they are portable. Multipolar electrocoagulation was used in the present study because its clinical efficacy has been documented at this institution (4,5), making it our standard form of therapy in patients with bleeding peptic ulcers. More recently, injection therapy has been evaluated for use in the treatment of nonvariceal upper gastrointestinal tract hemorrhage. Agents injected include absolute ethanol, polidocanol, and epinephrine. The attraction of injection therapy is its simplicity and low cost-only a sclerotherapy needle and the solution to be injected are necessary. Potential drawbacks to injection therapy are the significant tissue injury caused by sclerosant agents such as absolute ethanol or polidocanol (1~12) and the lack of an irrigation system in treating patients with active bleeding. Both multipolar electrocoagulation (4,5) and injection therapy (6,7) have been shown to be more effective than no endoscopic treatment in high-risk patients with bleeding ulcers. The present study, a prospective, randomized comparison of multipolar electrocoagulation and injection therapy in the treatment of patients with clinical and endoscopic features suggesting a high risk of further bleeding, shows that patients treated with either form of endoscopic therapy do extremely well, with no significant differences be-
GASTROENTEROLOGY
1306 LOREN LAINE
tween the two treatment groups. Less than 10% of the patients required surgery for bleeding, and the mortality rate was 3%. Bleeding was induced by both multipolar electrocoagulation and injection therapy in approximately one third of the nonbleeding visible vessels, but continued treatment stopped the bleeding in all cases. A delayed duodenal ulcer perforation was observed in one patient (hospitalized with hepatic and pulmonary failure] 9 days after multipolar electrocoagulation treatment. Previous reports of perforation after multipolar electrocoagulation treatment for upper gastrointestinal bleeding are extremely rare: Brearley et al. (13) report a patient who developed a duodenal ulcer perforation immediately after treatment [the patient did well without surgical treatment], and Rutgeerts et al. (14) report one patient with a perforated ulcer occurring 4 days after treatment. No perforations were noted in the patients receiving absolute ethanol. Because of the significant tissue injury induced by absolute alcohol (ll,l2), the maximum volume injected must be restricted; a maximum of 2 ml/patient was used in this study. In uncontrolled studies of absolute ethanol injection for hemostasis, the rate of perforation may approach 1% (15). Longterm sequelae of absolute ethanol injection in the gastroduodenal mucosa have not been well evaluated. In summary, both multipolar electrocoagulation and injection therapy seem to be safe and effective forms of treatment for selected high-risk patients with bleeding ulcers; results for the two treatments were comparable for all parameters assessed during hospitalization. Potential advantages of multipolar electrocoagulation over injection of absolute ethanol include induction of less tissue injury and availability of an irrigation system to wash bleeding lesions. The primary advantage of injection over electrocoagulation is that no initial capital outlay is required with injection therapy (the electrocoagulation unit costs $5995). Both multipolar electrocoagulation and injection therapy can be recommended for patients with clinical evidence of major bleeding and endoscopic evidence of an ulcer with active bleeding or a nonbleeding visible vessel. References 1. Gustavsson S, Kelly KA, Melton LJ III, Zinsmeister AR. Trends in peptic ulcer surgery: a population-based study in Rochester, Minnesota, 1956-1985. Gastroenterology 1988:94:688-694.
Vol. 99, No. 5
2, Christensen
A, Bousfield R, Christiansen J. Incidence of perforated and bleeding peptic ulcers before and after the introduction of Hz-receptor antagonists. Ann Surg 1988;207:4-6. 3, Therapeutic endoscopy and bleeding ulcers-NIH Consensus Conference. JAMA 1989;262:1369-1372, 4. Laine L. Multipolar electrocoagulation in the treatment of active upper gastrointestinal tract hemorrhage: a prospective controlled trial. N Engl J Med 1987;316:1613-1617. 5. Laine L. Multipolar electrocoagulation in the treatment of peptic ulcers with nonbleeding visible vessels: a prospective controlled trial. Ann Intern Med 1989;110:510-514. 6. Panes J, Viver J, Forne M. Garcia-Olivares E, Marco C, Garau J. Controlled trial of endoscopic sclerosis in bleeding peptic ulcers. Lancet 1987;2:1292-1294. 7. Chung SCS. Leung JWC, Steele RJC, Crofts TJ, Li AKC. Endoscopic injection of adrenaline for actively bleeding ulcers: a randomised trial. Br Med J 1988;296:1631-1633. 8. Johnston JH, Jensen DM, Auth D. Experimental comparison of endoscopic yttrium-aluminum-garnet laser, electrosurgery, and heater probe for canine gut arterial coagulation: importance of compression and avoidance of erosion. Gastroenterology 1987: 92:1101-1108. 9. Swain CP, Storey DW, Bown SG, Heath J, Mills TN, Salmon PR, Northfield TC, Kirkham JS, O’Sullivan JP. Nature of the bleeding vessel in recurrently bleeding gastric ulcers. Gastroenterology 1986;90:595-608. 10.Johnston JH, Jensen DM. Mautner W. Comparison of endoscopic electrocoagulation and laser photocoagulation of bleeding canine gastric ulcers. Gastroenterology 1982;82:904-910. 11.Randall GM, Jensen DM, Hirabayashi K, Machicado GA. Controlled study of different sclerosing agents for coagulation of canine gut arteries. Gastroenterology 1989;96:1274-1281, 12.Rutgeerts P, Geboes K, Vantrappen G. Experimental studies of injection therapy for severe nonvariceal bleeding in dogs. Gastroenterology 1989;97:810-621, 13.Brearley S, Hawker PC, Dykes PW, Keighley MRB. Perendoscopic bipolar diathermy coagulation of visible vessels using a 3.2mm probe-a randomised clinical trial. Endoscopy 1987;19: 160-163. 14.Rutgeerts P, Vantrappen G, Van Hootegem P, Broeckaert L, Janssens J. Coremans G, Geboes K. Neodymium-YAG laser photocoagulation versus multipolar electrocoagulation for the treatment of severely bleeding ulcers: a randomized comparison. Gastrointest Endosc 1987;33:199-202. 15.Asaki S. Endoscopic hemostasis by local absolute ethanol injection for upper G-I tract bleeding-a multi center study. In: Okabe H, Honda T. Ohshiba S. eds. Endoscopic Surgery. New York: Elsevier, 1984:105-116.
Received December 4,1989. Accepted May 10,199O. Address requests for reprints to: Loren Laine, M.D., Gastroenterology Section, Department of Medicine, University of Southern California School of Medicine, 2025Zonal Avenue, Los Angeles, California 90033. The author thanks the nursing and medical staffs of the Medical Admitting Area and the Gastrointestinal Unit of the Los Angeles CountyU.S.C. Medical Center for their assistance and Dr. Hartley Cohen for his thoughtful comments.
