A Comparison of Sucralfate with Placebo in the Treatment of Esophageal Ulcers Following Therapeutic Endoscopic Sclerotherapy of Esophageal Varices -A Prospective Controlled Randomized Trial KARL-JOSEPH PAQUET, M.D., PAUL KOUSSOURIS,M.D., ROLF KEINATH, M.D., WALTER RAMBACH, M.D., KALK, M.D., Bad Kmingen, Germany
JOHANN-FRIEDRICH
In 1984 Roark published the first report of a sucralfate treatment of esophageal ulcers after sclerotherapy. Because this was an uncontrolled trial we planned a prospective doubleblind placebo-controlled study with 60 patients. After therapeutic paravariceal injection-sclerotherapy of esophageal varices, patients were randomly treated with sucralfate suspension or placebo. Time of treatment was limited to a maximum of 3 weeks and the dosage of sucralfate was 1 g q.i.d. Wlcogant-Suspension). Healing was assessed by endoscopy at weekly intervals. Fifty-three patients (25 sucralfate, 28 placebo) were evaluable according to the protocol. No patient left the study because of side effects. At the start of the trial, the patients in the sucralfate group showed a larger ulcer area than the placebo group. There was a tendency to faster healing in the sucralfate group, especially in patients with deeper ulcerations. However, there was no significant difference in global healing between both treatment groups after 3 weeks. Sucralfate suspension may be of value in accelerating the healing process in esophageal ulcers after sclerotherapy, especially in patients with deep ulcers. These results should be confirmed in further trials, in which patients should be stratified with respect to their ulcer volume and severity of liver disease.
From the Department of Surgery and Medicine, Heinz-Kalk Hospital, Bad Kissingen, Germany M. A. Mercado received a scholarship from “Deutscher Akademischer Austauschdienst” (DAAD), Germany. Requests for reprints should be addressed to Karl-Joseph Paquet, M.D., Department of Surgery, Heinz-Kalk Hospital, W-8730 Bad Kissingen, Germany.
herapeutic endoscopicsclerotherapyis now a T widely used and acceptedmethod for treatment of bleeding esophagealvarices. Although a number of complicationshave been reported to be attributable to sclerotherapy,the benefitsof sclerotherapy appear to outweigh the risks. Although postsclerotherapyesophagealulceration is a relatively common complication [1,2], the true frequency of postsclerotherapy ulcers remains unknown. It dependson a number of variables,including the technique of injection (i.e., intra-, paravariceal,or combined),the sclerosantusedand its concentration, timing of injection (i.e., emergency, urgent, or elective), the experienceof the endoscopist,degreeof the liver disease(advanced or not advanced),the morbidity of the patient, and last but not least, whether the medicationhas been injected for the first time or on a long-term basis. Furthermore, the percentage with ulcers is also higher in studiesthat reevaluatepatients soonafter sclerotherapy compared with those who delay it. This fact suggeststhat the majority of the ulcers heal uneventfully. Although postsclerotherapy esophagealulcerations are usually asymptomatic and are considered to be a necessarysecondaryphenomenonof successfulsclerotherapy,they carry a potential risk of upper gastrointestinal (GI) bleeding or other complications, such as esophagealstricture or wall necrosis. Thus, they can delay the overall processof sclerotherapy. However, because of their rapid spontaneoushealing [2,3], it is unclear whether the presenceof postsclerotherapy esophagealulcerations requires any specifictherapy to acceleratethe healing process. Treatment of esophagealulcer after sclerotherapyhas beenmostly empirical with drugs used for peptic ulcer therapy. Sucralfate,the aluminum salt of sulfatedsucrose, is a well-tolerated drug for peptic ulcer disease,including duodenal and gastric ulcer and reflux esophagitis.Its mode of action is different from alkalinizing agents such as antacids or histamine-Z
August 8, 1!391
The American Journal of Medicine
Volume 91 (suppl 2A)
2A-147s
SYMPOSIUM
ON SUCRALFATE/
PAQUET ET AL
cralfate/5 mL) in healing esophageal ulcers after therapeutic, elective endoscopic sclerotherapy in a randomized, placebo-controlled, double-blind trial.
TABLE I Schedule of Paravariceal Endoscopic Sclerotherapy Elective a) 40 ml 0.5%pokdocanol in 40 portions in the termrnal esophagus, b) Repeat after 7 days if there are no ulcerations, use 1%pokdocanol; in the case of ulcerations wait or inject only 20 ml. c) Usually l-6 sessions usingthe same schedule until teleangiectasias have drsappeared and epithelium and mucosa are completely covered by fibrous tissue. d) Endoscopic assessment after 4 months and resclerosis if the vances have enlarged in size and teleangiectasias can be seen. Inject per session 0.5 or 1% polidocanol in 20-30 portions of 0.5-075 ml. e) Thereafter endoscopic assessment evety6-12months and reinjection accordrng to d.
TABLE II Basic Clinical Data of the Study Population Sucralfate Male/female Age (years) Portal hypertension Etiology Alcoholic cirrhosis Non-alcoholic cirrhosis Extrahepatic block Child-Pugh classification
Placebo
1916 53.3+ 13.93 25
16112 50.6+ 16.79 28
2
0
4.0-t 1.77
4.2+_1.13
Classificationof esophageal varices according to Paquet IV No. of varices at start [mean +_SD) No. of ulcers at start [mean + SD) Ulcer area at start (mm2) Median, minimum, maximum Ulcer volume at start (mm3) Median, minimum, maximum
2.08+ 0.99
1.79+_1.07
109.2 ? 111.47 65.9,20,443
98.5 c 98.36
278.82 336.04 127.2,39,1328
66.3,20,428 262.5f. 350.34 113.8,39,1543
TABLE Ill Healing Rates of Esopha eal Ulcers (Confirmative Evaluation, 2 x c Contingency Table s 7j) Sucralfate
PATIENTSAND METHODS From August 1986 to August 1989, 60 patients were randomized in a monocentric, double-blind prospective study of sucralfate (30 patients) or placebo (30 patients) to control the healing process of esophageal ulcers after paravariceal therapeutic injection sclerotherapy of esophageal varices. Sclerotherapy was carried out with 0.5-1.0% polidocanol as described elsewhere [6]. The schedule of paravariceal therapeutic endoscopic sclerotherapy is listed in Table I. The time of treatment was limited to a maximum of 3 weeks. The dosage of sucralfate or placebo was 5 mL q.i.d. Healing was assessed by endoscopy at weekly intervals. Patients whose ulcers healed before 3 weeks of treatment finished the study at the time of healing. The main objective of the trial was complete healing of the ulcers; the change in ulcer size at weekly intervals was also observed. Illcer size was calculated from the product of length and width (1 x w x 0.785), and the volume of the crater was estimated by multiplying the ulcer area with a score ranging from shallow to very deep (in four grades from 1 = shallow to 4 = very deep). The measurements of the ulcer diameters were estimated using the open tips of the biopsy forceps. Statistical analysis of ulcer healing was performed using the exact test for 2 x c contingency tables [7]. The differences between medications in terms of change in ulcer area or ulcer volume were compared using the Wilcoxon rank sum test. The number of patients recruited was calculated to detect a significant difference in healing rates of 35% (a = 5% one-sided, /3 = 20%).
Placebo
RESULTS Healing after 1week Healing after 2 weeks Healing after 3 weeks Not healed after 3 weeks
5 (20.8) 14 (58.3)
p=
0.505
9 (33.3) 10(37.0) 5 (18.5) 3 (11.1)
ncludes one patient who left the trial because of therapeutic failure and commencement of nonpermissible concomitant medication
(H&receptor blockers. It can be described as mucosa-protective because it strengthens the natural defense mechanisms of the GI tract and also because it protects the ulcerated area against attack by acid and pepsin [4]. In 1984 Roark reported the first successful treatment of postsclerotherapy esophageal ulcers with sucralfate [5]. The aim of the current study is to prove the efficacy of sucralfate suspension (1 g su2A-148
August 8,
1991
The American Journal of Medicine
Volume
Of 60 patients, seven had to be exciuded from evaluation for the following reasons: withdrawal of informed consent (3 sucralfate), violation of exclusion criteria (1 placebo), enlistment of the same patient twice (1 placebo), nonpermissible concomitant medication (2 sucralfate). A total of 53 patients (25 sucralfate, 28 placebo) were therefore evaluable for analysis of efficacy (Table II). The basic clinical data of the 53 evaluable patients are shown in Table II. There were no differences between treatment groups with respect to age, gender, severity of liver disease (Child-Pugh classification), classification of esophageal varices according to Paquet [8], and number of esophageal varices. In the sucralfate group there were larger and deeper esophageal ulcers.
91 (suppl
2A)
SYMPOSIUM
Table III presents the results of the study expressed as healing rates after 1, 2, or 3 weeks of treatment. Overall, there are no significant differences between sucralfate and placebo. Two patients (1 sucralfate, 1 placebo) stopped the trial because of therapeutic failure. They presented with deep or very deep ulcers and decompensated liver disease (Child-Pugh C). The patient from the sucralfate group had hematemesis on the 15th day of the trial and was treated successfully with metoclopramide and antacid in addition to sucralfate. The patient from the placebo group left the trial on day 4 because of bleeding. He was switched to ranitidine and metoclopramide, but bleeding persisted despite effective emergency sclerotherapy. Table IV shows the course of healing in terms of reduction in ulcer volume. An especially large reduction of ulcer volume was observed after the first week of therapy while patients received sucralfate (differences not significant). A subgroup analysis of patients with ulcers that were considered deep or extremely deep (13 sucralfate, 12 placebo) showed a trend for faster healing in the sucralfate group, but differences in healing rates at different treatment times (Table V) were again not significant. The three patients who healed within 1 week of sucralfate therapy presented with well-compensated liver disease (2 Child-Pugh A, 1 Child-Pugh B) as did the only such patient from the placebo group (Child-Pugh A). Nearly all patients with advanced liver disease (Child-Pugh C) showed deep or very deep ulcerations (4 of 5 sucralfate, 3 of 5 placebo). Three of four in the sucralfate group healed within the study period (one patient after 2 weeks, two at the end of 3 weeks) in comparison to two of three in the placebo group who healed at 2 weeks. In Figure 1 the cumulative changes in ulcer size
ON SUCRALFATE / PAQUET ET AL
TABLE IV Comparison of Healing of Esophageal Ulcers in Ulcer Volume (Mean 2 SD) Sucralfate [mm’l Start of trial After 1 week of treatment After 2 weeks of treatment After 3 weeks of treatment
Placebo [mm?
278.8A 336.04
262.5A 350.34 183.6? 390.93 38.82 99.18 10.9+ 45.7
92.3 k 124.62 15.6 ? 40.12
2.0t 9.45
Differences not significant (Wilcoxon test)
1
TABLE V Healing Rates of Deep and Very Deep Ulcers Sucralfate (n = 13) Healingafter 12 weeks week Healingafter 3 weeks Not healed lifferences
Placebo (n = 12)
3
ti
3 1
:
1
not significant (exact test for 2 x c contingency tables).
are expressed as percentage of initial ulcer area. This figure, as does Table IV, shows a clear tendency of faster ulcer healing in the sucralfate group compared to placebo but the results fail to reach statistical significance. A global final evaluation by the physician judged the treatment successful in 88% (22 of 25) of the sucralfate-treated patients compared to 78% (22 of 28) of those treated with placebo. Study medication was tolerated well and drugrelated side effects were not observed.
COMMENTS After the first encouraging report of Roark [51 on the efficacy of sucralfate in healing esophageal ulcers after sclerotherapy, there was a need for fur-
3 Sucralfate -4-. Placebo -..++-
Figure 1. Comparison of healing of esophageal ulcers (changes of ulcer area in percentage of basic value) in both groups.
doys of treatment
August 8,
1991 The American Journal of Medicine
Volume
91 (suppl 2A)
2A-149s
SYMPOSIUM
ON SUCRALFATE / PAQUET ET AL
ther controlled investigations. The mode of action of sucralfate [4,9] and its documented efficacy in reflux esophagi& [lo-131 supported the decision to start a placebo-controlled clinical investigation. The results of this trial show a clear tendency in favor of sucralfate suspension in the treatment of esophageal ulcers following therapeutic endoscopic sclerotherapy of esophageal varices. An especially large difference between sucralfate and placebo could be found in ulcer volume after the first week of treatment. However, no statistically significant difference could be detected in the overall healing rate. Analyzing the subgroup of patients with deeper ulcerations, there was a trend toward faster healing in the sucralfate group, but the differences again were not significant, due to the relatively low number of patients. The results of this study are in agreement with other clinical studies published during the course of the trial. Singal et al [14] reported that sucralfate (in tablet form) accelerated the healing of the esophageal ulcers after sclerotherapy in comparison to placebo, but the results were not significant. In 1989 Tabibian et al [15] published the results of a clinical trial with another sucralfate suspension (1 g/10 mL), different from that used in this trial. They found a higher healing rate in the sucralfate group compared to placebo (difference not signiticant, p = 0.17). However, there was an imbalance with regard to ulcer size between placebo and sucralfate that favored the sucralfate group. Polson et al reported the results of another trial with sucralfate in 1989 [16], in which sucralfate was used for the prevention of rebleeding following therapeutic injection therapy of esophageal varices. In patients with well-compensated liver disease, Child-Pugh A + B, sucralfate was significantly superior to placebo in preventing rebleeding. The number, extent, and duration of sclerotherapy-induced esophageal ulcers were not different between the two treatment groups. In conclusion, sucralfate suspension seems to be slightly superior to placebo in accelerating the healing process of esophageal ulcerations after sclerotherapy. Since no peptic ulcer medication has
2A-15OS
August 8, 1991
The American Journal of Medicine
proved its efficacy in healing this kind of ulceration to date, the use of sucralfate suspension for initial treatment could be a promising consideration, especially in patients with deep ulcerations. The results of this and other clinical trials suggest that this conclusion is worthy of further investigations. A better stratification of patients should be achieved by randomization after endoscopy according to area or depth of the ulcers and to underlying liver disease and its severity.
REFERENCES 1. Schumann BM, Beckman JW, Tedesco FJ, Griffin JW, Jr, Assad RT. Complications of endoscopic injection sclerotherapy: a review. Am J Gastroenterol 1987; 82: 82330. 2. Paquet KJ. Indications and early and long-term results of paravariceal immediate, elective and prophylactic Injection sclerotherapy. In: Treatment of esophageal varices. Amsterdam/New York/Oxford. ldezuki Y, ed. Exerpta Medica 1988: 1-22. 3. Soehendra N, de Heer K, Kempeneers K, Frommelt L. Morphological alterations of the esophagus after endoscopic sclerotherapy of varices. Endoscopy 1983; 15: 291-6. 4. Szabo S, Hollander D. Pathways of gastrointestinal protection and repair: mechanisms of acbon of sucralfate. Am J Med 1989; 86 (Suppl 6A): 23-31. 5. Roark G. Treatment of postsclerotherapy esophageal ulcers with sucralfate. Gastrointest Endosc 1984; 30: 9-10. 6. Paquet KJ, Oberhammer E. Sclerotherapy of bleeding esophageal varices by means of endoscopy. Endoscopy 1978; 10: 7-12. 7. Stucky W, Vollmar J. Verfahren zur Auswertung von 2 x c-Kontingenztafeln. Biometr Zeitschr 1975; 17: 147-62. 8. Paquet Kl. Prophylactic endoscoprc sclerosing treatment of the esophageal wall in varices-a prospective controlled randomized trial. Endoscopy 1982; 14: 4-5. 9. Nagashima R. Mechanrsms of action of sucralfate. J Clin Gastroenterol 1981; 3 (Suppl 2): 117-27. 10. Weiss W, Brunner H. Buttner GR, et al. Therapie der Refluxosophagltis mrt Sucralfat. Dtsch Med Wochenschr 1983; 10: 1706-11. 11. Evreux M. Sucralfate versus alginateiantacid rn the treatment of peptic esophagrtis. Am J Med 1987; 83 (Suppl 38): 48-50. 12. Hameeteman W, van den Boomgard DM, Dekker W, Schrijver M, Wesdorp IC, Tytgat GN. Sucralfate versus cimetidine in reflux esophagitis, a single blind multicenter study. J Clin Gastroenterol 1987; 9: 390-4. 13. Simon B, Beckenbach HP, Daake H. et al. RefluxosophagitisWlrksamkeitsvergleich von Sucralfat and Ranitidrn. Munch Med Wochenschr 1988; 9: 152-4. 14. Slngal AK, Sarin SK, Misra SP, Broor SL. Ulceration after esophageal and gastric variceal sclerotherapy-influence of sucralfate and other factors on healing. Endoscopy 1988; 20: 238-40. 15. Tabibian N, Smith JL, Graham DY. Sclerotherapy-associated esophageal ulcers: lessons from a double-blind, randomized comparison of sucralfate suspension versus placebo. Gastrointest Endosc 1989; 35: 312-5. 16. Polson RI, Westaby D, Gimson AE, et al. Sucralfate for the prevention of early rebleeding following injection sclerotherapy for esophageal varices. Hepatology 1989; 10: 279-82.
Volume 91 (suppl 2A)
JOHANN-FRIEDRICH
In 1984 Roark published the first report of a sucralfate treatment of esophageal ulcers after sclerotherapy. Because this was an uncontrolled trial we planned a prospective doubleblind placebo-controlled study with 60 patients. After therapeutic paravariceal injection-sclerotherapy of esophageal varices, patients were randomly treated with sucralfate suspension or placebo. Time of treatment was limited to a maximum of 3 weeks and the dosage of sucralfate was 1 g q.i.d. Wlcogant-Suspension). Healing was assessed by endoscopy at weekly intervals. Fifty-three patients (25 sucralfate, 28 placebo) were evaluable according to the protocol. No patient left the study because of side effects. At the start of the trial, the patients in the sucralfate group showed a larger ulcer area than the placebo group. There was a tendency to faster healing in the sucralfate group, especially in patients with deeper ulcerations. However, there was no significant difference in global healing between both treatment groups after 3 weeks. Sucralfate suspension may be of value in accelerating the healing process in esophageal ulcers after sclerotherapy, especially in patients with deep ulcers. These results should be confirmed in further trials, in which patients should be stratified with respect to their ulcer volume and severity of liver disease.
From the Department of Surgery and Medicine, Heinz-Kalk Hospital, Bad Kissingen, Germany M. A. Mercado received a scholarship from “Deutscher Akademischer Austauschdienst” (DAAD), Germany. Requests for reprints should be addressed to Karl-Joseph Paquet, M.D., Department of Surgery, Heinz-Kalk Hospital, W-8730 Bad Kissingen, Germany.
herapeutic endoscopicsclerotherapyis now a T widely used and acceptedmethod for treatment of bleeding esophagealvarices. Although a number of complicationshave been reported to be attributable to sclerotherapy,the benefitsof sclerotherapy appear to outweigh the risks. Although postsclerotherapyesophagealulceration is a relatively common complication [1,2], the true frequency of postsclerotherapy ulcers remains unknown. It dependson a number of variables,including the technique of injection (i.e., intra-, paravariceal,or combined),the sclerosantusedand its concentration, timing of injection (i.e., emergency, urgent, or elective), the experienceof the endoscopist,degreeof the liver disease(advanced or not advanced),the morbidity of the patient, and last but not least, whether the medicationhas been injected for the first time or on a long-term basis. Furthermore, the percentage with ulcers is also higher in studiesthat reevaluatepatients soonafter sclerotherapy compared with those who delay it. This fact suggeststhat the majority of the ulcers heal uneventfully. Although postsclerotherapy esophagealulcerations are usually asymptomatic and are considered to be a necessarysecondaryphenomenonof successfulsclerotherapy,they carry a potential risk of upper gastrointestinal (GI) bleeding or other complications, such as esophagealstricture or wall necrosis. Thus, they can delay the overall processof sclerotherapy. However, because of their rapid spontaneoushealing [2,3], it is unclear whether the presenceof postsclerotherapy esophagealulcerations requires any specifictherapy to acceleratethe healing process. Treatment of esophagealulcer after sclerotherapyhas beenmostly empirical with drugs used for peptic ulcer therapy. Sucralfate,the aluminum salt of sulfatedsucrose, is a well-tolerated drug for peptic ulcer disease,including duodenal and gastric ulcer and reflux esophagitis.Its mode of action is different from alkalinizing agents such as antacids or histamine-Z
August 8, 1!391
The American Journal of Medicine
Volume 91 (suppl 2A)
2A-147s
SYMPOSIUM
ON SUCRALFATE/
PAQUET ET AL
cralfate/5 mL) in healing esophageal ulcers after therapeutic, elective endoscopic sclerotherapy in a randomized, placebo-controlled, double-blind trial.
TABLE I Schedule of Paravariceal Endoscopic Sclerotherapy Elective a) 40 ml 0.5%pokdocanol in 40 portions in the termrnal esophagus, b) Repeat after 7 days if there are no ulcerations, use 1%pokdocanol; in the case of ulcerations wait or inject only 20 ml. c) Usually l-6 sessions usingthe same schedule until teleangiectasias have drsappeared and epithelium and mucosa are completely covered by fibrous tissue. d) Endoscopic assessment after 4 months and resclerosis if the vances have enlarged in size and teleangiectasias can be seen. Inject per session 0.5 or 1% polidocanol in 20-30 portions of 0.5-075 ml. e) Thereafter endoscopic assessment evety6-12months and reinjection accordrng to d.
TABLE II Basic Clinical Data of the Study Population Sucralfate Male/female Age (years) Portal hypertension Etiology Alcoholic cirrhosis Non-alcoholic cirrhosis Extrahepatic block Child-Pugh classification
Placebo
1916 53.3+ 13.93 25
16112 50.6+ 16.79 28
2
0
4.0-t 1.77
4.2+_1.13
Classificationof esophageal varices according to Paquet IV No. of varices at start [mean +_SD) No. of ulcers at start [mean + SD) Ulcer area at start (mm2) Median, minimum, maximum Ulcer volume at start (mm3) Median, minimum, maximum
2.08+ 0.99
1.79+_1.07
109.2 ? 111.47 65.9,20,443
98.5 c 98.36
278.82 336.04 127.2,39,1328
66.3,20,428 262.5f. 350.34 113.8,39,1543
TABLE Ill Healing Rates of Esopha eal Ulcers (Confirmative Evaluation, 2 x c Contingency Table s 7j) Sucralfate
PATIENTSAND METHODS From August 1986 to August 1989, 60 patients were randomized in a monocentric, double-blind prospective study of sucralfate (30 patients) or placebo (30 patients) to control the healing process of esophageal ulcers after paravariceal therapeutic injection sclerotherapy of esophageal varices. Sclerotherapy was carried out with 0.5-1.0% polidocanol as described elsewhere [6]. The schedule of paravariceal therapeutic endoscopic sclerotherapy is listed in Table I. The time of treatment was limited to a maximum of 3 weeks. The dosage of sucralfate or placebo was 5 mL q.i.d. Healing was assessed by endoscopy at weekly intervals. Patients whose ulcers healed before 3 weeks of treatment finished the study at the time of healing. The main objective of the trial was complete healing of the ulcers; the change in ulcer size at weekly intervals was also observed. Illcer size was calculated from the product of length and width (1 x w x 0.785), and the volume of the crater was estimated by multiplying the ulcer area with a score ranging from shallow to very deep (in four grades from 1 = shallow to 4 = very deep). The measurements of the ulcer diameters were estimated using the open tips of the biopsy forceps. Statistical analysis of ulcer healing was performed using the exact test for 2 x c contingency tables [7]. The differences between medications in terms of change in ulcer area or ulcer volume were compared using the Wilcoxon rank sum test. The number of patients recruited was calculated to detect a significant difference in healing rates of 35% (a = 5% one-sided, /3 = 20%).
Placebo
RESULTS Healing after 1week Healing after 2 weeks Healing after 3 weeks Not healed after 3 weeks
5 (20.8) 14 (58.3)
p=
0.505
9 (33.3) 10(37.0) 5 (18.5) 3 (11.1)
ncludes one patient who left the trial because of therapeutic failure and commencement of nonpermissible concomitant medication
(H&receptor blockers. It can be described as mucosa-protective because it strengthens the natural defense mechanisms of the GI tract and also because it protects the ulcerated area against attack by acid and pepsin [4]. In 1984 Roark reported the first successful treatment of postsclerotherapy esophageal ulcers with sucralfate [5]. The aim of the current study is to prove the efficacy of sucralfate suspension (1 g su2A-148
August 8,
1991
The American Journal of Medicine
Volume
Of 60 patients, seven had to be exciuded from evaluation for the following reasons: withdrawal of informed consent (3 sucralfate), violation of exclusion criteria (1 placebo), enlistment of the same patient twice (1 placebo), nonpermissible concomitant medication (2 sucralfate). A total of 53 patients (25 sucralfate, 28 placebo) were therefore evaluable for analysis of efficacy (Table II). The basic clinical data of the 53 evaluable patients are shown in Table II. There were no differences between treatment groups with respect to age, gender, severity of liver disease (Child-Pugh classification), classification of esophageal varices according to Paquet [8], and number of esophageal varices. In the sucralfate group there were larger and deeper esophageal ulcers.
91 (suppl
2A)
SYMPOSIUM
Table III presents the results of the study expressed as healing rates after 1, 2, or 3 weeks of treatment. Overall, there are no significant differences between sucralfate and placebo. Two patients (1 sucralfate, 1 placebo) stopped the trial because of therapeutic failure. They presented with deep or very deep ulcers and decompensated liver disease (Child-Pugh C). The patient from the sucralfate group had hematemesis on the 15th day of the trial and was treated successfully with metoclopramide and antacid in addition to sucralfate. The patient from the placebo group left the trial on day 4 because of bleeding. He was switched to ranitidine and metoclopramide, but bleeding persisted despite effective emergency sclerotherapy. Table IV shows the course of healing in terms of reduction in ulcer volume. An especially large reduction of ulcer volume was observed after the first week of therapy while patients received sucralfate (differences not significant). A subgroup analysis of patients with ulcers that were considered deep or extremely deep (13 sucralfate, 12 placebo) showed a trend for faster healing in the sucralfate group, but differences in healing rates at different treatment times (Table V) were again not significant. The three patients who healed within 1 week of sucralfate therapy presented with well-compensated liver disease (2 Child-Pugh A, 1 Child-Pugh B) as did the only such patient from the placebo group (Child-Pugh A). Nearly all patients with advanced liver disease (Child-Pugh C) showed deep or very deep ulcerations (4 of 5 sucralfate, 3 of 5 placebo). Three of four in the sucralfate group healed within the study period (one patient after 2 weeks, two at the end of 3 weeks) in comparison to two of three in the placebo group who healed at 2 weeks. In Figure 1 the cumulative changes in ulcer size
ON SUCRALFATE / PAQUET ET AL
TABLE IV Comparison of Healing of Esophageal Ulcers in Ulcer Volume (Mean 2 SD) Sucralfate [mm’l Start of trial After 1 week of treatment After 2 weeks of treatment After 3 weeks of treatment
Placebo [mm?
278.8A 336.04
262.5A 350.34 183.6? 390.93 38.82 99.18 10.9+ 45.7
92.3 k 124.62 15.6 ? 40.12
2.0t 9.45
Differences not significant (Wilcoxon test)
1
TABLE V Healing Rates of Deep and Very Deep Ulcers Sucralfate (n = 13) Healingafter 12 weeks week Healingafter 3 weeks Not healed lifferences
Placebo (n = 12)
3
ti
3 1
:
1
not significant (exact test for 2 x c contingency tables).
are expressed as percentage of initial ulcer area. This figure, as does Table IV, shows a clear tendency of faster ulcer healing in the sucralfate group compared to placebo but the results fail to reach statistical significance. A global final evaluation by the physician judged the treatment successful in 88% (22 of 25) of the sucralfate-treated patients compared to 78% (22 of 28) of those treated with placebo. Study medication was tolerated well and drugrelated side effects were not observed.
COMMENTS After the first encouraging report of Roark [51 on the efficacy of sucralfate in healing esophageal ulcers after sclerotherapy, there was a need for fur-
3 Sucralfate -4-. Placebo -..++-
Figure 1. Comparison of healing of esophageal ulcers (changes of ulcer area in percentage of basic value) in both groups.
doys of treatment
August 8,
1991 The American Journal of Medicine
Volume
91 (suppl 2A)
2A-149s
SYMPOSIUM
ON SUCRALFATE / PAQUET ET AL
ther controlled investigations. The mode of action of sucralfate [4,9] and its documented efficacy in reflux esophagi& [lo-131 supported the decision to start a placebo-controlled clinical investigation. The results of this trial show a clear tendency in favor of sucralfate suspension in the treatment of esophageal ulcers following therapeutic endoscopic sclerotherapy of esophageal varices. An especially large difference between sucralfate and placebo could be found in ulcer volume after the first week of treatment. However, no statistically significant difference could be detected in the overall healing rate. Analyzing the subgroup of patients with deeper ulcerations, there was a trend toward faster healing in the sucralfate group, but the differences again were not significant, due to the relatively low number of patients. The results of this study are in agreement with other clinical studies published during the course of the trial. Singal et al [14] reported that sucralfate (in tablet form) accelerated the healing of the esophageal ulcers after sclerotherapy in comparison to placebo, but the results were not significant. In 1989 Tabibian et al [15] published the results of a clinical trial with another sucralfate suspension (1 g/10 mL), different from that used in this trial. They found a higher healing rate in the sucralfate group compared to placebo (difference not signiticant, p = 0.17). However, there was an imbalance with regard to ulcer size between placebo and sucralfate that favored the sucralfate group. Polson et al reported the results of another trial with sucralfate in 1989 [16], in which sucralfate was used for the prevention of rebleeding following therapeutic injection therapy of esophageal varices. In patients with well-compensated liver disease, Child-Pugh A + B, sucralfate was significantly superior to placebo in preventing rebleeding. The number, extent, and duration of sclerotherapy-induced esophageal ulcers were not different between the two treatment groups. In conclusion, sucralfate suspension seems to be slightly superior to placebo in accelerating the healing process of esophageal ulcerations after sclerotherapy. Since no peptic ulcer medication has
2A-15OS
August 8, 1991
The American Journal of Medicine
proved its efficacy in healing this kind of ulceration to date, the use of sucralfate suspension for initial treatment could be a promising consideration, especially in patients with deep ulcerations. The results of this and other clinical trials suggest that this conclusion is worthy of further investigations. A better stratification of patients should be achieved by randomization after endoscopy according to area or depth of the ulcers and to underlying liver disease and its severity.
REFERENCES 1. Schumann BM, Beckman JW, Tedesco FJ, Griffin JW, Jr, Assad RT. Complications of endoscopic injection sclerotherapy: a review. Am J Gastroenterol 1987; 82: 82330. 2. Paquet KJ. Indications and early and long-term results of paravariceal immediate, elective and prophylactic Injection sclerotherapy. In: Treatment of esophageal varices. Amsterdam/New York/Oxford. ldezuki Y, ed. Exerpta Medica 1988: 1-22. 3. Soehendra N, de Heer K, Kempeneers K, Frommelt L. Morphological alterations of the esophagus after endoscopic sclerotherapy of varices. Endoscopy 1983; 15: 291-6. 4. Szabo S, Hollander D. Pathways of gastrointestinal protection and repair: mechanisms of acbon of sucralfate. Am J Med 1989; 86 (Suppl 6A): 23-31. 5. Roark G. Treatment of postsclerotherapy esophageal ulcers with sucralfate. Gastrointest Endosc 1984; 30: 9-10. 6. Paquet KJ, Oberhammer E. Sclerotherapy of bleeding esophageal varices by means of endoscopy. Endoscopy 1978; 10: 7-12. 7. Stucky W, Vollmar J. Verfahren zur Auswertung von 2 x c-Kontingenztafeln. Biometr Zeitschr 1975; 17: 147-62. 8. Paquet Kl. Prophylactic endoscoprc sclerosing treatment of the esophageal wall in varices-a prospective controlled randomized trial. Endoscopy 1982; 14: 4-5. 9. Nagashima R. Mechanrsms of action of sucralfate. J Clin Gastroenterol 1981; 3 (Suppl 2): 117-27. 10. Weiss W, Brunner H. Buttner GR, et al. Therapie der Refluxosophagltis mrt Sucralfat. Dtsch Med Wochenschr 1983; 10: 1706-11. 11. Evreux M. Sucralfate versus alginateiantacid rn the treatment of peptic esophagrtis. Am J Med 1987; 83 (Suppl 38): 48-50. 12. Hameeteman W, van den Boomgard DM, Dekker W, Schrijver M, Wesdorp IC, Tytgat GN. Sucralfate versus cimetidine in reflux esophagitis, a single blind multicenter study. J Clin Gastroenterol 1987; 9: 390-4. 13. Simon B, Beckenbach HP, Daake H. et al. RefluxosophagitisWlrksamkeitsvergleich von Sucralfat and Ranitidrn. Munch Med Wochenschr 1988; 9: 152-4. 14. Slngal AK, Sarin SK, Misra SP, Broor SL. Ulceration after esophageal and gastric variceal sclerotherapy-influence of sucralfate and other factors on healing. Endoscopy 1988; 20: 238-40. 15. Tabibian N, Smith JL, Graham DY. Sclerotherapy-associated esophageal ulcers: lessons from a double-blind, randomized comparison of sucralfate suspension versus placebo. Gastrointest Endosc 1989; 35: 312-5. 16. Polson RI, Westaby D, Gimson AE, et al. Sucralfate for the prevention of early rebleeding following injection sclerotherapy for esophageal varices. Hepatology 1989; 10: 279-82.
Volume 91 (suppl 2A)
