Original Papers Digestion 1990;46:193-198
©I990S. Karger AG. Basel 0012-2823/90/0464-0193S2.75/0
Efficacy of Oral Propranolol and Injection Sclerotherapy in the Long-Term Management ofVariceal Bleeding I luma Qureshi, Sarwar./. Zuberi, EjazAlam Pakistan Medical Research Council. Research Centre. Jinnah Postgraduate Medical Centre. Karachi. Pakistan
Key Words. Propranolol • Sclerotherapy • Cirrhosis • Idiopathic portal hypertension
Bleeding from oesophageal varices ac counts for a high mortality especially in those with poor hepatic reserve. No single most effective method to control acute bleeding has yet been defined. Sengstaken Blakemore tube is effective in 70% cases [ 1] while non-selective B blocker propranolol in doses that reduce the resting pulse rate by 25% has been found effective in the pre vention of rebleeding in 79-87% cases at 1 and 2 years [2-6]. but same results were not
supported by others [7], Sclerotherapy con trols active bleeding in 79-100% cases [810] and has a mortality of 8-33% which is directly related to the hepatic reserve. Scle rotherapy has been of value even in the long term prevention of bleeding from oesopha geal varices, with a 3-fold reduction in the chances of a rebleed [11-13]. Improved sur vival has been reported with both proprano lol [6] and sclerotherapy [II], but sclerother apy is superior to propranolol in patients
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Abstract. Three treatment regimens were tried in 145 portal hypertensives with bleeding oesophageal varices to test the efficacy of each regimen in the prevention of rebleeding. Fortv-seven cases received oral propranolol. 57 sclerotherapy, while 41 who. d.icLnot.'r.re.iy/v any treatment except conservative management served as controls. Patients were followed up at 6 weeks. 6 months. 1 and 2 years to see the frequency of bleeding with each regimen. No significant difference was seen at 6 weeks with either of the three regimens but at 6 months and 1 year the frequency of bleeding was significantly less (p < 0.05) in the propranolol group than in the other two groups while at 2 years results were significantly better with both propranolol (p < 0.05) and sclerotherpay (p < 0.001) than in controls. Efficacy of propran olol when compared with sclerotherapy showed similar results. The results of the present study are different from most of the western reports where sclerotherapy was found to be superior to propranolol. Variations in the results are likely to be due to differences in the etiology of portal hypertension in different countries.
194
Qureshi/Zuberi/Alam
Methods The present study is a retrospective analysis of patients with bleeding oesophageal varices who were treated with various regimens to control the bleeding. In the first phase of the study, no active treatment was available except conservative management, in the next phase propranolol was tried and lastly sclerother apy was applied. For the purpose of analysis only those patients were analysed who had similar age. sex. etiology and follow-up, the rest were excluded. One hundred and forty-five patients of similar age and sex. with endoscopically confirmed oesophageal varices, presenting with a history of upper gastroin testinal bleeding were thus selected for the study. Patients were classified according to their bleeding status as active bleeders who were actively bleeding at the time, and past bleeder who had bled 48-72 h prior to endoscopy. Children below 13 years of age. those with encephalopathy or hepatorenal syndrome were excluded from the study. Oesophageal varices were graded following the criteria of the Japanese Research Society [14], and severity of liver disease following Child's criteria [ 15]. Diagnosis in each case was based on histology but in cases where biopsy could not be done due to altered coagulation profile, massive ascites or early death, biochemical parameters and ultrasonographic find ings were taken into consideration to determine the etiology of portal hypertension. Diagnosis of idio pathic portal hypertension (IPH) was made following the criteria of the Japanese IPH committee [ 16], Oral propranolol was given to 47 cases in doses that reduced the resting pulse rate by 25%. Patients were followed up at 6 weeks. 6 months. 1 year and 2 years: and at each visit the pulse rate and blood pres
sure were recorded and any history of bleeding or side effects noted. Injection sclerotherapy using absolute alcohol was done in 57 patients. In case of bleeders, the active bleeding point was dealt with first followed by multiple intravariceal injections 2-4 cm above the cardio-oesophageal junction. Sclerotherapy was re peated at 7- to 15-day intervals until complete disap pearance of varices was achieved. End point of the study was either completion of 2 years of study with either therapy or a rebteed. Rcbleeding was defined as an upper gastrointestinal bleeding occurring 72 It af ter the cessation of active bleeding. Control group comprised patients who were not given any treatment except supportive therapy like blood transfusion, pitressin and plasma expanders. All the statistical anal ysis was done using the %2 test, life table analysis and log rank test.
Results A total of 145 patients with portal hyper tension who were either actively bleeding or had bled from the oesophageal varices in the past were selected for the study. Depending upon the availability of different treatment regimens. 47 were given propranolol. 57 sclerotherapy while 41 cases who were man aged conservatively served as controls. Of these, 6 patients in the control group and one each in the propranolol and sclerotherapy were lost to follow-up during the initial phase of the treatment and they were there fore excluded from the study. Comparison of the demographic profile of the three groups of patients at the entry into the trial is shown in table 1. Cirrhosis was twice more frequent than IPH as the cause of portal hypertension in the two treat ment groups, while the ratio of cirrhosis to IPH was almost similar in the control group. Rest of the parameters like age, sex, severity of disease and variceal size were almost sim ilar in all the three groups. A mean fall in the resting pulse rate of 32% was achieved in
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
with poor hepatic reserve (Child’s criteria). In developing countries like ours where ex pertise in portal hypertension surgery and sclerotherapy is limited to few centres, we have to develop our own strategy for the con trol of acute bleeding and for the prevention of rebleeding. In this study the efficacy of propranolol and sclerotherapy is compared with a control group to see which treatment is better.
Propranolol versus Sclerotherapy in Bleeding Varices
195
patients treated with propranolol, over a mean dose of 80 mg daily (20-140 mg). The cumulative percentage of patients free of rebleeding at the end of 1 and 2 years after inclusion were 92 and 97% in the pro pranolol group and 62 and 38%. respective ly, in the control group. The differences in the two groups were statistically significant (table 2). Further episodes of bleeding occurred in 19 of 46 cases, with maximum bleeds occur ring within 6 weeks to 6 months. In the pro pranolol group. 4 patients died of bleeding oesophageal varices. One patient developed bronchospasm while on propranolol, which subsided on discontinuation of therapy. None of the patients showed any deteriora tion in the hepatic function while on pro pranolol. In the sclerotherapy group, control of ac tive bleeding was achieved in 6 out of 8 cases (75%) while 2 continued to bleed and ex-
Table I. Demographic profile of the three groups Controls
Propran olol
Sclero therapy
Patients
41
47
57
Mean age ± SD
37± 12.4
39 ± 1 1.7
34+13.5
Male/female
21/20
19/28
29/28
Etiology Cirrhosis IPH
19 22
32 15
30 27
Child’s grade A B C
23 15 3
25 13 9
31 19 7
6
5
8
Old bleeders
37
42
49
Grade of varices I II III IV
0 19 18 4
0 25 20 2
0 15 33 8
Active bleeders
Table 2. Comparison of three treatment regimens Controls (A)
Propranolol (B)
Sclerotherapy (C)
p value
total
bled
never bled
total
bled
never bled
total
bled
never bled
(1)6 weeks
35
7 (20)
28 (80)
46
9 (20)
37 (80)
56
11 (20)
45 (80)
(2) 6 month
23
9 (39)
14 (61)
29
4 (14)
25 (86)
41
10 (24)
31 (76)
p < 0.05. A2 vs. B2
(3) 1 year
13
5 (38)
8 (62)
24
2 (8)
22 (92)
29
6 (21)
23 (79)
p < 0.05, A3 vs. B3
(4) 2 years
8
5 (62)
3 (38)
19
4 (21)
15 (79)
23
1 (4)
22 (96)
p < 0.05, A4 vs. B4 p < 0.001, A4 vs. C4
Figures in parentheses indicate percentage.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Duration
Qureshi/Zuberi/AIam
196
Mg. I. Life table analysis.
Table 3. Frequency of rebleeding in cirrhosis and IPH Controls
Cirrhosis IPH
Propranolol
Sclerotherapy
number
bled
number
bled
number
bled
19 22
12(63) 13(59)
32 15
16 (50) 3(20)
30 27
17(57) 10(37)
pired. Complete variceal obliteration was achieved in 26 of 56 cases (46.4) with a mean of 4.5 sessions over 8.4 weeks. Rebleeding occurred in 28 of 56 cases undergoing re peated sclerotherapy, of these 2 1 cases bled within 6 months of the study, before com plete sclerosis was achieved. These findings are similar to those seen in the propranolol group where 13 of 18 bleeders bled within 6 months. In the sclerotherapy group. 7 cases developed dysphagia of whom only one re
quired surgical dilatation while all others re sponded well to repeated dilatations with the endoscope. When the etiology of portal hypertension, i.e. cirrhosis vs. IPH. was compared to see if rebleeding frequency was different in differ ent groups, significantly lower bleeding rates were observed in IPH than in cirrhosis in both the treatment groups (table 3). Using life table analysis. 2 years’ survival was sim ilar (28%) in controls and those undergoing
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Figures in parentheses indicate percentage.
Propranolol versus Sclerotherapy in Bleeding Varices
197
sclerotherapy, while with propranolol a 40% survival was achieved, again showing better survival with propranolol (fig. 1). When the survival of patients with three treatment reg imens was compared using log rank analysis, the difference was not found to be statisti cally significant. Mortality was 14. 9 and 6% in the sclero therapy, propranolol and control group, re spectively. with bleeding as the major cause of death.
portal fibrosis who were given oral propran olol for 1 year [18], Similarly in the present study about 50% patients in both the treat ment groups had idiopathic portal hyperten sion and none of the patients had alcoholic cirrhosis: therefore it is likely that better results achieved in this study are due to both difference in etiology and severity of liver disease. Sclerotherpav. on the other hand, showed results almost similar to those of propranolol throughout 2 years of follow-up. In view of a similar response to both propranolol and sclerotherapy in patients with well compen sated liver function, it is merely the choice of the physician and the availability of the tech nique and expertise which would decide amongst the two available therapeutic mea sures.
The present study showed significantly less risk of rebleeding from oesophageal var ices in patients receiving either long-term oral propranolol or sclerotherapy than in controls, while no significant difference was seen in the prevention of rebleeding in those on propranolol or sclerotherapy. The cumulative percentages of patients free of recurrent bleeding at 1 and 2 years after inclusion were 92 and 79% in the pro pranolol group and 62 and 38% in the con trol group. These findings are similar to those reported by Lebrec et al. [6], but are in contrast to those of Westaby et al. [ 17] where only 40% patients were free of rebleeding. The possible explanation for this contro versy may be that non-selective B receptor blockers like propranolol are more effective in producing a fall in portal pressure than selective B receptor blocker like metoprolol. because of their added method of action mediated by B receptor blockade on splanch nic vessels [17], The difference in results could also be due to different causes and severity of liver disease in different countries like in Zimbabwe; rebleeding rates were sig nificantly low in patients with noncirrhotic
References 1 Novis BH. Duya P. Barbezat GG. et al: Fibreoptic endoscopy and use of Sengstaken tube in acute G.I. haemorrhage in patients with portal hyper tension and varices. Gut 1976:17:258-263. 2 Lebrec D. Nouel O. Corbie M. Benhamou .IP: Pro pranolol a medical treatment for portal hyperten sion. Lancet 1980:ii: 180—182. 3 Lebrec D. Poynord T. Hillon P. Benhamou .IP: Propranolol for prevention of recurrent gastroin testinal bleeding in patients with cirrhosis. N Engl .1 Med 1982:305:1371-1374. 4 Hillon P. Lebrec D. Munoz C. Jungers M. Goldfarb G. Benhamou JP: Comparison of the effects of cardioselective and non cardioselcctivc Bblocker on portal hypertension in patients with cirrhosis. Hepatology 1982:21:528-531. 5 Westaby D. Bihari D. Gimson A. Crossly I. Wil liams R: Selective and nonselective B receptor blockade in patients with cirrhosis and portal hy pertension. Gut 1984:25:121-124. 6 Lebrec D. Ponyard T. Bernuan J. BercoffE. Novel O. Capron JP. Poupon R. Bouvry M. Rueff B, Benhamou JP: A randomized controlled study of
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Discussion
propranolol for prevention of reccurent gastroin testinal bleeding in patients with cirrhosis. A final report. Hepatology 1984;4:355-358. 7 Burroughs AK. Jenkins WJ. Sherlock S. et al: Con trolled trial of propronolol for the prevention of recurrent gastrointestinal bleeding in patient with cirrhosis. N Engl J Med 1983;309:1539-1542. 8 Clark AW. MacDougall BRD. Westaby D: Pro spective clinical trial of injection sclerotherapy in cirrhotic patients with recent variceal haemor rhage. Lancet 1980;ii:552—554. 9 Sarin SK. Sachdeva GK, Nanda R. Vij JC. Anand BS: Endoscopic sclerotherapy using absolute alco hol. Gut 1985:26:120-124. 10 Westaby D. Macdougall BRD. Malia W. et al: A prospective randomized study of two sclerother apy techniques for oesophageal varices. Hepato logy 1983:3:681-684. I 1 MacDougall BRD. Theodossl A. Westaby D. et al: Incrased long term survival in variceal haemor rhage using injection sclerotherapy, results of a controlled trial. Lancet 1982:1:124-127. 12 Terbanche J. Borman PC. Jonker MAT. et al: Injection sclerotherapy of oesophageal varices. Semin Liver Dis 1982:2:233-241. 13 Blart AL. Larson AW. Radran GA, et al: A pro spective controlled trial of endoscopic sclerother apy in variceal bleeding progress report. Hepato logy' 1982:2:732. 14 Japanese Research Society for Portal Hyperten sion: General rules for recording endoscopic find ings on oesophageal varices. Jpn J Surg 1980:10: 84-93.
Qureshi/Zuberi/Alam
15 Child CG. Turcotte JG: Surgery in portal hyper tension: in Child CG (ed): Major Problems in Clinical Surgery. The Liverand Portal Hyperten sion. Philadelphia. Saunders. 1964. pp 1-84. 16 Thomas HC. Jones EA: Recent advances in Hepa tology: in Okuda K (ed): Idiopathic Portal Hyper tension. Churchill Livingstone, Edinburgh. 1986. pp 93-108. 17 Westaby D. Melia WM, MacDougall BRD. Hcgarty JE. Ginison AE. Williams R: Bj selective adrenoreceptor blockade for the long term man agement of variceal bleeding. A prospective ran domised trial to compare oral metoprolol with injection sclerotherapy in cirrhosis. Gut 1985:26: 421-425. 18 Küre CF: Controlled trial of propranolol to pre vent recurrent variceal bleeding in patients with non-cirrhotic portal fibrosis. Br Med J 1989:298: 1363-1365.
Received: October 5. 1989 Received in revised form: March 15. 1990 Huma Qureshi. MD Pakistan Medical Research Council Research Centre Jinnah Postgraduate Medical Centre Karachi (Pakistan)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
198
©I990S. Karger AG. Basel 0012-2823/90/0464-0193S2.75/0
Efficacy of Oral Propranolol and Injection Sclerotherapy in the Long-Term Management ofVariceal Bleeding I luma Qureshi, Sarwar./. Zuberi, EjazAlam Pakistan Medical Research Council. Research Centre. Jinnah Postgraduate Medical Centre. Karachi. Pakistan
Key Words. Propranolol • Sclerotherapy • Cirrhosis • Idiopathic portal hypertension
Bleeding from oesophageal varices ac counts for a high mortality especially in those with poor hepatic reserve. No single most effective method to control acute bleeding has yet been defined. Sengstaken Blakemore tube is effective in 70% cases [ 1] while non-selective B blocker propranolol in doses that reduce the resting pulse rate by 25% has been found effective in the pre vention of rebleeding in 79-87% cases at 1 and 2 years [2-6]. but same results were not
supported by others [7], Sclerotherapy con trols active bleeding in 79-100% cases [810] and has a mortality of 8-33% which is directly related to the hepatic reserve. Scle rotherapy has been of value even in the long term prevention of bleeding from oesopha geal varices, with a 3-fold reduction in the chances of a rebleed [11-13]. Improved sur vival has been reported with both proprano lol [6] and sclerotherapy [II], but sclerother apy is superior to propranolol in patients
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Abstract. Three treatment regimens were tried in 145 portal hypertensives with bleeding oesophageal varices to test the efficacy of each regimen in the prevention of rebleeding. Fortv-seven cases received oral propranolol. 57 sclerotherapy, while 41 who. d.icLnot.'r.re.iy/v any treatment except conservative management served as controls. Patients were followed up at 6 weeks. 6 months. 1 and 2 years to see the frequency of bleeding with each regimen. No significant difference was seen at 6 weeks with either of the three regimens but at 6 months and 1 year the frequency of bleeding was significantly less (p < 0.05) in the propranolol group than in the other two groups while at 2 years results were significantly better with both propranolol (p < 0.05) and sclerotherpay (p < 0.001) than in controls. Efficacy of propran olol when compared with sclerotherapy showed similar results. The results of the present study are different from most of the western reports where sclerotherapy was found to be superior to propranolol. Variations in the results are likely to be due to differences in the etiology of portal hypertension in different countries.
194
Qureshi/Zuberi/Alam
Methods The present study is a retrospective analysis of patients with bleeding oesophageal varices who were treated with various regimens to control the bleeding. In the first phase of the study, no active treatment was available except conservative management, in the next phase propranolol was tried and lastly sclerother apy was applied. For the purpose of analysis only those patients were analysed who had similar age. sex. etiology and follow-up, the rest were excluded. One hundred and forty-five patients of similar age and sex. with endoscopically confirmed oesophageal varices, presenting with a history of upper gastroin testinal bleeding were thus selected for the study. Patients were classified according to their bleeding status as active bleeders who were actively bleeding at the time, and past bleeder who had bled 48-72 h prior to endoscopy. Children below 13 years of age. those with encephalopathy or hepatorenal syndrome were excluded from the study. Oesophageal varices were graded following the criteria of the Japanese Research Society [14], and severity of liver disease following Child's criteria [ 15]. Diagnosis in each case was based on histology but in cases where biopsy could not be done due to altered coagulation profile, massive ascites or early death, biochemical parameters and ultrasonographic find ings were taken into consideration to determine the etiology of portal hypertension. Diagnosis of idio pathic portal hypertension (IPH) was made following the criteria of the Japanese IPH committee [ 16], Oral propranolol was given to 47 cases in doses that reduced the resting pulse rate by 25%. Patients were followed up at 6 weeks. 6 months. 1 year and 2 years: and at each visit the pulse rate and blood pres
sure were recorded and any history of bleeding or side effects noted. Injection sclerotherapy using absolute alcohol was done in 57 patients. In case of bleeders, the active bleeding point was dealt with first followed by multiple intravariceal injections 2-4 cm above the cardio-oesophageal junction. Sclerotherapy was re peated at 7- to 15-day intervals until complete disap pearance of varices was achieved. End point of the study was either completion of 2 years of study with either therapy or a rebteed. Rcbleeding was defined as an upper gastrointestinal bleeding occurring 72 It af ter the cessation of active bleeding. Control group comprised patients who were not given any treatment except supportive therapy like blood transfusion, pitressin and plasma expanders. All the statistical anal ysis was done using the %2 test, life table analysis and log rank test.
Results A total of 145 patients with portal hyper tension who were either actively bleeding or had bled from the oesophageal varices in the past were selected for the study. Depending upon the availability of different treatment regimens. 47 were given propranolol. 57 sclerotherapy while 41 cases who were man aged conservatively served as controls. Of these, 6 patients in the control group and one each in the propranolol and sclerotherapy were lost to follow-up during the initial phase of the treatment and they were there fore excluded from the study. Comparison of the demographic profile of the three groups of patients at the entry into the trial is shown in table 1. Cirrhosis was twice more frequent than IPH as the cause of portal hypertension in the two treat ment groups, while the ratio of cirrhosis to IPH was almost similar in the control group. Rest of the parameters like age, sex, severity of disease and variceal size were almost sim ilar in all the three groups. A mean fall in the resting pulse rate of 32% was achieved in
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
with poor hepatic reserve (Child’s criteria). In developing countries like ours where ex pertise in portal hypertension surgery and sclerotherapy is limited to few centres, we have to develop our own strategy for the con trol of acute bleeding and for the prevention of rebleeding. In this study the efficacy of propranolol and sclerotherapy is compared with a control group to see which treatment is better.
Propranolol versus Sclerotherapy in Bleeding Varices
195
patients treated with propranolol, over a mean dose of 80 mg daily (20-140 mg). The cumulative percentage of patients free of rebleeding at the end of 1 and 2 years after inclusion were 92 and 97% in the pro pranolol group and 62 and 38%. respective ly, in the control group. The differences in the two groups were statistically significant (table 2). Further episodes of bleeding occurred in 19 of 46 cases, with maximum bleeds occur ring within 6 weeks to 6 months. In the pro pranolol group. 4 patients died of bleeding oesophageal varices. One patient developed bronchospasm while on propranolol, which subsided on discontinuation of therapy. None of the patients showed any deteriora tion in the hepatic function while on pro pranolol. In the sclerotherapy group, control of ac tive bleeding was achieved in 6 out of 8 cases (75%) while 2 continued to bleed and ex-
Table I. Demographic profile of the three groups Controls
Propran olol
Sclero therapy
Patients
41
47
57
Mean age ± SD
37± 12.4
39 ± 1 1.7
34+13.5
Male/female
21/20
19/28
29/28
Etiology Cirrhosis IPH
19 22
32 15
30 27
Child’s grade A B C
23 15 3
25 13 9
31 19 7
6
5
8
Old bleeders
37
42
49
Grade of varices I II III IV
0 19 18 4
0 25 20 2
0 15 33 8
Active bleeders
Table 2. Comparison of three treatment regimens Controls (A)
Propranolol (B)
Sclerotherapy (C)
p value
total
bled
never bled
total
bled
never bled
total
bled
never bled
(1)6 weeks
35
7 (20)
28 (80)
46
9 (20)
37 (80)
56
11 (20)
45 (80)
(2) 6 month
23
9 (39)
14 (61)
29
4 (14)
25 (86)
41
10 (24)
31 (76)
p < 0.05. A2 vs. B2
(3) 1 year
13
5 (38)
8 (62)
24
2 (8)
22 (92)
29
6 (21)
23 (79)
p < 0.05, A3 vs. B3
(4) 2 years
8
5 (62)
3 (38)
19
4 (21)
15 (79)
23
1 (4)
22 (96)
p < 0.05, A4 vs. B4 p < 0.001, A4 vs. C4
Figures in parentheses indicate percentage.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Duration
Qureshi/Zuberi/AIam
196
Mg. I. Life table analysis.
Table 3. Frequency of rebleeding in cirrhosis and IPH Controls
Cirrhosis IPH
Propranolol
Sclerotherapy
number
bled
number
bled
number
bled
19 22
12(63) 13(59)
32 15
16 (50) 3(20)
30 27
17(57) 10(37)
pired. Complete variceal obliteration was achieved in 26 of 56 cases (46.4) with a mean of 4.5 sessions over 8.4 weeks. Rebleeding occurred in 28 of 56 cases undergoing re peated sclerotherapy, of these 2 1 cases bled within 6 months of the study, before com plete sclerosis was achieved. These findings are similar to those seen in the propranolol group where 13 of 18 bleeders bled within 6 months. In the sclerotherapy group. 7 cases developed dysphagia of whom only one re
quired surgical dilatation while all others re sponded well to repeated dilatations with the endoscope. When the etiology of portal hypertension, i.e. cirrhosis vs. IPH. was compared to see if rebleeding frequency was different in differ ent groups, significantly lower bleeding rates were observed in IPH than in cirrhosis in both the treatment groups (table 3). Using life table analysis. 2 years’ survival was sim ilar (28%) in controls and those undergoing
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Figures in parentheses indicate percentage.
Propranolol versus Sclerotherapy in Bleeding Varices
197
sclerotherapy, while with propranolol a 40% survival was achieved, again showing better survival with propranolol (fig. 1). When the survival of patients with three treatment reg imens was compared using log rank analysis, the difference was not found to be statisti cally significant. Mortality was 14. 9 and 6% in the sclero therapy, propranolol and control group, re spectively. with bleeding as the major cause of death.
portal fibrosis who were given oral propran olol for 1 year [18], Similarly in the present study about 50% patients in both the treat ment groups had idiopathic portal hyperten sion and none of the patients had alcoholic cirrhosis: therefore it is likely that better results achieved in this study are due to both difference in etiology and severity of liver disease. Sclerotherpav. on the other hand, showed results almost similar to those of propranolol throughout 2 years of follow-up. In view of a similar response to both propranolol and sclerotherapy in patients with well compen sated liver function, it is merely the choice of the physician and the availability of the tech nique and expertise which would decide amongst the two available therapeutic mea sures.
The present study showed significantly less risk of rebleeding from oesophageal var ices in patients receiving either long-term oral propranolol or sclerotherapy than in controls, while no significant difference was seen in the prevention of rebleeding in those on propranolol or sclerotherapy. The cumulative percentages of patients free of recurrent bleeding at 1 and 2 years after inclusion were 92 and 79% in the pro pranolol group and 62 and 38% in the con trol group. These findings are similar to those reported by Lebrec et al. [6], but are in contrast to those of Westaby et al. [ 17] where only 40% patients were free of rebleeding. The possible explanation for this contro versy may be that non-selective B receptor blockers like propranolol are more effective in producing a fall in portal pressure than selective B receptor blocker like metoprolol. because of their added method of action mediated by B receptor blockade on splanch nic vessels [17], The difference in results could also be due to different causes and severity of liver disease in different countries like in Zimbabwe; rebleeding rates were sig nificantly low in patients with noncirrhotic
References 1 Novis BH. Duya P. Barbezat GG. et al: Fibreoptic endoscopy and use of Sengstaken tube in acute G.I. haemorrhage in patients with portal hyper tension and varices. Gut 1976:17:258-263. 2 Lebrec D. Nouel O. Corbie M. Benhamou .IP: Pro pranolol a medical treatment for portal hyperten sion. Lancet 1980:ii: 180—182. 3 Lebrec D. Poynord T. Hillon P. Benhamou .IP: Propranolol for prevention of recurrent gastroin testinal bleeding in patients with cirrhosis. N Engl .1 Med 1982:305:1371-1374. 4 Hillon P. Lebrec D. Munoz C. Jungers M. Goldfarb G. Benhamou JP: Comparison of the effects of cardioselective and non cardioselcctivc Bblocker on portal hypertension in patients with cirrhosis. Hepatology 1982:21:528-531. 5 Westaby D. Bihari D. Gimson A. Crossly I. Wil liams R: Selective and nonselective B receptor blockade in patients with cirrhosis and portal hy pertension. Gut 1984:25:121-124. 6 Lebrec D. Ponyard T. Bernuan J. BercoffE. Novel O. Capron JP. Poupon R. Bouvry M. Rueff B, Benhamou JP: A randomized controlled study of
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
Discussion
propranolol for prevention of reccurent gastroin testinal bleeding in patients with cirrhosis. A final report. Hepatology 1984;4:355-358. 7 Burroughs AK. Jenkins WJ. Sherlock S. et al: Con trolled trial of propronolol for the prevention of recurrent gastrointestinal bleeding in patient with cirrhosis. N Engl J Med 1983;309:1539-1542. 8 Clark AW. MacDougall BRD. Westaby D: Pro spective clinical trial of injection sclerotherapy in cirrhotic patients with recent variceal haemor rhage. Lancet 1980;ii:552—554. 9 Sarin SK. Sachdeva GK, Nanda R. Vij JC. Anand BS: Endoscopic sclerotherapy using absolute alco hol. Gut 1985:26:120-124. 10 Westaby D. Macdougall BRD. Malia W. et al: A prospective randomized study of two sclerother apy techniques for oesophageal varices. Hepato logy 1983:3:681-684. I 1 MacDougall BRD. Theodossl A. Westaby D. et al: Incrased long term survival in variceal haemor rhage using injection sclerotherapy, results of a controlled trial. Lancet 1982:1:124-127. 12 Terbanche J. Borman PC. Jonker MAT. et al: Injection sclerotherapy of oesophageal varices. Semin Liver Dis 1982:2:233-241. 13 Blart AL. Larson AW. Radran GA, et al: A pro spective controlled trial of endoscopic sclerother apy in variceal bleeding progress report. Hepato logy' 1982:2:732. 14 Japanese Research Society for Portal Hyperten sion: General rules for recording endoscopic find ings on oesophageal varices. Jpn J Surg 1980:10: 84-93.
Qureshi/Zuberi/Alam
15 Child CG. Turcotte JG: Surgery in portal hyper tension: in Child CG (ed): Major Problems in Clinical Surgery. The Liverand Portal Hyperten sion. Philadelphia. Saunders. 1964. pp 1-84. 16 Thomas HC. Jones EA: Recent advances in Hepa tology: in Okuda K (ed): Idiopathic Portal Hyper tension. Churchill Livingstone, Edinburgh. 1986. pp 93-108. 17 Westaby D. Melia WM, MacDougall BRD. Hcgarty JE. Ginison AE. Williams R: Bj selective adrenoreceptor blockade for the long term man agement of variceal bleeding. A prospective ran domised trial to compare oral metoprolol with injection sclerotherapy in cirrhosis. Gut 1985:26: 421-425. 18 Küre CF: Controlled trial of propranolol to pre vent recurrent variceal bleeding in patients with non-cirrhotic portal fibrosis. Br Med J 1989:298: 1363-1365.
Received: October 5. 1989 Received in revised form: March 15. 1990 Huma Qureshi. MD Pakistan Medical Research Council Research Centre Jinnah Postgraduate Medical Centre Karachi (Pakistan)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 7:18:15 PM
198
