Twenty-five years of injection sclerotherapy for bleeding varices

Acute injection sclerotherapy has been used in Belfast for 25 years and the results are reviewed. During this period 264 patients had injection sclerotherapy for acute bleeding from oesophageal varices during 396 admissions; a rigid oesophagoscope was used and 447 injections were performed. The series includes 19 children who received 69 injections. Thirty-eight had extrahepatic portal venous hypertension and the remainder had intrahepatic disease. Overall, 81 were Child's grade A (including the 38 extrahepatics), 82 were grade B and 101 were grade C. Of the 396 admissions, acute injection sclerotherapy controlled bleeding in 362 instances (control rate 91.4 per cent); control rate in the children's group was 97.1 per cent and in the adults 90.2 per cent. The hospital mortality was 14.9 per cent (57 adults and 2 children). Nineteen deaths were due directly to bleeding oesophageal varices, two from bleeding gastric varices and seven directly or indirectly from oesophageal leaks. Most of the remaining deaths were due to liver failure. We consider that sclerotherapy is valuable in the control of variceal haemorrhage where bleeding is uncontrolled or recurs after vasopressin or tamponade in any admission.     

Rigid versus fiberoptic endoscopic injection sclerotherapy. A prospective randomized controlled trial in patients with bleeding esophageal varices.

During a 3-year period (June 1981-July 1984), 70 patients who presented with an endoscopically proven variceal bleed were randomized to receive either fiberoptic injection sclerotherapy (FIS, 36 patients) or a rigid scope technique (RIS, 34 patients), using ethanolamine oleate as the sclerosing agent. After discharge, patients entered into a chronic injection sclerotherapy program. Control of the acute bleeding episode (FIS, 91%, RIS, 95%) and the first hospital admission mortality (FIS, 34%, RIS, 44%) were similar. Eradication of varices was achieved in 16/19 and 13/16 cases in the two respective groups of patients who survived longer than 3 months, and only 2 of these patients (both of whom were in the RIS group) had a further major variceal bleed. The overall complication rate per injection was significantly lower in the FIS group (p less than 0.005). Twenty-six patients (14 FIS, 12 RIS) were alive at the end of the 4-year trial period. Of the total of 44 deaths, only 3 were not associated with a bleeding episode. Only five of the 29 patients who had their varices eradicated died during the trial period (median follow-up of 16 months). FIS has become the preferred method of treatment at the Groote Shuur Hospital, particularly in the long-term management of these patients. The procedure is safe, does not require a general anesthetic, and in long-term management, most patients can be treated on an outpatient basis. RIS should be reserved for the difficult recurrent acute bleeder, where the more controlled situation with a rigid scope under general anesthesia may provide safer and more effective sclerotherapy. The study stresses the importance of achieving eradication of all varices to prevent rebleeding and its attendant high mortality.     

A prospective evaluation of injection sclerotherapy in the treatment of acute bleeding from esophageal varices.

In a 25 month study of massive upper-gastrointestinal hemorrhage, 64 patients were shown to have esophageal varices on emergency endoscopy. Twenty-four patients were actively bleeding from varices and were treated with a Sengstaken tube, and in 22 this was followed by emergency injection sclerotherapy using a rigid esophagoscope and general anesthesia. These 22 patients were followed prospectively and had 51 episodes of endoscopically proven active bleeding from esophageal varices which required Sengstaken tube control of hemorrhage during 36 separate admissions. This group included our total experience of injection sclerotherapy in acute variceal bleeding. The majority (14 of 22 patients) had alcoholic cirrhosis. Definitive control of variceal bleeding during the period of hospitalization was achieved in 33 hospital admissions (92%), usually with a single injection (27 hospital admissions: 75%). The results were satisfactory in 26 hospital admissions (72%). There were nine deaths (41% overall patient mortality rate), but no patient died primarily of variceal bleeding, and exsanguinating variceal bleeding was no longer a problem. The mortality rate per injection was 18%, and the mortality rate per hospital admission was 25%. Injection sclerotherapy is proposed as the emergency treatment of choice for patients with proven bleeding esophageal varices who do not stop bleeding on initial conservative treatment.     

A 10-year prospective evaluation of balloon tube tamponade and emergency injection sclerotherapy for actively bleeding oesophageal varices

During a 10 year study period 234 patients were admitted on 371 occasions with a total of 566 acute variceal bleeding episodes. Of these, 173 patients had 343 variceal bleeds which required balloon tamponade to achieve initial control of bleeding during 229 admissions and were then referred for emergency injection sclerotherapy. Sixty-eight percent of these patients had alcoholic cirrhosis and 42% were poor risk Grade C patients. Injection sclerotherapy was performed initially using the rigid Negus oesophagoscope under general anaesthesia and later using the fibreoptic endoscope under light sedation. Definitive control of variceal bleeding was achieved with sclerotherapy during 197 hospital admissions (92%). Of the 17 failures of emergency sclerotherapy, 4 patients died from uncontrolled bleeding and 13 patients underwent major surgical intervention. Definitive control of variceal bleeding was achieved with a single injection treatment in 138 hospital admissions (70%). Complications were mostly of a minor nature and occurred at a rate of 6% per injection treatment. The overall hospital admission mortality was 36%. The majority of patients died due to liver failure. The mortality in patients who required 4 injection treatments to control variceal bleeding was 71%. Injection sclerotherapy is proposed as the emergency treatment of choice for patients whose variceal bleeding continues or recurs after initial conservative management. Patients whose variceal bleeding is not controlled by 2 injection treatments require more major emergency surgery.     

Limitations of multiple injection sclerotherapy sessions for acute variceal bleeding

Sixty-six patients with active bleeding (127 episodes) from oesophageal varices treated by balloon-tube tamponade followed by injection sclerotherapy with a rigid endoscope were followed up for at least 1 year and analysed to determine whether the number of acute injection sessions during each hospital admission (87) or any other known parameter of liver function, e.g. Child's grading, affected the outcome. Definitive control of bleeding was achieved with one or two injections during 75 of these admissions (86%) with a mortality rate of 21%. However, the mortality rate in those patients who received three or four injections was 66% and reached 89% when Child's category A patients were excluded. It is concluded that the mortality rate in poor risk patients becomes unacceptably high when more than two injection sessions are required during a single hospital admission. Other methods of treatment, such as emergency portacaval shunting or devascularization procedures, should be instituted in the small subgroup of patients whose variceal bleeding is not controlled by two injection sessions.     

Portal hypertension and bleeding esophageal varices

Summary Bleeding from esophageal varices exacts a high mortality and extraordinary societal costs. Prophylaxis—medication, sclerotherapy, or shunt surgery to prevent an initial bleeding episode—is ineffective. In patients who have bled from varices, endoscopic injection sclerotherapy can control acute bleeding in more than 90% of patients. Because recurrent bleeding frequently occurs and survival without definitive therapy is dismal, selection of a permanently effective treatment is mandatory once variceal bleeding has been controlled.Long-term injection sclerotherapy can be performed in compliant patients; it is relatively safe but is associated with a 30–50% rebleeding rate. Betablockers significantly reduce portal pressure and recurrent bleeding but have not been shown to diminish mortality from BEV. Portal decompressive surgery permanently halts bleeding in more than 90% of patients; the risk of operative mortality is high in decompensated cirrhotics, and long-term complications of encephalopathy and accelerated liver failure may limit indications for shunt surgery to good-risk cirrhotics who are not liver transplant candidates. Devascularization procedures have a low operative mortality and encephalopathy rate but unacceptably high rates of recurrent bleeding.Liver transplantation is curative therapy for bleeding esophageal varices and the associated underlying hepatic dysfunction; cost and availability of donor organs generally limit its use in this setting to variceal bleeders with end-stagè liver disease not associated with active alcoholism.     

Eradication of oesophageal varices recurring after portal non-decompressive surgery by injection sclerotherapy

The results of injection sclerotherapy for oesophageal varices which recurred after portal non-decompressive surgery were analysed retrospectively to evaluate its efficacy. We treated 60 consecutive patients with portal hypertension; 19 were treated on an emergency basis, seven electively and 34 on a prophylactic basis. All acute bleeding was controlled with one session of sclerotherapy using a transparent overtube. After eradication by sclerotherapy, no bleeding episodes occurred and there was no recurrence of the varices, except in three uncompliant patients, during a mean follow-up period of 33.1 months. Bleeding from a gastric ulcer and gastritis occurred in one patient each. Oesophageal stenosis occurred in nine (15 per cent) patients and gastric varices developed in two (3 per cent) patients. Twelve patients died, five from liver failure and six with hepatoma, but there was no bleeding from the gastrointestinal tract. The overall 4-year survival rate was 80 per cent. We recommend the use of sclerotherapy as the primary treatment for recurrent oesophageal varices.     

Sclerotherapy of esophageal varices: long-term results and determinants of survival

Between 1980 and 1986, 177 patients underwent sclerotherapy by means of the flexible fiberoptic endoscope for bleeding esophageal varices. Of these, 129 were treated with serial sclerotherapy alone. The remaining 48 patients underwent liver transplantation after sclerotherapy; these are reported separately. Patients were classified by Child's criteria, by the severity of the initial bleeding episode as reflected by the urgency of treatment, and by the nature of the underlying liver disease. Long-term survival rates were markedly influenced by Child's classification, with 83% of the patients in class A, 45% of those in class B, and 20% of those in class C surviving beyond 36 months (p less than 0.001). Urgent treatment was associated with a poorer survival than was elective treatment (p less than 0.001). Survival was not influenced by underlying alcoholic liver disease as compared to a nonalcoholic liver disease. The majority of deaths occurred within the first 100 days after the initial treatment. Child's class B and C patients had the highest early mortality rates, particularly in an acute treatment setting. The most frequent causes of death included progressive liver failure and persistent hemorrhage. Sclerotherapy for bleeding esophageal varices may successfully control hemorrhage, but the influence of this treatment on long-term survival is limited. Hepatic reserve, indicated by Child's classification, is the major determinant of survival. Significant improvements in survival after variceal bleeding are intimately linked to improvement in liver function.     

Rigid endoscopy under general anaesthesia is safe for chronic injection sclerotherapy

Injection sclerotherapy for acutely bleeding oesophageal varices has been used in Belfast since 1958. However, a chronic injection sclerotherapy programme with rigid oesophagoscopy under general anaesthesia commenced only in 1979. So far, 82 patients have entered the programme; 57 patients had already received 73 acute injections before commencing chronic sclerotherapy. Subsequently, the 82 patients received 221 chronic injections plus a further 29 acute injections for rebleeding episodes which occurred during the programme. There were 24 Child's grade A patients, 23 B and 35 C; 48 per cent had alcoholic cirrhosis. Forty-eight patients achieved variceal obliteration with a mean of four injections. During the programme 24 patients experienced 42 acute bleeds. There were only two bleeding episodes within 1 week of a chronic injection and eight within 4 weeks. In the 8-year period there have been four early deaths. One occurred 17 days after a chronic injection and three followed acute injections required for rebleeding during the programme. There were 21 late deaths, mostly due to progressive liver failure, and none from rebleeding. We conclude that chronic injection sclerotherapy using rigid oesophagoscopy under general anaesthesia is both safe and effective.     

Improved results with selective distal splenorenal shunt in a highly selected patient population. A prospective study

In a 5-year period 299 patients were admitted to the Heinz-Kalk Hospital with bleeding esophageal varices. Patients with acute bleeding were treated with endoscopic sclerotherapy. Sessions were performed as many times as needed for each individual case. One hundred seventy-eight patients in Child-Pugh class C were excluded from surgical treatment; the remaining 121 patients (Child AB) were selected using the following criteria: liver volume (ultrasound) between 1000 to 2500 ml, portal perfusion (sequential scintigraphy) more than 30%, no activity or progression of liver disease proved by biopsy, no stenosis of the hepatic arteries, and suitable anatomy to perform the Warren shunt. Only 32 patients fulfilled these criteria. In seven of these cases the shunt was technically impossible to perform. Operative mortality rate was 8% and the late mortality rate was 12%. No history of rebleeding, encephalopathy, and/or shunt thrombosis was recorded. Five-year survival rate, according to the method of Kaplan-Meier was 75%. We conclude that the Warren shunt is the treatment of choice for elective management of bleeding esophageal varices. The postoperative results can be improved with strict selection using the above criteria. The preoperative use of sclerotherapy has a positive influence. Prophylactic management to prevent encephalopathy is also recommended.     

Endoscopic sclerotherapy in acute variceal hemorrhage

Acute variceal hemorrhage in patients with alcoholic cirrhosis and poor liver function is associated with a high mortality. A nonoperative treatment, endoscopic sclerotherapy, was employed in 22 patients with cirrhosis and poor liver function who had 24 episodes of acute variceal hemorrhage over a 20 month period. Portal hypertension was secondary to alcoholic cirrhosis in 21 patients and cystic fibrosis in 1 patient. Of the 24 patient admissions, 21 were of patients in Child's class C and 3 were class B. Endoscopic sclerotherapy was performed under endotracheal general anesthesia using a modified Negus rigid esophagoscope. The sclerosant (5 percent sodium morrhuate) was injected into all visible varices near the gastroesophageal junction using a MacBeth needle. Definitive control of variceal hemorrhage for the entire hospitalization was achieved in 19 of 24 admissions (79 percent). The in-hospital mortality for acute variceal bleeding was 29 percent; 81 percent of the patients were discharged after control of hemorrhage. There were two major and five minor complications related to sclerotherapy. Based on this preliminary experience it is concluded that injection sclerotherapy controls bleeding and reduces mortality associated with acute variceal hemorrhage in patients with poor liver function.     

Injection sclerotherapy of oesophageal varices with the free-hand technique: experience in Hong Kong

A prospective study of the efficacy of injection sclerotherapy with the free-hand technique for acute bleeding oesophageal varices was conducted, to evaluate its use in the control of acute variceal bleeding and to assess long-term sclerotherapy as the definitive treatment. Between July 1981 and January 1985, a total of 108 patients (96 men, 12 women with mean age of 54.4 years) had intravariceal injection of 5 per cent ethanolamine oleate. The majority had non-alcoholic cirrhosis and alcoholism accounted for only 18.5 per cent. There were 22 Child's A, 42 Child's B and 44 Child's C patients. During the 411 sessions of injection, major complications occurred in 12 patients (11.1 per cent) with 3 deaths. Of the 145 episodes of acute variceal bleeding 91.7 per cent were successfully controlled. In episodes which required more than one injection to control the bleeding, there was a high mortality of 75 per cent. Over the three and a half year period, 33 out of the 93 patients on long-term sclerotherapy had re-bled (35.5 per cent). Varices were obliterated in 27 patients with a mean of 5.4 injections. From our experience, the procedure is safe and effective. However, its status as a definitive treatment when compared with conventional surgical treatment requires further controlled evaluation.     

Emergency sclerotherapy for control of acute oesophageal variceal bleeding using 3% aqueous phenol: a 15-year experience

BACKGROUND: The purpose of the present study was to compare the efficacy of emergency endoscopic variceal sclerotherapy (EVS) using 3% aqueous phenol as an initial procedure to control acute oesophageal variceal bleed. METHODS: One hundred and ninety-five consecutive patients presenting with acute variceal bleed were included in the study. Protocol based endoscopic sclerotherapy and management of bleeding oesophageal varices was done. Immediate cessation of bleeding, re-bleeding within the first 72 h, success of first injection, final success of EVS, short-term mortality, influence of aetiology of portal hypertension and severity of liver disease on these results were studied. RESULTS: Immediate cessation of bleeding was obtained in 191 out of 195 patients (97.9%). Twenty-seven (13.8%) patients re-bled within 72 h. On re-injection final success of EVS was 87.2% (170/195 patients). There was no significant difference between final success rate in cirrhotic versus non-cirrhotic patients (103/118 (87.3%) vs. 67/77 (87%)). Success of first sclerotherapy session was significantly higher than that of second sclerotherapy session. Surgical rescue was required in 25 (12.8%) patients. Mortality was 3.6%. Failure of EVS and mortality was significantly higher in Child's C group (P = 0.04, Relative risk = 0.5, confidence interval 0.22-1.16). CONCLUSION: EVS remains an effective and cost effective modality of treatment to control acute variceal bleeding irrespective of aetiology of portal hypertension provided strict protocol based management is followed. With timely surgical rescue for the failures, the overall mortality can be reduced to less than 5%.     

The Management of Persistent or Recurrent Variceal Bleeding After Injection Sclerotherapy by Somatostatin

Sixteen patients with persistent (n = 11) or recurrent (n = 5) variceal bleeding after injection sclerotherapy and balloon tamponade were treated with an intravenous infusion of somatostatin 250μg/ h. Somatostatin infusion successfully controlled the bleeding in 15 of the 16 patients but one rebled after 72 h of treatment. In one patient with poor liver function (Child’s C) bleeding was not controlled by somatostatin, further injection sclerotherapy or balloon tamponade of the oesophagus. The results of this study, although uncontrolled and with a small number of patients, suggest that somatostatin is a very effective treatment for the control of post-injection sclerotherapy variceal bleeding.     

Sclerotherapy superior to surgery for longer survival and less rebleeding in 103 cirrhotics with variceal bleeding

In the ten years period from 1976-1986, 103 cirrhotic Japanese with acute variceal bleeding underwent either surgical treatment (48) or endoscopic injection sclerotherapy (55) at Kyushu University Hospital. We retrospectively analysed the clinical records of these patients. The two groups were comparable with regard to clinical condition and liver function, except for the higher rate of Child's C patients in the sclerotherapy group than in the surgical group (29 vs. 16; P less than 0.05). Control of variceal bleeding was attained in all of the 48 surgical patients and in 54 (98.2%) of the 55 sclerotherapy patients. Bleeding recurred in five (10.4%) of the surgical group and in one (1.8%) of the sclerotherapy group, resulting in four and one deaths, respectively, during the hospital stay. Mortality rates at 30 days and six months were 16.7% (8/48), 43.8% (21/48) in the surgical group, and 9.1% (5/55), 14.5% (8/55) in the sclerotherapy group. The five-year cumulative survival rate was significantly higher (P less than 0.01) in the sclerotherapy group (53.3%) than in the surgical group (29.4%). Therefore, in our patients sclerotherapy led to a longer survival with fewer rebleedings, as compared to other patients who underwent conventional surgical treatment.     

Injection sclerotherapy of esophageal varices using ethanolamine oleate: A pilot study

Fifty-three patients with upper gastrointestinal bleeding and proven esophageal varices were treated by intravascular injection sclerotherapy of the varices using a mixture of ethanolamine oleate, bovine thrombin and cephalothin. An intraesophageal balloon was used to impede craniad flow during the injection. Except in three patients who failed to stop bleeding from nonvariceal lesions, sclerotherapy was 94 percent successful in controlling bleeding. The mortality rate was 21 percent, or less than half that in historical controls. The hospital mortality rate in sclerotherapy patients with ascites was 25 percent compared with 54 to 75 percent reported elsewhere. There has been no rebleeding from varices after the third treatment week in patients followed up for up to 14 months.     

Are child's class C patients with acute variceal bleeding worth treating?

In the ten year period January 1980 to December 1989, 102 patients with Child's Class C liver disease (Pugh's Modification) were admitted with acute variceal bleeding to one surgical unit with a policy of early sclerotherapy. There were 56 males and 46 females; the average age was 55 years (range 28-77). Fifty-three suffered from alcoholic cirrhosis. Four died before definitive treatment could be carried out, three from liver failure and one from uncontrolled bleeding. Of the remaining 98 patients, eight had urgent oesophageal transection with three deaths from hepatorenal failure; 90 had sclerotherapy with 19 hospital deaths, nine from recurrent bleeding, eight from liver failure often coupled with renal failure and two from respiratory complications. Of the 76 who survived to leave hospital, 52 received chronic injection sclerotherapy, 10 had elective oesophageal transection and 14 did not have further elective intervention for various reasons. Surviving patients have been followed up at a special Liver Clinic with minimum follow up of one year. Although no patient has yet survived ten years, the one, five and eight year survivals of 50%, 21% and 13% suggest that salvage of these patients is worthwhile.     

Endoscopic sclerotherapy

Various sclerotherapy techniques have proved successful in the management of acute variceal bleeding and in long-term control of patients after a variceal bleed. We prefer either an intravariceal or a combined intravariceal and paravariceal technique using ethanolamine oleate, but we advocate that individual units utilize the technique with which they have the most experience. The use of an unmodified flexible endoscope has been almost universally accepted. Once active variceal bleeding is diagnosed on emergency endoscopy, immediate emergency sclerotherapy should be performed. When this is not possible, bleeding should be controlled by balloon-tube tamponade with subsequent delayed emergency sclerotherapy after resuscitation. Patients with variceal bleeding that has stopped at the time of the diagnostic endoscopy can either be treated by immediate sclerotherapy or be observed initially and subsequently treated using the long-term management policy of the unit concerned. Over 90% of actively bleeding patients should be controlled using emergency sclerotherapy. Failures are defined as patients who have more than two acute variceal bleeds during a single hospital admission. Such patients should be identified early and treated either by simple staple-gun transection or by an emergency portosystemic shunt. Repeated injection sclerotherapy using a flexible endoscope and the technique with which the group concerned has the most experience is recommended as the primary form of treatment for the majority of patients after a proven esophageal variceal bleed. Repeat injection treatments should probably be performed at weekly intervals until the esophageal varices are eradicated, with follow-up at 6-month or yearly intervals thereafter. Recurrent varices should be treated similarly. Failures of sclerotherapy are defined as patients who have either recurrent bleeds or in whom varices are difficult to eradicate. They require either a portosystemic shunt or a devascularization and transection operation. All patients presenting with cirrhosis and variceal bleeding should be evaluated for liver transplantation; unfortunately, however, few variceal bleeders are candidates for transplantation. Prophylactic sclerotherapy in patients with esophageal varices that have not bled remains unjustified outside of controlled trials. Available trials have produced conflicting data.     

Injection sclerotherapy for bleeding varices: risk factors and complications

Risk factors for complications after injection sclerotherapy were studied in 163 patients undergoing 667 treatments for bleeding oesophageal varices. The overall mortality rate was 7 per cent per injection sclerotherapy session; 16 per cent per acute session and 2.4 per cent per elective session. Acute variceal bleeding was controlled by injection sclerotherapy in 91 per cent of patients. Complications occurred after 16 per cent of injection sclerotherapy sessions. Deaths and complications were significantly associated with poor modified Child's grading (P less than 0.001), the first variceal bleed (P less than 0.001), acute injection sclerotherapy (P less than 0.001) and the use of the rigid oesophagoscope (P less than 0.001).     

Elective-selective Warren splenorenal shunt operation. Prognosis of liver function and esophageal varices hemorrhage after repeat sclerotherapy in liver cirrhosis]

From March 1st, 1982 to March 1st 1990 399 patients were admitted to the Heinz-Kalk-Hospital with recurrent bleeding from esophageal varices. Therapy of first choice was acute or elective endoscopic sclerotherapy. Early recurrences and uncontrollable hemorrhage were treated by Linton-Nachlas tube or if unsuccessful by devascularisation procedure. Two early or late bleeding recurrences were defined as sclerotherapy failures and choosen after passing a selection analysis (liver volume 1000 to 2500 ml, portal perfusion more than 30%, liver biopsy without activity or progression, exclusion of stenosis in the arterial supply of the liver and Child-Pugh classification A and B) for a selective-elective splenorenal Warren shunt (SRS). In 10 of 44 selected patients (11%) with an underlying disease of intrahepatic block in 95%, mostly alcoholic origin (65%) intraoperatively the performance of an SRS was technically problematic or impossible. Therefore, a mesocaval interposition shunt was carried out. Early mortality of 34 SRS was 5.9% (2 patients) and late mortality 17.6% (6 patients). No encephalopathy and shunt thrombosis were recorded. Postoperative angio- and sequential scintigraphies proved that portal perfusion was preserved during the first two years, but diminished. Liver function remained stable, too. One case of early rebleeding could be successfully managed by emergency endoscopic sclerotherapy. Five- and eight-years survival rate, according to the method of Kaplan-Meier is about 70%. We conclude that the SRS is the treatment of choice for elective management of recurrent bleeding of esophageal varices refractory to sclerotherapy. Its performance should be not enforced; in case of technical difficulties narrow-lumen mesocaval interposition shunt is an excellent alternative.