Bucrylate Treatment of Bleeding Gastric Varices: 12 Years' Experience

• Introduction
• Patients and Methods
• Results
• Discussion
• References

Background and Study Aims: For several years now there has been an increasingly widespread use of a tissue adhesive in the treatment of bleeding gastric varices to achieve rapid, safe control of hemostasis and prevent rebleeding. In this study we report on our experience with the use of Bucrylate (Hystoacryl) for the treatment of gastric varices over a period of more than a decade.

Patients and Methods: Since 1988, 174 cirrhotic patients with actively bleeding gastric varices have been admitted to our department, where they received emergency treatment with injections of Bucrylate. Any associated nonbleeding esophageal varices were subjected to traditional sclerotherapy in combination with the Bucrylate treatment. The gastric varices were subdivided into four distinct groups according to the method advocated by Sarin in 1989. The patients underwent weekly sclerotherapy sessions until their varices were eradicated, and the follow-up with a mean of 36 months (range 9 - 90 months) consisted of endoscopy at 3, 6, and 12 months during the first year and then yearly checks to confirm obliteration of the varices.

Results: The hemostasis (97.1 %), early rebleeding (15.5 %), and hospital mortality (19.5 %) rates of the patients with bleeding gastric varices, treated with the tissue adhesive, were very similar to those of patients treated for esophageal varices over the same period (98.1 %, 13.0 %, and 16.4 %, respectively). The most frequent cause of death at 30 days was liver failure (76 % of cases), followed by hemorrhagic shock (8.8 %), and other less frequent causes. Sclerotherapy achieved obliteration rate for gastric varices (70 - 75 %) similar to that for esophageal varices in those patients with portal hypertension due to intrahepatic block (alcoholic and posthepatitis cirrhosis), but a rate of only 32 % in the group of patients with prehepatic block (splenoportomesenteric thrombosis), where surgery proved more effective (69 %). The medium- and long-term survival rates depended on the stability of the patients' liver conditions, on rapid, effective control of variceal hemostasis, and on complete, lasting obliteration of the gastric varices.

Conclusions: The use of Bucrylate in emergency sclerotherapy achieved results in bleeding gastric varices on a par with those obtained in esophageal varices in cases of alcoholic and posthepatitis cirrhosis. The group of patients with portal hypertension due to prehepatic block (splenoportal thrombosis) showed no benefit from sclerotherapy in terms of obliteration of gastric varices, but benefited from elective surgery. The choice of the obliterating treatment indicated may be facilitated by classifying gastric varices into distinct groups on the basis of anatomicotopographic criteria.

Introduction

Traditional sclerotherapy, which is effective in cases of hemorrhage from esophageal varices, has failed to afford satisfactory resolution of bleeding from gastric varices [1] [2] [3] . Some 13 years ago, Soehendra et al. [4] and Ramond et al. [5] were the first to attempt endoscopic sclerotherapy with N-butyl-2-cyanoacrylate (Histoacryl-L). In 1988, we ourselves adopted the use of Bucrylate in the management of gastric varices, and this study describes the results achieved in a group of 174 patients treated to date.

Patients and Methods

All patients with gastrointestinal hemorrhage referred to our hospital during more than 10 years' round-the-clock operating activity were admitted to the Endoscopic Emergency Surgery Service of the Verona University 1st Division of General Surgery. Since January 1 1988, 5523 patients with gastrointestinal bleeding have been treated, including 657 (11.8 %) with bleeding gastroesophageal varices.

The severity of hemorrhage at admission was assessed on the basis of clinical criteria (10 - 15 % drop in systolic blood pressure and 10 - 15 % increase in heart rate on switching from the supine to the upright position) and on the basis of hematocrit (Ht) and hemoglobin (Hb) values. Hemorrhage was regarded as severe when the patient's systolic blood pressure was below 90mmHg and heart rate > 110 beats/min, together with Hb and Ht below 6.5 mg/dl and 20 % respectively; as moderate, when Hb and Ht were below 8 mg/dl and 25 %, respectively; and mild, when Hb and Ht were below 10 mg/dl and 30 %, respectively.

Endoscopy and emergency injection sclerotherapy (EIS), using fiberoptic endoscopes or video endoscopes of the Olympus GIF IT 100 or Pentax EG 3400 type, were performed within an hour of admission, only after vital conditions had been restored and with continuous monitoring of heart rate and peripheral oxygen saturation. In the case of neurologically compromised patients or patients in a state of coma, the procedure was performed with anesthesiological assitance. Sengstaken-Blakemore tubes were only used in cases of torrential bleeding which could not be controlled by sclerotherapy, pending emergency surgery.

Emergency endoscopic sclerotherapy with 1 % Polidocanol for esophageal varices and with Bucrylate for gastric varices was performed in all patients with bleeding varices or with signs of recent bleeding (varices covered by clots or fibrin rings). The patients with nonbleeding varices (candidates for elective sclerotherapy) and those with bleeding clearly of nonvariceal origin (esophagogastroduodenal erosions/ulcers, Mallory-Weiss-type lesions, angiodysplasias, or benign or malignant tumors) were excluded from the study groups.

Underlying liver disease was classified according to the Child Pugh criteria. The gastric varices were subdivided as follows on the basis of the model proposed by Sarin and Kumar [6]: (a) GOV1, gastric varices of the subcardial ring, associated with esophageal varices; (b) GOV2, gastric varices of the fundus communicating with subcardial varices and associated with esophageal varices; (c) IGV1, isolated gastric varices of the fundus; and (d) IGV2, other isolated gastric varices (of the corpus antrum, sometimes associated with duodenal varices). There were no traces of subcardial varices in the IGV1 and IGV2 groups [6] [7] .

The esophageal varices were classed as types F1, F2, and F3 according to the guidelines of the Japanese Research Society for Portal Hypertension (JRSPH) [8].

The patients of the gastric and esophageal variceal groups were similar for age and sex, cause of portal hypertension, A, B, and C Child's grade, comorbidity, type of acute variceal bleeding, and type of emergency intensive-care treatment to restore patient's vital conditions before performing Immediate Endoscopic Sclerotherapy (IES). Portal congestive gastropathy was classified according to Hashizume et al [9]: mild, with hyperemic mucosa; moderate, with “mosaic”-type or “snake-skin” mucosa; and severe, with “leopard-skin” mucosa, covered with red patches. The bleeding gastric varices were treated by direct injection of the tissue adhesive N-butyl-2-cyanoacrylate (Histoacryl-L), diluted 50 % with Lipiodol, according to our own personal method described in previous studies [10]. The maximum dose injected was normally less than 3 ml per varix, without any limit on the total amount of Bucrylate. This adhesive was injected directly into every bleeding gastric varix in 174 patients and into the bleeding, huge, esophageal varices which were refractory to sclerotherapy in 32 patients; endoscopy was then completed with traditional 1 % Polidocanol sclerosing treatment of the other associated esophageal varices [11].

Esophageal bleeding varices were treated by the traditional freehand technique with combined intravariceal and paravariceal methods, using 1 % Polidaconol with a maximum of 35 - 45 ml material injected at each sclerosis session. Arrest of the bleeding was regarded as effective if hematocrit levels remained stable over the following 48 hours, with no drop compared with the value recorded at the end of sclerotherapy. At 48 hours after the initial sclerotherapy, endoscopic examination was performed to confirm initial hemostasis in each patient. The patients were then given weekly sclerosis sessions for esophageal and gastric varices until all varices were completely eradicated. After obliteration (eradication) of the varices was achieved, the patients underwent endoscopic follow-up examinations at 3, 6, and 12 months in the first year and then at yearly intervals thereafter. All rebleeds were treated according to the procedure described for acute bleeding. The follow-up extended over a mean period of 36 months, ranging from 9 to 90 months. Recurrence of gastric varices was defined as reappearance of gastric varices after their initial successful obliteration, as distinct from the definition of secondary gastric varices, which appear for the first time after obliteration of esophageal varices [7].

Statistics. The chi-squared test with Yates's correction was used to compare the data from the two variceal groups (hemostasis rate, early rebleeding, and 30-day mortality). The survival rates, calculated according to the Kaplan-Meier method, were tested by the logrank test. Significance was set at P < 0.05.

Results

Over the period from January 1 1988 to December 31 1998, 657 cirrhotic patients with bleeding due to rupture of gastroesophageal varices were admitted and treated. These comprised 483 patients with esophageal varices (73.5 %) and 174 with gastric varices (26.5 %) (123 men, 51 women; mean age 56.1, range 7 - 98), of which 159 were primary and 15 were secondary. In particular, the single gastric bleeds consisted of 29 cases of isolated gastric varices and 129 cases of gastric varices associated with nonbleeding esophageal varices (126 cases) or duodenal varices (three cases); double bleeds were from gastric and esophageal varices (15 cases) or from gastric and duodenal varices (one case), as shown in Table [1].

Gastric bleeding and nonbleeding varices associated with bleeding esophageal varices amounted to 174 and 227 cases, respectively, making a total of 401 gastric varices.

The type of bleeding and the Child A, B, C grades of the patients treated are shown in Table [2] for gastric varices and Table [3] for esophageal varices.

The etiology of portal hypertension involved alcoholic cirrhosis in 44.8 % of instances, posthepatitis cirrhosis in 37.5 %, primary biliary cirrhosis in 3.4 %, cystic fibrosis in 2.2 %, variously distributed spleno-porto-mesenteric thrombosis in 10.4 %, and prehepatic portal obstruction due to cavernoma in 1.7%. The blockage of portal circulation was intrahepatic in 87.9 % and prehepatic in 12.1 % (21 cases).

Bleeding and nonbleeding gastric and esophageal varices, classified according to the Sarin and the JSRPH methods, respectively, are listed in Tables [4] and [5], with the bleeding trends in each group: this highlights the overall tendency to bleeding of gastric varices, 43.3 %, compared with 79.8 % in the esophageal varices.

It should be noted that five cases of bleeding IGV2-group varices and ten cases of bleeding GOV2-group varices, making 15 cases out of 174 (8.6 %), bled, following eradication of pure esophageal varices. These were drawn from the 15 cases of IGV2-group (5/15, 33.3 %) appearance and from 14 cases of GOV2-group (10/14, 71.3 %) variceal relapse. These varices, defined as “secondary gastric varices”, appeared for the first time after sclerotherapy for esophageal varices.

The severity of bleeding from gastric varices was mild in 4 % of cases, moderate in 51 % and severe in 45 %; 75.7 % of patients with severe hemorrhage were Child grade C, bleeding from gastric varices of the GOV2 type in 69.7 % of cases.

Bleeding was stopped in 97.1 % of cases. In five cases in which emergency sclerotherapy was unable to stop torrential bleeding arising from indistinguishable fundal varices totally submerged in blood (three cases of GOV2 and two cases of IGV1), three patients required emergency surgery and the other two required emergency implantation of a transjugular intrahepatic portosystemic shunt (TIPS). Both patients with secondary GOV2 varices, one receiving emergency surgery and the other one treated with a TIPS, died in the postoperative period.

Table [6] gives the 30-day results of bleeding gastric varices after injection of Bucrylate, as compared with those in esophageal varices treated in the same period (P > 0.05 = n. s.).

Early rebleeds, within the first 30 days and in any event before variceal obliteration was achieved, occurred in 27 patients (15.5 %); in 26 patients these were variceal rebleeds (11 subcardial, 14 fundal, and one antral), while the other was a case of hemorrhagic hypertensive gastritis. Hemostasis of all rebleeds, even in cases of repeated episodes, was obtained with Bucrylate, except in the patient with hemorrhagic gastritis.

Late rebleeds, arising after eradication, occurred mainly in IGV1 (13 cases), in which complete obliteration of the varices was achieved only in one-third of cases (32.6 %), owing to regional portal hypertension due to prehepatic block caused by isolated thrombosis of the portal vein or pylethrombosis associated with thrombosis of the splenic and/or mesenteric vein. Another five rebleeds were from GOV1 (two cases) and GOV2 (three cases) varices in alcoholic and posthepatitis cirrhosis, making a total of 18 cases, corresponding to 12.8 % of the 141 surviving patients. In addition, there were three cases (1.7 %) of hemorrhagic hypertensive gastritis and five cases (2.9 %) of bleeding mucosal ulcers after extrusion of the cyanoacrylate-cast (Table [7]).

Of the 44 patients with rebleeds (early or late) of exclusively variceal origin, 70.4 % were Child C patients, 40.9 % had hiatal hernia of the gastric fundus, with bleeding varices within the hernia, and 36.3 % were suffering from moderate to severe hypertensive gastritis.

Over a 10-year period, no procedure-related mortality was encountered by the authors. Early and late complications were represented by nine cases of chestpain (5.2 %), without pyrexia, and by five cases of Histoacryl cast extrusion-related bleeding ulcer (2.9 %), respectively. More severe were the early and late complications of esophageal sclerotherapy encountered by the authors during the same period; three cases of complete esophageal wall perforation (0.6 %) with two deaths, 21 cases of postsclerosis bleeding mucosal ulcer (4.3 %), and, finally, 17 cases of esophageal mild stricture (2.8 %), easily cured by endoscopic dilations. The mortality at 30 days (34 patients, 19.5 % of cases) was closely related to the severity of the cirrhosis (30 Child C patients (88.2 %), three Child B patients (8.8 %), and one Child A patient (2.9 %) and also depended on the severity of the bleeding at admission. In cases with mild hemorrhage, the mortality was nil, while in patients with moderate hemorrhage it was 29.4 % (10 cases), and in patients with severe hemorrhage it was 70.6 % (24 cases).

The most frequent cause of death was liver failure (26 cases, 14.9 %), followed by three cases of hemorrhagic shock (1.7 %), two cases of infectious lung complications of cystic fibrosis (1.1 %), one case of septic shock in an HIV-positive patient (0.6 %), and two cases of cardiovascular accidents (1.1 %).

Obliteration of gastric varices was achieved on average in 5.7 sclerosis sessions (range 2.5 - 45) using Bucrylate in a mean total amount of 1113.9 ml (with a mean of 7.9 ml for each patient) and within a period of 2.6 to 26 weeks in a total of 98 of the 141 surviving patients (69.5 %). These comprised 38/51 GOV1 cases (74.5 %) within a time of 2.6 to 6 weeks, 46/63 GOV2 cases (73.0 %) within a period of 3.2 to 9.6 weeks, 6/19 IGV1 cases (31.6 %) within 7.5 to 26 weeks and 8/8 IGV2 cases (100 %) within 2.7 to 7.6 weeks; the mean amount of Bucrylate for each patient was 4.2 ml, 6.3 ml, 14.7 m, and 5.6 ml, respectively, in the groups mentioned.

Sclerosant material such as 1 % Polidocanol has never been used by the authors in elective sclerotherapy for gastric variceal eradication.

In the group of 43 patients with nonobliterated varices, 25 (four with Child B and 21 with Child C cirrhosis) died during follow-up as a result of liver failure (16 cases), hemorrhagic shock (two cases), cancer-cirrhosis (three cases), paralytic stroke (two cases), and myocardial infarction (two cases).

In another 13 cases, of which six were Child A, five were Child B, and two were Child C, failure to eradicate the gastric varices was secondary to a condition of portal hypertension due to prehepatic block, caused by isolated thrombosis of the portal vein (four cases) or of the splenic vein (three cases), or by thrombosis of the portal vein associated with thrombosis of the splenic vein (three cases) or the mesenteric vein (two cases), or thrombosis of the splenoportomesenteric tract (one case). Of the 13 patients in this group, 11 underwent elective surgical treatment and the other two patients, both Child C, underwent TIPS implantation, including one 32-year-old patient with gastric varices which could not be eradicated and were associated with alcoholic cirrhosis; this patient had been abstaining for more than 6 months and had been put on the waiting list for a liver transplant.

The other five patients with nonobliterated gastric varices (two Child B and three Child C IGV1 cases) were suffering from portal thrombosis as a result of alcoholic cirrhosis (two cases) and posthepatitis cirrhosis (three cases), and are still being treated with endoscopic follow-up.

The cumulative survival rates of the 174 patients at 1, 2, and 3 years were 68.2 %, 57.3 %, and 45.7 %, respectively. In the Child A patients, the 1-year survival rate was 87.5 % as against 75 % at years 2 and 3; the Child B patients had one-, two-, and three-year survival rates of 76.1 %, 65.5 %, and 57.6 %, respectively, while the corresponding rates in the Child C patients were 57.2 %, 42.8 %, and 31.3 %.

In addition, the mean survival time in Child B patients with eradicated varices and in those with noneradicated varices was 36.5 months (95 % confidence interval [CI] = 31.5 - 41.5 months) and 22.7 months (95 % CI = 8.7 - 36.66 months), respectively; in Child C patients with eradicated varices and noneradicated varices it was 28.1 months (95 % CI = 21.9 - 34 months) and 15 months (95 % CI = 9.1 - 20.9 months), respectively (Figures [1] and [2]).

The six surviving Child A patients, with varices fully eradicated following the surgical treatment, were not included in the long-term survival table.

The long-term survival of the patients in our experience appears to be affected not only by the severity of the underlying liver disease, but also by the eradication of gastric varices, whether achieved by sclerotherapy (61 cases) or by surgery (11 cases).

Discussion

The incidence of gastric varices in patients with portal hypertension reported in the literature varies considerably: ranging from 2 % in the study by Feldman and Feldman [12] to 5 - 6 % reported by Hosking and Johnson [13], 10 % by Trudeau and Prindeville [1], 16 % by Sarin and Kumar [6], 37 % by Hashizume et al. [14], and 60 - 70 % by Paquet and Oberhammer [15], while Watanabe et al. [16], in their study of portal hemodynamics, report an incidence of gastric varices identical to that of esophageal varices.

This substantial variability of incidence rates probably depends on the fact that the patient populations considered are poorly matched in terms both of grade of cirrhosis and, particularly, the presence of bleeding and nonbleeding varices.

As reported by Sarin, gastric varices are far more frequent in patients with active variceal bleeding than in nonbleeding cases, which clearly demonstrates that actively bleeding gastric varices form at a more advanced stage of portal hypertension and cirrhosis [17].

The presence of isolated varices of the gastric fundus is a rare event, which we observed mainly in cases of regional portal hypertension (12.1 %) caused by prehepatic obstruction of the splenoportomesenteric tract.

Bleeding from gastric varices is less frequent than from esophageal varices, accounting for approximately 8 to 36 % of all variceal bleeds in cirrhotic patients, according to the data reported by the various authors, whose findings tend to be more concordant in this respect [1] [7] [13] .

The tendency to bleed (bleeding frequency) percentages reported in our study in the various types of gastric varices classified according to the Sarin method - i. e. the probability that cirrhotic patients with varices will bleed during a certain time period - are similar to the figures reported by Sarin in the GOV2 and IGV1 groups (60.3 % and 63.6 % vs. 55 % and 78 %, respectively), but significantly greater than Sarin's figures in the GOV1 and IGV2 groups (32.2 % and 25.8 % vs. 12 % and 9 % respectively) [6] [7] . The overall risk of bleeding from gastric varices is also greater in our data when compared to Sarin's figures (43.3 % vs. 24.5 %).

The greater tendency to bleed found in our study as compared to that by Sarin may be due to the fact that our patients mostly belonged to the Child C category (> 58 %) and were suffering from alcoholic and/or posthepatitis cirrhosis in 75 % of cases, whereas more than 30 % of Sarin's patients presented portal hypertension due to noncirrhogenic hepatic fibrosis, with good residual liver function.

On the basis of our results obtained, which are consistent with those reported by other authors, it emerges that Bucrylate injections in the treatment of variceal bleeding allow gastric and esophageal varices to become treatable in the same way, by endoscopy [4] [5] [18] .

In the past, bleeding due to the rupture of gastric varices was always considered a very serious event, complicated by frequent rebleeds and associated with high postoperative mortality rates. At the end of the 1980s, endoscopic sclerotherapy, which had become the treatment of choice for bleeding esophageal varices, failed to show the same degree of efficacy in gastric varices, so much so, indeed, that some authors, at the first rebleed after sclerotherapy, would resort to surgical treatment or, at the beginning of the 1990s, to implantation of a TIPS [1] [3] [19] [20] [21] .

The treatment of gastric varices remains controversial today; noteworthy is the injection of gastric varices with thrombin as an effective method for bleeding control, as suggested in some reports [22] [23] [24] . Williams et al. clearly detail the use of direct intravariceal injection of bovine thrombin as a means of thrombosing gastric varices. Initial experience with gastric varices suggests that thrombin is an important addition to the hemostatic options in the treatment of bleeding fundal gastric varices which cannot be injected with conventional sclerosants. In addition, thrombin has a good safety profile with no evidence of thrombosis distant from the site of injection [23] [24] . Furthermore, this approach does not produce mucosal ulceration, which is a major limitation of sclerosing agents (also occurring with the tissue adhesive) and may be a cause of subsequent rebleeding [5].

The data in our study population of more than 650 patients treated by emergency sclerotherapy for bleeding varices, including 174 gastric varices, clearly show that Bucrylate injection achieved similar results in gastric and esophageal varices, in terms of hemostasis, early rebleeding, and 30-day mortality.

This high degree of efficacy is associated with a low incidence of procedure-related complications and, similarly to the studies of Feretis et al. [25] and Oho et al. [26], no episodes of cerebropulmonary or peripheral thromboembolism and no formation of septicemic abscesses, of the types mentioned by some authors [27] [28] [29] , were shown in our experience.

Bleeding gastric varices treated with Bucrylate occlude after only a single injection, thus reducing the rebleeding rate, which is lowered even further if injection of cyanoacrylate into the gastric varices is followed by traditional sclerotherapy (with 1 % Polidocanol) of the associated nonbleeding esophageal varices [25] [26] [30] .

Combining Bucrylate with sclerotherapy does not necessarily result in a higher complication rate than sclerosis used alone; furthermore, as with the results reported by Thakeb et al., the authors' experience showed that combined treatment may even reduce the complication rate of sclerotherapy [18] [30] . Nonetheless, isolated but serious cases of post-injection Bucrylate embolization-related complications have occurred and been reported by some authors [28] [29] [30] ; overall, in patients where there is suspicion of presumably anomalous right-to-left shunting, care must be taken with the amount of Bucrylate injected per varix at each session [18].

Early gastric rebleeds before variceal eradication occurred at practically the same rate (15.5 %) as esophageal rebleeds (13 %), and in both cases appeared to correlate closely with Child C classification and with the severity of hemorrhage at admission.

On the basis of our analysis of the mortality at 30 days, which was similar in both gastric (19.5 %) and esophageal (16.4 %) varices, it clearly emerges that the use of Bucrylate, which guarantees immediate, stable hemostasis, makes it possible to reduce the mortality due to hemorrhagic shock to less than 10 % of cases. The cirrhotic patients who fail to survive the variceal bleeding, even when controlled successfully, are mainly those who develop liver failure (more than 76 % of the mortality at 30 days), with irreversible coma and hepatorenal or ascitogenic syndrome [31] [32] [33] [34] .

The usefulness of subdividing gastric varices into four groups, as advocated by Sarin, is evident if we consider that the 75 - 100 % variceal obliteration rates achieved in the three types, GOV1, GOV2, and IVG2, were comparable to those achieved in esophageal varices (78 - 81 %). After eradication, these patients presented a mean 36-month survival rate of 62.2 %, which is very similar to that achieved in esophageal varices (68.6 %) [35].

On the other hand, in the group with isolated varices of the gastric fundus (IGV1), eradication was achieved only in one-third of cases (32.6 %). Failure to obliterate the varices or inadequate eradication, which are associated with a reduced long-term survival rate (16.6 %), were also the cause of the high incidence of late rebleeds in the 13 patients with portal hypertension due to prehepatic block. It proved advantageous to be able to identify in advance that these cases belonged to the IGV1 category.

These patients underwent elective anastomotic surgery or resection owing to the long-term inefficacy of sclerotherapy. Follow-up endoscopies after surgery revealed radical healing of the gastric varices, and the patients are still alive in 63.2 % of cases after periods ranging from 12 to 48 months.

On the basis of our experience with Bucrylate, we believe there is no significant difference in the post-sclerotherapy course of patients with alcoholic or post-hepatitis portal hypertension, whether such therapy is given for gastric or for esophageal bleeding varices. After the achievement of rapid, effective hemostasis, the absence of rebleeds and the complete, lasting obliteration of the varices, together with the stability of the patient's liver condition, are the factors mainly responsible for the good medium- to long-term survival rate.

There can be little doubt that the use of cyanoacrylate in emergency sclerotherapy for the control of hemostasis and in elective sclerotherapy for obliteration of varices has made it possible to achieve results (hemostasis, postoperative mortality, rebleed rates, obliteration) in gastric varices on a par with those obtained in esophageal varices in patients with portal hypertension due to intrahepatic block.

In conclusion, these results arise from a personal experience, lacking the methodological rigor inherent in a prospective randomized study, but nevertheless still confirm studies by other authors, already reported in the literature [17] [18] [25] [26] [30] [36] [37] .

The outcome of tissue adhesive injection of bleeding gastric varices in our cirrhotic patients affected by different types of portal hypertension has led to the following conclusions.

• Emergency sclerotherapy with Bucrylate injection of bleeding gastric varices has proved safe and very effective.

• In portal hypertension due to intrahepatic block, gastric varices behave like esophageal varices (GOV1 and 2 group) when injected with Bucrylate for variceal eradication.

• In portal hypertension due to prehepatic block, Bucrylate injection yields no particular benefit in cases of bleeding gastric varices (IGV1 group) where surgery would appear to be more effective for curative aims.

• The choice of treatment indicated (sclerotherapy or surgery) to obtain eradication may be facilitated by classifying gastric varices into four distinct groups on the basis of simple anatomicotopographic criteria.

References

• 1 Trudeau W, Prindiville T. Endoscopic injection sclerosis in bleeding gastric varices.  Gastrointest Endosc. 1986;  32 264-268
  CrossrefPubMedSearch in Google Scholar
• 2 Bretagne J F, Duducourt J C, Morisot D, et al. Is endoscopic variceal sclerotherapy effective for the treatment of gastric varices? (Abstract).  Dig Dis Sci. 1986;  31 5055
  PubMedSearch in Google Scholar
• 3 Paquet K J, Mercado M A, Gad H A. Surgical procedures for bleeding esophago-gastric varices when sclerotherapy fails: a prospective study.  Am J Surg. 1990;  160 43-47
  PubMedSearch in Google Scholar
• 4 Soehendra N, Nam Y C, Grimm H, et al. Endoscopic obliteration of large esophagogastric varices with Bucrylate.  Endoscopy. 1986;  18 25-26
  PubMedSearch in Google Scholar
• 5 Ramond M J, Valla D, Gotlib J P, et al. Obturation endoscopique des varices oeso-gastriques par le Bucrylate. I Etude clinique de 49 malades.  Gastroenterol Clin Biol. 1986;  25 56-60
  PubMedSearch in Google Scholar
• 6 Sarin S K, Kumar A. Gastric varices: profile, classification, and management.  Am J Gastroenterol. 1989;  84 1244-1249
  PubMedSearch in Google Scholar
• 7 Sarin S K, Lahoti D, Saxena S, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients.  Hepatology. 1992;  16 1343-1349
  CrossrefPubMedSearch in Google Scholar
• 8 Japanese Research Society for Portal Hypertension. The general rules for recording endoscopic findings of esophageal varices.  Jpn J Surg. 1980;  10 84-87
  PubMedSearch in Google Scholar
• 9 Hashizume M, Ohta M, Veno K, et al. Endoscopic ligation of oesophageal varices compared with injection sclerotherapy: a prospective randomized trial.  Gastrointest Endosc. 1993;  39 (Suppl 2) 123-160
  PubMedSearch in Google Scholar
• 10 Inaspettato G, Rodella L, Pratticò F, et al. La scleroterapia delle varici esofago-gastriche sanguinanti.  Acta Chir Ital. 1991;  47 988-990
  PubMedSearch in Google Scholar
• 11 Inaspettato G, Kind R, Rodella L, et al. Immediate endoscopic sclerotherapy (I.E.S.) in the treatment of esophago-gastric bleeding varices; technique, complications and results of 312 patients in the 1988 - 1993's years (abstract).  Madrid; IGSC Congress 25 - 29 October 1993
• 12 Feldman M, Feldman M, Jr. Gastric varices.  Gastroenterology. 1956;  30 318-321
  PubMedSearch in Google Scholar
• 13 Hosking S W, Johnson A G. Gastric varices: a proposed classification leading to management.  Br J Surg. 1988;  75 195-196
  PubMedSearch in Google Scholar
• 14 Hashizume M, Kitano S, Tamega H, et al. Endoscopic classification of gastric varices.  Gastrointest Endosc. 1990;  36 276-280
  PubMedSearch in Google Scholar
• 15 Paquet K-J, Oberhammer E. Sclerotherapy of bleeding oesophageal varices by means of endoscopy.  Endoscopy. 1978;  10 7
  Thieme ConnectPubMedSearch in Google Scholar
• 16 Watanabe K, Kimma K, Matsutani S, et al. Portal hemodynamics in patients with gastric varices.  Gastroenterology. 1988;  95 434-440
  CrossrefPubMedSearch in Google Scholar
• 17 Sarin S K. Long term follow-up of gastric variceal sclerotherapy: an eleven-year experience.  Gastrointest Endosc. 1997;  46 8-13
  CrossrefPubMedSearch in Google Scholar
• 18 Binmoeller K F, Soehendra N. “Superglue”: the answer to variceal bleeding and fundal varices.  Endoscopy. 1995;  27 392-396
  PubMedSearch in Google Scholar
• 19 Gimson A ES, Westaby D, Williams R. Endoscopic sclerotherapy in the management of gastric variceal haemorrhage.  J Hepatol. 1991;  13 274-278
  PubMedSearch in Google Scholar
• 20 Merican I, Burroughs A K. Gastric varices.  Eur J Gastroenterol Hepatol. 1992;  4 511-520
  PubMedSearch in Google Scholar
• 21 Conn H O. Transjugular intrahepatic portal-systemic shunts: the status of the art.  Hepatology. 1993;  17 148-158
  CrossrefPubMedSearch in Google Scholar
• 22 Williams S G, Peters R A, Westaby D. Thrombin - an effective treatment for gastric variceal haemorrhage.  Gut. 1994;  35 (Suppl 9) 1287-1289
  CrossrefPubMedSearch in Google Scholar
• 23 Fujii Y, Sugawa C, Ozawa T, et al. Hemostasis activation during esophageal variceal sclerotherapy with thrombin in cirrhotics.  Am Surg. 1991;  57 (4) 222-225
  PubMedSearch in Google Scholar
• 24 Snobl J, Van Buuren H R, Van Blankenstein M. Endoscopic injection therapy using thrombin: an effective and safe method for controlling oesophagogastric variceal bleeding.  Gastroenterology. 1992;  102 A891
  PubMedSearch in Google Scholar
• 25 Feretis C, Dimopoulos C, Benakis P, et al. N-butyl-2-cyanoacrylate (Histoacryl) plus sclerotherapy versus sclerotherapy alone in the treatment of bleeding esophageal varices; a randomized prospective study.  Endoscopy. 1995;  27 355-357
  PubMedSearch in Google Scholar
• 26 Oho K, Iwao T, Sumiao M, et al. Ethanolamine oleate versus butyl cyanoacrylate for bleeding gastric varices: a non randomized study.  Endoscopy. 1995;  27 349-354
  PubMedSearch in Google Scholar
• 27 Lehman R A, Hayes G J. The toxicity of 2-cyanoacrylate tissue adhesives: brain and blood vessels.  Surgery. 1967;  61 915-922
  PubMedSearch in Google Scholar
• 28 See A, Florent C, Lamy P, et al. Cerebrovascular accidents after endoscopic obliteration of esophageal varices with bucrylate in two patients.  Gastroenterol Clin Biol. 1986;  10 (Suppl 8 - 9) 604-607
  PubMedSearch in Google Scholar
• 29 Zambelli A. The New Italian Endoscopic Club: Complications of endoscopic sclerotherapy of esophageal varices.  G Ital Endosc Dig. 1991;  14 103-119
  PubMedSearch in Google Scholar
• 30 Thakeb F, Salama Z, Salama H, et al. The value of combined use of N-butyl-2-cyanoacrylate and ethanolamine oleate in the management of bleeding oesophagogastric varices.  Endoscopy. 1995;  27 358-364
  PubMedSearch in Google Scholar
• 31 Prindiville T, Miller M, Trudeau W. Prognostic indicators in acute variceal hemorrhage after treatment by sclerotherapy.  Am J Surg. 1987;  72 655-659
  PubMedSearch in Google Scholar
• 32 Infante-Rivard C, Esnaola S, Villenueve J P. Clinical and statistical validity of conventional prognostic factors in predicting short-term survival among cirrhotics.  Hepatology. 1987;  7 (Suppl 4) 660-664
  PubMedSearch in Google Scholar
• 33 Heresbach D, Bretagne J F, Raoul J L, et al. Pronostic et facteurs pronostiques de l'hemorragie par rupture de varice chez le cirrhotique à l'ère de la sclérose endoscopique.  Gastroenterol Clin Biol. 1991;  15 838-844
  PubMedSearch in Google Scholar
• 34 Sumino M, Toyonaga A, Tanikawa K. Evaluation of patient outcome following sclerotherapy for esophageal varices.  Gastroenterology. 1996;  31 387-393
  PubMedSearch in Google Scholar
• 35 Inaspettato G, Kind R, Rodella L, et al. Immediate endoscopical sclerotherapy of acutely bleeding esophageal and gastric varices. A seven-yrs' (1988 - 1995) experience on 498 treated patients.  Ital J Dig Endosc. 1997;  20 21-28
  PubMedSearch in Google Scholar
• 36 Westaby S GJ, Westaby D. Management of variceal haemorrhage.  Br Med J. 1994;  308 1213-1217
  PubMedSearch in Google Scholar
• 37 Tran K M, Halter F. Endoscopic management of gastroesophageal varices.  Problems in General Surgery. Philadelphia: Lippincott Williams & Wilkins,. 1998;  15 (Suppl 4) 63-76
  PubMedSearch in Google Scholar

R. Kind, M.D.

I Divisione Clinicizzata di Chirurgia Generale Ospedale Civile Maggiore

Piazzale Stefani I 37126 Verona, Italy

Fax: Fax:+ 39-045-8345355

Email: E-mail:chirurgica.endoscopia@mail.azosp.vr.it

References

• 1 Trudeau W, Prindiville T. Endoscopic injection sclerosis in bleeding gastric varices.  Gastrointest Endosc. 1986;  32 264-268
  CrossrefPubMedSearch in Google Scholar
• 2 Bretagne J F, Duducourt J C, Morisot D, et al. Is endoscopic variceal sclerotherapy effective for the treatment of gastric varices? (Abstract).  Dig Dis Sci. 1986;  31 5055
  PubMedSearch in Google Scholar
• 3 Paquet K J, Mercado M A, Gad H A. Surgical procedures for bleeding esophago-gastric varices when sclerotherapy fails: a prospective study.  Am J Surg. 1990;  160 43-47
  PubMedSearch in Google Scholar
• 4 Soehendra N, Nam Y C, Grimm H, et al. Endoscopic obliteration of large esophagogastric varices with Bucrylate.  Endoscopy. 1986;  18 25-26
  PubMedSearch in Google Scholar
• 5 Ramond M J, Valla D, Gotlib J P, et al. Obturation endoscopique des varices oeso-gastriques par le Bucrylate. I Etude clinique de 49 malades.  Gastroenterol Clin Biol. 1986;  25 56-60
  PubMedSearch in Google Scholar
• 6 Sarin S K, Kumar A. Gastric varices: profile, classification, and management.  Am J Gastroenterol. 1989;  84 1244-1249
  PubMedSearch in Google Scholar
• 7 Sarin S K, Lahoti D, Saxena S, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients.  Hepatology. 1992;  16 1343-1349
  CrossrefPubMedSearch in Google Scholar
• 8 Japanese Research Society for Portal Hypertension. The general rules for recording endoscopic findings of esophageal varices.  Jpn J Surg. 1980;  10 84-87
  PubMedSearch in Google Scholar
• 9 Hashizume M, Ohta M, Veno K, et al. Endoscopic ligation of oesophageal varices compared with injection sclerotherapy: a prospective randomized trial.  Gastrointest Endosc. 1993;  39 (Suppl 2) 123-160
  PubMedSearch in Google Scholar
• 10 Inaspettato G, Rodella L, Pratticò F, et al. La scleroterapia delle varici esofago-gastriche sanguinanti.  Acta Chir Ital. 1991;  47 988-990
  PubMedSearch in Google Scholar
• 11 Inaspettato G, Kind R, Rodella L, et al. Immediate endoscopic sclerotherapy (I.E.S.) in the treatment of esophago-gastric bleeding varices; technique, complications and results of 312 patients in the 1988 - 1993's years (abstract).  Madrid; IGSC Congress 25 - 29 October 1993
• 12 Feldman M, Feldman M, Jr. Gastric varices.  Gastroenterology. 1956;  30 318-321
  PubMedSearch in Google Scholar
• 13 Hosking S W, Johnson A G. Gastric varices: a proposed classification leading to management.  Br J Surg. 1988;  75 195-196
  PubMedSearch in Google Scholar
• 14 Hashizume M, Kitano S, Tamega H, et al. Endoscopic classification of gastric varices.  Gastrointest Endosc. 1990;  36 276-280
  PubMedSearch in Google Scholar
• 15 Paquet K-J, Oberhammer E. Sclerotherapy of bleeding oesophageal varices by means of endoscopy.  Endoscopy. 1978;  10 7
  Thieme ConnectPubMedSearch in Google Scholar
• 16 Watanabe K, Kimma K, Matsutani S, et al. Portal hemodynamics in patients with gastric varices.  Gastroenterology. 1988;  95 434-440
  CrossrefPubMedSearch in Google Scholar
• 17 Sarin S K. Long term follow-up of gastric variceal sclerotherapy: an eleven-year experience.  Gastrointest Endosc. 1997;  46 8-13
  CrossrefPubMedSearch in Google Scholar
• 18 Binmoeller K F, Soehendra N. “Superglue”: the answer to variceal bleeding and fundal varices.  Endoscopy. 1995;  27 392-396
  PubMedSearch in Google Scholar
• 19 Gimson A ES, Westaby D, Williams R. Endoscopic sclerotherapy in the management of gastric variceal haemorrhage.  J Hepatol. 1991;  13 274-278
  PubMedSearch in Google Scholar
• 20 Merican I, Burroughs A K. Gastric varices.  Eur J Gastroenterol Hepatol. 1992;  4 511-520
  PubMedSearch in Google Scholar
• 21 Conn H O. Transjugular intrahepatic portal-systemic shunts: the status of the art.  Hepatology. 1993;  17 148-158
  CrossrefPubMedSearch in Google Scholar
• 22 Williams S G, Peters R A, Westaby D. Thrombin - an effective treatment for gastric variceal haemorrhage.  Gut. 1994;  35 (Suppl 9) 1287-1289
  CrossrefPubMedSearch in Google Scholar
• 23 Fujii Y, Sugawa C, Ozawa T, et al. Hemostasis activation during esophageal variceal sclerotherapy with thrombin in cirrhotics.  Am Surg. 1991;  57 (4) 222-225
  PubMedSearch in Google Scholar
• 24 Snobl J, Van Buuren H R, Van Blankenstein M. Endoscopic injection therapy using thrombin: an effective and safe method for controlling oesophagogastric variceal bleeding.  Gastroenterology. 1992;  102 A891
  PubMedSearch in Google Scholar
• 25 Feretis C, Dimopoulos C, Benakis P, et al. N-butyl-2-cyanoacrylate (Histoacryl) plus sclerotherapy versus sclerotherapy alone in the treatment of bleeding esophageal varices; a randomized prospective study.  Endoscopy. 1995;  27 355-357
  PubMedSearch in Google Scholar
• 26 Oho K, Iwao T, Sumiao M, et al. Ethanolamine oleate versus butyl cyanoacrylate for bleeding gastric varices: a non randomized study.  Endoscopy. 1995;  27 349-354
  PubMedSearch in Google Scholar
• 27 Lehman R A, Hayes G J. The toxicity of 2-cyanoacrylate tissue adhesives: brain and blood vessels.  Surgery. 1967;  61 915-922
  PubMedSearch in Google Scholar
• 28 See A, Florent C, Lamy P, et al. Cerebrovascular accidents after endoscopic obliteration of esophageal varices with bucrylate in two patients.  Gastroenterol Clin Biol. 1986;  10 (Suppl 8 - 9) 604-607
  PubMedSearch in Google Scholar
• 29 Zambelli A. The New Italian Endoscopic Club: Complications of endoscopic sclerotherapy of esophageal varices.  G Ital Endosc Dig. 1991;  14 103-119
  PubMedSearch in Google Scholar
• 30 Thakeb F, Salama Z, Salama H, et al. The value of combined use of N-butyl-2-cyanoacrylate and ethanolamine oleate in the management of bleeding oesophagogastric varices.  Endoscopy. 1995;  27 358-364
  PubMedSearch in Google Scholar
• 31 Prindiville T, Miller M, Trudeau W. Prognostic indicators in acute variceal hemorrhage after treatment by sclerotherapy.  Am J Surg. 1987;  72 655-659
  PubMedSearch in Google Scholar
• 32 Infante-Rivard C, Esnaola S, Villenueve J P. Clinical and statistical validity of conventional prognostic factors in predicting short-term survival among cirrhotics.  Hepatology. 1987;  7 (Suppl 4) 660-664
  PubMedSearch in Google Scholar
• 33 Heresbach D, Bretagne J F, Raoul J L, et al. Pronostic et facteurs pronostiques de l'hemorragie par rupture de varice chez le cirrhotique à l'ère de la sclérose endoscopique.  Gastroenterol Clin Biol. 1991;  15 838-844
  PubMedSearch in Google Scholar
• 34 Sumino M, Toyonaga A, Tanikawa K. Evaluation of patient outcome following sclerotherapy for esophageal varices.  Gastroenterology. 1996;  31 387-393
  PubMedSearch in Google Scholar
• 35 Inaspettato G, Kind R, Rodella L, et al. Immediate endoscopical sclerotherapy of acutely bleeding esophageal and gastric varices. A seven-yrs' (1988 - 1995) experience on 498 treated patients.  Ital J Dig Endosc. 1997;  20 21-28
  PubMedSearch in Google Scholar
• 36 Westaby S GJ, Westaby D. Management of variceal haemorrhage.  Br Med J. 1994;  308 1213-1217
  PubMedSearch in Google Scholar
• 37 Tran K M, Halter F. Endoscopic management of gastroesophageal varices.  Problems in General Surgery. Philadelphia: Lippincott Williams & Wilkins,. 1998;  15 (Suppl 4) 63-76
  PubMedSearch in Google Scholar

R. Kind, M.D.

I Divisione Clinicizzata di Chirurgia Generale Ospedale Civile Maggiore

Piazzale Stefani I 37126 Verona, Italy

Fax: Fax:+ 39-045-8345355

Email: E-mail:chirurgica.endoscopia@mail.azosp.vr.it