|
Keywords
Band litigation (EVL), endoscopy, gastric varices, sclerotherapy (EST).
Introduction
|
Figure 1 |
|
Figure 2 |
|
Figure 3 |
|
Figure 4 |
|
Figure 5 |
|
Figure 6 |
|
Video Clip 1: Combination therapy for gastric varices.
|
Gastric varices are caused most commonly by cirrhosis with associated portal hypertension. However, patients with pancreatic disease, especially inflammatory pancreatic disease, can develop splenic vein thrombosis with subsequent formation of gastric varices, which can be isolated only to the stomach. Portal vein thrombosis can cause gastric varices, and there have been several reports of gastric varices developing after endoscopic therapy for bleeding esophageal varices, particularly sclerotherapy (EST) with a lower incidence after band ligation (EVL). Gastric varices are classified into four categories: GOV1, GOV2, IGV1, and IGV2. GOV1 and GOV2 are gastric varices that extend from esophageal varices across the gastroesophageal junction (1,2). In GOV1, they only extend 5 cm or less, while in the GOV2 group, the varices extend into the fundus. IGV stands for isolated gastric varices. IGV1 are varices found only in the fundus and IGV2 are isolated non-fundic varices. The junctional (GOV) varices are usually on the lesser curvature. They are deep submucosal veins that arise from the left gastric vein. Fundic varices (IGV) develop from the short gastric and posterior gastric veins or via direct anastomoses with retroperitoneal veins. They are often associated with large gastrorenal vein shunts.
Gastric varices can be found in 15-20% of patients with portal hypertension. The rate may be higher in those who have already had an episode of bleeding. About 14-36% will bleed and the associated mortality is 30-52%, with a 30% rebleeding rate. Larger varices, those with associated red spots and those found in patients with advanced liver disease as defined by the Child’s classification are more likely to bleed (3).
Treatment Options
Treating gastric varices can be very challenging. Though criteria for varices that are at high risk for bleeding have been developed, there are no reports of attempts at prophylactic treatment using endoscopic based therapy. Medical management is similar to the treatment of esophageal varices. Vasopressin and octreotide are useful for control of acute bleeding, and beta-blockers and long-acting nitrates may prevent subsequent bleeding. The Sengstaken-Blakemore tube may have some utility for bleeding from junctional (GOV1 or 2) varices, but would have little utility for varices in the fundus or further down in the stomach.
Successful hemostasis and obliteration of gastric varices has been reported with intravariceal injections of sclerosant, absolute alcohol, thrombin, and cyanoacrylate, which is not currently available in the United States. There is less experience with EVL in this setting. In an initial report, EVL stopped acute bleeding in 16 of 18 (88 percent) patients. Six patients died and bleeding recurred in three. However, many of these patients had noncirrhotic portal hypertension due to schistosomiasis. It is uncertain if similar results would be obtained in patients with advanced hepatic synthetic dysfunction.
EST has been used with modest success particularly for junctional varices using a variety of sclerosants. Control of acute bleeding with EST has ranged from 62-100%, while long-term control with successful obliteration of varices has ranged from 0-94% (4,5). However, the high success rate has been in patients with GOV1. In one study, the obliteration rates were 94.4%, 70.4%, and 41% respectively for GOV1, GOV2, and IGV1 type varices. The same study recorded rebleeding rates of 5.5%, 19%, and 53% in these subgroups; 50% of a group treated with alcohol, as the sclerosant, required surgery to ultimately control their bleeding.
EVL alone or in combination can be effective for controlling bleeding from gastric varices. In a study comparing cyanoacrylate to EVL for gastric varices, control of active bleeding was achieved in 40-50% of patients depending on the type of gastric varix (most were GOV) (6). The physicians achieved a 45% obliteration rate in an average of 2 sessions. There was a 42-55% rebleeding rate. In a small study using only EVL, there was only a 25% rebleeding rate with 22% mortality, and they achieved obliteration with an average of three sessions (7).
Combination therapy using EVL and EST for esophageal varices has been used and has been found to have variable success rates. Reveille et al. had a 30% early rebleeding rate and were able to achieve variceal eradication in 76% of their patients with a mean of 3.2 sessions (8). Saeed et al. compared EVL alone to combination therapy and were unable to show any benefit for combination therapy for esophageal varices and in fact showed there were more complications with the combination therapy (9). A Japanese study using EVL and polidocanol injection for gastric varices reported 100% control of acute bleeding (10). They had a 12.5% variceal recurrence rate with only a 3.6% rebleeding rate. My personal anecdotal experience is that combination therapy can be used to treat gastric varices with acceptable control of bleeding and eradication in selected patients.
Combination methods have been used to treat varices in the gastric cardia and fundus. The varices are banded just below the gastroesophageal junction, usually with the endoscope in the retroflexed position. One or two bands are applied to each visible varix. Large varices in the fundus, well away from the esophageal junction, should not be banded. These are often fed by huge vessels just below the surface; as a result, they are difficult to control with any endoscopic method and have a high risk of rebleeding.
Combination therapy can be used in any patient with documented bleeding from varices in the cardia. This is best performed when there is not too much blood in the area, since clear vision of the varices is required. Combination therapy can be tried as initial therapy, but a low threshold should be maintained for moving on to shunting (radiologic or surgical) if there is recurrent bleeding or the acute bleeding cannot be controlled. Intravenous octreotide should be used in these patients in the acute setting to augment the effect of the endoscopic intervention.
Several variations in the types of sclerosants and the protocol for administering EST in combination with EVL have been described. As an example, one technique involves injection of 1 mL of 5 percent ethanolamine oleate into each visible varix above EVL sites during the initial session. Subsequent sessions are performed at the same time intervals as for EVL alone, but both procedures are done as indicated by the findings at each endoscopy. Another study involving a small number of patients suggested that better results might be obtained by placing a band near the GE junction, followed by injection of sclerosant in the varix proximally, and then ligating again just above the injection site. The technique I use involves injecting 1-2 ml of 5% ethanolamine oleate proximal to the ligation sites using a standard 23-gauge sclerotherapy injection needle, with the intent of achieving intravariceal injection of the sclerosant.
Surgical management of bleeding esophageal varices includes various types of portocaval and mesocaval shunts, varix over-sewing, and devascularization procedures (Segura technique, etc.). These are rarely used due to a combination of lack of experience by most surgeons, moderately high mortality, substantial rates of encephalopathy after most shunt procedures in patients with advance liver disease, and lack of efficacy of non-shunt procedures except in the hands of devoted experts.
The major salvage or perhaps even primary treatment approach for gastric varices in the United States is the TIPS procedure (transjugular intrahepatic portasystemic shunt) (11). Chau et al. in 1998 reported on the utility of TIPS for salvage therapy in patients with uncontrolled bleeding from either gastric or esophageal varices (12). They were able to control bleeding acutely in 98.8% of the patients with esophageal varices and 96.4% of the patients with gastric varices. Rebleeding rates were 24% and 29% respectively. Most of the early rebleeding was due to sclerotherapy ulcerations. However, Escorsell et al. did a cost analysis comparing TIPS to medical treatment with propranolol and isosorbide-5-monnitrate, which found that TIPS patients rebled less, but had more encephalopathy and a greater cost of treatment while not increasing the patient’s survival (13). Consequently, TIPS is probably the best salvage procedure.
Therapies currently not readily available in the United States or minimally studied include the following: cyanoacrylate injections, thrombin injections, detachable snares, endoclips, Doppler assisted injections of sclerosant, and balloon occlusion retrograde transvenous obliteration (14). The most important and well studied of these is cyanoacrylate injection. Many investigators in numerous countries around the world have used this quickly solidifying glue. In randomized trials, it has proved to be better than EST or EVL in controlling bleeding and preventing rebleeding (15,16). In a non-controlled trial, Huang et al. showed that they could achieve definitive hemostasis in 93.3% of their 90 patients using an average of 1.3 cyanoacrylate injection sessions (17). There was a 23.3% recurrent bleeding rate. They achieved obliteration of the varices in 36.1% of the patients. They noted that 22 of 61 evaluable patients extruded their cyanoacrylate cast in an average of 6 months. The survival rate in these patients was best if they were Child’s class A or B, did not have malignancy, and did not have portal vein thrombosis. Lo et al. compared cyanoacrylate injection to band ligation for bleeding gastric varices (6). They were able to control 87% of their acutely bleeding patients. They achieved obliteration in 51% with an average of 2.7 sessions per patient. The rebleeding rate was 31% with the high percentage in the IGV1 group. The overall mortality was 29%.
There are two fully published reports of using thrombin injections to control gastric variceal bleeds, both of which came from England. Przemioslo et al. achieved initial hemostasis in 94% of their 52 patients (18). The 6 week rebleed rate was 18% in the surviving 49 patients. Yang et al. reviewed their results with 12 patients treated with thrombin injection (19). They had a 25% rebleeding rate after initial treatment. There is no intravenous thrombin approved for use in the United States, only topical forms.
Detachable snares have been reported to be effective for treating large gastric varices in one small study (20). Lee et al. had a 2.9% overall hemostatic rate in their 41 patients treated with detachable snares (21). Endoloops (detachable snares) are available from the Olympus Corporation for use in the United States. Endoclips have been reported to possible be effective for treating gastric varices, but the size of the endoclip bite limits their use to smaller varices (22).
It is imperative that all patients treated with any of the above mentioned interventions, except medical management, also receive treatment with a PPI to suppress acid secretion and prevent complications related to acid interaction with bands, injection sites, and treatment related ulcers.
Summary
Gastric varices represent a significant challenge for the gastroenterologist/endoscopist since they are a more serious complication of portal hypertension and are more difficult to control, especially through endoscopic means. Today, in the United States, endoscopic therapy is limited to EST or EVL, but the available information seems to support the use of combination therapy as the best method for endoscopic treatment of gastric varices. Currently, endoscopic therapy should be limited to GOV1 and some GOV2 with only limited extension into the fundus. We could use more information from controlled trials in this area, but the information will be difficult to obtain because of the limited number of patients who have this problem. Any clinical trial will need to be a multicenter trial. There is a possibility that cyanoacrylate will eventually enter the US market, which would be a significant aid for endoscopically managing these patients.
References
1. Sarin SK, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertensive patients. Hepatology 1992;16:1343-9. <Related link>
2. Korula J, et al. Demonstration of two distinct subsets of gastric varices. Observations during a seven-year study of endoscopic sclerotherapy. DDS 191;36:303-9. <Related link>
3. Kim T, et al. Risk factors for hemorrhage from gastric fundal varices. Hepatology 1997;25:307-12. <Related link>
4. Trudeau W and Prindiville T. Endoscopic injection sclerosis in bleeding gastric varices. Gastro Endo 1986;32:264-8. <Related link>
5. Sarin SK. Long-term follow-up of gastric variceal sclerotherapy: an eleven-year experience. Gastro Endo 1997;46:8-14. <Related link>
6. Lo G-H, et al. A prospective, randomized trial of butyl cyanoacrylate injection versus band ligation in the management of bleeding gastric varices. Hepatology 2001;33:1060-4. <Related link>
7. Shiha G and El-Sayed SS. Gastric variceal ligation: a new technique. Gastro Endo 1999;49:437-41. <Related link>
8. Reveille RM, et al. Combination endoscopic variceal ligation (EVL) and low-volume endoscopic sclerotherapy (ES) for bleeding esophageal varices: a faster route to variceal eradication. Gastro Endo 1991;37:243.
9. Saeed ZA, et al. Endoscopic variceal ligation is superior to combined ligation and sclerotherapy for esophageal varices: a multicenter prospective randomized trial. Hepatology. 1997 Jan;25(1):71-4. <Related link>
10. Arakaki Y, Murakami K, Thahashi K, et al. Clinical evaluation of combined endoscopic variceal ligation and sclerotherapy of gastric varices in liver cirrhosis. Endoscopy 2003, 35:940. <Related link>
11. Boyer TD. Transjugular intrahepatic portosystemic shunt: current status. Gastroenterology 2003;124:1700-10. <Related link>
12. Chau TN, et al. "Salvage" transjugular intrahepatic portosystemic shunts: Gastric fundal compared with esophageal variceal bleeding. Gastroenterology 1998;114:981-7. <Related link>
13. Escorsell A. et al. TIPS versus drug therapy in preventing variceal rebleeding in advanced cirrhosis: a randomized controlled trial. Hepatology 2002;35:385-92. <Related link>
14. Shiba M, et al. Efficacy and safety of balloon-occluded endoscopic injection sclerotherapy as a prophylactic treatment for high-risk gastric fundal varices: a prospective, randomized, comparative trial. Gastro Endo 2002;56:522-8. <Related link>
15. Jutabha R, et al. Randomized, prospective study of cyanoacrylate injection sclerotherapy, or rubber band ligation for endoscopic hemostasis of bleeding canine gastric varices. Gastro Endo 1995;41:201-5. <Related link>
16. Sarin SK, et al. A randomized controlled trial of cyanoacrylate versus alcohol injection in patients with isolated fundic varices. AJM 2002;97:1010-5. <Related link>
17. Huang YH, et al. Endoscopic treatment of bleeding gastric varices by N-butyl-2-cyanoacrylate (Histoacryl) injection: long-term efficacy and safety. Gastro Endo 2000;52:160-7. <Related link>
18. Przemioslo RT, et al. Thrombin is effective in arresting bleeding from gastric variceal hemorrhage. DDS 1999;44:778-81. <Related link>
19. Yang WY, et al. Endoscopic use of human thrombin in bleeding gastric varices. AJG 2002;97:1381-5. <Related link>
20. Lee MS, et al. Use of detachable snares and elastic bands for endoscopic control of bleeding from large gastric varices. Gastro Endo 2002;56:83-8. <Related link>
21. Koutsomanis D. Endoscopic clipping for bleeding varices. Gastro Endo 1994;40:126-7. <Related link>
22. Yoshida T, et al. Endoscopic ligation of gastric varices using a detachable snare. Endoscopy 1994;26:502-5. <Related link>
|
