Use of Thrombin for the Treatment of Esophageal Varices

Esophageal varices are varicosities of the branches of the azygous vein which anastomoses with the tributaries of the portal vein in the lower esophagous (1). Esophageal varices occur in patients with portal hypertension most often secondary to liver cirrhosis with the cirrhosis most commonly associated with alcohol abuse. These varicosities are occasionally referred to as uphill varices to differentiate this pathology from the much rarer downhill varices caused by superior vena cava obstruction. Varices of either origin have substantial risk of rupture and hemorrhage (2-4). Patients with liver cirrhosis may have esophageal varices at the time of presentation with the incidence dependent on the severity of the cirrhosis (2-6). Patients with compensated cirrhosis have 30% incidence while 70% of the patients with decompensated cirrhosis have esophageal varices. Individuals who do not present with varices have a substantial risk of development of varices following initial presentation. Bleeding from the varices is a common problem and quite difficult to control. In one of the earlier studies in this area, Graham and Smith (4) suggested a mortality rate in excess of 30% during the initial hospitalization with a substantial risk of rebleeding. Wu and Sun have reported that variceal hemorrhage is responsible for one-third of the mortality associated with cirrhosis (7). Ferguson and colleagues noted a mortality rate of 30-50% for variceal hemorrhage (8). Krige and colleagues (9) reported that even with control of variceal bleeding, there was only a 26% survival in their treatment group after five years. Despite these gloomy statistics, there has been a significant decrease in the reduction of mortality in this population in the last twenty years (10). This positive news must be balanced against a perception of increased alcohol abuse (11).

The treatment of bleeding gastric varices presents a considerable challenge to the physician (5,6, 12). While a pharmacological approach using beta-blockers and other anti-hypertension drugs is available (13), most interest is directed at endoscopic approaches to the management of bleeding gastric varices (5,6,14) including banding or variceal ligation and injection sclerotherapy. These two techniques are in addition to the use of a transjugular intrahepatic portosystemic stent shunt (TIPS/TIPSS) (15-19) or a distal spleno-renal shunt (DSRS)(20,21).

Injection sclerotherapy for esophageal varices was introduced by Crafoord and Frenckner in 1939 (22). These investigators administered quinine-uretan with an oesophagealscope to treat esophageal variceal bleeding. Subsequently Hunt and coworkers used ethanolamine oleate as the sclerosing agent with a MacBeth needle as the administration device (23). Subsequent studies (24-26) validated this therapeutic approach. Studies in the last twenty years have focused on the development of more effective sclerosing agents.

The focus of this article is the use of thrombin in the treatment of esophageal varices as a form of injection sclerotherapy. The purpose of injection sclerotherapy is to "harden" the blood vessels leading to their eventual resolution (27-30). Basic injection sclerotherapy uses a variety of sclerosing agents including ethanolamine oleate, saline, tetradecyl sulfate, or alcohol, or a combination (31,32) . In addition to the basis injection sclerotherapy, there has been the use of tissue adhesives or glue including cyanoacrylates (10, 33-35) and fibrin glue (36, 37). A discussion of fibrin glue/fibrin sealant is shown below. Cyanoacrylates are popular but will polymerize prior to entrance into the varicosity (32, 38). There is also the issue of the cyanoacrylate plug embolizing into the circulation (39). There are also other complications arising from cyanoacrylate use including sepsis (40) and formation of an inflammatory tissue tumor (41)

Thrombin, while well accepted for the treatment of vascular pseudoaneurysms via percutaneous administration, has a somewhat less-substantial status in the treatment of gastric varices (32, 35, 42-44). This situation exists despite a significant history which seems longer than the record of thrombin use in vascular pseudoaneurysms. Sunderquist and Vang injected thrombin in hypertonic glucose into esophageal varices in 1974 (45, 46). A number of studies followed (47-50) including some which used gel foam and thrombin (51, 52).

The current use of thrombin for the treatment of esophageal bleeding appears to be based on studies by Williams and coworkers (42) in 1994. These investigators used bovine thrombin (1000 U/mL) in 11 patients with bleeding gastric varices. A mean volume of 5.5 mL (5,500 U) was used and produced initial hemostasis in all subjects. Varices were thrombosed and/or obliterated after several injections. There was one incident of a rebleed during the nine month followup period. A subsequent study using human material(43) also demonstrated the effectiveness of thrombin in the treatment of gastric varices. In another study, Kojima and colleagues (44) als0 used human thrombin (described as thrombin glue and referred to in units; this might reflect the use of thrombin "cannibalized" from a fibrin sealant kit) to seal the scleroscent injection site rather than a thrombosing agent within the varicosity. To the best of my knowledge, a free-standing human thrombin commercial product has not been available since the second World War. Physicians who have used thrombin have either canabilized the material from a fibrin sealant kit or used autologous thrombin.

Westaby (53) has reviewed variceal injection sclerotherapy and concluded that while cyanoacrylates are effective in sealing the varicosities, there are technical challenges with the use of this material in the injection process which can result in irreversible damage to the endoscope. Westaby notes that thrombin does not present this problem, does not cause mucosal damage as is the case with cyanoacrylates, and can be injected with the sclerosing agent.

The injection of thrombin for endoscopic sclerotherapy has not, to the best of my knowledge, been associated with any adverse event of clinical significance. Gardner and Brooks (54) used a mixture of sodium morrhuate, topical thrombin, and cephalothin sodium with 50% dextrose as a sclerosing agent in 11 patients with esophageal varices. There was no effect on panel of blood coagulation factors including factor V and factor VIII. These investigators also reported the lack of effect on the prothrombin time, the partial thromboplastin time, platelet count, or fibrinogen split products. The studies, as with most of the studies, were performed with bovine topical thrombin which has been associated with immunological side reactions (55). Several subsequent studies on the use of bovine thrombin in the treatment of esophageal varices were equivocal in theirs results. de Franchis and colleagues (56) reported that global coagulation assays such as the prothrombin time and partial thromboplastin time did not change. There was a marked increase in fibrinopeptide A levels after treatment. This effect was not sustained with FPA levels returning to normal after 24 hours. These investigators did report abnormally high basal levels of FPA in approximately half of the patients in this study. Fugii and colleagues (57) used thrombin with tetradecyl sodium and cafzolium as a sclerosing solutions in 24 patients with esophageal varices. There was a transient elevation of fibrinogen D-dimer and thrombin-antithrombin in the treated patients. There were no thrombotic or bleeding complications. These investigators also reported elevation of D-dimer, tissue plasminogen activator, and plasminogen activator inhibitor-1 in most patients prior to treatment. Koyama and colleagues (58) also observed an increase in thrombin-antithrombin, D-dimer, and plasminogen activator inhibitor-1 following the administration of thrombin for the treatment of esophageal varices. Elevation of thrombin-antithrombin and D-dimer was more persistent than that observed by others lasting seven days. Biagi and colleagues (59) observed transient changes in β-thromboglobulin, platelet factor 4, and fibrinopeptide A following sclerotherapy with ethanolamine oleate without thrombin. In another study (60) which also used ethanolamine oleate without thrombin, Takada and coworkers observed a transient increase in thrombin-antithrombin complex following sclerotherapy. The extent of the increased was dependent on the amount of sclerosant used.

In one of the exceptions to the use of bovine thrombin, Kitano and coworkers (61) used human thrombin (100-150 U) in 5% ethanolamine oleate for endoscopic sclerotherapy of esophageal varices and did not observed any signs of coagulation disorders such as disseminated intravascular coagulation. The quality of the human thrombin used in these studies is not known but it would be unlikely to contain factor V/Va which is notable contaminant of bovine thrombin preparations (55).

Pseudoanerysms secondary to cardiac catheterization is a problem with some similarity to gastric varices in that hemorrhage is an unwanted consequence. In a study on the use of thrombin for the treatment of pseudoaneurysms, Kruger and colleague (62) reported an increase in thrombin-antithrombin complexes after ultrasonographically (US)guided thrombin injection. There were no other laboratory or clinical changes. In a subsequent larger study with 240 patients with simple and complex pseudoanerysms (63), there were two thrombotic episodes which resolved spontaneously. A mild allergic response was observed in one subject. This study used bovine thrombin which, as noted above, has been associated with immunological side reactions. It is also likely that the amount of thrombin used with the sclerosing agent can be reduced as has been observed for the use of thrombin with the percutaneous treatment of vascular pseudoaneurysms (64).

It is noted that another hemostatic agent, recombinant factor VIIa, has been used without success (65) and in one case with an accompanying cerebrovascular event (66). It is not clear as to why recombinant VIIa would have a functional advantage over thrombin in this clinical situation. Recombinant factor VIIa would also likely be more expensive than thrombin.

The various investigators who have used thrombin have been mostly successful in the use of the thrombin as a hemostatic agent for gastic variceal bleeding but there is the sense that a randomized clinical trial would be useful which would compare the various approaches.

A final note on thrombin and fibrin sealant

Thrombin is an enzyme which catalyzes the cleavage of peptide bonds in fibrinogen results in the formation of fibrin and aggregates blood platelets. Both these reactions occur on the addition of thrombin to blood. With a few exceptions, such as disseminated intravascular coagulation, there is sufficient fibrinogen and platelets in blood to form an adequate thrombus.

Thrombin is rapidly inactivated by plasma serpins and, to the best of my knowledge, there is no thrombotic or hemorrhagic event associated with the topical use OR with the use in the treatment of pseudoaneurysms or esophageal varices. It is noted that thrombin bound to fibrin resulting from fibrin sealant is protected from inactivation by serpins.

Thrombin is a free-standing biological product AND a component of fibrin sealant.

Thrombin and fibrin sealant are different products with somewhat different indications.

Fibrin sealant forms a polymer network independent of contact with blood or tissue. In this respect fibrin sealant is similar to cyanoacrylate glues (35, 67-69). Fibrin sealant is considered to be a biological glue while cyanoacrylates are non-biological glues. In principle, fibrin sealant will not polymerize until the two components are mixed and extruded from the delivery device. Cyanoacrylates will spontaneously polymerize upon contact with water


References Cited

1. Levine, M.S. and Rubesin, S.E., Radiology of the pharynx and esophagus, in The Esophagus, 4th edn., ed. D.O. Castell and J.E. Richter, Lippincott, Williams and Wilkins, Philadelphia, PA, Chapter 3, ps 47-105, 2004.

2. Brick, I.B. and Palmer, E.D., Incidence and diagnosis of esophageal varices in cirrhosis of the liver: an esophageal study, Gastroenterology 25, 378-384, 1953

3. Sanz, C.J. and Suarez, L.B., Incidence of esophageal varices in cirrhois of the liver, Am.J.Gastroenterol. 29, 156-162, 1958

4. Graham, D.Y. and Smith, J.L., The course of patients after variceal hemorrhage, Gastroenterology 80, 800-809, 1981.

5. Elta, G.H., Approach to the patient with gross gastrointestinal bleeding, in Textbook of Gastroenterology, 4th edn., ed. T. Yamada, D.H. Alpers, et al., Lippincott, Williams and Wilkins, Philadelphia, PA, Chapter 33, pps. 698-723, 2004

6. Sung, J.J.Y. and Lau, J.Y.W., Endoscopic therapy for upper gastrointestinal variceal hemorrhage, in Textbook of Gastroenterology, 4th edn., ed. T. Yamada, D.H. Alpers, N. Kaplowitz, et al., Lippincott, Williams, and Wilkins, Philadelphia, PA., Chapter 137, pps 3008-3027, 2003.

7. Wu, J.C. and Sun, J.J., Update on treatment of variceal hemorrhage, Dig.Dis. 20, 134-144, 2002.

8. Ferguson, J.W., Tripathi, D., and Haynes, C., The management of acute variceal hemorrhage, Aliment.Pharmacol.Ther. 18, 253-262, 2003

9. Krige, J.E.J., Kotze, U.K., Bornman, P.C., et al., Variceal recurrence, rebleeding, and survival after endoscopic injection sclerotherapy in 287 alcoholic cirrhotic patients with bleeding esophageal varices, Ann.Surg. 244, 764-770, 2006.

10. Chung, S., Management of bleeding in the cirrhotic patients, J.Gastrointestinal Hepatol. 17, 355-360, 2002.

11. Sargent, S., The aetiology, management, and complications of alcoholic hepatitis, Br.J.Nurs. 14, 556-562, 2005.

12. Berry, P.A. and Wendon, J.A., The management of severe alcoholic liver disease and variceal bleeding in the intensive care unit, Curr.Opin.Crit.Care 12, 171-177, 2006.

13. Lopez-Mendez, E. and Uribe, M., Beta blockers in portal hypertension. Are they really a good option?, Ann.Hepatol. 5, 86-91, 2006.

14. Krige, J.E.J., Shaw, J.H., and Bornman, P.C., The evolving use of endoscopic treatment for bleeding esophageal varices, World J. Surg. 29, 966-973, 2005.

15. Fleig, W.E., Non-surgical treatment of variceal bleeding, Adv.Surg. 33, 439-458, 1999.

16. Conn, H.O., Portal hypertension, varices, and transjugular intrahepatic portosystemic shunts, Clin.Liver Dis. 4, 133-150, 2000.

17. Bosch, J. and Abraldes, J.G., Management of gastrointestinal bleeding in patients with cirrhosis of the liver, Semin.Hematol. 41(1 Suppl 1), 8-12, 2004

18. Tripathi, D., Ferguson, J., Barkell, H., et al., Improved clinical outcome with transjugular intrahepatic portosystemic stent-shunt utilizing polytetrafluoroethylene-covered stents, Eur.J.Gastroenterol.Hepatol. 18, 225-232, 2006

19. Kochar, N., Tripathi, D., Ireland, H.. et al., Transjugular intrahepatic portosytemic stent shunt (TIPSS) modification in the management of post-TIPSS refractory hepatic encephalopathy, Gut 55, 617-623, 2006.

20. Henderson, J.M., Boyer, T.D., Kudner, M.H., et al., Distal splenorenal shunt versus transjugular intrahepatic portal systematic shunt for variceal bleeding: a randomized trial. Gastroenterology 130, 1643-1651, 2006.

21. Koniaris, L.G., Perez, E.A., and Livingstone, A.S., DSRS versus TIPS for variceal bleeding: Lessons from late follow-up of 507 DSRS patients, Gastroenterology 131, 978, 2006.

22. Crafoord, C. and Frencken, P., New surgical treatment of varicous veins of the oesophagus, Acta Otolaryngology 27, 422-429, 1939.

23. Hunt, P.S., Johnston, G.W., and Rodgers, H.W., The emergency management of bleeding oesophageal varices with sclerosing injection, Brit.J.Surg. 56, 305-307, 1969.

24. Westaby, D., MacDougall, B.R.D., and Williams, R., Improved survival following injection sclerotherapy for esophageal varices. Final analysis of a controlled trial, Hepatology 5, 827-830, 1985

25. Terblanche, J., The long-term management of patients after an oesophageal variceal bleed: the role of sclerotherapy, Brit.J.Surg. 72, 88-90, 1985.

26. Sarin, S.K., Sachdev, G., and Nanda, R., Follow-up of patients after variceal eradication. A comparison of patients with cirrhosis, noncirrhotic portal fibrosis, and extrahepatic obstruction, Ann.Surg. 204, 78-82, 1986

27. Westaby, D. and Williams, R., Elective sclerotherapy—techniques and results, Endoscopy 18(Suppl 2), 28-31, 1986.

28. Westaby, D., Long-term injection sclerotherapy for the primary and secondary management of variceal haemorrhage, Z.Gastroenterol. 26(Suppl 2), 36-39, 1988.

29. Terblanche, J., The use of sclerotherapy for the management of oesophageal varices in portal hypertension, Surg.Endosc. 2, 149-155, 1988.

30. Van Stiegmann, G., Techniques for endoscopic obliteration of esophageal varices, Surg.Annu. 23, 175-202, 1991.

31. Terblanche, J., Bornman, P.C., and Kirsch, R.E., Sclerotherapy for bleeding esophageal varices, Annu.Rev.Med. 35, 83-94, 1984.

32. Tripathi, D., Ferguson, J.W., Therapondos, G., et al., Review article: recent advances in the management of bleeding gastic varices, Aliment.Pharmacol.Ther. 24, 1-17, 2006.

33. Kind, R., Gugliemi, A., Rodella, L., et al., Bucrylate treatment of bleeding gastric varices; 12 years experience, Endoscopy 32, 512-519, 2000.

34. Petersen, B., Barkun, A., Carpenter, S., et al.,(ASGE Technology Assessment Committee), Tissue adhesives and fibrin glues, Gastrointestinal Endoscopy 60, 327-333, 2004.

35. Gostout, C.J., What we need is a reliable plug, Am.J.Gastroenterol. 97, 1281-1283, 2002.

36. Heneghan, M.A., Byrne, A., and Harrison, P.M., An open pilot study of the effects of a human fibrin glue for endoscopic treatment of patients with acute bleeding from gastric varices, Gastrointestinal Endoscopy 56, 422-426, 2002.

37. Datta, D., Vlavianos, P., Alisa, A., and Westaby, D., Use of fibrin glue (Beriplast) in the management of bleeding gastric varices, Endoscopy 35, 675-678, 2003.

38. Binmoeller, K.F., Glue for gastric varices: some sticky issues, Gastrointest.Endosc. 52, 298-300, 2000.

39. Kok, K., Bond, R.P., Duncan, I.C., et al., Distal embolization and local wall ulceration after gastric variceal obliteration with N-butryl-2-cyanoacrylate: a case report and review of the literature, Endoscopy 36, 442-446, 2004.

40. Wahl, P., Lammer, F., Conen, D., et al., Septic complications after injection of N-butyl-2-cyanoacrylate: report of two cases and review, Gastrointes.Endosc. 61, 911-916, 2004

41. Sato, T., Yamazaki, K., Toyota, J., et al., Inflammatory tumor in pancreatic trail induced by endoscopic ablation with cyanoacrylate glue for gastric varices, J.Gastroenterol. 39, 475-478, 2004.

42. Williams, S.G.J., Peters, R.A., and Westaby, D., Thrombin – an effective treatment for gastric variceal hemorrhage, Gut 35, 1287-1289, 1994.

43. Yang, W.Y., Tripathi, D., Therapondos, G., et al., Endoscopic use of human thrombin in bleeding gastric varices, Amer.J.Gastroenterol. 97, 1381-1385, 2002.

44. Kojima, K., Imazu, H., Matsumara, M., et al., Sclerotherapy for gastric fundal variceal bleeding: Is complete obliteration possible without cyanoacrylate?, J.Gastroenterology Hepatology 20, 1701-1706, 2005.

45. Lunderquist, A. and Vang, J., Sclerosing injection of esophageal varices through transhepatic selective catheterization of the gastric coronary vein, Acta Radiologica Scand. 15, 546-550, 1974

46. Lunderquist, A. and Vang, J., Transhepatic catheterization and obliteration of the coronary vein in patients with portal hypertension and esophageal varices, New Engl.J.Med. 291, 646-649, 1974.

47. Sivak, M.V., Jr., Stout, D.J. and Skipper, G., Endoscopic injection sclerosis (EIS) of esophageal varices, Gastrointestinal Endoscopy 27, 52-57, 1981.

48. Hedberg, S.E., Fowler, D.L., and Ryan, R.L.R., Injection sclerotherapy of esophageal varices using ethanolamine oleate. A pilot study, Amer.J.Surg. 143, 426-431, 1982.

49. Lyons, S.D., Sugawa, C.,Geller, I.R., and Vandenberg, D.M., Comparison of 1% sodium tetradecyl sulfate to thrombogenic sclerosant cocktail for endoscopic sclerotherapy, Am.Surg. 54, 81-84, 1988.

50. Nakamura, R., Bucci, L.A., Sugawa, C., et al., Sclerotherapy of bleeding esophageal varices using a thombogenic cocktail, Am.Surg. 57, 226-230, 1991.

51. Scott, J., Dick, R., Lang, R.G., and Sherlock, S., Percutaneous transhepatic obliteration of gastro-oesophageal varices, Lancet II, 53-55, 1976.

52. Ansell, J.E., Widrich, W.C., Johnson, W.C., et al., Gelfoam and autologous clot embolization: Effect on coagulation, Invest.Radiol. 13, 115-120, 1978.

33. Westaby, D., Indications and results of variceal injection sclerotherapy, in Gastroenterologic Endoscopy, 2nd edn., ed. MV. Sivak, Jr., W.B. Saunders, Philadelphia, PA, Chapter 33,pps. 389-402, 2000.

54. Gardner, E.C. and Brooks, W.S., Absence of disseminated intravascular coagulation with encloscopic sclerosis of esophageal varices, Gastroenterol.Endoscopy 28, 67-69, 1982.

55. Lundblad, R.L., Bradshaw, R.A., Gabriel, D., et al., A review of the therapeutic uses of thrombin, Thromb.Haemostas. 91, 851-860, 2004.

56. de Franchis, R., Cipolla, M., Primignani, M., et al., Activation of coagulation in cirrhotics after endoscopic variceal schlerotherapy, Amer.J.Gastroenterol. 82, 1287-1291, 1987.

57. Fujii, Y., Sugawa, C., Ozawa, T., et al., Hemostasis activation during esophageal variceal sclerotherapy with thrombin in cirrhotics, Amer.Surg. 57, 222-225, 1991.

58. Koyama, T., Kakishita, E., Nakai, Y., and Okamoto, E., Significance of hemostatic molecular markers during disseminated intravascular coagulation in patients with liver cirrhosis treated by endoscopic embolization for esophageal varices, Amer.J.Hematol. 38, 90-94, 1991.

59. Biagi, G., Grauso, F., Bernardi, M. et al., Platelet and blood clotting activation after endoscopic sclerotherapy of oesophageal varices in liver cirrhosis, Thromb.Res. 44, 411-416, 1986.

61. Kitona, S., Hashizume, M., Yamaga, H., et al., Human thrombin plus 5 percent ethanolamine oleate injected to sclerose oesophageal varices: a prospective, randomized trial, Brit.J.Surg. 78, 715-718, 1989.

62.Kruger, K., Zahringer, H., Sohngen, F.D., et al., Femoral pseudoaneurysms: management with percutaneous thrombin injection—success rates and effects on systemic coagulation, Radiology 226, 452-458, 2003

63. Kruger, K., Zachringer, M., Strohe, D., et al., Postcatherization pseudoaneurysm: Results of US-guided percutaneous thrombin injection in 240 patients, Radiology 236, 1104-1110, 2005.

64. Olsen, D.M., Rodriguez, J.A., Vrana, M., et al., A prospectus of ultrasound scan-guided thrombin injection of femoral pseudoaneurysm: a trend toward minimal medication, J.Vasc.Surg. 36, 779-782, 2002.

65. Bosch, J., Thabut, D., Bendtsen, F., et al., Recombinant factor VIIa for upper gastrointestinal bleeding in patients with cirrhosis: A randomized, double-blind trial, Gastroenterology 127, 1123-1130, 2004.

66. Akyildiz, M., Turan, I., Ozutemiz, O., et al., A cerebrovascular event after single-dose administration of recombinant factor VIIa in a patient with esophageal variceal bleeding, Dig.Dis.Sci. 51, 1647-1649, 2006.

67. Cushieri, A., Tissue adhesives in endosurgery, Semin.Laperosc.Surg. 8, 63-68, 2000.

68. Reece, T.B., Maxey T.S., and Kran, I.L., A prospectus on tissue adhesives, Amer. .J.Surg. 182, 405-445, 2001.

69. Gosain, A.R. and Lyon, V.B.(Plastic Surgery Education Foundation DATA Committee), The current status of tissue glues. Part II., For adhesion of soft tissues, Plast.Reconstruct.Surg. 110, 1581-1584, 2002.