GI bleeding risk in patients undergoing dialysis
Article Outline
All patients with end-stage kidney disease with peptic ulcer bleeding who were undergoing dialysis should be managed as high-risk patients.
Early autopsy studies of patients who died of uremia revealed that GI abnormalities were common,1 and edema, congestion, or ulcerations with hemorrhage were seen in two thirds of such patients. In addition, abnormal GI bleeding is a common and potentially serious complication of renal failure, and although dialysis has helped to reduce the incidence of severe hemorrhage, the high risk of morbidity and mortality associated with abnormal GI bleeding persists.
The sources of upper GI hemorrhage in patients with uremia are similar to those in patients with normal renal function2 and include gastric and duodenal ulcers, vascular ectasia, gastric erosions, esophagitis, Mallory-Weiss tears, and gastric cancer. Gastroduodenal ulcers constitute the most common lesions viewed on endoscopy, and they are responsible for nearly 60% of major upper GI bleeding episodes in patients with chronic renal failure.
The true incidence of upper GI hemorrhage in uremic patients undergoing maintenance hemodialysis is not known. A retrospective, Canadian study by Holden et al3 of patients undergoing long-term outpatient hemodialysis reported a 2.5% per person-year incidence rate of major bleeding episodes. In this study, the overwhelming majority of bleeding episodes originated from the GI tract. In such patients, GI bleeding may be the result of a uremic milieu, but it also can be related to causative disorders (eg, vasculitis with GI mucosal ulcerations), concurrent medication use (eg, aspirin, warfarin), or systemic anticoagulation with heparin during dialysis.
In this issue of Gastrointestinal Endoscopy, Cheung et al4 review the outcomes of peptic ulcer disease in patients with end-stage renal disease and compared them with rebleeding in patients with moderate, stage 3, chronic kidney disease, and patients with normal kidney function. All patients in this study received acid suppression treatment. Endoscopic therapy in the form of injection or bipolar cautery or endoclip placement (or combination) was performed within 24 hours for patients with endoscopic high-risk stigmata. The overall recurrent bleeding rates were significantly higher for the end-stage renal disease group compared with the moderate chronic kidney disease group and the normal kidney function group (38% vs 14% vs. 12%, respectively). More interestingly, rebleeding rates were significantly higher in the end-stage kidney disease group even when these patients had low-risk ulcer stigmata on endoscopy. The authors concluded that all patients with end-stage kidney disease with peptic ulcer bleeding who were undergoing dialysis should be managed as high-risk patients.
Although the pathogenesis of excessive bleeding in patients with end-stage renal disease is multifactorial,5 a defect of primary hemostasis caused by platelet dysfunction in the form of impaired platelet adhesiveness and altered platelet-vessel–wall interaction is thought to play a major role. Moderate thrombocytopenia is also seen frequently in uremic patients, although the thrombocytopenia is not severe enough to cause bleeding. Dialysis removes from the vascular compartment several toxins that contribute to impaired platelet function and partially corrects, but does not eliminate, platelet function abnormalities. The adequacy of hemodialysis should be confirmed in patients with major bleeding complications.
Yet, hemodialysis itself is known to contribute to increased bleeding. Recurrent and prolonged exposure of blood to the artificial surface of the dialyzer membrane and blood tubing may induce chronic activation of platelets, leading to platelet exhaustion and dysfunction. In contrast, peritoneal dialysis is not associated with exposure to the artificial membrane and can be considered in patients at high risk for bleeding. In addition, peritoneal dialysis does not require the use of systemic anticoagulation.
There are a number of agents that can be used to achieve hemostasis in patients with renal insufficiency when there is active GI bleeding. Cryoprecipitate can cause resolution of bleeding in uremic patients, but the result is short-lived and not effective in 50% of patients. Hence, it is usually used only for life-threatening GI hemorrhage. Desmopressin acetate (1-deamino-8-D-arginine-vasopressin), a synthetic derivative of antidiuretic hormone, is a possible therapeutic alternative that acts by stimulating the release of high-molecular-weight multimers of von Willebrand factor from endothelial cells. After its administration in patients with renal failure, bleeding time improves in 1 hour, and its effect lasts for up to 8 hours. However, repeated administration of desmopressin acetate results in tachyphylaxis. There are also anecdotal observations of diminished GI bleeding in uremic patients treated with conjugated estrogens, and such therapy is especially useful in patients who need sustained control of bleeding, because its effectiveness lasts for up to 2 weeks.
Tranexamic acid, a potent inhibitor of the fibrinolytic system, was studied as an adjunctive therapy to standard management in patients with GI bleeding. It stabilizes clots by preventing the binding of plasminogen to fibrin and the activation of plasminogen to plasmin. It was effective in controlling chronic bleeding from colonic angiodysplasias in dialysis patients.6 In a small study, tranexamic acid use resulted in reduced rebleeding rates in dialysis patients with upper GI bleeding.7 However, because tranexamic acid accumulates in renal failure, it should be considered only in acute settings when satisfactory responses have not been obtained with other treatments.
Systemic anticoagulation with heparin is a common practice during hemodialysis treatments and is used to prevent clotting of the extracorporeal circuit. Heparin can also induce thrombocytopenia by an immune mechanism. Multiple strategies to minimize systemic anticoagulation have been used in patients with an increased bleeding risk.5 These include hemodialysis without heparin, the use of low-dose or low-molecular-weight heparin, regional anticoagulation with heparin and protamine, or regional anticoagulation with citrate. All of these methods may reduce the bleeding risk, albeit with limitations and complications.
Anemia is a common finding in patients undergoing dialysis, and anemia potentiates hemorrhagic diathesis by affecting platelet function. Circulating red blood cells facilitate the interaction of platelets with the subendothelium at the area of injury by displacing platelets toward the vessel wall, away from the axial flow. Correction of anemia with erythropoietin or red blood cell transfusions improves bleeding time in uremic patients.
Several drugs increase the GI bleeding risk in patients with renal failure. Beta-lactam antibiotics and third-generation cephalosporins accumulate in advanced renal failure and may alter platelet membrane function by interfering with the adenosine diphosphate receptor. Aspirin at moderate doses and nonsteroidal anti-inflammatory drugs, such as ibuprofen, naproxen, indomethacin, phenylbutazone, and sulfinpyrazone inhibit platelet cyclooxygenase and disturb platelet function. In the Canadian study by Holden et al3 of patients undergoing long-term outpatient hemodialysis, the frequency of major bleeding events increased with the use of aspirin. In dialysis patients with active bleeding, the use of such medications should be avoided.
One mechanism that may contribute to the increased risk of GI bleeding in patients on hemodialysis is inadvertent systemic anticoagulation resulting from leaking of heparin used as a locking solution for dialysis catheters. Pepper et al8 showed that a significant leak of heparin from dialysis catheters occurs immediately after a catheter lock procedure is performed, as demonstrated by a prolonged activated partial thromboplastin time. In the same study, the systemic anticoagulation was evident for up to 3 hours after the catheter lock with heparin was placed. In vitro studies have shown that more than 50% of the catheter lock volume of heparin leaked out of the catheter. Yevzlin et al,9 in a small, retrospective study, showed that use of a catheter lock with concentrated heparin (5000 units/mL) was associated with increased bleeding complications as compared with use of low-dose heparin (1000 units/mL) or citrate locks. In the study by Cheung et al,4 the authors indicate that heparin was not used during hemodialysis treatments in their patients with peptic ulcer bleeding. However, the rate of dialysis catheter use or information regarding type of solution used for catheter locks in their patients has not been provided. According to a U.S. renal data system annual report from 2005, central venous catheters were used for hemodialysis in 60% of patients undergoing incident hemodialysis and 30% of patients undergoing prevalent hemodialysis.
Patients with end-stage renal disease develop hemostatic disorders mainly in the form of bleeding diathesis. Treatment goals in such patients are twofold: prevention of bleeding in high-risk patients before invasive procedures or surgery and prompt control of active bleeding. The strategy used depends on the severity of bleeding and its response to previous therapy.
Disclosure
The author disclosed no financial relationships relevant to this publication.
References
- . The gastrointestinal tract in uremia. Dig Dis Sci. 1993;38:257–268
- . Upper gastrointestinal bleeding in patients with chronic renal failure: role of vascular ectasia. Am J Gastroenterol. 1996;91:2329–2332
- Major bleeding in hemodialysis patients. Clin J Am Soc Nephrol. 2008;3:105–110
- Peptic ulcer bleeding outcomes adversely affected by endstage renal disease. Gastrointest Endosc. 2010;71:44–49
- . Treatment of bleeding in dialysis patients. Semin Dial. 2009;22:279–286
- . Successful treatment of bleeding from colonic angiodysplasias with tranexamic acid in hemodialysis patient. Am J Kidney Dis. 1998;31:536–538
- . Tranexamic acid is beneficial as adjunctive therapy in treating major upper gastrointestinal bleeding in dialysis patients. Nephrol Dial Transplant. 2003;18:1388–1391
- Inadvertent postdialysis anticoagulation due to heparin line locks. Hemodial Int. 2007;11:430–434
- Concentrated heparin lock is associated with major bleeding complications after tunneled hemodialysis catheter placement. Semin Dial. 2007;20:351–354
PII: S0016-5107(09)02441-9
doi:10.1016/j.gie.2009.09.005
© 2010 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.
