A Korean Multi-Center Survey about Warfarin Management before Gastroenterological Endoscopy in Patients with a History of Mechanical Valve Replacement Surgery.

BACKGROUND: Guidelines for esophagogastroduodenoscopy (EGD) in the West allow the continued use of warfarin under therapeutic international normalized ratio (INR) level. In Korea, no guidelines have been issued regarding warfarin treatment before EGD. The authors surveyed Korean cardiac surgeons about how Korean cardiac surgeons handle warfarin therapy before EGD using a questionnaire. Participants were requested to make decisions regarding the continuation of warfarin therapy in two hypothetical cases.
METHODS: The questionnaire was administered to cardiac surgeons and consisted of eight questions, including two case scenarios.
RESULTS: Thirty-six cardiac surgeons at 28 hospitals participated in the survey, and 52.7% of the participants chose to stop warfarin before EGD in aortic valve replacement patients without risk factors for thromboembolism. When the patient's INR level was 2, 31% of the participants indicated that they would choose to continue warfarin therapy. For EGD with biopsy, 72.2% of the participants chose warfarin withdrawal, and 25% of the participants chose heparin replacement. In mitral valve replacement patients, 47.2% of the participants chose to discontinue warfarin, and 22.2% of the participants chose heparin replacement. For EGD with biopsy in patients with a mitral valve replacement, 58.3% of the participants chose to stop warfarin, and 41.7% of the participants chose heparin replacement.
CONCLUSION: This study demonstrated that attitudes regarding warfarin treatment for EGD are very different among Korean surgeons. Guidelines specific to the Korean population are required.

Introduction

During esophagogastroduodenoscopy (EGD) examinations and/or the treatment of patients on antithrombotic therapy, it is necessary to balance the risk of hemorrhage induced by antithrombotic therapy with the risk of thromboembolism after antithrombotic drug withdrawal [1,2]. The risk of hemorrhage depends on the nature of the endoscopic procedure, and the risk of thromboembolism depends on comorbidities [2]. Guidelines for EGD in the USA, UK, and Europe allow the continued use of warfarin at the time of an EGD examination, including examinations involving a mucosal biopsy, because the risk of hemorrhage induced by diagnostic EGD with mucosal biopsy is low [3-6]. It is believed, without solid evidence, that Caucasians are more susceptible to thromboembolism than Asians [7]. In 2005, the Japan Gastroenterological Endoscopy Society (JGES) recommended three or four days of withdrawal from warfarin, three days from aspirin, and five days from ticlopidine, even for low-risk endoscopic procedures requiring biopsy [2,8]. However, the JGES guidelines were changed in 2014 and no longer recommended warfarin discontinuation for diagnostic EGD [2]. Furthermore, warfarin discontinuation is not required for EGD with a mucosal biopsy if the international normalized ratio (INR) is in the therapeutic range [2]. In Korea, no ethnicity-specific warfarin management guidelines have been issued for EGD. This multicenter questionnaire-based survey was performed to evaluate current practice with respect to warfarin management for EGD in mechanical valve replacement patients.

Methods

The questionnaire was administered to cardiac surgeons (all board-certified) during the 2015 Korean Society for Thoracic and Cardiovascular Surgery annual meeting. It consisted of eight questions. Two questions gathered general information (participant age and affiliation), four addressed each surgeon’s personal judgment regarding warfarin treatment before EGD on the basis of two hypothetical cases, and two requested opinions about ethnic differences regarding bleeding and the risk of thromboembolism. The full questionnaire is provided in Appendix 1.

Results

Thirty-six cardiac surgeons from 28 hospitals participated in the survey. Eleven of the hospitals were in Seoul, 10 were in Incheon and Gyeonggi-do, three were in Busan and Gyeongsangnam-do, four were in Daegu and Gyeongsangbuk-do, one was in Daejeon and Chungcheongnam-do, and two were in Gwangju and Jeollanam-do. Five of the participants were in their thirties, 21 were in their forties, eight were in their fifties, and two were over sixty.

We asked surgeons to indicate their policies regarding warfarin treatment before EGD conducted during annual health check-ups (Fig. 1). The first case presented in the questionnaire (case 1) was a 60-year-old male on warfarin with an INR level of 2 who had undergone aortic valve replacement with a mechanical valve. Nineteen of the 36 surgeons responded that they would stop warfarin and not replace it with another medication for two to five days (median, three days) prior to EGD. Four stated that they would replace warfarin with subcutaneous enoxaparin, and two that they would replace warfarin with intravenous heparin. Eleven surgeons responded that they would perform EGD without discontinuing warfarin.

Fig. 1

Policy before EGD for annual health check-up. Policies regarding warfarin discontinuation before an EGD for an annual health check-up in an AVR patient with no risk factors for thromboembolism. Case: 60-year-old male, AVR (mechanical), received warfarin, present international normalized ratio of 2, no risk factors of thromboembolism. EGD, gogastroduodenoscopy; AVR, aortic valve replacement.

Moreover, if EGD with a biopsy was indicated for the patient in case 1, 26 surgeons answered that they would stop warfarin without a replacement for two to seven days (median, three days) prior to EGD (Fig. 2), seven that they would replace warfarin with subcutaneous enoxaparin, and two that they would replace warfarin with intravenous heparin. One surgeon stated that EGD would be performed without discontinuing warfarin.

Fig. 2

Policy before EGD with biopsy. Policies regarding warfarin discontinuation before EGD with a biopsy for an AVR patient with no risk factors for thromboembolism. Case: 60-year-old male, AVR (mechanical), received warfarin, present international normalized ratio of 2, no risk factors of thromboembolism. EGD, esophagogastroduodenoscopy; AVR, aortic valve replacement.

Case 2 involved a 60-year-old male on warfarin with an INR level of 2.5–3 who had undergone mitral valve replacement with a mechanical valve. Before EGD for an annual health check-up, 17 surgeons answered that they would stop warfarin for two to give days (median, three days) prior to EGD (Fig. 3), five that they would replace warfarin with subcutaneous enoxaparin, and three that they would replace warfarin with intravenous heparin. Eleven surgeons stated that they would perform EGD without discontinuing warfarin. If the patient in case 2 underwent EGD with a biopsy, 21 surgeons responded that they would stop warfarin without a replacement for two to seven days (median, three days) prior to EGD (Fig. 4), 10 responded that they would replace warfarin with subcutaneous enoxaparin, and five responded that they would replace warfarin with intravenous heparin. No surgeon responded that EGD would be performed without discontinuing warfarin.

Fig. 3

Policy before EGD for annual health check-up. Policies regarding warfarin discontinuation before EGD in MVR patients with no risk factors for thromboembolism. Case: 60-year-old male, MVR (mechanical), received warfarin, present international normalized ratio of 2.5–3, no risk factors of thromboembolism. EGD, esophagogastroduodenoscopy; MVR, mitral valve replacement.

Fig. 4

Policy before EGD with biopsy. Policies regarding warfarin discontinuation before EGD with a biopsy in MVR patients with no risk factors for thromboembolism. Case: 60-year-old male, MVR (mechanical), received warfarin, present international normalized ratio of 2.5–3, no risk factors of thromboembolism. EGD, esophagogastroduodenoscopy; MVR, mitral valve replacement.

Regarding the relative risks of thromboembolism after warfarin withdrawal in Koreans and Caucasians, two surgeons believed that the risk was higher in Koreans, eight believed that the risk was the same, and 26 thought that the risk was lower in Koreans (Fig. 5). Regarding the relative risks of bleeding during EGD in Koreans and Caucasians without discontinuing warfarin at an INR level of 2–3, 18 surgeons believed the risk was higher in Koreans, 17 believed the risk was the same, and one considered the risk to be lower in Koreans.

Fig. 5

Cardiac surgeons’ opinions regarding the risks of thromboembolism and bleeding during esophagogastroduodenoscopy without warfarin discontinuation. (A) Risk of thromboembolism. (B) Risk of bleeding.

Discussion

The suspension of warfarin therapy causes reversion to a state of hypercoagulability [2], and it has been reported that in 1% of cases, warfarin withdrawal results in thromboembolic complications and poor prognoses [2, 9–11]. In addition to drug withdrawal, the risk of thromboembolism may be increased by dehydration caused by preparation for endoscopic examination, and this risk should be addressed by adequate fluid replacement [2]. In the 2014 JGES guidelines, the conditions with a high risk of thromboembolism associated with discontinuing anticoagulant therapy are a history of cardiogenic brain embolism, mechanical mitral valve replacement, thromboembolism following mechanical valve replacement, atrial fibrillation accompanying valvular heart disease, atrial fibrillation without valvular heart disease but with a high risk of stroke, anti-phospholipid antibody syndrome, and deep vein thrombosis/pulmonary thromboembolism [2]. However, the guidelines state that the risk of thromboembolism associated with anticoagulant discontinuation varies considerably [2]. Once thromboembolic complications have occurred, they are often serious, and all patients on anticoagulant therapy should therefore be treated as high-risk patients [2].

The 2014 JGES guidelines classify gastroenterological endoscopic procedures into the following four categories: diagnostic gastroenterological endoscopy without biopsy, endoscopic mucosal biopsy, gastroenterological endoscopy with a low risk of bleeding, and gastroenterological endoscopy with a high risk of bleeding [2]. The incidence of bleeding complications after mucosal biopsy is 0.002% in the stomach and 0.09% in the large intestine, regardless of antithrombotic use [12,13]. Several studies, even in Japan, have shown that the bleeding risk after biopsy was not increased by antithrombotics [14,15]. Furthermore, since antithrombotics do not increase the risk of bleeding after diagnostic EGD including a mucosal biopsy, the 2014 JGES guidelines indicate that discontinuing warfarin is not required if the INR is under the therapeutic range [2].

In the present study, 52.7% of the participants responded that they would discontinue warfarin before EGD in the aortic valve replacement patient (case 1) without risk factors for thromboembolism, while 31% answered that they would maintain warfarin therapy with at an INR level of 2. If EGD with a biopsy was to be performed, 72.2% stated that they would stop warfarin, and 25% stated they would replace warfarin with subcutaneous enoxaparin or intravenous heparin. In case 2, the mitral valve replacement patient, 47.2% of participants chose to discontinue warfarin, and 22.2% chose to replace warfarin with intravenous heparin or subcutaneous enoxaparin. These results demonstrate that the percentage of participants who chose heparin replacement was higher for mitral valve replacement patients than for aortic valve replacement patients. In case 2, if EGD was to be performed with a biopsy, 58.3% of participants chose to discontinue warfarin without a replacement and 41.7% chose to replace warfarin with heparin. These findings indicate that Korean cardiac surgeons are prepared to stop warfarin, with varying attitudes towards heparin replacement, before EGD with or without a biopsy, which is not in line with Western or Japanese guidelines.

One study explored differences between the management practices of Korean and Western endoscopists regarding patients on anticoagulation or antiplatelet medications [16]. The authors found that, although Korean endoscopists appeared to be concerned about the risk of bleeding, they did not seem to be as concerned about thromboembolism, which may have been due to the belief held by Korean endoscopists that Caucasians are more susceptible than Asians to thromboembolism [16].

In the present study, most Korean cardiac surgeons (72%) indicated that the risk of thromboembolism after warfarin withdrawal is lower in Koreans than Caucasians. Half of the participants (50%) believed the risk of bleeding during endoscopy without warfarin withdrawal at an INR of 2–3 to be higher in Koreans, while 47.2% believed the risk to be equivalent in Koreans and Caucasians.

Although some studies have supported the proposal that Caucasians are more susceptible than Asians to thromboembolism [17], the risk of thromboembolism in Asians may be substantial [16], and the cessation of warfarin thus requires careful consideration. Furthermore, the mortality rate associated with embolism is greater than that associated with bleeding [17]. Asians may be predisposed to severe forms of embolism, such as cerebrovascular events [16,18]. Thromboembolism in Caucasians may be more likely to be of the cardiovascular variety, which has less severe clinical consequences than central nervous system thromboembolism, whereas the cerebrovascular system is more likely to be involved in Asians [16,18]. Indeed, a Korean survey indicated that six of 81 endoscopists (7.4%) have experienced a case of embolism in a patient after discontinuing anticoagulation and antiplatelet medications for an endoscopic procedure, and five of these six were cerebrovascular embolisms [19].

Guidelines are useful as references, but the risks of hemorrhage and thromboembolism after withdrawal should be assessed on a case-by-case basis. The 2014 JGES guidelines recommend that when anticoagulants need to be discontinued before gastroenterological endoscopy, the prescribing doctor should be consulted beforehand, because it is important to develop a management plan optimized for the individual based on a consultation between the endoscopist and the physician responsible for prescribing antithrombotics [2]. In Korea, no guidelines have been issued regarding warfarin management before EGD.

This study has several limitations. First, the number of participants was small, and its methodology involved a questionnaire based on hypothetical clinical cases. Nonetheless, this study revealed that preferences regarding warfarin treatment at the time of EGD vary considerably among Korean surgeons. This is a preliminary study to confirm the necessity of developing new guidelines. Further collaborative studies between cardiac surgeons and endoscopists are necessary to develop new guidelines that take into account ethnic differences and current practice.