Digestive Endoscopy 2014; 26: 424–429
doi: 10.1111/den.12208
Original Article
Antithrombotic drug does not affect the positive predictive value of an immunochemical fecal occult blood test Yosuke Tsuji,1,3 Toshiaki Gunji,4 Hajime Sato,5 Akiko Ono,3 Takafumi Ito,3 Ken Ohata,3 Nobutake Yamamichi,1 Mitsuhiro Fujishiro,1,2 Nobuyuki Matsuhashi3 and Kazuhiko Koike1 Departments of 1Gastroenterology and 2Endoscopy and Endoscopic Surgery, Graduate School of Medicine, The University of Tokyo, 3Department of Gastroenterology, 4Center for Preventive Medicine, NTT Medical Center Tokyo, Tokyo, and 5Department of Health Policy and Technology Assessment, National Institute of Public Health, Wako, Japan Background and Aim: Antithrombotic drugs may affect the diagnostic performance of immunochemical fecal occult blood test (iFOBT) for colorectal cancer (CRC) screening. The aim of the present study was to assess the effect of antithrombotic drugs on the diagnostic performance of iFOBT. Methods: We analyzed 1016 patients who underwent colonoscopy for positive iFOBT. Patients were classified as follows: patients who had advanced neoplasms detected and those who did not; patients who had cancers detected and those who did not; patients who had any neoplasms detected and those who did not. We compared the following factors between two paired groups: sex, age, endoscopists’ experience, and antithrombotic drug usage. Results: A total of 139 patients were taking antithrombo-
Conclusion: The present study demonstrated that the positive predictive value of iFOBT was not affected by ongoing antithrombotic therapy.
tic drugs (13.7%). Advanced neoplasms, cancers, and any neoplasms were detected in 196 (19.3%), 59 (5.8%), and 490 (48.2%)
Key words: antithrombotic agent, cancer screening, colonic neoplasm, colonoscopy, false-positive reaction
INTRODUCTION
based FOBT, but immunochemical FOBT (iFOBT) was later developed as a better screening test6 and a CRC screening program using iFOBT was incorporated into public health policy in Japan in 1992.7 Antithrombotic drugs, such as aspirin or warfarin, are widely used for prophylaxis of cardiocerebrovascular diseases, but their use is associated with a higher incidence of both occult and overt gastrointestinal bleeding.8 Therefore, it might be inferred that antithrombotic drugs affect the diagnostic performance of FOBT. However, cessation of antithrombotic drugs might cause serious thrombotic events. Therefore, if antithrombotic drugs do not affect the diagnostic performance of iFOBT, it is desirable that they should be continued. Concerning this issue, there are scant and controversial data.9–18 Moreover, most of these studies were conducted with guaiac-based FOBT.9,10,12,15–18 Levi et al. investigated the effect of aspirin, non-steroidal antiinflammatory drugs and anticoagulants on the performance of
I
N JAPAN, THE incidence of colorectal cancer (CRC) is very high1 and CRC is a major cause of cancer death. Among all cancers, the age-standardized ratio of CRC mortality in Japan is the highest in females and the third highest in males.2 Randomized control studies have revealed that populationbased screening for CRC using a fecal occult blood test (FOBT) is effective.3–5 These data were based on guaiac-
Corresponding: Nobuyuki Matsuhashi, Department of Gastroenterology, NTT Medical Center Tokyo, 5-9-22 Higashi-Gotanda, Shinagawa-ku, Tokyo 141-8625, Japan. Email: nmatuha-tky @umin.ac.jp The contents of this paper were presented, in part, at the 20th United European Gastroenterology Week Meeting, 12–16 October, 2012, Amsterdam, The Netherlands. Received 29 July 2013; accepted 15 October 2013.
424 bs_bs_banner
patients, respectively. There were no higher detection rates in the antithrombotic drug (–) group than in the (+) group (advanced neoplasms: 19.3% vs 19.4%, P = 1.000; cancers: 5.8% vs 5.8%, P = 1.000; any neoplasms: 48.4% vs 47.5%, P = 0.856). Multivariate logistic regression analysis revealed that none of aspirin, warfarin, or other antithrombotic drugs was a significant factor for advanced neoplasms (95% CI 0.350–1.216, P = 0.179; 95% CI 0.421–1.899, P = 0.772; 95% CI 0.323–1.810, P = 0.764, respectively). As to cancers and any neoplasms, no antithrombotic drug also proved to be a significant factor.
© 2013 The Authors Digestive Endoscopy © 2013 Japan Gastroenterological Endoscopy Society
Digestive Endoscopy 2014; 26: 424–429
Antithrombotic drugs and iFOBT 425
iFOBT, but that study was not carried out in a screening setting.14 Hence, as to iFOBT, there are still few data on the effect of antithrombotic drugs. The aim of the present study was to assess the effect of antithrombotic drugs on the diagnostic performance of iFOBT.
METHODS Study protocol
T
HIS STUDY PROTOCOL was approved by the Institutional Review Board of the NTT Medical Center Tokyo, and written informed consent for colonoscopy examination from the cecum to the rectum (TCS) was obtained from all patients before the original procedures were conducted. The study was carried out in accordance with the Declaration of Helsinki. This study was registered in the University Hospital Medical Network Clinical Trials Registry (UMIN000010100).
Patients Electronic medical charts of all patients who underwent TCS for positive iFOBT at NTT Medical Center Tokyo in 2011 were reviewed. In Japan, national colorectal cancer screening using iFOBT has been incorporated into public health policy since 1992.7 The recommended screening program is as follows. Screening by health facilities in residential areas is offered to persons aged 40 years or older. A 2-day iFOBT is conducted, and if one or two samples show positive results, a detailed examination is recommended. TCS is recommended as the diagnostic examination, but, if impossible, sigmoidoscopy in combination with a double-contrast barium study is permitted.7 Exclusion criteria in the present study were as follows: inflammatory bowel disease, colectomy history, and incomplete colonoscopy.
Data collection Drug usage information was obtained from electronic medical charts. Before carrying out TCS, it was necessary to know whether the patients were regularly taking antithrombotic drugs in order to avoid polypectomy during antithrombotic treatment. Therefore, doctors ordering TCS checked the antithrombotic medication status and recorded it in the electronic medical chart. Patients were categorized as antithrombotic drug takers based on the medication status within 30 days before iFOBT. Aspirin, clopidogrel, ticlopidine, warfarin, cilostazol and dabigatran were defined as antithrombotic drugs. We also searched each endoscopist’s experience of TCS, and defined examiners with ≥1000 cases of TCS as ‘experienced endoscopists’, and those with
doi: 10.1111/den.12208
Original Article
Antithrombotic drug does not affect the positive predictive value of an immunochemical fecal occult blood test Yosuke Tsuji,1,3 Toshiaki Gunji,4 Hajime Sato,5 Akiko Ono,3 Takafumi Ito,3 Ken Ohata,3 Nobutake Yamamichi,1 Mitsuhiro Fujishiro,1,2 Nobuyuki Matsuhashi3 and Kazuhiko Koike1 Departments of 1Gastroenterology and 2Endoscopy and Endoscopic Surgery, Graduate School of Medicine, The University of Tokyo, 3Department of Gastroenterology, 4Center for Preventive Medicine, NTT Medical Center Tokyo, Tokyo, and 5Department of Health Policy and Technology Assessment, National Institute of Public Health, Wako, Japan Background and Aim: Antithrombotic drugs may affect the diagnostic performance of immunochemical fecal occult blood test (iFOBT) for colorectal cancer (CRC) screening. The aim of the present study was to assess the effect of antithrombotic drugs on the diagnostic performance of iFOBT. Methods: We analyzed 1016 patients who underwent colonoscopy for positive iFOBT. Patients were classified as follows: patients who had advanced neoplasms detected and those who did not; patients who had cancers detected and those who did not; patients who had any neoplasms detected and those who did not. We compared the following factors between two paired groups: sex, age, endoscopists’ experience, and antithrombotic drug usage. Results: A total of 139 patients were taking antithrombo-
Conclusion: The present study demonstrated that the positive predictive value of iFOBT was not affected by ongoing antithrombotic therapy.
tic drugs (13.7%). Advanced neoplasms, cancers, and any neoplasms were detected in 196 (19.3%), 59 (5.8%), and 490 (48.2%)
Key words: antithrombotic agent, cancer screening, colonic neoplasm, colonoscopy, false-positive reaction
INTRODUCTION
based FOBT, but immunochemical FOBT (iFOBT) was later developed as a better screening test6 and a CRC screening program using iFOBT was incorporated into public health policy in Japan in 1992.7 Antithrombotic drugs, such as aspirin or warfarin, are widely used for prophylaxis of cardiocerebrovascular diseases, but their use is associated with a higher incidence of both occult and overt gastrointestinal bleeding.8 Therefore, it might be inferred that antithrombotic drugs affect the diagnostic performance of FOBT. However, cessation of antithrombotic drugs might cause serious thrombotic events. Therefore, if antithrombotic drugs do not affect the diagnostic performance of iFOBT, it is desirable that they should be continued. Concerning this issue, there are scant and controversial data.9–18 Moreover, most of these studies were conducted with guaiac-based FOBT.9,10,12,15–18 Levi et al. investigated the effect of aspirin, non-steroidal antiinflammatory drugs and anticoagulants on the performance of
I
N JAPAN, THE incidence of colorectal cancer (CRC) is very high1 and CRC is a major cause of cancer death. Among all cancers, the age-standardized ratio of CRC mortality in Japan is the highest in females and the third highest in males.2 Randomized control studies have revealed that populationbased screening for CRC using a fecal occult blood test (FOBT) is effective.3–5 These data were based on guaiac-
Corresponding: Nobuyuki Matsuhashi, Department of Gastroenterology, NTT Medical Center Tokyo, 5-9-22 Higashi-Gotanda, Shinagawa-ku, Tokyo 141-8625, Japan. Email: nmatuha-tky @umin.ac.jp The contents of this paper were presented, in part, at the 20th United European Gastroenterology Week Meeting, 12–16 October, 2012, Amsterdam, The Netherlands. Received 29 July 2013; accepted 15 October 2013.
424 bs_bs_banner
patients, respectively. There were no higher detection rates in the antithrombotic drug (–) group than in the (+) group (advanced neoplasms: 19.3% vs 19.4%, P = 1.000; cancers: 5.8% vs 5.8%, P = 1.000; any neoplasms: 48.4% vs 47.5%, P = 0.856). Multivariate logistic regression analysis revealed that none of aspirin, warfarin, or other antithrombotic drugs was a significant factor for advanced neoplasms (95% CI 0.350–1.216, P = 0.179; 95% CI 0.421–1.899, P = 0.772; 95% CI 0.323–1.810, P = 0.764, respectively). As to cancers and any neoplasms, no antithrombotic drug also proved to be a significant factor.
© 2013 The Authors Digestive Endoscopy © 2013 Japan Gastroenterological Endoscopy Society
Digestive Endoscopy 2014; 26: 424–429
Antithrombotic drugs and iFOBT 425
iFOBT, but that study was not carried out in a screening setting.14 Hence, as to iFOBT, there are still few data on the effect of antithrombotic drugs. The aim of the present study was to assess the effect of antithrombotic drugs on the diagnostic performance of iFOBT.
METHODS Study protocol
T
HIS STUDY PROTOCOL was approved by the Institutional Review Board of the NTT Medical Center Tokyo, and written informed consent for colonoscopy examination from the cecum to the rectum (TCS) was obtained from all patients before the original procedures were conducted. The study was carried out in accordance with the Declaration of Helsinki. This study was registered in the University Hospital Medical Network Clinical Trials Registry (UMIN000010100).
Patients Electronic medical charts of all patients who underwent TCS for positive iFOBT at NTT Medical Center Tokyo in 2011 were reviewed. In Japan, national colorectal cancer screening using iFOBT has been incorporated into public health policy since 1992.7 The recommended screening program is as follows. Screening by health facilities in residential areas is offered to persons aged 40 years or older. A 2-day iFOBT is conducted, and if one or two samples show positive results, a detailed examination is recommended. TCS is recommended as the diagnostic examination, but, if impossible, sigmoidoscopy in combination with a double-contrast barium study is permitted.7 Exclusion criteria in the present study were as follows: inflammatory bowel disease, colectomy history, and incomplete colonoscopy.
Data collection Drug usage information was obtained from electronic medical charts. Before carrying out TCS, it was necessary to know whether the patients were regularly taking antithrombotic drugs in order to avoid polypectomy during antithrombotic treatment. Therefore, doctors ordering TCS checked the antithrombotic medication status and recorded it in the electronic medical chart. Patients were categorized as antithrombotic drug takers based on the medication status within 30 days before iFOBT. Aspirin, clopidogrel, ticlopidine, warfarin, cilostazol and dabigatran were defined as antithrombotic drugs. We also searched each endoscopist’s experience of TCS, and defined examiners with ≥1000 cases of TCS as ‘experienced endoscopists’, and those with
