BACKGROUND AND AIMS: The association between gastrointestinal (GI) bleeding and subsequent detection of GI cancer in patients using antiplatelet/anticoagulant medications is unclear. We investigated the association between the occurrence of GI bleeding and the detection of GI cancer and assessed whether this association differs in patients treated with clopidogrel or warfarin compared to non-treated patients. METHODS: A claims analysis was conducted using the Truven Health MarketScan(®) Research databases. Patients were grouped into the treatment cohort if they received a prescription for clopidogrel or warfarin or into the non-treatment cohort if they did not receive these medications. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for GI cancer diagnosed after GI bleeding. RESULTS: Overall, in the treatment cohort, patients who experienced a GI bleed were 6 times (HR: 5.64, 95% CI, 5.12, 6.21) more likely to be diagnosed with GI cancer compared with those without bleeding. In the non-treatment cohort patients were 13 times (HR: 13.34, 95% CI, 12.21, 14.58) more likely to be diagnosed with GI cancer after GI bleeding. The HRs of GI cancer were higher within 6 months of the first GI bleed and decreased remarkably thereafter. CONCLUSIONS: This study suggests that an episode of GI bleeding increased the rates of detection of GI cancers.
BACKGROUND AND AIMS: The association between gastrointestinal (GI) bleeding and subsequent detection of GI cancer in patients using antiplatelet/anticoagulant medications is unclear. We investigated the association between the occurrence of GI bleeding and the detection of GI cancer and assessed whether this association differs in patients treated with clopidogrel or warfarin compared to non-treated patients. METHODS: A claims analysis was conducted using the Truven Health MarketScan(®) Research databases. Patients were grouped into the treatment cohort if they received a prescription for clopidogrel or warfarin or into the non-treatment cohort if they did not receive these medications. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for GI cancer diagnosed after GI bleeding. RESULTS: Overall, in the treatment cohort, patients who experienced a GI bleed were 6 times (HR: 5.64, 95% CI, 5.12, 6.21) more likely to be diagnosed with GI cancer compared with those without bleeding. In the non-treatment cohort patients were 13 times (HR: 13.34, 95% CI, 12.21, 14.58) more likely to be diagnosed with GI cancer after GI bleeding. The HRs of GI cancer were higher within 6 months of the first GI bleed and decreased remarkably thereafter. CONCLUSIONS: This study suggests that an episode of GI bleeding increased the rates of detection of GI cancers.
Authors: Matthew T Roe; Derek D Cyr; Debra Eckart; Phillip J Schulte; Michael A Morse; Kimberly L Blackwell; Neal E Ready; S Yousuf Zafar; Anne W Beaven; John H Strickler; Jane E Onken; Kenneth J Winters; Lisa Houterloot; Dmitry Zamoryakhin; Stephen D Wiviott; Harvey D White; Dorairaj Prabhakaran; Keith A A Fox; Paul W Armstrong; E Magnus Ohman Journal: Eur Heart J Date: 2015-12-05 Impact factor: 29.983