Deciding not to resume warfarin therapy after an episode of gastrointestinal bleeding raises the risk of thrombosis by a factor of 10 and the risk of death threefold, according to a retrospective cohort study published online September 17 in Archives of Internal Medicine.
"For many patients who have experienced gastrointestinal bleeding, the benefits of resuming warfarin therapy will outweigh the risks," said Daniel M. Witt, Pharm.D., of the clinical pharmacy anticoagulation service at Kaiser Permanente of Colorado, Aurora, and his associates.
Dr. Daniel Witt
"Surprisingly little is known about warfarin therapy and resumption" following a GI bleed, and there is no consensus as to the optimal timing or the risks of restarting anticoagulation. Dr. Witt and his colleagues used Kaiser’s administrative and clinical databases to study the incidence of thrombosis, recurrent bleeding episodes, and death from any cause in 442 adults who presented to a hospital or emergency department with warfarin-associated GI bleeding in 2005-2009 and who were followed for 90 days.
The mean patient age was 74 years, and the study population was equally comprised of men and women. Half of the study subjects were taking warfarin to prevent atrial fibrillation–related stroke or systemic embolization. One-quarter of patients used it to treat or prevent a second venous thrombosis, 10% were on it to prevent thromboembolic complications from prosthetic heart valves, and the remainder took it for other indications.
After an index GI bleed, 260 patients (59%) resumed warfarin therapy, usually within a week. The median time to resumption of warfarin was 4 days. In 41 of these patients, warfarin therapy was never suspended. It was suspended and never resumed in the remaining 182 patients.
During the 90-day follow-up, 11 patients (2.5%) had a thrombotic event. There were six arterial events, including five strokes and one systemic embolus, and five venous events, including three pulmonary embolisms and two deep vein thromboses (DVTs).
The rate of thrombotic events was 0.4% among the patients who resumed warfarin therapy (one DVT), compared with 5.5% among those who did not resume warfarin (five strokes, one systemic embolus, three pulmonary embolisms, and one DVT), a significant difference, the investigators said (Arch. Intern. Med. 2012 [doi:10.1001/archinternmed.2012.5261]).
"Patients who either never interrupted warfarin therapy or resumed therapy within 14 days of the index GI bleed experienced no thromboses," they added.
GI bleeding recurred in 36 patients (8.4%) overall. A numerically higher proportion of patients who resumed warfarin therapy had recurrent GI bleeding (10%) than those who did not resume warfarin (5.5%), but this difference was not statistically significant.
In addition, a multivariable analysis that accounted for numerous possible confounders – including patient age, sex, propensity for complications; the INR at admission; acute treatment for the GI bleed; and location of GI bleed – also showed that the risk for rebleeding was not significantly greater in patients who resumed warfarin therapy than in those who did not.
Moreover, recurrent GI bleeding was never fatal. Fatal strokes did occur, however, in three patients with atrial fibrillation whose warfarin therapy was withdrawn and never resumed, Dr. Witt and his associates said.
A total of 52 patients (12%) died during follow-up. The most common cause of death was related to malignancy (29% of deaths), infection (19% of deaths), or cardiac disease (17%).
Resumption of warfarin therapy was strongly associated with a threefold decrease in the risk of death from any cause.
The investigators also performed a post hoc analysis excluding all patients who died within 1 week of the index GI bleed to rule out those who may not have had an opportunity to resume warfarin therapy. In this analysis, the strong association between resumption of warfarin and decreased mortality persisted.
Mortality was lowest among patients who resumed warfarin therapy within 15-90 days of the index GI bleed, the researchers said.
"Our results provide some guidance regarding the optimal timing of warfarin therapy resumption following GI bleeding, but clinical judgment remains a critical factor in this difficult decision," they noted.
This study was funded by CSL Behring. Dr. Witt’s associates, but not Dr. Witt, reported ties to numerous industry sources.