Patients With A-fib Still Have High Mortality Risk After Appropriate Anticoagulation Is Started
Only a fraction of deaths that occur in anticoagulated patients with A-fib are related to stroke or bleeding, according to a new meta-analysis. The finding underscores the need to explore approaches other than modifications in anticoagulation to address the lingering high risk of dying in this population.
Nearly half of deaths (46%) in four pivotal trials of non-vitamin K antagonist oral anticoagulants (NOACs) versus warfarin were cardiac in nature and included sudden death, heart failure, and MI, lead author Antonio Gómez-Outes, MD, PhD (Spanish Agency for Medicines and Medical Devices, Madrid, Spain), and colleagues show.
Another 30% were nonvascular deaths, whereas only about 6% were due to ischemic stroke or systemic embolism and about 6% were related to hemorrhage, the investigators report in a study published in the December 13, 2016, issue of the Journal of the American College of Cardiology. The rest of the deaths were due to other vascular reasons or unknown causes.
Overall, NOACs lowered mortality risk relative to warfarin, mostly in connection to prevention of fatal bleeds, which were infrequent in both groups.
“From a global health care perspective, further measures to reduce mortality in this population, beyond appropriate anticoagulation not incurring an excessive increase in major bleeding, have to include an improvement in the management of relevant comorbidities, mainly heart failure, coronary artery disease, and diabetes, together with proven global CV risk-reduction measures and healthy lifestyle changes,” the authors write.
Christian Ruff, MD (Brigham and Women’s Hospital, Boston, MA), who was not involved in the study, told TCTMD that the analysis raises an important point regarding the residual high risk of death in appropriately anticoagulated A-fib patients and highlights the fact that some of the mortality advantage for NOACs is attributable to their better safety profile.
But, he said, the reduction in fatal bleeds observed with NOACs is not enough to fully explain the lower rate of all-cause mortality, a point also made in an accompanying editorial written by Stuart Connolly, MD (McMaster University, Hamilton, Canada).
The difference in the risk of dying can also be attributed to problems caused by more minor types of bleeding, even if a patient fully recovers from that event, Ruff said. That could be related to a patient either permanently or temporarily stopping anticoagulation after a bleed or to the inflammatory, prothrombotic state that results from the hemorrhage, he noted.
“These more minor bleedings that we say are ‘minor’ are actually really important because they lead to either changes in therapy, particularly discontinuing drugs such as anticoagulants, or they lead to other complications that that patient has weeks or months down the road that put them on a course to having worse outcomes,” Ruff said.
There are still a lot of unknowns, however, about why these patients continue to have such a high risk of death even after initiating anticoagulation and what should be done about it, he said.
As the NOACs have been introduced into practice, clinicians have been using reduced doses in a greater proportion of patients than would be expected based on how the drugs were used in the pivotal trials, Ruff noted. One of the reasons is that physicians are “very keen” to avoid bleeding, and that instinct appears to be correct because bleeds tend to worsen subsequent clinical outcomes, he said.
The current study “reinforces the notion that safety is really intricately tied to efficacy,” Ruff said, and “it shows that we probably have to pay much more attention to patients’ other comorbid conditions . . . that are likely amplifying this mortality signal and aren’t really directly treated at all by their anticoagulant.”
There are currently four NOACs approved for stroke prevention in patients with nonvalvular A-fib: dabigatran (Pradaxa; Boehringer Ingelheim), apixaban (Eliquis; Bristol-Myers Squibb), rivaroxaban (Xarelto; Janssen Pharmaceuticals), and edoxaban (Savaysa; Daiichi Sankyo). Gómez-Outes et al examined causes of death among the participants in the pivotal trials supporting the approval of those agents; the analysis included a total of 71,683 patients treated with a NOAC or warfarin followed for a median of 1.8 to 2.8 years.
The average annual all-cause mortality rate was lower in patients receiving a NOAC instead of warfarin (4.46% vs 4.87%; RR 0.90; 95% CI 0.86-0.95). That difference was mostly due to a halving of fatal bleeding (0.19% vs 0.38% per year; RR 0.49; 95% CI 0.40-0.61), with no significant impacts on non-bleeding-related vascular deaths or nonvascular deaths.
“The pernicious effects of anticoagulant-related bleeding are often underappreciated by the medical community, and the [study] nicely draws our attention to the beneficial effect of minimizing these bleeding events,” Connolly writes in his editorial. But like Ruff, he points out that the difference in fatal bleeds cannot fully explain the NOAC advantage.
“Several analyses of the NOAC trials suggest that there are pernicious long-term consequences of anticoagulant-related major bleeding that are not captured by the immediate bleeding-related case fatality rate,” he writes. In addition to discontinuation of anticoagulation, other factors like adverse effects of transfusion, continuing anemia, subclinical organ damage sustained during the bleed, and an induced prothrombotic state could be at play, he says.
“These findings remind us that we should pay particular attention to protecting our patients from bleeding when we use oral anticoagulants or other antithrombotic agents,” Connolly says. “The common idea that strokes and myocardial infarctions leave permanent damage, but patients recover without long-term consequences from major bleeding, is an intuition that is not supported by current data.”
Gómez-Outes A, Lagunar-Ruíz J, Terleira-Fernández A-I, et al. Causes of death in anticoagulated patients with atrial fibrillation. J Am Coll Cardiol. 2016;68:2508-2521.
Connolly SJ. Anticoagulant-related bleeding and mortality. J Am Coll Cardiol. 2016;68:2522-2524.
- Gómez-Outes reports no relevant conflicts of interest.
- Connolly reports relationships with Bristol-Myers Squibb, Pfizer, Bayer, Portola, Janssen, Boehringer Ingelheim, and Daiichi Sankyo.
- Ruff reports having served on advisory boards for manufacturers of NOACs approved for stroke prevention in A-fib.