Atrial fibrillation (AF) is common and occurs in 2% to 4% of adults 60 years of age or older.1 Thromboembolic events, including stroke and non-central nervous system (CNS) systemic embolism (SE), are common complications. Non-CNS SE accounts for ~10% of all thromboembolic events2 and are important to identify because they are associated with high morbidity and mortality. Using data from ROCKET AF (Rivaroxaban Once daily, Oral, Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation; NCT00403767), we describe the incidence, location, diagnosis, treatment, and outcomes in patients with non-CNS SE. Baseline characteristics of patients with non-CNS SE are presented and discussed to identify those at increased risk of such an event.
Methods and Results
The design and methods of ROCKET AF have been described.3,4 Briefly, it was a multicenter, randomized, double-blind, event-driven trial conducted at 1178 participating sites in 45 countries.3 Included patients had atrial fibrillation and were at moderate-to-high risk for stroke as defined by a CHADS2 score ≥2. A total of 14,264 patients were randomly assigned to receive fixed-dose rivaroxaban 20 mg daily (15 mg daily in patients with creatinine clearance 30–49 mL/min) or dose-adjusted warfarin (target international normalized ratio 2.0–3.0). Patients were intended to continue study drug throughout the trial unless discontinuation was clinically indicated (e.g., safety concern, pregnancy, stroke or non-CNS SE, HIV diagnosis, abnormal liver function, creatinine clearance <25 mL/min on 2 consecutive measurements, or need for excluded medication).4
Non-CNS SE was defined as abrupt vascular insufficiency associated with clinical or radiologic evidence of arterial occlusion in the absence of other likely mechanisms. In the presence of atherosclerotic peripheral artery disease (PAD), diagnosis of embolism to the lower extremities required angiographic demonstration of acute arterial occlusion.4 For patients who met the definition of non-CNS SE, the anatomical location, method of diagnosis, presence of atherosclerosis, and recent trauma or instrumentation of the affected arterial territory were documented.
Institutional review board approval was obtained at all sites and all patients provided written informed consent.
The current analysis is a post-hoc subgroup analysis of the intention-to-treat population. Categorical variables are summarized as counts (percentages), and continuous variables are summarized as medians (25th, 75th percentiles) or means (standard deviations). All analyses were performed with SAS software version 9.2 (SAS Institute, Inc, Cary, NC).
Of 14,264 randomized patients in ROCKET AF, 47 experienced non-CNS SE, corresponding to a rate of 0.183 (95% confidence interval [CI] 0.136–0.241) per 100 patient-years of follow-up. (Patient characteristics shown in Table 1.)
Table 1.
Baseline characteristics of patients with and without non-CNS SE
| Characteristic | Patients with non-CNS SE (N=47) | Patients without non-CNS SE (N=14,124) |
|---|---|---|
| Treatment | ||
| Rivaroxaban | 20 (42.6%) | 7061 (50.0%) |
| Warfarin | 27 (57.4%) | 7063 (50.0%) |
| Age, yrs | 75 (68, 80) | 73 (65, 78) |
| Female | 24 (51.1%) | 5581 (39.5%) |
| Type of atrial fibrillation | ||
| Persistent | 39 (83.0%) | 11446 (81.0%) |
| Paroxysmal | 8 (17.0%) | 2482 (17.6%) |
| New onset / newly diagnosed | 0 (0.0%) | 196 (1.4%) |
| CHADS2 score, mean (SD) | 3.7 (1.1) | 3.5 (0.9) |
| CHADS2 score | ||
| 1 | 0 (0.0%) | 3 (0.0%) |
| 2 | 6 (12.8%) | 1851 (13.1%) |
| 3 | 13 (27.7%) | 6156 (43.6%) |
| 4 | 19 (40.4%) | 4048 (28.7%) |
| 5 | 5 (10.6%) | 1792 (12.7%) |
| 6 | 4 (8.5%) | 274 (1.9%) |
| Presenting Characteristics | ||
| Creatinine clearance*, mL/min | 55 (44, 75) | 67 (52, 87) |
| Baseline Comorbidities | ||
| Prior stroke, TIA, or non-CNS embolism | 32 (68.1%) | 7735 (54.8%) |
| Peripheral arterial disease | 6 (12.8%) | 826 (5.8%) |
| Hypertension | 43 (91.5%) | 12781 (90.5%) |
| Diabetes | 14 (29.8%) | 5633 (39.9%) |
| Prior myocardial infarction | 16 (34.0%) | 2430 (17.2%) |
| Congestive heart failure | 2 (7.4%) | 1190 (13.5%) |
| Chronic obstructive pulmonary disease | 7 (14.9%) | 1474 (10.4%) |
| Medications | ||
| Prior vitamin K antagonist use | 33 (70.2%) | 8820 (62.4%) |
| Prior chronic aspirin use | 17 (36.2%) | 5167 (36.6%) |
Categorical variables are shown as a number (%), and continuous variables are shown as a median (25th, 75th percentiles), unless otherwise noted.
Calculated using the Cockcroft-Gault equation.
Abbreviations: CNS, central nervous system; SD, standard deviation; SE, systemic embolism; TIA, transient ischemic attack.
Of the 47 non-CNS SE events, 29 occurred in lower extremities, 8 in mesenteric arteries, 6 in upper extremities, 2 in renal arteries, and 1 in the splenic artery; location was not recorded for 1 non-CNS SE.
Methods of diagnosis of non-CNS SE varied, but the most common were ultrasound (34%), computed tomography/angiography (21%), invasive angiography (17%), and magnetic resonance imaging/angiography (2%). Multiple diagnosis methods could be used for a single patient or patients could be diagnosed clinically. The initial diagnostic test was not standardized or dictated by the study protocol. There were 16 patients in whom no diagnostic test was performed, and while clinical diagnosis was not a pre-specified method, this can be inferred.
The use of study medication (rivaroxaban or warfarin) was variable at the time of non-CNS SE. Prior to the occurrence of non-CNS SE, 21 patients had permanently discontinued study drug and 2 had temporarily discontinued. Reasons for permanent discontinuation included bleeding adverse events (n=3), non-bleeding adverse events (n=2), non-compliance with study medication (n=1), withdrawn consent (n=5), investigator decision (n=2 [not protocol-related]), or the clinical efficacy endpoint (i.e., stroke) was reached (n=8). The median time from permanent discontinuation to non-CNS SE was 24 days (25th, 75th percentiles: 6, 92). However, 168 of the 534 patients who experienced stroke in ROCKET AF permanently stopped study drug before the stroke (31.5%).
Following non-CNS SE, 14 patients stopped study medication permanently, 1 temporarily discontinued use, and 7 continued study medication throughout the study. Study medication use was not adequately characterized in 2 patients.
There were 53 interventions in 47 patients; multiple interventions could be attempted in the same patient when clinically indicated. The most common interventions were surgery (47%), medical management (28%), and percutaneous intervention (13%); 11% of interventions were not described.
A total of 11 patients died (n=6 rivaroxaban, n=5 warfarin); 7 died within 30 days of non-CNS SE, 2 died between 30 days and 6 months, and 2 died >6 months after the non-CNS SE event. Of the 11 fatal events, 5 occurred in lower extremities, 5 in mesenteric arteries, and 1 in upper extremities.
Among those who survived the non-CNS SE event, 2 patients had a subsequent stroke, 1 required amputation, and 3 experienced major bleeding events. Of those who experienced major bleeding, 1 patient had multiple major bleeding events and 2 patients had major bleeding on the same day as the non-CNS SE event.
Comment
In patients enrolled in ROCKET AF, non-CNS SE occurred infrequently, at a rate of 0.183 (95% CI 0.136–0.241) per 100 patient-years. Importantly, non-CNS SE accounted for only 8.2% of the total thromboembolic events, while stroke accounted for the remainder, which is similar to prior estimates.2 The difference in frequency between stroke and non-CNS SE is multifactorial. One explanation for this discrepancy is that the higher prevalence of symptomatic embolization of carotid arteries versus subclavian arteries can be associated with the low resistance and increased volume of blood flow in cerebral arteries.5 Alternatively, it is reasonable to assume that the disparity is at least partly due to the difficulty in detecting non-CNS events. Cerebral tissue is particularly sensitive to ischemia as compared with other organs. Visceral or mesenteric SE detection is impacted by asymptomatic or non-specific symptoms, and many of these locations have a rich vascular network leading to collateral circulation and compensation. Consequently, the prevalence of non-CNS SE events likely will remain understated.5 The vast majority of non-CNS SE events occurred in lower extremity arterial distributions, which is consistent with prior data.5–7 PAD and prior atherosclerosis in the affected arterial distribution appear to be associated with the occurrence of non-CNS SE, which may suggest that non-CNS SE occurs more readily in areas of diseased vasculature. The mechanism explaining this finding may be that occlusion of the vasculature due to embolic material occurs at areas of atherosclerotic plaque, causing total obstruction of blood flow.8 Alternatively, some of the association with PAD may be related to thrombosis of a diseased vessel rather than a true embolic event.5,8
Bleeding was a major adverse event associated with non-CNS SE (n=3 [6.4%]), and while this sample size is small, the rate of major bleeding seems to be increased compared with the intention-to-treat population (3.6% of patients on rivaroxaban, 3.4% of patients on warfarin).3 Given the small sample size it is not possible to draw a conclusion, although patients with non-CNS SE likely received additional procedures and antithrombotic therapy as treatment, thus placing them at increased bleeding risk. There was also 1 patient (2.8%) who required amputation after lower extremity SE (n=35 for all extremity SE). Amputation is required in ~7% of limb ischemia cases due to PAD.9
There was a high rate of mortality associated non-CNS SE, as 11 patients (23.4%) who experienced it died. Significantly, the majority of patients died within 30 days of the event. Even more striking was the difference in mortality based on location of SE. For lower extremity SE, the rate of mortality in ROCKET AF was 17.2%, which is similar to previous reports for acute limb ischemia; previous studies suggest that the mortality rate of acute limb ischemia in the community is approximately 9% in-hospital and 42.5% at 1 year post-event.2,10 Visceral SE (including mesenteric, renal, and splenic) has a much higher rate of mortality, with 45.5% of patients experiencing such an event in ROCKET AF eventually dying. This is representative of prior literature, in which reported mortality rates of visceral or mesenteric ischemia are 55% to 60%.2
Non-CNS SE is less commonly recognized clinically than ischemic stroke, but is associated with significant morbidity and mortality. While the majority of these events occur in the extremities, visceral events also occur and are associated with high mortality rates. Many patients permanently stopped study medication prior to non-CNS SE, emphasizing the importance of uninterrupted anticoagulation in patients with atrial fibrillation. There is some suggestion that PAD and prior atherosclerosis may predispose patients to non-CNS SE; additional studies are needed to identify patients at risk and guide appropriate treatment.
Acknowledgments
Sources of Funding: ROCKET AF was supported by Johnson & Johnson Pharmaceutical Research and Development and Bayer HealthCare. David Huberman was supported by the National Institutes of Health (grant T32 HL079896).
Footnotes
Disclosures: Orgel, Wojdyla: None. Huberman: National Institutes of Health (grant T32 HL079896). Halperin: Consultant/Advisory board: Bayer AG HealthCare, Boehringer Ingelheim, Daiichi Sankyo, Johnson & Johnson, Ortho-McNeil-Janssen Pharmaceuticals, Pfizer, Sanofi Aventis, Biotronik, Boston Scientific, Janssen, Medtronic. Breithardt: Institutional research grant: BMS/Pfizer, Sanofi-Aventis, St. Jude; Speaker bureau: Bayer, BMS/Pfizer, Daiichi Sankyo; Consultant/Advisory board: Bayer, BMS/Pfizer. Singer: Institutional research grant: Johnson & Johnson, Bristol-Myers Squibb, Boehringer Ingelheim, Medtronic; Consultant/Advisory board: Boehringer Ingelheim, Bristol-Myers Squibb, CVS Health, Johnson & Johnson, Merck, Pfizer, St. Jude Medical. Fox: Institutional research grant: Bayer, Janssen, AstraZeneca; Honoraria: Bayer, AstraZeneca, GlaxoSmithKline, Janssen, Sanofi; Consultant/Advisory board: Bayer, Lilly, AstraZeneca, Sanofi. Hankey: Speaker bureau: Bayer; Consultant/Advisory board: Bayer, Sanofi; Other: AC Immune (Chair, Data Monitoring Committee). Mahaffey: Consultant: Eli Lilly, ACC, AstraZeneca, BAROnova, Bayer, Bio2 Medical, Boehringer Ingleheim, Bristol-Myers Squibb, Cubist, Elsevier (AHJ), Epson, Forest, GlaxoSmithKline, Johnson & Johnson, MEDTRONIC, Merck, Mt. Sinai, MyoKardia, Omthera, Portola, Purdue Pharma, Spring Publishing, The Medicines Company, Vindico, WebMD; Ownership interest: BioPrint Fitness; Institutional research grants: Amgen, Boehringer Ingelheim, Daiichi Sankyo, Johnson & Johnson, MEDTRONIC, St. Jude, Tenax. Jones: Institutional research grant: American Heart Association, AstraZeneca, Bristol-Myers Squibb, Boston Scientific Corp.; Honoraria: American College of Physicians, American College of Radiology, American Physician Institute. Patel: Institutional research grant: AstraZeneca, CSL, HeartFlow, Janssen Research & Development, Johnson & Johnson, Maquet, Medtronic, NHLBI; Consultant: AstraZeneca, Bayer, CSL, Genzyme, Janssen Research & Development, Medtronic, Merck.
References
- 1.Petersen P. Thromboembolic complications in atrial fibrillation. Stroke. 1990;21:4–13. doi: 10.1161/01.str.21.1.4. [DOI] [PubMed] [Google Scholar]
- 2.Bekwelem W, Connolly SJ, Halperin JL, Adabag S, Duval S, Chrolavicius S, Pogue J, Ezekowitz MD, Eikelboom JW, Wallentin LG, Yusuf S, Hirsch AT. Extracranial systemic embolic events in patients with nonvalvular atrial fibrillation: Incidence, risk factors, and outcomes. Circulation. 2015;132:796–803. doi: 10.1161/CIRCULATIONAHA.114.013243. [DOI] [PubMed] [Google Scholar]
- 3.Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Breithardt G, Halperin JL, Hankey GJ, Piccini JP, Becker RC, Nessel CC, Paolini JF, Berkowitz SD, Fox KA, Califf RM ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638. [DOI] [PubMed] [Google Scholar]
- 4.ROCKET AF Study Investigators. Rivaroxaban-once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation: rationale and design of the ROCKET AF study. Am Heart J. 2010;159:340–347. doi: 10.1016/j.ahj.2009.11.025. [DOI] [PubMed] [Google Scholar]
- 5.Wasilewska M, Gosk-Bierska I. Thromboembolism associated with atrial fibrillation as a cause of limb and organ ischemia. Adv Clin Exp Med. 2013;22:865–873. [PubMed] [Google Scholar]
- 6.Geller BJ, Giugliano RP, Braunwald E, Murphy SA, Hanyok JJ, Jin J, Mercuri M, Antman EM, Ruff CT. Systemic, noncerebral, arterial embolism in 21,105 patients with atrial fibrillation randomized to edoxaban or warfarin: results from the Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction Study 48 trial. Am Heart J. 2015;170:669–674. doi: 10.1016/j.ahj.2015.06.020. [DOI] [PubMed] [Google Scholar]
- 7.Chatterjee NA, Lubitz SA. Systemic Embolic Events (SEE) in Atrial Fibrillation: SEEing Embolic Risk More Clearly. Circulation. 2015;132:787–789. doi: 10.1161/CIRCULATIONAHA.115.018172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Frost L, Engholm G, Johnsen S, Møller H, Henneberg EW, Husted S. Incident thromboembolism in the aorta and the renal, mesenteric, pelvic, and extremity arteries after discharge from the hospital with a diagnosis of atrial fibrillation. Arch Intern Med. 2001;161:272–276. doi: 10.1001/archinte.161.2.272. [DOI] [PubMed] [Google Scholar]
- 9.Korabathina R, Weintraub AR, Price LL, Kapur NK, Kimmelstiel CD, Iafrati MD, Ali Tahir SM. Twenty-year analysis of trends in the incidence and in-hospital mortality for lower-extremity arterial thromboembolism. Circulation. 2013;128:115–121. doi: 10.1161/CIRCULATIONAHA.113.003543. [DOI] [PubMed] [Google Scholar]
- 10.Baril DT, Ghosh K, Rosen AB. Trends in the incidence, treatment, and outcomes of acute lower extremity ischemia in the United States Medicare population. J Vasc Surg. 2014;60:669–677. doi: 10.1016/j.jvs.2014.03.244. [DOI] [PMC free article] [PubMed] [Google Scholar]