Table 1. Selected studies, including AF-HF, comparing NOACs vs. warfarin.
| First Author | Type of
study |
Study
duration |
Setting | Comparators | Number of patients in each comparator | Patients with HF | Comments |
| AF: atrial fibrillation; HF: heart failure; LVEF: left ventricle ejection fraction; MACEs: major adverse cardiac event; NOACs: non-vitamin-K oral anticoagulants; RCT: randomized-control trials; VTE: vein thromboembolism. | |||||||
| Fereira, et al.[29] | Analysis of RCT | 2005−2011 | Global | Dabigatran (110 and 150 mg) vs. warfarin | 1641/
1640/ 1623 |
100% | The relative effects of dabigatran vs. warfarin on the occurrence of stroke or SE and major bleeding were consistent among those with and without HF and those with low (≤ 40%) or preserved (> 40%) LVEF (P interaction not significant). |
| Van Diepen, et al.[31] | Analysis of RCT | 2006−2010 | Global | Rivaroxaban vs. warfarin | 4530/
4503 |
100% | Rivaroxaban is an efficacious and safe alternative to VKAs in the population with HF with AF. Treatment-related outcomes were similar in patients with and without HF and across HF subgroups. |
| McMurray, et al.[32] | Analysis of RCT | 2006−2011 | Global | Apixaban vs. warfarin | 9120
(total AF-HF patients) |
100% | Apixaban reduced the risk for both stroke or systematic embolism and death more than warfarin in patients with HF independently of LVEF (P > 0.05). |
| Magnani, et al.[33] | Analysis of RCT | 2008−2013 | Global | Edoxaban vs. warfarin | 3097/
4048 |
100% | The efficacy and safety of edoxaban compared with controlled warfarin in AF patients with HF were similar to those without HF. |
| Laliberte, et al.[56] | Retrospective, Observational | 2011−2012 | US | Rivaroxaban vs. warfarin | 3654/
14616 |
19.6% | Real world data show that rivaroxaban and warfarin are equivalent regarding safety and efficacy, while rivaroxaban is associated with less VTE and better implementation. |
| Hecker, et al.[57] | Prospective, Observational | 2011−2013 | Germany | Rivaroxaban | 1204 | 37.2% | Effectiveness and safety of rivaroxaban is confirmed in real-world clinical practice. |
| Friberg, et al.[58] | Retrospective, observational | 2011−2014 | Sweden | Apixaban/Dabigatran/Rivaroxaban/Warfarin | 6547/
6651/ 5440/ 49418 |
19.5% | NOACs provided a safer profile than warfarin. |
| Yoshiha, et al.[59] | Retrospective, observational | 2011−2015 | US | Apixaban/Edoxaban/Dabigatran/Rivaroxaban vs. warfarin | 52/35/
33/30/257 |
100% | All-cause mortality was significantly lower in the NOACs group than in the warfarin group inthe post-matched cohort (12.3% vs. 35.1%, log-rank P = 0.038) |
| Amin, et al.[60] | Retrospective, observational | 2012−2015 | US | Apixaban/Dabigatran/Rivaroxaban/warfarin | 10615/
4927/ 15921/ 32373 |
100% | Apixaban was safer, regarding major bleeding and more effective regarding MACEs, comparing to other NOACs and warfarin |
| Lip, et al.[61] | Retrospective, observational | 2013 | US | Apixaban/Dabigatran/Rivaroxaban/Warfarin | 2402/
4173/ 10050/ 12713 |
20.2%/
20./ 19.%/ 27.3% |
Among newly anticoagulated AF patients in the real‐world practice, initiation with rivaroxaban or warfarin was associated with a significantly greater risk of major bleeding compared with initiation on apixaban. |
| Hohnloser, et al.[62] | Retrospective, observational | 2013−2015 | Germany | Apixaban/Dabigatran/Rivaroxaban/Phenprocoumon | 3633/
3138/ 12063/ 16179 |
37.1%/
31.7%/ 34.6%/ 40.4% |
Apixaban is associated with a significantlylower risk for bleeding compared to phenprocoumon, dabigatran was equivalent to phenprocoumon bleeding risk with rivaroxaban washigher. |
| Von Lueder, et al.[63] | Retrospective, observational | 2015 | US | Apixaban/Edoxaban/Dabigatran/Rivaroxaban vs. warfarin | 666/
32/ 1361/ 1005/ 8260 |
100% | NOACs were superior in all-cause mortality and MACEs, vs. warfarin. |