Safety of Intravenous Thrombolysis Among Patients Taking Direct Oral Anticoagulants

A Systematic Review and Meta-Analysis

Shima Shahjouei, MD, MPH; Georgios Tsivgoulis, MD, PhD, MSc, FESO; Nitin Goyal, MD; Alireza Sadighi, MD; Ashkan Mowla, MD; Ming Wang, MS, PhD; David J. Seiffge, MD; Ramin Zand, MD, MPH


Stroke. 2020;51(2):533-541. 

In This Article


The result of this study suggests that the risk of symptomatic hemorrhagic transformation after IVT is not increased among selected AIS patients pretreated with DOACs when it is compared with patients taking warfarin with INR <1.7 or those who do not take any anticoagulation. Finding of our review is independent of the elapsed time between last DOACs intake (less or more than 48 hours) and thrombolysis. Further studies should be conducted to determine the risk of thrombolysis early after DOACs intake.

We performed this study based on 3 notable assumptions. First, the only Food and Drug Administration–approved rout for the administration of alteplase in acute ischemic stroke patients is intravenous.[34,35] On this basis, we excluded all patients who underwent intraarterial thrombolysis.

Second, under normal renal function, the plasma concentration of DOACs after 48 hours from the last intake is low enough that high-risk surgical interventions would be possible.[36]

Despite that each of the DOACs has its unique pattern of pharmacodynamic and pharmacokinetic, all of them have rapid absorption, time to maximum concentration of <4 hours and short half-life (5–17 hours).[37] Thus, AHA/ASA guideline in 2018 allowed IV tPA after 48 hours following the last DOAC intake. As the purpose of this study was to evaluate the risk of sICH in the presence of DOAC consumption, we limited this review to patients who received IVT within 48 hours of last DOAC intake. One excluded study by these criteria was a report of the Get With The Guidelines-Stroke (GWTG-Stroke) program by Xian et al.[7] Among 42 887 stroke patients of this study, 251 individuals were treated with DOACs before IVT—dabigatran: 87, rivaroxaban: 129, and apixaban: 35. Among these 251 patients who received DOACs, 12 (4.8%) patients experienced sICH in <36 hours after the intervention. Similar to our study, Xian et al[7] could not find any significantly increased risk of sICH in patients treated with DOACs in comparison with patients on warfarin or those who were not anticoagulated before the stroke. The other excluded study due to the longer interval between last DOACs intake and IVT was from Japanese National Cerebral and Cardiovascular Center Stroke registry. Toyoda et al[15] presented 8 stroke patients under DOACs—dabigatran: 3, apixaban: 2, edoxaban:3—who received IVT without sICH complications in 36 hours. The interval between last DOAC intake and IVT was 15.8 (5.6–83.3) hours. This consensus guide on stroke thrombolysis for anticoagulated patients in Japan allowed IVT if the time span from the last dose of DOACs exceeds 4 hours, INR <1.7, and PTT <1.5× the baseline value (≤40 seconds as a guide). Based on our findings and the result of previous studies it seems that protocol violation and administration of tPA to selected patients with prior intake of DOACs do not carry the additional risk of symptomatic hemorrhagic transformation in the shorter interval from the last DOACs intake.

In the third assumption, we excluded all patients who underwent thrombolysis after receiving anticoagulation reversal with the specific antidote. Idarucizumab (a humanized monoclonal antibody fragment with very high affinity for dabigatran) is an Food and Drug Administration–approved reversal agent for DOACs. However, adverse outcome and mortality due to hemorrhagic transformation after IVT were reported in patients pretreated with Idarucizumab.[38,39] In the current review, when we considered all available reports of IVT in AIS patients treated with DOACs (81 patients who received Idarucizumab and 378 patients without pretreatment with reversal agents), there was an increased risk of hemorrhagic transformation and early mortality in patients who received the antidote. This result was in spite of higher comorbidities (diabetes mellitus, hypertension, hyperlipidemia, prior stroke/TIA, and coronary artery diseases) in the patients without pretreatment with reversal agents. Another reversal agent is andexanet alfa (a specific antidote for the anti-Xa inhibitors)[40–42] that received approval for use in patients treated with rivaroxaban and apixaban in the United States.[43] There is scarce data on clinical outcome of IVT in AIS patients who received andexanet alfa. In addition, the anticoagulation reversal agents are not available and affordable in many countries.

As physicians become more confident with thrombolysis under AHA/ASA guidelines, more reports on the safety of IVT protocol violation are being published.[44,45] Results of a prospective multicenter study by Matute et al[46] pointed out that IVT has no advanced risk of sICH in patients with INR <2. Likewise, Lyerly et al[44] reported that elevated prothrombin time (PT)/PTT is not associated with a higher risk of either sICH or any hemorrhagic transformation following IVT. Brunner et al[47] demonstrated that the risk of sICH and in-hospital mortality do not vary between patients with abnormal baseline coagulation status (warfarin/heparin therapy, coagulation disorders or idiopathic) and patients with normal coagulation state.

Similarly, recent systematic reviews and meta-analyses do not provide any causal evidence of poor outcomes after IVT among AIS patients with antiplatelet pretreatment.[48,49] For instance, Luo et al[49] on an analysis of 19 studies with a total of 108 588 patients could not reveal any independent association of unfavorable outcome or mortality following administration of systemic thrombolysis in AIS patients taking different antiplatelet agents—aspirin, dipyridamole, clopidogrel, ticlopidine, prasugrel, ticagrelor, cilostazol, abciximab, eptifibatide, and tirofiban. Likewise, a meta-analysis of 7 clinical trials by Tsivgoulis et al[48] on 4376 patients treated with aspirin, clopidogrel, or dipyridamole failed to picture the advanced risk of IVT-induced fatal sICH, 3-month functional adverse outcomes or mortality.

The strength of this study is direct communication with authors to get unpublished data according to our study protocol, reduce the duplicate reports, and minimize overall risk of bias. However, details of the clinical guideline vary among different centers, and this may cause bias. Our study considered the 48 hours as the cut point of last DOACs intake. Considering the exact time of last DOAC intake may present advance risk of thrombolytic therapy in shorter intervals. The sample size is still a limiting factor, as thrombolysis of the patients in <48 hours following the last DOAC intake is regarded as protocol violation, and many of the published cases were reported sporadically, and we failed to include them in the meta-analysis. In addition, the sample size of the patients who met our inclusion/exclusion criteria was considerably smaller than the control groups (366 versus 505 374) and this might affect the power of the study. Also, the present meta-analysis evaluated the safety of IVT in selected (by anticoagulation tests or decision made by treating physician) AIS patients pretreated with DOACs. Thus, the present findings are not easily generalizable.