Physicians' Academy for Cardiovascular Education

NOACs do not increase drug-induced liver injury

Literature - Caldeira D et al. Heart. 2014 - Heart. 2014 Jan 29


Risk of drug-induced liver injury with the new oral anticoagulants: systematic review and meta-analysis

Caldeira D, Barra M, Santos AT, et al.
Heart. 2014 Jan 29. doi: 10.1136/heartjnl-2013-305288


Some cardiovascular (CV) drugs have been associated with drug-induced liver injury (DILI), and the oral direct thrombin (IIa) inhibitor ximelagatran was withdrawn from the market due to the risk of DILI [1]. These safety warnings only emerge with postmarketing experience, because hepatic adverse drug reactions due to CV drugs are quite uncommon.
Closer surveillance and reporting of hepatic adverse events has been performed during the clinical trials of new oral anticoagulants (NOACs). All NOACs except for dabigatran, are metabolised by the liver. These drugs are associated with increases in transaminases and abnormal liver function [2-4]. The 2013 European guidance for the use of NOACs recommends annual monitoring of liver function [5].
This systematic review and meta-analysis of phase III randomised-controlled trials (RCTs) evaluated the risk of hepatic adverse drug reactions associated with NOACs. 29 studies met eligibility criteria, enrolling 152116 patients in total, 83513 of whom were treated with NOACs during a mean duration of 16 months.

Main results

  • Pooled analysis of 25 studies showed that NOAC use does not increase the risk of DILI (transaminases >3x upper limit of normal (ULN), with total bilirubin >2x ULN), with RR: 0.90 (95%CI: 0.72-1.13).
  • Individual NOACs did not show an increased risk of DILI, and no differences in risk of DILI were found between NOACs vs. control.
  • The incidence of DILI was <1% in intervention (0.22%) and control arms (0.24%).
  • When analysed separately for type of control group (vitamin K antagonists (VKAs), low-molecular weight heparin (LMWH) and placebo or non-pharmacological treatment), NOACs did not show an increased risk of DILI in any of the comparisons.
  • NOACs showed a lower likelihood than controls to have increased transaminases >3x ULN (RR: 0.79, 95%CI: 0.79-0.90). This effect was stronger in studies with LMWH as controls (RR: 0.71, 95%CI: 0.59-0.85), while the pooled results of the other trials showed a non-significant risk reduction of transaminase elevation (VKAs: RR: 0.81, 95%CI: 0.64-1.02).
  • No differences were seen between NOAC and control treatment in the risk of bilirubin elevations >2x ULN (RR: 0.93, 95%CI: 0.59-1.48).


These pooled estimates of large RCTs show that NOACs are not associated with an increased risk of drug-induced liver injury.
NOACs were found to be less likely associated with increased transaminases than controls. Since this putative ‘protective’ effect was mostly evident in studies that used LMWH control treatment, it may not reflect a true protective effect of NOAC, but rather LMWH-associated hepatotoxicity.
Find this article in Pubmed


1. Keisu M, Andersson TB. Drug-induced liver injury in humans: the case of ximelagatran. Handb Exp Pharmacol 2010;196:407–18.
2. Xarelto: Summary of Products Characteristics. (accessed 31 Oct 2013).
3. Eliquis: Summary of Products Characteristics. (accessed 31 Oct 2013).
4. Pradaxa: Summary of Products Characteristics. (accessed 31 Oct 2013).
5. Heidbuchel H, Verhamme P, Alings M, et al. European Heart Rhythm AssociationPractical Guide on the use of new oral anticoagulants in patients with non-valvularatrial fibrillation. Europace 2013;15:625–51.