Literature Review
By Kyle Hobbs, MD, and Sara Stern-Nezer, MD
Due to their efficacy and favorable rates of complications when
compared to vitamin K antagonist therapy, direct or novel oral
anticoagulant (NOAC/DOAC) use is now widespread, and
reversal agents exist for many of these medications. The studies
reviewed below analyzed how different disease outcomes
(traumatic brain injury and subdural hematoma) were affected
by the use of DOAC therapy.
The novel oral anticoagulants have worse outcomes compared
with warfarin in patients with intracranial hemorrhage after TBI.
J Trauma Acute Care Surg 2018; 85: 915-20.
Summary: NOAC use in trauma patients is becoming increasingly
prevalent, but with limited understanding of effect on outcomes
in the TBI population. This single-center observational analysis
of prospectively collected data included all adult trauma patients
admitted with a diagnosis of traumatic brain injury (TBI) who had
intracerebral hemorrhage (ICH) on initial head CT and were taking
preinjury oral anticoagulants (warfarin, direct thrombin inhibitors,
or Xa inhibitors). Patients were stratified by type of preinjury oral
anticoagulant (warfarin vs NOAC) and by severity of TBI. Propensity
score matching was performed for the two cohorts in a 1:2 ratio
(NOAC:warfarin). NOAC patients had a higher Injury Severity Score
(ISS) on admission. After propensity score matching (70 NOACs,
140 warfarin), mean age was 58.7 + 15.2 years, 67.6% male,
median GCS 14 (IQR, 8-15), with falls being the most common
mechanism of injury. In the NOAC group, 54% were taking oral
Xa inhibitors and 46% were taking dabigatran (direct thrombin
inhibitor). There were no differences in demographic characteristics,
comorbidities, mechanism of injury, GCS on admission, or rate of
epidural, subdural, or subarachnoid hemorrhage. 58% of patients
were taking antiplatelet agents in addition to their anticoagulant,
evenly matched between the two groups. NOAC patients had
higher rates of progression of ICH on repeat CT scan (p=0.03),
neurosurgical intervention (p=0.04), longer ICU length of stay
(p=0.04), and higher mortality (p=0.04). There were no differences
in hospital length of stay or discharge GOS-E. Subanalysis showed
that mild and moderate TBI patients were significantly more likely
to be discharged to rehab/SNF, and had higher mortality and need
for neurosurgical intervention, while these differences were not
significant for patients with severe TBI.
Commentary: This study showed worse outcomes in TBI patients
who were being treated with NOAC therapy when compared to
patients taking warfarin. This analysis has limited generalizability
as it was a single center study, and the reason for anticoagulation
was not included; however, the propensity score matching did
attempt to account for confounding variables between groups.
Data was lacking on the reversal agents used in both the NOAC
and warfarin groups, which likely had an impact on progression
of hemorrhage, as did the relatively high rate of concomitant
antiplatelet use. Further study in this population is needed,
particularly with the approval of adexanet alfa.
Improved outcomes in patients taking DOACs over VKA in
acute subdural hemorrhage.
Neurocrit Care 2018; epub ahead of print.
Summary: While therapy with vitamin K antagonists (VKA) has
been shown to worsen outcomes in acute SDH, there is little data
on the effect of DOAC therapy in this population. This singlecenter
LITERATURE REVIEW
retrospective analysis included all patients treated for acute
SDH who were on DOAC or VKA therapy on admission. Patients
with purely chronic SDH were excluded, but acute on chronic
SDH was allowed. The neurosurgical consultant determined
whether to administer pro-hemostatic substances and/or perform
operative intervention. Prothrombin complex concentrate
(PCC) was administered in the majority of cases, as well as IV
phytomenadione if patients were on VKA therapy. 128 patients
were included (65 DOAC, 63 VKA). PCC was administered more
frequently in the VKA group and at higher mean dose. Patients
with VKA had a lower GCS score on admission, fewer VKA patients
had GCS 13-15, and more VKA patients were intubated prior to
hospital arrival. Recent trauma was reported more often in the
DOAC group (89% vs 60%, p<0.001). There were no significant
differences in the rate of rehemorrhage in the DOAC group,
and overall 30-day mortality was the same for DOAC vs VKA
(26% vs 27%, p=1.00). Rate of neurosurgical procedures was
not significantly different between groups. DOAC patients had
significantly higher Glasgow Outcome Scores (GOS) at hospital
discharge, and mean hospital length of stay was shorter (5.9 vs 7.2
days, p=0.024). Only 7 DOAC patients were taking dabigatran, of
which 4 (57%) received idarucizumab, with no deaths at 30 days.
One patient on apixaban who received PCC on admission died of
an ST-elevation MI 5 days later.
Commentary: This study suggests that patients on preinjury
DOAC therapy may have improved outcomes compared to those
taking VKAs in the event of acute subdural hemorrhage. While
mortality was the same between groups, the DOAC group had
shorter hospital LOS and better 30-day outcomes, as well as
similar rates of neurosurgical intervention and rehemorrhage
compared to warfarin. This study is limited in its small size and
that it was conducted at a single center. In addition, there was
imbalance between groups on admission, with higher mean
admission GCS scores in the DOAC group and lower rates of
prehospital intubation. It is possible, however, that the benefit
seen from being on DOAC therapy would have been higher with
andexanet alfa for Xa inhibitor reversal, as only 58% DOAC
patients received PCC in this study. Whether andexanet alfa use
translates to better clinical outcomes remains to be seen.
23