1674

Short-term Tranexamic Acid Treatment in

Aneurysmal Subarachnoid Hemorrhage
Eelco F.M. Wijdicks, MD, Djo Hasan, MD, Kenneth W. Lindsay, PhD, FRCS,
Paul J.A.M. Brouwers, MD, Richard Hatfield, FRCS, Gordon D. Murray, PhD,
Jan van Gijn, MD, and Marinus Vermeulen, MD

Antinbrinolytic treatment for 4 weeks after a subarachnoid hemorrhage has been shown to have
no effect on outcome since a reduction in the rate of rebleeding was offset by an increase in
ischemic events. To determine if a shorter course (4 days) of antinbrinolytic treatment before
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017

the expected onset of ischemic complications might reduce the rate of rebleeding yet avoid
ischemic complications, we prospectively studied a series of 119 patients with subarachnoid
hemorrhage; 479 patients with subarachnoid hemorrhage from our previous randomized
double-blind study (238 treated with placebo, 241 with long-term tranexamic acid) served as
historical control groups. At 3 months' follow-up, the outcome of patients treated with
short-term tranexamic acid was not different from that of patients treated with long-term
tranexamic acid. The rate of rebleeding (24 of 119, 20%) was near that with placebo (56 of 238,
24%). In contrast, the rate of cerebral infarction (33 of 119, 28%) was almost identical to that
after long-term tranexamic acid (59 of 241, 24%), although mortality from cerebral infarction
was reduced. Compared with historical control groups, treatment with tranexamic acid for 4
days fails to reduce the incidence of rebleeding but still increases the rate of cerebral infarction.
{Stroke 1989;20:1674-1679)

I n a recent randomized controlled trial, we dem-

onstrated that treating patients with aneurys-
mal subarachnoid hemorrhage (SAH) with tran-
examic acid (TEA) significantly reduced the
incidence of rebleeding. This benefit, however, was
offset by an increase in the incidence of cerebral
infarction.1 In this previous trial, the duration of
treatment (4 weeks or until surgery or death) was
based on studies showing raised levels of fibrin
degradation products in the cerebrospinal fluid (CSF)
for up to 5 weeks after SAH.2>3 However, we
subsequently found reasons to doubt whether CSF
levels of fibrin degradation products represent any-
thing other than the permeability of the blood-brain
barrier.4
In the previous study, TEA was given during the
period of maximum risk for cerebral infarction. The
From the University Department of Neurology, Utrecht
(E.F.M.W., P.J.A.M.B., J.v.G.), the University Hospital Rot-
terdam, Rotterdam (D.H., M.V.), the Netherlands, the Depart-
ment of Neurosurgery, Royal Free Hospital, London (K.W.L.,
R.H.), and the Department of Statistics, University of Glasgow,
Glasgow (G.D.M.), United Kingdom.
Supported by Kabivitrum, Stockholm, Sweden.
Address for reprints: Eelco F.M. Wijdicks, University Hospi-
tal, Department of Neurology, Heidelberglaan 100, 3584 CX
Utrecht, The Netherlands.
Received March 20, 1989; accepted June 27, 1989.
rate of cerebral infarction in the TEA-treated group
started to diverge from that of the control group on
Day 6 of treatment. This raised the possibility that a
shorter period of treatment might avoid the concom-
itant increase in cerebral infarction and yet maintain
the inhibition of rebleeding. In this study, we chose
a treatment period of 4 days, with a maximum
interval between SAH and initiation of treatment of
72 hours. Before considering a full-scale random-
ized trial, we prospectively studied the effect of
short-term treatment with TEA on the rate of
rebleeding and cerebral infarction in a series of 119
consecutive patients with SAH. We compared the
results with those of two historical control groups
from the previous randomized controlled trial.1
Subjects and Methods
From January 1986 to May 1987, we prospec-
tively studied 119 consecutive patients with SAH
admitted <72 hours after the first clinical symptoms
to one of three hospitals. The criteria for eligibility
were symptoms and signs of SAH and computed
tomographic (CT) evidence of aneurysmal hemor-
rhage in the basal cisterns or fissures.5-6 In three
patients in whom CT scan showed no abnormality,
lumbar puncture demonstrated blood, confirmed by
blood pigments on spectrophotometry. As control
 Wijdicks et al Short-term TEA in SAH 1675

TABLE 1. Patient Characteristics at Entry Into Trials of TEA After Aneurysmal SAH
Current study Previous trial
Short-term TEA Long-term TEA Placebo
(n = 119) (n=241) (n=238)
Characteristic No. % No. % No. %
Sex
Male 45 38 95 39 94 39
Female 74 62 146 61 144 61
Mean age (yr) 52.9 50.3 50.2
Interval from SAH to entry
<24hr 78 66 95 39 91 38
25-48 hr 23 19 86 36 75 32
49-72 hr 18 15 60 25 72 30
Loss of consciousness at ictus
Yes 60 50 110 46 111 47
No 59 50 131 54 127 53
Glasgow Coma Scale score
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017

<12 33 28 50 21 46 19
12 or 13 27 23 63 26 70 30
14 59 49 128 53 122 51
Clinical grade on Hunt and Hess Scale
I 10 8 42 17 37 15
II 52 44 93 39 87 37
III 30 25 61 25 71 30
IVor V 27 23 45 19 43 18
Aneurysm confirmed
Yes
Angiography 68 57 130 54 155 65
Autopsy 9 8 22 9 19 8
No
Normal angiogram 19 16 51 21 34 14
Angiography nor autopsy performed 23 19 38 16 30 13
Site of ruptured aneurysm (if present)
Anterior communicating artery 21 31 64 42 70 40
Carotid artery 23 34 48 32 53 30
Middle cerebral artery 18 26 26 17 34 20
Posterior circulation 6 9 14 9 17 10
TEA, tranexamic acid; SAH, subarachnoid hemorrhage. % of total for characteristic.

groups we considered 238 patients treated with
placebo and 241 patients treated with long-term
TEA.1
Treatment with TEA commenced <72 hours after
the presenting SAH; 6 g/day were given for a
maximum of 96 hours. TEA was administered by
intravenous bolus in six doses. Treatment was
discontinued at surgery, if a diagnosis other than
aneurysm was established, if the angiogram was
negative, or if venous thrombosis or pulmonary
infarction developed. Exclusion criteria were the
presence or history of recent deep-vein thrombosis,
coagulation disorders, renal insufficiency, preg-
nancy, or negative angiography carried out before
the start of treatment. If death appeared imminent,
entry into the current study was delayed.
In addition, patients were randomized to receive
fludrocortisone,7 and fluid and sodium balances and
plasma volumes were compared in patients with
and without this treatment; these results have been
reported in a separate paper.8 During the study
period all patients had a fluid intake of at least 3,000
ml/day. If a patient was on antihypertensive treat-
ment before the SAH, the same dosage was contin-
ued. Hypertension developing in a patient without a
history of hypertension was not treated.
CT was carried out on admission and after clini-
cal deterioration. Four-vessel angiography was per-
formed at the clinician's discretion. The amount of
subarachnoid blood on CT scan was graded from 0
(no blood) to 3 (completely filled with blood) for the
frontal interhemispheric fissure, the quadrigeminal
cisterns, each suprasellar cistern, each ambient
cistern, each basal sylvian fissure, and each lateral
fissure; therefore, the maximum score was 30. Sim-
ilarly, the amount of intraventricular blood on the
 1676 Stroke Vol 20, No 12, December 1989

TABLE 2. Outcome at 3 Months in Trails of TEA After Aneurysmal Subarachnoid Hemorrhage

Current study Previous trial
Short-term TEA Long-term TEA Placebo
(n =119) (n=241) (n=238)
Outcome No. % No. % No. %
Death 45 38 84 35 89 37
Rebleeding 19 42 19 23 45 51
Delayed cerebral ischemia 10 22 38 45 24 27
Other causes 16 36 27 32 20 22
Severely disabled 9 8 30 12 23 10
Good recovery or moderately disabled 65 54 127 53 126 53
TEA, tranexamic acid.

admission CT scan was graded for each ventricle,
with a maximum score of 30.
In the current study (short-term TEA group) 113
(95%) and in the previous trial (long-term TEA or
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017
Outcome was assessed at 3 months according to
the five-point Glasgow Outcome Scale,9 and the
presence of a limb or speech deficit was noted. If a
patient died <3 months after the SAH, the cause of

placebo groups) 352 (73%) of the admission CT death was recorded.

scans were available for review.
All patients remained under continuous observa-
tion in an intensive care or high-dependency unit for
4 weeks or until the patient underwent surgery or
died. Any deterioration in the level of conscious-
ness or the development of focal signs was reported,
in which case the patient was reexamined. When-
ever possible these events were investigated with
CT. Intracranial complications were recorded as
rebleeding, infarction, hydrocephalus, local edema
from a hematoma, epilepsy, or other. Definite
rebleeding was denned as sudden deterioration,
with increased hemorrhage on CT scan or at autopsy
compared with a previous CT scan; probable rebleed-
ing was denned as sudden deterioration and death,
without the possibility of confirmation by CT and
when autopsy was refused; definite infarction was
defined as gradual development of focal signs, with
or without deterioration in consciousness level and
with CT or autopsy confirmation; probable infarc-
tion was defined as gradual development of focal
signs, with or without deterioration in conscious-
ness level, without confirmation by CT or autopsy,
but with exclusion of other causes that could explain
the deterioration; and extracranial events were
recorded as cardiorespiratory, metabolic, gastroin-
testinal bleeding, or other.
Results
Of the 119 patients with SAH enrolled in the
current study, 57 (48%) were from Rotterdam, 36
(30%) were from London, and 26 (22%) were from
Utrecht. Table 1 compares patient characteristics at
entry in the current study and in the previous
randomized trial. Only the distribution of interval
from SAH to entry differed significantly. This dif-
ference resulted from the large contribution of neu-
rologic centers with primary referral to the current
study; neurosurgical centers contributed more
heavily to the previous trial. Differences in the
interval from SAH to entry did not prove to be an
important prognostic factor, at least not within the
first 3 days.10 The incidence of near-maximal sub-
arachnoid blood scores (21-30) was slightly higher
in the current study, but the difference did not reach
significance (32 of 113, 28% vs. 66 of 352, 19%).
Frank intraventricular hemorrhage (score >2) was
distributed similarly (21% vs. 23%). Other charac-
teristics were closely similar, and no difference
reached significance.
In the current study, angiography was carried out
in 87 patients (73%). An aneurysm was demon-
strated in 77 patients (65%) on angiography or at
autopsy. More than one aneurysm was demon-

TABLE 3. Rate of Events in Trials of TEA After Aneurysmal Subarachnoid Hemorrhage

Current study Previous trial
Short-term TEA Long-term TEA Placebo
{n = \19) (n=241) (n=238)
Event No. % 95% CI No. % 95% CI No. % 95% CI
Rebleeding 24 20 13-27 21 9 5-13 56 24 19-29
Delayed cerebral ischemia 33 28 19-35 59 24 19-29 36 15 10-20
Hydrocephalus requiring shunt 21 18 11-25 35 15 10-20 29 12 8-16
Deep-vein thrombosis 0 0 — 12 5 — 9 4 —
Pulmonary embolism 2 2 8 3 9 4
TEA, tranexamic acid; CI, confidence interval. % of n.
 Wijdicks et al Short-term TEA in SAH 1677

f
E

days from presenting hemorrhage 0 5 10 15

FIGURE 1. Histogram of distribution of patients with
rebleeding by days after initial aneurysmal subarachnoid
hemorrhage. Hatched bars indicate rebleeding during
treatment with tranexamic acid.
strated in 16 patients (13%) in the current study.
Fifty-one patients (43%) underwent surgical clip-
ping of the aneurysm.
In the current study, short-term TEA treatment
lasted 96 hours in 100 patients (83%). Treatment
was discontinued earlier in two patients who had
surgery, in seven who died, in four who had a
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017
days from presenting hemorrhage
FIGURE 2. Histogram of distribution of patients with
delayed cerebral ischemia by days after initial aneurys-
mal subarachnoid hemorrhage. Hatched bars indicate
ischemic events during treatment with tranexamic acid.
severely disabled, and 44 (65%) had a good recov-
ery or remained moderately disabled.
Short-term treatment with TEA in the current
study did not reduce the rate of rebleeding com-
pared with treatment with placebo in the previous
trial (Table 3), although marked differences occurred

negative angiogram, in two in whom a cause for the
SAH other than an aneurysm was identified, and in
four for a variety of other reasons. In no patient did
side effects make it necessary to stop treatment.
Because the current study was designed to look at
patients with signs and symptoms of SAH on an
intention-to-treat basis, all 119 patients were included
in the final analysis.
Short-term treatment with TEA did not result in a
better outcome than long-term treatment with TEA
or treatment with placebo (Table 2). Of the patients
who died, almost half died of rebleeding. Mortality
from delayed cerebral ischemia was less in the
current study after short-term TEA than in the
previous trial after long-term TEA. Other causes of
death included ventriculitis following shunting in
three patients, the direct effect of the initial SAH in
one patient, and operative complications in three
patients. Restricting analysis of outcome to the 68
patients in the current study with an aneurysm
demonstrated on angiography did not alter these
findings; 19 patients (28%) died, five (7%) became
among the three centers; in Rotterdam 11 of 57
patients (19%), in London two of 36 (6%), and in
Utrecht 11 of 26 (42%) rebled. We failed to identify
factors other than chance to account for this varia-
tion. In one patient rebleeding was classified as
probable; in all other rebleeding was definite. Two
thirds of the patients who rebled did so after Day 4
of treatment (Figure 1). Six patients had more than
one definite additional hemorrhage; only one of
these later hemorrhages occurred during TEA treat-
ment. Only the time of the first rebleed is indicated
in Figure 1. The results did not change significantly
when only patients with a proven aneurysm were
considered (11 of 68, 16%).
The rate of delayed cerebral ischemia in the
current study closely resembled that after long-term
TEA treatment in the previous trial (Table 3).
Similar results were obtained when the analysis was
restricted to patients with a confirmed aneurysm (24
of 68, 35%). Delayed cerebral ischemia occurred in
most patients between Days 2 and 10 after SAH
(Figure 2). Of the 33 episodes of cerebral infarction,

TABLE 4. Events and Outcome After Angiographically Confirmed Aneurysmal Subarachnoid Hemorrhage by Time to Onset of Treatment
With TEA
Current study Previous trial
Short-term TEA ]Long-term TEA Placebo
0-24 hr 24-76 hr 0-2't h r 24-76 hr 0-24 hr 24-76 hr
(» =43) (n=25) (#i=46) (#i=84) («=64) (n=91)
No. % No. % No. % No. % No. % No. %
Events
Rebleeding 9 21 2 8 6 13 7 8 16 25 19 21
Delayed cerebral ischemia 15 35 9 36 19 41 18 21 7 11 12 13
Outcome
Death 12 28 7 28 18 39 19 22 21 33 29 32
Severe disability 3 7 2 8 7 15 9 11 7 11 11 12
Good recovery or moderate disability 28 65 16 64 21 46 56 67 36 56 51 56
TEA, tranexamic acid.
 1678 Stroke Vol 20, No 12, December 1989
21 were classified as definite. No marked differ-
ences in the incidences of hydrocephalus, deep-vein
thrombosis, or pulmonary embolism were noted
between the current study and the previous trial.
When all patients with a positive angiogram were
grouped according to the onset of treatment, the
distributions of events and outcome among the
three treatments were nearly identical. Thus, the
interval from SAH to treatment did not contribute
to any bias (Table 4).
Discussion
The hypothesis underlying our study was that
shortening the duration of antifibrinolytic treatment
after SAH might still prevent rebleeding without the
risk of ischemic complications. Unfortunately, the
reverse proved to be the case. The results of our
current prospective study clearly demonstrate that
TEA given for only 4 days no longer protects against
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017
rebleeding. Most second hemorrhages occurred after
stopping TEA; 4 days appears to be insufficient to
permit a firm plug to develop. The risk of rebleeding
can therefore be diminished only by continued anti-
fibrinolytic treatment throughout the high-risk period
in the second week.11 In contrast, the most important
side effect (development of cerebral infarction)
remained, even though treatment was stopped before
the period during which vasospasm is most fre-
quently encountered. Therefore, TEA may have a
harmful effect during the first days after SAH. Why
outcome in the current study was not worse than that
in the previous trial can be explained by two factors.
First, the number of patients in the current study was
relatively small. Second, mortality from infarction in
the previous trial was confounded by fluid restriction
in patients with hyponatremia12 and by antihyperten-
sive treatment.
How antifibrinolytic treatment precipitates delayed
cerebral ischemia remains unknown. Experimental
studies support the idea that antifibrinolytic agents
lead to more persistent thrombi.13-14 This effect
might add to the impairment of the microcirculation
caused by vasospasm.1S Another explanation for
the role of TEA in producing cerebral infarction is
that the drug might delay the disappearance of
blood from the basal cisterns and prolong contact of
vasoconstrictive substances with the vessel wall.
CSF obtained from patients with SAH significantly
inhibits endothelium-dependent relaxation of canine
basilar arteries.16-17
The use of historical controls carries substantial
disadvantages, but this study was designed to indi-
cate whether we should proceed with a randomized
controlled trial. A relatively short period elapsed
between the two studies, making radical changes in
management policies unlikely, but recognized dif-
ferences existed. No antihypertensive medication
was used in the current study, and all patients
received a large fluid intake; this may explain the
reduced mortality from cerebral infarction. Our
findings leave little doubt that TEA treatment for 4
days has no beneficial and only adverse effects. A
full-scale clinical trial of this regimen is therefore no
longer attractive. Antifibrinolytic treatment in gen-
eral is unlikely to be of value until other measures
have convincingly been demonstrated to reduce the
incidence of cerebral ischemia.
Acknowledgments
We thank the nursing staff and the residents for
their efforts, Mrs. E. Budelman-Verschuren for
excellent secretarial help, and Dr. Allen H. Ropper
for reviewing the manuscript.
References
1. Vermeulen M, Lindsay KW, Murray GD, Cheah F, Hijdra
A, Muizelaar JP, Schannong M, Teasdale GM, van Crevel
H, van Gijn J: Antifibrinolytic treatment in subarachnoid
hemorrhage. N EnglJ Med 1984;311:432-437
2. Maurice-Williams RS, Gordon YB, Sykes A: Monitoring
fibrinolytic activity in the cerebrospinal fluid after aneurys-
mal subarachnoid hemorrhage: A guide to the risk of rebleed-
ing? J Neurol Neurosurg Psychiatry 1980;43:175-181
3. Tovi D, Nilsson JM, Thulin CA: Fibrinolytic activity of the
cerebrospinal fluid after subarachnoid hemorrhage. Ada
Neurol Scand 1973;49:l-9
4. Vermeulen M, van Vliet HHDM, Lindsay KW, Hijdra A,
van Gijn J: Source offibrin/fibrinogendegradation products
in the CSF after subarachnoid hemorrhage. / Neurosurg
1985;63:573-577
5. Van Gijn J, van Dongen KJ: Computerized tomography in
subarachnoid hemorrhage: Difference between patients with
and without an aneurysm on angiography. Neurology 1980;
30:538-539
6. Van Gijn J, van Dongen KJ, Vermeulen M, Hijdra A: Perimes-
encephalic hemorrhage: A nonaneurysmal and benign form of
subarachnoid hemorrhage. Neurology 1985;35:493-497
7. Wijdicks EFM, Vermeulen M, van Brummelen P, van Gijn
J: The effect of fludrocortisone acetate on plasma volume
and natriuresis in aneurysmal subarachnoid hemorrhage.
Clin Neurol Neurosurg 1988;90:209-214
8. Hasan D, Lindsay KW, Wijdicks EFM, Murray GD, Brou-
wers PJAM, Bakker WH, van Gijn J, Vermeulen M: Effect
of fludrocortisone acetate in patients with subarachnoid
hemorrhage. Stroke 1989;20:1156-1161
9. Teasdale GM, Jennett B: Assessment of coma and impaired
consciousness. A practical scale. Lancet 1974;2:81-84
10. Hijdra A, van Gijn J, Nagelkerke NJD, Vermeulen M, van
Crevel H: Prediction of delayed cerebral ischemia, rebleed-
ing, and outcome after aneurysmal subarachnoid hemor-
rhage. Stroke 1988;19:1250-1256
11. Vermeulen M, van Gijn J, Hijdra A, van Crevel H: Causes of
acute deterioration in patients with a ruptured intracranial
aneurysm. / Neurosurg 1984;60:935-939
12. Wijdicks EFM, Vermeulen M, Hijdra A, van Gijn J: Hypona-
tremia and cerebral infarction in patients with ruptured
intracranial aneurysm: Is fluid restriction harmful? Ann
Neurol 1985;17:137-140
13. Mullan S, Raimondi AJ, Dobben G, Vailati G, Hekmatpanah
J: Electrically induced thrombosis in intracranial aneurysms.
J Neurosurg 1965;22:539-547
14. Patterson RH, Harpel P: The effect of epsilon amino caproic
acid and tranexamic acid on thrombus size and strength in a
simulated arterial aneurysm. J Neurosurg 1971;34:365-371
15. Meyer CHA, Lowe D, Meyer M, Richardson PL, Neil-
Dwyer G: Progressive change in cerebral blood flow during
the first three weeks after subarachnoid hemorrhage. Neu-
rosurgery 1983; 12:58-76
16. Kanamaru K, Waga S, Kojima T, Fujimoto K, Itoh H:
Endothelium-dependent relaxation of canine basilar arteries.
Part 1: Difference between acetylcholine- and A23187-
 Wijdicks et al Short-term TEA in SAH 1679

induced relaxation and involvement of lipoxygenase metab-
olite(s). Stroke 1987;18:932-937
17. Kanamaru K, Waga S, Kojima T, Fujimoto K, Niwa S:
Endothelium-dependent relaxation of canine basilar arteries.
Part 2: Inhibition by hemoglobin and cerebrospinal fluid
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017
from patients with aneurysmal subarachnoid hemorrhage,
Stroke 1987;18:938-943
KEY WORDS • antifibrinolytic agents • cerebral infarction •
subarachnoid hemorrhage
 Short-term tranexamic acid treatment in aneurysmal subarachnoid hemorrhage.
E F Wijdicks, D Hasan, K W Lindsay, P J Brouwers, R Hatfield, G D Murray, J van Gijn and M
Vermeulen

Stroke. 1989;20:1674-1679
doi: 10.1161/01.STR.20.12.1674
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Downloaded from http://stroke.ahajournals.org/ by guest on May 10, 2017

Copyright © 1989 American Heart Association, Inc. All rights reserved.

Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/20/12/1674

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:

http://stroke.ahajournals.org//subscriptions/