Original Paper
The Clinical Safety of High-Dose Piracetam -
I t s Use in the Treatment of Acute Stroke
1. De Reuckl, B. Van Vleymen2
I Department of Neurology. University Hospital. Ghent. Belgium
International Development. UCB Pharma, Andertecht. Belgium
R Recent post-marketing surveillance reports have con-
firmed the benign safety profile and lack of organ toxicity
shown by piracetam during i t s 25 years of clinical usage. Toler-
ance has proved equally good with the more recent use of larg-
er doses (up to 24glday) for the long-term control of cortical
myoclonus and when given intravenously to patients with
acute stroke. This paper provides a brief review of these find-
ings and records the safety of piracetam as found in the Pirace-
tam in Acute Stroke Study (PASS). a randomized multicenter
placebo-controlled study in 927 patients with acute ischemic
stroke. Patients receive one intravenous bolus injection of pla-
cebo or 12 g piracetam, piracetam 12 g daily for 4 weeks and
maintenance treatment for 8 weeks. The major results have
been reported (De Deyn et al., Stroke 28 I19971 2347-2352).
Safety was assessed taking into account adverse events includ-
ing abnormal laboratory test results and mortality. Death
within 12 weeks occurred more frequently in the piracetam
group but the difference from placebo was not significant. Of
many potential risk, prognostic and treatment-related factors
examined by logistic regression, 6 contributed significantly to
death of which the most important were initial severity of
stroke and age. Neither treatment nor any treatment-related
factor contributed significantly to death. Adverse events were
similar in frequency, type and severity in piracetam and place-
bo groups. Events of cerebral, non-cerebral and uncertain ori-
gin likewise occurred with similar frequency. Few patients dis-
continued because of adverse events. There was no difference
between treatments in the frequency of events associated with
bleeding, including hemorrhagic transformation of infarction.
An important finding was that, of 31 patients with primary
hemorrhagic stroke enrolled, 3 plracetarn-treated patients died
compared with 6 on placebo. The results suggest that pirace-
tam in high dosage may be given to patients with acute stroke
without significant adverse effects.
Introduction
During its 25 years of clinical use, piracetam has consistently
shown a benign safety profile (Abuhazzabab et a].. 1978: Oos-
terveld. 1980; Reisberg et al., 1982; Delaere. 1997a). Adverse ef-
fects have almost always been mild and an infrequent reason
Pharmacopsychiat. 1999; 32 (Supplement): 33-37
Q Ceorg Thieme Verlag Stuttgart .New York
ISSN 0176-3679
33
for discontinuation of treatment. Consistent with the results
of animal studies which have shown a wide margin of safety
Downloaded by: University of British Columbia. Copyrighted material.
(UCB Internal Report. 1967). there has been no evidence of hu-
man organ toxicity. Post-marketing surveillance has con-
firmed the benign profile of piracetam and shown that adverse
events occur only slightly more often than during placebo ad-
ministration (Delaere, 1997b).
These observations are pertinent to the more recent use of
larger doses of piracetam (up to 24 g/day), with similarly good
acceptance, in the long-term treatment of cortical myoclonus
(Brown et a!., 1993; Koskiniemi et al.. 1998) and of their intra-
venous administration to patients with acute stroke (Herr-
schaft, 1988; Platt et al., 1992).
Most treatments under recent or current evaluation for the
treatment of acute stroke interfere with different pathophysio-
logical mechanisms involved in i t s pathogenesis. The need to
demonstrate an acceptable balance between efficacy and safe-
ty in randomized trials remains a major concern (Wahlgren,
1997).
In addition to earlier reports that piracetam may improve pa-
tients with acute stroke (Herrschaft, 1988: Platt et a!.. 1992).
the methodology and results of the Piracetam Acute Stroke
Study (PASS). a multicenter. randomized, double-blind, place-
bo-controlled trial in 927 patients from 10 European countries,
have been published (De Deyn et al., 1997). Neurologic out-
come was assessed by the Orgogozo scale (Orgogozo and Dar-
tigues. 1986) and function by the Barthel Index (Wade and
Hewer. 1987). There was no significant improvement in pirace-
tam-treated patients in neurologic or functional status when
treatment was begun within 12 hours of stroke onset but a
post hoc analysis i n a large subgroup of patients with moder-
ate and severe stroke treated within 7 hours demonstrated a
better outcome after piracetam. A further trial is in progress
which aims to confirm these findings (PASS 11).
The purpose of this paper is to provide a detailed record of the
safety of piracetam as found in PASS.
Patients and Methods
Briefly, patients aged 40- 85 years with a clinical diagnosis of
acute ischemic supratentorial stroke received an intravenous
bolus of 12 g piracetam within 12 hours of stroke onset. 12g
34 Pharrnacopsychiat. 1999; 32 (Supplement)
daily for 4 weeks and maintenance dosage with 4.8 g daily for
up to 12 weeks after the stroke.
Only patients whose initial Orgogozo scale scores were be-
tween 10 and 65 were included, so that those in whom strokes
were very severe and very mild were excluded. Computed to-
mography within 24 hours of admission, repeated within 8
days if necessary, was required to confirm the diagnosis and al-
low exclusion of patients with cerebral hemorrhage, infraten-
torial lesions or a significant mass effect. Aspirin or other anti-
platelet drugs were permitted as concomitant therapy (al-
though starting medication was not recommended during the
first few days). Also allowed were heparin or low-molecular-
weight heparin in low dosage for the prevention of deep vein
thrombosis and in full dosage to prevent cardiac embolism.
calcium antagonists for hypertension or ischemic heart dis-
ease and osmotic therapy for a space-occupying lesion.
Safety was assessed taking into account adverse events, in-
cluding abnormal laboratory test results, and mortality. All ad-
verse events encountered, regardless of cause, were reported
on a separate dedicated page of the case report form on which
investigators recorded symptoms or the presumptive diagno-
sis and indicated any possible relationship to study medica-
tion. Adverse events were coded using the WHO classification
and terminology (WHO, 1989).
All serious adverse events were to be reported to the study
monitor for immediate transmission (within 24 hours) to the
company drug safety department. Those adverse events con-
sidered serious were death, rehospitalization or prolonged
hospitalization, other life-threatening conditions not resulting
in the above, any congenital anomaly, the occurrence of malig-
nancy or one of the adverse events classified as critical by
WHO.
Steering and Safety Committees supervised the study. Mem-
bers of the latter were otherwise unconnected with the trial
and their functions included revjew of deaths and serious ad-
verse events and, termination of the project in the event of un-
toward findings.
Statistics
All analyses reported were based on the intention to treat and
thus included data from all randomized patients including
those subsequently excluded because they failed to meet all
inclusion criteria, e.g. inappropriate CT scan findings. All anal-
yses of Orgogozo scale scores were performed after stratifica-
tion to control for baseline values in order to minimize imbal-
ance in the initial severity of stroke between treatment groups.
Deaths were scored using the last recorded Orgogozo scale val-
ue. Mortality rates were compared and risk rates for death
were estimated and compared using the Cochran-Mantel-
Haenszel test after controlling for baseline Orgogozo scale
scores as described above. Logistic regression analysis was per-
formed to determine which of many potential factors predict-
ed mortality. Adverse events in the two treatment groups were
compared using the Chi square test.
1. De Reuck. B. Van Vleymen
Results
Of 927 patients randomised. all of whom received treatment.
464 were treated with piracetam and 463 with placebo. All
were assessed for safety. Initial neurologic deficit determined
by the mean values of the Orgogozo scale, blood pressure and
risk factors for stroke present on admission were similar in the
two groups.
Mortality
Death within 12 weeks occurred more often in the piracetam
group (23.9%;111/464) as compared with placebo (19.2%;891
463). The difference was not significant ( P = 0.15)after control-
ling for initial Orgogozo scale scores which was done to mini-
mize differences between treatment groups in the initial se-
verity of stroke. Of many potential risk, prognostic and treat-
ment-related factors examined by logistic regression. 6 con-
tributed significantly to death (Table 1)of which the most im-
portant were initial severity of stroke and age. With an initial
Downloaded by: University of British Columbia. Copyrighted material.
Orgogozo scale value below the median (35 in each treatment
group), the conditional relative risk for mortality was 6.01
(95%CI. 3.25 to 11.09) and was 3.51 (95%C1.1.89 to 6.51) with
age 72 years or more. Neither treatment nor any treatrnent-
related factor, including time to start of treatment and interac-
tion between treatment and any other factor, contributed sig-
nificantly to death.
Table 1 Results of logistic regression analysis of factors influencing
mortality
Baseline Factor Conditional 95%
Relative Risk Confidence
for Survival Intervals
- Severity of stroke
- Orgogozo scale 2 3 5 7 6.01 (3.25- 1 1.09)
- Glasgow Coma scale > 12' 1.72 (1.20 -2.46)
- Age<72 3.51 (1.89-6.51)
- Female sex 1.79 (1.26 - 2.54)
- Myocardial infarction 1.65 (1.06-2.57)
- Diabetes 1.6 (1.08- 2.38)
Median values
f An interaction between age and Orgogozo scale score was present. The corn-
bined effects of age 272 and Orgogozo scale < 35 were more than additive.
Although the mean baseline Orgogozo score was similar in the
two treatment groups (piracetam 37.2. placebo 38.6). the ini-
tial stroke was more severe in that the baseline Orgogozo scale
score was below the median value of 35 in 214 piracetam-
treated patients compared to 195 in the placebo group, a dif-
ference of 19. In these patients age distribution and mortality
(piracetam 36%[77/214]: placebo 33%[64/195)) were similar
in each group.
Adverse events
The adverse event profile was similar in both treatment groups
with no significant differences between them in the numbers
of patients with adverse events, the frequency of adverse
events overall or of those considered serious (Table 2). Nor
did the types of adverse event differ between groups. When
The Clinical Safety o f Hiqh-Dose Piracetam Pharmacopsychiat. 1999; 32 (Supplement) 'hs
Table 2 Overall summary of adverse events Table 3 Events of cerebral. non-cerebral or uncertain origin

Patients with adverse events Placebo Piracetam Cerebral Non-cerebral Uncertain origin
(n = 463) (n = 464)
Placebo Piracetam Placebo Piracetam Placebo Piracetam
Number (%of patients) a) All adverse events
- with adverse events 363 (78.4) 358 (77.2) (n = 593) (n = 623) (n = 30)
- with serious adverse events 222 (47.9) 223 (48.1) 305 288 846 777 15 15
- withdrawn due to adverse events 6 (1.3) 9 (2.0) b) Serious adverse events
(n = 274) (n = 388) (n = 15)
135 139 189 199 10 5

Table 4 Adverse events of cerebral origin Table 5 All hemorrhagic adverse events

Event Placebo Piracetam Placebo Piracetam

(n = 305) (n = 288) (n = 30) (n = 38)

Hemorrhagic transformation of infarction 16 17 Castro-intestinal hemorrhage 7 10

Condition aggravated 32 37 Hemorrhagic transformation of infarction 16 17
Cerebral edema 14 19 Retinal hemorrhage 0 2

Downloaded by: University of British Columbia. Copyrighted material.

Recurrent stroke 15 18 Hematuria 0 6
Stuporlcoma 9 8 Hematoma (cerebral and non-cerebral) 5 1
Seizures 16 5 Abnormal platelet count or coagulation defect 2 2
Muscle contractions 4 3
Headache 27 19
Psychiatric 158 148 cannot be defined more precisely. Seizures were reported in 5
Others 14 14
piracetam-treated and 16 placebo-treated patients.

Adverse events associated w i t h bleeding were analysed in de-

adverse events, both i n total and those labeled serious, could
be classified as o f cerebral, non-cerebral o r uncertain origin.
distribution was similar i n piracetam- and placebo-treated pa-
tients (Table 3). For serjous adverse events, in the piracetam
group 139 of 343 events were ofcerebral origin compared w i t h
135 o f 334 in the placebo group. Nine piracetam-treated and 6
placebo-treated patients discontinued treatment because of
adverse events, all considered either unrelated or unlikely to
be due t o study medication.
When individual adverse events o f cerebral origin were con-
sidered (Table 4). the frequency and nature o f each event o r
related clusters o f events were similar in both piracetam and
placebo groups. Thus hemorrhagic transformation o f infarc-
tion was observed in 17 piracetam-treated and 16 placebo-
treated patients. The events reported according r o W H O termi-
nology as "condition aggravated". "cerebral oedema" and "ce-
rebrovascular disorder'' are taken t o mean progression or re-
currence o f stroke. The terms are not mutually exclusive b u t
tail and are listed in Table 5. Of 33 patients w i t h hemorrhagic
transformation o f infarction, 17 were receiving piracetam (of
w h o m 4 died) compared w i t h 16 o n placebo (of w h o m 3 died).
Other hemorrhagic events occurred w i t h similar frequency in
the t w o groups.
Concurrent ontithrombotic medication
The numbers of patients receiving aspirin o r heparin. either i n
l o w or high dosage, were similar in b o t h treatment groups (Ta-
bles 6 and 7 ) .Outcome i n patients receiving heparin is shown
i n Table 6 and in those o n aspirin inTable 7. Mortality rate was
l o w i n both piracetam and placebo groups i n patients receiving
concurrent aspirin.
Patients with primary hemorrhagic stroke
Enrolled i n the trial and included in the intention-to-treat
analysis were 31 patients w i r h prjmary cerebral hemorrhage
o n initial CTscan (Table 8). Although the initial stroke was sub-

Table 6 Outcome in patients receiving concomitant heparin

Placebo Piracetam
(n = 463) (n = 464)
Heparin Heporin
None Low-dose High-dose None Low-dose High-dose
(n=394) (n=49) (n=20) (n = 389) (n = 52) (n = 23)

Age 70.9 74.8 60.9 70.1 69.3 68.7

Time to treatment (h) 7 7.9 8 7 7.7 7.3
Mean Orgogozo score at baseline 38 42.9 39.8
Mean Orgogozo score at 4 weeks 58.1 55.7 52.3
Mortality a t 12 weeks (%) 19 22 25 25 21 13
36 Pharmacopsychiat. 1999; 32 (Supplement) I. De Reuck. B. Van Vlevmen

Table 7 utcome in patients receiving concomitant aspirin within 24 Several findings in the present study have particular relevance
hours of stroke to the treatment of acute stroke. There was no increase in the
frequency of hemorrhagic transformation of infarction or in
Placebo Piracetam bleeding from other sites. This is important both on general
(n = 463) (n = 464)
No aspirin aspirin No aspirin aspirin grounds and because piracetam has been reported to possess
- -
( n 395) (n 68) -
(n 406) (n = 58) antithrombotic (Stockmans et al., 1998) and rheologic effects
(Nalbandian et al., 1983; Crotemeyer e t a]., 1986; Moriau et
Age 70.1 70 70.4 66.4 al., 1993).
Onset of treatment (h) 7.1 7 7 7.3
Mean Orgogozo score 37.6 44.6 36.3 43.1 Of a t least equal importance is the finding that piracetam did
a t baseline not harm those patients with primary hemorrhagic stroke ini-
Mean Orgogozo score 56.1 62 57.2 61.6 tially included in the trial. Although the initial stroke was sub-
a t 4 weeks stantially more severe in the piracetam group. the mortality
Mortality at 12 weeks (%) 20 13 26 7 rate was somewhat lower in these patients compared to those
given placebo. This important finding provides evidence that
piracetam may be administered in acute stroke prior to hospi-
tal admission and CT scan.
Table 8 Patients with cerebral hemorrhage on initial CTscan
Placebo Piracetam Seizures were fewer in the piracetam group as compared with
placebo. Although the numbers were small this finding is in

Downloaded by: University of British Columbia. Copyrighted material.

(n = 463) (n = 464)
accord with the well-documented antimyoclonic effects of pi-
Patients with evidence of cerebral 16 15 racetam and its beneficial effects in patients with progressive
haemorrhage on initial CTscan myoclonus epilepsy (Koskiniemi et al., 1998). Piracetam has
Orgogozo scale score. Mean (SD) been reported to possess anticonvulsant properties when giv-
Baseline 38.4 (14.3) 28.7 (1 1 .l) en in combination with antiepileptic drugs in animals (Trausch
4 weeks 40.3 (26.4) 29.3 (12.8) and I<eller,1993) and man (Monadori and Schmutz, 1986).
1 2 weeks 46.3 (30.6) 39.3 (26.0)
Mortality Patients dead N (%) The low mortality rate in patients receiving aspirin in both
4 weeks 5 (31.3) 3 (20.0) treatment groups is of interest. Although the initial stroke
12 weeks 6 (37.5) 3 (20.0) was less severe in these patients and their number small, the
findings are in accord with the results of supplementary anal-
yses reported by the International Stroke Trial Collaborative
stantially more severe in the 15 piracetam-treated patients Group (1997) which consisted of a reduction of approximately
than in the 16 receiving placebo (mean Orgogozo scale scores: 10 per 1000 deaths or recurrent strokes in the early weeks.
piracetam 28.7; placebo 38.4). only 3 piracetam-treated pa-
tients died within 12 weeks compared with 6 on placebo. In conclusion, w e have shown the use of piracetam in high
dosage given to patients with acute stroke to be without signi-
Discussion ficant adverse effects. Hemorrhagic transformation of infarc-
tion was no more frequent in piracetam-treated patients than
Piracetam in a dose of 12 g and placebo, given as an intrave- in the placebo group. That piracetam had no adverse effect on
nous bolus injection within 12 hours of the onset of stroke patients with primary hemorrhagic stroke indicates that it
and then in similarly large daily doses for 4 weeks, showed may be suitable for acute administration prior to hospital ad-
closely comparable tolerability profiles. There was no signifi- mission and CT scan.
cant difference in mortality between pjracetam and placebo
groups. References

The present study provides the first documentation of the
safety profile of piracetam given for acute stroke, as an initial
bolus injection and subsequently in high dosage for 4 weeks.
The findings are important both because they further confirm
the unusually good toleration of piracetam and because of
their relevance for the potential use of piracetam in the treat-
ment of acute stroke.
lnitial severity of stroke and age are the strongest predictors of
mortality after stroke (Grotta, 1988). We confirmed this using
logistic regression and also found no significant correlation be-
tween any treatment-related factor and death. The greater
number of deaths in piracetam-treated patients was not statis-
tically significant and occurred largely because the initial
stroke was more severe in more patients in the piracetam
group than on placebo. It is most unlikely that piracetam influ-
ences mortality in acute stroke.
Abuhazzabab Sr. FS, Merwin CE, Zimmermann RL. Sherman MC: A
double-blind investigation of piracetam (Nootropil) in the mem-
ory of geriatric inpatients. Psychopharmacol Bull 1978; 14: 23 -
26.
Brown P. Steiger MJ, Thompson PD, Rothwell JC, Day BL, Salina M.
Waegemans T, Marsden CD: Effectiveness of piracetam in corti-
cal myoclonus. Mov Disord 1993; 8: 63 - 68.
Delaere A: Piracetam cumulative safety report. April 1971 -April
1997. UCB Internal Report, August 1997 (1997a).
Delaere A: Piracetam safety profile in double-blind studies. UCB In-
ternal Report, September 1997 (1997b).
De Deyn PP, De Reuck J , Deberdt W, Vlietinck R Orgogozo J-M:
Treatment of acute ischemic stroke with piracetam. Stroke
1997; 28: 2347-2352.
Grotemeyer I<H, Hofferberth G, Hirschberg M: Normalisierung hy-
perreaktiver Thrombozyten bei Patienten mit TIAs unter Pirace-
tam? Nervenarzt 1986; 57(3): 180- 183.
The
- -
Clinical Safety of High-Dose Piracetam Pharmacopsychiat. 1999; 32 (Supplement)

Grotta JC: Post-stroke management concernsand outcomes. Geriat- Corresponding Author

rics 1988; 43: 40 -47.
Herrschaft H: Die Wirksamkeit von Piracetam bei der akuten zere-
bralen [schamie des Menschen. Klinisch kontrollierte Doppel- Department of Neurology
blindstudie Piracetam/lO% Dextran 40 versus 10%Dextran 401 University Hospital
Placebo. Med I<lin 1988: 83: 667-677. 8-9000 Ghent
International Stroke Trial Collaborative Group: The International Belgum
Stroke Trial (In): a randomised trial of aspirin. subcutaneous
heparin, both or neither among 19435 patients with acute
ischaemic strolie. Lancet 1997: 349: 1569-1581.
Koskiniemi M. Van Vleymen B. Hakarnies L, Lamusuo S, Taalas 1: Pi-
racetam relieves symptoms in progressive myoclonus epilepsy: a
multicentre, randomised, double blind, crossover study compar-
ing rhe efficacy and safety of three dosages of oral piracetam
with placebo. J Neurol Neurosurg Psychiatry 1998: 64: 344 -348.
Moriau M, Crasborn L, Lavenne-Pardonge E, Von Frenclcell R, Col-
Debeys C: Platelet anti-aggregant and rheoIogical properties of
piracetam. A pharmacodynamic study in normal subjects. A n -
neimittelforschung 1993; 43(1)2: 110- 118.
Monadori C, Schmutz M: Synergisric effects of oxiracetam and pira-
cetam in combinarion with antiepileptic drugs. Acta Neurol

Downloaded by: University of British Columbia. Copyrighted material.

Scand 1986: 74 (Suppl): 113- 116.
Nalbandian RM, Henry RL, Burek CZ, Diglio CA. Goldman AL, Taylor
GW, Hoffman WH: Diminished adherence of sickled erythro-
cytes to cultured vascular endothelium by piracetam. Amer J
Hematol. 1983: 15: 147-151.
Oosterveld WJ: The efficacy of piracetam in vertigo. A double-blind
study in patients with vertigo of central origin. Arzneimittelfor-
schung 1980; 30(11)/11: 1947-1949.
Orgogozo J-M, Dartigues JF: Clinical trials in brain infarction. In:
Battistini N, Fiorani P. Courbier R Plum F, Fieschi C (eds.): Acute
Brain Ischemia: Medical and Surgical Therapy. Raven Press. New
York 1986; 32: 201 - 208.
Platt D. Horn J. Summa J-D. Schmitt-Riith B. Reintern B. [<aunt2J.
Kronert E: Zur Wirksamkeit von Piracetam bei geriatrischen Pa-
tienten mit akuter zerebraler Ischamie. Eine klinisch kontrol-
lierte Doppelblindstudie. Medwelt 1992: 43: 181 -190.
Reisberg B. Ferris SH, Schneck M[<.Corwin J. Mir P. Friedrnan E. Sher-
man KA, McCarthy M, Bartus RT: Piracetarn in the treatment of
cognitive impairment in the elderly. Drug Devel Res 1982; 2:
475 - 480.
Stockmans F, Deberdt W, Nystrorn A, NystrBm E, Stassen JM, Vermy-
len J, Deckmijn H : Inhibitory effect of piracetam on platelet-rich
thrombus formation in an animal model. Thromb Haemost
1998; 7911 : 222 - 227.
Trausch 3, Keller G: The effects of the noocmpics piracetam and
meclofenoxate alone or in combination with antiepileptic drugs
on acute pentylenetetratole induced seizures in rats. Arch Phar-
macol 1993; 347 (Suppl 139): Abs 556.
UCB Internal Report: ucb 6215: Oral toxicity studies in pedegree
dogs. 1967.
Wade DT, Hewer RL: Functional abilities after stroke: measurement.
natural history and prognosis. J Neurol Neurosurg Psychiatry
1987; 50: 177 - 182.
Wahlgren NG: Pharmacological treatment of acute stroke. Cerebro-
vasc Dis 1997; 7 (Suppl3): 27 - 30.
WHO: Uppsala Monitoring Center. International monitoring of ad-
verse reactions to drugs. Adverse reaction terminology. Decem-
ber 1989.