Original Paper
Cerebrovasc Dis 2006;21:380–385 Received: September 21, 2005
DOI: 10.1159/000091547
Citicoline in Intracerebral Haemorrhage:
A Double-Blind, Randomized, Placebo-
Controlled, Multi-Centre Pilot Study
Julio J. Secades a José Álvarez-Sabín b Francisco Rubio c Rafael Lozano a
Antoni Dávalos d José Castillo e for the Trial Investigators1
a
Medical Department, Grupo Ferrer SA, b Department of Neurology, Hospitals de la Vall d’Hebrón, and
c
Department of Neurology, Hospital de Bellvitge, Barcelona, d Department of Neurosciences,
Hospital Universitari Germans Trias i Pujol, Badalona, and e Department of Neurology,
Complejo Hospitalario Universitario, Santiago de Compostela, Spain
Key Words
Intracerebral haemorrhage
Neuroprotection

Citicoline
Abstract
Background: In experimental models citicoline has
shown beneficial effects in intracerebral haemorrhage.
Citicoline is a neuroprotectant drug with some beneficial
effects in human ischaemic stroke and with an excellent
safety profile. We decided to carry out a pilot study to
test its safety and efficacy in human intracerebral haem-
orrhaging. Methods: In this double-blind, placebo-con-
trolled pilot study, patients had to be previously indepen-
dent, aged between 40 and 85 years, and had to be
admitted within 6 h after onset of symptoms of an acute
primary supratentorial hemispheric cerebral haemor-
rhage diagnosed by neuroimaging (CT or MRI). Baseline
1
List of centres and investigators: Complejo Hospitalario Universitario, San-
tiago de Compostela (Spain): J. Castillo, R. Leira, M. Blanco, M. Rodríguez-
Yáñez; Hospitals de la Vall d’Hebrón, Barcelona (Spain): J. Álvarez-Sabín,
J. Montaner, C. Molina, P. Delgado, E. Santamarina, R. Huertas, F. Purroy;
Hospital Universitari Dr. Trueta, Girona (Spain): A. Dávalos, J. Serena, M.
Castellanos, Y. Silva; Hospital de Bellvitge, Barcelona (Spain): F. Rubio.
© 2006 S. Karger AG, Basel
1015–9770/06/0216–0380$23.50/0
Fax +41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/ced
Accepted: December 6, 2005
Published online: February 15, 2006
severity was defined as patients with a score larger than
8 points on the Glasgow Coma Scale and larger than 7
on the National Institutes of Health Stroke Scale. Patients
received either a placebo or 1 g/12 h citicoline for 2 weeks
(orally or intravenously). The primary aim was to evalu-
ate safety with respect to the number of adverse events
that occurred. The efficacy endpoint was the percentage
of patients with a modified Rankin Score (mRS) at 3
months. Results: 19 patients in each group were includ-
ed in the study. The incidence of serious adverse events
was not different among groups (4 patients in each
group). One patient in the placebo group was catego-
rised as independent (mRS ^ 2) in comparison with 5
patients in the citicoline group (OR, 5.38; 95% CI, 0.55–
52). Conclusions: Citicoline seems to be a safe drug in
human intracerebral haemorrhage with a positive trend
regarding efficacy. These data should be confirmed in a
larger trial.
Copyright © 2006 S. Karger AG, Basel
Introduction
Intracerebral haemorrhage (ICH) has a worse progno-
sis than cerebral infarction [1]. In general, the treatment
of ICH is controversial and the role of surgery still re-
130.240.43.43 - 8/9/2013 3:40:45 PM
Julio J. Secades
Lulea Tekniska Universitet
Medical Department, Ferrer Grupo SA
Av. Diagonal 549 5ª planta
ES–08029 Barcelona (Spain)
Downloaded by:
Tel. +34 936 003 837, Fax +34 934 907 078, E-Mail jsecades@ferrergrupo.com
mains unclear [1, 2]. There is some evidence that brain
ischaemia plays a role in the secondary brain injury seen
in some experimental models of the ICH [3], but the find-
ings are controversial in humans, at least in the acute
phase [4–6]. Neurological deterioration occurs in 23% of
ICH cases, and its presence results in a worse prognosis
[7]. The main contributor to early neurological worsening
is haematoma growth, but cell death and brain oedema
in the periphery of the ICH may play an important role.
Animal models have shown that brain injury in the pe-
riphery of the haematoma is due to a vascular compres-
sion, a local inflammatory reaction, and the release of
vasoactive substances [8, 9]. In clinical practice, inflam-
mation and high plasmatic levels of ischaemia biomark-
ers have been associated with poor outcome [10].
Citicoline is a neuroprotectant drug with positive ef-
fects in experimental models [11] and in the treatment of
the acute phase of cerebral ischaemia [12, 13], showing,
in some cases, a significant reduction in the volume of
cerebral infarction [14]. Citicoline also has a positive ef-
fect on brain oedema [11] and in ICH models [15, 16]. In
these models, citicoline reduced the volume of the isch-
aemic lesion associated with the haematoma.
Based on this evidence, we decided to carry out a pilot
study to test the safety and trends on the efficacy of citi-
coline in human ICH.
Subjects and Methods
Study Design
This was a multi-centre, double-blind, randomised, placebo-
controlled pilot study conducted in four Spanish university hospi-
tals. The first patient was admitted in January 2001 and the last in
July 2003. Patients were randomly assigned to citicoline or to a
placebo. Being a pilot study, no sample size calculation was made.
The trial received approval from independent ethics committees
and from the Spanish health authorities, and was conducted in ac-
cordance with Guidelines for Good Clinical Practice, and the Dec-
laration of Helsinki.
Patients
Patients had to be previously independent, aged between 40 and
85 years, and had to be admitted within 6 h after the onset of symp-
toms of an acute supratentorial hemispheric cerebral haemorrhage
diagnosed by neuroimaging (CT or MRI). Baseline severity was
defined as patients with a score larger than 8 points on the Glasgow
Coma Scale and larger than 7 on the National Institutes of Health
Stroke Scale (NIHSS), with at least 2 points coming from motor
deficits. Written informed consent was mandatory. Exclusion cri-
teria were as follows: intraventricular or subarachnoid haemor-
rhages, haemorrhages secondary to anticoagulant therapy or to sub-
jacent pathologies (such as arteriovenous malformations, angioma,
brain tumours, …), coma (Glasgow Coma Scale less than or equal
Citicoline in Intracerebral Haemorrhage
to 8), no motor deficits in the NIHSS score, patients with INR (in-
ternational normalised ratio) larger than 1.7, any condition or treat-
ment at baseline that would interfere with the efficacy or safety
assessments, and expectancy of life less than 3 months due to co-
morbidity.
Study Treatment
For randomisation a table of random digits was used, creating
random permuted blocks of 2 patients. The randomisation list was
deposited in the pharmacy department of the participating centres.
The pharmacy department was responsible for delivering the blind-
ed treatments. Both, the placebo and the active drug ampoules were
completely indistinguishable to the investigators and patients. As
the investigators were blinded until the end of the study, prediction
to which group the next patient was allocated was not possible.
The treatment schedule was a 1-gram citicoline ampoule or a
placebo every 12 h by continuous intravenous perfusion. If the pa-
tient was able to swallow, the drug was given orally at intervals of
12 h. The total time of treatment was 2 weeks.
Outcome Measures
The primary endpoint was safety. The safety of the treatment
was assessed with respect to the number of adverse events that oc-
curred. Safety assessments included routine haematology and clin-
ical chemistry tests. The efficacy endpoint was the percentage of
independent patients at 3 months, assessed with the modified
Rankin Score (mRS; independent patient was defined as a patient
with an mRS score ^ 2 at 3 months). Secondary efficacy endpoints
included the evolution of neurological deficits assessed with the
NIHSS, and the volume of the residual lesion. NIHSS scores were
measured at baseline, days 2, 7 and 14, and week 12. The mRS
scores were recorded at day 14 and week 12. Neuroimaging data
were available at baseline, week 2 and week 12.
Statistical Analysis
The monitoring of the study, the data management, and the
statistical analysis were conducted by Biometrica (Barcelona,
Spain), an external clinical research organisation.
Statistical analysis was conducted according to the intention-to-
treat principle. The intention-to-treat population was defined as
patients randomized with at least one efficacy evaluation after re-
ceiving at least one medication dose, and fulfilling inclusion and
exclusion criteria of this protocol. If no data were recorded during
analysis at week 12, then data from the most recent visit (last ob-
servation carried forward) were carried forward.
The descriptive variables and number of adverse events are ex-
pressed as a percentage in function of the treatment group, and all
comparisons with these variables were conducted using the
2
test.
Previously, the descriptive and the clinical variables were ana-
lysed at baseline to test the homogeneity between groups.
For the analysis of ordinal variables we used non-parametric
statistical tests. In the comparisons between groups we used the
Mann-Whitney U test for independent data. For the analysis of the
evolution in each group we used the Wilcoxon test.
The p values presented in this work were two-tailed. Values of
p ! 0.05 were considered significant.
130.240.43.43 - 8/9/2013 3:40:45 PM
Cerebrovasc Dis 2006;21:380–385 381
Lulea Tekniska Universitet
Downloaded by:
Table 1. Demographic data, risk factors and baseline characteris- Table 2. Study completion status
tics
Status Placebo Citicoline
Placebo Citicoline (n = 19) (n = 19)
(n = 19) (n = 19)
Patients completed 12 (63.2) 15 (78.9)
Sex Patients discontinued 7 (36.8) 4 (21.1)
Male 8 (42.1) 11 (57.9) Reasons for discontinuation
Female 11 (57.9) 8 (42.1) Patient request 2 (10.5) 0 (0)
Age (mean 8 SD), years 67.0813.1 74.5812.0 Change of hospital 1 (5.3) 1 (5.3)
Weight (mean 8 SD), kg 66.487.2 69.889.7 Death 2 (10.5) 2 (10.5)
Height (mean 8 SD), cm 161.783.8 165.887.6 Intercurrent disease 1 (5.3) 0 (0)
Therapeutic window (mean 8 SD), h 4.781.0 4.581.7 Forbidden medication 1 (5.3) 0 (0)
Baseline NIHSS Other1 0 (0) 1 (5.3)
Mean 8 SD 13.786.7 10.685.2
Median 15 9 Figures in parentheses are percentages.
1
Baseline GCS (mean8SD) 13.881.5 13.981.7 Not specified in the CRF.
Haematoma localization
Right hemisphere 8 (42.1) 5 (26.3)
Left hemisphere 11 (57.9) 14 (73.7)
Haematoma topography
Caudate 0 (0) 1 (5.3)
Table 3. Number of adverse events recorded during the trial
Putamen 9 (47.4) 10 (52.6)
Thalamus 7 (36.8) 5 (26.3)
Other 3 (15.8) 3 (15.8) Adverse event Placebo Citicoline
Haematoma volume, mean 8 SD, cm3 24.1823.5 24.189.7
Risk factors Dyspepsia 1 (5.3) –
Smoking 4 (21.1) 2 (10.5) Gastric bleeding – 1 (5.3)
Alcohol 5 (26.3) 5 (26.3) Pancreatitis 1 (5.3) –
High blood pressure 10 (52.6) 11 (61.1) Convulsion – 1 (5.3)
Previous ICH 3 (15.8) 1 (5.3) Agitation – 1 (5.3)
Antiplatelet treatment 0 (0) 2 (10.5) Bronchospasm – 1 (5.3)
Cough – 1 (5.3)
Figures in parentheses are percentages; GCS = Glasgow Coma Upper respiratory tract infection – 1 (5.3)
Scale. Dyspnoea – 1 (5.3)
Brain haemorrhage 1 (5.3) 1 (5.3)
Non-haemorrhagic stroke 1 (5.3) –
Urinary tract infection – 1 (5.3)
Pyelonephritis – 1 (5.3)
Thromboembolism 1 (5.3) –
Thrombopenia – 1 (5.3)
Results
Figures in parentheses are percentages. Note that patients could
have had more than one adverse event.
A total of 38 patients were included in the study, 19 in
each group (fig. 1). The patient baseline characteristics
are shown in table 1. There were no differences between
groups with reference to risk factors, concomitant treat-
ments, or concomitant diseases. Table 4. Results in the primary efficacy outcome at weeks 2 and 12
(intention-to-treat analysis)
Study Completion Status
Table 2 summarizes the study completion status of the mRS Placebo Citicoline
patients included in this trial. No differences were found week 2 week 12 week 2 week 12
concerning reasons of study discontinuation between the
groups. Four patients in the placebo group and 1 in the 0–2 1 (6.7) 1 (6.7) 4 (22.2) 5 (27.8)
citicoline group discontinued the study before day 14, so 3–6 14 (93.3) 14 (93.3) 14 (77.8) 13 (72.2)
there is no data for the efficacy endpoint (mRS) to make Figures in parentheses are percentages.
the last observation carried forward.
130.240.43.43 - 8/9/2013 3:40:45 PM

382 Cerebrovasc Dis 2006;21:380–385 Secades /Álvarez-Sabín /Rubio /Lozano /

Lulea Tekniska Universitet

Dávalos /Castillo
Downloaded by:
Fig. 1. Flow diagram of the trial.

Safety Analysis
Four patients in each group exhibited adverse events
during the study. Table 3 shows a description of adverse
events recorded during the trial. None of them were
related to the treatment by the investigators. There
were no statistically significant differences between
groups in the frequency of adverse events as well as
changes in vital signs and laboratory tests during the
study. The mortality rate was similar in both groups,
with 2 deaths in each group during the study. In addi-
tion, 1 brain haemorrhage was reported as an adverse
event in each group, which resulted from re-bleeding of
the same ICH.

Efficacy Analysis
Primary Outcome. Table 4 shows the results by group
according to the primary outcome. One patient in the
placebo group was categorised as having no disability
Fig. 2. Evolution of the lesion volume. Bars represent the SD.
(mRS ^ 2) in comparison with 5 patients in the citicoline
group (OR, 5.38; 95% CI, 0.55–52.4).
Secondary Outcomes. The evolution of the NIHSS
scores changed from 13.2 8 6.5 at baseline to 7.8 8 6.8 evolution of NIHSS scores between baseline and week 12,
at week 12 in the placebo group, and from 10.6 8 6.2 to without differences between groups. The volume of the
5.2 8 5.7 in the citicoline group. There was a significant residual lesion was 6.3 8 10.5 cm3 in the placebo group
improvement in both groups (Friedman; p ! 0.01) in the and 5.8 8 6.8 cm3 in the citicoline group (fig. 2).
130.240.43.43 - 8/9/2013 3:40:45 PM

Citicoline in Intracerebral Haemorrhage Cerebrovasc Dis 2006;21:380–385 383

Lulea Tekniska Universitet
Downloaded by:
 Discussion
Regarding the controversy about the optimal treat-
ment for ICH [2], we can suggest the use of neuroprotec-
tants to ‘protect’ the brain against the ischaemic insult
observed in these cases. We also have to consider that the
pathophysiology of brain injury after ICH shares several
mechanisms with ischaemic injury [3, 8–10, 17]. This
could therefore be an opportunity to contribute to the
treatment of ICH patients using neuroprotectant drugs.
Recently, argatroban, a thrombin inhibitor, has shown
positive effects in reducing secondary brain damage in a
rat ICH model[18]. Up to now, the efficacy and safety of
neuroprotectant drugs has only been tested in patients
with ICH in subgroup analyses from the CLASS study
[19, 20] and from the GAIN studies [21], despite experi-
mental data. In the CLASS study it was demonstrated
that clomethiazole was safe, but there was no evidence of
efficacy due to the small sample size. However, clome-
thiazole failed to demonstrate efficacy in the treatment of
acute ischaemic stroke [20]. In the case of gavestinel, the
results suggest that the drug is not of substantial benefit
or harm to patients with primary ICH [21]. These results
come from 571 patients with spontaneous ICH diagnosed
by baseline CT scan of the head who participated in both
GAIN studies. In addition, the GAIN studies did not
show positive results for acute ischaemic stroke patients
[22, 23]. In our case, we are checking the safety and the
effects of citicoline on the outcome of patients with pri-
mary ICH in a pilot study. Citicoline is a drug with some
evidence of effectiveness in the treatment of acute isch-
aemic stroke [12–14], and is available in some countries
for this condition. In experimental models, citicoline was
shown to improve electroencephalogram changes after
the induction of an ICH in dogs [15]. In addition, in an
ICH mouse model, citicoline-treated animals had a small-
er surrounding volume of ischaemic injury than placebo-
treated animals [16]. Citicoline is a very safe drug in acute
ischaemic stroke patients [13]. We can advance a similar
safety profile of the drug in ICH patients with the safety
data registered in this trial, and with no differences in the
rate of adverse events in comparison with placebo. The
results of this pilot trial, despite limitations concerning
the sample size and statistical significance, are similar to
those seen in other ischaemic stroke studies. Our data
shows a positive trend in favour of citicoline in compari-
son with placebo as regards the final outcome. If we con-
duct a sensitivity analysis, including all the drop-outs be-
fore the second week and considering the worst scenario,
the results are more positive when regarding the active
treatment.
In conclusion, citicoline seems to be a safe drug when
used in ICH patients. The efficacy for this condition
should be confirmed in a larger clinical trial.
Acknowledgement
We thank Erik Cobo and Francesc Miras, from the Department
of Statistics and Operational Research of the UPC of Barcelona for
their statistical advice.

References

1 Kase CS: Intracerebral hemorrhage. Baillieres
Clin Neurol 1995;4:247–278.
2 Mendelow AD, Gregson BA, Fernandes HM,
Murray GD, Teasdale GM, Hope DT, Karimi
A, Shaw MDM, Barer DH: Early surgery ver-
sus conservative treatment in patients with
spontaneous supratentorial intracerebral hae-
matomas in the International Surgical Trial in
Intracerebral Haemorrhage (STICH): a ran-
domised trial. Lancet 2005;365:387–397.
3 Gong C, Boulis N, Qian J, Turner DE, Hoff JT,
Keep RF: Intracerebral hemorrhage-induced
neuronal death. Neurosurgery 2001; 48: 875–
882.
4 Schellinger PD, Fiebach JB, Hoffmann K,
Becker K, Orakcioglu B, Kollmar R, Juttler E,
Schramm P, Schwab S, Sartor K, Hacke W:
Stroke MRI in intracerebral hemorrhage: is
there a perihemorrhagic penumbra? Stroke
2004;34:1674–1680.
5 Rosand J, Eskey C, Chang Y, Gonzalez RG,
Greenberg SM, Koroshetz WJ: Dynamic sin-
gle-section CT demonstrates reduced cerebral
blood flow in acute intracerebral hemorrhage.
Cerebrovasc Dis 2002;14:214–220.
6 Siddique MS, Fernandes HM, Wooldridge
TD, Fenwick JD, Slomka P, Mendelow AD:
Reversible ischemia around intracerebral
hemorrhage: a single-photon emission com-
puterized tomography study. J Neurosurg
2002;96:736–741.
7 Leira R, Dávalos A, Silva Y, Gil-Peralta A, Te-
jada J, Garcia M, Castillo J: Early neurologic
deterioration in intracerebral hemorrhage: pre-
dictors and associated factors. Neurology
2004;63:461–467.
8 Andaluz N, Zuccarello M, Wagner KR: Exper-
imental animal models of intracerebral hemor-
rhage. Neurosurg Clin N Am 2002; 13: 385–
393.
9 Belayev L, Saul I, Curbelo K, Busto R, Belayev
A, Zhang Y, et al: Experimental intracerebral
hemorrhage in the mouse: histological, behav-
ioral, and hemodynamic characterization of a
double-injection model. Stroke 2003;34:2221–
2227.
10 Castillo J, Dávalos A, Alvarez-Sabín J, Pumar
JM, Leira R, Silva Y, et al: Molecular signa-
tures of brain injury after intracerebral hemor-
rhage. Neurology 2002;58:624–629.
11 Secades JJ: CDP-choline: update and review of
its pharmacology and clinical use. Methods
Find Exp Clin Pharmacol 2002;24(suppl B):1–
53.
12 Saver JL, Wilterdink J: Choline precursors in
acute and subacute human stroke: a meta-anal-
ysis. Stroke 2002;33:353.
130.240.43.43 - 8/9/2013 3:40:45 PM

384 Cerebrovasc Dis 2006;21:380–385 Secades /Álvarez-Sabín /Rubio /Lozano /

Lulea Tekniska Universitet

Dávalos /Castillo
Downloaded by:
13 Dávalos A, Castillo J, Alvarez-Sabín J, Secades
JJ, Mercadal J, Lopez S, Cobo E, Warach S,
Sherman D, Clark WM, Lozano R: Oral citico-
line in acute ischemic stroke: an individual pa-
tient data pooling analysis of clinical trials.
Stroke 2002;33:2850–2857.
14 Warach S, Harnett K: Dose dependent reduc-
tion in infarct growth with citicoline treatment:
evidence of neuroprotection in human stroke?
Stroke 2002;33:354.
15 Kase M, Ono J, Yoshimasu N, Hiyamuta E,
Sugiera K: Effect of CDP-choline in experi-
mental cerebral hemorrhage. Curr Ther Res
Clin Exp 1974;16:483–502.
16 Clark W, Gunion-Rinker L, Lessov N, Hazel
K: Citicoline treatment for experimental intra-
cerebral hemorrhage in mice. Stroke 1998;29:
2136–2140.
17 Fatar M, Stroick M, Griebe M, Hennerici M:
Matrix metalloproteinases in cerebrovascular
diseases. Cerebrovasc Dis 2005;20:141–151.
Citicoline in Intracerebral Haemorrhage
18 Nagatsuna T, Nomura S, Suehiro E, Fujisawa
H, Koizumi H, Suzuki M: Systemic adminis-
tration of argatroban reduces secondary brain
damage in a rat model of intracerebral hemor-
rhage: histopathological assessment. Cerebro-
vasc Dis 2005;19:192–200.
19 Wahlgren NG, Díez-Tejedor E, Teitelbaum J,
Arboix A, Leys D, Ashwood T, Grossman E:
Results in 95 hemorrhagic stroke patients in-
cluded in CLASS, a controlled trial of clome-
thiazole versus placebo in acute stroke pa-
tients. Stroke 2000;31:82–85.
20 Lyden P, Shuaib A, Ng K, Levin K, Atkinson
RP, Rajput A, Wechsler L, Ashwood T, Claes-
son L, Odergren T, Salazar-Grueso E: Clome-
thiazole Acute Stroke Study in ischemic stroke
(CLASS-I): final results. Stroke 2002; 33: 122–
128.
Cerebrovasc Dis 2006;21:380–385
21 Haley EC Jr, Thompson JLP, Levin B, Davis
S, Lees KR, Pittman JG, DeRosa JT, Ordron-
neau P, Brown DL, Sacco RL: Gavestinel does
not improve outcome after acute intracerebral
hemorrhage. An analysis from GAIN Interna-
tional and GAIN Americas studies. Stroke
2005;36:1006–1010.
22 Lees KR, Asplund K, Carolei A, Davis SM,
Diener HC, Kaste M, Orgogozo JM, White-
head J, for the GAIN International Investiga-
tors: Glycine Antagonist (gavestinel) in Neuro-
protection (GAIN International) in patients
with acute stroke. Lancet 2000; 355: 1949–
1954.
23 Sacco RL, DeRosa JT, Haley EC, Levin B, Or-
dronneau P, Phillips SJ, Rundek T, Snipes RG,
Thompson JLP, for the GAIN Americas Inves-
tigators: Glycine antagonist in neuroprotection
for patients with acute stroke. GAIN Americas:
a randomized controlled trial. JAMA 2001;
285:1719–1728.
130.240.43.43 - 8/9/2013 3:40:45 PM
385
Lulea Tekniska Universitet
Downloaded by: