Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/303509170

CiticolineforAcuteIschemicStroke:ASystematicReviewandFormalMeta-analysisofRandomized,Double-Blind,andPlacebo-ControlledTrials

ArticleinJournalofstrokeandcerebrovasculardiseases:theofficialjournalofNationalStrokeAssociation·May2016

DOI:10.1016/j.jstrokecerebrovasdis.2016.04.010

CITATIONS

0

READS

73

7authors,including:

JulioJSecades

FerrerGroupS.A.

46PUBLICATIONS835CITATIONS

SEEPROFILE

JoséAlvarez-Sabín

UniversityHospitalValld'Hebron

482PUBLICATIONS11,051CITATIONS

SEEPROFILE

JoseCastillo

UniversidadAntonioRuizdeMontoya

176PUBLICATIONS5,258CITATIONS

SEEPROFILE

Availablefrom:JulioJSecades

Retrievedon:17October2016

Citicoline for Acute Ischemic Stroke: A Systematic Review andFormal Meta-analysis of Randomized, Double-Blind, and

Placebo-Controlled Trials

Julio J. Secades, MD, PhD,* José Alvarez-Sabín, MD, PhD,† José Castillo, MD, PhD,‡Exuperio Díez-Tejedor, MD, PhD,§ Eduardo Martínez-Vila, MD, PhD,‖

José Ríos, MSc,¶ and Natalia Oudovenko, PhD*

Background: Citicoline is a drug approved for the treatment of acute ischemic stroke.Although evidence of its efficacy has been reported, recently published results ofa large placebo-controlled clinical trial did not show differences. This study aimsto assess whether starting citicoline treatment within 14 days after stroke onsetimproves the outcome in patients with acute ischemic stroke, as compared withplacebo. Methods: A systematic search was performed to identify all published,unconfounded, randomized, double-blind, and placebo-controlled clinical trials ofciticoline in acute ischemic stroke. Results: Ten randomized clinical trials met ourinclusion criteria. The administration of citicoline was associated with a signifi-cant higher rate of independence, independently of the method of evaluation used(odds ratio [OR] 1.56, 95% confidence interval [CI] = 1.12-2.16 under random effects;OR 1.20, 95% CI = 1.06-1.36 under fixed effects). After studying the cumulativemeta-analysis, and with the results obtained with the subgroup of patients whowere not treated with recombinant tissue plasminogen activator (rtPA) (OR 1.63,95% CI = 1.18-2.24 under random effects; OR 1.42, 95% CI = 1.22-1.66 under fixedeffects), our hypothesis of dilution of the effect of citicoline was confirmed. Whenwe analyzed the effect of citicoline in patients who were not treated with rtPAand were receiving the highest dose of citicoline started in the first 24 hours after

From the *Scientific Department, Ferrer Group, Barcelona, Spain; †Department of Neurology, Hospital Universitari Vall d’Hebrón, Barce-lona, Spain; ‡Department of Neurology, Hospital Clínico Universitario de Santiago, Santiago de Compostela, Spain; §Department of Neurologyand Stroke Centre, Hospital Universitario La Paz, Madrid, Spain; ‖Department of Neurology, Clínica Universitaria de Navarra, Pamplona,Spain; and ¶Biostatistics and Data Management Core Facility, IDIBAPS (Hospital Clinic), Faculty of Medicine, Universitat Autònoma de Bar-celona, Barcelona, Spain.

Received March 15, 2016; revision received April 11, 2016; accepted April 13, 2016.Funding: Ferrer Grupo funded this systematic review and provided the publications, translations, and, if needed, the databases of the avail-

able studies.Competing interests: J.J.S. and N.O. are full-time employees of the company funding this research. J.A.S., J.C., E.D.T., and E.M.V. received

consultant fees and honoraria from Ferrer as member of the trial steering committee of the ICTUS trial and received honoraria for givinglectures for Ferrer.

ICMJE Forms for Disclosure of Potential Conflicts of Interest from all authors are provided.Authors’ contributions: J.J.S. and N.O. reviewed the abstracts of articles retrieved in the search. J.J.S., J.A.S., J.C., E.D.T., and E.M.V. re-

viewed all retrieved papers to identify those trials meeting the inclusion criteria for the review. J.J.S. and N.O. independently extracted theefficacy data from eligible trials. J.A.S., J.C., E.D.T., and E.M.V. resolved discrepancies through discussion by referencing the original report.J.J.S. and J.R. performed separate analyses.

Address correspondence to Julio J. Secades, MD, PhD. Scientific Department, Ferrer Group, Avda. Diagonal 549; 08029 Barcelona, Spain.E-mail: jsecades@ferrer.com.

1052-3057/$ - see front matter© 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.04.010

ARTICLE IN PRESS

Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2016: pp ■■–■■ 1

onset, based on more recent trials, there was no heterogeneity, and the size ofthe effect has an OR of 1.27 (95% CI = 1.05-1.53). Conclusions: This systematic reviewsupports some benefits of citicoline in the treatment of acute ischemic stroke. But,on top of the best treatment available (rtPA), citicoline offers a limited benefit.Key Words: Acute ischemic stroke—therapy—neuroprotection—pharmacologicaltreatment—CDP-choline—citicoline—systematic review—meta-analysis.© 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.

Introduction

Despite global and regional health crises, global life ex-pectancy has increased continuously and substantially overthe past 40 years.1 This is associated with an increase inincidences of age-related diseases, such as ischemic stroke.Stroke is the second leading cause of death in the world,and since 1990, a 26.5% increase in ischemic stroke mor-tality has been reported,2 with a parallel increase in therates of disability.3,4

Only in recent years have advances allowed for rele-vant improvement in the outcome of this devastatingdisease. A new era in acute stroke care began in 1995,when it was shown that early intravenous administra-tion of recombinant tissue plasminogen activator (rtPA)improved outcome in a carefully selected patient groupwith acute ischemic stroke.5 Together with the use of rtPA,6-9

treatment in stroke units (SUs),10 the early use of aspirin,11

and decompressive surgery12 are recognized as thera-pies with proven efficacy in the management of acuteischemic stroke. Thus, very early treatment with rtPA inan SU can be considered the gold standard,13 but can onlybe applied in a qualified center, in developed countries.14

However, the use of rtPA remains very low in the world,15

and there are substantial differences in the standard ofcare for acute ischemic stroke.16-20 Obviously, there is anurgent need for available therapeutics for acute stroke.

Citicoline, or CDP-choline, a drug that combines neu-rovascular protection and repair effects,21-23 has been usedto treat acute ischemic stroke and other neurologicaldisorders,21 with an excellent safety profile.24 In 2002, a formalmeta-analysis of trials of CDP-choline in acute and sub-acute stroke25 suggested a beneficial and substantial treatmenteffect, with absolute reductions of 10%-12% in the rates oflong-term death and disability. In a pooled data analysis(PDA), the odds ratio (OR) of complete functional and neu-rological recovery in patients with moderate-to-severe acuteischemic stroke treated with oral citicoline for 6 weeks was1.33 (95% confidence interval [CI] = 1.10-1.62) when com-pared with placebo.26 A further updated meta-analysis oftabulated data confirmed these previous results, with a re-duction in death or dependency (OR .65, 95% CI = .54-.77;P = .00001; Number Needed to Treat 9.5).27

The recent ICTUS trial28 could not confirm the effica-cy of citicoline against placebo as an add-on therapy ontop of the best standard of care. Also in the ICTUS trial,it was hypothesized that the beneficial effect of citicoline

over time was diluted in parallel with improved stan-dards of care for acute ischemic stroke.

For this reason, we decided to perform an exhaustiveand systematic search for all double-blind and placebo-controlled clinical trials performed with citicoline worldwide,and to conduct an updated meta-analysis to assess if treat-ment with citicoline (started within 14 days after strokeonset) improves the outcome (measured as a modified RankinScale [mRS] score of 0-2 or equivalent) in patients withacute ischemic stroke when compared with placebo.

Methods

Protocol and Registration

The present systematic review and meta-analysis arebased on a pre-existing protocol, adapted from a pub-lished protocol in The Cochrane Library.29 The protocolhas been submitted and registered at the PROSPERO reg-ister, with registration number CRD42013005070.

Eligibility Criteria

Clinical trials that were double-blind, randomized, andcontrolled with placebo were included. No restrictionsregarding language, publication date, or publication statuswere applied. Only trials in which the active study agentwas CDP-choline were included, and trials comparing CDP-choline treatment with any other active treatment wereexcluded.

All trials randomizing patients of any age or sex wereincluded, with index events of ischemic stroke and/orpresumed ischemic stroke (in studies without neuroimaging),and with randomization occurring within 14 days afterstroke onset. This broad treatment time window of 14days was chosen because citicoline may have effects atdifferent stages in the evolution of ischemic brain injuryand repair. No restrictions were applied in regard to doses,route of administration, or duration of treatment.

The primary efficacy measure was patient indepen-dence at the end of a scheduled follow-up period.

Information Sources

Studies were identified by searching electronic data-bases, scanning reference lists of articles, and consultingwith experts in the field and at the drug company (FerrerGroup).

ARTICLE IN PRESS2 J.J. SECADES ET AL.

The search was applied to

- Cochrane Central Register of Controlled Trials(CENTRAL) (The Cochrane Library, latest issue);

- MEDLINE (1966 to present);- PubMed (1940s to present);- Embase (1974 to present);- Internet Stroke Center (stroke trials registry); and- National Institute of Health (ClinicalTrials.gov).

Also, an exhaustive search on the Internet and in thebibliographic database of Ferrer Group was performedto detect studies published in journals that were notindexed or included in databases.

Search Strategy

The following search strategy for MEDLINE was usedand modified for the other databases when necessary.

MEDLINE

1) cerebrovascular disorders/ or basal ganglia cere-brovascular disease/ or exp brain ischemia/ orcarotid artery diseases/ or carotid artery thrombosis/or intracranial arterial diseases/ or cerebral arte-rial diseases/ or exp “intracranial embolism andthrombosis”/ or stroke/ or exp brain infarction/or cerebral infarction/

2) ((maintain ischaemic or ischemic) adj5 (stroke$ orapoplex$ or cerebral vasc$ or cerebrovasc$ or cva)).tw.

3) ((brain or cerebr$ or cerebell$ or vertebrobasil$ orhemispher$ or intracran$ or intracerebral orinfratentorial or supratentorial or middle cerebr$or mca$ or anterior circulation) adj5 (isch?emi$or infarct$ or thrombo$ or emboli$ or occlus$ orhypoxi$)).tw.

4) 1 or 2 or 35) CDP-choline or cytidine diphosphate choline or

citicoline6) (choline or (cytidine adj5 diphosphocholine) or cy-

tidine diphosphate choline or CDP-choline orciticoline or citicholine or cyticholine or cidifos).tw.

7) 5 or 68) 4 and 79) limit 8 to humansThe search in Cochrane CENTRAL, Clinicaltrials.gov,

and Internet Stroke Center was performed using the term“CDP-choline” or “citicoline.” The broad search on theInternet and PubMed was performed using the term “CDP-choline OR citicoline OR cytidine-5′-diphosphocholine ORcytidine-5′-diphosphate OR citicholine OR citocoline ORcitocholine.” The search in the bibliographic database ofthe Ferrer Group was performed using the terms “CDP-choline or citicoline and stroke.”

Selection of Studies

Two review authors (J.J.S. and N.O.) reviewed the ab-stracts of articles retrieved in the search. Full-length papers

for any possible abstract that met the inclusion criteriawere obtained. Five review authors (J.J.S., J.A.S., J.C., E.D.T.,and E.M.V.) reviewed all retrieved papers to identify thosetrials that met the inclusion criteria for the review.

Data Extraction and Management

Two review authors (J.J.S. and N.O.) independently ex-tracted the efficacy data from eligible trials. These reviewauthors resolved discrepancies through discussion withother authors (J.A.S., J.C., E.D.T., and E.M.V.) and by ref-erencing the original report.

Data Items

The primary efficacy measure was patient indepen-dence at the end of the scheduled clinical trial follow-up. If available, the mRS was used for this measure (mRSscore of 0-2). In studies where the mRS measure was notavailable, we used the most comprehensive measure ofdisability or handicap available from the trial.

The information was checked for the doses used, theroute of administration, and the duration of treatmentwith CDP-choline.

Assessment of Risk of Bias in the Included Studies

The methodological quality of the selected studies wasassessed using The Cochrane Collaboration’s tool for as-sessing the risk of bias.30,31 Each of the following pointswas scored as “yes,” “no,” or “unclear” (where “yes” in-dicates that the study is less open to bias):

1) Method of randomization (selection bias)2) Concealment of allocation (indication bias)3) Blinding of investigators and patients (perfor-

mance bias)4) Blinding of outcome assessment (detection bias)5) Adequate follow-up (attrition bias)6) Other possible biasBased on these criteria, we divided the studies into the

following 3 categories:

• A—all quality criteria met: low risk of bias,• B—one or more of the quality criteria only partly

met: moderate risk of bias, and• C—one or more criteria not met: high risk of bias.To avoid selection bias, only studies categorized as C

were rejected. Other studies not included were those per-formed for other indications, such as hemorrhagic strokeand those not reporting independence as an outcome.

Measures of Treatment Effect

Dichotomous statistical meta-analytic methods were used,and the Mantel–Haenzel method was applied for the es-timation of OR in fixed effects and DerSimonian–Lairdfor random effects to calculate the global treatment effect.

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 3

Data Synthesis

The formal meta-analysis was performed using the rmetapackagea of the R softwareb.

Hypothesis of statistical heterogeneity was quantifiedby means of an I2 value defined as 100% × (Q − df)/Q,where Q is Cochran’s heterogeneity statistic and df thedegrees of freedom and their P value. I2 values of therandom effects represent the percentage contribution ofa given random effect to the overall heterogeneity. Valuesof I2 around 25%, 50%, and 75% can be considered aslow, moderate, and high levels of heterogeneity,respectively.30 End points from individual studies wereanalyzed to compute individual and pooled OR with their95% CI, by means of a random-effects model (which ac-commodates clinical and statistical variations better).

To explore the hypothesis of a diluted effect of citicolinein parallel with the improvement of the standard of careof acute ischemic stroke, we used the cumulative meta-analysis. In cumulative meta-analysis, experiments areaccumulated from the earliest to the latest, where eachsuccessive experiment includes a synthesis of all previ-ous experiments. This chronological combination of theexperiments allows us to show if there is a consistencyin the results of consecutive experiments and indicate ifthe results continually favor 1 process, product, ortreatment.

The analyses were performed separately by J.J.S. andJ.R. Then the results were compared for discrepancies.

Risk of Bias across Studies

As presumed, heterogeneity existed among the studiesperformed over 4 decades; so, the main analysis used therandom-effects model to determine if the effects of CDP-choline were statistically and significantly different fromthe control group. As there is evidence of between-study heterogeneity (I2 > 50%), we compared the fixed-effects and random-effects estimates of the interventioneffect.31

Additional Analyses

There were nonprespecified subgroup analyses. In thesensitivity analysis, we performed the same analyses withthe rtPA-treated patients excluded, as rtPA can be con-sidered responsible for the ceiling effect and because, inthe ICTUS trial,28 a significant interaction with the useof rtPA was detected. Because the percentage of strokepatients who are treated with rtPA is, in general, verylow,15 this analysis is convenient and allows a more re-

alistic estimation of the effect of citicoline when comparedto the most usual acute care. Another additional analy-sis was based on patients not treated with rtPA and whobegan the highest dose of citicoline (2 g/day/6 weeks)within the first 24 hours after onset, which was the doseshowing efficacy in the PDA26 and the dose used in theICTUS trial,28 reflecting a more contemporary standardof care, as the studies are more recent.

To report this systematic review, we followed thePRISMA statement.32

Results

Study Selection

Searching the Embase database provided 1219 docu-ments, with 45 documents on acute stroke. Searching theMEDLINE database provided 350 documents, with 27documents on acute stroke. Searching the CochraneCENTRAL provided 166 documents, with 29 docu-ments on acute stroke without duplicates. Searching thePubMed database provided 1983 documents, with 24 docu-ments on acute stroke. Searching ClinicalTrials.gov provided33 documents, with only 1 completed study on acute isch-emic stroke. Searching the Internet Stroke Center provided10 documents, with 5 completed studies on acute isch-emic stroke. Searching the Ferrer Group database provided243 documents, with 33 documents on acute stroke withoutduplicates. The studies were selected after reviewing thecitation, the abstracts, and the full papers when avail-able. Compiling all the results, 63 studies on acute strokewere selected to be fully reviewed for inclusion in thesystematic review (Fig 1). Among the 63 studies se-lected, only 10 fulfilled the criteria to be included in thesystematic review and meta-analysis. The reasons for ex-cluding the 53 studies were as follows:

- Study categorized as C: 3733-69

- Duplicated publications: 370-72

- Study not C but compared with an active drug: 473-76

- Study not C but no independence data were avail-able: 377-79

- Study not C but in hemorrhagic stroke: 280,81

- Study not C but other investigational product: 182

- Study not C but in chronic phase: 183

- A preliminary report: 184

- Not a report of a clinical trial: 185

Study Characteristics

The 10 studies selected28,86-94 were double-blind, ran-domized, and placebo-controlled clinical trials studyingthe effect of citicoline on the recovery of patients withacute ischemic stroke (Table 1). The oldest study was pub-lished in 1980 and the most recent study was publishedin 2012; thus, there is a gap of 32 years between the firstand the last studies.

a Lumley T. rmeta: Meta-Analysis. R package version 2.16; 2009.Available at: http://CRAN.R-project.org/package=rmeta.

b R Development Core Team. R: A language and environment forstatistical computing. R Foundation for Statistical Computing, Vienna,Austria; 2005. ISBN 3-900051-07-0, URL http://www.Rproject.org.[RRID:OMICS_00245].

ARTICLE IN PRESS4 J.J. SECADES ET AL.

The included studies involved 4436 patients, but thenumber of patients evaluated was 4420. All the patientswere 18 years or older, suffering from an acute isch-emic stroke and treated within a time window of between8 hours and 14 days.

Four trials were single centered,86-88,94 whereas all theother trials were multicentered. The treatments tested wereeither citicoline, with doses ranging from 250 to 2000 mgdaily, or placebo. The duration of the treatment rangedfrom 10 days to 6 weeks. Citicoline or placebo was ad-ministered intravenously in 4 studies,86-89 orally in another4 studies,90-93 and both intravenously and orally in 2studies.28,94 Four of the included trials were performedin Europe,28,86-88 four studies were performed in the USA,90-93

one study was performed in Japan,94 and one study inVenezuela.94

The primary end point was different among the trials,but all have information about independence as at leastas secondary outcome, allowing patient independence tobe evaluated in the efficacy of the intervention. The timingof outcome measures varies between 10 days and 12 weeks.Safety was evaluated in all of the studies.

Table 2 shows the methodological quality of theselected studies. One major difference between thetrials is the standard of care applied. In the trialsperformed in the 1980s, the standard of care comprised

the use of activators of cerebral metabolism (i.e.,hydergine), cerebral vasodilators (i.e., papaverine), vita-mins, and methods for reducing blood pressure. Insome cases, patients were also treated with hypertonicsolutions, adrenocorticotropic hormone, or corticoste-roids. Many of these interventions are not valid today.During the 1990s, the studies were performed in theUnited States, and the standard of care was morehomogeneous, including anticoagulants, antiplatelets,and the use of rtPA in a few cases (less than 13% ofcases). The standard of care used in the study per-formed in Venezuela94 included the use of antiplatelets,calcium channel blockers, and vitamins. In the ICTUStrial,28 the standard of care included a very early andaggressive treatment (stroke code), with more than 80%of patients treated in SU and more than 46% of patientsreceiving rtPA. Other differences to consider among thetrials are the therapeutic window, with 2 studies havinga therapeutic window of more than 48 hours, includingthe baseline severity of the patients (Table 1).

Synthesis of the Results

The effect estimates and the confidence intervals arepresented as a forest plot in Figure 2. All the results arereported directly, as obtained from the original publication.

Figure 1. Flow diagram of the study selection. Abbreviation: CDP-choline, cytidine diphosphate choline.

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 5

Table 1. Summary of the included studies evaluating the efficacy of citicoline in acute ischemic stroke

SourceNo. of

patientsAge (mean,

range) Baseline severityTherapeutic

window Dose Route Follow-up

Boudouresques and Michel86 52 63.6 Mild–moderate 48 h 750 mg/day/10 days i.v. 10 daysGoas et al87 64 62.8 Mild–moderate 48 h 250-750 mg/d/20 days i.v. 90 daysCorso et al88 33 71.5 Mild–moderate 7-10 days 1 g/day/30 days i.v. 30 daysTazaki et al89 272 29-90 Mild–moderate 14 days 1 g/day/14 days i.v. 14 daysClark et al90 (USA1) 259 67.8 NIHSS score higher than 4 24 h 500-2000 mg/day/6 weeks p.o. 12 weeksClark et al91 (USA2) 394 70.5 NIHSS score higher than 4 24 h 500 mg/day/6 weeks p.o. 12 weeksWarach et al92 (USA3) 100 70.3 NIHSS score higher than 4 24 h 500 mg/day/6 weeks p.o. 12 weeksClark et al93 (USA4) 899 67.5 NIHSS score higher than 7 24 h 2000 mg/day/6 weeks p.o. 12 weeksAlviarez and Gonzalez94 65 69.8 NIHSS score higher than 4 8 h 2000 mg/day/6 weeks i.v. (3 days) then p.o. 6 weeksDávalos et al28 (ICTUS) 2298 72.8 NIHSS score higher than 7 24 h 2000 mg/day/6 weeks i.v. (3 days) then p.o. 12 weeks

Abbreviations: i.v., intravenous; NIHSS, National Institutes of Health Stroke Scale; p.o., oral.

Table 2. Quality measures of randomized controlled trials included

Trial Classification

Method ofrandomization

and concealmentof allocation

RCTstopped early(attrition bias)

Patientsblinded

Health careprovider blinded

Datacollectorsblinded

Outcomeassessorsblinded

Boudouresques and Michel86 A Yes No Yes Yes Yes YesGoas et al87 A Yes No Yes Yes Yes YesCorso et al88 B Unclear No Yes Yes Yes YesTazaki et al89 A Yes No Yes Yes Yes YesClark et al90 (USA1) A Yes No Yes Yes Yes YesClark et al91 (USA2) A Yes No Yes Yes Yes YesWarach et al92 (USA3) A Yes No Yes Yes Yes YesClark et al93 (USA4) A Yes No Yes Yes Yes YesAlviarez and Gonzalez94 B Unclear No Yes Yes Yes YesDávalos et al28 (ICTUS) A Yes Yes Yes Yes Yes Yes

Abbreviation: RCT, randomized clinical trial.

ARTIC

LEIN

PRESS

6J.J.

SECADESET

AL.

The administration of citicoline was associated with asignificant higher rate of independence (mRS score of 0-2),independently of the method of evaluation used (OR 1.56,95% CI = 1.12-2.16 under random effects; OR 1.20, 95%CI = 1.06-1.36 under fixed effects). As the estimates aresimilar, then we can consider that any small-study effecthas little effect on the intervention effect estimate.

Heterogeneity across Studies

As expected, a significant level of heterogeneity acrossstudies was identified (I2 = 72.1%; τ2 = .149; P = .0002),

reflecting the time gap of 32 years between the studiesincluded in the meta-analysis. To explore this heteroge-neity, a funnel plot was drawn. The funnel plot in Figure 3shows evidence of asymmetry, and this asymmetry re-flects the progressive improvement of the standard of care(placebo group) over time. The cumulative meta-analysis presented in Figure 4 shows how the effect ofthe intervention decreases over time in parallel with ex-pected improvements in the standard of care. There aresubstantial differences between the standard of care foracute ischemic stroke patients in the 1980s compared withthe standard of care now. All the included trials were

Study

Fixed effect modelRandom effects modelHeterogeneity: I−squared=72.1%, tau−squared=0.149, p=0.0002

Events Total

2348

ExperimentalEvents Total

2072

Control Odds Ratio

OR

1.20 1.56

95%−CI

[1.06; 1.36][1.12; 2.16]

W(fixed)

100%−−

W(random)

−−100%

Figure 2. Effect estimates and confidence intervals of the intervention on the rates of independence (mRS score of 0-2 or equivalent) in comparison withthe placebo. Abbreviations: CI, confidence interval; mRS, modified Rankin Scale; OR, odds ratio.

Figure 3. Funnel plot showing evidence of asymmetry.

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 7

performed with the highest quality in their time, as allthe studies were performed in qualified centers.

In the sensitivity analysis, we performed the same anal-ysis excluding those patients treated with rtPA from thestudies USA496 and ICTUS.28 For this analysis, heteroge-neity was lower (I2 = 64.5%; P = .004), still reflecting thetime gap between the studies, but showing the appro-priateness of this additional analysis. The independencerate in patients treated with citicoline but who did notreceive rtPA was higher (OR 1.63, 95% CI = 1.18-2.24 underrandom effects; OR 1.42, 95% CI = 1.22-1.66 under fixedeffects) (Fig 5). Of the patients not treated with rtPA andbeginning the highest dose of citicoline (2 g/day/6 weeks)within the first 24 hours after stroke onset (Fig 6), theanalysis also shows a positive result, with an OR of 1.27(95% CI = 1.05-1.53), and, in this case, heterogeneity wasnot significant (I2 = 29.1%, τ2 = .0193, P = .2375). Regard-ing safety, none of the studies detected any safety concernsrelated to the intervention.

Discussion

In recent years, several clinical trials were conductedto find effective therapies,97 but no convincing results havebeen obtained. Although the randomized clinical trial hasbecome the gold standard of evidence, there are someconcerns about its overvaluation.98 In the field of acuteischemic stroke trials, a substantial change in the designand conduct of the trials has been demonstrated,99 butthis has been accompanied by a reduction in external va-lidity (or generalizability) of the results.96,98 There is concernamong clinicians that external validity is often poor, andFigure 4. Cumulative meta-analysis under random-effects (a) and fixed-

effects (b) models showing the dilution of the effect of the intervention overtime. Abbreviations: CI, confidence interval; OR, odds ratio.

Study

Fixed effect modelRandom effects modelHeterogeneity: I−squared=64.5%, tau−squared=0.1326, p=0.0026

Events Total

1757

ExperimentalEvents Total

1459

Control Odds Ratio

OR

1.42 1.63

95%−CI

[1.22; 1.66][1.18; 2.24]

W(fixed)

100%−−

W(random)

−−100%

Figure 5. Effect estimates and confidence intervals of the intervention on the rates of independence (mRS score of 0-2 or equivalent) in comparison withplacebo in patients not treated with rtPA. Abbreviations: CI, confidence interval; mRS, modified Rankin Scale; OR, odds ratio.

ARTICLE IN PRESS8 J.J. SECADES ET AL.

this has led to the underuse of effective treatments. Inthe most recent trials, the findings may only be appli-cable to SUs that are participating in clinical trials andmay not be applicable in day-to-day clinical practice.100

The present systematic review and meta-analysis offerdata on the effect of citicoline in the treatment of acuteischemic stroke in comparison with placebo. The studiesincluded in the review are double-blind and placebo-controlled, and were conducted in academic centers. Thestudies are of the highest quality for the time they wereperformed. We tried to minimize as much as possible therisk of bias.95,101 We did not include 3 studies77-79 becauseonly the abstracts of these studies were available (thestudies were never published), so we could not verifythe quality of the studies. It has to be considered thatour estimates coincide with those reported in previousmeta-analyses25,27 showing that the use of citicoline wasassociated with a decrease on the rates of long-term deathand disability. Also, the results obtained in the PDA26 co-incide with our results, showing how citicoline increasedthe rates of total recovery in patients with moderate tosevere ischemic stroke.

By definition, a systematic review is a summary of themedical literature that uses explicit methods to performa thorough literature search and critical appraisal of in-dividual studies, and which also uses appropriate statisticaltechniques to combine these valid studies.102 It is rea-sonable to suggest that a systematic review should beperformed (or updated) after each and every study inthe development of a new intervention, and that doingso will enhance decision making when deciding whetherto proceed with further development, as well as whenimplementing the positive findings for a newintervention.102,103 This process inevitably brings togeth-er studies that are diverse in their design, but it isincontrovertible that treatment decisions should be basedon evidence when it exists and that good quality sys-

tematic reviews provide an essential mechanism inreviewing available evidence.104

As mentioned before, the major differences between thetrials included in this systematic review are related withthe changes of the standard for acute ischemic stroke treat-ment. This finding is relevant when we are trying toevaluate today the efficacy of a product that has beenon the market for more than 35 years. Furthermore, thechanges in the standard of care across time is a well-known source of heterogeneity.95,105 Thus, we have toconsider that when an active treatment is given, the overallresponse is the result of the treatment itself and the contextin which it is given.106

Following the results obtained in the ICTUS trial,28 theefficacy of citicoline could be questioned, but we have toconsider the poor external validity of the trial. In fact, thelesson we learnt with the ICTUS trial is that, in the caseof a patient treated very early with rtPA in a multidisci-plinary SU, citicoline is not able to provide any additionalbeneficial effect. This can be considered the gold standardfor the treatment of acute ischemic stroke but differs verymuch from actual clinical practice in the world.15,107

Another point to consider is the controversy in re-cruiting rtPA-treated patients in stroke trials,108 as we cannotrule out a ceiling effect resulting from an already maximalimprovement due to rtPA use. This fact has been arguedas a possible explanation for the failure of some recentacute ischemic stroke trials.28,109,110 One argument in favorof this hypothesis is the higher odds obtained in our sen-sitivity analysis, which includes only patients not treatedwith rtPA.

We investigated the source of heterogeneity among thetrials included in this review and we considered that theexplanation of this heterogeneity was the difference inthe standard of care for patients, rather than the qualityof the studies. This effect is evaluable in the cumulativemeta-analysis. Thus, our conclusion that the effect of

Study

Fixed effect modelRandom effects modelHeterogeneity: I−squared=29.1%, tau−squared=0.0193, p=0.2375

USA 1 1997USA 4 2001Alviarez 2007ICTUS 2012

Events

29154 13152

Total

1103

65 396 29

613

ExperimentalEvents

22107 10

147

Total

1077

64 368 30

615

Control

0.5 1 2

Odds Ratio

OR

1.271.30

1.541.551.621.05

95%−CI

[1.05; 1.53][1.01; 1.68]

[0.76; 3.13][1.15; 2.10][0.57; 4.66][0.81; 1.36]

W(fixed)

100%−−

6.3%34.6% 2.8%

56.4%

W(random)

−−100%

11.0%38.6% 5.4%

45.0%

Figure 6. Effect estimates and confidence intervals of the intervention on the rates of independence (mRS score of 0-2 or equivalent) in comparison withthe placebo in patients not treated with rtPA and who began the highest dose of citicoline (2 g/day/6 weeks) within the first 24 hours after stroke onset.Abbreviations: CI, confidence interval; mRS, modified Rankin Scale; OR, odds ratio; rtPA, recombinant tissue plasminogen activator.

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 9

citicoline could be diluted over time in parallel with pro-gressive improvements in the standard of care for acuteischemic stroke patients appears to be valid. The anal-ysis performed with more recent studies (Fig 6), in patientswho received the highest dose of citicoline in the first24 hours without rtPA treatment, should be consideredthe real size of the effect of citicoline in a more contem-porary context. This could help in defining what kindof patients could benefit from treatment with citicoline.Further advantages of citicoline include the long timewindow compared to rtPA and the administration of thetreatment not only in well-staffed, well-trained, and well-equipped stroke services administering thrombolysis butalso in much less advanced stroke services.

It can be concluded that the current systematic reviewand meta-analysis support some benefits of citicoline inthe treatment of acute ischemic stroke. It is confirmedthat the effect of citicoline has been diluted over timein parallel with improvements in the standard of care.Consequently, on top of the best treatment available (rtPA),citicoline offers a limited benefit.

Acknowledgments: The authors thank Georgina Casa-novas (Statistician, Asociación Hipótesis Alternativa [H1]) fortechnical assistance in the statistical analyses and Matt Elmorefor assistance with the technical writing.

References

1. Wang H, Dwyer-Lindgren L, Lofgren KT, et al. Age-specific and sex-specific mortality in 187 countries,1970-2010: a systematic analysis for the Global Burdenof Disease Study 2010. Lancet 2012;380:2071-2094.doi:10.1016/S0140-6736(12)61719-X.

2. Lozano R, Naghavi M, Foreman K, et al. Global andregional mortality from 235 causes of death for 20 agegroups in 1990 and 2010: a systematic analysis for theGlobal Burden of Disease Study 2010. Lancet2012;380:2095-2128. doi:10.1016/S0140-6736(12)61728-0.

3. Vos T, Flaxman AD, Naghavi M, et al. Years lived withdisability (YLDs) for 1160 sequelae of 289 diseases andinjuries 1990-2010: a systematic analysis for the GlobalBurden of Disease Study 2010. Lancet 2012;380:2163-2196.doi:10.1016/S0140-6736(12)61729-2.

4. Murray CJ, Vos T, Lozano R, et al. Disability-adjustedlife years (DALYs) for 291 diseases and injuries in 21regions, 1990-2010: a systematic analysis for the GlobalBurden of Disease Study 2010. Lancet 2012;380:2197-2223.doi:10.1016/S0140-6736(12)61689-4.

5. Tissue plasminogen activator for acute ischemic stroke.The National Institute of Neurological Disorders andStroke rt-PA Stroke Study Group. N Engl J Med1995;333:1581-1587.

6. Jauch EC, Saver JL, Adams HP Jr, et al. Guidelines forthe early management of patients with acute ischemicstroke: a guideline for healthcare professionals from theAmerican Heart Association/American StrokeAssociation. Stroke 2013;44:870-947. doi:10.1161/STR.0b013e318284056a.

7. European Stroke Organisation (ESO) ExecutiveCommittee, ESO Writing Committee. Guidelines for

management of ischaemic stroke and transient ischaemicattack 2008. Cerebrovasc Dis 2008;25:457-507. doi:10.1159/000131083.

8. ESO News: Update Guidelines January 2009. Specifictreatment: thrombolysis. Cerebrovasc Dis 2009;27:619-620.

9. Shinohara Y, Yanagihara T, Abe K, et al. II. Cerebralinfarction/transient ischemic attack (TIA). J StrokeCerebrovasc Dis 2011;20(Suppl 1):S31-S73. doi:10.1016/j.jstrokecerebrovasdis.2011.05.004.

10. Fuentes B, Díez-Tejedor E. Stroke units: many questions,some answers. Int J Stroke 2009;4:28-37. doi:10.1111/j.1747-4949.2009.00244.x.

11. Chen ZM, Sandercock P, Pan HC, et al. Indications forearly aspirin use in acute ischemic stroke: a combinedanalysis of 40 000 randomized patients from the ChineseAcute Stroke Trial and the International Stroke Trial.On behalf of the CAST and IST collaborative groups.Stroke 2000;31:1240-1249.

12. Vahedi K, Hofmeijer J, Juettler E, et al. Early decom-pressive surgery in malignant infarction of the middlecerebral artery: a pooled analysis of three randomisedcontrolled trials. Lancet Neurol 2007;6:215-222.

13. Berglund A, Svensson L, Sjöstrand C, et al. Higherprehospital priority level of stroke improves thrombolysisfrequency and time to stroke unit: the Hyper AcuteSTroke Alarm (HASTA) study. Stroke 2012;43:2666-2670.

14. Kleindorfer D, de Los Rios La Rosa F, Khatri P, et al.Temporal trends in acute stroke management. Stroke2013;44:S129-S131. doi:10.1161/STROKEAHA.113.001457.

15. Roth JM. Recombinant tissue plasminogen activator forthe treatment of acute ischemic stroke. Proc (Bayl UnivMed Cent) 2011;24:257-259.

16. Ali M, Atula S, Bath PM, et al. Stroke outcome in clinicaltrial patients deriving from different countries. Stroke2009;40:35-40. doi:10.1161/STROKEAHA.108.518035.

17. Gray LJ, Sprigg N, Bath PM, et al. Significant variationin mortality and functional outcome after acute ischaemicstroke between Western countries: data from thetinzaparin in acute ischaemic stroke trial (TAIST). JNeurol Neurosurg Psychiatry 2006;77:327-333.

18. Heidrich J, Heuschmann PU, Kolominsky-Rabas P, et al.Variations in the use of diagnostic procedures after acutestroke in Europe: results from the BIOMED II study ofstroke care. Eur J Neurol 2007;14:255-261.

19. Masjuan J, Alvarez-Sabín J, Arenillas J, et al. [Strokehealth care plan (ICTUS II. 2010)]. Neurologia2011;26:383-396. doi:10.1016/j.nrl.2010.05.008.

20. Ribo M, Molina CA, Pedragosa A, et al. Geographicdifferences in acute stroke care in Catalunya: impact ofa regional interhospital network. Cerebrovasc Dis2008;26:284-288. doi:10.1159/000147457.

21. Secades JJ. Citicoline: pharmacological and clinical review,2010 update. Rev Neurol 2011;52(Suppl 2):s1-s62.

22. Gutiérrez-Fernández M, Rodríguez-Frutos B, Fuentes B,et al. CDP-choline treatment induces brain plasticitymarkers expression in experimental animal stroke.Neurochem Int 2012;60:310-317. doi:10.1016/j.neuint.2011.12.015.

23. Alvarez-Sabin J, Roman GC. The role of citicoline inneuroprotection and neurorepair in ischemic stroke. BrainSci 2013;3:1395-1414. doi:10.3390/brainsci3031395.

24. Citicoline Periodic Safety Updated Report (Jan2011-Dec2011).

25. Saver JL, Wilterdink J. Choline precursors in acute andsubacute human stroke: a meta-analysis. Stroke2002;33:353. [27th International Stroke Conference, SanAntonio TX, Feb 2002].

ARTICLE IN PRESS10 J.J. SECADES ET AL.

26. Dávalos A, Castillo J, Alvarez-Sabín J, et al. Oral citicolinein acute ischemic stroke: an individual patient datapooling analysis of clinical trials. Stroke 2002;33:2850-2857.

27. Lee M, Towfighi A, Saver JL. Choline precursors in acuteand subacute ischemic and hemorrhagic stroke: anupdated meta-analysis of randomized controlled trials.Stroke 2010;41:e263.

28. Dávalos A, Alvarez-Sabín J, Castillo J, et al. Citicolinein the treatment of acute ischaemic stroke: aninternational, randomised, multicentre, placebo-controlledstudy (ICTUS trial). Lancet 2012;380:349-357.doi:10.1016/S0140-6736(12)60813-7.

29. Lee M, Towfighi A, Saver JL. Choline precursors foracute and subacute ischemic and hemorrhagic stroke.Cochrane Database Syst Rev 2010;(3):Art. No.: CD008401.doi:10.1002/14651858.CD008401.

30. Higgins JPT, Thompson SG, Deeks JJ, et al. Measuringinconsistency in meta-analyses. BMJ 2003;327:557-560.

31. Higgins JPT, Green S. eds. Cochrane Handbook forSystematic Reviews of Interventions Version 5.1.0[updated March 2011]. The Cochrane Collaboration, 2011.Available at: http://handbook.cochrane.org/. AccessedFebruary 26, 2015.

32. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMAstatement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions:explanation and elaboration. PLoS Med 2009;6:e1000100.doi:10.1371/journal.pmed.1000100.

33. De Gregorio G, Armeli G, Restivo S, et al. Valutazionereografica degli effetti angio-emodinamici cerebralidella citidin-difosfocolina nelle sindromi ictaliarterioesclerotiche [Rheographic evaluation of the cerebralangio-hemodynamic effects of cytidine diphosphatecholine in arteriosclerotic ictal syndromes]. Boll Soc ItalCardiol 1977;22:341-345.

34. Colonna JP, Poletto B. Étude multicentrique de laCDP-choline en Neurologie hospitalière. La Vie Médicale1980;28-29:2317-2318.

35. Schott B, Joyeux O. Valutazione dell’impiego dellaciticolina nella terapia di accidenti ischemici cerebrali.Giorn Ital Ricerche Clin Terap 1982;3:56-60.

36. Bellagamba G, Gentili G, Caratelli L, et al. Modificazionidelle plache ateromasiche carotidee e delle condizionidi flusso in pazienti affetti da patologia cerebrovascolaredopo trattamento con CDP-colina [Changes in the carotidatheromatous plaques and in flow conditions in patientswith cerebrovascular pathology after treatment with CDPcholine. Clinico-instrumental study]. MinervaCardioangiol 1984;32:483-488.

37. Centrone G, Ragno G, Calicchio G. Uso della citicolinaad alti dosaggi nelle affezioni acute cerebro-vascolari[Use of citicoline in high dosages in acute cerebrovasculardisease]. Minerva Med 1986;77:371-373.

38. Iovine C, Ricci A, Neri G, et al. [Remarks on cyticholinein subjects affected by stroke’s outcome and psycho-organic syndrome]. Rass Inter Clin Terap 1986;66:1449-1453.

39. Lebedeva NV, Lunev DK, Shvedkov VV, et al.[Recognan therapy of motor disturbances in patients withstroke]. Klin Med (Mosk) 1991;69:83-84.

40. Di Mascio R, Marchioli R, Vitullo F, et al. La gestioneospedaliera del paziente cerebrovascolare acuto. Risultatidi uno studio condotto nella Regione Abruzzo. RecentiProg Med 1996;87:523-529.

41. Bruhwyler J, Van Dorpe J, Geczy J. Multicentric open-label study of the efficacy and tolerability of citicoline

in the treatment of acute cerebral infarction. Curr TherRes 1997;58:309-316.

42. Liu RH, Liu HL. Clinical investigation on curative effectof dengzhanhua plus citicoline for acute cerebralinfarction. Clin Med (Northfield Il) 1999;19:36-37.

43. Yu W, Shen F, Ma Z. [Observation of therapeutic effectof Salviae miltiorrhiza and cytosine diphosphate-cholineinjection on patients with hypertensive cerebralhemorrhage]. Zhongguo Zhong Xi Yi Jie He Za Zhi2000;20:94-96.

44. Shu GC, Zhou WH, Li YM. Approach the timing andefficacy of citocoline in the treatment of apoplexy.Shanghai Med Pharm J 2000;21:33-34.

45. Fridman EA, Ottaviano F, Fiol M, et al. Neuroproteccionen el ICTUS isquemico agudo. Factibilidad de unprotocolo terapeutico [Neuroprotection in acute ischemicstroke. Practicability of guidelines for treatment]. RevNeurol 2001;32:818-821.

46. Yusuf M, Sadick A. Efficacy of citicoline in ischemicstroke as a supportive therapy. Pak J Med Sci2001;17:106-111.

47. Shi F, Zhou J, Meng D. [Curative effect of soybeanlecithin on cerebral infarction]. Zhonghua Yi Xue Za Zhi2001;81:1301-1303.

48. Zhou XG, Chen SS, Xiang XZ. The clinical observationof magnesium sulfate injection in the treatment of acutestroke. J Chin Gen Med 2001;2:18-19.

49. Zhang LD, Liu CY, Gu YX, et al. A study of relationshipbetween therapeutic efficacy and therapeutic durationof cranial nerve Growthine (CNG) on patients withintracerebral hemorrhage acute cerebral infarction. JPractic New Med 2001;3:677-678.

50. Liu HR. Observation of clinical curative effect on acutestage of mild cerebral infarction used cytidinediphosphate choline (CDPC). Hainan Med J 2001;12:2-3.

51. Zheng Y-P, He M-L. A prospect, random, single blindand control study on recovered effect of sermiortreatment on motor function of patients with acutecerebral infarction. Chin J Clin Rehabil 2002;6:2328-2329.

52. Zhang GZ, Dong SQ, Dou W, et al. Effect observationof nerve functional rehabilitation after treatment withcytridini triphophatis on acute ischemia cerebral vasculardisease. Chin J Clin Rehabil 2002;6:3292.

53. Li JY, Peng YZ, Yang F. [Clinical observation on effectof tongnao huoluo acupuncture therapy in treating acutecerebral infarction at ultra-early or acute stage].Zhongguo Zhong Xi Yi Jie He Za Zhi 2003;23:736-739.

54. Garcia A, Diaz F, Gil-Nuñez AC, et al. Tolerability andefficacy of the combination of piracetam and citicolinein acute ischemic stroke. A randomized comparativeopen study. 5th World Stroke Congress, Vancouver,Canada, 2004.

55. Zhong SQ, Sun LJ, Yan YZ, et al. Effect of Xuesaitongsoft capsule on hemorrheology and in auxiliarily treatingpatients with acute cerebral infarction. Chin J Integr Med2005;11:128-131.

56. Jiang H, Wang M, Guo J, et al. The midnight-noonebb-flow point selection for 30 cases of acute ischemiccerebrovascular diseases. J Tradit Chin Med 2008;28:193-197.

57. Cho HJ, Kim YJ. Efficacy and safety of oral citicolinein acute ischemic stroke: drug surveillance study in 4191cases. Methods Find Exp Clin Pharmacol 2009;31:171-176.doi:10.1358/mf.2009.31.3.1364241.

58. Mousavi S, Zare M, Khorvash F, et al. Evaluation ofciticholine effect in acute phase of supratentorial

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 11

hemorrhage: a randomized placebo-controlled clinicaltrial. J Cell Tiss Res 2009;9:1773-1776.

59. Mousavi SA, Zare M, Khorvash F, et al. Evaluation ofciticholine effect in acute phase of stroke: a randomizedclinical trial. J Cell Tiss Res 2009;9:1785-1788.

60. Panteleienko L, Sokolova L. Patients treated withciticoline in acute phase of ischemic stroke reportbetter quality of life. Eur J Neurol 2009;16(3 Suppl):412.

61. Leon-Jimenez C, Chiquete E, Cantu C, et al. Citicolinefor acute ischemic stroke in Mexican hospitals: aretrospective postmarketing analysis. Methods Find ExpClin Pharmacol 2010;32:325-330. doi:10.1358/mf.2010.32.5.1465004.

62. Shamalov NA, Stakhovskaia LV, Shetova IM, et al.[Efficacy and safety of the combined therapy withciticholine and actovegin in the acute period of ischemicstroke]. Zh Nevrol Psikhiatr Im S S Korsakova 2010;110(9Pt 2):13-17.

63. Makrotova TA, Chefranova Z, Titiva L. Combinationof thrombolytic therapy with citicoline in treatment acuteischaemic stroke. 20th European Stroke Conference,Hamburg, Germany, 2011.

64. Sobrino T, Rodríguez-González R, Blanco M, et al.CDP-choline treatment increases circulating endothelialprogenitor cells in acute ischemic stroke. Neurol Res2011;33:572-577. doi:10.1179/016164110X12807570510176.

65. Ershov VI. [Comparative aspects of using neuroprotectorsin the management of patients with ischemic stroke].Zh Nevrol Psikhiatr Im S S Korsakova. 2011;111(8 Pt2):41-44.

66. Martynov MI, Boiko AN, Kamchatnov PR, et al.[Neuroprotective treatment with citicoline (Ceraxon) inpatients with ischemic stroke.]. Zh Nevrol Psikhiatr ImS S Korsakova. 2012;112(3 Stroke):21-26.

67. Piradov MA, Sergeev DV, Krotenkova MV. Citicoline(Ceraxon) in acute stroke: assessment of clinical efficacyand effects on cerebral perfusion. Ann Clin Exp Neurol2012;6:31-36.

68. Mitta M, Goel D, Bansal KK, et al. Edaravone—citicolinecomparative study in acute ischemic stroke (ECCS-AIS).J Assoc Physicians India 2012;60:36-38.

69. Vaizova OE, Zautner NA, Alifirova VM, et al. [Influenceof neuroprotectors with choline-positive action on thelevel of brain-injury markers during acute ischemicstroke]. Eksp Klin Farmakol 2012;75:7-9.

70. Tazaki Y, Otomo E, Kutsuzawa N, et al. Clinical effectof CDP-choline (Nicholin) on disturbance ofconsciousness in acute stage of stroke (1st report): pre-testby the multicenter double-blind placebo-controlledmethod. Ther Res 1985;3:883-903.

71. Tazaki Y, Oyomo E, Kutsuzawa T, et al. Clinical effectof CDP-choline (Nicholin) on disturbance ofconsciousness at the acute stage of stroke. Multicenter,double-blind, placebo controlled study. Igaku No Ayumi.1986;136:719-817.

72. Pardo Cid P, Andres AI, Fernandez N, et al. StrokeCDP-choline trial: difficulties of patient selection forrandomised intervention trials. Z Gerontol Geriatr1999;32(2 Suppl):II-263.

73. Dereux JF, Gallois P. Resultats comparatifs ACTH/citicoline dans la phase initiale des infarctus cerebraux.Gaz Med 1987;94:82-85.

74. Ortolani O, Caggiano M, Mannelli R, et al. Protectionfrom ischemia-reperfusion damage in patients withstroke: the role of rutin and GSH. Transplant Proc1995;27:2877-2878.

75. Zhang L, Li HY, Li YZ. Citicoline capsules vs piracetamtablets in treatment of cerebral infarction. Chin J NewDrugs Clin Rem 2002;21:202-204.

76. Zhou ME, He L, Kong SY, et al. Efficacy and safety ofsodium citicoline tablets in the treatment of acute cerebralhemorrhage: a randomized controlled trial. Chin J EvidBased Med 2005;5:818-821.

77. Djoenaidi W, Hendro S, Adril AH, et al. A double-blindcontrolled trial of cytidine-diphosphate-choline (CDP-choline) on thrombotic stroke patients. 12th WorldCongress of Neurology, Kyoto (Japan), September 1981.

78. Martínez SM, Porras MA, Alvarado MA. Papel del5′-citidin-difosfato de colina en la isquemia cerebral delanciano: Estudio a doble ciego en pacientes con comadebido a accidente vascular cerebral agudo de tipooclusivo. Congreso por el Año Jubilar de la SociedadPeruana de Psiquiatría, Neurología y Neurocirugía, 1988.

79. Guillén F, Buendia C, Herrera J. CDP-choline in thetreatment of acute ischaemic stroke. 5th Meeting of ENS,Munich, 1995.

80. Secades JJ, Alvarez-Sabín J, Rubio F, et al. Citicoline inintracerebral haemorrhage: a double-blind, randomized,placebo-controlled, multi-centre pilot study. CerebrovascDis 2006;21:380-385.

81. Chua R. Role of intravenous citicoline for supratentorialhemorrhage. Cerebrovasc Dis 2007;23(4 Suppl):73.

82. Wang YD, Lu RY, Huang XX, et al. Human tissuekallikrein promoted activation of the ipsilesionalsensorimotor cortex after acute cerebral infarction. EurNeurol 2011;65:208-214. doi:10.1159/000325735.

83. Bisantis C, Colosi P, Fiorentino F. [Double blind clinicalexperimentation of CDP-coline on carotid circle byophthalmodinamometric valuation]. Ann Ottalmol ClinOcul 1982;108:865-872.

84. Kameyama M, Ohtomo E, Koshijima S, et al. Effect ofciticoline (CDP-choline) on cerebrovascular diseases. Acontrolled clinical trial with a double-blind method. ClinEvaluation/Rinsho Hyoka 1974;2:429-439.

85. Pardo P, Andres AI, Castro C, et al. ACVA: dificultadesde selección de pacientes en estudios de intervencionrandomizados. Rev Esp Geriatr Gerontol 1998;33(S1):57.

86. Boudouresques AB, Michel B. Therapeutic conduct inlight of a cerebral vascular accident and the use ofCDP-choline [in French]. Presented at: InternationalSymposium: Brain Suffering and Precursors ofPhospholipids. Paris; January 18, 1980:109-121.

87. Goas JY, Bastard J, Missoum A. Results after 90 daysof stroke treatment with CDP choline, concerning adouble blind trial [in French]. Presented at: InternationalSymposium: Brain Suffering and Precursors ofPhospholipids. Paris; January 18, 1980:123-128.

88. Corso EA, Arena M, Ventimiglia A, et al. La CDPCcolinanelle vasculopatie cerebrali: valutazioni cliniche e disemeiologia strumentale [CDP choline in cerebralvasculopathy: clinical evaluation and instrumentalsemeiology]. Clin Ter 1982;102:379-386.

89. Tazaki Y, Sakai F, Otomo E, et al. Treatment of acutecerebral infarction with a choline precursor in amulticenter double-blind controlled study. Stroke1988;19:211-216.

90. Clark WM, Warach SJ, Pettigrew LC, et al. A randomizeddose-response trial of citicoline in acute ischemic strokepatients. Citicoline Stroke Study Group. Neurology1997;49:671-678.

91. Clark WM, Williams BJ, Selzer KA, et al. A randomizedefficacy trial of citicoline in patients with acute ischemicstroke. Stroke 1999;30:2592-2597.

ARTICLE IN PRESS12 J.J. SECADES ET AL.

92. Warach S, Pettigrew LC, Dashe JF, et al. Effect ofciticoline on ischemic lesions as measured by diffusion-weighted magnetic resonance imaging. Ann Neurol2000;48:713-722.

93. Clark WM, Wechsler LR, Sabounjian LA, et al. A phaseIII randomized efficacy trial of 2000 mg citicoline in acuteischemic stroke patients. Neurology 2001;57:1595-1602.

94. Alviarez E, Gonzalez M. Efectividad y tolerabilidad dela Citicolina en el ICTUS Isquémico agudo, estudioaleatorizado, doble ciego comparado con placebo.Archivos Venezolanos de Farmacologia y Terapeutica2007;26:127-130.

95. Juni P, Witschi A, Bloch R, et al. The hazards of scoringthe quality of clinical trials for meta-analysis. JAMA1999;282:1054-1060.

96. Rothwell PM. External validity of randomised controlledtrials: “to whom do the results of this trial apply?”.Lancet 2005;365:82-93.

97. Martinez-Vila E, Irimia P. Challenges of neuroprotectionand neurorestoration in ischemic stroke treatment.Cerebrovasc Dis 2005;20(2 Suppl):148-158.

98. Kaplan BJ, Giesbrecht G, Shannon S, et al. Evaluatingtreatments in health care: the instability of a one-leggedstool. BMC Med Res Methodol 2011;11:65. doi:10.1186/1471-2288-11-65.

99. Kidwell CS, Liebeskind DS, Starkman S, et al. Trendsin acute ischemic stroke trials through the 20th century.Stroke 2001;32:1349-1359.

100. Lopez GA, Afshinnik A, Samuels O. Care of the strokepatient: routine management to lifesaving treatmentoptions. Neurother 2011;8:414-424. doi:10.1007/s13311-011-0061-1.

101. Moher D, Jadad AR, Nichol G, et al. Assessing thequality of randomized controlled trials: an annotated

bibliography of scales and checklists. Control Clin Trials1995;16:62-73.

102. Bath PM, Gray LJ. Systematic reviews as a tool forplanning and interpreting trials. Int J Stroke 2009;4:23-27.doi:10.1111/j.1747-4949.2009.00235.x.

103. Ioannidis JPA, Cappelleri JC, Lau J. Issues in comparisonsbetween meta-analyses and large trials. JAMA1998;279:1089-1093.

104. Thompson SG, Higgins JP. Treating individuals 4: canmeta-analysis help target interventions at individualsmost likely to benefit? Lancet 2005;365:341-346.

105. LeLorier J, Grégoire G, Benhaddad A, et al. Discrepanciesbetween meta-analyses and subsequent largerandomized, controlled trials. N Engl J Med 1997;337:536-542.

106. Finniss DG, Kaptchuk TJ, Miller F, et al. Biological,clinical, and ethical advances of placebo effects. Lancet2010;375:686-695. doi:10.1016/S0140-6736(09)61706-2.

107. Feigin VL, Lawes CM, Bennett DA, et al. Worldwidestroke incidence and early case fatality reported in 56population-based studies: a systematic review. LancetNeurol 2009;8:355-369. doi:10.1016/S1474-4422(09)70025-0.

108. Hermann DM, Chopp M. Promoting brain remodellingand plasticity for stroke recovery: therapeutic promiseand potential pitfalls of clinical translation. Lancet Neurol2012;11:369-380. doi:10.1016/S1474-4422(12)70039-X.

109. Shuaib A, Lees KR, Lyden P, et al. NXY-059 for thetreatment of acute ischemic stroke. N Engl J Med2007;357:562-571.

110. Ringelstein EB, Thijs V, Norrving B, et al. Granulocytecolony-stimulating factor in patients with acute ischemicstroke: results of the AX200 for Ischemic Stroke Trial.Stroke 2013;44:2681-2687. doi:10.1161/STROKEAHA.113.001531.

ARTICLE IN PRESSCITICOLINE IN ACUTE ISCHEMIC STROKE 13