Article Text

Cilostazol and isosorbide mononitrate for the prevention of progression of cerebral small vessel disease: baseline data and statistical analysis plan for the Lacunar Intervention Trial-2 (LACI-2) (ISRCTN14911850)
  1. Philip M Bath1,
  2. Iris Mhlanga1,
  3. Lisa J Woodhouse1,
  4. Fergus Doubal2,
  5. Katherine Oatey3,
  6. Alan A Montgomery4,
  7. Joanna M Wardlaw2
  8. LACI-2 Trial Investigators
  1. 1Stroke Trials Unit, University of Nottingham, Nottingham, UK
  2. 2Centre for Clinical Brain Sciences, UK Dementia Research Institute Centre, University of Edinburgh, Edinburgh, UK
  3. 3Edinburgh Clinical Trials Unit, University of Edinburgh, Edinburgh, UK
  4. 4Nottingham Clinical Trials Unit, University of Nottingham, Nottingham, UK
  1. Correspondence to Professor Philip M Bath; philip.bath{at}


Background Cerebral small vessel disease (SVD) causes lacunar strokes (25% of all ischaemic strokes), physical frailty and cognitive impairment and vascular and mixed dementia. There is no specific treatment to prevent progression of SVD.

Methods The LACunar Intervention Trial-2 is an investigator-initiated prospective randomised open-label blinded-endpoint phase II feasibility study assessing cilostazol and isosorbide mononitrate for preventing SVD progression. We aimed to recruit 400 patients with clinically evident lacunar ischaemic stroke and randomised to cilostazol, isosorbide mononitrate, both or neither, in addition to guideline secondary ischaemic stroke prevention, in a partial factorial design. The primary outcome is feasibility of recruitment and adherence to medication; key secondary outcomes include: drug tolerability; recurrent vascular events, cognition and function at 1 year after randomisation; and safety (bleeding, falls, death). Data are number (%) and median (IQR).

Results The trial commenced on 5 February 2018 and ceased recruitment on 31 May 2021 with 363 patients randomised, with the following baseline characteristics: average age 64 (56.0, 72.0) years, female 112 (30.9%), stroke onset to randomisation 79.0 (27.0, 244.0) days, hypertension 267 (73.6%), median blood pressures 143.0 (130.0, 157.0)/83.0 (75.0, 90.0) mm Hg, current smokers 67 (18.5%), educationally achieved end of school examinations (A-level) or higher 118 (32.5%), modified Rankin scale 1.0 (0.0, 1.0), National Institutes Health stroke scale 1.0 (1.4), Montreal Cognitive Assessment 26.0 (23.0, 28.0) and total SVD score on brain imaging 1.0 (0.0, 2.0). This publication summarises the baseline data and presents the statistical analysis plan.

Summary The trial is currently in follow-up which will complete on 31 May 2022 with results expected in October 2022.

Trial registration number ISRCTN14911850.

  • stroke
  • clinical trial
  • statistics

Data availability statement

No data are available. Not applicable since main paper/results yet to be analysed and published.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


lCerebral small vessel disease (SVD) is a major cause of stroke (lacunar ischaemic stroke), intracerebral haemorrhage and vascular and mixed dementias.1 It is most commonly due to an intrinsic disorder of the brain’s small perforating arterioles with endothelial dysfunction manifesting as impaired vasoreactivity, vascular stiffness, blood–brain barrier leakage and perivascular inflammation.1 Currently, there is no specific secondary prevention for lacunar ischaemic stroke or SVD-associated cognitive decline.2 Combined aspirin and clopidogrel increased bleeding3 and intensive antihypertensive therapy did not reduce recurrence or prevent cognitive decline4 in lacunar ischaemic stroke.

However, there are other potential interventions that might prevent SVD or reduce its development and progression, including: drugs that modulate the blood–brain barrier or lower blood lipids, immunosuppressive agents, neurotrophins, peroxisome proliferator-activated receptor-gamma agonists, rho-kinase antagonists, vitamins, anti-inflammatory agents and xanthine oxidase inhibitors.2 5 Additionally, drugs that modulate the nitric oxide (NO)-cyclic guanylate cylase-phosphodiesterase-5 and prostacyclin (PGI2)-cyclic adenylate cylase-phosphodiesterase-3 systems are attractive since they mimic key endogenous vaso-regulators.2 Both NO and PGI2 have vasodilator, antiplatelet, antileucocyte,6 antismooth muscle and proendothelial effects. NO may be administered orally as substrate (L-arginine or inorganic nitrate), organic nitrates (such as isosorbide mononitrate, ISMN) or drugs that inhibit PDE5 (such a dipyridamole7 or sildenafil) to increase and preserve levels of the NO second messenger, cyclic guanylate monophosphate. Although PGI2 has to be administered intravenously, PDE3 may be inhibited with cilostazol,8 which preserves levels of the PGI2 second messenger, cyclic adenylate monophosphate; cilostazol has showed promise in secondary prevention trials in ischaemic stroke, which included large proportions of patients with lacunar stroke subtype8; and is widely used in East Asia. In addition, in experimental models, cilostazol reversed impaired oligodendrocyte precursor cell maturation which occurs when endothelial cells are dysfunctional thus potentially reducing myelin damage and facilitating its repair,9 and increased the astrocyte-to-neuron lactate shuttle thus increasing energy supply and prolonging neuronal survival.10 Both actions are potentially relevant to reducing brain damage in SVD.

SVD can present clinically as lacunar ischaemic stroke, cognitive impairment, mobility and/or mood disorders, or be covert and detected on incidentally on brain imaging.1 These presentations are associated with increased short and long-term risk of recurrent stroke, cognitive decline, dementia and functional impairments.11–15 However, although these clinical outcomes are highly relevant to patients and clinical services,15 16 the data on their rates long term, individually or in combination, are sparse, precluding reliable sample size estimations for clinical trials in SVD.17 Furthermore, it was unclear if patients with clinically evident lacunar ischaemic stroke could be identified with key prognostic variables determined accurately in routine clinical practice, or recruited in sufficient numbers rapidly enough, to make a clinical trial in stroke presentations of SVD feasible in a practical time period. This is because lacunar ischaemic stroke may be confused clinically with mild cortical ischaemic stroke,18 MRI is not always available acutely nor positive for small subcortical stroke,19 and long-term outcomes vary with severity of SVD lesions on brain imaging,13–15 making it necessary to minimise the randomisation on SVD severity.20 Since lacunar ischaemic stroke is currently managed according to guidelines covering the whole of secondary prevention of ischaemic stroke (representing best medical care) and there is no justification currently for withholding secondary prevention, any novel drugs should be tested against a background of guideline secondary ischaemic stroke prevention.

The LACunar Intervention Trial-2 (LACI-1/2) series of trials are assessing the feasibility of recruiting patients with clinically evident lacunar ischaemic stroke and tolerability of treatment with ISMN and cilostazol,17 21–24 while also gathering data on outcome event rates, with the aim of testing these agents in a large phase III efficacy trial. In LACI-1, short-term administration of cilostazol and ISMN in addition to best medical care were well tolerated when the dose was escalated, without safety concerns, in patients with clinically evident lacunar ischaemic stroke23; additionally, these drugs appeared to reduce arterial stiffness without effects on platelet function22 and appeared to improve cerebrovascular function.24

The present trial, LACI-2, is testing the long-term feasibility and tolerability of administering cilostazol and ISMN to patients with clinically evident lacunar ischaemic stroke given on top of guideline secondary ischaemic stroke prevention while assessing safety and gathering data on clinically relevant outcome event rates.17 Here, we present the statistical analysis plan (SAP) and baseline data of LACI-2.


The LACunar Intervention Trial-2 (LACI-2) is a UK-based investigator-initiated, prospective, randomised, partial factorial, open-label, blinded-endpoint phase II feasibility trial of cilostazol and/or ISMN. Patients with clinically evident lacunar ischaemic stroke, with no limit on the time interval since the stroke, with capacity to consent and who were independent in activities of daily living, were randomised to cilostazol, ISMN, both or neither in a partial factorial design. Patients with contraindications to one of the trial drugs could be randomised to the other drug alone. Randomised treatment was given in all patients in addition to guideline stroke prevention therapies, typically clopidogrel or aspirin, antihypertensive drugs and statins. Patients requiring oral anticoagulation were excluded.

Full details of regulatory approvals, patient assessment, inclusion and exclusion criteria, randomisation, minimisation criteria (age, baseline modified Rankin scale (mRS), National Institutes Health stroke scale (NIHSS), time from stroke to randomisation, highest educational attainment, smoking, systolic BP), dose-escalation scheme, short-term and long-term assessments are given in the protocol paper.17

The presence of an infarct relevant to the clinical presentation and a simplified estimate of total SVD lesion severity score, adapted for use on CT or MRI and by people not expert in brain imaging of SVD,17 21 were determined at the recruiting site. All diagnostic CT or MRI, scans obtained at any stroke recurrence and 1-year follow-up MRI, were collected by the trial office for central adjudication.

The primary outcome is feasibility of recruitment and adherence to medication. Key secondary outcomes include recruitment of hospital sites and participants, drug tolerability, symptoms (such as headache, nausea, palpitations) which might deter drug adherence, safety (bleeding, falls, death), recurrent vascular events (including stroke and myocardial infarction), cognition and function over 1 year of follow-up. Imaging outcomes (white matter hyperintensity (WMH) severity, new cortical or subcortical infarcts, lacunes, haemorrhage, microbleeds) are assessed with MRI at 1 year. Details of central CT and MRI adjudication are given in the LACI-2 protocol paper.17

LACI-2 will provide outcome event rates and hence information to estimate sample size, recruitment and trial procedures for a phase III trial based in clinical outcomes. Further information is given in online supplemental file and published protocol.17

Supplemental material

We present here a complete listing of baseline data; data are number (%) and median (IQR), unless otherwise stated.

The full details of the SAP25 26 are given in the accompanying online supplemental file, Appendix S1 and is presented prior to locking of the study database so that analyses are not data driven or reported selectively.27 The SAP also lists planned secondary analyses and substudies, and follows the layout suggested in guidelines.25 26


The trial commenced in February 2018; recruitment was suspended due to COVID-19 restrictions on 17 March 2020 by the Sponsor and reopened on 10 June 2020; however, the additional delays in reinstating site approvals meant that there was a total of 4 months without recruitment. Following a 6-month extension, recruitment stopped on 31 May 2021 to allow a year of follow-up prior to close, a total of 40 months (36 months excluding COVID-19 suspension, figure 1).

Of a planned 400 participants, 363 (91%) were recruited from 26 UK hospitals. At baseline, the average age was 64 (range 31–87) years (table 1). There were 112 (31%) female patients, the median NIHSS was 0 (0, 0), the mRS was 2 in 85 (23%) patients, and the time from stroke onset to randomisation was 79.0 (27.0, 244.0) days (table 1). The median age for completing education was 16 years, with 118 (33%) participants having one or more A-level equivalent or above qualifications. Vascular risk factor rates included median blood pressure of 143.0 (130.0, 157.0)/83.0 (75.0, 90.0) mm Hg, current smoking in 67 (19%), drug-treated hypertension in 258 (71%), drug-treated hyperlipidaemia in 278 (77%), diabetes mellitus in 80 (22%), atrial fibrillation in 5 (1%), carotid stenosis of >50% left 3 (0.8%), right 6 (1.7%) and history of previous stroke in 25 (7%) (table 2).

Table 1

Baseline characteristics by small vessel disease score as determined by recruiting site

Table 2

Baseline clinical and investigation characteristics by small vessel disease score as determined by recruiting site

Baseline clinical findings included persistent unilateral weakness in 288 (79%), sensory change in 148 (41%), neglect/inattention in 6 (2%), dysphasia in 19 (5%) and visual disturbance in 11 (3%) (table 2). On cognitive testing, the median Montreal Cognitive Assessment score was 26.0 (23.0, 28.0) with 119 (33%) having a score below 25 signifying cognitive impairment. The median score on the trail making test, part B, a test of executive function, was 25.0 (23.0, 25.0) points.

A majority of participants, 219 (60%), had both an admission CT scan and MRI brain imaging (320, 88%, had CT and 229, 65%, had MRI) (table 2). Imaging findings as reported by the recruiting site included an acute ischaemic stroke lesion thought to be consistent with the lacunar stroke symptoms in 296 (82%), WMHs present in 179 (49%) and median modified total SVD score of 1.0 (0.0, 2.0).

On central adjudication, a visible index infarct was present in 319 (88%) and averaged 10×8×10 mm in size (table 3). Only one scan was considered to be normal, that is, to have neither an acute infarct responsible for the recent stroke symptoms or any background changes of SVD or atrophy. Microbleeds were present in 20% of patients in mainly lobar (28%), deep (41%) or mixed lobar/deep (31%) locations (table 4). Atrophy was present in 75% of participants. WMHs were present in 95% of participants: 76% had up to Fazekas 2 periventricular and 82% had up to Fazekas 2 deep WMH scores. Sixty-two per cent of participants had one or more old vascular lesions including lacunes present in 96% (table 5).

Table 3

Adjudicated baseline imaging characteristics by small vessel disease score as determined by recruiting site

Table 4

Adjudicated baseline imaging characteristics (microhaemorrhages, atrophy, white matter hyperintensities and perivascular spaces) by small vessel disease score as determined by recruiting site

Table 5

Adjudicated baseline imaging characteristics (old vascular and non-stroke lesions) by small vessel disease score as determined by recruiting site

In univariate analyses and in comparison with participants with lower SVD scores (table 1), those with moderate or severe SVD scores were older and more likely to have had a previous stroke, be taking antihypertensive drugs and have ataxia, and less likely to have visual loss. Additionally, participants with more severe SVD scores were more likely to have had MRI at diagnosis and to have a relevant acute ischaemic stroke lesion present on imaging (table 3). Similarly, in comparison with participants with a low SVD score, those with a moderate/high score had more atrophy, WMHs and their severity (table 4), and the presence of old vascular lesions (table 5).


SVD is common worldwide and, so far, whether presenting covertly or with stroke, cognitive or physical impairment, or neuropsychiatric symptoms,28 has no demonstrated interventions that prevent or limit its development and progression.2–4 16 The LACI trials are assessing the tolerability and feasibility of recruiting patients, the outcome event rates, and tolerability and safety of treatment with ISMN and cilostazol17 21–24 in addition to guideline secondary stroke prevention, with the aim of testing these agents in a large phase III efficacy trial.

The baseline characteristics presented here show that LACI-2 is representative of patients with clinically evident lacunar ischaemic stroke, a marker of SVD. In particular, the average age was younger than for other types of ischaemic stroke at 64 years,29 there was a preponderance of males (69%)30 and most participants had relatively few physical signs or impairments resulting from their stroke.15 A majority of participants were diagnosed with hypertension and taking antihypertensive medication, and most were taking antiplatelet or lipid lowering drugs as guideline secondary prevention of ischaemic stroke. Many were ex- smokers or current smokers. Very few had atrial fibrillation or carotid stenosis (about 1%) reflecting that, while emboli can enter and intracranial large artery atheroma may obstruct perforating cerebral arterioles, these are rare in, and uncommon causes of, lacunar ischaemic stroke.29 31

A minority of participants had clinical cortical features such as dysphasia (19, 5.2%), neglect (6, 2%) or hemianopia (11, 3%), reflecting the overlap in symptoms between cortical and lacunar ischaemic stroke,18 and potential difficulty of recruiting pure lacunar syndrome populations on the basis of clinical syndrome and CT scanning in busy regional hospital stroke services. Nonetheless, it is reassuring that very few patients were found to have a primary cortical infarct on central expert scan read (3%), the remaining 97% having either a recent small subcortical (ie, acute lacunar) infarct as the primary cause of symptoms, or no definite visible relevant lacunar infarct but no alternative cause for their symptoms.

On adjudicated brain imaging, most participants (264, 73%) were judged to have a moderate to severe SVD score by the recruiting hospital, consistent with patients with lacunar ischaemic stroke, and demonstrating that a simplified version of the SVD score, suitable for use on CT or MRI, can be applied in busy stroke services.32

The trial is in follow-up and once this is completed, the database will be cleaned and locked. Therefore there may be minor changes in the baseline data between that provided here and in subsequent publications. Analysis will follow the SAP given here as a online supplemental file. If reasonable drug adherence and safety are confirmed, then a phase III trial will be designed and submitted for funding using the vascular and cognitive outcome event rates to power the next phase of the trial.

Supplemental material

Data availability statement

No data are available. Not applicable since main paper/results yet to be analysed and published.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by East Midlands Nottingham 2 Research Ethics Committee of the Health Research Authority number 17/EM/0077 on 10/05/2017. Participants gave informed consent to participate in the study before taking part.


We thank the patients and their relatives for their time and effort to participate in LACI-2; and the Trial Steering Committee, Sponsor, Data Monitoring Committee, Edinburgh Clinical Trials Unit staff, UK Clinical Research Network, Scottish Stroke Research Network, International Advisory Panel, and all staff at participating sites (listed in the protocol paper (Ref 17)), for their support. For further details on LACI 2 Investigator Group, refer to online supplemental file 2 in this article.


Supplementary material

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors JMW is the chief investigator, obtained ethics and regulatory approvals; PMB is Director of the Nottingham Stroke Trials Unit. KO is the current trial manager, responsible for daily running of the trial including regulatory compliance; FD is a Principle Investigator; AAM provides statistical expertise; IM and LJW are the trial statisticians; PMB drafted the manuscript; all other authors commented and edited it; all authors approved the final version for submission. JMW is the guarantor for the trial.

  • Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by British Heart Foundation (CS/15/5/31475), the Alzheimer’s Society (AS-PG-14–033), EU Horizon 2020 SVDs@Target (666881), MRC UK DRI, Fondation Leducq (16/05 CVD), NHS Research Scotland, The Stroke Association and Garfield-Weston Foundation, Chief Scientist Office (UC), and National Institute of Health Research. PMB is Stroke Association Professor of Stroke Medicine and an Emeritus NIHR Senior Investigator.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.