Clinical Investigation of the Protective Effects of Palm Vitamin E Tocotrienols on Brain

White Matter
Yogheswaran Gopalan, Ibrahim Lutfi Shuaib, Enrico Magosso, Mukhtar Alam Ansari, Mohd
Rizal Abu Bakar, Jia Woei Wong, Nurzalina Abdul Karim Khan, Wei Chuen Liong, Kalyana
Sundram, Bee Hong Ng, Chinna Karuthan and Kah Hay Yuen

Stroke. published online April 3, 2014;

Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2014 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/early/2014/04/03/STROKEAHA.113.004449

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

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:

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

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

 Original Contribution

Clinical Investigation of the Protective Effects of Palm

Vitamin E Tocotrienols on Brain White Matter
Yogheswaran Gopalan, MPharm; Ibrahim Lutfi Shuaib, FRCR; Enrico Magosso, PhD;
Mukhtar Alam Ansari, PhD; Mohd Rizal Abu Bakar, MD; Jia Woei Wong, PhD;
Nurzalina Abdul Karim Khan, PhD; Wei Chuen Liong, FRCR; Kalyana Sundram, PhD;
Bee Hong Ng, PhD; Chinna Karuthan, PhD; Kah Hay Yuen, PhD

Background and Purpose—Previous cell-based and animal studies showed mixed tocotrienols are neuroprotective, but the
effect is yet to be proven in humans. Thus, the present study aimed to evaluate the protective activity of mixed tocotrienols
in humans with white matter lesions (WMLs). WMLs are regarded as manifestations of cerebral small vessel disease,
reflecting varying degrees of neurodegeneration and tissue damage with potential as a surrogate end point in clinical trials.
Methods—A total of 121 volunteers aged ≥35 years with cardiovascular risk factors and MRI-confirmed WMLs were
randomized to receive 200 mg mixed tocotrienols or placebo twice a day for 2 years. The WML volumes were measured
from MRI images taken at baseline, 1 year, and 2 years using a validated software and were compared. Fasting blood
samples were collected for full blood chemistry investigation.
Results—According to per-protocol (88 volunteers) and intention-to-treat (121 volunteers) analyses, the mean WML volume
of the placebo group increased after 2 years, whereas that of the tocotrienol-supplemented group remained essentially
unchanged. The mean WML volume change between the 2 groups was not significantly different (P=0.150) at the end
of 1 year but was significant at the end of 2 years for both per-protocol and intention-to-treat analyses (P=0.019 and
P=0.018). No significant difference was observed in the blood chemistry parameters between the 2 groups.
Conclusions—Mixed tocotrienols were found to attenuate the progression of WMLs.
Clinical Trial Registration—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00753532.   
(Stroke. 2014;45:00-00.)
Key Words: magnetic resonance imaging ◼ tocotrienols

V itamin E is a general term representing 2 classes of

compounds, namely tocopherols and tocotrienols.
Tocotrienols, but not tocopherols, have been reported to pos-
circulation to the ischemic area. However, it is important to
assess these findings in a prospective clinical trial.
Many compounds have been shown to display neuroprotec-
sess neuroprotective properties at concentrations within the tive effects in animal models of stroke but failed in clinical
physiological relevant range. Using hippocampal HT4 neu- trials.5 The use of patients with stroke for such studies has its
ronal cells and immature primary cortical neurons, Sen et al1 limitations. The short treatment time window from the onset
and Khanna et al2 showed that α-tocotrienol at nanomolar of clinical stroke makes recruitment of subject and logistics
concentrations was able to block glutamate-induced neuro- challenging. Also, during an ischemic stroke, poor perfusion
nal cell death by suppressing early activation of c-Src kinase may severely limit the amount of the administered compound
and 12-lipoxygenase pathway, which is independent of its from reaching the target tissues.6 It should be noted that in the
antioxidant property. In a later study using a stroke-induced animal studies that yielded positive outcomes,3,4 the test com-
rat model, rats supplemented with tocotrienols were found pounds were usually administered before stroke induction.
to have reduced infarct volume and brain injury after stroke Other reasons for clinical failure have been attributed to the
induction compared with matched controls.3 This effect was difference in the proportion of gray and white matter between
similarly demonstrated in stroke-induced canine by Rink human and animal brain. For example, rats have ≈10%
et al.4 Moreover, they also reported that tocotrienols were of white matter in the brain, whereas humans have ≈50%.7
capable of preventing loss of white matter fiber tract connec- Thus, positive outcome in studies using animals, such as rats,
tivity after stroke with improved cerebrovascular collateral is mainly associated with protection of the gray matter, but

Received December 8, 2013; final revision received February 27, 2014; accepted March 4, 2014.
From the Faculty of Pharmacy, Universiti Teknologi Mara, Selangor, Malaysia (Y.G.); Advanced Medical and Dental Institute, Penang, Malaysia (I.L.S.,
E.M., M.A.A., M.R.A.B.); School of Pharmaceutical Sciences, Universiti Sains Malaysia, Penang, Malaysia (N.A.K.K., K.H.Y.); Hovid Research Sdn
Bhd, Ipoh, Malaysia (J.W.W., B.H.N.); Malaysian Palm Oil Council (K.S.); Faculty of Medicine, Universiti Malaya, Kuala Lumpur, Malaysia (C.K.);
Buckinghamshire Healthcare, NHS Trust, Buckinghamshire, United Kingdom (W.C.L.).
Correspondence to Kah Hay Yuen, PhD, School of Pharmaceutical Sciences, Universiti Sains Malaysia, 11800 Minden, Penang, Malaysia. E-mail
[email protected]
© 2014 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.113.004449

1
Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014
2  Stroke  May 2014

the tested compounds may not necessarily be protective also liver function, renal function, and fasting blood glucose. Demographic
of the white matter, which is a major component affected in data including age, weight, height, ethnicity, social history, medical
history, and blood pressure were also recorded. Those with normal re-
human stroke.6,8,9 In view of this, a suitable model focusing on
nal and liver function were invited to undergo MRI for WMLs if they
the white matter is warranted. Moreover, injury of the white fulfilled ≥1 of the following criteria: total cholesterol level between
matter has been reported to be the major cause of functional 5.2 and 6.2 mmol/L and low-density cholesterol level between 2.6 and
disability in cerebrovascular disease and hence the importance 4.2 mmol/L, body mass index of >25 kg/m2, hypertension, or diabetes
of white matter protection.10 mellitus under medical supervision and treatment. Individuals with
the above conditions were reported to be predisposed to the develop-
White matter lesions (WMLs) are abnormal hyperintense
ment of WMLs19–25 and hence were selected to increase our chances
regions observed in MRI T2-weighted sequences.11 Often of recruiting volunteers with WMLs. Exclusion criteria included cur-
detected in elderly people, these hyperintense lesions are rent use of vitamin E supplementation (tocopherols or tocotrienols)
thought to be attributed to arteriosclerosis of small blood or within the past 3 months at the time of recruitment, self-reported
vessels in the brain, leading to chronic hypoperfusion and intolerance/hypersensitivity to vitamin E, pregnant female, presence
or history of neurological impairment, psychiatric disorders, stroke,
ischemic damage in the white matter.12 Pathological findings transient ischemic attack, malignancy, and hematological disorders.
corresponding to WMLs include myelin pallor, tissue rarefac- Those taking cholesterol-lowering drugs or showing signs of alcohol
tion associated with loss of myelin and axon, mild gliosis, as and drug dependence or those who have contraindications to undergo
well as fiber degeneration.11 They reflect a spectrum of neu- MRI were also excluded.
ropathological changes with tissue damage of varying sever- A total of 390 volunteers who fulfilled the above inclusion and ex-
clusion criteria underwent an MRI scan but were only recruited into
ity.13 It has been suggested that WML progression be used to the study if they had ill-defined hyperintensities ≥5 mm in diameter
complement the currently used primary outcome of cognitive on both T2 and proton density/fluid-attenuated inversion recovery
impairment and dementia in neuroprotection trial. The poten- images. Lacunes were differentiated by well-defined areas with sig-
tial of WMLs as a surrogate end point in intervention trials11,14 nal characteristics similar to cerebrospinal fluid. Lesions with these
is mainly attributed to the fact that WMLs progress over characteristics of ≤2 mm in diameter were excluded because they are
perivascular spaces, including those around the anterior commissure,
time.14 Moreover, they have been shown to be associated with where perivascular spaces can be large.26 Based on the MRI scan, 177
cardiovascular risk factors such as hypertension15 and diabe- volunteers were considered as having WMLs and thus were eligible
tes mellitus.16 Furthermore, WMLs can be measured quantita- for recruitment, but only 121 agreed to participate in the study. On en-
tively and noninvasively on a large population, and thus WML rollment, the volunteers were dispensed study supplementations with
progression has been advocated as a marker to evaluate the instructions to take 1 capsule in the morning and another 1 capsule
in the evening. They were also advised to maintain their habitual diet
efficacy of various pharmacological substances in research and physical activity and to report if they have been diagnosed with
and clinical settings.17,18 diseases or illnesses in between follow-up visits.
Hence, the present study was conducted to investigate
the protective effects of mixed tocotrienols in volunteers Clinical Evaluation Procedures
with WMLs. Volunteers were followed up every 3 months for a duration of 2 years.
During each visit, body mass index, blood pressure, and lifestyle hab-
Materials and Methods its (such as food intake and physical activity) and adverse events were
recorded. Fasting blood samples were obtained at 3, 6, 9, 12, and 24
Study Protocol months and analyzed for serum total cholesterol, low-density lipopro-
The present study was designed as a randomized, double-blind, and tein cholesterol, high-density lipoprotein cholesterol, triglycerides,
placebo-controlled trial. The trial was investigator initiated and un- apolipoprotein B, lipoprotein A, high-sensitivity C-reactive protein,
dertaken at Universiti Sains Malaysia facilities in Hospital Kepala fasting glucose, serum creatinine, alkaline phosphatase, alanine
Batas, Penang, Malaysia. The study commenced in November 2007 transaminase, aspartate transaminase, and γ-glutamyl transpeptidase.
with the first and the last patient randomized on November 6, 2007, Brain MRI for WMLs was conducted at baseline, 1 year, and at the
and March 27, 2010, respectively. All volunteers with MRI-confirmed end of the 2-year study period. The magnetic resonance images were
WMLs were subjected to a 2-year intervention period and were given only analyzed on completion of the study by an independent radiolo-
either mixed tocotrienols (T3) 200 mg or matching placebo (edible gist who was blinded to the study medications.
palm oil) twice daily. Both placebo and T3 (Tocovid Suprabio) were Study supplementations sufficient for 3 months were dispensed to
soft gelatin capsules and were similar in terms of color, size, shape, the volunteers during the follow-up visits. Compliance to the study
and surface texture. They were purchased from Hovid Bhd (Ipoh, supplements was assessed through patient recall, returned capsule
Malaysia). count, and plasma levels of T3. Total plasma concentrations of T3
The volunteers were randomly assigned to study medications us- (α-, γ-, and δ-T3) were measured using a validated high-performance
ing a permuted block randomization. Each block of specified size liquid chromatographic method27 after completion of the study, until
contained allocation ratio of 1:1 (placebo:tocotrienols). The random then the samples were kept in frozen condition at −80°C.
allocation sequence was not made known to the researchers who were
involved in the enrollment, management, and evaluation of volun-
teers. They were blinded throughout the study. Brain MRI
The study was approved by the by the Research Ethics Committee
for Human Studies of Universiti Sains Malaysia, Malaysia (http:// MRI Parameters
www.crp.kk.usm.my/pages/jepem.htm), and the volunteers provided Volunteers were imaged with a 1.5-T MRI machine (Model Signa
informed consent before participation in the trial. HDx, General Electric, Milwaukee, WI) using an HD 8 ch NV ar-
ray (Invivo corporation, Pewaukee, WI) radiofrequency coil. Padding
was used to immobilize the head without causing discomfort. The
Study Population scanner alignment light was used to align at the glabella. A rapid sag-
Through leaflet distribution and public talks regarding this trial, 1300 ittal localizer scan was acquired as a reference image for planning.
volunteers aged ≥35 years from the northern region of peninsular Brain MRI images were taken at an axial, sagittal, and coronal
Malaysia were screened for their full blood chemistry, lipid profile, plane of T1-weighted, T2-weighted, and a 3-dimensional volumetric

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

 Gopalan et al   Tocotrienols and Brain White Matter    3

T2 fluid-attenuated inversion recovery sequences with slice thick- and 3 were excluded because they started on antihyperlipid-
ness of 5 mm. The pulse sequence parameters for T1-weighted im- emic drugs (Figure 1). Among the 33 who dropped out from
ages were repetition time, 440; echo time, 14; field of view, 24×18
cm; number of excitations, 2; and matrix size, 320× 224. For the T2-
the study, 5 from the T3 group and 7 from the placebo group
weighted sequence parameters, the pulse sequence were repetition had undergone only the baseline brain MRI scan and came
time, 4440; echo time, 100.4; field of view, 24×24 cm; number of for ≥1 consecutive follow-up, whereas 11 from the T3 group
excitations, 1; and matrix size, 512×256, followed by the T2 fluid- and 10 from the placebo group had repeated their brain MRI
attenuated inversion recovery whereby the sequences were repetition scan at the 1-year follow-up period. Thus, 109 volunteers
time, 8002; echo time, 129; field of view, 24×24 cm; number of exci-
tations, 1; and matrix size 320×224. completed the first year of study and 88 volunteers completed
the entire study period of 2 years. Hence, 121 volunteers were
WML Volumetric Measurement included in the intention-to-treat analysis, whereas 88 in the
WML volumetric measurement was conducted using a validated soft- per-protocol analysis.
ware, Endeavor.28 The software provided a quantitative analysis tools
to quantify lesion loads and reconstruct brain tissue and lesions for Baseline Characteristics
3-dimensional visualization. It is a fully automated adaptive technique Baseline characteristics of the 121 volunteers are shown in
to detect and segment WMLs in cranial magnetic resonance images
without requiring any atlas or elaborated training procedures. It gen- Table 1. No statistically significant difference was observed
erates WML measurements in area (square millimeter) and volume between the baseline data of the T3 and placebo groups.
(cubic millimeter) when the axial T1 and T2 fluid-attenuated inversion Furthermore, the 2 groups have comparable baseline mean
recovery images are loaded in the software. There are 3 main stages in- WML volume of 1627±154 mm3 and 1732±197 mm3, respec-
volved: (1) the preprocessing stage, which includes skull stripping and
inhomogeneity correction; (2) the WMLs segmentation stage; and (3)
tively (P=0.67). The baseline total plasma T3 levels were
the postprocessing stage, which includes normal brain tissue classifi- below detection limit (0.04 μg/mL).
cation and morphological processing to remove ­false-positive WMLs.
WML Volume Changes
Outcome Measures In the per-protocol analysis, 46 were on T3 group and 42 on pla-
The primary outcome measure was changes in WML volume or cebo. The mean WML volume (±SEM) of the T3 group remained
load after 1 year and 2 years from baseline. Secondary outcomes essentially unchanged with values of 1523±182 mm3 at baseline,
included an improvement in the total lipid profile and other markers 1532±167 mm3 at 1 year, and 1494±179 mm3 at 2 years. In con-
associated with increased cardiovascular risks including apolipo-
protein B, h­ igh-sensitivity C-reactive protein and lipoprotein A, and trast, the mean WML volume of the placebo group showed a
total lipid profile.

Statistical Analysis
All primary and secondary outcome measures were analyzed us-
ing the per-protocol approach which encompassed all 88 volun-
teers who have completed the 2-year follow-up period. However,
­intention-to-treat analysis was also performed on all randomized
volunteers for WML changes. Missing data were imputed using last
value carried forward.
The changes in WML volume at 1 year and 2 years from baseline
were calculated for each volunteer. The data obtained were analyzed
using an ANOVA procedure for a 2-factor split-plot experimental de-
sign.29 If statistically significant difference was observed, a post hoc
simple main effect test was used to locate the pair that gave rise to the
difference observed.
Effect size analysis was also conducted in the WML volume
change for the 2 groups.30 Homogeneity of the baseline characteris-
tics between the 2 groups was assessed using an independent Student
t test. A statistically significant difference was considered at P<0.05.
All the statistical analyses were conducted using Statistical Package
for Social Sciences (IBM SPSS) software version 19.

Results
Volunteers Flow and Follow-Up
Of the 121 volunteers who participated in the study, 62
received T3 and 59 received placebo. During the study period,
16 (25.8%) volunteers from the T3 group and 17 (28.8%) from
the placebo group dropped out, with a total dropout of 27.3%.
Of the 16 who dropped out in the T3 group, 9 withdrew their
consent, 2 were noncontactable, 4 were excluded because of
protocol violation by taking antihyperlipidemic medications,
and 1 became pregnant. Of the 17 who dropped out in the pla- Figure 1. Participant flowchart. T3 indicates tocotrienol; and
cebo group, 13 withdrew their consent, 1 was noncontactable, WML white matter lesion.

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

4  Stroke  May 2014

Table 1. Baseline Characteristics of Volunteers Recruited

Into the Study (Mean±SD)
Total T3 Group Placebo Group
Characteristics (n=121) (n=62) (n=59) P Value
Sex (male) 48 (39.7%) 27 (43.5%) 21 (35.6%)
Age, y 52±8.5 52±8.8 52±8.2 0.66
WML volume, mm3 1678±1361 1627±1211 1732±1511 0.67
BMI, kg/m2 26±4.30 26.12±4.32 26.16±4.28 0.95
TC, mmol/L 5.70±0.70 5.76±0.63 5.63±0.67 0.27
LDL, mmol/L 3.60±0.60 3.57±0.50 3.56±0.62 0.90
HDL, mmol/L 1.42±0.32 1.43±0.34 1.41±0.30 0.81
TG, mmol/L 1.57±0.90 1.69±1.15 1.45±0.64 0.16
Systolic blood 129±16 126±16 131±16 0.08
pressure, mm Hg
Diastolic blood 78±9 76±9 79±9 0.07
pressure, mm Hg
Lp(A), mg/dL 19.50±16.20 19.61±16.60 19.47±15.90 0.96
ApoB, g/L 1.30±0.20 1.29±0.19 1.23±0.21 0.18
hs-CRP, mg/L 3.50±6.60 3.50±7.40 3.45±5.70 0.97
Serum 82.80±13.90 83.2±13.50 82.4±14.40 0.76
creatinine, μmol/L
ALP, IU/L 74.20±19.10 73.7±21.20 74.7±16.70 0.77
AST, IU/L 32.20±10.20 32.2±11.40 32.2±8.70 0.97
ALT, IU/L 30.70±15.90 30.0±14.60 31.4±17.30 0.65
GGT, IU/L 25.70±15.40 25.5±13.70 26.0±17.20 0.83 Figure 2. A, Mean white matter lesion (WML) volume change
Fasting 5.60±1.23 5.37±0.70 5.74±1.60 0.10 from baseline of tocotrienol and placebo groups using per-
glucose, mmol/L protocol analysis. B, Mean WML volume change from baseline of
tocotrienol and placebo groups using intention-to-treat analysis.
Smokers, current 18 (14.9%) 11 (17.7%) 7 (11.9%) 0.45
Hypertension only 17 (14.0%) 8 (12.9%) 9 (15.3%) Analysis according to intention-to-treat principle also
Diabetes mellitus only 8 (6.6%) 3 (4.8%) 5 (8.5%) yielded similar outcomes. As shown in Figure 2B, the T3
Hypertension and 5 (4.1%) 3 (4.8%) 2 (3.4%) group showed a negligible change in the mean WML volume
diabetes mellitus from baseline at 1 year (−17.1±56 mm3; −1.1%) and at 2
ALP indicates alkaline phosphatase; ALT, alanine transaminase; ApoB, years (−46.1±92 mm3; −2.8%), whereas the placebo group
apolipoprotein B; AST, aspartate transaminase; BMI, body mass index; GGT, γ- recorded an increase of 122±71 mm3 (7.0%) and 270±95
glutamyl transpeptidase; HDL, high-density lipoprotein; hs-CRP, high-sensitivity mm3 (15.6%), respectively. Again, the increase in the mean
C-reactive protein; LDL, low-density lipoprotein; Lp(A), lipoprotein A; T3, of WML volume was significantly different statistically at
tocotrienol; TC, total cholesterol; TG, triglycerides; and WML, white matter lesion.
the end of 2 years (P=0.018) but was not statistically signifi-
cant after 1 year (P=0.120). Effect size attributed to the pla-
gradual increase over time (1633±232 mm3 at baseline, 1804±246 cebo group was 0.2 (P=0.09) at 1 year and 0.4 (P=0.006) at
mm3 at 1 year, and 2013±305 mm3 at 2 years), being in accord 2 years, whereas insignificant effect size was observed with
with the progression of WML reported in other studies.14,18,20,31 the T3 group with respective values of 0.04 (P=0.76) and
As observed from Figure 2A, the T3 group showed a negligible 0.07 (P=0.62).
change in the mean WML volume from baseline at 1 year (9±67
mm3; 0.6%) and 2 years (−29±119 mm3; −1.9%). For the placebo Clinical Parameters
group, the mean WML volume was increased by 171±81 mm3 Table 2 shows the blood chemistry profiles of the 88 volun-
(10.5%) after 1 year and 380±123 mm3 (23.3%) after 2 years. teers at baseline, 3, 6, 9, 12, and 24 months. For all parameters,
Increase in the mean WML volume was significantly different there was no significant change in the mean levels from base-
statistically between the 2 groups at end of 2 years (P=0.019), but line throughout the study period for both treatment groups,
was not significant after 1 year (P=0.150). This could be because albeit an increase in the mean high-sensitivity C-reactive pro-
of the slow, natural progression of WMLs.24,25 tein was observed at 24th month in the placebo group. This
A moderate effect size of 0.3 (P=0.06) at 1 year and 0.5 was because of unexplained sudden increase observed in
(P=0.004) at 2 years from baseline was obtained with the pla- the high-sensitivity C-reactive protein of 2 volunteers in the
cebo group, being in accord with the observed WML progres- placebo group at this interval. The results suggested that T3
sion in this group. In comparison, the effect size in T3 group did not seem to have any influence on all these biochemical
was 0.02 (P=0.88) at 1 year and 0.06 (P=0.81) at 2 years, parameters. Moreover, there was also no statistically signifi-
inferring that WML progression was negligible. cant difference between the 2 groups in all the parameters at

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

 Gopalan et al   Tocotrienols and Brain White Matter    5

Table 2. Blood Biochemistry Profiles of Volunteers (Per-Protocol Compliance

Analysis; Mean±SD) Compliance with the study protocol and supplementation reg-
Parameters Month T3 (n=46) Placebo (n=42) P Value imen was satisfactory, as evidenced by the returned capsule
count. In addition, the mean total T3 plasma levels of the T3
TC, mmol/L Baseline 5.79±0.68 5.64±0.69 0.32
group were markedly and consistently higher compared with
3 5.76±0.61 5.52±0.63 0.08
those of the placebo group which were below detection limit
6 5.84±0.67 5.56±0.69 0.06 (Figure 3).
9 5.75±0.69 5.52±0.57 0.08
12 5.78±0.51 5.60±0.56 0.11 Adverse Events
24 5.84±0.65 5.70±0.62 0.29 Throughout the study period, no adverse event related to the
HDL, mmol/L Baseline 1.44±0.37 1.42±0.32 0.77 consumption of T3 was reported except for 5 complaints of
3 1.45±0.38 1.47±0.35 0.77 mild diarrhea or loose stools during the first week of supple-
6 1.46±0.43 1.48±0.43 0.81 mentation (T3: n=5; placebo: n=0). Their conditions were
9 1.42±0.32 1.40±0.33 0.76
fully resolved without the need to initiate drug intervention or
the volunteers withdrawing from the study.
12 1.41±0.33 1.38±0.33 0.68
24 1.31±0.30 1.40±0.36 0.21
Discussion
LDL, mmol/L Baseline 3.58±0.53 3.57±0.65 0.96
This is the first clinical trial to investigate the protective effect
3 3.55±0.51 3.39±0.67 0.19 of T3 on brain white matter, and the results showed that T3
6 3.68±0.56 3.44±0.58 0.06 could attenuate the progression of WML. It was previously
9 3.68±0.57 3.48±0.49 0.07 reported that the neuroprotective properties of T3 was essen-
12 3.66±0.45 3.47±0.64 0.10 tially via attenuation of glutamate-induced excitotoxicity.1,2,32
24 3.79±0.57 3.61±0.72 0.19 Recent studies tend to suggest that compounds including T3,
TG, mmol/L Baseline 1.71±1.23 1.43±0.72 0.20 which can attenuate the neurotoxic effects of glutamate, are
3 1.62±1.10 1.40±0.57 0.25
protective not only of neuron bodies (gray matter) but also
of the white matter region. These studies demonstrated that
6 1.58±1.05 1.36±0.62 0.24
glutamate is also released in the white matter region during
9 1.43±0.76 1.37±0.71 0.69
cerebral ischemia33 and exerting similar excitotoxicity via
12 1.63±1.19 1.30±0.58 0.11 N-methyl-D-aspartate glutamate receptors which are now
24 1.64±0.96 1.37±0.37 0.10 known to be present in the white matter.34
ApoB, g/L Baseline 1.30±0.19 1.22±0.21 0.06 Dufouil et al35 and Godin et al36 demonstrated that antihy-
3 1.27±0.16 1.26±0.13 0.61 pertensive treatment reduced WML progression. In addition,
6 1.31±0.17 1.26±0.16 0.17 Mok et al37 also showed that antihyperlipidemic treatment
9 1.33±0.16 1.24±0.16 0.01 could slow down the progression of WML in those with severe
WML at baseline. Hence, in the present study, only those vol-
12 1.27±0.18 1.24±0.17 0.40
unteers with well-controlled hypertension and diabetes melli-
24 1.24±0.22 1.22±0.22 0.61
tus were recruited. Moreover, their blood pressure and fasting
Lp(A), mg/dL Baseline 18.4±13.5 17.4±15.9 0.76 blood glucose were closely monitored and were found to be
3 19.1±12.9 16.9±15.1 0.46 essentially unchanged throughout the study period (systolic
6 19.6±13.6 16.0±16.5 0.27 blood pressure was <140 mm Hg, diastolic blood pressure
9 19.0±12.6 16.1±15.2 0.33
12 19.9±13.0 16.5±15.2 0.25
24 19.4±12.6 17.5±14.7 0.50
hs-CRP, mg/L Baseline 2.94±6.35 3.32±6.44 0.78
3 2.28±2.62 3.58±4.59 0.10
6 2.46±3.11 3.39±8.17 0.47
9 2.30±2.91 3.54±4.44 0.12
12 2.48±3.16 3.43±4.95 0.29
24 2.37±3.15 5.44±15.2 0.18
ApoB indicates apolipoprotein B; HDL, high-density lipoprotein; hs-CRP, high-
sensitivity C-reactive protein; LDL, low-density lipoprotein; Lp(A), lipoprotein A;
T3, tocotrienol; TC, total cholesterol; and TG, triglycerides.

each follow-up period, with the exception of apolipoprotein B

levels at the ninth month follow-up interval. Hence, the above
Figure 3. Mean total plasma tocotrienol concentrations of volun-
findings tend to suggest that T3 have no effect on cardiovas- teers at baseline, 3, 6, 9, 12, and 24 months for tocotrienol and
cular risk factors. placebo groups.

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

6  Stroke  May 2014

<90 mm Hg, and fasting blood glucose was <7 mmol/L). In References
the case of mildly hypercholesterolemic volunteers, their total 1. Sen CK, Khanna S, Roy S, Packer L. Molecular basis of vitamin E
cholesterol levels were <5.84 mmol/L and the low-density action. Tocotrienol potently inhibits glutamate-induced ­ pp60(c-Src)
kinase activation and death of HT4 neuronal cells. J Biol Chem.
lipoprotein levels were <3.79 mmol/L and they were not on
2000;275:13049–13055.
any antihyperlipidemic medications. Those who were started 2. Khanna S, Roy S, Ryu H, Bahadduri P, Swaan PW, Ratan RR,
on antihyperlipidemic treatment were withdrawn from the et al. Molecular basis of vitamin E action: tocotrienol modulates
study. All these measures were aimed to avoid any confound- ­12-lipoxygenase, a key mediator of glutamate-induced neurodegenera-
tion. J Biol Chem. 2003;278:43508–43515.
ing effects on the results of our study. 3. Khanna S, Roy S, Slivka A, Craft TK, Chaki S, Rink C, et al.
The present investigation must also be differentiated from Neuroprotective properties of the natural vitamin E α-tocotrienol. Stroke.
several recent studies evaluating antihyperlipidemic38,39 and 2005;36:e144–e152.
antihypertensive drugs35,36,40 on WML progression. These 4. Rink C, Christoforidis G, Khanna S, Peterson L, Patel Y, Khanna S,
et al. Tocotrienol vitamin E protects against preclinical canine isch-
studies were basically investigating WML progression via emic stroke by inducing arteriogenesis. J Cereb Blood Flow Metab.
controlling the risk factors of hyperlipidemia and hyper- 2011;31:2218–2230.
tension rather than the protective effects of these agents. In 5. O’Collins VE, Macleod MR, Donnan GA, Horky LL, van der Worp BH,
Howells DW. 1,026 experimental treatments in acute stroke. Ann Neurol.
contrast, our study was primarily aimed at investigating the
2006;59:467–477.
protective effects of T3 on WML progression. 6. Gladstone DJ, Black SE, Hakim AM; Heart and Stroke Foundation of
Our volunteers have a relatively smaller mean baseline Ontario Centre of Excellence in Stroke Recovery. Toward wisdom from
WML volume compared with those of Godin et al36 (1.68 ver- failure: lessons from neuroprotective stroke trials and new therapeutic
directions. Stroke. 2002;33:2123–2136.
sus 5.31 cm3) in a white population. This could be attributable 7. Zhang K, Sejnowski TJ. A universal scaling law between gray mat-
to the fact that our volunteers were younger (mean age of 52 ter and white matter of cerebral cortex. Proc Natl Acad Sci U S A.
versus 72 years). Moreover, a relatively low percentage of our 2000;97:5621–5626.
volunteers were presented with hypertension (14%), diabetes 8. Correa F, Gauberti M, Parcq J, Macrez R, Hommet Y, Obiang P, et al.
Tissue plasminogen activator prevents white matter damage following
mellitus (6.6%), or both (4.2%), whereas in the study by Godin stroke. J Exp Med. 2011;208:1229–1242.
et al,36 75% of the volunteers were hypertensive. However, 9. Ho PW, Reutens DC, Phan TG, Wright PM, Markus R, Indra I, et al. Is
the mean baseline WML volume reported by Mok et al37 in white matter involved in patients entered into typical trials of neuropro-
tection? Stroke. 2005;36:2742–2744.
an Asian population was comparable with that of our study,
10. Goldberg MP, Ransom BR. New light on white matter. Stroke.
although the median age of their volunteers was 63 years. 2003;34:330–332.
A validated, fully automated plug-in software28 was used to 11. Debette S, Markus HS. The clinical importance of white matter hyper-
compute the volume of WMLs. It has the advantage of mini- intensities on brain magnetic resonance imaging: systematic review and
meta-analysis. BMJ. 2010;341:c3666.
mizing discrepancies that might occur using methods which 12. Yamauchi H, Fukuda H, Oyanagi C. Significance of white matter high
are semiautomated or those based on visual rating scales intensity lesions as a predictor of stroke from arteriolosclerosis. J Neurol
to assess the WMLs.35,39 In general, visual rating scales are Neurosurg Psychiatry. 2002;72:576–582.
insensitive for measuring changes in WML severity and are 13. Gouw AA, Seewann A, Vrenken H, van der Flier WM, Rozemuller
JM, Barkhof F, et al. Heterogeneity of white matter hyperintensities in
subjected to significant ceiling or floor effects during the grad- Alzheimer’s disease: post-mortem quantitative MRI and neuropathology.
ing of WMLs. Limitations of visual scales to evaluate WML Brain. 2008;131(pt 12):3286–3298.
severity, as well as WML changes over time, have been exten- 14. Gouw AA, van der Flier WM, Fazekas F, van Straaten EC, Pantoni L,
Poggesi A, et al; LADIS Study Group. Progression of white matter
sively discussed.35,36,41–44
hyperintensities and incidence of new lacunes over a 3-year period: the
The findings from our present study indicated that T3 Leukoaraiosis and Disability study. Stroke. 2008;39:1414–1420.
could attenuate the progression of WMLs. Because they 15. Liao D, Cooper L, Cai J, Toole JF, Bryan NR, Hutchinson RG, et al.
are natural vitamins and shown to be well tolerated, they Presence and severity of cerebral white matter lesions and hypertension,
its treatment, and its control. The ARIC Study. Atherosclerosis Risk in
can be used as a long-term supplement that may help to Communities Study. Stroke. 1996;27:2262–2270.
minimize injury of brain tissues particularly in the white 16. Perros P, Deary IJ, Sellar RJ, Best JJ, Frier BM. Brain abnormalities
matter region during an ischemic event. Also, the present demonstrated by magnetic resonance imaging in adult IDDM patients
study demonstrated that WMLs can be used for investigat- with and without a history of recurrent severe hypoglycemia. Diabetes
Care. 1997;20:1013–1018.
ing the prophylactic effects of potential neuroprotectants 17. Kuller LH, Longstreth WT Jr, Arnold AM, Bernick C, Bryan RN,
with objective measurements. Beauchamp NJ Jr, et al. White matter hyperintensity on cra-
nial magnetic resonance imaging: a predictor of stroke. Stroke.
2004;35:1821–1825.
Acknowledgments 18. Schmidt R, Berghold A, Jokinen H, Gouw AA, van der Flier WM,
We thank all individuals who have participated in the trial and the Barkhof F, et al; LADIS Study Group. White matter lesion progression in
research staff for their great support. LADIS: frequency, clinical effects, and sample size calculations. Stroke.
2012;43:2643–2647.
19. Breteler MM, van Swieten JC, Bots ML, Grobbee DE, Claus JJ, van
Sources of Funding den Hout JH, et al. Cerebral white matter lesions, vascular risk factors,
The study was supported by a grant from the Malaysian Palm Oil and cognitive function in a population-based study: the Rotterdam Study.
Board. Neurology. 1994;44:1246–1252.
20. Schmidt R, Fazekas F, Kleinert G, Offenbacher H, Gindl K, Payer F,
et al. Magnetic resonance imaging signal hyperintensities in the deep and
Disclosures subcortical white matter. A comparative study between stroke patients
Dr Wong and Ng are employees of Hovid Research Sdn Bhd, a sub- and normal volunteers. Arch Neurol. 1992;49:825–827.
sidiary of Hovid Bhd. Dr Yuen is an Executive Director of Hovid Bhd. 21. Fazekas F, Niederkorn K, Schmidt R, Offenbacher H, Horner S, Bertha
The other authors report no conflicts. G, et al. White matter signal abnormalities in normal individuals:

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014

 Gopalan et al   Tocotrienols and Brain White Matter    7

correlation with carotid ultrasonography, cerebral blood flow measure- 35. Dufouil C, Chalmers J, Coskun O, Besançon V, Bousser MG, Guillon
ments, and cerebrovascular risk factors. Stroke. 1988;19:1285–1288. P, et al; PROGRESS MRI Substudy Investigators. Effects of blood
22. Ylikoski A, Erkinjuntti T, Raininko R, Sarna S, Sulkava R, Tilvis R. pressure lowering on cerebral white matter hyperintensities in patients
White matter hyperintensities on MRI in the neurologically nondiseased with stroke: the PROGRESS (Perindopril Protection Against Recurrent
elderly. Analysis of cohorts of consecutive subjects aged 55 to 85 years Stroke Study) Magnetic Resonance Imaging Substudy. Circulation.
living at home. Stroke. 1995;26:1171–1177. 2005;112:1644–1650.
23. Schmitz SA, O’Regan DP, Fitzpatrick J, Neuwirth C, Potter E, Tosi 36. Godin O, Tzourio C, Maillard P, Mazoyer B, Dufouil C. Antihypertensive
I, et al. White matter brain lesions in midlife familial hypercholester- treatment and change in blood pressure are associated with the progres-
olemic patients at 3-Tesla magnetic resonance imaging. Acta Radiol. sion of white matter lesion volumes: the Three-City (3C)-Dijon Magnetic
2008;49:184–189. Resonance Imaging Study. Circulation. 2011;123:266–273.
24. Taylor WD, MacFall JR, Provenzale JM, Payne ME, McQuoid DR, 37. Mok VC, Lam WW, Fan YH, Wong A, Ng PW, Tsoi TH, et al. Effects
Steffens DC, et al. Serial MR imaging of volumes of hyperintense white of statins on the progression of cerebral white matter lesion: Post hoc
matter lesions in elderly patients: correlation with vascular risk factors. analysis of the ROCAS (Regression of Cerebral Artery Stenosis) study.
AJR Am J Roentgenol. 2003;181:571–576. J Neurol. 2009;256:750–757.
25. Schmidt R, Fazekas F, Kapeller P, Schmidt H, Hartung HP. MRI white 38. ten Dam VH, van den Heuvel DM, van Buchem MA, Westendorp
matter hyperintensities: three-year follow-up of the Austrian Stroke RG, Bollen EL, Ford I, et al; PROSPER Study Group. Effect of
Prevention Study. Neurology. 1999;53:132–139. pravastatin on cerebral infarcts and white matter lesions. Neurology.
26. Wahlund LO, Barkhof F, Fazekas F, Bronge L, Augustin M, Sjögren M, 2005;64:1807–1809.
et al; European Task Force on Age-Related White Matter Changes. A 39. Bernick C, Katz R, Smith NL, Rapp S, Bhadelia R, Carlson M, et al;
new rating scale for age-related white matter changes applicable to MRI Cardiovascular Health Study Collaborative Research Group. Statins
and CT. Stroke. 2001;32:1318–1322. and cognitive function in the elderly: the Cardiovascular Health Study.
27. Yap SP, Julianto T, Wong JW, Yuen KH. Simple high-performance liquid Neurology. 2005;65:1388–1394.
chromatographic method for the determination of tocotrienols in human 40. Weber R, Weimar C, Blatchford J, Hermansson K, Wanke I,
plasma. J Chromatogr B Biomed Sci Appl. 1999;735:279–283. Möller-Hartmann C, et al; PRoFESS Imaging Substudy Group.
­
28. Ong KH, Ramachandram D, Mandava R, Shuaib IL. Automatic white Telmisartan on top of antihypertensive treatment does not prevent pro-
matter lesion segmentation using an adaptive outlier detection method. gression of cerebral white matter lesions in the prevention regimen for
Magn Reson Imaging. 2012;30:807–823. effectively avoiding second strokes (PRoFESS) MRI substudy. Stroke.
29. Kirk RE. Split- plot design- factorial design with block-treatment confound- 2012;43:2336–2342.
ing. In: Kirk RE, ed. Experimental Design: Procedures for the Behavioural 41. van Straaten EC, Fazekas F, Rostrup E, Scheltens P, Schmidt R, Pantoni
Sciences. Pacific Grove, CA: Cole Publishing Company; 1968:245–318. L, et al; LADIS Group. Impact of white matter hyperintensities scor-
30. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. ing method on correlations with clinical data: the LADIS study. Stroke.
Hillsdale, NJ: Erlbaum; 1988. 2006;37:836–840.
31. Whitman GT, Tang Y, Lin A, Baloh RW, Tang T. A prospective study of 42. Scheltens P, Erkinjunti T, Leys D, Wahlund LO, Inzitari D, del Ser T,
cerebral white matter abnormalities in older people with gait dysfunc- et al. White matter changes on CT and MRI: an overview of visual rat-
tion. Neurology. 2001;57:990–994. ing scales. European Task Force on Age-Related White Matter Changes.
32. Khanna S, Parinandi NL, Kotha SR, Roy S, Rink C, Bibus D, et al. Eur Neurol. 1998;39:80–89.
Nanomolar vitamin E alpha-tocotrienol inhibits glutamate-induced acti- 43. Kapeller P, Barber R, Vermeulen RJ, Adèr H, Scheltens P, Freidl W, et al;
vation of phospholipase A2 and causes neuroprotection. J Neurochem. European Task Force of Age Related White Matter Changes. Visual rat-
2010;112:1249–1260. ing of age-related white matter changes on magnetic resonance imaging:
33. Tekkök SB, Ye Z, Ransom BR. Excitotoxic mechanisms of isch- scale comparison, interrater agreement, and correlations with quantita-
emic injury in myelinated white matter. J Cereb Blood Flow Metab. tive measurements. Stroke. 2003;34:441–445.
2007;27:1540–1552. 44. Pantoni L, Simoni M, Pracucci G, Schmidt R, Barkhof F, Inzitari D.
34. Lipton SA. NMDA receptors, glial cells, and clinical medicine. Neuron. Visual rating scales for age-related white matter changes (leukoaraiosis):
2006;50:9–11. can the heterogeneity be reduced? Stroke. 2002;33:2827–2833.

Downloaded from http://stroke.ahajournals.org/ by guest on April 28, 2014