вход по аккаунту


Evidence for a central cholinergic effect of high-dose thiamine.

код для вставкиСкачать
Evidence for a Central Cholinergx
Effect of High-Dose Thramine
K. J. Meador, MD, M. E. Nichols, MD, P. Franke, MD, M. W. Durkin, M D , R. L. Oberzan, BA,
E. E. Moore, BA, and D . W. Loring, P h D
I n vitro animal studies have suggested that thiamine is involved in the presynaptic release of acetylcholine. Total
thiamine content in cholinergic nerve terminals is comparable with that of acetylcholine, and the phosphorylation
state of thiamine changes with release of acetylcholine. Thiamine binds to nicotinic receptors and may exhibit anticholinesterase activity. Based on these observations, we investigated the effects of pharmacological doses of thiamine on
the cognitive deficits induced by the anticholinergic scopolamine in healthy young adults using a randomized, doubleblind, placebo-controlled, double-crossover design. Drug effects were assessed by P3 event-related potential, quantitated electroencephalography, and free recall memory. Conditions included (1) baseline, (2) thiamine 5 gm p.0. and
scopolamine 0.007 mg/kg IM, and ( 3 ) lacrose PO and scopolamine 0.007 mgikg IM. Thiamine significantly reduced
adverse effects of scopolamine on P3 latency, spectral components of electroencephalography, and memory recall. T h e
results are consistent with a cholinomimetic effect of thiamine in the central nervous system. Additional studies are
needed to delineate the basic mechanisms and possible therapeutic efficacy of thiamine at pharmacological dosages.
Meador KJ, Nichols ME, Franke P, Durkin MW, Oberzan RL,Moore EE, Loring DW. Evidence
for a central cholinergic effect of high-dose thiamine. Ann Neurol 1993;34:724-726
Thiamine appears to be involved in the process of acetylcholine release from the presynaptic junction [ 1, 21.
High concentrations of phosphorylated thiamine are
found in cholinergic nerve terminals 11, 21, and its
phosphorylation state changes with release of acetylcholine and thiamine into the synaptic cleft [2). Thiamine binds reversibly to nicotinic receptors 131 and
may exhibit anticholinesterase activity [21. Further,
some studies have reported a mild beneficial effect of
high dose thiamine hydrochloride (3-8 gm Poiday) in
dementia of Alzheimer’s type 14, 51. Based on these
observations, we investigated the effects of pharmacological dosages of thiamine o n cognitive deficits induced by the anticholinergic scopolamine in healthy
adults using a randomized, double-blind, placebocontrolled, double-crossover design.
blind crossover fashion separated by 1 week. Subjects received thiamine 5 gm PO or placebo (lactose) followed by
scopolamine 0.007 mgikg IM 1.5 hours later. Testing was
begun 3 hours post thiamine or placebo orally and 1.5 hours
post scopolamine intramuscularly.
P3 Ewn t - r e h d Potentid Kecordings
Testing was performed at the same time of day on a baseline
day and on 2 drug treatment days in a randomized double-
Encephalograms (EEGs) were recorded from scalp locations
Pz, Cz, and Fz referred to linked mastoids (10-20 system)
using zinc-lead electrodes. Impedances were maintained below 3 kR. P3 potentials were obtained using the “oddball”
paradigm. Tones were 1 kHz (nontarget, 80%) or 2 kHz
(target, 2Oc/r), 50-msec duration ( 5 msec riseifall times and
40 msec plateau), and presented binaurally over headphones
at 60 dB SL with 2.5-second interstimulus interval. Approximately 50 targets were averaged per condition, in 750-msec
epochs with a 25% prestimulus baseline, filtered 0.045 to
70 Hz (3 dB down, 12 dB per octave).
Averaging epochs contaminated by eye movements were
automatically rejected. Subjects were instructed to minimize
eye movements and to avoid finger movements in order to
avoid time-locked movement potentials. Prior to recording,
the subjects were familiarized with the two types of tones
and instructed to count silently the target tones. Following
each session, the subject’s count was recorded to assure accuracy of task performance, and all subjects performed the task
adequately (ix., <5% error) in each treatment state.
Latency and amplitude measurements were made from the
Pz electrode tracing of the target average. Latency values
From the Section of Behavioral Neurology, Department of Neurology, Medical College of Georgia, Augusta, GA.
Address correspondence to Dr Meador, Section of Behavioral Neurology, Department of Neurology, Medical College of Georgia, Au-
Su & e m
Thirteen paid, healthy volunteers (6 men, 7 women; age,
23-35 yr; mean, 30 yr) with informed consent were free of
drugs, including over-the-counter medications. for 2 weeks
prior to testing. Caffeine and ethanol were prohibited for 24
hours prior to testing.
Kereived Jan 25, 1993, and in revised form Apr 20 and May 27.
Accepted for publication Jun 3, 1993.
Copyright 0 1993 by the American Neurological Association
GA 30912-328”.
were obtained from the peak amplitude of the major positive
component between 280 and 480 msec. Amplitude measurements were taken from this maximum peak relative to the
prestimulus baseline.
Means ( i SD) f o v P 3 L t e n q (msec). P 3 Amplztude ipV2),
and Percentage of Recull of Verbul Paragruphs an the Three
Drag Condztzons
Electroencephalographic Recordings and Memory Task
EEGs were collected during an eyes-closed activation task
(i.e., presentation of a prose passage) to control for nondrug-related variances in cognitive state and attention. The
same montage, impedance, and filter settings as for P3 recordings were used to obtain 6 0 seconds of EEG in each
drug state for each subject for spectral analysis. The EEG
was digitized at 256 Hz and stored for off-line analysis.
The paragraph stories were originally developed for an
investigation of memory in stroke [ b } and later shown to be
sensitive to effects of antiepileptic drugs 171.The stories are
detailed narratives of visual scenes, which were designed to
be of comparable difficulty. Presentation was counterbalanced and pseudorandomized across drug states. Each story
is approximately 300 words in length and has 100 scoring
units, which is a reapportionment of units from the original
study. Memory performance is scored as a percentage of
total units correctly recalled. Free recall was obtained upon
completion of the story presentation and EEG recording.
EEG was visually inspected off-line for the presence of
artifacts. At least 8 artifact-free, 4-second epochs were required for each subject. Two subjects were excluded from
the spectral EEG analysis due to an inadequate number of
artifact-free epochs. Artifact-free EEG underwent Fourier
transformation to calculate total power (pV’) in the delta
(0.25-3.75 Hz), theta (4-7.75 HZ), alpha (8-12.75 Hz),
and beta (13-30 Hz) bandwidths for each subject in each
drug state.
A naLyses
Separate three-level (baseline, thiamineiscopolamine, placeboiscopolamine) repeated measures analyses of variance
(ANOVAs) were performed for P3 latency, P3 amplitude,
and percentage of recall for the paragraph memory task. For
EEG spectral data, a 3 (drug condition) x 4 (bandwidth)
ANOVA was performed. Post hoc analyses using Bonferroni
criteria were conducted as indicated.
P3 latency differed statistically across drug states
t F ( 2 , 2 4 ) = 18.98, p 5 0.00011. Means ( 2SD) and
results of follow-up analyses are depicted in the Table.
An example of P3 waveforms is given in Figure 1.
P3 latency was slower for placeboiscopolamine than
thiamineiscopolamine, and slower for thiaminelscopolm i n e than baseline.
P3 amplitude varied significantly across drug states
[F(2,24) = 7.88, p 5 0.0031. Placebo/scopolamine
lowered P3 amplitude relative to baseline, and thiamine/scopolamine ( p = 0.10) exhibited a trend toward
reduced P3 amplitude relative to baseline. Placebo/
scopolamine and thiamine/scopolamine P3 amplitudes
did not differ statistically.
Story percentage of recall was also affected by drug
state [ F(2,24) = 18.11, p 5 0.0001); see Table. Mem-
P3 Latency
P3 Amplitude
9 of Recall
+ thiamine
(73 7 )
+ placebo
12 87J
13 90
“/I 5 0.01, relative to baseline.
b/I 5 0.05, relative to baseline.
‘ p 5 0.01, relative to scopolamine
‘ p 5 0.004, relative to baseline.
+ thiamine.
1? 5.25JLV
Placebo/Scoplamine ............
Fig 1. An example of P3 event-related waveforms during baseh e . thiamineiscopolamlne,and placeboiscopolamine conditions.
ory was poorer for placeboiscopolamine than thiamineiscopolamine, and for thiamineiscopolamine than
EEG spectral power data exhibited a significant
bandwidth x drug state interaction { F(6,60) = 76.27,
p 5 0.0051, a main effect of bandwidth [F(3,60) =
4.57, p 5 0.0091, and a trend for main effect of drug
state IF(2,GO) = 3.40, p = 0.0543. The bandwidth
means by drug state are depicted in Figure 2. As can
be seen in this figure, the significant interaction is the
result of drug state differences in total power for the
alpha bandwidth. Placeboiscopolamine had reduced
alpha power compared with thiamine/scopolamine ( p
I0.0002), but thiamine/scopolamine did not differ
from baseline.
The cognitive effects of the anticholinergic scopolamine were partially reversed by high dose thiamine
as evidenced by alterations in the P3 event-related potential, spectral EEG, and memory performance. The
results suggest that high dose oral thiamine has an active pharmacological effect on central cholinergic systems. In prior studies, the P3 event-related potential
has proven sensitive to the effects of anticholinergics
{8, 97. These effects can be partially reversed by the
Meador et al: High-Dose Thiamine
Fig 2. Mean total pouzv ( p V 2 )for delra, theta, alpha, und beta
collected during tierbandwidths of spectral ele~troencephalograms
bal memory activation i n baseline, thiaminelscopolamine,and
placebol.!cupolamine conditions.
(i.e., some gastrointestinal upset at the upper end of
this range) [5}. Thus, high dose thiamine treatment
may play a role in neurological disorders with few
other therapeutic options. Further, thiamine’s effect
might be enhanced by the use of allithiamine derivatives, which are absorbed more readily than thiamine
hydrochloride [13]. It should be noted that thiamine
hydrochloride in the range of 3 to 8 gm POiday appears safe over an extended period IS] but that inrravenous thiamine at 20 mgikg in cats can produce neuromuscular and ganglionic blockade { 14, 151.
In conclusion, high dose thiamine partially reverses
the cognitive and electroencephalographic effects of
the centrally active anticholinergic scopolamine. Future
studies should seek to further delineate the underlying
mechanisms and possible therapeutic efficacy of thiamine at these dosages.
We thank J. Crosby for typing the manuscript.
anticholinesterase physostigmine but not the adrenergic methylphenidate [S}. In the present study, high
dose thiamine produced an effect similar to physostigmine. It is possible that thiamine exerts its effect
through a different mechanism, but a cholinomimetic
effect of thiamine would be consistent with animal
studies, which have revealed an important role for thiamine in cholinergic mechanisms [ 1-31.
The present study employed only a single thiamine
dosage ( 5 gm PO). This dosage was chosen because
our prior patient study employing dosages from 3 to
8 gm Poiday suggested that cholinergic transmission
at the neuromuscular junction is altered at dosages of
5 gm PO/day or greater, as evidenced by statistically
significant changes in single fiber electromyography
[S]. The contention that higher dosages of thiamine
exert a differential effect is also supported by recent
unpublished daca in our laboratory using a rat model,
which demonstrated partial reversal of scopolamine
effects at 100 to 200 mgikg IP but not at 1.5 to 50
Present recommended therapeutic dosages of thiamine (100 mgiday) [lo] for Wernicke-Korsakoff’s syndrome are based upon early treatment and absorption
studies but have never been investigated systematically. Although current recommended dosages may be
adequate within certain parameters, they may not be
optimal. Cholinomimetics can reduce the behavioral
deficits of thiamine deprivation and enhance thiamine
activity Cll]. Because only mild cognitive enhancement is seen in patients with dementia of Alzheimer’s
type treated with either high dose thiamine [4, 51 or
with an anticholinesterase 121, the higher dosages of
thiamine may be simply producing a cholinomimetic
effect. However, unlike other cholinomimetic drugs,
high dose thiamine hydrochloride in the range of 3 to
8 gm POiday has produced only minimal side effects
726 Annals of Neurology Vol 34
No 5
The study was presented in part at the 1772 annual meeting of the
American Neurological Association, Toronto, Canada.
1. Eder L, Hirt L, Dunant Y. Possible involvement of thiamine in
acetylcholine release. Nature 1976;264:186-188
2. Eder L, Dunant Y .Thiamine and cholinergic transmission in the
electric organ of torpedo. J Neurochem 1980;35:1278-1296
3. Waldelind L Studies on thiamine and neuromuscular transmission. Acta Physiol Scand 1978;459(suppl): 1-35
4. Bkdss JP, Gleason P, Brush D, et al. Thiamine and Alzheimer’s
disease. Arch Neurol 1988;45:833-835
5. Meador KJ, Loring DW, Nichols ME, et al. Preliminary findings
of high close thiamine iri dementia of Alzheimer’s type. J Geriatr
Psychiatr Neurol 1993 (in press)
6. Martin RC, Meador K J , Loring DW, et al. The contribution of
an imagery code to verbal memory. Cortex 1770;26:443-450
7 . Meador KJ, Loring DW. Allen ME, et al. Comparative cognitive
effects of carbamazepine and phenyroin in healthy adulrs. Neurology 1991;41:1537-1 540
8. Meador KJ, Loring DW, Adams RJ, et al. Central cholinergic
systems and the P3 evoked potential. Int J Neurosci 1987;33:
9 Meador KJ. Cholinergic, serotonergic, and GABAnergic effects
on the ERP. Electroencephdogr Clin Neurophysiol 1973 (in
10. Wilson JD, Braunwald E, Isselbacher KJ, et al, eds. Harrison’s
principles of internal medicine. 12th ed. New York: McGrawHill, 15)5)1:2047
11 Gibson G, Barclay L, Mass JP. The role of the cholinergic system in thiamin deficiency. Ann N Y Acad Sci 1982;378:382-
November 1993
12 Davis KL., Thal L1,Gamzu ER, et al. A double-blind, placebocontrolled multicenter study of racrine for Alzheimer’s disease.
N Engl J Med 1772;327:1253-1253
13 Baker H, Frank 0. Absorption, utilization, and clinical effectiveness of allithiamines compared to water-soluble thiamines. J
Nurr Sci Vitaminol 1976;22:63-68
14 DiPalma JR, Hirchcock P. Neuromuscular and ganglionic
blocking action of thiamine and its derivatives. Anesrhesiology
15. Enomoto K. Edwards C . Thiamine blockade of neuromuscular
transmission. Brain Res 1985;358:316-323
Без категории
Размер файла
328 Кб
central, effect, thiamine, dose, evidence, high, cholinergic
Пожаловаться на содержимое документа