Download Lower limb flexor reflex: Comparisons between - e

Accepted Manuscript
Title: Lower limb flexor reflex: Comparisons between
drug-induced akathisia and restless legs syndrome
Authors: Aysegul Gunduz, Baris Metin, Sinem Metin, Burc
Cagri Poyraz, Derya Karadeniz, Gunes Kiziltan, Meral E.
Kiziltan
PII:
DOI:
Reference:
S0304-3940(17)30062-9
http://dx.doi.org/doi:10.1016/j.neulet.2017.01.042
NSL 32584
To appear in:
Neuroscience Letters
Received date:
Revised date:
Accepted date:
2-7-2016
15-1-2017
17-1-2017
Please cite this article as: Aysegul Gunduz, Baris Metin, Sinem Metin, Burc Cagri
Poyraz, Derya Karadeniz, Gunes Kiziltan, Meral E.Kiziltan, Lower limb flexor reflex:
Comparisons between drug-induced akathisia and restless legs syndrome, Neuroscience
Letters http://dx.doi.org/10.1016/j.neulet.2017.01.042
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Lower limb flexor reflex: comparisons between drug-induced
akathisia and restless legs syndrome
Aysegul Gunduz1, Baris Metin1, Sinem Metin2, Burc Cagri Poyraz2, Derya Karadeniz1,
Gunes Kiziltan1, Meral E. Kiziltan1
1
Department of Neurology, Cerrahpasa Medical School, Istanbul University, Turkey
2
Department of Psychiatry, Cerrahpasa Medical School, Istanbul University, Turkey
Correspondence to: Aysegul Gunduz
I.U. Cerrahpasa Medical Faculty Department of Neurology, Istanbul, Turkey
e-mail: [email protected]
Cerrahpasa Medical Faculty, Department of Neurology 34098
K.M.Pasa /Istanbul/Turkey
Tel. Number: +902124143162
Fax number: +902126330176
Highlights

Excitability of LLFR pathway is increased in both akathisia and RLS.

Spatial spread of LLFR in akathisia is similar to healthy subjects and differs from RLS
patients.

Our results suggest a different origin for increased spinal excitability in akathisia and
RLS.
Abstract
Background and objective: Akathisia is characterized by restlessness and crawling sensations
similar to restless legs syndrome (RLS). Long latency flexor reflex (LLFR) which has helped
to advance RLS pathophysiology has never been investigated in akathisia. Due to the clinical
commonalities of akathisia and RLS, we investigated the behavior of LLFR in patients with
akathisia aiming to understand pathophysiology of akathisia.
Patients and methods: Seven patients with neuroleptic-induced akathisia, 12 drug-naïve
patients with primary RLS and 17 healthy subjects were prospectively enrolled in the study.
LLFR was recorded from unilateral tibialis anterior (TA) and long head of biceps femoris (BF)
muscles after stimulating the sole by trains of electrical stimuli. We measured amplitude,
latency, duration, presence of response and compared between three groups.
Results: One-way ANOVA showed mean durations of early and late responses recorded over
TA were the longest in akathisia group compared to both RLS group and healthy subjects
(p=0.012). The spatial spread of LLFR in akathisia patients was comparable to those of healthy
subjects whereas presence of response on BF was significantly less in akathisia than RLS group.
Conclusions: Our findings indicate increased excitability of LLFR pathway in akathisia group.
These findings are probably due to lack of inhibition originated in regions other than those
known to downregulate in RLS.
Keywords: Akathisia, restless legs syndrome, lower limb flexor reflex
Introduction
Akathisia is a movement disorder characterized by urge, inner restlessness and constant
requirement to move the limbs. Akathisia symptoms may fluctuate throughout the day. The
restlessness decreases by passive or active movements [1]. This neural disorder may be
idiopathic or primary, as well as secondary to a number of factors including neuroleptic drugs.
Interestingly, restlessness and crawling sensations are also clinical aspects of restless legs
syndrome (RLS) [2]. Restlessness in RLS is also relieved by movement of the affected limb
[3], and symptoms in RLS also fluctuate. In fact, RLS symptom are less prominent during the
morning and worsen during late afternoon [3].
The long latency flexor reflex (LLFR), also called flexor, flexion or withdrawal reflex,
is a protective reflex characterized by synergistic flexion movement of lower extremities after
receiving painful lower limb stimulus. RI component is currently not considered. RII is elicited
by stimulation of A-beta fibers [4], and it appears at about 55 ms post electrical stimulation.
RIII is elicited by stimulation of A-delta fibers, and has a latency ranging from 85 to 180 ms
[4,5]. LLFR is modulated by interneurons controlled by dopaminergic, cholinergic and
GABAergic neural transmission [6-8]. LLFR has been used to investigate functional status of
nociceptive spinal pathways as well as pathways contributing to the locomotion and
sensorimotor integration at the spinal level [4].
Germane to this research, LLFR test has helped to advance RLS pathophysiology [9,10].
However, this reflex has never been investigated in akathisia. Due to the clinical commonalities
of akathisia and RLS, we planned to investigate the behavior of LLFR in patients with akathisia
aiming to disentangle akathisia pathophysiology. To accomplish such aim, we compared
akathisia results with data obtained from RLS patients, and from a group of healthy individuals.
We hypothesized that some, but not all, of the LLFR measures would differ among these three
groups. The abnormal interneuronal activity in akathisia found in this study by measuring the
LLFR will offer novel biological explanations for better understanding this neural disorder.
Participants and methods:
Participants: Seven patients with akathisia were studied who were diagnosed as neurolepticinduced akathisia by psychiatrists based on the Diagnostic and Statistical Manual of Mental
Disorders [11]. Five of these patients had schizophrenia, one had psychotic depression and one
had bipolar disorder. All patients were receiving dopamine receptor antagonist medications
(quetiapine, olanzapine, clozapine, zuclopenthixol or amisulpride). Three patients were also
using selective serotonin reuptake inhibitors (SSRI), one patient also received valproic acid.
The duration of medication use ranged from one month to 2 years. Akathisia was diagnosed
one month before this study was made. The Barnes Akathisia Rating Scale (BARS) [12] was
used to score akathisia. Two patients had mild, three patients had moderate and two patients
had severe akathisia.
Twelve drug-naïve patients with primary RLS were also studied. RLS patients fulfilled
the International RLS Study Group criteria [2,3]. Patients underwent detailed interviews that
included the average number of days with RLS in the last four weeks. Neurological examination
and routine biochemical investigations (serum ferritin, vitamin B12, folic acid, urea, and
glucose) were also made. Polysomnography was performed when history indicated symptoms
of concomitant sleep disorder such as periodic limb movement disorders (PLMD), insomnia or
obstructive sleep apnea syndrome. The range of disease duration in RLS group was between 1–
20 years. We asked patients about the average number of days with RLS symptoms in the last
four weeks and calculated the mean frequency of symptoms per week for each patient.
Afterwards, we calculated mean frequency of symptoms per week for the group which was 5.5
± 2.3 days/week.
Seventeen healthy subjects were also studied. These participants were matched to
patients as close as possible regarding age and gender. Table 1 shows demographical findings
of groups.
Electrophysiological assessments: Studies were done in the afternoon, between 13:30-15:30
hours, to minimize circadian influences [13]. Neuropack Sigma MEB-5504k machine (Nihon
Kohden Medical, Tokyo, Japan) was used to perform the electrophysiological studies, which
were made in a quiet room. Individuals were studied in supine position with both legs semiflexed. Participants were asked to remain awake and relaxed as much as possible during the
test.
The stimulating electrode for eliciting the LLFR was positioned on the sole of the right foot
(Figure 1). The recording electrodes were placed over the right-sided tibialis anterior (TA) and
long head of biceps femoris (BF) muscles. A 20 ms electrical stimulation train of four 0.5 ms
square-wave pulses duration was applied. Four recording trials were obtained to assess
consistency of the responses. Unilateral extremity testing was done since no asymmetry of the
symptoms was reported in akathisia or RLS patients. The EMG activity was filtered (low-pass
filters: 3 kHz, high-pass filters: 2 Hz), stored and analyzed off-line. Sweep time was 50
ms/division and sensitivity was set to 100-200 µV/ division. EMG activities lower than 50 µV
amplitude were rejected. Responses with latencies between 40-60 ms from TA muscle were
classified as early responses. Responses with latencies between 90-150 ms from TA and BF
muscles were named as late responses [4]. Throughout the manuscript, we will refer them as
early and late instead of RII and RIII to avoid confusion. The parameters used in the
electrophysiological studies of LLFR are stimulus intensity, onset latency, amplitude, duration,
spatial spread, recovery curve and habituation [4].
Statistical analysis:
1) Latency was defined as the duration between stimulus artifact and onset of EMG activity
bigger than 50 µV.
2) Amplitude between highest positive and negative peaks was measured.
3) Duration was defined as the time from the onset of relevant EMG activity till its total
abolishment.
4) Presence of response was also noted and probability was calculated as patients with response
/ total number of patients.
5) Spatial spread was defined as the topography of muscles showing EMG activity.
6) In some cases, multiple EMG responses appeared, we labeled them as ‘after-discharges’.
1) Onset latencies, 2) amplitudes, 3) duration, and presence of early TA muscle responses were
measured and mean values were calculated. Since no consistent early BF responses were
detected, we did not include them in the analysis. 4) Onset latencies, 5) duration and 6) presence
of late TA muscle responses and 7) onset latencies, and 8) presence of BF muscle responses
were also measured and mean values were calculated.
Data analyses were performed using the SPSS 11.5 software statistical package (SPSS Inc.,
Chicago, IL, USA).
1) Latency, and duration of responses were compared using two-way ANOVA combining two
factors (three groups: akathisia, RLS and healthy subjects and two different muscles: TA and
BF). One-way ANOVA was used to evaluate differences between the groups of akathisia, RLS
and healthy subjects.
2) Latency, amplitude and duration of were also compared using Kruskal-Wallis test. MannWhitney U test was used for post-hoc analysis because the data distributed non-homogenously.
3) Presence of response or presence of after-discharges were compared by chi-square test.
Statistical significance was established at p < 0.05. We also performed Bonferroni correction
to avoid type I error.
Results
1) Comparisons using two-way ANOVA showed both being in different groups and
responses of different muscles had an impact on durations of responses (F=4.379,
p=0.018). The effect of muscle-group interaction was not significant (p=0.104).
One-way ANOVA showed mean durations of early and late responses recorded over
TA were longest in akathisia group (p=0.012) whereas latency was shortest in RLS
group (p=0.027). Mean duration of early TA responses was significantly longer than
other responses (F=7.410; p=0.009).
2) Kruskal-Wallis test showed results similar to one-way ANOVA:
A. Mean durations of early and late TA responses were significantly prolonged in
akathisia group compared to other groups, RLS and healthy subjects (Figure 2,
Table 2).
B. Mean latencies of early TA and BF responses in akathisia group were similar to
healthy subjects whereas latency of late TA response was shorter in RLS group
(p=0.031). Post-hoc analysis revealed a significant difference between RLS
group and healthy subjects (p=0.013).
C. Although insignificant, the mean amplitude of early TA responses tended to be
higher in groups of akathisia and RLS compared to healthy subjects.
3) A. The spatial spread in akathisia patients was comparable to those of healthy
subjects whereas presence of BF response was significantly less in akathisia group
than RLS group (Figure 3, p=0.008).
B. Three patients with akathisia had continuous TA responses without development
of any silent period whereas two patients (16.7%) in the RLS group and one patient
(14.3%) in the akathisia group had after-discharges.
Discussion
The major results of our study were the increased duration of LLFR in akathisia, presence of
after-discharges and longer duration responses. These findings indicate abnormal inhibitory
mechanisms in akathisia. The spatial spread of LLFR as well as normal response latencies in
akathisia compared to healthy individuals indicate that excitability alterations in akathisia exist;
however, clearly differ from RLS patients.
Shorter latency, higher amplitudes, prolonged durations, and greater spatial spread
suggest increased excitability and transmissibility throughout the pathway.
Latency mostly reflects capability of transmission and structural integrity and, to a lesser
extent excitability whereas amplitude and duration reflect excitability of the related reflex
circuit. Spatial spread reflects the number of muscles recruited in the response [9].
In our patient population, latencies of akathisia group were comparable to healthy
subjects confirming absence of structural changes in this group. Duration of responses strongly
suggested presence of increased excitability in akathisia. Presence of after-discharges or
prolonged and even continuous responses is suggestive of increased excitability of LLFR circuit
in akathisia. Although there was no difference regarding amplitudes, the lack of significance
may also be related to the large intraindividual and interindividual differences. Generator of the
LLFR is at the spinal level and is formed by a complex network of interneurons which
constitutes a filtering mechanism by integrating suprasegmental commands and peripheral
sensory feedback and is under the control of a more complex supraspinal organization [4,14].
Therefore, increased excitability suggests loss of inhibitory drive on the generator of LLFR at
the spinal level in akathisia.
Suprasegmental pathways which may have influence on LLFR include dopaminergic,
GABAergic and cholinergic systems as well as reticulospinal, corticospinal and rubrospinal
tracts. For example, baclofen, a GABAergic drug is known to attenuate amplitudes of LLFRs
and to increase the threshold for them in patients with enhanced responses [15]. Therefore, our
findings of enhanced LLFR in akathisia may also be the result of loss of or reduction in
inhibitory GABAergic drive. But there exists a major difference in this aspect. Because patients
with akathisia exhibited enhancement in both early and late components whereas GABA seems
to be more in charge of RIII component [4]. Therefore, we may conclude that loss of
GABAergic inhibitory drive may not be the sole underlying factor at least for our cohort of
akathisia. On the other hand, it is suggested that abnormal dopaminergic transmission may be
the underlying pathophysiology in akathisia [1]. LLFR is known to abolish by the effect of
dopaminergic drugs [16] and it is enhanced in the well-known examples of abnormal
dopaminergic systems such as PLMD and RLS [9,10]. However, different pattern and extent of
spinal excitability in akathisia compared to RLS also suggests contribution of other pathways
in akathisia.
The stimulation of the sole triggers activity of TA in almost all healthy subjects whereas
contraction of BF is less evident. Withdrawal pattern mainly depends on the afferent input and
encoded information which is probably abnormal in RLS. Despite the low BF response in
akathisia similar to healthy subjects, patients with RLS had a higher number of BF response
similar to the previous studies [9] which was attributed to the abnormal sensorimotor integration
at the spinal level. On contrary, in akathisia, sensorimotor integration at the spinal level
represented by LLFR, perception of afferent inputs and learned motor task, seems to be
unchanged.
A recent study investigating prepulse inhibition after different types of conditioning
stimuli in RLS reported that there were no changes in patients with RLS compared to healthy
subjects suggesting reserved functional integrity of forebrain interneurons which have roles in
mediation of prepulse inhibition of blink reflex [17] whereas dopamine antagonists are known
to restore prepulse inhibition in patients with schizophrenia [18] and to modulate it in healthy
humans [19]. This difference between akathisia and RLS may worth further investigations
which may contribute further understanding of underlying changes in akathisia.
This study is not without limitations. The lack of patient group using dopamine
antagonist without exhibiting akathisia which may more clearly show the underlying changes
in akathisia and changes due to use of medications. Investigating patients using neuroleptic
medications before development of akathisia may contribute to understanding of its
development and deserves further analysis to be a possible tool to prevent this unwanted effect.
Sample size is another factor, a larger sample size with patients receiving different medications
may also allow subgroup analysis regarding different medications. Although the effects of
various cognitive states have long been known, the effect of anxiety and anxious state is
controversial [4] and there was no impact of anxiety on flexion reflex threshold level or it did
not lead to its enhancement [20,21].
In conclusion, we proved that LLFR is enhanced in akathisia; however, the excitability
changes clearly differ from those found in RLS. These findings are probably due to lack of
inhibition originated in regions other than those known to downregulate in RLS.
References
1. 1. N. Patel, J. Jankovic, M. Hallett, Sensory aspects of movement disorders. Lancet.
Neurol. 13 (2014) 100-112.
2. R.P. Allen, D. Picchietti, W.A. Hening, C. Trenkwalder, A.S. Walters, J. Montplaisi,
Restless Legs Syndrome Diagnosis and Epidemiology workshop at the National
Institutes of Health; International Restless Legs Syndrome Study Group. Restless legs
syndrome: diagnostic criteria, special considerations, and epidemiology. A report from
the restless legs syndrome diagnosis and epidemiology workshop at the National
Institutes of Health, Sleep. Med. 4 (2003) 101-119.
3. A.S. Walters, Restless legs syndrome and periodic limb movements in sleep, in: C.
Guilleminault (ed), Clinical neurophysiology of sleep disorders, Elsevier, Stanford,
2005, p. 273–280.
4. G. Sandrini, M. Serrao, P. Rossi, A. Romaniello, G. Cruccu, J.C. Willer, The lower limb
flexion reflex in humans, Prog. Neurobiol. 77 (2005) 353-395.
5. W.F. Brown, C.F. Bolton, M.J. Aminoff. Neuromuscular Function and Disease, Basic,
Clinical, and Electrodiagnostic Aspects, Elsevier, 2002.
6. R. Taugner, W. Culp, Effect of nicotine on spinal cord of cat; effect of nicotine on
patellar, flexor and extensor reflex before and after administration of interneuron
poisons, Naunyn. Schmiedebergs. Arch. Exp. Pathol. Pharmakol. 220 (1953) 423-432.
7. E. Carstens, J.G. Campbell, Parametric and pharmacological studies of midbrain
suppression of the hind limb flexion withdrawal reflex in the rat, Pain. 33 (1988) 201213.
8. M.A. Hoving, V.H. van Kranen-Mastenbroek, E.P van Raak, G.H. Spincemaille, E.L.
Hardy, J.S. Vles JS; Dutch Study Group On Child Spasticity, Placebo controlled utility
and feasibility study of the H-reflex and flexor reflex in spastic children treated with
intrathecal baclofen, Clin. Neurophysiol. 117 (2006) 1508-1517.
9. W. Bara-Jimenez, M. Aksu, B. Graham, S. Sato, M. Hallett, Periodic limb movements
in sleep: state-dependent excitability of the spinal flexor reflex, Neurology. 54 (2000)
1609-1616.
10. M. Aksu, W. Bara-Jimenez, State dependent excitability changes of spinal flexor reflex
in patients with restless legs syndrome secondary to chronic renal failure, Sleep. Med.
3 (2002) 427-430.
11. American Psychiatric Association. Diagnostic and Statistical Manual of Mental
Disorders, DSM-IV-TR 4th Edition, Wilson Boulevard, Arlington, VA, 2000
12. T.R. Barnes, A rating scale for drug-induced akathisia, Br. J. Psychiatry. 154 (1989)
672-676.
13. A. Gündüz, N.U. Adatepe, M.E. Kiziltan, D. Karadeniz, O. Uysal, Circadian changes in
cortical excitability in restless legs syndrome, J. Neurol. Sci. 316 (2012) 122-125.
14. B.T. Shahani, R..R Young, Human flexor reflexes, J. Neurol. Neurosurg. Psychiatry. 34
(1971) 616-627.
15. M. Parise, L. García-Larrea, P. Mertens, M. Sindou, F. Mauguière, Clinical use of
polysynaptic flexion reflexes in the management of spasticity with intrathecal baclofen,
Electroencephalogr. Clin. Neurophysiol. 105 (1997) 141-148.
16. G. Paradiso, F. Khan, R. Chen, Effects of apomorphine on flexor reflex and periodic
limb movement, Mov. Disord. 17 (2002) 594-597.
17. F.E. Leon-Sarmiento, E. Peckham, D.S. Leon-Ariza, W. Bara-Jimenez, M. Hallett,
Auditory and Lower Limb Tactile Prepulse Inhibition in Primary Restless Legs
Syndrome: Clues to Its Pathophysiology, J. Neurophysiol. 32 (2015) 369-374.
18. J.K. Wynn, M.F. Green, J. Sprock, G.A. Light, C. Widmark, C. Reist, S. Erhart, S.R
Marder, J. Mintz, D.L. Braff, Effects of olanzapine, risperidone and haloperidol on
prepulse inhibition in schizophrenia patients: a double-blind, randomized controlled
trial, Schizophr. Res. 95 (2007) 134-142.
19. P.A. Csomor, R.R. Stadler, J. Feldon, B.K. Yee, M.A. Geyer, F.X. Vollenweider,
Haloperidol differentially modulates prepulse inhibition and p50 suppression in healthy
humans stratified for low and high gating levels, Neuropsychopharmacology. 33 (2008)
497-512.
20. D.J. French, C.R. France, J.L. France, L.F. Arnott, The influence of acute anxiety on
assessment of nociceptive flexion reflex thresholds in healthy young adults, Pain 114
(2005) 358-363.
21. E.L. Terry, K.L. Kerr, J.L. DelVentura, J.L. Rhudy, Anxiety sensitivity does not
enhance pain signaling at the spinal level, Clin J Pain. 28 (2012) 505-510.
Figure 1. Representative example of the positioning of stimulation electrode on the sole.
Stimulation of sole leads to stimulation of nociceptive afferents which in turn activates motor
neurons in spinal cord providing dorsiflexion of foot and flexion of knee.
Figure 2. Representative examples of lower limb flexor reflex in a healthy subject (A), in a
patient with akathisia (B) and in a patient with restless legs syndrome (C). Prolonged responses
are remarkable in patients with akathisia or restless legs syndrome (Arrows indicate the time of
stimulation; TA, tibialis anterior muscle; BF, biceps femoris muscle).
100
90
80
70
60
50
40
30
20
10
0
Akathisia
RLS
R/Early TA
R/Late TA
Healthy subjects
R/BF-LH
Figure 3. Spatial spread of lower limb flexor reflex. The probability of responses were
comparable between patients with akathisia and healthy subjects suggesting similar spatial
spread in these two groups (TA, tibialis anterior; BF-LH, biceps femoris-long head).
Table 1. Demographical findings of all participants.
Age, year
Gender, M/F, n
Akathisia
n=7
38.0±9.7
4/3
RLS
n=12
42.0±11.4
4/8
Healthy subjects
n=17
47.1±11.9
8/9
p
0.183
0.576
Table 2. Mean onset latency, amplitude and probability of lower limb flexor reflex
responses.
Early TA response
Latency, ms
Amplitude, µV
Duration, ms
Probability, n (%)
Late TA response
Latency, ms
Duration, ms
Probability, n (%)
BF response
Latency, ms
Probability, n (%)
Akathisia
n=7
RLS
n=12
Healthy subjects
n=17
p
42.2±12.2
1577.6±951.9
312.5±209.0
6 (85.7)
41.3±10.9
1957.3±1377.7
190.4±194.5
11 (91.7)
44.3±8.5
1097.0±649.7
58.6±20.8
16 (94.1)
0.689
0.124
0.011*
0.795
111.0±21.8
235.0±229.5
5 (71.4)
96.3±11.1
190.6±188.5
9 (75.0)
127.8±24.7
56.7±20.0
10 (58.8)
0.031**
0.023*
0.632
144.0±49.6
3 (42.9)
103.3±32.1
11 (91.7)
139.3±55.3
6 (35.3)
0.085
0.008
* posthoc analysis revealed the major difference was between akathisia groups and healthy subjects as well as RLS group and healthy
subjects.
**posthoc analysis revealed the major difference was between RLS group and healthy subjects.
TA, tibialis anterior; BF, biceps femoris; restless legs syndrome, RLS