Pathophysiology of Restless Legs Syndrome and Periodic Limb Movements in Sleep

1
Pathophysiology of Restless Legs Syndrome and
Periodic Limb Movements in Sleep

• Arthur S. Walters, M.D.
• Professor of Neurology
• Associate Director Sleep Medicine
• Vanderbilt University School of Medicine
• Nashville, Tennessee

2
Disclosures for last year

• Consultant to UCB Pharma on adult and pediatric
  RLS.

3
The International Restless Legs Syndrome Study
Group (IRLSSG)

• Consists of over 130 physicians and scientists
  from 17 countries dedicated to research on
  Restless Legs/Periodic Limb Movements in Sleep.
• Developed a consensus definition of RLS published
  originally in 1995 and updated for better clarity
  and republished in 2003.
• Same definition also was accepted by
  International Classification of Sleep Disorders
  Version 2.

4
Obligate Clinical Features RLS

• An urge to move the legs usually accompanied or
  caused by uncomfortable and unpleasant sensations
  in the legs.
• The urge to move or unpleasant sensations begin
  or worsen during periods of rest or inactivity
  such as lying or sitting.
• The urge to move or unpleasant sensations are
  partially or totally relieved by movement such as
  walking or stretching, at least as long as the
  activity continues.
• The urge to move or unpleasant sensations are
  worse in the evening or night than during the day
  or only occur in the evening or night.

5
Supportive Clinical Features of RLS

• Positive Family History of RLS.
• Improvement with dopaminergic therapy.
• Periodic Limb Movements In Sleep (PLMS).
• At least 4 movements in a row at least 8 mcv
  high, 0.5-10 seconds in duration and 5-90
  seconds apart.
• Periodic Limb Movements in Wakefulness (PLMs).

6
---

• PLMS

7
Associated Features of RLS

• Sleep Disturbance.
• Neurological Examination Normal in idiopathic
  or familial cases. Evidence of Peripheral
  Neuropathy or Radiculopathy in secondary cases.
• Serum ferritin lt 50 mcg/L.
• Clinical course
• Most patients middle to older age, but may be
  seen in children.
• Usually progressive but static course may be
  seen. Remissions of a month or more may be seen
  in 15 of cases.

8
Therapy of RLS

• First line Dopaminergic Agents
• L-Dopa, ropinirole, pramipexole
• Second line
• Opioids
• Anticonvulsants gabapentin, pregabilin
• Benzodiazepines clonazepam, diazepam
• Iron Therapy

9
Pathogenetic Mechanisms and Hypotheses
10
Summary of talk

• Therapy has led to pathogenetic hypotheses-
  Iron, Dopaminergic agonists,Opioids
• We will also review
• Genetics
• Immunologic Hypothesis

11
Summary of talk

• Circadian Control of RLS/PLMS
• Brief Summary of Neurophysiology of RLS/PLMS
• Relationship of RLS/PLMS to Cardiovascular
  Disease

12
Iron pathogenesis and RLS

• Some RLS patients present with iron deficiency.
  Their serum and CSF ferritin levels are low
  (Earley et al, 2000).
• With MRI, Allen et al. (2001) revealed that RLS
  patients iron level is low at the Substantia
  Nigra (SN) area and the MRI iron values of this
  area correlate with RLS severity.
• Conner, et al. (2003) found SN dopaminergic cells
  intracellular Fe level is low in both primary and
  secondary RLS patients with post-mortem autopsy.

13
Dopaminergic system pathogenesis in RLS

• A-11 dopaminergic diencephalo spinal lesions lead
  to a restless rat. DA agonist treatment with
  pramipexole improves restlessness (Ondo et al Mov
  Disord 2000).
• In human RLS in basal ganlia circuit there is
  down regulation of D2 receptors and up regulation
  of Tyrosine Hydroxylase, the rate limiting step
  for dopamine synthesis (Connor et al Brain 2009)

14
Dopaminergic system pathogenesis in RLS

• L-DOPA treatment of RLS pushes symptoms to an
  earlier time of day (Augmentation).
• When augmentation occurs the Dim Light Melatonin
  onset (DLMO) occurs earlier (Garcia-Borreguero et
  al. 2004)
• This supports a role for the circadian pacemaker
  in the production of RLS symptoms and supports a
  role for dopaminergic deficit in the triggering
  of the onset and timing of the onset of RLS
  symptoms.

15
Dopaminergic system pathogenesis in RLS

• L-DOPA
• Suppresses Prolactin
• Increases Growth Hormone
• However, in RLS patients this response
  occurs more at night when RLS symptoms are
  expected to be maximum (Garcia-Borreguero et al
  2004).
• This suggests that a circadian dip in
  dopamine levels at night triggers the symptoms of
  RLS.

16
RLS Genetics
17
(No Transcript)
18
RLS Genetics

• Linkage studies have yielded 5 chromosomes but
  no genes
• Allelic Association studies have yielded 4
  variants of common genes that account for over
  50 of all RLS
• ( Simultaneous publication by group of David Rye,
  NEJM and group of Juliane Winkelmann Nat Genet
  2007)

19
Genetics of RLS/PLMS

• The gene found in common by both Juliane
  Winkelmann and David Rye on chromosome 6 is the
  BTBD9 gene which stands for Broad complex-
  tramtrack-bric-a-brac - domain 9.
• May have something to do with iron metabolism
  since there was a drop is serum ferritin of 13
  for each copy of the variant of the gene.

20
Genetics of RLS/PLMS

• The Meis 1 gene (one of the 6 genes associated
  with RLS) may also have something to do with
  general iron metabolism
• ( Silver et al SLEEP 2010)

21
Genetics of RLS/PLMS

• Another two genes that convey genetic risk
  for RLS were recently discovered Protein
  tyrosine phosphatase receptor type delta gene
  (PTPRD) and the Nitric Oxide Synthase gene (NOS).
• ADHD is more common in RLS/PLMS and vice versa.
• These genes also convey genetic risk for ADHD

22
The Immunologic Hypothesis

• About 15 of RLS patients undergo remissions of a
  month or more that are independent of therapy.
• This is reminiscent of Multiple Sclerosis, an
  immunologically mediated Neurological Disease
  also characterized by exacerbations and
  remissions.

23
The immunologic hypothesis

• RLS is more common in Rheumatoid Arthritis (up
  to 1/3) but not osteoarthritis (4) (Auger et al
  2005 Salih et al 1994).
• RLS is more common in Multiple Sclerosis (up to
  1/3). (Auger et al 2005 Manconi et al 2007).
• Multiple Sclerosis and Rheumatoid Arthritis but
  not osteoarthritis are thought to have an
  immunologic diathesis.

24
RLS and the immunologic hypothesis

• Hornyak M et al. Neurology 2008 70 1620-2.
• Double blind crossover study.
• 10 RLS patients either hydrocortisone 40 mg iv or
  saline placebo.
• Statistically significant Improvement in
  sensory leg discomfort (p 0.032).

25
The immunologic hypothesis

• Gamignani et al Mov Disord 2006
• RLS in 29 of 97 consecutive patients with
  polyneuropathy
• More often sensory neuropathy of small fiber
  type (15 of 29 vs. 16 of 68P 0.009).
• In the RLS group, dysimmune neuropathies,
  significantly more frequent (11 of 29 vs. 10 of
  68 P 0.016).

26
Weinstock and Walters

• Background
• Irritable Bowel Syndrome (IBS) is frequently
  associated with Small Intestinal Bacterial
  Overgrowth (SIBO).
• Small Intestinal Bacterial Overgrowth can be
  detected by a positive Lactulose Breath Test
  (LBT).

27
Weinstock and Walters

• RLS Group - 33
• General Population Controls -25
• 2nd control group were GI disease was excluded -
  30
• 27 RLS patients had Irritable Bowel Syndrome
  (IBS) as opposed to 4 in the general population
  controls (p.0326)

28
Weinstock and Walters

• A positive LBT was found in 67 of RLS patients
  as opposed to 28 of General Population controls
  (p .0074) and10 of Completely Healthy
  controls
• This indicates that a full 57 of positive LBTs
  and therefore Small Intestinal Bacterial
  Overgrowth can be attributed to RLS

29
Prevalence of SIBO in RLS

• RLS gt controls with 2003 and 2009 criteria

Plt0.001 vs. controls for both criteria
(30/39 patients screened randomized for
treatment)
30
The immunologic Hypothesis

• SIBO ---?cytokines and/or translocation of
  lipopolysaccharides ----? increased Hepcidin---?
  abnormal central processing of iron.
• Alternatively SIBO may induce a direct
  auto-immune attack on the brain and/or peripheral
  nervous system in RLS.

31
RLS and the immunologic hypothesis

• We did a survey of the 40 conditions frequently
  associated with RLS.(Secondary forms of RLS)
• 30 of the 40 associated with inflammation or
  immune dysfunction.
• Other Than Multiple Sclerosis and Rheumatoid
  Arthritis other examples include Narcolepsy and
  Celiac Disease.

32
(No Transcript)
33
(No Transcript)
34
Secondary RLS 43 disorders/risk factors

• 21 can have peripheral iron deficiency
• 34 can have systemic inflammation or immune
  disorders 1
• 17 can have small intestinal bacterial overgrowth
  (SIBO) 2,3

Bacterial Overgrowth
Iron deficiency

1. Weinstock et al. Inflamm Bowel Dis. 2009 (in
Press).. 2. Weinstock et al. Dig Dis Sci.
2008531252-1256. 3. Weinstock et al. Dig Dis
Sci. 2009 (in Press).
35
Dopaminergic and opiate system pathogenesis in
RLS

• The effect of the dopaminergic agents and the
  opioids is probably specific to the dopaminergic
  and opiate receptors since blockade of the
  effects of these drugs in RLS treated patients
  brings back the symptoms of RLS.

36
Opioids may act through the Dopaminergic System

• The effect of the opioids is probably mediated
  through the dopaminergic system since opiate
  receptor blockers only reverse the therapeutic
  effects of the opioids in RLS treated patients,
  but dopaminergic receptor blockers reverse the
  therapeutic effect of either the opioids or
  dopaminergic agents in RLS treated patients.
  (Akpinar, et al.1987 Montplaisir, et al. 1990)

Naloxone Pimozide V
V V V
Opioids gtgtgtgt Dopaminegtgtgtgt RLS
37
Opiate Receptor Pet Scanning and RLS

• Abnormalities in Post-synaptic Receptor Binding
  in Medial Pain Pathways (von Spiczak et. al.
  2005).
• Degree of binding correlated negatively with
  severity of RLS
• This further implicates the endogenous opiate
  system in the pathogenesis of RLS and suggests
  that RLS symptoms trigger the release of
  endogenous opiates in the medial pain system.

38
(No Transcript)
39
Opioid Hypothesis for RLS

• Walters AS, Ondo WG, Zhu W, Le W. Journal of the
  Neurological Sciences 279 62-65 2009. .
• Post-mortem study of 5 RLS patients and 6
  controls.
• In thalamus Beta endorphin reduced by 37.5 (p
  .006, effect size 2.16).
• In thalamus Met enkephalin reduced by 26.4 (p
  .028, effect size 1.58).

40
Possible in vitro model of RLS(Sun et al. APSS
2003Winner of honorable mention Young
Investigator Award)

• Iron deficiency causes cell death in the
  substantia nigra in rats
• The dying cells in the substantia nigra are
  primarily dopaminergic although glial cells are
  affected as well.
• Opioids protect against the dopaminergic cell
  death under conditions of iron deprivation.

41
(No Transcript)
42
(No Transcript)
43
Summary of animal model of RLS

• Low Fe High opioids and Low Fe
• V V
• V V
• V V
• DA Cell Death Less DA Cell Death

44
Mu-Opioid Receptor Knockout Mice Genotyping

• Building up a colony at UAB
• Genotypes identified by PCR.
• Mating heterozygous mu-Opioid receptor knockout
  mice to obtain homozygous mutant mice.

/- /- / /-
230 bp
PGK
Mutant Allele
Primer 1
Primer 2
230 bp
WT Allele
Primer 2
45
Results to date of Mu opiate receptor deficient
mouse 1

• Animals are more hyperactive during the sleep
  period as is characteristic of RLS (in mice this
  is during the day).
• Mice are more sensitive to pain as has been
  previously described in human RLS by Stiasny
  Kolster et al 2004

46
Results to date of Mu opiate receptor deficient
mouse 2

• Serum iron is low compared to control mice.
• Next Step is to sacrifice brains to see if there
  is brain iron deficiency, downregulation of D2
  receptors and upregulation of Tyrosine
  Hydroxylase as in human RLS.

47
New Project

• CSF study of RLS patients and controls to look
  for alterations in
• Beta endorphin
• Met enkephalin
• Leu enkephalin

48
Early issue on definition of RLS

• Are RLS patients worse at night only because they
  are lying down or are the symptoms under
  circadian control?

49
(No Transcript)
50
(No Transcript)
51
Our Early Circadian Studies

• Patients were still worse at night even though
  they were lain down semi-continuously both day
  and night.
• This effect was largely independent of sleep or
  sleep deprivation.

52
Our Early Circadian Studies

• This suggests that criteria 3 and 4 for RLS are
  largely independent of one another and should be
  kept separate and not combined
• Worsening at rest, e.g. sitting or lying
• Worsening at night

53
New Project

• Bright Light at night to push the symptoms of
  RLS/PLMS to a later time of night.
• This will more definitively prove that RLS/PLMS
  is controlled by a circadian rhythm
• May be used as a therapy in situations where
  daytime symptoms are prominent.

54
Neuroanatomy

• What structures are involved in RLS/PLMS?
• RLS patients have more gray matter in the
  pulnivar of the thalamus by high resolution
  T1-weighted MRI (Etgen et. al. 2005).
• By functional MRI during the sensory symptoms of
  RLS the thalamus and cerebellum are activated and
  during the PLMs the brain stem is activated near
  the red nucleus (group of Claudia Trenkwalder).

55
Peripheral and Central Mechanisms in RLS

• There was decreased temperature perception in
  RLS patients with and without peripheral
  neuropathy.
• This suggests that there is a peripheral
  processing abnormality in patients with RLS due
  to peripheral neuropathy and a central sensory
  processing abnormality in patients with
  idiopathic RLS that may lead to the sensory
  symptoms of RLS (Schattschneider et al J Neurol
  2004).

56
Peripheral and Central mechanisms in RLS (cont)

• Static hyperalgesia as tested by pricking pain
  was abnormal in idiopathic RLS. This is normally
  thought to be suggestive of abnormalities in the
  peripheral nervous system. The peripheral
  processing abnormality may take place by some
  unknown mechanism since these patients did not
  have peripheral neuropathy.
• No peripheral neuropathy in these patients and
  the static hyperalgesia was normalized by
  long-term dopaminergic treatment implicating a
  central processing abnormality in RLS
  (Stiasny-Kolster et. al. 2004).

57
Neurophysiology

• Blink reflex
• H reflex
• Elicitation of the flexor response by
  stimulation of the foot and comparison to PLMS.
• Brainstem auditory evoked responses
• Somatosensory evoked responses
• Central magnetic stimulation
• Paired transcranial magnetic stimulation
• Examination of the cortical silent period.

58
Neurophysiology

• Brain stem generator inhibits lumbosacral
  generator for PLMS under normal circumstances.
• Lumbosacral generator becomes disinhibited under
  conditions of complete spinal cord transection or
  when sensory input from RLS overcomes the
  inhibition from the brain stem. PLMS ARE THEN
  PRODUCED.

59
Integrated Motor and Sensory Model

• Brain Stem ( - )
• V
• V
• --------------------------spinal cord
  transection V
• V Neuropathy/Radiculopathy
• V V
• V V
• LS spinal generator for PLMS ltltltltltlt RLS ()

60
SLEEP, Vol. 32, No. 5, 2009 589-597
61
Relationship of RLS to Medical Conditions

• Four large epidemiologic studies from groups of
  Jan Ulfberg, John Winkelman and Barb Phillips all
  show
• Hypertension
• 1.5 x more likely in RLS--
• Heart Disease
• 2.5 x more likely in RLS--

62
RLS and Stroke

• Elwood et al., J Epidemiol Community Health
  20066069-73.
• 1986 men aged 55-69 years
• After 10 years, 107 with RLS experienced an
  ischemic stroke
• Compared to controls relative odds of an ischemic
  stroke was 1.67 (1.07-2.60) P 0.024

63
PLMS and daytime hypertension

• In patients with daytime hypertension, there is a
  direct correlation between the number of PLMS and
  the severity of the hypertension (Espinar-Sierra
  1997).
• Recently reproduced in a much larger cohort by
  group of David Rye.

64
(No Transcript)
65
Autonomic Nervous System and PLMS

• Group of Jacques Montplaisir Pennestri et al.
  (Neurology 2007)
• Our group -- Siddiqui et al (Clin Neurophysiol).
• Fake PLMs little rise in BP. Real PLMs
  associated with rises in BP that are much greater
  (on up to avg 19 mm Hg for RRLMS) suggesting that
  autonomic nervous system may play a role in
  pathogenesis of PLMS.
• RRLMSgtPLMS with arousalsgt PLMS without arousalsgt
  PLMWgt Fake PLMs.

66
Blood pressure changes associated with PLMS
Pennestri et al., Neurology (2007)
67
Autonomic arousals occur in tandem with PLMS (in
RLS)
Pennestri et al., Neurology (2007)
68
Autonomic Nervous System and RLS/PLMS continued

• 80 of patients with RLS have PLMS.
• Previous literature shows that hypertension and
  heart disease are associated with RLS.
• Perhaps rise in BP associated with PLMS are the
  cause of heart disease and perhaps stroke
  associated with RLS.

69
Hypothetical spinal cord positive feedback
mechanism mediating dopamine responsive Restless
Legs Syndrome

1. DA inhibits preganglionic sympathetics, thus, in
   its absence, basal sympathetic tone may increase.
2. Increased adrenaline via innervation of skeletal
   muscle, in turn, might irritate muscle spindles.
3. The resulting enhanced input from pain-encoding
   high threshold muscle afferents in lamina I are
   insufficiently suppressed in the absence of DA or
   D2-like receptors.

Clemens, Rye and Hochman, Neurology 67 125-130
(2006)
70
Hypertension and RLS/PLMS

• Interestingly vasodilators were among the first
  treatments used by Ekbom in the 1940s and 1950s
  for RLS
• Phenoxybenzamine an alpha adrenergic receptor
  blocker was used to treat PLMS and normalized the
  peripheral pulse responses (Ware et al Sleep 1988)

71
Hypertension and RLS/PLMS

• Controlled studies are needed to
• Verify the response of RLS to antihypertensives.
• Verify the response of PLMS to
  antihypertensives.

72
Stroke and RLS

• RLS patients with no prior history of stroke
• Tendency toward an increased number of silent
  strokes compared to controls.
• However, study was small and much need for
  statistical correction due to other stroke risk
  factors

73
Stroke and RLS/PLMS

• Is RLS/PLMS mediated by its increased
  association with hypertension or heart disease ?
• Or is RLS/PLMS an independent risk factor for
  stroke?

74
New Project

• Repeat study looking for silent stroke by MRI in
  RLS patients
• However, recruit patients with no other stroke
  risk factors.
• Will also correlate PLMS with number/size of
  silent strokes.

75
(No Transcript)
76
Summary of pathophysiologic mechanisms for
RLS/PLMS

• Low dopamine, iron and opioids may contribute to
  the pathogenesis of RLS/PLMS
• The discovery of susceptibility genes for RLS
  will give us new hypotheses for the pathogenesis
  of RLS.
• Immunologic mechanisms represent a possible new
  frontier for RLS research.

77
Summary of pathophysiologic mechanisms for
RLS/PLMS

• RLS/PLMS are under circadian control
• Suprasegmental Disinhibition may contribute to
  the pathogenesis of PLMS
• Increased sympathetic tone in the absence of
  dopaminergic innervation may explain the
  increased prevalence of hypertension, heart
  disease and perhaps stroke in RLS/PLMS probably
  more relevant to adult RLS.