Evolution of Neurological, Neuropsychological and

Sleep-Wake Disturbances After Paramedian

Thalamic Stroke
Dirk M. Hermann, MD; Massimiliano Siccoli, MD; Peter Brugger, PhD; Karen Wachter, MD;
Johannes Mathis, MD; Peter Achermann, PhD; Claudio L. Bassetti, MD

Background and Purpose—The clinical features and natural course of paramedian thalamic stroke is poorly known. The
aim of this study was to characterize the evolution of neurological, neuropsychological, and sleep–wake deficits after
paramedian thalamic stroke.
Methods—Forty-six consecutive patients, aged 48.4⫾16.6 years, were studied. Fourteen had bilateral, 16 left-sided, and
16 right-sided lesions. Assessment included neurological examinations, estimation of sleep needs, formal neuropsy-
chological tests (n⫽27), and polysomnographies (n⫽31). Functional outcome was followed up over 1 year in 31 patients
with the modified Rankin Scale and Barthel index.
Results—Oculomotor palsy (76% of patients), mild gait ataxia (67%), deficits of attention (63%), fluency and error control
(59%), learning and memory (67%), and behavior (67%) were common in the acute stroke phase. Outcome was
excellent with right-sided infarcts but mostly incomplete with bilateral and left-sided lesions. This was mainly related
to persistent frontal lobe-related and cognitive deficits found in 100% bilateral and 90% left-sided, but only 33%
right-sided strokes. Initially, hypersomnia was present in all patients associated with increased stage 1 sleep, reduced
stage 2 sleep, and reduced sleep spindles. Sleep needs improved in patients with bilateral and almost disappeared with
unilateral lesions after 1 year. Sleep architecture remained abnormal with the exception of sleep spindles that increased.
Conclusions—Whereas neurological deficits and hypersomnia recover to large extent in patients with paramedian thalamic
stroke, the frontal lobe-related and cognitive deficits, which are mainly linked with bilateral and left-sided lesions, often
persist. As such, stroke outcome is better in right-sided than bilateral or left-sided infarcts. (Stroke. 2008;39:62-68.)
Key Words: attention 䡲 cognition 䡲 hypersomnia 䡲 ischemic brain infarct 䡲 thalamus

P aramedian thalamic stroke presents with a stereotypical

clinical picture often initiated by a short-lasting loss of
consciousness followed by a hypersomnolent state in which
sleep needs are markedly increased and the patient’s ability to
maintain attention is reduced.1– 4 Patients typically also ex-
hibit neurological symptoms such as vertical gaze palsy and
other oculomotor deficits, Horner’s syndrome, dysarthria,
gait ataxia, and mild sensorimotor deficits.3,5–10
With the improvement of consciousness, neuropsycholog-
ical problems such as executive, learning, and memory
deficits become more evident.1,3,5,6,9,11–15 Sleep studies have
shown an increased sleep duration with predominance of
superficial sleep stages, increased sleep fragmentation, and
decreased sleep spindles.2– 4 Hypersomnia has been consid-
ered to result from the disruption of thalamic sleep-generating
and arousal-maintaining mechanisms.1–3
Few reports suggested that neuropsychological abnormal-
ities and sleep–wake disturbances may persist over several
months poststroke.3,5,15 Overall, the prevalence of neurolog-
ical, neuropsychological, and neurophysiological abnormali-
ties as well as the long-term consequences of paramedian
thalamic stroke is poorly known. The aim of this study was to
analyze the development of clinical, neuropsychological, and
sleep–wake abnormalities in the acute and postacute stroke
phase and to identify factors influencing stroke recovery.
Materials and Methods
Neurological Examinations
Forty-six consecutive patients with ischemic paramedian thalamic
stroke referred to the University Hospitals of Berne (1992 to 2000)
and Zurich (2001 to 2004) were studied. All patients were seen by
one of the authors (C.L.B.). A detailed medical history and neuro-
logical examination were always performed. Severity of neurological
impairment was estimated by the Scandinavian Stroke Scale. In all
but 2 patients, MRI was performed within 2 weeks after stroke.
Diagnostic workup always included blood tests, electrocardiogram,

Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz.
Received May 24, 2007; accepted June 7, 2007.
From the Department of Neurology (D.M.H., M.S., P.B., K.W., C.L.B.), University Hospital, Zurich, Switzerland; the Department of Neurology (J.M.),
Inselspital Berne, Berne, Switzerland; and the Institute of Pharmacology and Toxicology (P.A.), University of Zurich, Zurich, Switzerland.
Correspondence to Dirk M. Hermann, MD, Department of Neurology, University Hospital Zurich, Frauenklinikstrasse 26, CH-8091 Zürich,
Switzerland. E-mail dirk.hermann@usz.ch
© 2007 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.107.494955

Downloaded from http://stroke.ahajournals.org/62

at NY UNIV MED CTR LIBR on May 25, 2015
 Hermann et al
chest x-ray, Doppler/duplex sonography, and transesophageal
echocardiography.
Cardiovascular risk factors were assessed according to the follow-
ing criteria: (1) arterial hypertension: RR repeatedly ⬎140/90
mm Hg or treatment with antihypertensive drugs; (2) diabetes:
morning glucose ⱖ120 mg/dL/HbA1c ⱖ7% or treatment with
antidiabetics; and (3) hypercholesteremia: cholesterol ⱖ250 mg/dL
or treatment with cholesterol-lowering drugs. A positive family
history was considered when at least a first-degree relative had
experienced a stroke or cardiac infarction at the age of ⬍60
(males)/⬍65 (females) years.
Neuropsychological Tests
Thirty-nine neuropsychological examinations were conducted in 27
patients. Eight patients had bilateral, 10 left-sided, and 9 right-sided
lesions. In all patients, a first examination took place in the first
month after stroke (between days 5 and 30; mean interval from
stroke: 14.3⫾10.5 days). Follow-up studies were done in 12 patients
(mean interval: 13.8⫾11.2 months). Four patients had bilateral, 5
left-sided, and 3 right-sided infarcts.
Neuropsychological examinations focused on 5 cognitive do-
mains: (1) attention (Attention Test Battery16; d2 cancellation test17);
(2) fluency and error control (verbal fluency: Thurstone and Thur-
stone18; figural fluency: Regard et al19; both tests evaluate repetitive
namings/rule breaks; color–word interference: Stroop20); (3) mem-
ory with separate aspects of learning and active recall (both verbal
and figural; Rey21; Osterrieth22); (4) speech and language (assessed
clinically, noting presence/absence of dysarthria, word finding def-
icits, phonematic/semantic paraphrasias, language comprehension
difficulties, reading/writing problems). In cases with abnormalities,
the Aachen Aphasia Test23 was used for quantitative and qualitative
deficit grading; and (5) visuospatial functions were evaluated by a
patient’s ability to recognize subjective contours, overlapping figure,
line drawings, faces, and colors. Visuoconstructive abilities and
presence/absence of hemispatial neglect were assessed by having
patients copy simple geometric line drawings and a complex geo-
metric figure.22 In addition to cognition, abnormalities in the
following 6 behavioral–affective domains were recorded: somno-
lence, psychomotor slowing, euphoria/emotional lability, anosogno-
sia, confabulations, and impulsivity. Deficits in the 5 cognitive
domains were rated retrospectively by 2 senior neuropsychologists as
absent, mild, or moderate–severe. Both raters were blinded to an
individual patient’s stroke localization, and in cases of disagreement
(⬍3% of scorings), a consensus rating was obtained.
Sleep History and Studies
A detailed history of sleep habits before stroke was always asked for,
including estimated sleep needs (in hours/day). After the stroke,
sleep needs were also repeatedly assessed at every consultation. In
addition, a total of 56 overnight video polysomnographies were
recorded in 31 patients using Neurofax (Nihon Kohden, Tokyo,
Japan; 1992 to 2000) and Embla (Flaga, Reykjavik, Iceland; 2001 to
2004) sleep recording systems from 10 PM to 5 AM. Eleven patients
had bilateral, 10 left-sided, and 10 right-sided lesions. Twenty
patients received both polysomnographic and formal neuropsycho-
logical studies. Sleep scoring was done visually on 2 computer-based
electroencephalographic (EEG) analysis systems (Nicolet-Ultrasom
2.1, Madison, Wis; Somnologica, Reykjavik, Iceland) according to
Rechtschaffen and Kales.24 Sleep spindles per hour of sleep stage 2
were counted automatically both ipsilateral and contralateral to the
infarct in patients assessed with Nicolet-Ultrasom 2.1.3 In all 31
patients, a first polysomnography took place in the first month after
stroke (starting on day 3; mean interval from stroke: 9.7⫾9.4 days).
In 6 patients, 2 additional follow-up polysomnographies were rec-
orded 12 and 24 months thereafter. These patients were considered
for a more detailed analysis of the evolution of sleep architecture
changes.
In 4 patients (2 bilateral, one left-sided, one right-sided stroke),
nocturnal EEG power spectra were computed during stage 2 sleep
from sleep recordings obtained with a portable polygraphic amplifier
(PS1).25 All patients were recorded once during the first 9 days after
Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
Evolution of Paramedian Thalamic Stroke 63
stroke. In 2 patients (one bilateral and the left-sided stroke), sleep
recordings were repeated in the postacute phase (after 2 and 12
months). In the power spectra obtained, sleep spindle peaks were
calculated.25 The spindle peak size was determined in the F3C3
derivation.
In 15 patients, actigraphies were performed (6 actigraphies in the
first week, 3 after 1 month, 3 after 1 year, and 3 after 2 years).
Actigraphies were registered over 7 days (Motion-Logger; Ambula-
tory Monitoring Inc, Ardsley, NY), and over this recording period,
the average “time asleep”/24 hours was calculated.26 The average
“time asleep” was then correlated with the subjectively estimated
sleep needs.
Follow-Up Examinations
Thirty-one patients were followed up by clinical examination over at
least 12 months (mean duration: 37.6⫾33.4 months; range, 12 to 131
months) and were included in a more detailed analysis of stroke
recovery. Nine patients had bilateral, 10 left-sided, and 12 right-
sided lesions (age 45.7⫾17.4, 42.0⫾14.2, and 42.6⫾16.7 years). All
patients were examined at least once in the subacute phase (3 to 6
months after stroke) and after ⱖ1 year, in most cases more often.
During all control examinations, neurological symptoms, neuropsy-
chological deficits, and sleep–wake habits were carefully inquired
and detailed neurological examinations performed. Outcome was
evaluated using the modified Rankin Scale (mRS; ranging from
0⫽normal to 5⫽requiring constant and full supervision and care)
and Barthel index (0 to 100 points).
Statistics
Clinical scores, sleep, and sleep architecture changes were analyzed
by one-way analysis of variance followed by least significant
differences tests or by 2-way analysis of variance. Follow-up
examinations were evaluated by repeated measurement analysis of
variance. Neuropsychological results were examined by Fisher exact
tests. Relationships between subjectively estimated sleep needs and
actigraphically determined “time asleep” were evaluated by Pear-
son’s correlations. All data are expressed as means⫾SD. Probability
values ⬍0.05 were considered significant.
Results
Patients’ Characteristics
Thirty-four male and 12 female patients with a mean age of
48.4⫾16.6 years (range, 16 to 83 years) were prospectively
included. There were 14 bilateral, 16 left-sided, and 16
right-sided strokes. Three of 46 patients had a personal
history of strokes (2 striatal infarcts, one centrum semiovale
infarct) and 4 patients a history of transient ischemic attacks.
Six patients revealed silent lesions in MRI scans (2 cortical,
one striatal, one anterior thalamic, 2 cerebellar lesions).
Vascular Risk Factors
Nine patients had arterial hypertension and 12 hypercholes-
teremia. Twenty-one patients were smokers (24.5⫾16.5
pack-years) and 16 habitual snorers. Nine patients had a
positive family history for vascular diseases and one patient
diabetes mellitus.
Stroke Etiology
The presumed stroke etiology was arterioarterial embolism in
4 patients (macroangiopathy in 3, vertebral dissection in one
patient) and cardioembolism in 3 patients (atrial fibrillation in
2, mechanical valve replacement in one patient). In 20
patients, a patent foramen ovale was found, 9 of which were
associated either with atrial septum aneurysm or a spontane-
64 Stroke January 2008

ous interatrial shunt. In all other patients, stroke etiology

remained unknown.

Neurological Observations
In 9 patients, stroke onset was accompanied by a short-lasting
loss of consciousness (7 of 14 bilateral, one of 16 left-sided,
one of 16 right-sided strokes), which usually lasted less than
1 hour.
The most common focal neurological findings were ocu-
lomotor deficits (most commonly vertical gaze palsy; n⫽35
of 46⫽76% of patients), anisocoria (14 of 46⫽30%), facial
sensory deficits (4 of 46⫽9%), facial nerve palsy (19 of
46⫽41%), glossopharyngeal palsy/swallowing difficulties (2
of 46⫽4%), lingual palsy (4 of 46⫽9%), dysarthria (24 of
46⫽52%), mild gait ataxia (31 of 46⫽67%), sensory deficits
of arms/ legs (6 of 46⫽13%), mild motor deficits of arms/legs
(24 of 46⫽52%), hypogeusia/hyposmia (4 of 46⫽9%), and
visual deficits (2 of 46⫽4%). The prevalence of focal
symptoms/signs did not depend on the stroke side.

Evolution of Neurological Deficits/Stroke Outcome

Focal neurological deficits mostly recovered within a few
weeks to months from stroke. Only 12 of 31 patients had
focal neurological abnormalities after 1 year (oculomotor
deficits in 6, gait ataxia in one, arm motor deficits in one,
dysarthria/swallowing difficulties in 4 patients). In 4 of these
patients, neurological abnormalities were very mild and not
recognized any more by the patients. The Scandinavian
Stroke Scale, which was 47.6⫾5.2 for bilateral, 49.1⫾3.8 for
left-sided, and 50.3⫾5.2 for right-sided infarcts immediately
after stroke, largely normalized within 1 year (Scandinavian
Stroke Scale: 56.9⫾1.2 for bilateral, 57.4⫾1.3 for left-sided,
and 58.0⫾0 for right-sided injuries; n⫽31).
Stroke recovery was excellent in patients with right-sided
stroke (mRS at 12 months: 0.3⫾0.6; Figure 1). Eight of 11
Figure 1. Evolution of neurological deficits, assessed by the
patients with right-sided lesions exhibited no residual deficits mRS and Barthel index, as well as sleep needs/24 hours in 31
after 1 year. On the other hand, recovery was incomplete after patients with paramedian thalamic stroke (9 bilateral, 10 left-
bilateral (mRS: 2.8⫾1.1) and left-sided (1.7⫾1.2) infarcts sided, and 12 right-sided infarcts), which were followed up over
(Figure 1). The poor recovery was accompanied by persisting at least 12 months poststroke. Note that deficits in daily-life
activities, assessed by the mRS and Barthel index, improve bet-
everyday life deficits in the latter groups (Barthel index after ter in patients with right-sided than bilateral or left-sided lesions.
1 year: 75.6⫾21.1 in bilateral, 91.0⫾16.1 in left-sided, Sleep needs are increased in all patients with immediately after
98.3⫾5.8 in right-sided infarcts; Figure 1). the stroke, although hypersomnia is more pronounced in bilat-
eral than unilateral infarcts. Hypersomnia continuously improves
over 12 months from stroke, but sleep needs remain elevated
Neuropsychological Findings with bilateral infarcts. Data are means⫾SD. n.d., not deter-
mined. *P⬍0.05/**P⬍0.01 compared with bilateral lesions;
Cognitive Deficits †P⬍0.05/††P⬍0.01 compared with left-sided lesions.
Results are summarized in Table 1. Deficits in attention (17
of 27⫽63% of patients), fluency and error control (16 of
27⫽59%), and learning and memory (18 of 27⫽67%) were Serious difficulties in visuospatial processing were almost
most prevalent. All patients with bilateral and most patients absent. A tactile or visual hemineglect was seen in 2 patients
with left-sided infarcts revealed attention deficits. On the with bilateral and one patient with right-sided stroke. One
other hand, only a few patients with right-sided lesions had patient with left-sided stroke revealed constructional apraxia
attributed to visuospatial deficits.
attention abnormalities.
Fluency and error control as well as learning and memory Behavioral Changes
recall were affected in most patients with bilateral and Findings are summarized in Table 2. Eighteen of 27 patients
left-sided infarcts but were rarely compromised after right- (67%) revealed behavioral deficits, patients with bilateral
sided stroke. Speech and language were abnormal in half of stroke always revealing abnormalities in at least one entity.
the patients with bilateral and left-sided lesions, but was Some changes (anosognosia, confabulation) were found only
never affected with right-sided infarcts. in patients with bilateral infarcts, and others (decreased
Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
 Hermann et al Evolution of Paramedian Thalamic Stroke 65

Table 1. Neuropsychological Deficits in the First Month After Stroke in 27 Patients With Paramedian
Thalamic Infarcts
Stroke Laterality

Bilateral Left Right

Moderate to Moderate to Moderate to

Deficit None Mild Severe None Mild Severe None Mild Severe
Attention 0/8 5/8 3/8 4/10 3/10 3/10 6/9 3/9 0/9
Fluency and error control 2/8 3/8 3/8 1/10 3/10 6/10 8/9 1/9 0/9
Memory
Learning 2/8 4/8 2/8 3/10 3/10 4/10 8/9 1/9 0/9
Recall 1/8 4/8 3/8 1/10 3/10 6/10 7/9 1/9 1/9
Speech/language 4/8 3/8 1/8 5/10 2/10 3/10 9/9 0/9 0/9
Visuospatial processing 6/8 2/8 0/8 9/10 1/10 0/10 7/9 1/9 1/9
Data are prevalence numbers for each stroke topography. Numbers in bold significantly differ from the other stroke localizations
(P⬍0.05 on Fisher exact tests).

impulsivity) only after left-sided lesions. Other changes
(psychomotor slowing, euphoria/emotional lability) were
similarly present in all three stroke localizations.
Neuropsychological Findings in the Recovery Phase
Although 3 of 3 patients with right-sided stroke subjected to
neuropsychological follow-up tests exhibited normal results
in their second test, zero of 4 patients with bilateral and only
2 of 5 patients with left-sided infarcts exhibited full recovery.
Attention, fluency, and error control as well as learning and
memory remained abnormal. Partial improvements were seen
in 2 of 4 patients with bilateral and one of 5 patients with
left-sided stroke.
Sleep–Wake Disturbances (sleep needs/hypersomnia)
All patients exhibited increased sleep needs within the first 24
hours poststroke (Figure 1). Hypersomnia was more pro-
nounced in patients with bilateral than left- or right-sided
infarcts (Figure 1). Hypersomnia often markedly improved
within 12 months poststroke. Self-estimated sleep needs
similarly decreased in patients with bilateral, with left-sided,
and right-sided infarcts (Figure 1). After 1 year, sleep needs
remained elevated above prestroke values by 3.1⫾2.1 hours/
day in patients with bilateral (n⫽9 patients), but only by
1.2⫾0.9 hours/day in left-sided (n⫽10) and 0.8⫾1.5 hours/
day in right-sided (n⫽12) stroke.
In patients investigated over more than 12 months (n⫽16
patients with follow-ups ⱖ2 years), sleep needs further
decreased. Sleep needs finally normalized in 9 of 12 patients
with unilateral stroke, but remained elevated in 4 of 4 patients
with bilateral infarcts (abnormal sleep needs: ⱖ1 hour/24
hour above prestroke values).
Validation studies in 15 patients confirmed that the subjec-
tively estimated sleep needs correlated well with the actigraphi-
cally determined “time asleep} (y⫽⫺0.505⫹1.003x; r⫽0.975,
P⬍0.001; y⫽actigraphical ”time asleep,“ x⫽subjectively re-
ported sleep duration).
Sleep–Wake Studies
Polysomnographies
Compared with control subjects examined in our hospital
because of peripheral neurological diseases (n⫽12), patients
with thalamic stroke revealed a nonsignificantly lower sleep
efficiency, a significantly higher proportion of stage 1, and a
significantly lower proportion of stage 2 sleep (Table 3).
Stages 3/4 and rapid eye movement sleep were not signifi-
cantly altered. Sleep spindles were decreased in patients with
paramedian thalamic stroke when compared with control
subjects (Table 3). Patients with unilateral stroke had higher
sleep spindles counts than patients with bilateral infarcts
(P⬍0.05). In 6 patients recorded at least 3 times after stroke,
the proportion of sleep stages did not significantly change
from the acute to chronic phase. Sleep spindles, however,
increased over 2 years from stroke (Table 4).
Electroencephalographic Power Spectra
Spectral EEG analysis performed in the first week after stroke
revealed either a small (0.31 [lg power/␮V2/0.25 Hz]; patient
1) or absent (0.00; patient 2) frontocentral sleep spindle peak
in 2 patients with bilateral paramedian thalamic infarcts

Table 2. Behavior Changes in the First Month After Stroke in 27 Patients With Paramedian Thalamic Infarcts
Psychomotor Euphoria/Emotional Decreased Increased No
Somnolence Slowing Lability Anosognosia Confabulations Impulsivity Impulsivity Changes
Bilateral 2/8 1/8 2/8 1/8 2/8 0/8 1/8 0/8
Left 0/10 3/10 1/10 0/10 0/10 3/10 0/10 3/10
Right 0/9 1/9 1/9 0/9 0/9 0/9 1/9 6/9
Data are prevalence numbers for each stroke topography. Numbers shown in bold significantly differ from the other stroke localizations (P⬍0.05
in Fisher exact tests).

Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
66 Stroke January 2008

Table 3. Effects of Stroke Topography on Sleep Architecture Table 4. Evolution of Sleep Architecture During Stroke
in the Acute Phase (first month) After Stroke in 31 Patients Recovery in 6 Patients With Paramedian Thalamic Infarcts
With Paramedian Thalamic Infarcts and 12 Control Subjects
First Month 12 Months 24 Months
Control Stroke Laterality Recording time, min 382⫾56 389⫾53 406⫾34
Subjects§ Bilateral Left Right Sleep efficiency, % 88⫾10 92⫾8 87⫾9
(n⫽12) (n⫽11) (n⫽10) (n⫽10)
Stage 1, % 44⫾22 44⫾19 40⫾24
Age 47.0⫾4.0 49.2⫾15.9 40.0⫾13.1 52.4⫾14.8
Stage 2, % 32⫾8 36⫾14 31⫾17
Recording time, min 453⫾42 402⫾40 444⫾97 435⫾64
Stage 3/4, % 15⫾15 8⫾9 10⫾13
Sleep efficiency, % 88⫾9 86⫾9 89⫾10 82⫾11
Rapid eye movement sleep, % 9⫾4 10⫾4 10⫾9
Stage 1, % 15⫾7 31⫾24* 29⫾23 35⫾14*
Spindle density, hours⫺1
Stage 2, % 57⫾9 34⫾16† 43⫾21* 36⫾13†
Lesion side 129⫾113 164⫾105 206⫾101*
Stage 3/4, % 9⫾8 9⫾11 13⫾8 9⫾9
Contralateral 117⫾94 157⫾76 200⫾100
Rapid eye movement 19⫾5 19⫾14 11⫾6 12⫾6
Data are means⫾SD.
sleep, %
*Repeated-measurement analysis of variance with significant time effect
Spindle density, (F1,10⫽6.319; P⬍0.05).
hours⫺1㛳
Lesion side 308⫾13 80⫾56¶ 169⫾123‡ 140⫾88 turbances after paramedian thalamic stroke. Our main find-
Contralateral 168⫾98‡ 122⫾81 ings are (1) long-term recovery after paramedian thalamic
Data are means⫾SD. stroke is significantly better in right-sided than in bilateral
*P⬍0.05/†P⬍0.01 compared with control subjects. and left-sided infarcts; (2) bilateral and left-sided but not
‡P⬍0.05 compared with bilateral stroke patients (one-way analysis of right-sided strokes regularly present with deficits in executive
variance followed by least significant difference tests).
functions and memory, which may persist in the subacute
§Control group of hospitalized patients with peripheral neurological
diseases. phase and explain the unfavorable outcome; and (3) post-
㛳n⫽6 for bilateral, n⫽8 for left-sided, and n⫽7 for right-sided infarcts stroke hypersomnia improves considerably within months,
(recordings done with Neurofax; Nihon Kohden); values in control subjects whereas sleep EEG changes, besides a moderate improve-
taken from Bassetti C, Mathis J, Gugger M, Lovblad KO, Hess CW. Hypersomnia ment in sleep spindle activity, may remain unchanged for
following paramedian thalamic stroke: a report of 12 patients. Ann Neurol. years.
1996;39:471– 480.
¶Mean value of both sides.
In this study, patients were consecutively included. As a
typical university department of neurology diagnoses be-
(Figure 2). In both patients, slow wave activity was also tween 2 and 5 paramedian thalamic strokes each year,
markedly reduced (Figure 2). Two patients with unilateral patients in this study were collected over more than a decade.
stroke (one left-sided, one right-sided lesion) revealed a During this time, we systematically investigated neurological,
slightly higher spindle peak above the infarcted hemisphere neuropsychological, and sleep–wake deficits wherever possi-
(both 0.40). Follow-up EEGs 2 and 12 months, respectively, ble in a routine clinical setting. We are therefore confident
after the stroke (n⫽2) exhibited an increase of the spindle that our data set is representative for this stroke syndrome. In
peak both in one patient with bilateral and left-sided infarct our study, not all patients received all 3 kinds of data analysis
(both 0.64; Figure 2). Average spindle peak size in fronto- (ie, clinical recovery, neuropsychological, and sleep assess-
central derivations of hospitalized controls was 0.59⫾0.06.25 ments), which was impossible for logistical reasons related to
patient care. We compensated this lack of a uniform patient
Predictors of Stroke Recovery collective by sampling sufficiently high patient numbers in all
To identify factors predicting stroke recovery, we next 3 data subsets providing the statistical power to draw reliable
compared patients exhibiting complete or almost-complete conclusions regarding the previously mentioned hemispheric
(mRS ⱕ1) and incomplete (mRS ⱖ2) recovery 1 year after influences on stroke outcome, neuropsychological deficits,
stroke. Compared with patients exhibiting complete or and sleep.
almost-complete recovery, patients with poor outcome were
older (P⬍0.05) and more frequently exhibited stroke involve- Stroke Recovery Is Better in Right-Sided Than in
ment of the left thalamus (P⬍0.01). Besides, stroke presented Bilateral and Left-Sided Infarcts
more often with an initial loss of consciousness (P⬍0.05). Our finding that stroke recovery, assessed by the mRS,
Patients with incomplete recovery also revealed higher sleep depends on the lesion side is new. In fact, detrimental
needs immediately after their stroke (P⬍0.01) and after 1 outcomes have previously been reported mainly after bilateral
year (P⬍0.01). The severity of neurological symptoms, paramedian thalamic lesions.9,13,15,27 The unfavorable out-
measured by the Scandinavian Stroke Scale, sleep efficiency, come in bilateral and left-sided strokes in our study was
and sleep architecture, did not correlate with stroke outcome. clearly not attributed to persistent focal neurological abnor-
malities. In fact, only 12 of 31 patients exhibited focal
Discussion neurological signs after 1 year, and abnormalities in these
Our article offers a detailed and comprehensive analysis of patients were often mild and no more recognized by the
focal neurological, neuropsychological, and sleep–wake dis- patients. These observations challenge previous reports, in
Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
 Hermann et al
which persistent neurological deficits have been
emphasized.28,29
Cognitive Deficits Are More Prominent After
Bilateral and Left-Sided Infarcts
The poor outcome in patients with left-sided paramedian
thalamic lesions was well explained by impairments of
frontal lobe-related functions such as fluency and error
control as well as of learning, memory, and language, which
were more severe in patients with bilateral and left-sided
lesions in whom they also recovered incompletely. Previous
studies, mostly in small series, have emphasized severe
memory deficits after bilateral infarcts.9,13,15,27 A few case
reports have also described persistent verbal and visual
amnesia after left-sided lesions.30 –32
Cognitive changes in patients involving the left-sided
thalamus may anatomically be explained by lesions of the
hippocampal memory loop (also known as Papez circuit),
which connects the hippocampus with the mammillary bodies
and anterior thalamus.33 Brain lesions involving the hip-
pocampal loop induce a failure of information storage, ie,
amnesia.34,35 The observation that fluency and error control,
learning, and memory were most severely affected after
infarcts involving the left thalamus is in line with a laterality
of the hippocampal loop for these functions.
Attention and Behavioral Deficits Are Most
Prominent After Bilateral Infarcts
Attention and behavioral changes have been reported after
bilateral and, less commonly, unilateral paramedian thalamic
stroke. Patients may exhibit “pseudopsychiatric” syndromes
with emotional lability, disinhibition, frontal-like behavior,
personality changes, or delirium.9,13,15 In our series, attention
and behavioral deficits were in fact present in all patients with
bilateral and unilateral paramedian thalamic lesions. In pa-
tients with bilateral infarcts, behavioral changes were some-
times profound and attention deficits mostly did not recover
completely in the postacute phase. Unfortunately, neuropsy-
chological follow-up studies were performed only in a rela-
Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
Evolution of Paramedian Thalamic Stroke 67
Figure 2. Hypnogram and average EEG
power spectrum of stage 2 sleep (F3C3
derivation) of a 62-year-old male patient
with a bilateral paramedian thalamic
infarct in the acute (4 days after stroke)
and subacute (2 months later) stroke
phase. The size of the spindle peak was
determined according to Gottselig et
al.25 It increased from 0.31 (lg[power/
␮V2/0.25Hz]) at 14.75 Hz (acute phase)
to 0.64 at 14.0 Hz (subacute phase).
tively small number of patients. This precludes more detailed
conclusions about the recovery of the cognitive and behav-
ioral deficits.
Although attention deficits in patients with bilateral stroke
are most likely related to the interruption of ascending
reticular arousal systems (see also subsequently), the behav-
ior changes may reflect a dysfunction of the lateral limbic
loop, which links the dorsomedial thalamus with the orbito-
frontal cortex and amygdala.33 In contrast to hippocampal
loop lesions causing pure amnesia,34,35 lesions of the lateral
limbic loop typically evoke amnesia associated with behavior
changes and confabulation tendency.36 Our observation that
behavioral changes were particularly prominent after bilateral
infarcts does not exclude the possibility reported in the
literature that in single patients, unilateral lesions of the
lateral limbic loop may exhibit behavioral abnormalities at
least in the peracute phase (up to 4 days from stroke), which
was not assessed by our clinical neuropsychologists.
Sleep Architecture Does Not Recover Despite
Improvement of Sleep Needs/Sleep Spindles
Sleep architecture did not show any improvement over time
in this study despite a decrease in sleep needs and some
increase in sleep spindle activity. In fact, stage 1 sleep, which
was increased in the acute stroke phase independent of the
stroke hemisphere affected, as well as stage 2 sleep and sleep
spindles, which were reduced, remained abnormal represent-
ing a sort of “sleep EEG signature” of paramedian thalamic
infarcts. Our data are in line with previous observations in a
small patient series in which persistent disturbances in sleep
patterns were seen.3 The absent recovery of sleep architecture
may in parts reflect an insufficient improvement of mecha-
nisms underlying sleep spindles production and more gener-
ally sleep consolidation.
Relationship Between Sleep–Wake Disturbances
and Neuropsychological Deficits
The increased sleep needs in patients with bilateral parame-
dian thalamic infarcts persisting over months or years2– 4
68 Stroke January 2008
together with the long-lasting daytime apathy1,3,5–7,15 previ-
ously suggested that sleep disturbances and neuropsycholog-
ical changes in this syndrome are linked. The dissociation in
our study between sleep architecture and neuropsychological
disturbances challenges the hypothesis of a simple relation-
ship between sleep EEG and cognitive impairments.
However, elevated sleep needs were predictive for a
moderate to poor stroke outcome, which in turn was associ-
ated with persistent attention disturbances, besides impair-
ments of executive functions, learning, and memory. This
observation could give support to the hypothesis that hyper-
somnia is linked with attention rather than cognitive changes,
pointing toward a common dysfunction in arousal systems.
As such, the integrity of ascending reticular arousal networks
rather than that of sleep-generating systems may be relevant
for the recovery from paramedian thalamic stroke.
Sources of Funding
Supported by the University priority program “Integrative Human
Physiology” and the Swiss National Center of Competence “Neural
plasticity.”
Disclosures
None.
References
1. Catsman-Berrevoets CE, von Harskamp F. Compulsive pre-sleep
behavior and apathy due to bilateral thalamic stroke: response to bro-
mocriptine. Neurology. 1988;38:647– 649.
2. Guilleminault C, Quera-Salva MA, Goldberg MP. Pseudo-hypersomnia
and pre-sleep behaviour with bilateral paramedian thalamic lesions.
Brain. 1993;116:1549 –1563.
3. Bassetti C, Mathis J, Gugger M, Lovblad KO, Hess CW. Hypersomnia
following paramedian thalamic stroke: a report of 12 patients. Ann
Neurol. 1996;39:471– 480.
4. Santamaria J, Pujol M, Orteu N, Solanas A, Cardenal C, Santacruz P,
Chimeno E, Moon P. Unilateral thalamic stroke does not decrease ipsi-
lateral sleep spindles. Sleep. 2000;23:333–339.
5. Guberman A, Stuss D. The syndrome of bilateral paramedian thalamic
infarction. Neurology. 1983;33:540 –546.
6. Castaigne P, Lhermitte F, Buge A, Escourdolle R, Hauw JJ, Lyon-Caen
O. Paramedian thalamic and midbrain infarcts: clinical and neuropatho-
logical study. Ann Neurol. 1981;10:127–148.
7. Tatemichi TK, Steinke W, Duncan C, Bello JA, Odel JG, Behrens MM,
Hilal SK, Mohr JP. Paramedian thalamopeduncular infarction: clinical
syndromes and magnetic resonance imaging. Ann Neurol. 1992;32:
162–171.
8. Ackermann H, Ziegler W, Petersen D. Dysarthria in bilateral thalamic
infarction. A case study. J Neurol. 1993;240:357–362.
9. Chatterjee A, Yapundich R, Mennemeier M, Mountz JM, Inampudi C,
Pan JW, Mitchell GW. Thalamic thought disorder: on being ‘a bit
addled.’ Cortex. 1997;33:419 – 440.
10. Wiest G, Mallek R, Baumgartner C. Selective loss of vergence control
secondary to bilateral paramedian thalamic infarction. Neurology. 2000;
54:1997–1999.
11. Baumgartner RW, Regard M. Bilateral neuropsychological deficits in
unilateral paramedian thalamic infarction. Eur Neurol. 1993;33:195–198.
Downloaded from http://stroke.ahajournals.org/ at NY UNIV MED CTR LIBR on May 25, 2015
12. Stuss DT, Guberman A, Nelson R, Larochelle S. The neuropsychology of
paramedian thalamic infarction. Brain Cogn. 1988;8:348 –378.
13. Bogousslavsky J, Regli F, Delaloye B, Delaloye-Bischof A, Assal G,
Uske A. Loss of psychic self-activation with bithalamic infarction. Neu-
robehavioural, CT, MRI and SPECT correlates. Acta Neurol Scand.
1991;83:309 –316.
14. Raymer AM, Moberg P, Crosson B, Nadeau S, Rothi LJ. Lexical–
semantic deficits in two patients with dominant thalamic infarction.
Neuropsychologia. 1997;35:211–219.
15. Engelborghs S, Marien P, Pickut BA, Verstraeten S, De Deyn PP. Loss of
psychic self-activation after paramedian bithalamic infarction. Stroke.
2000;31:1762–1765.
16. Becker M, Sturm W, Willmes K, Zimmermann P. A normative study on
the attention test battery (TAP) by Zimmermann and Fimm. Zeitschr
Neuropsychol. 1996;7:3–15.
17. Brickenkamp R. Test d2. Aufmerksamkeits-Belastungstest. Göttingen:
Hogrefe; 1981.
18. Thurstone LL, Thurstone TG. Chicago Tests of Primary Mental Abilities.
Chicago: University of Chicago Press; 1941.
19. Regard M, Stuss E, Knapp P. Children’s production on verbal and
nonverbal fluency tasks. Percept Mot Skills. 1982;55:839 – 844.
20. Stroop JR. Studies of interference in serial verbal reactions. Exp Psychol.
1935;18:643– 662.
21. Rey A. L’examen clinique en psychologie. Paris: Presses Universitaires
de France; 1964.
22. Osterrieth PA. Le test d’une figure complexe. Arch Psychol. 1944;30:
206 –356.
23. Huber W, Poeck K, Weniger D, Willmes K. Der Aachener Aphasie-Test.
Göttingen: Hogrefe; 1983.
24. Rechtschaffen A, Kales A. A Manual of Standardized Terminology,
Techniques and Scoring System for Sleep Stages of Human Subjects. Los
Angeles: UCLA Brain Information Service, Brain Research Institute;
1968.
25. Gottselig JM, Bassetti CL, Achermann P. Power and coherence of sleep
spindle frequency activity following hemispheric stroke. Brain. 2002;
125:373–383.
26. Bassetti C, Gugger M, Bischof M, Mathis J, Sturzenegger C, Werth E,
Radanov B, Ripley B, Nishino S, Mignot E. The narcoleptic borderland:
a multimodal diagnostic approach including cerebrospinal fluid levels of
hypocretin-1 (orexin A). Sleep Med. 2003;4:7–12.
27. Szirmai I, Vastagh I, Szombathelyi E, Kamondi A. Strategic infarcts of
the thalamus in vascular dementia. J Neurol Sci. 2002;15:91–97.
28. Büttner T, Schilling G, Hornig CR, Dorndorf W. Thalamic infarcts—
clinical aspects, neuropsychological findings, prognosis. Fortschr Neurol
Psychiatr. 1991;59:479 – 487.
29. Reilly M, Connolly S, Stack J, Martin EA, Hutchinson M. Bilateral
paramedian thalamic infarction: a distinct but poorly recognized stroke
syndrome. Q J Med. 1992;82:63–70.
30. Fukutake T, Akada K, Ito S, Okuda T, Ueki Y. Severe personality
changes after unilateral paramedian thalamic infarct. Eur Neurol. 2002;
47:156 –160.
31. Edelstyn NM, Ellis SJ, Jenkinson P, Sawyer A. Contribution of the left
dorsomedial thalamus to recognition memory: a neuropsychological case
study. Neurocase. 2002;8:442– 452.
32. Segal JB, Williams R, Kraut MA, Hart J Jr. Semantic memory deficit with
a left thalamic infarct. Neurology. 2003;61:252–254.
33. Schnider A. Spontaneous confabulation, reality monitoring, and the
limbic system—a review. Brain Res Rev. 2001;36:150 –160.
34. Gaffan D, Gaffan EA. Amnesia in man following transsection of the
fornix. Brain. 1991;114:2611–2618.
35. Schnider A, Regard M, Landis T. Anterograde and retrograde amnesia
following bitemporal infarction. Behav Neurol. 1994;7:87–92.
36. Schnider A. Spontaneous confabulation and the adaptation of thought to
ongoing reality. Nat Neurosci Rev. 2003;4:662– 671.
 Evolution of Neurological, Neuropsychological and Sleep-Wake Disturbances After
Paramedian Thalamic Stroke
Dirk M. Hermann, Massimiliano Siccoli, Peter Brugger, Karen Wachter, Johannes Mathis, Peter
Achermann and Claudio L. Bassetti

Stroke. 2008;39:62-68; originally published online November 29, 2007;

doi: 10.1161/STROKEAHA.107.494955
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2007 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/39/1/62

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/ at NY UNIV MED CTR LIBR on May 25, 2015