Sleep Quality Perception in the Chronic Fatigue Syndrome: Correlations with Sleep Efficiency, Affective Symptoms and Intensity of Fatigue

Sleep Laboratory, Department of Psychiatry, University Hospital Brugmann, Brussels, Belgium.
Neuropsychobiology (Impact Factor: 2.26). 02/2007; 56(1):40-6. DOI: 10.1159/000110727
Source: PubMed
One of the core symptoms of the chronic fatigue syndrome (CFS) is unrefreshing sleep and a subjective sensation of poor sleep quality. Whether this perception can be expressed, in a standardized questionnaire as the Pittsburgh Sleep Quality Index (PSQI), has to our knowledge never been documented in CFS. Furthermore, correlations of subjective fatigue, PSQI, affective symptoms and objective parameters such as sleep efficiency are poorly described in the literature.
Using a cross-sectional paradigm, we studied subjective measures like PSQI, Fatigue Severity Scale scores and intensity of affective symptoms rated by the Hamilton Depression and Anxiety scales as well as objective sleep quality parameters measured by polysomnography of 28 'pure' (no primary sleep and no psychiatric disorders) CFS patients compared to age- and gender-matched healthy controls.
The PSQI showed significantly poorer subjective sleep quality in CFS patients than in healthy controls. In contrast, objective sleep quality parameters, like the Sleep Efficiency Index (SEI) or the amount of slow-wave sleep did not differ significantly. Subjective sleep quality showed a correlation trend with severity of fatigue and was not correlated with the intensity of affective symptoms in CFS.
Our findings indicate that a sleep quality misperception exists in CFS or that potential nocturnal neurophysiological disturbances involved in the nonrecovering sensation in CFS are not expressed by sleep variables such as the SEI or sleep stage distributions and proportions.


Available from: Olivier Le Bon
Fax +41 61 306 12 34
Original Paper
Neuropsychobiology 2007;56:40–46
DOI: 10.1159/000110727
Sleep Quality Perception in the Chronic Fatigue
Syndrome: Correlations with Sleep Efficiency,
Affective Symptoms and Intensity of Fatigue
Daniel Neu
a, b
Olivier Mairesse
Guy Hoffmann
Amirouche Dris
Luc J. Lambrecht
Paul Linkowski
Paul Verbanck
Olivier Le Bon
Sleep Laboratory, University Hospital Brugmann, Université Libre de Bruxelles,
Sleep Laboratory,
University Hospital Erasme, Université Libre de Bruxelles, Brussels , and
Department of Internal Medicine,
Biomed Flanders, Gent , Belgium;
Sleep Unit, Department of Oto-Rhino-Llaryngology, Kirchberg Hospital,
Luxembourg , Luxembourg
and was not correlated with the intensity of affective symp-
toms in CFS. Conclusion: Our findings indicate that a sleep
quality misperception exists in CFS or that potential noctur-
nal neurophysiological disturbances involved in the non-
recovering sensation in CFS are not expressed by sleep
variables such as the SEI or sleep stage distributions and pro-
Copyright © 2007 S. Karger AG, Basel
The chronic fatigue syndrome (CFS) is still a debated
clinical entity with an estimated prevalence rate ranging
from 0.2 to 0.7%
[1] . CFS is a nosologically defined syn-
[2] of unknown etiology and pathogenesis. Fur-
thermore, lifetime prevalence of a 6-month-long fatigue
episode has been estimated to be up to 9%
[3] . The Amer-
ican Sleep Disorders Association
[4] and the American
Psychiatric Association in its Diagnostic and Statistical
Manual of Mental Disorders (DSM)
[5] do not presently
recognize CFS as a distinct clinical entity. During the last
two decades, CFS has received several different confusing
labels; it has also been called neurasthenia, a label first
used in 1903 by Pierre Janet
[6] . The International Clas-
sification of Diseases
[7] defines neurasthenia as a neu-
rotic disorder with muscle weakness accompanied by
Key Words
Sleep quality Pittsburgh Sleep Quality index Chronic
fatigue syndrome Affective symptoms
Background/Aims: One of the core symptoms of the chron-
ic fatigue syndrome (CFS) is unrefreshing sleep and a subjec-
tive sensation of poor sleep quality. Whether this perception
can be expressed, in a standardized questionnaire as the
Pittsburgh Sleep Quality Index (PSQI), has to our knowledge
never been documented in CFS. Furthermore, correlations of
subjective fatigue, PSQI, affective symptoms and objective
parameters such as sleep efficiency are poorly described in
the literature. Methods: Using a cross-sectional paradigm,
we studied subjective measures like PSQI, Fatigue Severity
Scale scores and intensity of affective symptoms rated by
the Hamilton Depression and Anxiety scales as well as objec-
tive sleep quality parameters measured by polysomnogra-
phy of 28 ‘pure’ (no primary sleep and no psychiatric disor-
ders) CFS patients compared to age- and gender-matched
healthy controls. Results: The PSQI showed significantly
poorer subjective sleep quality in CFS patients than in
healthy controls. In contrast, objective sleep quality param-
eters, like the Sleep Efficiency Index (SEI) or the amount of
slow-wave sleep did not differ significantly. Subjective sleep
quality showed a correlation trend with severity of fatigue
Received: January 12, 2007
Accepted after revision: August 18, 2007
Published online: November 6, 2007
Dr. Daniel Neu, MD
Brugmann University Hospital U.L.B./V.U.B.
Sleep Laboratory U78 – Department of Psychiatry
4, Arthur Van Gehuchten Plaza, BE–1020 Brussels (Belgium)
Tel. +32 2 477 2554, Fax +32 2 477 2162, E-Mail
© 2007 S. Karger AG, Basel
Accessible online at:
Page 1
Sleep Quality Perception in CFS
Neuropsychobiology 2007;56:40–46
functional disorders (digestive or endocrinal), headaches
and anxiety. The American Centers for Disease Control
and Prevention criteria
[2] define CFS as a more than 6-
month-long fatigue, not being the result of ongoing exer-
tion nor being alleviated by rest with an induced disabil-
ity of 50% regarding usual daily activities.
Associated psychosomatic conditions such as fibro-
myalgia (FM) and irritable bowel syndrome (IBS) are
very frequent in CFS
[1] . There are similarities between
[8] and FM can be confused with CFS [1] .
Wessley et al.
[9] suggest that CFS, FM and IBS represent
different manifestations of the same somatic and psycho-
logical disturbances.
Treatment options are varied and range from hormon-
al treatments (dehydroepiandrosterone, corticosteroids),
immunotherapy and antidepressants to vitamins and
alimentary complements. Besides cognitive behavioral
therapy and graded aerobic exercise, no treatment option
has shown evident reproducible benefits.
The current mainstream view about chronic fatigue is
that the syndrome is either the result of an undetected
primary sleep disorder, or is the psychosomatic expres-
sion of an anxiety or depression disorder, perhaps linked
to a particular personality pattern. A more modern ap-
proach considers CFS as a multifactor clinical condition
in a biopsychosocial model
[1] .
Sleep is at the crossroads of the suspected syndrome,
as most patients complain of poor sleep quality and un-
refreshing nights as mentioned above. Reports of sleep
disturbances in CFS did not isolate a clear picture, large-
ly because of different selection criteria, type of control
group or characteristics of the recorded night (first night
only, home recordings), which makes comparisons and
reproducibility difficult. Several reports showed poor
sleep efficiency (Sleep Efficiency Index, SEI)
[10–14] .
Some mentioned increased sleep onset latency (SOL) [10,
, and increased number or duration [10, 12, 13] of in-
termittent awakenings. Rapid eye movements have been
shown to decrease
[10, 14] or increase [15] . A recent study
reported reduced slow-wave sleep (SWS) but increased
delta band power
[16] .
On the subjective side, the Pittsburgh Sleep Quality
Index (PSQI) is one of the most used and well-validated
questionnaires of subjective sleep quality
[17] . It mea-
sures the average sleep quality of the last month, with
higher scores indicating worse sleep. It has been applied
in sleep disorders, general clinical conditions and psy-
chosomatic medicine. Recently, a study using the PSQI
showed no correlation of subjective and objective sleep
quality in IBS
[18] . These results were consistent with pre-
viously published results about sleep quality in IBS
[19] .
Assessment of sleep quality perception with the PSQI has
to our knowledge never been published in CFS.
The objective of the present study was to test whether
PSQI scores significantly express poorer sleep quality
and whether this subjective sleep quality perception is
correlated with objective sleep efficiency in CFS. Patients
were compared to a group of healthy controls to confirm
differences between the conditions. Descriptively, rela-
tions between PSQI and a panel of other objective sleep
variables were tested. The relationships between PSQI se-
verity of fatigue and intensity of affective symptoms were
also measured.
To avoid potential biases in a sleep study, patients with
primary sleep disorders such as sleep apnea-hypopnea
syndrome (SAHS), periodic limb movement disorder
(PLMD) or narcolepsy were excluded. Also, as mood and
anxiety disorders are known to be associated with nu-
merous sleep abnormalities, patients with such comorbid
disorders were excluded. Patients presenting only with
CFS are labelled ‘pure’ CFS.
The null hypothesis was that there would be no cor-
relation relation between higher PSQI scores and de-
creased SEI. Considering the sample size, this study fo-
cuses on sleep efficiency measures, with and without
SOL. The other objective parameters are analyzed de-
M e t h o d s
Patients and Controls
During a 12-month period (from October 2005 to September
2006), patients with ‘pure’ CFS were selected prospectively among
the population admitted to our sleep laboratory. The recruited
CFS patients had fatigue and unrefreshing sleep as their main
complaint. They were referred to the sleep unit by the medical
department of another institution (tertiary care setting) after a
full medical checkup. Center for Disease Control criteria were
used for a first selection of CFS patients
[2] : (1) clinically evalu-
ated, unexplained or relapsing chronic fatigue – that is of new or
definite onset – is not the result of ongoing exertion nor is it al-
leviated by rest, and results in the substantial reduction of previ-
ous levels of occupational, educational, social or personal activi-
ties; (2) concurrent occurrence of 4 or more of the following
symptoms, all of which must have persisted or recurred during 6
or more consecutive months of illness and must not have predat-
ed the fatigue: self-reported impairment in short-term memory or
concentration; sore throat; tender cervical or axillary lymph
nodes; muscle pain; multijoint pain without joint swelling or red-
ness; headaches of a new type, pattern or severity; unrefreshing
sleep; postexertional malaise lasting more than 24 h; (3) exclusion
of any active medical condition that may explain the presence of
chronic fatigue; any previous condition which might explain fa-
Page 2
Neu /Mairesse /Hoffmann /Dris /
/Linkowski /Verbanck /Le Bon
Neuropsychobiology 2007;56:40–46
tigue and which has not come to an end; substance abuse within
2 years prior to onset; severe obesity. In order to avoid potential
overlaps with primary sleep or psychiatric disorders, further ex-
clusion criteria were: (1) Apnea-Hypopnea Index or Periodic Limb
Movement Index 6 5/h, clinical criteria for narcolepsy or idio-
pathic hypersomnia; (2) all DSM-IV axis I diagnoses. In unclear
cases, rejection prevailed.
All patients had been free of neuropsychopharmacological
treatment for at least 2 weeks prior to recording, and, in most
cases, this period exceeded 3 months. Daytime napping was not
allowed in the sleep unit. All patients in our lab completed ques-
tionnaires about lifestyle and drinking habits. Patients with a
consumption of more than two units of alcohol per day were ex-
Controls were locally recruited healthy volunteers. They were
paid EUR 100 by private funding for their participation. Regular
sleep-wake schedules were required and no working shifts were
allowed. No significant somatic condition and no current or past
mental disorder were allowed in the control group. Further exclu-
sion criteria were identical to those of the patient group.
All patients and controls filled out a sleep diary 2 weeks prior
to sleep recording to assess the regular sleep-wake schedule; they
received a standard physical examination (D.N.), the structured
clinical interview for DSM-IV
[5] (D.N., G.H., O.L.) and psycho-
metric assessment (D.N.).
All patients and controls gave written informed consent; the
study was approved by the local ethics committee and conducted
in accordance with the rules and regulations for the conduct of
clinical trials stated by the World Medical Assembly in Helsinki.
M a t e r i a l
All participants admitted to the sleep unit were recorded for 2
consecutive nights without habituation. Recordings were per-
formed between Mondays and Fridays. Participants were pre-
pared for the recordings between 10 p.m. and 11 p.m. and allowed
to retire when they wished. Morning arousal was spontaneous.
Polysomnography included 3 electroencephalograms recorded
from Fp2-A1, C4-A1, and O2-A1 sites, 2 electrooculograms, sub-
mental and bilateral anterior tibial electromyograms. Oral and
nasal airflow were recorded by thermoresistors (Healthdyne
Technologies USA), and respiratory effort was measured by tho-
racic and abdominal belts (Pro-Tech CT2
, Mukilteo, Wash.,
USA). Capillary oxygen saturation was monitored by photosensi-
tive finger oximetry (Nonin Flexi-Form II 7000A, Nonin Med-
ical Inc., Minneapolis, Minn., USA, and LINOP Adt, Masimo
Corp., Irvine, Calif., USA).
P o l y s o m n o g r a p h y
All recordings were randomly analyzed by one of two well-
trained technicians, amidst other clinical work and without
knowledge of the study goals, on 21-inch screens displaying 30-
second polysomnogram epochs (Respironics Inc.
Alice 4 and
5, Murrysville, Pa., USA). Respiratory events were reviewed visu-
ally by the technicians. Classical criteria were used for sleep stage
[20] . Interrater reliability (kappa) exceeded 0.89 for all
SOL was defined as the time between lights out and the first
period of stage 2. Wake time did not include sleep latency. SEI 1
was defined by the total sleep time/time in bed ratio. SEI 2 was
defined by the total sleep time/period of sleep time ratio. Non-
rapid eye movement sleep included sleep stages 1–4. Rapid eye
movement sleep latency was defined as the time between the first
epoch of stage 2 and the first of rapid eye movement sleep. Light
sleep was the sum of stages 1 and 2. SWS was the sum of stages 3
and 4. The ratios are expressed in percentages. An episode of ap-
nea was defined as more than an 80% reduction in airflow for at
least 10 s during sleep. Hypopnea was defined as a 50–80% reduc-
tion of airflow amplitude accompanied by either a 3% or greater
reduction in oxygen saturation or an arousal. Microarousals were
defined according to the American Sleep Disorders Association
criteria [21] . The Microarousal Index (MAI) represented the
number of microarousals
per hour of sleep.
PSQI and Fatigue Scale
All participants answered the self-reporting questionnaires
on the first day of their stay in our unit at the same daytime (be-
tween 5 p.m. and 7 p.m.) before their first night of polysomno-
graphic recording.
Scoring of the PSQI was performed according to the guide-
lines provided by the authors
[17] . The 19 items are grouped into
7 component scores, each weighted equally on a scale from 0 to 3.
These components are subjective sleep quality, sleep latency, sleep
duration, habitual sleep efficiency, sleep disturbance, use of sleep
medication and daytime dysfunction. The component scores
were then summed to give the global PSQI score. In validation
studies, a global PSQI score 1 5 indicates that a subject is having
severe difficulties in at least 2 areas, or moderate difficulty in
more than 3 areas
[17] .
The Fatigue Severity Scale (FSS) is a self-report instrument
used to assess levels of fatigue and its effect on daily functioning.
The FSS was first used on individuals with multiple sclerosis and
systemic lupus erythematosus. It has later been used in studies on
obstructive sleep apnea, aerobic exercise
[22] , fatigue, obesity,
Parkinson disease, hepatitis C infection or CFS
[23] .
Affective Symptoms Scales
All patients completed the 13-item short form of the Beck De-
pression Inventory (BDI) on the first day of their stay in our labo-
ratory before the first night of polysomnography between 5 p.m.
and 7 p.m. All subjects were rated with the 21-item Hamilton De-
pression Rating Scale (HAMD-21) and the Hamilton Anxiety
Scale (HAMA) between 11 a.m. and 13 p.m. by the same inter-
viewer (D.N.) on the second day of their stay in our laboratory
after their first night of polysomnographic recording.
All variables in each of the groups were compatible with the
use of parametric tests, except for the PSQI subscales. Between-
group comparisons involving continuous data were computed us-
ing MANOVAs with Bonferroni correction: one MANOVA was
performed for the sleep parameters and another one for the psy-
chological tests. When equality of variances was not met, Mann-
Whitney nonparametric tests were used. Association between
continuous variables within the CFS group was tested using the
Pearson product-moment r. Hypothesis tests were two-sided and
carried out at the 5% significance level. Trends are noted at a 5
10% level. All results are expressed as mean 8 standard error of
the mean. The statistical analyses were computed using SPSS 10
(SPSS Inc., Chicago, Ill., USA).
Page 3
Sleep Quality Perception in CFS
Neuropsychobiology 2007;56:40–46
R e s u l t s
Our sample included 28 ‘pure’ CFS female patients
and 12 female controls. Table 1 shows descriptive and
polysomnographic variables. Sleep data examined in sta-
tistical analyses were from the second recorded night
only, in order to avoid potential first-night effects
[24] , in
particular in the CFS
[25] . Age was not significantly dif-
ferent between CFS patients and healthy controls (34.2 8
9.2 years vs. 32.3 8 9.6 years).
As equality of variances was not met for the MAI, the
variable was removed from the MANOVA on objective
sleep variables. No significant difference was evidenced
between groups. The MAI was shown to be higher in the
CFS than in the control group using the Mann-Whitney
U test (p = 0.001).
Mann-Whitney tests showed significant differences
between patients and controls for all global scores in sub-
jective measurements (p = 0.001 in each case). PSQI sub-
scales showed significantly higher scores (worse sleep
quality) in the patient group for sleep quality, sleep la-
tency, sleep efficiency, sleep disturbance and daytime
dysfunction. Sleep duration and use of medication were
not statistically distinct ( table 2 ).
Correlations between the PSQI and its subscales, and
objective sleep quality parameters, including MAI, were
not significant in the patient group (r = –0.195; r = –0.161).
Correlations between the FSS and objective sleep mea-
sures were not significant either. Correlations between
the HAMA and sleep measures showed a trend for SWS
(r = –0.364, p = 0.057). The BDI was correlated with sleep
efficiency 1 (r = –0.391; p = 0.040), SOL (r = 0.478; p =
0.010) and a trend for SWS (r = –0.359; p = 0.061).
Not surprisingly, average correlations appeared be-
tween the scales measuring affective states. The same
comparisons in the healthy control group showed quite
weak associations, except for links between both Hamil-
ton scales (data not shown).
This study compared objective and subjective param-
eters of sleep as well as the affective state of 28 female
‘pure’ CFS patients and 12 female healthy controls. The
main objective was the assessment of relationships be-
tween sleep efficiency and subjective evaluation of sleep
quality. Several other analyses were performed descrip-
tively using other objective sleep parameters and subjec-
tive scales other than the PSQI.
Table 1. Descriptive variables and objective sleep parameters
Variable CFS
(n = 28)
(n = 12)
Age, years
34.289.2 32.389.6
23.184.1 21.382.0
TIB, min
508.6858.0 485.2850.5
SPT, min
472.1859.4 457.9850.7
TST, min
433.1871.8 406.1832.5
SOL, min
26.5815.6 21.2811.3
SEI 1, %
84.688.2 84.185.9
SEI 2, %
91.487.2 89.186.2
8.687.2 10.986.1
LS, %
53.9811.5 54.288.6
SWS, %
24.0811.6 22.0811.8
77.987.5 76.288.3
13.686.1 13.385.9
RL, min
83.6851.9 80.4864.6
AHI (per h)
2.382.4 1.181.3
30.8818.5 11.986.1
Values are presented as means 8 standard deviation. Ratios
are presented as means 8 standard deviation and expressed as
TIB = Time in bed; SPT = sleep period time; TST = total sleep
time = SPT – WASO; SEI 1 = TST/TIB; SEI 2 = TST/SPT;
WASO = wake after sleep onset; LS = light sleep = sleep stages 1
and 2; REMS = rapid eye movement sleep in percent of SPT; RL =
REM sleep latency; AHI = Apnea-Hypopnea Index.
Tabl e 2. Comparison of subjective parameters: PSQI, FSS, HAMA,
Variable CFS
(n = 28)
(n = 12)
p (Mann-
PSQI global score
9.5484.0 2.4281.2
Sleep quality
1.780.7 0.780.5
Sleep latency
1.881.0 0.580.6
Sleep duration
0.781.0 0.280.4
Sleep efficiency
1.081.1 0.180.3
Sleep disturbance
1.580.7 0.780.4
Use of medication
0.881.3 0.080.0
Daytime dysfunction
2.080.7 0.280.4
FSS mean score
6.280.8 1.781.1
17.284.6 3.181.8
10.283.9 1.781.2
9.185.5 0.380.6
Scores are presented as means 8
standard deviation.
Page 4
Neu /Mairesse /Hoffmann /Dris /
/Linkowski /Verbanck /Le Bon
Neuropsychobiology 2007;56:40–46
Sleep efficiency, whether including the SOL or not,
was not shown to differ significantly between the CFS
and the control group. Comparisons on sleep quality per-
ception (PSQI) unambiguously showed poorer ratings in
CFS patients than in controls. A mismatch seems thus to
be present between the subjective perception and the ob-
jective analysis of CFS sleep. The possibility of a misper-
ception had also previously been mentioned in an inter-
esting study about twins who were discordant for CFS
[26] .
As Elsenbruch et al.
[19] proposed in a similar paper
about IBS, this mismatch may mean either that (1) the
perception of sleep quality is affected by the same mech-
anisms that trigger CFS: such a mechanism may lower
patients’ perceptual threshold for both CFS and sleep
evaluation; (2) the attention paid to internal stimuli in
general may be affected, so that slight deviations from
normalcy, such as a brief awakening or perception of fa-
tigue, are more likely to be remembered and interpreted
as an abnormality; (3) anxiety creates or enhances a re-
sponse bias towards reporting both CFS and sleep prob-
lems, as the influence of psychological states on percep-
tion have been emphasized. Insomniacs are known to un-
derestimate their sleep, whereas healthy subjects tend to
overestimate it
[27, 28] . In other words, these biases can
either be of psychological or biological origin, if we accept
a Cartesian division of mind and body.
Because of sample size issues, all analyses except for
the correlations between the PSQI and objective sleep ef-
ficiency measures are to be considered descriptive, as has
been said above. The substantial quantitative differences
in microarousals observed here between patients and
controls were also described in another study by our
group in a distinct group of CFS patients
[29] . What is
particularly striking is that there is no identifiable cause
for triggering these microarousals in CFS, in contrast to
microarousals in SAHS, PLMD, or experimental noise
exposure disturbances, for example. In general, micro-
arousals can be related to wake-triggering phenomena
such as changes in airflow and blood gasometry in SAHS
or myoclonias in PLMD. Furthermore, an increased
number of microarousals in primary sleep disorders is
generally related to a relative and significant reduction of
SWS. No such reduction of SWS was apparent here or in
our retrospective study
[29] . Thus, respiratory or subcor-
tical events may explain these microarousals in other
conditions, but there is no identifiable cause for them in
CFS. Microarousals can thus partially account for the
nonrecovering and unrefreshing sleep sensations in
Miscellaneous subjective measures, on the other hand,
showed more daytime fatigue (FSS), more depression
(HAMD and BDI) and more anxiety (HAMA) in the CFS
group. Most of the recent research studies on CFS tend to
consider CFS as distinct from major depressive disorder
when considering clinical diagnostic criteria
[30, 31] , but
also when considering opposite hypothalamic-pituitary-
adrenal axis regulation profiles
[32] and diminished SWS
in major depressive disorder
[3335] . The findings in our
sample, where duration and proportion of SWS showed
no significant difference compared to controls, confirm
these differences. Although no patient in the sample
could be diagnosed with DSM-IV major depression, de-
pression ratings were not negligible. Statistical trends
were even found for a negative association between SWS
and intensity of depression (HAMD and BDI). However,
despite this relatively high intensity of depression, SWS
was not reduced as a whole. Also, intensity of depression
was not associated with intensity of fatigue.
Although both the MAI and the PSQI differed be-
tween the patient and the control groups, they were not
correlated with each other. A high level of microarousals
is therefore not translated into poorer sleep quality per-
The substantially more elevated scores on the PSQI
found in CFS patients in comparison to controls indicate
that the PSQI can also be used in CFS as a clinical routine
measure for the severity of patients’ complaints about
sleep quality.
There are some limitations that have to be considered.
The absence of significant differences in the remaining
objective sleep parameters could be due to the way these
variables are assessed. Sleep stages, proportions and sleep
architecture (classical scoring) are global concepts with
limited meanings. The main difficulty arises from the
usual 30-second epoch staging, which may hide short-
duration events. Different underlying physiological or
pathological brain processes could express themselves,
according to the used sleep staging criteria, with the same
sleep stage proportions
[36] . Quantitative approaches like
sleep EEG power spectral analysis and analysis of sleep
microstructure, notably the cycling alternating pattern,
may, in the future, prove more promising when seeking
for sleep-related abnormalities that could account for the
daytime symptoms in CFS
[16] . However, increases in
specific power bands (ultraslow, or delta) cannot pres-
ently be related to sleep efficiency or refreshing sleep.
This being a sleep study, we thought it was crucial to
isolate patients not suspected of suffering from comorbid
disorders which may interfere with sleep, such as SAHS,
Page 5
Sleep Quality Perception in CFS
Neuropsychobiology 2007;56:40–46
PLMD or major depression. Our sample is thus superse-
lected. This means that the present patient group fulfills
all the Center for Disease Control criteria but that all pa-
tients with Center for Disease Control criteria may not
display the present findings. The conclusions as such
cannot therefore be generalized to them. It is, however,
our conviction that including patients with identifiable
sleep disorders or disorders well known to be associated
with sleep anomalies only confuses the matter and we
advocate the analysis of ‘pure’ CFS in sleep studies at
In conclusion, our findings indicate either that a sleep
quality misperception may be present in CFS or that the
neurophysiological disturbances involved in the nonre-
covering sensation in CFS are not expressed by sleep vari-
ables such as SEI or sleep stage distributions and propor-
tions. Future studies investigating sleep in CFS should
perhaps focus upon sleep microstructure and quantita-
tive sleep EEG measurements. The present findings are
reinforced by the strict selection criteria used here for
CFS patients, where confounding sleep disorders such as
PLMD or SAHS as well as comorbid axis I diagnoses were
excluded, while adaptation issues to the lab were reduced
by using only results from the second recorded night.
A c k n o w l e d g m e n t s
We would like to express our gratitude to Julie Martina for her
devoted work during the data acquisition and recording. We
would also like to thank Robert Moutrier, MD, for reviewing the
manuscript. D.N. was supported by a research grant from the Na-
tional Funding for Scientific Research from the Ministry of Re-
search, Culture and Superior Education of the Grand-Duchy of
Luxembourg. P.L. was supported by the National Funding for
Scientific Research (FNRS), Belgium. O.L. was supported by
SOMALCPE, a private fund dedicated to research in psychiatric
neurosciences and sleep medicine.
1 Maquet D, Demoulin C, Crielaard JM:
Chronic fatigue syndrome: a systematic re-
view. Ann Readapt Med Phys 2006;
49: 337–
347, 418–427.
2 Fukuda K, Straus SE, Hickie I, Sharpe MC,
Dobbings JG, Komaroff A, the International
Chronic Fatigue Syndrome Study Group:
The chronic fatigue syndrome: a compre-
hensive approach to its definition and study.
Ann Intern Med 1994;
121: 953–959.
3 Scott LV, Medbak S, Dinan TG: Blunted ad-
renocorticotropin and cortisol responses to
corticotropin-releasing hormone stimula-
tion in chronic fatigue syndrome. Acta Psy-
chiatr Scand 1998;
97: 450–457.
4 American Sleep Disorders Association: In-
ternational Classification of Sleep Disorders,
Revised. Rochester, American Sleep Disor-
ders Association, 1997.
5 American Psychiatric Association: Diagnos-
tic and Statistical Manual of Mental Disor-
ders, ed 4. Washington, American Psychiat-
ric Association Press, 1994.
6 Janet P: Obsessions and Psychasthenia. Par-
is, Alcan, 1903.
7 World Health Organization: International
Classification of Diseases, ed 10. Geneva,
World Health Organization, 1990.
8 Gomborone JE, Gorard DA, Dewsnap PA,
Libby GW, Farthing MJ: Prevalence of irri-
table bowel syndrome in chronic fatigue. J R
Coll Physicians Lond 1996;
30: 512–513.
9 Wessely S, Nimnuan C, Sharpe M: Function-
al somatic syndromes: one or many? Lancet
354: 936–939.
10 Whelton CH, Salit I, Moldofsky H: Sleep, Ep-
stein-Barr, virus infection, musculoskeletal
pain and depressive symptoms in chronic fa-
tigue syndrome. J Rheumatol 1992;
19: 6:939–
11 Morris R, Sharpe M, Sharpley AL, Cowen PJ,
Hawton K, Morris J: Abnormalities of sleep
in patients with the chronic fatigue syn-
drome. BMJ 1993;
306: 1161–1164.
12 Sharpley A, Clements A, Hawton K, Sharpe
M: Do patients withpure chronic fatigue
syndrome (neurasthenia) have abnormal
sleep? Psychosom Med 1997;
59: 592–596.
13 Fischler B, Le Bon O, Hoffmann G, Cluydts
R, Kaufman L, De Meirleir K: Sleep anoma-
lies in the chronic fatigue syndrome. Neuro-
psychobiology 1997;
35: 115–122.
14 Stores G, Fry A, Crawford C: Sleep abnor-
malities demonstrated by home polysom-
nography in teenagers with chronic fatigue
syndrome. J Psychosom Res 1998;
45; 1: 85
15 Ball N, Buchwald D, Schmidt D, Goldberg G,
Ashton S, Armitage R: Monozygotic twins
discordant for chronic fatigue syndrome.
Objective measures of sleep. J Psychosom
Res 2004;
56: 207212.
16 Guilleminault C, Poyares D, Rosa A, Kiriso-
glu C, Almeida T, Lopes MC: Chronic fa-
tigue, unrefreshing sleep and nocturnal poly-
somnography. Sleep Med 2006;
7: 513–520.
17 Buysse DJ, Reynolds CF 3rd, Monk TH, Ber-
man SR, Kupfer DJ: The Pittsburgh Sleep
Quality Index: a new instrument for psychi-
atric practice and research. Psychiatry Res
28: 193–213.
18 Heitkemper M, Jarrett M, Burr R, Cain KC,
Landis C, Lentz M, Poppe A: Subjective and
objective sleep indices in women with irri-
table bowel syndrome. Neurogastroenterol
Motil 2005;
17: 523–530.
19 Elsenbruch S, Harnish MJ, Orr WC: Subjec-
tive and objective sleep quality in irritable
bowel syndrome. Am J Gastroenterol 1999;
94: 2447–2452.
20 Rechtschaffen A, Kales A (eds): A Manual of
Sta nda rdiz ed Terminolog y Tech niques and
Scoring System for Sleep Stages of Human
Subjects. Washington, US Government
Printing Office, 1968.
21 American Sleep Disorders Association: EEG
arousals: scoring rules and examples. A pre-
liminary report from the Sleep Disorders At-
las Task Force of the American Sleep Disor-
ders Association. Sleep 1992;
15: 174–184.
22 Krupp LB, LaRocca NG, Muir-Nash J, Stein-
berg AD: The fatigue severity scale. Applica-
tion to patients with multiple sclerosis and
systemic lupus erythematosus. Arch Neurol
46: 1121–1123.
23 Olson LG, Ambrogetti A, Sutherland DC: A
pilot randomized controlled trial of dexam-
phetamine in patients with chronic fatigue
syndrome. Psychosomatics 2003;
44: 38–43.
24 Agnew HW, Webb WB, Williams RL: The
first-night effect: an EEG study of sleep. Psy-
chophysiology 1966;
2: 263–266.
25 Le Bon O, Minner P, Van Moorsel C, Hoff-
mann G, Gallego S, Lambrecht L, Pelc I,
Linkowski P: First-night effect in the chron-
ic fatigue syndrome. Psychiatry Res 2003;
120: 191–199.
Page 6
Neu /Mairesse /Hoffmann /Dris /
/Linkowski /Verbanck /Le Bon
Neuropsychobiology 2007;56:40–46
26 Watson NF, Kapur V, Arguelles LM, Gold-
berg J, Schmidt DF, Armitage R, Buchwald
D: Comparison of subjective and objective
measures of insomnia in monozygotic twins
discordant for chronic fatigue syndrome.
Sleep 2003;
26: 324–328.
27 Coates TJ, Killen JD, Silverman S, George J,
Marchini E, Hamilton S, Thoresen CE: Cog-
nitive activity, sleep disturbance, and stage
specific differences between recorded and
reported sleep. Psychophysiology 1983;
28 Carskadon MA, Dement WC, Mitler MM,
Guilleminault C, Zarcone VP, Spiegel R:
Self-reports versus sleep laboratory findings
in 122 drug-free subjects with complaints of
chronic insomnia. Am J Psychiatry 1976;
29 Le Bon O, Neu D, Valente F, Linkowski P:
Paradoxical NREMS distribution in ‘pure’
chronic fatigue patients: a comparison with
sleep apnea-hypopnea patients and healthy
control subjects. J Chronic Fatigue Syndr, in
30 Caplan C: Chronic fatigue syndrome or just
plain tired? CMAJ 1998;
159: 519520.
31 Powell R, Dolan R, Wessely S: Attributions
and self-esteem in depression and chronic
fatigue syndromes. J Psychosom Res 1990;
34: 665–673.
32 Demitrack MA, Crofford LJ: Evidence for
and pathophysiologic implications of hypo-
thalamic-pituitary-adrenal axis dysregula-
tion in fibromyalgia and chronic fatigue
syndrome. Ann NY Acad Sci 1998;
840: 684–
33 Hubain P, Le Bon O, Vandenhende F, Van
Wijnendaele R, Linkowski P: Major depres-
sion in males: effects of age, severity and ad-
aptation on sleep variables. Psychiatry Res
145: 169–177.
34 Adrien J: Neurobiological bases for the rela-
tion between sleep and depression. Sleep
Med Rev 2002;
6: 341–351.
35 Jones D, Gershon S, Sitaram N, Keshavan M:
Sleep and depression. Psychopathology 1987;
20(suppl 1):20–31.
36 Rechtschaffen A: Current perspectives on
the function of sleep. Perspect Biol Med
41: 359–390.
Page 7
    • "Hence, self-reported good sleepers, may present with objective sleep variables (sleep fragmentation, respiratory disturbance e.g.) similar to poor sleepers. It has previously been mentioned that classical PSG variables may not be able to express potential underlying disturbances involved in the nonrecovering sensations of reported poor sleep quality (Neu et al., 2007). Consequently, the present findings raise questions about what influences complex perceptions like sleep quality and which dimensional parts of it are within the effective reach of standard polysomnography (Buysse et al., 2008). "
    [Show abstract] [Hide abstract] ABSTRACT: Fatigue and sleepiness are ubiquitous symptoms in various conditions and are frequently associated to impaired sleep quality. While separate fatigue and sleepiness scales exist, both constructs are often confused. Unraveling this issue requires estimating the instruments’ measurement properties, potential scale recalibration and re-evaluation of symptom intensities on a comparable basis. This study aims at improving the assessment of these symptoms and quantifying their degree of overlap using common-person-equating (CPE). One hundred fifty-nine patients, either with complaints of fatigue, sleepiness and/or non-restorative sleep, addressed to an academic sleep unit for a full-night polysomnography (PSG), enrolled in the study. Symptom levels were measured with the Fatigue Severity (FSS) and Epworth Sleepiness (ESS) scales. Sleep quality was assessed by the Pittsburgh Sleep Quality Index, defining’‘good’ and’‘poor’ sleeper groups. Good and poor sleepers did not differ statistically regarding demographics and PSG parameters. Rasch analysis revealed that, considering proper calibration, the ESS and FSS generate reliable and valid, unidimensional linear measures and to be invariant to perceived sleep quality. CPE showed predominantly fatigued, rather than sleepy patients, being more likely to present as poor sleepers. A concordance diagram based on scale scores is provided, in order to improve the differentiation of both symptoms.
    No preview · Article · Feb 2016
    • "The meta-analytic component of this review was performed in accordance with the Cochrane Handbook for Systematic Reviews[32]with the purpose of identifying possible differences in PSQI global and subscale scores between clinical and non-clinical samples (Tables S2a and S2b for raw data). Seven studies[22,46,49,51,56,68,79], six of fair quality[22,46,49,51,56,79]and one of good quality[68], provided global scores for both groups. The clinical samples combined equated to 801 individuals, and the non-clinical samples comprised 3433 persons. "
    No preview · Article · Dec 2015
  • Source
    • "The PSQI has been used in several psychometric studies, including in patients with bone marrow transplant, renal transplant, breast cancer, benign breast problems, and primary insomnia [15, 16]. Moreover, the PSQI is also used as a surrogate of perception of sleep quality in patients of chronic fatigue syndrome, myasthenia gravis, dementia , healthy women, and even in patients with OSA1718192021. Therefore, this retrospective study was conducted to investigate factors that are associated with the perception of sleep quality by analyzing the PSQI, focusing on the role of hypoxemia, in OSA patients. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Perceived sleep quality may play an important role in diagnosis and therapy for obstructive sleep apnea (OSA). However, few studies have assessed factors that are associated with perceived sleep quality in OSA patients. Hypoxemia depresses the central nervous system and attenuates the perceived respiratory load in asthmatic patients. This study aimed to investigate the factors related to perceived sleep quality, focusing on the role of hypoxemia. Methods Polysomnography studies of 156 OSA patients were reviewed. Traditional polysomnographic parameters, including parameters of oxy-hemoglobin saturation (SpO2), were calculated, and the sleep questionnaire and scales were used. Considering the possible pitfalls of absolute values of SpO2 and individualized responses to hypoxemia, the amplitude of desaturation was further computed as “median SpO2 minus lowest 5 % SpO2 “and “highest 5 % SpO2 minus median 5 % SpO2”. Correlations between these parameters and perceived sleep quality, represented as the Pittsburgh sleep quality index (PSQI), were performed. Multiple linear regression analysis was also conducted to investigate the factors associated with the PSQI. Results Although the PSQI was not correlated with the apnea-hypopnea index (r = −0.113, p = 0.162) and oxygen desaturation index (r = −0.085, p = 0.291), the PSQI was negatively correlated with “median SpO2 minus lowest 5 % SpO2” (r = −0.161, p = 0.045). After adjusting for age, total sleep time, the periodic limb movements index, tendency of depression, and the lowest 5 % SpO2, the “median SpO2 minus lowest SpO2” was still a significant predictor for a lower PSQI (β = −0.357, p = 0.015). Conclusions More severe hypoxemia is associated with better perceived sleep quality among OSA patients. This paradox may be associated with hypoxemia-related impairment of perception. The effect of hypoxemia did not appear to be significant in relatively mild hypoxemia but become significant in severe hypoxemia.” Median SpO2 minus lowest 5 % SpO2” may also be a better predictor of perceived sleep quality than the apnea-hypopnea index because of the disproportionate effects of hypoxemia. Additionally, further studies are necessary to confirm the role of hypoxemia on perceived sleep quality and identify the possible threshold of hypoxemia in OSA patients.
    Full-text · Article · Dec 2015 · BMC Pulmonary Medicine
Show more