Available via license: CC BY 4.0
Content may be subject to copyright.
R E S E A R C H Open Access
Self-reported fatigue following intensive
care of chronically critically ill patients:
a prospective cohort study
Gloria-Beatrice Wintermann
1*
, Jenny Rosendahl
2,3
, Kerstin Weidner
1
, Bernhard Strauß
3
, Andreas Hinz
4
and Katja Petrowski
1
Abstract
Background: Protracted treatment on intensive care unit (ICU) sets the patients at increased risk for the development
of chronic critical illness (CCI). Muscular and cardio-respiratory deconditioning are common long-term sequelae, going
along with a state of chronic fatigue. At present, findings regarding the frequency, long-term course, and associated
factors of self-reported fatigue following ICU treatment of CCI patients are lacking.
Methods: CCI patients with the diagnosis of critical illness polyneuropathy/myopathy (CIP/CIM) were assessed at three
time points. Four weeks following the discharge from ICU at acute care hospital (t1), eligibility for study participation
was asserted. Self-reported fatigue was measured using the Multidimensional Fatigue Inventory (MFI-20) via telephone
contact at 3 (t2, n= 113) and 6 months (t3, n= 91) following discharge from ICU at acute care hospital.
Results: At both 3 and 6 months, nearly every second CCI patient showed clinically relevant fatigue symptoms (t2/t3:
n=53/n= 51, point prevalence rates: 46.9%/45.1%). While total fatigue scores remained stable in the whole sample,
female patients showed a decrease from 3 to 6 months. The presence of a coronary heart disease, the perceived fear
of dying at acute care ICU, a diagnosis of major depression, and the perceived social support were confirmed as
significant correlates of fatigue at 3 months. At 6 months, male gender, the number of medical comorbidities, a
diagnosis of major depression, and a prior history of anxiety disorder could be identified. A negative impact of
fatigueontheperceivedhealth-relatedqualityoflifecouldbeascertained.
Conclusions: Nearly every second CCI patient showed fatigue symptoms up to 6 months post-ICU. Patients at
risk should be informed about fatigue, and appropriate treatment options should be offered to them.
Trial registration: The present study was registered retrospectively at the German Clinical Trials Register (date of
registration: 13th of December 2011; registration number: DRKS00003386). Date of enrolment of the first participant to
the present trial: 09th of November 2011.
Keywords: Fatigue, Multidimensional Fatigue Inventory (MFI-20), Intensive care unit (ICU), Chronic critical illness (CCI),
Sepsis, Health-related quality of life, Posttraumatic stress
* Correspondence: gloria.wintermann@uniklinikum-dresden.de
1
Department of Psychotherapy and Psychosomatic Medicine, Medizinische
Fakultät Carl Gustav Carus, Technische Universität Dresden, Dresden
Fetscherstraße 74, 01307 Dresden, Germany
Full list of author information is available at the end of the article
© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Wintermann et al. Journal of Intensive Care (2018) 6:27
https://doi.org/10.1186/s40560-018-0295-7
Background
Subjective fatigue refers to an overwhelming, sustained
sense of physical, emotional, and/or cognitive exhaustion
that is not proportional to recent activity [1–3]. Fatigue
has been most frequently related to cancer, its treat-
ment, and chronic illnesses [4–7]. Beyond that, current
research revealed that fatigue is one of the most preva-
lentanddebilitatingproblem,emergingduringcritical
illness and following the treatment on intensive care
unit (ICU) [8–10].
Particularly, the long-term stay on ICU (> 72 h) along
with prolonged mechanical ventilation may lead to a
state of muscular/cardio-respiratory deconditioning, in-
creasing the risk for chronic fatigue [11,12]. Between 5
and 10% of acutely ill patients who require ongoing mech-
anical ventilation develop a syndrome named chronic crit-
ical illness (CCI) [13–15]. This comprises distinct clinical
features (e.g., myopathy, neuropathy, loss of lean body
mass, anasarca, vulnerability to infection and sepsis, delir-
ium, coma). Characteristical is that CCI patients are not
expected to be weaned from the ventilator in the immedi-
ate future, necessitating ongoing, tight cardio-pulmonary
monitoring and long-term ventilator dependency [14]. As
long-term sequelae, CCI patients might suffer from a
chronic state of exhaustion which may interfere with the
patient’s usual functional capacity and ability to participate
in one’s own rehabilitation [16,17].
At present, there is only preliminary evidence for the
occurrence of fatigue in CCI patients [18–20]. In long-
term treated ICU patients, chronic fatigue affected be-
tween more than one third and three quarter within
1 year after ICU discharge [18,21,22]. In line, Puntillo
et al. showed that 75% of patients at high risk of dying
reported “being tired”after prolonged ICU treatment
[19]. However, the studies mentioned above did not
apply either multidimensional and valid measures of fa-
tigue, used short time frames, small sample sizes, or
quite heterogeneous ICU samples.
Considering the etiopathogenetic mechanisms of fatigue,
a multifactorial approach has been proclaimed, assuming
the influence of sociodemographic, pathophysiological,
and environmental factors, the impact of therapeutic in-
terventions, and medication [16]. Above, fatigue goes
along with emotional stress, a decreased health-related
quality of life [23], and greatly overlaps with symptoms of
depression, anxiety, and disordered sleep [16,24–26].
However, there is still a need for studies unraveling sub-
jective fatigue in CCI patients, especially its related factors
[16]. This is of clinical relevance since an early identifica-
tion would allow the referral of affected patients to ap-
propriate symptom management programs. Therefore,
the main aims of the present study were the following:
first, the assessment of the rate and course of self-
reported fatigue using the Multidimensional Fatigue
Inventory (MFI-20) [27] at 3 and 6 months following
ICU discharge; second, the identification of associated
variables; and third, the impact of fatigue on psycho-
logical outcomes (e.g. posttraumatic stress, health-
related quality of life) in CCI patients.
Methods
Setting and procedure
A homogeneous sample of patients with a principal
diagnosis of critical illness polyneuropathy (CIP), critical
illness myopathy (CIM), or combined CIP/CIM was con-
secutively enrolled during its treatment at a large re-
habilitation hospital between November 2011 and May
2013. CIP and CIM have been shown to be important
causes of the ICU-acquired weakness and failed weaning
from the ventilator (e.g., [28]). The patients were
assessed at three time points within an observational,
prospective cohort study. The patient’s informed consent
for study participation was obtained within 4 weeks fol-
lowing the transfer from ICU at acute care hospital to
post-acute ICU at a rehabilitation hospital (t1). Subse-
quently, the patient’s cognitive status was assessed using
the Confusion Assessment Method for the Intensive
Care Unit (CAM-ICU; [29,30]) vis-à-vis at bedside. The
diagnosis of a delirium or a positive evaluation of the
two CAM-ICU subtasks disorganized thinking and at-
tention screening examination precluded the study par-
ticipation and necessitated the repetition of the CAM-
ICU after 2 weeks. In case that patients were not able to
adequately communicate, their next-of-kin or designated
power of attorney for health care was contacted to get
the informed consent. At t2, 3 months following the
transfer from acute care ICU to post-acute ICU, and at
t3, 6 months post-transfer, the intensity of fatigue symp-
toms, posttraumatic stress symptoms, and the health-
related quality of life were assessed using questionnaires
via a telephone contact.
The present report was nested within a prospective-
longitudinal cohort study with the primary goal to assess
the rate and predictors of stress disorders following pro-
longed critical illness [31].
Participants
Patients with a principal diagnosis of CIP (ICD-10: G62.
80), CIM (ICD-10: G72.80), or CIP/CIM with or without
sepsis were eligible for a study participation. Patients
had to fulfill the following further inclusion criteria: age
between 18 and 72 years, a minimum length of ICU stay
of 6 days, mechanical ventilation, sufficient German lan-
guage skills, informed consent, and absence of a current
delirium. Above, data on the intensity of fatigue symptoms
had to be available at 3 months following the discharge
from ICU at acute care hospital. Participants who were
not alert showed fluctuating attention or consciousness or
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 2 of 12
could not communicate because of sedation or blocked
tracheal cannula were not included.
Measures
The Confusion Assessment Method for the Intensive Care
Unit (CAM-ICU) [29,30] is an instrument for the as-
sessment of delirium in the ICU. In the present study, it
was applied in order to screen the patients for the pres-
ence of acute cognitive dysfunctions. The number of
correctly identified letters or pictures in the auditory or
visual components of the Attention Screening Examin-
ation and the number of right answers in the subtask
disorganized thinking were summed up to an achieve-
ment score.
The Multidimensional Fatigue Inventory-20 (MFI-20)
[23,27] is a 20-item self-report measurement of fatigue
severity. The MFI-20 was applied via telephone inter-
view at 3 (t2) and 6 months (t3) following the discharge
from acute care ICU. It covers the five dimensions
General Fatigue, Physical Fatigue, Mental Fatigue, Re-
duced Motivation, and Reduced Activity. Items are
summed up to a simple total score with a minimum
value of 4 (absence of fatigue) and a maximum value of
20 for each subscale. A total fatigue score is calculated
as the sum of the subscale scores (range 20–100). Higher
total scores indicate higher levels of fatigue. Validity has
been shown for different participant populations, e.g.,
cancer patients, army recruits, and chronic fatigue syn-
drome [27]. Internal consistency has been shown to be
good for the General, Physical, and Mental Fatigue di-
mensions (Cronbach’s alpha .84) and adequate for the
subscales Reduced Activity and Reduced Motivation
(Cronbach’s alpha > .65) [27]. In the present study,
Cronbach’sαwas .91 at 3 and .93 at 6 months. The 75th
percentile was chosen as cutoff value for high fatigue
based on the subscale General Fatigue and the total
score (53+) of the MFI-20 total score [23,32,33].
Additionally, a current diagnosis of an affective dis-
order was assessed by an experienced clinical psycholo-
gist using the Structured Clinical Interview for the
Diagnostic and Statistical Manual of Mental Disorders
DSM-IV (SCID) [34] at 3 months (t2) and 6 months (t3)
following the discharge from ICU at acute care hospital.
The assessment of a lifetime history of an affective dis-
order was only realized at 6 months.
The Posttraumatic Stress Syndrome Scale (PTSS-10)
[35,36] is a 10-item self-report questionnaire for the es-
timation of the intensity of posttraumatic stress symp-
toms (e.g., sleep disturbance, nightmares, frequent
changes in mood) at 3 and 6 months following the dis-
charge from acute care ICU. The total score is received
by summing up the scores of all items (range 10–70).
The internal consistency and test-retest reliability of the
PTSS-10 can be regarded as high (Cronbach’sα= .92,
test-retest reliability r= .89) [37]. In the present study,
Cronbach’sαwas .82 at 3 and .87 at 6 months.
The Multidimensional Scale of Perceived Social Support
(MSPSS) [38] consists of 12 items and was applied to as-
sess the perceived support from three social sources (fam-
ily members, friends, significant others) at 3 and 6 months
post-ICU. Scores of all items are summed up to a total
score (range 12–84). Internal reliability has been shown to
be high (Cronbach’sα.89–.93) [39]. In the present study,
Cronbach’sαwas .90 at t2 and .89 at t3.
The health-related quality of life was measured with
the questionnaire Euro-Quality of Life (EQ-5D-3L) [40]
at 3 and 6 months. The EQ-5D-3L assesses five dimen-
sions (mobility, self-care, usual activities, pain/discom-
fort, and anxiety/depression). Additionally, the subjective
state of health is assessed by a visual analogue scale
(VAS) ranging from 0 (worst) to 100 (best). A single
one-dimensional index value was generated based on a
simple sum score according to Hinz et al. [41]. In the
present study, Cronbach’sαfor health-related quality of
life was .74 at t2 and .75 at t3.
Sociodemographic characteristics (e.g., age, marital sta-
tus, education), medical history (e.g., sepsis, location of
sepsis, number of sepsis episodes, duration of invasive
ventilation, duration of ICU stay, presence of somatic
comorbidities, severity of medical illness), and a life-time
diagnosis of a psychological disorder were obtained from
the patient records at t1. The Barthel index (BI)was
applied at admission and discharge from post-acute
ICU by a trained study nurse. The BI is a measure of
performance in activities of daily living in 11 domains
(e.g., fecal/urinary incontinence) with values ranging
between 0 and 100. Additionally, the early rehabilita-
tion BI (range −325–0) was used assessing seven further
domains (e.g., intensive care supervision, tracheostomy
tube management, mechanical ventilation, confusion, se-
vere impairment of communication, dysphagia) [42]. The
scores of both Barthel scales were summed up, yielding a
minimum value of −325 and a maximum value of 100. A
higher value indicates a better performance. Interrater re-
liability is very high (r= .95). Test-retest reliability is good
as well (r=.89) [43].
Statistical methods
For normally distributed data, arithmetic means, and
standard deviations, otherwise, medians and interquar-
tile ranges are reported. For categorical variables, fre-
quencies and percents are shown. Bivariate correlational
analyses were calculated depending on the level of meas-
urement between sociodemographic, clinical, and psy-
chological variables and fatigue symptoms at t2 and t3.
The course of fatigue symptoms was examined with the
analysis of covariance (ANCOVA) for repeated measures
for normally distributed subscales of the MFI-20
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 3 of 12
(General Fatigue) and the total fatigue score. As covari-
ates, age and gender were included. In case of non-
normally distributed subscales of the MFI-20, the sign
test was applied. A missing value in one item of the MFI
at t3 was recorded in one participant and replaced by
the median of the subscale.
MFI-20 scores were compared with age- and gender-
stratified subgroups of a normative German sample [23].
Standardized mean differences (Hedges’g) with 95%
confidence interval (CI) were calculated for these com-
parisons. As the cutoff value for high fatigue, the 75th
percentile (53+) was used according to Kuhnt et al. [32].
For the classification with respect to high vs. low Gen-
eral Fatigue, the age- and gender-adjusted 75th percen-
tiles from a German representative community sample
were applied [23,33].
Clinical (e.g., sepsis-related characteristics, length of
mechanical ventilation/ICU stay, severity of medical illness/
Barthel index, number of medical comorbidities), psycho-
logical (e.g., prior psychiatric history, diagnosis of PTSD/
major depression, perceived fear of dying/helplessness at
ICU), and sociodemographic (e.g., age, gender, family/edu-
cation status) factors which were correlated with the
dependent variable (MFI-20 total score) with a pvalue < .2
[44] were entered in a multivariable linear regression ana-
lysis. Correlational analyses using Spearman’srankcorrel-
ation were calculated between the MFI-20 total score, the
health-related quality of life (EQ-5D-3L), and posttraumatic
stress symptomatology (PTSS-10). For all analyses, a signifi-
cance level of α= 0.05 (two-sided) was applied. All data
were analyzed using SPSS 24 (SPSS Inc., Chicago, IL, USA).
Results
Descriptive data
Of the N= 352 potentially to be enrolled CCI patients
with the primary diagnosis of CIP, CIM or CIP/CIM,
n= 157 (44.6%) patients could not be enrolled for dif-
ferent reasons (e.g., death, positive CAM-ICU, refusal
of study participation; see Fig. 1). Finally, data of n=113
patients were available at t2 and of n=91 patients at t3
(Fig. 1). Table 1displays the major sociodemographic,
clinical, and psychological characteristics of the sample at
Fig. 1 Flow chart including the dropped out patients and final sample of CCI patients. CAM-ICU: Confusion Assessment Method for the Intensive
Care Unit; CIP/CIM: Critical Illness Polyneuropathy/Critical Illness Myopathy; SCID: Structured Clinical Interview for DSM (Diagnostic and Statistical
Manual of Mental Disorders)–IV disorders
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 4 of 12
Table 1 Descriptive characteristics of chronically critically ill (CCI) patients (n= 113) and the subsamples of patients with high (n= 61) vs.
low fatigue (n= 52) at 3 months (t2) following the discharge from ICU at acute care hospital
Characteristic Patients
(n= 113)
High fatigue
(n= 61)
a
Low fatigue
(n= 52)
a
U/χ
2
(p)
b
Sociodemographic variables
Age, years median (IQR) 61.1 (55.7–65.6) 61.5 (56.3–65.6) 58.7 (54.6–65.5) 1351.000 (.176)
c
Gender, n(%)
Male 82 (72.6) 45 (73.8) 37 (71.2)
Female 31 (27.4) 16 (26.2) 15 (28.8) .097 (.756)
d
Family status, n(%)
Single 10 (8.8) 5 (8.2) 5 (9.6)
Married/cohabited 78 (69.0) 48 (78.7) 30 (57.7)
Divorced/living apart 16 (14.2) 3 (4.9) 13 (25.0)
Widowed 9 (8.0) 5 (8.2) 4 (7.7) 10.594 (.032*)
d
Partnership
Yes 78 (69.0) 48 (78.7) 30 (57.5)
No 35 (31.0) 13 (21.3) 22 (42.3) 5.788 (.016*)
d
Education, n(%)
e
< 10 years 35 (31.0) 17 (27.9) 33 (63.5)
≥10 years 72 (63.7) 39 (63.9) 18 (34.6) .296 (.587)
d
Clinical variables
Sepsis, n(%)
No sepsis 36 (31.9) 18 (29.5) 18 (34.6)
Sepsis 42 (37.2) 26 (42.6) 16 (30.8)
Severe sepsis or septic shock 35 (31.0) 17 (27.9) 18 (34.6) 1.704 (.427)
d
Number of sepsis episodes, median (IQR) 1.0 (0.0–1.0) 1.0 (0.0–1.0) 1.0 (0.0–1.0) 1512.500 (.911)
d
Site of infection, n(%)
Respiratory 56 (49.6) 32 (52.5) 24 (46.2) .446 (.504)
d
Urinary/genitals 12 (10.6) 5 (8.2) 7 (13.5) .820 (.365)
d
Abdominal 10 (8.8) 4 (6.6) 6 (11.5) .863 (.509)
f
Bones/soft tissue 6 (5.3) 4 (6.6) 2 (3.8) .410 (.685)
f
Wound infection 2 (1.8) 1 (1.6) 1 (1.9) .013 (1.000)
f
Heart 1 (.9) 1 (1.6) 0 (0.0) .860 (1.000)
f
Multiple 13 (11.5) 5 (8.2) 8 (15.4) 1.425 (.233)
d
Others
g
8 (7.1) 2 (3.3) 6 (11.5) 2.911 (.140)
f
Unknown 4 (3.5) 2 (3.3) 2 (3.8) .026 (1.000)
f
Barthel index, median (IQR)
At admission at post-acute ICU −200.0 (−225.0–125.0) −185.0 (−225.0–100.0) −200.0 (−225–128.8) 1538.500 (.781)
c
At discharge from post-acute ICU −35.0 (−82.5–7.5) −25.0 (−80.0–35.0) −40.0 (−85.0–0.0) 1319.500 (.124)
c
At discharge from rehabilitation hospital 65.0 (35.0–85.0) 65.0 (0.0–80.0) 75.0 (60.0–88.8) 1226.500 (.038*)
c
ICU stay, days median (IQR) 66.0 (49.0–93.5) 69.0 (46.0–87.0) 62.0 (49.0–111.5) 1585.000 (.977)
c
Mechanical ventilation, days median (IQR) 47.0 (33.0–70.0) 45.0 (30.0–71.5) 50.5 (33.5–69.8) 1405.500 (.298)
c
Number of medical comorbidities,
median (IQR)
9.0 (7.0–12.0) 10.0 (8.0–13.0) 8.0 (6.3–11.0) 1098.000 (.005**)
c
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 5 of 12
3 months (t2). CCI patients with complete data of the
MFI-20 (n= 113) had a median age of 61.1 years. 72.
6% (n= 82) were men. Acute respiratory insufficiency
(n= 87, 77.0%), left heart failure (n= 40, 35.4%), and
diabetes (n= 43, 38.1%) occurred as the most common
medical comorbidities (Additional file 1:TableS1).
Non-participants were significantly more severely ill than
patients being followed up as shown by a lower Barthel
index at discharge from post-acute ICU/rehabilitation
hospital (data not shown, p< .001). Non-participants were
less often educated 10 years or longer (non-participants
vs. participants: 37.7 vs. 63.7%, p= .024). Above, signifi-
cantly more non-participants suffered from hypertension
(non-participants vs. participants: 21.8 vs. 12.4%, p=.036),
organic brain syndrome (53.1 vs. 38.9%, p=.013), neuro-
logical disorders (37.7 vs. 23.9%, p=.010), or pneumonia
(31.4 vs. 17.7%, p=.007).
At t2, CCI patients with high fatigue showed a signifi-
cantly lower Barthel index at discharge from rehabilitation
hospital, a higher number of medical comorbidities, lived
more often in a partnership and had more often a current
diagnosis of major depression than patients with low fa-
tigue (see Table 1for detailed information). Patients with
high fatigue more often showed certain medical comor-
bidities (e.g., left heart failure, coronary heart disease, or-
ganic brain syndrome, Additional file 1: Table S1). At t3,
CCI patients with high fatigue were significantly older and
had more often a diagnosis of major depression, posttrau-
matic stress disorder (PTSD), or prior history of an
anxiety disorder (Additional file 2: Table S2).
Point prevalence rates and intensity of fatigue
According to the total score/the subscale General Fatigue
of the MFI, 54.0% (n= 61)/46.9% (n= 53) and 49.5%
(n= 45)/45.1% (n= 41) presented with clinically rele-
vant symptoms of self-reported fatigue at t2 and t3,
respectively. There was no significant difference with
respect to the classification between t2 and t3 (McNemar
test: χ
2
=.000, p> .824). At both time points and in all
MFI subscales, CCI patients reported fatigue scores about
one standard deviation above the scores of a German
general population [23]. High effect sizes were obtained
for the MFI subscales Physical Fatigue, Reduced Activity,
and General Fatigue (see Fig. 2a,b).
Course of fatigue from 3 (t2) to 6 (t3) months following
acute care ICU
According to the total fatigue score, there was a significant
time × gender effect (F(1, 88.0) = 5.604, p= .020, η
2
= .060).
While female patients showed a significant decrease over
time, men remained stable from t2 to t3 (see Fig. 3). All
other effects were non-significant (main effect of time:
Table 1 Descriptive characteristics of chronically critically ill (CCI) patients (n= 113) and the subsamples of patients with high (n= 61) vs.
low fatigue (n= 52) at 3 months (t2) following the discharge from ICU at acute care hospital (Continued)
Characteristic Patients
(n= 113)
High fatigue
(n= 61)
a
Low fatigue
(n= 52)
a
U/χ
2
(p)
b
Psychological variables at (post-acute) ICU
Perceived fear of dying at ICU
h
, median (IQR) 1.0 (1.0–6.0) 2.0 (1.0–6.0) 1.0 (1.0–5.0) 1375.500 (.321)
c
Perceived social support according to MSPSS
h
,
median (IQR)
6.3 (5.4–6.9) 6.3 (5.4–6.9) 6.3 (5.8–7.0) 1448.500 (.425)
c
Diagnosis of major depression according to SCID I
h
,
n(%)
9 (8.0) 8 (13.1) 1 (1.9) 5.023 (.035*)
f
Diagnosis of posttraumatic stress disorder (PTSD)
according to SCID I, n(%)
18 (15.9) 13 (21.3) 5 (9.6) 3.137 (.077)
d
Prior psychiatric history
History of harmful alcohol consumption, n(%) 22 (19.5) 12 (19.7) 10 (19.2) .003 (.953)
d
History of anxiety disorders, n(%) 8 (7.1) 5 (8.2) 3 (5.8) .251 (.724)
f
History of depressive disorders, n(%) 23 (20.4) 11 (18.0) 12 (23.1) .441 (.507)
d
History of psychological disorder, n(%) 70 (61.9) 34 (55.7) 36 (69.2) 2.168 (.141)
d
IQR interquartile range, MSPSS Multidimensional Scale of Perceived Social Support, PTSD posttraumatic stress disorder, SCID I Structured Clinical Interview
according to DSM IV
*p≤.05, **p≤.01
a
Subsamples were generated using the cutoff score 53+ suggested by Kuhnt et al. [32]
b
Statistical value and pvalue refer to the comparison between the subsamples of patients with high vs. low fatigue
c
pvalue from Mann-Whitney Utest
d
pvalue from chi-squared test
e
n= 6 miss ing values; high fatigue: n= 5, low fatigue: n=1
f
pvalue from Fisher’s exact test
g
n= 1 brain , n= 5 centr al venous catheter, n= 1 port system, n= 1 urinary catheter; high fatigue: n= 1 brain, n= 1 central venous cathete r, low fatigue: n= 1 port
system, n= 1 urinary catheter, n= 4 central venous catheter
h
n= 2 missing values
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 6 of 12
F(1, 88.0) = 1.262, p= .264, η
2
= .014; main effect age:
F(1, 88.0) = 2.749, p= .101, η
2
= .030; main effect
gender: F(1, 88.0) = 1.537, p= .218, η
2
= .017; time ×
age: F(1, 88.0) = .010, p= .920, η
2
= .000). Results were
similar for the subscale General Fatigue. With respect
to the other MFI subscales, there was no change from
t2 to t3 (all pvalues > .512).
Risk factors of fatigue in CCI patients
In univariate regression analyses, older age and living in
a partnership were significant sociodemographic corre-
lates of an increased fatigue score at t2. As clinical
factors, the Barthel index at discharge from the rehabili-
tation hospital, the number of medical comorbidities,
and the presence of a coronary heart disease could be
identified. A SCID-I diagnosis of major depression or
PTSD, the perceived fear of dying, and the perceived social
support could be elucidated as significant psychological
factors (Additional file 3: Table S3). In the multivariable
regression model, controlling for age and gender, the pres-
ence of a coronary heart disease, the perceived fear of
dying at ICU, a diagnosis of major depression, and the
perceived social support explained 20.1% of the total
variance (Table 2).
At t3, partnership, female gender, a diagnosis of
major depression/PTSD, the number of medical comor-
bidities, and a prior history of an anxiety disorder were
significant correlates of an increased fatigue score
(Additional file 4: Table S4). In a multivariable regres-
sion analysis, these factors explained a total variance of
31.8% (Table 3).
Association with posttraumatic stress and health-related
quality of life
The total fatigue score was significantly positively corre-
lated with the posttraumatic symptom score of the PTSS-
10 at both t2 (Spearman’srho=.656,p< .001) and t3
(Spearman’srho=.600,p< .001). Fatigue was significantly
negatively correlated with the health-related quality of
life as measured using the EQ-5D-3L (t2: Spearman’s
rho = −.648, p< .001; t3: Spearman’srho=−.663, p< .001).
Discussion
The main aim of the present study was to assess the
frequency and the course of fatigue in CCI patients fol-
lowing protracted treatment on ICU. Furthermore, the
investigation of related sociodemographic, clinical, and
psychological factors was of interest in these patients.
Currently, no study exists characterizing CCI patients
with respect to a chronic state of exhaustion as after-
math following the survival of long-term mechanical
ventilation. The present study finding is of clinical rele-
vance, since fatigue is one of the most debilitating and
distressing complaint in survivors of intensive care treat-
ment [9,18–20]. Unrecognized and untreated fatigue
may have impeding effects on the patients’health-
related quality of life and rehabilitation process [16].
Fig. 2 a,bEffect sizes with 95% CI comparing patients and the general population at t2 and t3. CI: confidence interval; German general
population (N= 2037) according to Schwarz et al. [23]
Fig. 3 The course of the MFI-20 total score for male and female
patients at 3 (t2) and 6 (t3) months following the discharge from ICU
at acute care hospital. MFI-20: Multidimensional Fatigue Inventory
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 7 of 12
The findings of the present study elucidate that nearly
every second patient suffered from an overwhelming
sense of tiredness at rest up to 6 months post-ICU. This
is in line with Steenbergen et al. [18] and Chaboyer and
Grace [20] showing reports of clinically relevant fatigue
between 37% and more than 50% even 1 year after ICU
discharge. Moreover, CCI patients presented with fatigue
scores one standard deviation above a representative
sample of the adult German population which is in ac-
cordance with findings by Rosendahl et al. [45] in pa-
tients surviving severe sepsis.
According to the impact of gender on self-reported fa-
tigue in CCI patients, female patients had lower values
and presented a decline up to 6 months post-ICU. This
is in contrast to other findings in the general population
(e.g., [23,46]). In the present sample, male patients were
more severely ill than female patients as shown in a
higher number of medical comorbidities (Mann-Whitney
U= 934.500, p= .030). Thus, the symptoms of multiple
concomitant diseases and the necessitating intensified
treatment options may have led to a higher vital exhaus-
tion in male CCI patients. However, our result should be
Table 2 Multivariable linear regression (stepwise) showing significant clinical and psychological variables of total fatigue as measured
with the MFI-20 in chronically critically ill patients (n= 113) 3 months following the discharge from ICU at acute care hospital. The final
model was controlled for age and gender
Multivariable linear regression
a
Beta CI pvalue
Clinical variables
Coronary heart disease .27 .23–1.03 .002**
Psychological variables at (post-acute) ICU
Perceived fear of dying at ICU .25 .08–.42 .005**
Psychological variables 3 months following ICU
Diagnosis of major depression according to SCID I .26 .31–1.56 .004**
Perceived social support according to MSPSS −.18 −.35–(−).01 .043*
R
2
(corrected): .201 (F(4, 108) = 7.771, p< .001)
a
Method stepwise; PTSD at t2 and perceived fear of dying at ICU were significantly correlated (point biserial coefficient =.242, p= .011); family status and MSPSS
at t2 were significantly correlated (point biserial coefficient = −.264, p= .005). Number of medical comorbidities and diagnosis of major depression/coronary heart
disease were significantly correlated (point biserial coefficient = .279, p= .003/.305, p= .001). For parsimony of the final model and to prevent multicollinearity,
PTSD at t2, family status and number of medical diagnoses were not considered in the final model. Tolerance/variance inflation factor and condition number test
did not indicate multicollinearity
MFI-20 Multidimensional Fatigue Inventory, MSPSS Multidimensional Scale of Perceived Social Support, PTSD posttraumatic stress disorder, SCID I Structured
Clinical Interview according to DSM IV
*p≤.05, **p≤.01
Table 3 Multivariable linear regression (stepwise) showing significant sociodemographic, clinical, and psychological variables of total
fatigue as measured with the MFI-20 in chronically critically ill patients (N= 91) 6 months following the discharge from ICU at acute
care hospital. The final model was controlled for age and gender
Multivariable linear regression
a
Beta CI pvalue
Sociodemographic variables
Gender (male vs. female) −.23 −.91–(−).11 .013*
Clinical variables
Number of medical comorbidities .18 .00–.35 .045*
Psychological variables 6 months following ICU
Diagnosis of major depression according to SCID I .44 .80–1.87 < .001**
Prior psychiatric history
History of anxiety disorder .32 .55–1.85 < .001**
R
2
(corrected): .318 (F(4, 90) = 11.497, p< .001)
a
Method stepwise; number of medical comorbidities and PTSD at t3/family status were significantly correl ated (point biserial coefficient = .251, p= .016/.380,
p< .001). For parsimony of the final model and to prevent multicollinearity, PTSD at t3 and family status were not considered in the final model. Tolerance/
variance inflation factor and condition number test did not indicate multicollinearity
MFI-20 Multidimensional Fatigue Inventory, SCID I Structured Clinical Interview according to DSM IV
*p≤.05, **p≤.001
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 8 of 12
interpreted thoughtfully, taking into account the peculiar-
ities of the present sample; for instance, three quarter
were male CCI patients. Nonetheless, gender differences
in self-reported fatigue have not been consistently shown
in the previous literature either (e.g., [47]). Beyond, in the
present study, fatigue seemed to improve over time in fe-
male CCI patients whereas men showed a rather stable
course. Former results pointed out that people affected by
chronic fatigue improve with time but most remain func-
tionally impaired for several years, independent from gen-
der [48]. The course of fatigue has not yet been investigated
before in CCI patients. Future studies should therefore
more differentially consider the multiple aspects of fatigue
in men and women suffering from CCI. Following, our re-
sult needs replication in a larger sample of CCI patients
with an even distribution of male/female patients and a
higher representation of patients 40 years or younger.
Regarding the impact of age, older CCI patients
showed significantly higher fatigue values than younger
CCI patients, particularly at t2. This result is consistent
with data based on a German general population [23].
Increasing age goes along with a reduction in muscle
strength, sarcopenia, and an increasing variability in
motor neuron firing rates [2]. Above, the risk for chronic
diseases, multi-morbidity, and a loss of psychosocial
functioning increases with age and might thus contribute
to higher fatigue scores [46].
A higher number of medical comorbidities and
greater illness severity were associated with a higher
self-reported fatigue. This is in line with the findings by
Chlan and Savik [49] in mechanically ventilated pa-
tients. One may assume that more severely ill patients
are exposed to an intensified medical therapy, an in-
creased number of medications, and a heightened bur-
den of concomitant diseases, entailing an increased
fatigue level.
Three months post-ICU, the presence of a coronary
heart disease turned out as one of the most prominent
risk factor for increased fatigue values. Fatigue has been
widely studied in patients with acute or chronic cardio-
vascular diseases (e.g., see [50,51]). In these patients, fa-
tigue constitutes one of the most distressing health
complaints. It goes along with the inability to perform
physical or intellectual efforts, a decreased health-related
quality of life, and impedes participation in physical ac-
tivity [51]. In the present sample of CCI patients, one
quarter was affected by a chronic coronary heart disease
(CHD). In these patients, an intricate interplay between
neuroendocrine and hemodynamic dysfunction may lead
to a mismatch of cardiac output to the patient’s need
during exercise and goes along with peripheral muscle
deconditioning. Together with the CIP/CIM as compli-
cation of critical illnesses such as sepsis, systemic inflam-
matory response syndrome, and multiple organ failure,
difficulties in weaning from mechanical ventilation oc-
curred in these patients and consequently led to pro-
longed ICU stays (median 47.0), long-term disabilities,
and a muscular/cardio-respiratory deconditioning [52].
The perceived fear of dying at ICU, a SCID-based
diagnosis of major depression or PTSD, and a prior anx-
iety disorder were identified as further prominent risk
factors for increased self-reported fatigue in CCI pa-
tients. Moreover, fatigue levels were significantly related
to posttraumatic symptomatology 3 and 6 months post-
ICU. According to the DSM-IV/V, fatigue shows consid-
erable conceptual overlap with some mood and anxiety
disorders [24,53,54]. For instance, core features of a
major depression are a loss of energy/vitality nearly
every day, anhedonia, psychomotor retardation, and di-
minished ability to concentrate. These symptoms are
also comprised in the MFI subscales, especially Reduced
Activity/Reduced Motivation and Mental Fatigue. In line,
Bunevicius et al. [51] showed that self-reported fatigue is
strongly related to symptoms of anxiety and depression
in patients with coronary artery disease. Above, Eckhardt
et al. [7] have confirmed this result, proving depressive
symptoms as sole predictor of fatigue intensity/interfer-
ence from fatigue in patients with coronary heart
disease.
Moreover, the fear of dying has not yet been proven as
risk factor for increased fatigue in CCI patients. In a
study by Wade et al. [55], the acute psychological reac-
tion in the ICU displayed the strongest risk factor for
poor psychosocial outcomes after ICU treatment. It can
be supposed that patients with higher perceived fear of
dying experienced a higher number of traumatic inter-
ventions because of a higher severity of medical ill-
ness. A correlational analysis with the Barthel index
confirmed this assumptiononlybytrend(Spearman’s
r=−.172, p= .071). Otherwise, the peri-ICU stress re-
action shares common features with anxiety and mood
disorders and may constitute an indicator of a psycho-
logical vulnerability or a prior mental disorder predis-
posing the patient to the development of specific
subfacets of fatigue (e.g., Mental Fatigue, Reduced
Activity, Reduced Motivation) [7]. Furthermore, a
shared pathophysiologic mechanism between affective
disorders and fatigue with respect to the autonomous
nervous system, the hypothalamic-pituitary-adrenal
axis, or immunological functions has been demon-
strated [56].
While the perceived social support turned out to be
an essential resource and seems to have a buffering ef-
fect on self-reported fatigue in former studies (e.g.,
[46]), being married or living in a partnership could be
identified as a risk factor for increased self-reported
fatigue values in our present study. Patients living in a
partnership do not necessarily receive greater social
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 9 of 12
support than patients living alone. Survivors of CCI
show profound changes in their family/friend relation-
ships and their ability to participate in social roles and
activities [8]. In our sample of CCI patients, partners
were often affected by a chronic disease or disabilities
themselves, keeping them from visiting the patient in
the study center or granting adequate relief. Future
studies should also implement a multidimensional
measure of perceived social support and not solely rely
on the simple assessment of partnership status via yes/
no answering. Above, our finding of a salutogenetic im-
pact of social support by family, friends, or significant
others led us to conclude that a multidisciplinary ap-
proach is recommended in the treatment of self-reported
fatigue in CCI patients, taking into account the needs of
the patient’s whole social or family system.
An interesting and new finding of our present study
displays the fact that major depression was the only
common factor associated with fatigue at both t2 and t3.
One reason for this might be the composition and selec-
tion of variables considered in the final models. For in-
stance, some variables were excluded for reasons of
parsimony of the final model and multicollinearity.
Otherwise, there is evidence that certain variables have
short-term effects on psychiatric outcomes like anxiety
or depression while others, particularly related to con-
tinuing behaviors associated with prior psychiatric his-
tory, have long-term effects [57].
The results of our present study should be thoroughly
evaluated in the context of methodological limitations.
Generally, univariate analyses should be interpreted cau-
tiously because they may be due to confounding. Above,
the present study was primarily designed to assess a
posttraumatic stress disorder in CCI patients following
prolonged treatment on ICU. Fatigue was measured as
secondary outcome. Nonetheless, a post hoc power ana-
lysis with Cohen’sf
2
of .25 and .47 for t2 and t3,
respectively; an αlevel of 5%; a sample size of n= 113/91;
and eight regressors revealed a power of nearly 100%.
Another shortcoming concerns the lacking applica-
tion of objective laboratory tests in order to validate the
self-reported fatigue scores. Future studies should addi-
tionally determine exercise capacity measuring peak
VO
2
, e.g., during bicycle ergometry, in order to specify
the self-reported fatigue on an objective level and to
differentiate between the physical and mental compo-
nents of fatigue [50]. Also, the characterization of the
functional status by measuring the activity of daily living
and neuropsychological functioning should be realized in
future studies. Furthermore, the impact of the current
general medical conditions (e.g., cardiac, pulmonary, hep-
atic, renal, physical) as well as medication (e.g., sedatives,
analgetics; [49]) on long-term fatigue intensity following
protracted ICU treatment should be considered.
The validity of the point prevalence rates of fatigue
reported in our present study has to be carefully con-
sidered since the use of the 75th percentile as cutoff
criteria for high fatigue seems to be a bit arbitrary. The
cutoff value for the total fatigue score was retrieved
from a sample of cancer patients, not the general popu-
lation [32]. When a more conservative cut-off criterion
(90th percentile) for the classification of high fatigue is
applied, after all, only n= 34 (30.1%) of the CCI patients
could be still identified as cases at t2 and n=22 (24.2%)
at t3.
Another limiting fact concerns the mixture of patients
without and with sepsis. The latter displays one of the
most prominent risk factors for the development of CIP
or CIM [52]. Mixing up both subsamples may have led
to an inadequate estimation of the point prevalence rate
of fatigue. We have decided to merge both subsamples
because the validity of the sepsis diagnoses made by the
acute care hospitals was questionable. Above, both sub-
samples did not significantly differ with respect to the
main descriptive characteristics. In line, our results
could be, by and large, replicated when only patients
with sepsis were considered.
No information regarding the self-reported fatigue was
available in the forefront of the ICU admission. Thus,
we cannot causatively attribute high fatigue levels to the
protracted treatment on ICU. Furthermore, patients of
the present sample reported multiple reasons for fatigue;
among them were morbidity, difficulties in breathing,
medication, sleep disorders, and pain. Thus, profound
diagnostics of fatigue are required leading to individually
tailored interventions which are targeted on the multiple
reasons of fatigue.
Finally, the present study sample of CCI patients can
be regarded as a selective sample due to the high drop-
out rate of nearly 58%. Although the latter mirrors the
daily clinical situation and is in line with other studies
(e.g., [58]), it cannot be ruled out that the present fatigue
values might be underestimated since more severely ill
patients or patients with a lower functional status have
not been followed up.
Conclusion
To conclude, self-reported fatigue is a common symp-
tom in CCI patients surviving intensive care after pro-
longed invasive ventilation. Male gender, the illness
severity, a diagnosis of coronary heart disease, major de-
pression, and the fear of dying at ICU were most intim-
ately related to increased fatigue while the perceived
social support turned out as salutogenetic factor. ICU
survivors at risk should be regularly evaluated in routine
clinical care following ICU discharge, and specialized in-
terventions should be offered to them.
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 10 of 12
Additional files
Additional file 1: Table S1. Medical comorbidities of chronically
critically ill (CCI) patients (n= 113) and the subsamples of patients with
high (n= 61) vs. low fatigue (n= 52) at 3 months (t2) following the
discharge from ICU at acute care hospital.
a
pvalue from chi-squared test;
b
pvalue from Fisher’s exact test; *p≤.05, **p≤.01. (DOCX 18 kb)
Additional file 2: Table S2. Descriptive characteristics of chronically
critically ill (CCI) patients (n= 91) and the subsamples of patients with
high fatigue (n= 45) vs. low fatigue (n= 46) at 6 months (t2) following
the discharge from ICU at acute care hospital.
a
Subsamples were generated
using the cutoff score 53+ suggested by Kuhnt et al. [32];
b
Statistical value
and pvalue refer to the comparison between the subsamples of patients
with high fatigue vs. low fatigue;
c
pvalue from Mann-Whitney Utest;
d
pvalue from chi-squared test;
e
n= 5 missing values; high fatigue: n=2,
low fatigue: n=3;
f
pvalue from Fisher’s exact test;
g
n= 1 brain, n= 3 central
venous catheter, n= 1 urinary catheter; high fatigue: n=1 brain,n=1
central venous catheter, low fatigue: n= 1 urinary catheter, n= 2 central
venous catheter; IQR = interquartile range, *p≤.05. (DOCX 22 kb)
Additional file 3: Table S3. Univariate linear regression for the identification
of sociodemographic, clinical, and psychological predictors of total fatigue as
measured with the MFI-20 in chronically critically ill patients (N= 113) 3 months
following the discharge from ICU at acute care hospital.
1
n= 7 missing values;
2
n= 1 missing value;
3
n= 2 missing values; ASDS = Acute Stress Disorder Scale;
ASD = Acute Stress Di sorder; CAM-ICU = Confusion Assessment Method for the
Intensive Care Unit; MFI-20 = Multidimensional Fatigue Inventory; MSPSS =
Multidimensional Scale of Perceived Social Support; PTSD = Posttraumatic Stress
Disorder; SCID I = Structured Clinical Interview according to DSM IV; *p≤.05,
**p≤.01, ***p≤.001. (DOCX 18 kb)
Additional file 4: Table S4. Univariate linear regression for the
identification of sociodemographic, clinical, and psychological
predictors of total fatigue as measured with the MFI-20 in chronically
critically ill patients (N= 91) 6 months following the discharge from
ICU at acute care hospital.
1
n= 7 missing values;
2
n=1missing value;
3
n= 2 missing values; ASDS = Acute Stress Disorder Scale; ASD = Acute S tress
Disorder; CAM-ICU = Confusion Assessment Method for the Intensive Care
Unit; MFI-20 = Multid imensional Fatigue Inventory; MSPSS = Multidimensional
Scale of Perceived Social Support; PTSD = P osttraumatic Stress Disorder;
SCID I = Structured Clinical Interview according to DSM IV; *p≤.05, **p≤.01,
***p≤.001. (DOCX 18 kb)
Abbreviations
CAM-ICU: Confusion assessment method for the intensive care unit;
CCI: Chronic critical illness; CIM: Critical illness myopathy; CIP: Critical illness
polyneuropathy; DSM-IV: Diagnostic and Statistical Manual of Mental Disorders;
EQ-5D-3L: Euro-Quality of Life; ICU: Intensive care unit; MFI-20: Multidimensional
fatigue inventory; MSPSS: Multidimensional Scale of Perceived Social Support;
PTSS-10: Posttraumatic Symptom Scale; SCID: Structured Clinical Interview for
the Diagnostic and Statistical Manual of Mental Disorders DSM-IV
Acknowledgements
We would like to kindly thank Stefan Rueckriem, Sara Wuestemann, Clara L.
Buck, Christine Schier, and Corinna Klotzsche for their dedicated support in
the patient enrollment and data assessment.
Funding
This study was supported by the German Federal Ministry of Education and
Research grant 01EO1002.
Availability of data and materials
Data from the present trial can be obtained on request by emailing the
corresponding author (GBW).
Authors’contributions
GBW realized the statistical analyses and wrote the manuscript. KP and JR
supervised the data collection. AH proof-read the manuscript and provided
expertise in fatigue. BS and KW gave methodological and statistical advice
on the study design, enrollment, and data analysis. All authors read and
approved the final manuscript.
Ethics approval and consent to participate
The present longitudinal, prospective cohort study was approved by the
local ethics committee of the Friedrich-Schiller University, Jena, Germany
(no. 3278-10/11). All participants have signed written informed consent and
provided verbal consent before the telephone interview.
Consent for publication
All authors have provided consent for publication of the present manuscript.
Competing interests
The authors declare that they have no competing interests.
Publisher’sNote
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Author details
1
Department of Psychotherapy and Psychosomatic Medicine, Medizinische
Fakultät Carl Gustav Carus, Technische Universität Dresden, Dresden
Fetscherstraße 74, 01307 Dresden, Germany.
2
Center for Sepsis Control and
Care, Jena University Hospital, Friedrich-Schiller University, Jena, Germany.
3
Institute of Psychosocial Medicine and Psychotherapy, Jena University
Hospital, Friedrich-Schiller University, Jena, Germany.
4
Department of Medical
Psychology and Medical Sociology, University of Leipzig, Leipzig, Germany.
Received: 1 February 2018 Accepted: 4 April 2018
References
1. Cella D, Peterman A, Passik S, Jacobsen P, Breitbart W. Progress toward
guidelines for the management of fatigue. Oncology (Williston Park). 1998;
12(11A):369–77.
2. Finsterer J, Mahjoub SZ. Fatigue in healthy and diseased individuals. Am J
Hosp Palliat Care. 2014;31(5):562–75.
3. Gulanick M, Myers JL. Nursing care plans: diagnoses, interventions, and
outcomes. 7th ed. St.Louis: Elsevier/Mosby; 2011.
4. Biberacher V, Schmidt P, Selter RC, Pernpeinter V, Kowarik MC, Knier B, et al.
Fatigue in multiple sclerosis: associations with clinical, MRI and CSF
parameters. Mult Scler. 2017; https://doi.org/10.1177/1352458517712078.
5. Farragher JF, Polatajko HJ, Jassal SV. The relationship between fatigue
anddepressioninadultswithend-stagerenaldiseaseonchronicin-
hospital hemodialysis: a scoping review. J Pain Symptom Manag.
2017;53(4):783–803 e781.
6. Lou JS, Kearns G, Oken B, Sexton G, Nutt J. Exacerbated physical fatigue and
mental fatigue in Parkinson’s disease. Mov Disord. 2001;16(2):190–6.
7. Eckhardt AL, Devon HA, Piano MR, Ryan CJ, Zerwic JJ. Fatigue in the
presence of coronary heart disease. Nurs Res. 2014;63(2):83–93.
8. Hashem MD, Nallagangula A, Nalamalapu S, Nunna K, Nausran U, Robinson
KA, et al. Patient outcomes after critical illness: a systematic review of
qualitative studies following hospital discharge. Crit Care. 2016;20(1):345.
9. Spadaro S, Capuzzo M, Valpiani G, Bertacchini S, Ragazzi R, Dalla Corte F,
et al. Fatigue in intensive care survivors one year after discharge. Health
Qual Life Outcomes. 2016;14(1):148.
10. Cajanding RJ. Causes, assessment and management of fatigue in critically ill
patients. Br J Nurs. 2017;26(21):1176–81.
11. Wessely S, Hotopf M, Sharpe M. Chronic fatigue and its syndromes. Oxford:
Oxford University Press; 1998.
12. Bissett B, Leditschke IA, Neeman T, Boots R, Paratz J. Weaned but weary:
one third of adult intensive care patients mechanically ventilated for 7 days
or more have impaired inspiratory muscle endurance after successful
weaning. Heart Lung. 2015;44(1):15–20.
13. Girard K, Raffin TA. The chronically critically ill: to save or let die? Respir Care.
1985;30(5):339–47.
14. Nelson JE, Cox CE, Hope AA, Carson SS. Chronic critical illness. Am J Respir
Crit Care Med. 2010;182(4):446–54.
15. Kahn JM, Le T, Angus DC, Cox CE, Hough CL, White DB, et al. The
epidemiology of chronic critical illness in the United States. Crit Care Med.
2015;43(2):282–7.
16. Matthews EE. Sleep disturbances and fatigue in critically ill patients. AACN
Adv Crit Care. 2011;22(3):204–24.
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 11 of 12
17. Griffiths RD, Jones C. Recovery from intensive care. BMJ. 1999;319(7207):
427–9.
18. Steenbergen S, Rijkenberg S, Adonis T, Kroeze G, van Stijn I, Endeman H.
Long-term treated intensive care patients outcomes: the one-year mortality
rate, quality of life, health care use and long-term complications as reported
by general practitioners. BMC Anesthesiol. 2015;15:142.
19. Puntillo KA, Arai S, Cohen NH, Gropper MA, Neuhaus J, Paul SM, et al.
Symptoms experienced by intensive care unit patients at high risk of dying.
Crit Care Med. 2010;38(11):2155–60.
20. Chaboyer W, Grace J. Following the path of ICU survivors: a quality-
improvement activity. Nurs Crit Care. 2003;8(4):149–55.
21. Choi J, Hoffman LA, Schulz R, Tate JA, Donahoe MP, Ren D, et al. Self-
reported physical symptoms in intensive care unit (ICU) survivors: pilot
exploration over four months post-ICU discharge. J Pain Symptom Manag.
2014;47(2):257–70.
22. Needham DM, Dinglas VD, Bienvenu OJ, Colantuoni E, Wozniak AW, Rice
TW, et al. One year outcomes in patients with acute lung injury randomised
to initial trophic or full enteral feeding: prospective follow-up of EDEN
randomised trial. BMJ. 2013;346:f1532.
23. Schwarz R, Krauss O, Hinz A. Fatigue in the general population. Onkologie.
2003;26(2):140–4.
24. Addington AM, Gallo JJ, Ford DE, Eaton WW. Epidemiology of unexplained
fatigue and major depression in the community: the Baltimore ECA follow-
up, 1981-1994. Psychol Med. 2001;31(6):1037–44.
25. Andrykowski MA, Donovan KA, Laronga C, Jacobsen PB. Prevalence,
predictors, and characteristics of off-treatment fatigue in breast cancer
survivors. Cancer. 2010;116(24):5740–8.
26. Horneber M, Fischer I, Dimeo F, Ruffer JU, Weis J. Cancer-related fatigue:
epidemiology, pathogenesis, diagnosis, and treatment. Dtsch Arztebl Int.
2012;109(9):161–71. quiz 172
27. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue
Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J
Psychosom Res. 1995;39(3):315–25.
28. Latronico N, Herridge M, Hopkins RO, Angus D, Hart N, Hermans G, et al.
The ICM research agenda on intensive care unit-acquired weakness.
Intensive Care Med. 2017;43(9):1270–81.
29. Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, et al. Evaluation of
delirium in critically ill patients: validation of the Confusion Assessment
Method for the Intensive Care Unit (CAM-ICU). Crit Care Med. 2001;29(7):
1370–9.
30. Klugkist M, Sedemund-Adib B, Schmidtke C, Schmucker P, Sievers HH,
Huppe M. Confusion Assessment Method for the Intensive Care Unit (CAM-
ICU): diagnosis of postoperative delirium in cardiac surgery. Anaesthesist.
2008;57(5):464–74.
31. Wintermann GB, Brunkhorst FM, Petrowski K, Strauss B, Oehmichen F, Pohl
M, et al. Stress disorders following prolonged critical illness in survivors of
severe sepsis. Crit Care Med. 2015;43(6):1213–22.
32. Kuhnt S, Ernst J, Singer S, Ruffer JU, Kortmann RD, Stolzenburg JU, et al.
Fatigue in cancer survivors—prevalence and correlates. Onkologie. 2009;
32(6):312–7.
33. Singer S, Kuhnt S, Zwerenz R, Eckert K, Hofmeister D, Dietz A, et al. Age-
and sex-standardised prevalence rates of fatigue in a large hospital-based
sample of cancer patients. Br J Cancer. 2011;105(3):445–51.
34. Wittchen H-U, Zaudig M, Fydrich T. Strukturiertes Klinisches Interview für
DSM-IV. Göttingen: Hogrefe; 1997.
35. Raphael B, Lundin T, Weisaeth L. A research method for the study of
psychological and psychiatric aspects of disaster. Acta Psychiatr Scand
Suppl. 1989;353:1–75.
36. Stoll C, Kapfhammer HP, Rothenhausler HB, Haller M, Briegel J, Schmidt M,
et al. Sensitivity and specificity of a screening test to document traumatic
experiences and to diagnose post-traumatic stress disorder in ARDS patients
after intensive care treatment. Intensive Care Med. 1999;25(7):697–704.
37. Thulesius H, Hakansson A. Screening for posttraumatic stress disorder
symptoms among Bosnian refugees. J Trauma Stress. 1999;12(1):167–74.
38. Zimet GD, Dahlem NW, Zimet SG, Farley GK. The multidimensional scale of
perceived social support. J Pers Assess. 1988;52(1):30–41.
39. Canty-Mitchell J, Zimet GD. Psychometric properties of the
multidimensional scale of perceived social support in urban adolescents.
Am J Community Psychol. 2000;28(3):391–400.
40. Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol
group. Ann Med. 2001;33(5):337–43.
41. Hinz A, Klaiberg A, Brahler E, Konig HH. The quality of life questionnaire EQ-
5D: modelling and norm values for the general population. Psychother
Psychosom Med Psychol. 2006;56(2):42–8.
42. Rollnik JD. The Early Rehabilitation Barthel Index (ERBI). Rehabilitation
(Stuttg). 2011;50(6):408–11.
43. SB O’S, Schmitz TJ. Physical rehabilitation. Philadelphia: F.A. Davis; 2007.
44. Myhren H, Ekeberg O, Toien K, Karlsson S, Stokland O. Posttraumatic stress,
anxiety and depression symptoms in patients during the first year post
intensive care unit discharge. Crit Care. 2010;14(1):R14.
45. Rosendahl J, Brunkhorst FM, Jaenichen D, Strauss B. Physical and mental
health in patients and spouses after intensive care of severe sepsis: a dyadic
perspective on long-term sequelae testing the actor-partner
interdependence model. Crit Care Med. 2013;41(1):69–75.
46. Watt T, Groenvold M, Bjorner JB, Noerholm V, Rasmussen NA, Bech P.
Fatigue in the Danish general population. Influence of sociodemographic
factors and disease. J Epidemiol Community Health. 2000;54(11):827–33.
47. Buchwald D, Pearlman T, Kith P, Schmaling K. Gender differences in patients
with chronic fatigue syndrome. J Gen Intern Med. 1994;9(7):397–401.
48. Lorusso L, Mikhaylova SV, Capelli E, Ferrari D, Ngonga GK, Ricevuti G.
Immunological aspects of chronic fatigue syndrome. Autoimmun Rev. 2009;
8(4):287–91.
49. Chlan LL, Savik K. Contributors to fatigue in patients receiving mechanical
ventilatory support: a descriptive correlational study. Intensive Crit Care
Nurs. 2015;31(5):303–8.
50. Casillas JM, Damak S, Chauvet-Gelinier JC, Deley G, Ornetti P. Fatigue in
patients with cardiovascular disease. Ann Readapt Med Phys. 2006;49(6):
309–19. 392-402
51. Bunevicius A, Stankus A, Brozaitiene J, Girdler SS, Bunevicius R. Relationship
of fatigue and exercise capacity with emotional and physical state in
patients with coronary artery disease admitted for rehabilitation program.
Am Heart J. 2011;162(2):310–6.
52. Zhou C, Wu L, Ni F, Ji W, Wu J, Zhang H. Critical illness polyneuropathy and
myopathy: a systematic review. Neural Regen Res. 2014;9(1):101–10.
53. Aaronson LS, Teel CS, Cassmeyer V, Neuberger GB, Pallikkathayil L, Pierce J,
et al. Defining and measuring fatigue. Image J Nurs Sch. 1999;31(1):45–50.
54. APA. Diagnostic and statistical manual of mental disorders (DSM-4®-TR). 4th
ed. Washington, DC: American Psychiatric Association; 2000.
55. Wade DM, Howell DC, Weinman JA, Hardy RJ, Mythen MG, Brewin CR, et al.
Investigating risk factors for psychological morbidity three months after
intensive care: a prospective cohort study. Crit Care. 2012;16(5):R192.
56. Silverman MN, Heim CM, Nater UM, Marques AH, Sternberg EM.
Neuroendocrine and immune contributors to fatigue. PM R. 2010;2(5):338–46.
57. Hopkins RO, Key CW, Suchyta MR, Weaver LK, Orme JF Jr. Risk factors for
depression and anxiety in survivors of acute respiratory distress syndrome.
Gen Hosp Psychiatry. 2010;32(2):147–55.
58. Jubran A, Lawm G, Duffner LA, Collins EG, Lanuza DM, Hoffman LA, et al.
Post-traumatic stress disorder after weaning from prolonged mechanical
ventilation. Intensive Care Med. 2010;36(12):2030–7.
Wintermann et al. Journal of Intensive Care (2018) 6:27 Page 12 of 12
