Effect of critical illness polyneuropathy on the withdrawal from

mechanical ventilation and the length of stay in septic patients*

Jose Garnacho-Montero, MD, PhD; Rosario Amaya-Villar, MD; Jose Luis García-Garmendía, MD, PhD;
Juan Madrazo-Osuna, MD; Carlos Ortiz-Leyba, MD, PhD

Objectives: No previous study has demonstrated whether crit-
ical illness polyneuropathy itself lengthens mechanical ventilation
or whether this prolonged duration of ventilatory support is ex-
plained by concomitant risk factors for weaning failure. Our
objectives were to evaluate the impact of critical illness polyneu-
ropathy on the length of mechanical ventilation after controlling
for coexisting risk factors for weaning failure and to assess the
impact of critical illness polyneuropathy on the length of the stay
in a cohort of septic patients.
Design: Prospective cohort study.
Setting: Intensive care unit of a tertiary hospital.
Patients: All patients with severe sepsis or septic shock who
required mechanical ventilation for >7 days who were considered
ready to discontinue mechanical ventilation.
Interventions: Patients underwent a neurophysiologic evalua-
tion at onset of weaning from mechanical ventilation.
Measurements and Main Results: Sixty-four critically ill septic
patients were enrolled, and 34 developed critical illness polyneu-
ropathy (53.1%; 95% confidence interval, 40.2– 65.7%). Length of
mechanical ventilation was significantly higher in patients who
had developed critical illness polyneuropathy (median 34 days vs.
14 days, p < .001). The duration of the weaning period was also
significantly greater in patients with critical illness polyneurop-
athy (median 15 days vs. 2 days, p < .001) even though factors
suspected to influence the weaning process did not differ be-
tween these two groups. Multiple logistic regression analysis
indicated that critical illness polyneuropathy was the only risk
factor independently associated with weaning failure (odds ratio,
15.4; 95% confidence interval, 4.55, 52.3; p < .001). Lengths of
intensive care unit and hospital stays were significantly higher in
patients with critical illness polyneuropathy.
Conclusions: In critically ill septic patients, critical illness
polyneuropathy significantly increases the duration of mechanical
ventilation and prolongs the lengths of intensive care unit and
hospital stays. (Crit Care Med 2005; 33:349 –354)
KEY WORDS: sepsis; critical illness polyneuropathy; weaning;
mechanical ventilation

C omplications that can delay
recovery from critical illness
are a major concern to critical
care physicians. Neuromuscu-
lar disturbances are now well recognized
as complications that can affect long-
term ventilated patients, and the number
of reports has risen markedly in the last
decades. Although diverse disorders can
afflict the peripheral nervous system in
these patients, critical illness polyneu-
ropathy (CIP) is the most clearly defined
neuromuscular complication in these pa-
tients (1).
*See also p. 452.
From the Intensive Care Unit (JG-M, RA-V, CO-L)
and Department of Clinical Neurophysiology (CO-L),
Hospital Universitario Virgen del Rocío, Sevilla. Spain;
and the Intensive Care Unit (JLG-G), Hospital San Juan
de Dios del Aljarafe, Sevilla, Spain.
Supported, in part, by grant 154/97 from Conse-
jería de Salud de la Junta de Andalucía.
Copyright © 2005 by the Society of Critical Care
Medicine and Lippincott Williams & Wilkins
DOI: 10.1097/01.CCM.0000153521.41848.7E
Commonly, CIP is manifest by a fail-
ure to wean from mechanical ventilation.
In fact, it was originally described in pa-
tients with difficult weaning from venti-
latory support (2, 3). Various studies have
attempted to determine whether CIP pro-
longs the duration of mechanical ventila-
tion, and conflicting data have been ob-
tained. We and others found that the
duration of mechanical ventilation was
significantly longer in patients suffering
from CIP compared with those who did
not (4 –7). In contrast, other studies
found that CIP did not prolong the dura-
tion of mechanical ventilation or the
length of the weaning time (8, 9). Never-
theless, to the best of our knowledge,
none of these previous studies have dem-
onstrated that CIP itself prolongs me-
chanical ventilation or whether this pro-
longation of mechanical ventilation is the
effect of concurrent risk factors for wean-
ing failure.
Until now, little information has been
available concerning the impact of CIP on
in-hospital mortality and the length of
hospital stay, especially if patients are
evaluated only after improvement from
the acute illness and initiation of wean-
ing. In this subgroup of patients we can
certainly assess the clinical consequences
of this neurologic disorder.
We performed a prospective clinical
study enrolling patients with severe sep-
sis or septic shock who required mechan-
ical ventilation for ⱖ7 days who were
considered ready to discontinue mechan-
ical ventilation. Our primary objective
was to evaluate the impact of CIP on the
length of mechanical ventilation after
controlling for coexisting risk factors for
weaning failure. Our secondary objectives
were to assess the impact of CIP on the
length of the stay as well as to determine
the costs associated with this neurologic
complication.
MATERIALS AND METHODS
Hospital. This is a prospective study car-
ried out in the intensive care unit (ICU) of the
Hospital Virgen del Rocio from July 1, 1999, to
December 31, 2002. This is a medical-surgical

Crit Care Med 2005 Vol. 33, No. 2 349

unit in a large university hospital. Trauma and
neurosurgical patients are admitted to an-
other unit, and these patients were not in-
cluded in this protocol. Written consent was
obtained from patients’ relatives, and the Eth-
ical Committee of the Hospital Virgen del Ro-
cio approved this study.
Patients. All patients with severe sepsis or
septic shock following American College of
Chest Physicians/Society of Critical Care Med-
icine criteria (10) who required mechanical
ventilation for ⱖ7 days were evaluated. A pa-
tient was included in this study when the
attending physician considered that the pa-
tient was stabilized and ready to be weaned
from the ventilator. The patients included in
this protocol were followed up until death or
hospital discharge.
Excluded were patients ⬍18 yrs old; preg-
nant patients; patients intubated for exacerba-
tion of chronic obstructive pulmonary disease;
patients with a history of neuromuscular dis-
ease, liver cirrhosis, and end-stage renal dis-
ease; and patients infected with human immu-
nodeficiency virus. Routinely, the patient or,
most frequently, close relatives were asked
about signs and symptoms of preexisting neu-
romuscular disease, and the patient was ex-
cluded from the study if previous symptoms
were reported.
All patients received standard supportive
treatment including surgical treatment of the
focus of infection if necessary, fluid resuscita-
tion, vasoactive drugs, antimicrobial therapy,
and nutritional support. The sedation protocol
included intravenous midazolam (Dormicum,
Roche, Madrid, Spain) plus morphine both in
continuous drips. Muscle relaxant use was left
to the discretion of the physician in charge of
the patient.
Study Design. Weaning from the ventila-
tory support was initiated when the attending
physician considered that the patient was sta-
bilized and ready to be weaned from the ven-
tilator. Daily the attending physician assessed
the patient’s readiness for liberation from me-
chanical ventilation.
General criteria to be enrolled in this study
were adequate gas exchange (arterial oxygen
saturation ⱖ90% for an inspired oxygen frac-
tion ⱕ0.4, with a positive end-expiratory pres-
sure ⱕ5 cm H2O), body temperature ⬍38°C,
hemoglobin level ⬎9 g/dL, and ability to fol-
low simple commands. Patients were sitting in
their beds at 45° from the horizontal, and
weaning was performed by a 2-hr trial of spon-
taneous (T-piece) breathing following the pro-
tocol described by Esteban et al. (11). If extu-
bation was not achieved with this method, the
weaning could be continued either with T-
piece trials or with pressure-support ventila-
tion: progressive reduction of pressure sup-
port in steps of 2 cm of water every 2 hrs until
a pressure of 5 cm of water was reached. Two
different ventilators were used throughout the
study period: Servo Ventilator 300 (Siemens,
Madrid, Spain) and EVITA2 dura (Dräger, Ma-
drid, Spain). When discontinuation from me-
350
chanical ventilation was not possible, a percu-
taneous tracheostomy was performed (12).
The duration of mechanical ventilation
was recorded. The day of the beginning of
mechanical ventilation, the date of the onset
of weaning from mechanical ventilation, the
date of disconnection from the ventilator, and
the need of reintubation were recorded. The
reason for reintubation was also noted (13).
Extubation or spontaneous ventilation
through a tracheostomy marked the end of
mechanical ventilation. The weaning failure
group consisted of all patients in whom, after
trials of extubation, a tracheostomy was per-
formed because of the impossibility of discon-
tinuation from mechanical ventilation and
those patients in whom extubation failed and
who required reintubation (14).
Demographic variables of all patients en-
rolled in this study were recorded: age, gender,
and dates of admission and discharge from the
ICU and the hospital. Severity of illness was
evaluated by the Acute Physiology and
Chronic Health Evaluation (APACHE) II score
(15) considering the worst data point of the
first 24 hrs in the ICU. At the time of admis-
sion, severity of organ failure was evaluated by
the Sequential Organ Failure Assessment
(SOFA) score (16). Both scores were also re-
corded at the onset of weaning.
Detailed information of total doses of sed-
ative drugs administered since admission to
the ICU was recorded. Intravenous midazolam
was interrupted when the patient’s clinical
condition had improved and weaning was con-
sidered indicated. The following conditions
that were judged to be clinically valuable as
factors that could hinder the weaning from
mechanical ventilation were also noted: previ-
ous history of chronic obstructive pulmonary
disease as defined by the APACHE II score
(15), active smoker (17), cardiac insufficiency
(as defined by the APACHE II score), morbid
obesity (body mass index ⬎40), and under-
weight (body mass index ⬍20) (18). Body mass
index was defined as the ratio of weight in
kilograms divided by the square of height in
meters.
The occurrence of ventilator-associated
pneumonia was also recorded. Ventilator-
associated pneumonia diagnosis required ra-
diographic appearance of a new and persistent
pulmonary infiltrate and at least two of the
following criteria: temperature ⬎38°C or
⬍35.5°C, leukocytosis ⬎12,000 cells/mm3 or
leukopenia ⬍4,000 cells/mm3, and purulent
bronchial secretions. In these patients, a bron-
choscopy with protected brush or a quantita-
tive tracheal aspirate was carried out to obtain
microbiological documentation (19).
Neurophysiologic Studies. A neurophysio-
logic study (electroneurogram and needle
electromyography) was performed at onset of
weaning from mechanical ventilation. The
same investigator (JMO), who was unaware of
the patient’s medical condition, carried out all
these studies.
The protocol of study has been described
elsewhere (7). Briefly, the study included mea-
surement of motor nerve conduction (median,
peroneal, and tibial nerves); calculation of
conduction velocities, amplitude, and shape of
the compound muscle action potential; mea-
surement of distal latencies; and repetitive
nerve stimulation. In addition, sensory nerve
action potential and velocity conduction of the
sural nerve and the sensory median nerve
were also recorded. Signs of denervation, such
as fibrillation potentials and positive waves in
muscles, were also sought in four muscles
(deltoid, quadriceps femoris, first dorsal inter-
osseus, and tibialis anterior muscle) using
needle electromyography.
CIP was diagnosed when signs of acute
axonal injuries were present: reduced com-
pound muscle action potential and sensory
nerve action potential amplitudes, preserva-
tion of the speed of impulse conduction and
the distal latencies, and signs of denervation
in at least one of the explored muscles (20).
Statistical Analysis. Comparisons between
patients with CIP and without CIP were ac-
complished using unpaired Student’s t-test for
parametric continuous variables after correc-
tion for equality of variance (Levene’s test) and
Mann-Whitney U test for nonparametric con-
tinuous variables. Categorical variables were
analyzed with Pearson’s chi-square test, ex-
cept where small samples required the use of
Fisher’s exact test. The ␣ level was set at .05.
All p values were two-sided.
The Kaplan-Meier method was used to
compare the duration of weaning period and
the length of hospital stay between patients
with and without CIP; comparisons between
groups were made using log-rank tests.
Univariate analysis comparing weaning
failure group with successful weaning group
was performed as previously indicated. A mul-
tivariate analysis using logistic regression
analysis was used to evaluate independent risk
factors for weaning failure. The variables in-
cluded in the multivariate analysis were those
with significance levels of p ⬍ .10 in univari-
ate analysis. The model was constructed using
a forward stepwise method with the likelihood
ratio test. The presence of collinearity in the
model was evaluated using the tolerance and
variance inflation factor analysis followed by
testing the condition indexes. No condition
index ⬎15 was found, so the variance propor-
tions were not examined. The odds ratio and
corresponding 95% confidence intervals were
also calculated (21).
RESULTS
Description of Patients. During the
study period, 139 patients fulfilled inclu-
sion criteria and underwent mechanical
ventilation for ⱖ7 days. Sixty-nine pa-
tients (44 males and 25 females) died
before discontinuation from mechanical
ventilation could be started. The mean
Crit Care Med 2005 Vol. 33, No. 2
APACHE II score on admission of these dian, 22 hrs; range, 1, 96) whereas only cation: 27 of 34 (79.4%) vs. six of 30
69 patients was 22.2 (6.5). Therefore, six- four (13.3%) of the patients without CIP (20%, p ⬍ .0001).
ty-eight patients were assessed although required reintubation (median, 24 hrs; ICU mortality was not statistically dif-
only 64 were finally analyzed (39 males range, 5, 46) in the same period (p ⬍ .05). ferent (seven of 34 [20.6%] patients with
and 25 females): One patient was prema- Reasons for reintubation in these 18 pa- CIP vs. three of 30 [10%] patients with-
turely transferred to another hospital, tients are listed in Table 2. Tracheostomy out CIP; p ⫽ .31), whereas in-hospital
and in three patients the weaning process was performed in 21 patients (61.8%) mortality was significantly higher in pa-
was abandoned because of a complication with CIP and only in four (13.3%) of the tients with CIP (16 of 34 [47.1%)] vs. six
not related to disconnection from me- patients without CIP (p ⬍ .05). Weaning of 30 [20%]; p ⫽ 0.03).
chanical ventilation (suture leakage in failure was significantly more frequent in Lengths of ICU and hospital stays were
two patients and a pancreatic abscess that
patients who developed CIP than in those significantly higher in patients with CIP
required surgical drainage in one pa-
subjects who did not present this compli- (Table 3). Duration of hospitalization af-
tient).
All patients presented severe sepsis or
septic shock. The primary diagnoses were
abdominal sepsis (n ⫽ 30), pneumonia (n Table 1. Univariate analysis comparing patients with and without critical illness polyneuropathy (CIP)
⫽ 21), urosepsis (n ⫽ 5), sepsis of un-
known origin (n ⫽ 5), and others (n ⫽ 3). CIP (n ⫽ 34) No CIP (n ⫽ 30) p
The mean APACHE II score on admission
Age, yrsa 61.1 (14.9) 61.5 (13.1) .9
was 18.6 (6.5) and the mean SOFA score APACHE II at admissiona 19.3 (6.9) 17.7 (5.9) .3
of the first 24 hrs in the unit was 7.2 (4). SOFA at admissiona 7.2 (4.3) 7.2 (3.5) 1
Comparison of CIP and Control Pa- APACHE II (onset of weaning)a 12.5 (4.6) 12.5 (4.4) 1
tients. CIP occurred in 34 (53.1%; 95% SOFA (onset of weaning)a 4 (2.5) 4 (2.4) 1
Total dose of midazolama,b 4306 (7267) 1728 (2589) .06
confidence interval, 40.2– 65.7%) of the Total dose of morphinea,b 616 (1114) 182 (141) .03
prospectively monitored critically ill pa- Muscle relaxantsa,b 10 (29.4)c 3 (10)d .055
tients. Patients with CIP exhibited typical COPDe 3 (8.8) 6 (20) .3
signs of an acute sensorimotor axonal Active smokere 9 (26.5) 11 (36.7) .4
neuropathy. Cardiac insufficiencye 1 (2.9) 2 (6.7) .6
Morbid obesitye 5 (14.7) 2 (6.7) .4
The mean time (SD) elapsed between Underweighte 2 (5.9) 1 (3.3) 1
the onset of mechanical ventilation and VAPe 17 (50) 8 (26.7) .056
the beginning of weaning was 16.4 (8.1)
days in patients with CIP and 11.3 (4.5) APACHE, Acute Physiology and Chronic Health Evaluation; SOFA, Sequential Organ Failure
days in patients without CIP (p ⬍ .05). Assessment; COPD, chronic obstructive pulmonary disease; VAP, ventilator-associated pneumonia.
a
Patients in both groups had similar age, Results are expressed as mean (SD); bdoses in milligrams; cseven patients received vecuronium,
two patients received vecuronium and atracurium, and one patient received exclusively atracurium;
APACHE II score, and SOFA score in the d
these three patients received vecuronium; eresults are expressed as n (%).
first 24 hrs. The total dose of midazolam
received did not differ between the two
groups, but the amount of morphine was
significantly higher in patients with CIP
(p ⬍ .05). The other conditions evaluated
as factors that may impede weaning did
not differ between patients with or with-
out CIP (Table 1). The method of weaning
was a 2-hr trial of spontaneous (T-piece)
breathing in 56 patients and the gradual
reduction of pressure support in eight
patients (four with CIP and four without
CIP).
Length of mechanical ventilation was
significantly higher in patients who de-
veloped CIP: median 34 (range 12–99)
days vs. 14 (7– 44) days (p ⬍ .001). The
duration of the weaning period was also
significantly greater in patients with CIP:
median 15 (1–74) days vs. 2 (0 –29) days
(p ⬍ 0.001). Figure 1 shows that the
comparison of the length of weaning pe-
riod by Kaplan-Meier method was statis- Figure 1. Kaplan-Meier curves of probability of remaining under mechanical ventilation after the
tically significant (log-rank test, p ⬍ initiation of weaning for patients with critical illness polyneuropathy (solid lines) and without critical
.0001). illness polyneuropathy (dashed lines). The probability of weaning failure was significantly higher for
Of the 34 patients who developed CIP, patients with critical illness polyneuropathy. Log -rank test, p ⬍ .0001. Time denotes the period from
14 (41.2%) required reintubation (me- the onset of weaning process to final withdrawal of mechanical ventilation.

Crit Care Med 2005 Vol. 33, No. 2 351

Table 2. Reasons for reintubation in 18 patientsa
Hypoxemia
Increased respiratory work
Impaired clearance of secretions
Decreased consciousness
Atelectasis
Respiratory acidosis
Upper airway obstruction
Cardiac failure
CIP, critical illness polyneuropathy.
a
More than one cause could be diagnosed in a single patient.
Table 3. Comparisons of the lengths of stays of patients with and without critical illness polyneurop-
athy (CIP)
DOH before admission to ICU
DOH before the onset of weaning
DOH after the onset of weaning
Length of ICU stay
Length of hospital stay
DOH, duration of hospitalization; ICU, intensive care unit.
ter the onset of weaning was also signif-
icantly higher in patients with CIP than
in those without CIP. Figure 2 shows that
the length of hospital stay was statisti-
cally different in patients with and with-
out CIP using the Kaplan-Meier analysis
(log-rank test, p ⬍ .0001).
Comparison of Failed vs. Successful
Weaning Groups. The failed weaning
group consisted of 33 patients: 18 patients
who required reintubation (eight of them
were finally extubated and in ten patients
the tracheostomy was necessary because
successful extubation was not achieved)
and 15 patients in whom the extubation
was not possible and tracheostomy was in-
dicated because of prolonged mechanical
ventilation and impossibility of liberation
from the ventilator. Twelve of these pa-
tients (eight with CIP and four without
CIP) were discharged from the ICU with the
tracheostomy tube in place. This failed
weaning group was compared with the rest
of the patients (n ⫽ 31) who were suc-
cessfully weaned from the ventilator. Uni-
variate analysis is presented in Table 4.
Using multiple logistic regression analy-
sis, CIP was the only risk factor indepen-
dently associated with weaning failure
(odds ratio, 15.4; 95% confidence inter-
val, 4.55–52.3; p ⬍ .001).
DISCUSSION
In this prospective study, we establish
that CIP increases the duration of me-
352
CIP (n ⫽ 14) No CIP (n ⫽ 4)
8 3
7 1
6 1
3 1
2 0
2 1
0 1
1 0
CIP (n ⫽ 34) No CIP (n ⫽ 30) p
1.5 (0–73) 1 (0–60) .09
15.5 (7–42) 12 (7–29) ⬍.01
56 (6–342) 18.5 (8–60) ⬍.0001
46.5 (8–134) 22.5 (10–60) ⬍.0001
85 (20–370) 33 (14–106) ⬍.0001
chanical ventilation and, after controlling
for other confounding variables, is an in-
dependent cause of weaning failure. Our
results also provide compelling data
about the increase in the duration of hos-
pital stay caused by this neurologic com-
plication.
Discontinuation from mechanical
ventilation begins when the precipitating
cause is at least partially reversed. Never-
theless, this process can fail in patients
meeting weaning criteria, prolonging the
time of intubation and ventilatory sup-
port. Diverse underlying conditions or
complications occurring during the pe-
riod of mechanical ventilation can lead to
this weaning failure.
CIP is a well-recognized cause of
weaning failure. Most investigators have
concluded that the length of mechanical
ventilation was significantly higher in pa-
tients with CIP than in patients who did
not develop this complication although
the duration of weaning time was not
assessed (4, 6, 7). Two studies specifically
evaluated the length of weaning period in
patients with CIP obtaining conflicting
results (5, 8). Likewise, the severity of
conduction abnormalities did not corre-
late with the length of mechanical venti-
lation or the duration of stay in the ICU
in 62 patients with the diagnosis of CIP
(22).
Moreover, many of the patients in-
cluded in these studies had not recovered
O
ur study con-
firms that criti-
cal illness poly-
neuropathy significantly
prolongs the duration of me-
chanical ventilation and is
an independent risk factor
for weaning failure in criti-
cally ill septic patients.
from the acute illness and were not ready
to be weaned off the ventilatory support.
A recent study attempted to solve this
question by evaluating exclusively pa-
tients with a satisfactory level of con-
sciousness. The authors concluded that
mechanical ventilation was more pro-
longed in those patients who developed
sensorimotor axonal neuropathy com-
pared with those who did not present this
complication (23).
None of these studies was able to ad-
just for risk factors of weaning failure or
severity of illness. It is well recognized
that the characteristics of patients and
the severity of illness at admission to the
ICU are determinants of the duration of
mechanical ventilation (24).
Our results confirm that CIP signifi-
cantly prolongs the duration of mechan-
ical ventilation and weaning time. In our
series, the duration of ventilatory support
before the onset of weaning was signifi-
cantly longer in patients with the diagno-
sis of CIP than in those subjects who did
not present this complication, which is in
agreement with other studies (3, 23).
However, given that the weaning period
was also significantly prolonged in pa-
tients with CIP, the longer duration of
ventilatory support observed in CIP pa-
tients can be attributed to this neurologic
abnormality and not to the period elapsed
between the onset of mechanical ventila-
tion and the beginning of weaning.
The occurrence of reintubation in the
group of patients who did not present CIP
(13.5%) is similar to the rate reported
previously (14). This rate was signifi-
cantly greater in patients with CIP, which
may influence the poor outcome of these
patients. Reintubation has been indepen-
dently associated with mortality during
Crit Care Med 2005 Vol. 33, No. 2
 discharged alive from the ICU. This find-
ing is of extraordinary importance in view
of the economic burden that it may rep-
resent. Moreover, we cannot overlook
that symptoms due to CIP can persist for
a considerable period of time, and there-
fore these patients may require medical
care for years (4, 27, 28).
Our study has several limitations.
First, only patients with severe sepsis and
septic shock were included in this study.
Consequently, it is unknown whether our
findings can be generalized to other crit-
ically ill patients. Second, we did not
measure lung function variables, which
may have influenced the weaning pro-
cess. Third, we did not assess the associ-
ation between the prolongation of me-
chanical ventilation and the severity of
CIP. Despite all these limitations, our re-
sults shed light on the precise contribu-
Figure 2. Kaplan-Meier curves comparing the length of hospital stay in patients with critical illness tion of CIP to delayed weaning from me-
polyneuropathy (solid lines) and without critical illness polyneuropathy (dashed lines). The length of chanical ventilation in critically ill septic
hospital stay in the whole group was significantly longer in patients with critical illness polyneurop- patients.
athy than in patients who did not develop this neurologic complication. Log-rank test, p ⬍ .0001.
CONCLUSIONS
Table 4. Comparisons of patients who failed weaning with patients successfully extubated
Our study confirms that CIP signifi-
Failed Weaning Successful Weaning cantly prolongs the duration of mechan-
(n ⫽ 33) (n ⫽ 31) p ical ventilation and is an independent risk
factor for weaning failure in critically ill
Age, yrsa 62.1 (13.8) 60.4 (14.3) .6
APACHE II at admissiona 19.5 (6.9) 17.7 (6) .3 septic patients. These findings suggest
SOFA at admissiona 7.2 (3.9) 7.1 (4.1) .9 that CIP should be taken into account in
APACHE II (onset of weaning)a 12.5 (4.2) 12.7 (4.7) .9 clinical trials comparing different wean-
SOFA (onset of weaning)a 3.7 (2.3) 4.3 (2.5) .4 ing modes or in studies evaluating com-
Total dose of midazolama,b 3106 (6267) 2128 (3081) .4
Total dose of morphinea,b 610 (1136) 203 (148) .09
plications of mechanical ventilation in
Muscle relaxantsa,b 8 (24.2)c 5 (16.1)d .42 septic patients. CIP also accounts for ex-
CIPe 27 (81.8) 7 (22.6) ⬍.0001 tended ICU and hospital stays. Further
COPDe 5 (15.2) 4 (12.9) .8 studies should focus on specific interven-
Active smokere 11 (33.3) 9 (29) 7 tions (e.g., weaning protocols or rehabil-
Cardiac insufficiencye 1 (3) 2 (6.5) 5
Morbid obesitye 4 (12.1) 3 (9.7) 7 itation programs) that could help to im-
Underweighte 2 (6.1) 1 (3.2) 1 prove the poor prognosis of these
VAPe 18 (54.5) 7 (22.6) .009 patients.
APACHE, Acute Physiology and Chronic Health Evaluation; SOFA, Sequential Organ Failure
Assessment; CIP, critical illness polyneuropathy; COPD, chronic obstructive pulmonary disease; VAP, REFERENCES
ventilator-associated pneumonia. 1. Garnacho-Montero J, Madrazo-Osuna J,
a
Results are expressed as mean (SD); bdoses in milligrams; cfive patients received vecuronium, two García Garmendia JL, et al: Neuromuscular
patients received vecuronium and atracurium, and one patient received exclusively atracurium; dthese disorders of the critically ill patient. Clin
five patients received vecuronium; eresults are expressed as n (%). Pulm Med 2001; 8:354 –359
2. Bolton CF, Gilbert JJ, Hahn AF, et al: Poly-
neuropathy in critically ill patient. J Neurol
hospitalization (25). However, we consid- Moreover, after adjustment for con- Neurosurg Psych 1984; 47:1223–1231
ered reintubation in the ICU, whereas the founding variables, CIP was the only inde- 3. Witt NJ, Zochodne DW, Bolton CF, et al:
majority of the previous studies only con- pendent risk factor for weaning failure. In Peripheral nerve function in sepsis and mul-
sidered reintubation in the following 48 this cohort of critically ill septic patients, tiple organ failure. Chest 1991; 99:176 –184
4. Leijten FSS, Harinck-De Weerd JE, Poort-
or 72 hrs. Our rationalization was that CIP increases ⬎15 times the risk of wean-
vliet DCJ, et al: The role of polyneuropathy in
CIP persists for a large period of time, ing failure. This finding may also explain
motor convalescence after prolonged me-
and CIP has even been described as the the poor outcome of patients with CIP, be- chanical ventilation. JAMA 1995; 274:
cause of ICU readmission because of cause weaning failure has been previously as- 1221–1225
acute neuromuscular respiratory failure sociated with an increased risk of death (14). 5. Druschky A, Herkert M, Radespiel-Tröger M,
after ICU discharge in patients success- CIP significantly lengthens the dura- et al: Critical illness polyneuropathy: clinical
fully weaned from ventilation (26). tion of hospital stay in those patients findings and cell culture assay of neurotox-

Crit Care Med 2005 Vol. 33, No. 2 353

 icity assessed by a prospective study. Inten-
sive Care Med 2001; 27:686 – 693
6. Thiele RI, Jakob H, Hund E, et al: Sepsis and
catecholamine support are the major risk
factors for critical illness polyneuropathy af-
ter open heart surgery. Thorac Cardiovasc
Surg 2000; 48:145–150
7. Garnacho-Montero J, Madrazo-Osuna J,
García Garmendia JL, et al: Critical illness
polyneuropathy: Risk factors and clinical
consequences. A cohort study in septic pa-
tients. Intensive Care Med 2001; 27:
1288 –1296
8. Leijten FSS, De Weerd AW, Poortvliet DCJ, et
al: Critical illness polyneuropathy in multi-
ple organ dysfunction syndrome and wean-
ing from the ventilator. Intensive Care Med
1996; 22:856 – 861
9. Sander HW, Saadeh PB, Chandswang N, et al:
Diaphragmatic denervation in intensive care
unit patients. Electromyogr Clin Neuro-
physiol 1999; 39:3–5
10. American College of Chest Physicians/
Society of Critical Care Medicine Consensus
Committee: Definition for sepsis and organ
failures and guidelines for the use of innova-
tive therapies in sepsis. Chest 1992; 101:
1658 –1662
11. Esteban A, Frutos F, Tobin MJ, et al: A com-
parison of four methods of weaning patients
from mechanical ventilation. N Engl J Med
1995; 332:345–350
12. Kollef MH, Aherns TA, Shanon W: Clinical
354
predictions and outcomes for patients re-
quiring tracheostomy. Crit Care Med 1999;
27:1714 –1720
13. Esteban A, Alia I, Tobin MJ, et al: Effect of
spontaneous breathing trial duration on out-
come of attempts to discontinue mechanical
ventilation. Am J Respir Crit Care Med 1999;
159:512–518
14. Epstein SK, Ciubotaru RL, Wong JB: Effect
of failed extubation on the outcome of me-
chanical ventilation. Chest 1997; 112:
186 –192
15. Knaus WA, Draper EA, Wagner DP, et al:
APACHE II, a severity of disease classification
system. Crit Care Med 1985; 13:818 – 829
16. Vincent JL, Moreno R, Takala J, et al: The
SOFA (Sepsis-related Organ Failure Assess-
ment) score to describe organ dysfunction/
failure. Intensive Care Med 1996; 22:707–710
17. Pittet D, Thiévent B, Wenzel RP, et al: Im-
portance of pre-existing co-morbidities for
prognosis of septicemia in critically ill pa-
tients. Intensive Care Med 1993; 19:265–272
18. Willet WC, Dietz WH, Colditz GA: Guidelines
for healthy weight. N Engl J Med 1999; 341:
427– 434
19. Meduri GU: Diagnosis and differential diag-
nosis of ventilator-associated pneumonia.
Clin Chest Med 1995; 16:61–93
20. Bolton CF: Sepsis and the systemic inflam-
matory response syndrome: Neuromuscular
manifestations. Crit Care Med 1996; 24:
1408 –1416
21. Hosmer DW, Lemeshow SA. 1989. Applied
logistic regression. New York: John Wiley.
22. Zifko UA, Zipko HT, Bolton CF: Clinical and
electrophysiological findings in critical ill-
ness polyneuropathy. J Neurol Sci 1998; 159:
186 –193
23. De Jonghe B, Sharshar T, Lefaucheur JP, et
al: Paresis acquired in the intensive care
unit. A prospective multicenter study. JAMA
2002; 288:2859 –2867
24. Seneff MG, Zimmerman JE, Knaus WA, et al:
Predicting the duration of mechanical venti-
lation. The importance of disease and pa-
tients characteristics. Chest 1996; 110:
469 – 479
25. Epstein SK, Vuong V: Lack of influence of
gender on outcomes of mechanically venti-
lated medical ICI patients. Chest 1999; 116:
732–739
26. Latronico N, Guarneri B, Alongi S, et al:
Acute neuromuscular respiratory failure af-
ter ICU discharge. Intensive Care Med 1999;
25:1302–1306
27. De Seze M, Petit H, Wiart L, et al: Critical
illness polyneuropathy. A 2-year follow-up
study in 19 severe cases. Eur Neurol 2000;
43:61– 69
28. Fletcher SN, Kennedy DD, Ghosh IR, et al:
Persistent neuromuscular and neurophysio-
logic abnormalities in long-term survivors of
prolonged critical illness. Crit Care Med
2003; 31:1012–1106
Crit Care Med 2005 Vol. 33, No. 2