The final is that you critique “A Comparison of Four Methods of Weaning Patients from Mechanical Ventilation” article. You can use the information that you have already collected from the weekly assignments.
Copyright, 1995, by the Massachusetts Medical Society
Volume 332
FEBRUARY 9, 1995
Number 6
A COMPARISON OF FOUR METHODS OF WEANING PATIENTS FROM MECHANICAL
VENTILATION
ANDRÉS ESTEBAN, M.D., PH.D., FERNANDO FRUTOS, M.D., MARTIN J. TOBIN, M.D., INMACULADA ALÍA, M.D.,
JOSÉ F. SOLSONA, M.D., INMACULADA VALVERDÚ, M.D., RAFAEL FERNÁNDEZ, M.D.,
MIGUEL A. DE LA CAL, M.D., SALVADOR BENITO, M.D., PH.D., ROSER TOMÁS, M.D.,
DEMETRIO CARRIEDO, M.D., SANTIAGO MACÍAS, M.D., AND JESÚS BLANCO, M.D.,
FOR THE S PANISH L UNG F AILURE C OLLABORATIVE G ROUP *
Abstract Background. Weaning patients from mechanical ventilation is an important problem in intensive care
units. Weaning is usually conducted in an empirical manner, and a standardized approach has not been developed.
Methods. We carried out a prospective, randomized,
multicenter study involving 546 patients who had received
mechanical ventilation for a mean (SD) of 7.56.1 days
and who were considered by their physicians to be ready
for weaning. One hundred thirty patients had respiratory
distress during a two-hour trial of spontaneous breathing.
These patients were randomly assigned to undergo one
of four weaning techniques: intermittent mandatory ventilation, in which the ventilator rate was initially set at a
mean (SD) of 10.02.2 breaths per minute and then
decreased, if possible, at least twice a day, usually by 2 to
4 breaths per minute (29 patients); pressure-support ventilation, in which pressure support was initially set at
18.06.1 cm of water and then reduced, if possible, by
2 to 4 cm of water at least twice a day (37 patients);
intermittent trials of spontaneous breathing, conducted
two or more times a day if possible (33 patients); or a
once-daily trial of spontaneous breathing (31 patients).
Standardized protocols were followed for each technique.
Results. The median duration of weaning was 5 days
for intermittent mandatory ventilation (first quartile, 3 days;
third quartile, 11 days), 4 days for pressure-support ventilation (2 and 12 days, respectively), 3 days for intermittent (multiple) trials of spontaneous breathing (2 and
6 days, respectively), and 3 days for a once-daily trial of
spontaneous breathing (1 and 6 days, respectively). After
adjustment for other covariates, the rate of successful
weaning was higher with a once-daily trial of spontaneous
breathing than with intermittent mandatory ventilation
(rate ratio, 2.83; 95 percent confidence interval, 1.36 to
5.89; P0.006) or pressure-support ventilation (rate ratio,
2.05; 95 percent confidence interval, 1.04 to 4.04;
P0.04). There was no significant difference in the rate
of success between once-daily trials and multiple trials of
spontaneous breathing.
Conclusions. A once-daily trial of spontaneous breathing led to extubation about three times more quickly than
intermittent mandatory ventilation and about twice as
quickly as pressure-support ventilation. Multiple daily trials of spontaneous breathing were equally successful.
(N Engl J Med 1995;332:345-50.)
A
weaning, it is surprising that the process continues to
be managed empirically and that a standardized approach has not been developed.
Weaning techniques differ considerably from one another.3 Traditionally, intermittent trials of spontaneous
breathing, conducted one or more times a day, have
been used. Intermittent mandatory ventilation was
introduced amid claims that it was superior to the traditional weaning approach. It allows the patient to
breathe spontaneously between ventilator-delivered
breaths4; thus, weaning can be considered to begin with
the institution of mechanical ventilation. In the 1980s,
pressure-support ventilation became available5; it provides a titratable pressure boost to every inspiratory effort, and weaning is accomplished by gradually decreasing the level of the pressure boost.
Efficacy studies of weaning techniques can be faulted
for having a retrospective design, inappropriate study
LTHOUGH often lifesaving, mechanical ventilation
causes numerous life-threatening complications,1
making it important to discontinue ventilator support
at the earliest possible time. More than 40 percent of
the time that a patient receives mechanical ventilation
is spent trying to wean the patient from the ventilator.2
Considering the proportion of staff time devoted to
From the Hospital Universitario de Getafe, Madrid (A.E., F.F., I.A., M.A.C.);
Loyola University, Chicago, and Hines Veterans Affairs Hospital, Hines, Ill.
(M.J.T.); Hospital del Mar (J.F.S.) and Hospital Santa Creu i Sant Pau (I.V., S.B.),
Barcelona, Spain; Hospital Parc Tauli, Sabadell, Spain (R.F.); Hospital Germans
Trias i Pujol, Badalona, Spain (R.T.); Complejo Hospitalario de León, León,
Spain (D.C.); Hospital General de Segovia, Segovia, Spain (S.M.); and Hospital
del Río Ortega, Valladolid, Spain (J.B.). Address reprint requests to Dr. Esteban
at the Servicio de Cuidados Intensivos, Hospital Universitario de Getafe, Ctra. de
Toledo km 12’500, 28905 Getafe, Madrid, Spain.
Supported in part by a grant from the Veterans Affairs Research Service.
*The other members of the Spanish Lung Failure Collaborative Group are listed in the Appendix.
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
346
THE NEW ENGLAND JOURNAL OF MEDICINE
populations, and poorly standardized protocols; in addition, most were conducted before the use of pressuresupport ventilation became widespread.6-8 Accordingly,
we performed a prospective, randomized study involving patients who were deemed ready to discontinue mechanical ventilation. In a subgroup of patients who
were difficult to wean we compared the length of time
required for weaning with the use of four techniques:
intermittent mandatory ventilation, pressure-support
ventilation, intermittent trials of spontaneous breathing
conducted several times a day, and a once-daily trial of
spontaneous breathing.
METHODS
Patients
The study was conducted between October 1992 and October
1993 in the medical–surgical intensive care units of 14 teaching hospitals in Spain. The study population consisted of 546 patients (378
men and 168 women), with a mean (SD) age of 58.218.4 years.
All received mechanical ventilation for more than 24 hours because
of acute respiratory failure. The following underlying conditions were
present: chronic obstructive pulmonary disease with acute respiratory failure in 128 patients, acute lung injury in 319, neurologic or neuromuscular disorders in 85, and miscellaneous causes in 14. The
acute lung injury was a result of surgery in 74 patients, infection in
73, heart failure in 69, multiple trauma in 51, adult respiratory distress syndrome in 23, and other pulmonary causes in 29. On admission to the intensive care unit, the patients had a mean score of
18.77.0 on the Acute Physiology and Chronic Health Evaluation
(APACHE II) scale.9 Until the first attempt was made to discontinue
ventilator support, all patients received assist–control ventilation.
The patients received mechanical ventilation for a mean of 7.56.1
days before weaning was started. No hospital contributed more than
10 percent of the study population.
To be enrolled in the study the patients had to have an improvement in or resolution of the underlying cause of acute respiratory failure; adequate gas exchange, as indicated by a ratio of the partial
pressure of arterial oxygen (PaO2) to the fraction of inspired oxygen
(FiO2) above 200 with a positive end-expiratory pressure of 5 cm
of water; a core temperature below 38ºC; a hemoglobin level above
10 g per deciliter; and no further need for vasoactive and sedative
agents. In addition, the attending physician had to agree that the patient was in stable condition and ready to be weaned from the ventilator. Patients with a tracheostomy were excluded. The study was approved by the ethics committees of the hospitals, and the patients
provided informed consent.
Protocol
Feb. 9, 1995
extubation, the patients received supplemental oxygen by face mask.
If a patient had signs of poor tolerance at any time during the trial,
assist–control ventilation was reinstituted. For the purpose of the
study, these patients were designated as being difficult to wean from
mechanical ventilation.
Even if there were no signs of distress by the end of this trial, extubation could be postponed for a maximum of 24 hours if the primary physician thought that a patient might not be able to clear secretions or protect the airway against aspiration. Patients continued
to breathe spontaneously through the T-tube circuit. If they met criteria for poor tolerance, mechanical ventilation was reinstituted.
These patients were not included in the weaning-protocol group.
Patients who were designated as being difficult to wean from mechanical ventilation were stratified according to center and randomly
assigned with the use of a random-number table10 to be weaned in
one of four ways: intermittent mandatory ventilation, pressure-support ventilation, intermittent trials of spontaneous breathing, and a
once-daily trial of spontaneous breathing. The patients were assigned
to the groups in a blinded fashion with the use of opaque, sealed,
numbered envelopes, which were opened only when a patient did not
successfully complete the two-hour trial of spontaneous breathing.
All adjustments for each weaning technique were made by the primary physician.
Intermittent Mandatory Ventilation
In the group that received intermittent mandatory ventilation, the
ventilator rate was initially set at half the frequency used during assist–control ventilation; this initial rate was 10.02.2 breaths per
minute, and mechanical breaths were synchronized with inspiratory
effort. We attempted to decrease the ventilator rate, usually by two
to four breaths per minute, at least twice a day. The ventilator rate
was decreased more rapidly if tolerated by the patient, as reflected by
clinical assessment and blood gas monitoring. Patients who tolerated
a ventilator rate of five breaths per minute for two hours without
signs of distress were extubated. A continuous positive airway pressure of 5 cm of water was permitted.
Pressure-Support Ventilation
In the group that received pressure-support ventilation, pressure
was titrated to achieve a frequency of 25 breaths per minute. Pressure support was initially set at 18.06.1 cm of water, and we attempted to reduce this level of support by 2 to 4 cm of water at least
twice a day. The pace was increased if the patient did not have signs
of distress (the same criteria were applied as in the initial trial of
spontaneous breathing, except that a respiratory frequency of 25
breaths per minute was required). Patients who tolerated pressure
support at a setting of 5 cm of water for two hours with no apparent
ill effects were extubated. A continuous positive airway pressure of
5 cm of water was permitted.
Intermittent Trials of Spontaneous Breathing
After patients were enrolled in the study, assist–control ventilation
was stopped and the patients breathed spontaneously for three minutes through a T-tube circuit, with the FiO2 set at the same level
(0.380.05) as that used during mechanical ventilation. Tidal volume and respiratory frequency were measured with a spirometer during this period. Maximal inspiratory pressure was measured three
times in succession, and the most negative value was selected. Patients who met at least two of the following criteria underwent a trial
of spontaneous breathing lasting up to two hours: maximal inspiratory pressure below 20 cm of water, tidal volume above 5 ml per
kilogram of body weight, and a respiratory frequency of less than 35
breaths per minute. Weaning was considered to have begun with the
onset of this trial. During this trial, patients received humidified oxygen-enriched gas through a T-tube circuit. The primary physician
terminated the trial if a patient had any of the following signs of distress: a respiratory frequency of more than 35 breaths per minute, arterial oxygen saturation below 90 percent, heart rate above 140 beats
per minute or a sustained increase or decrease in the heart rate of
more than 20 percent, systolic blood pressure above 180 mm Hg or
below 90 mm Hg, agitation, diaphoresis, or anxiety. Patients who had
none of these features at the end of the trial were extubated. After
Patients assigned to intermittent trials of spontaneous breathing
were disconnected from the ventilator and allowed to breathe spontaneously through either a T-tube circuit or a continuous-flow circuit
designed to provide a continuous positive airway pressure of 5 cm
of water. The duration of the trials was gradually increased, and they
were attempted at least twice a day. Between the trials, assist–control ventilation was provided for at least one hour. Patients able to
breathe on their own for at least two hours without signs of distress
were extubated.
Once-Daily Trial of Spontaneous Breathing
Patients assigned to a once-daily trial of spontaneous breathing
were disconnected from the ventilator and allowed to breathe spontaneously through a T-tube circuit for up to two hours each day. If
signs of intolerance developed, assist–control ventilation was reinstituted for 24 hours, at which time another trial was attempted. Patients who tolerated a two-hour trial without signs of distress were
extubated.
For all four methods, weaning was considered to have failed if
reintubation was necessary within 48 hours after extubation or if
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Vol. 332
No. 6
COMPARISON OF METHODS OF WEANING PATIENTS FROM MECHANICAL VENTILATION
extubation was not possible after 14 days of
weaning. Weaning was considered successful
if extubation was achieved within the 14-day
period and reintubation was not required
within 48 hours of extubation.
Statistical Analysis
347
Table 1. Characteristics of the Study Population at Base Line.*
CHARACTERISTIC
INTERMITTENT
MANDATORY
VENTILATION
(N 29)
PRESSURESUPPORT
VENTILATION
(N 37)
INTERMITTENT
SPONTANEOUSBREATHING TRIALS
(N 33)
ONCE-DAILY
SPONTANEOUSBREATHING TRIAL
(N 31)
Age — yr
64.213.3
59.916.4
59.116.4
65.014.3
APACHE II score
20.87.0
18.97.6
20.16.8
18.36.6
The chi-square test was used to compare
Chronic obstructive pulmonary
8 (27.6)
18 (48.6)
12 (36.4)
14 (45.2)
categorical data, and the Kruskal–Wallis test
disease —no. (%)
was used to compare continuous variables
Acute lung injury — no. (%)
19 (65.5)
17 (45.9)
18 (54.5)
14 (45.2)
among the groups. The Kaplan–Meier methNeurologic disorder — no. (%)
2 (6.9)
2 (5.4)
3 (9.1)
3 (9.7)
od was used to determine the probability of
the success of a particular method of weaning
243.557.9
242.360.3
223.661.8
229.265.6
Ratio of PaO2 to FiO2
over time.11 The relative probability of sucMaximal inspiratory pressure
25.911.9
30.716.7
31.418.7
30.813.4
cess over time was examined by a Cox pro— cm of water
12
portional-hazards model. Base-line covariTidal volume — ml/kg
5.30.9
6.61.7
5.21.8
7.42.1
ates included in the model were the weaning
Respiratory frequency — breaths/
28.45.4
26.86.4
28.95.4
29.98.4
technique, age, APACHE II score, ratio of
min
PaO2 to FiO2, maximal inspiratory pressure,
Duration of ventilator support be6.54.5†
10.88.6
11.57.4
8.45.3
spontaneous respiratory frequency, spontanefore weaning begun — days
ous tidal volume per kilogram, duration of
Time to failure of 1st spontaneous- 48.533.2
52.334.6
46.523.6
52.532.7
previous ventilator support, and the length of
breathing trial — min
time to the failure of the initial trial of spon*Plus–minus values are means SD.
taneous breathing. Backward elimination was
used to reduce the model to the subgroup of
†P 0.037 for the comparison with the other three groups.
factors that made statistically significant contributions to variation in the time required
for successful weaning. Data were censored on 2 patients who died
ration of ventilatory support before weaning was beduring the study, 2 patients in whom weaning was interrupted begun, which was shorter in the patients who received incause of intercurrent illness, 23 patients who required reintubation
termittent mandatory ventilation than in the other
within 48 hours of extubation, and 11 patients who were still receivgroups (Table 1).
ing ventilator support on day 14. We calculated that 31 patients were
Kaplan–Meier plots of the probability of successful
needed in each group to detect at a power of 80 percent a difference
in weaning time between groups of two days, with a two-tailed alpha
weaning with the use of each technique are shown
error of 0.05. Data are presented as means SD, medians, or proporin Figure 1, and the associated median times to suctions, as appropriate.
Of the 546 study patients, 416 (76.2 percent) successfully underwent a two-hour trial of spontaneous
breathing, and 372 (89.4 percent) of them were immediately extubated. Of these 372 patients, 58 (15.6 percent) required reintubation within 48 hours. Extubation was postponed for 24 hours in 44 patients,
primarily because of concern about their ability to
maintain clear airways. These patients breathed
through a T-tube circuit for up to 24 hours, but 16 (36.4
percent) required reinstitution of mechanical ventilation during this period. The remaining 28 (63.6 percent) were extubated within this 24-hour period, and
only 2 required reintubation within the subsequent 48
hours.
One hundred thirty patients (23.8 percent) had signs
of poor tolerance during the initial trial of spontaneous
breathing, which lasted a mean (SD) of 50.131.2
minutes (range, 5 to 110). These patients were randomly assigned to intermittent mandatory ventilation (29
patients), pressure-support ventilation (37), intermittent trials of spontaneous breathing (33) involving the
use of a T-tube (27) or continuous positive airway pressure (6) interspersed with assist–control ventilation, or
a once-daily trial of spontaneous breathing alternating
with assist–control ventilation (31). The groups were
similar with respect to the patients’ characteristics, the
indications for mechanical ventilation, and respiratory
function; the only significant difference was in the du-
cessful extubation are listed (with first and third quartiles) in Table 2. Cox proportional-hazards regression
analysis revealed four factors that predicted the time
1.0
Probability of Successful Weaning
RESULTS
0.9
0.8
0.7
0.6
0.5
Intermittent trials
0.4
0.3
Once – daily trial
0.2
Pressure – support ventilation
0.1
Intermittent mandatory
ventilation
0.0
2
4
6
8
10
12
14
Duration of Weaning (days)
Figure 1. Kaplan–Meier Curves of the Probability of Successful
Weaning with Intermittent Mandatory Ventilation, Pressure-Support Ventilation, Intermittent Trials of Spontaneous Breathing,
and a Once-Daily Trial of Spontaneous Breathing.
After adjustment for base-line characteristics in a Cox proportional-hazards model, the rate of successful weaning with a
once-daily trial of spontaneous breathing was 2.83 times higher
than that with intermittent mandatory ventilation (P0.006) and
2.05 times higher than that with pressure-support ventilation
(P0.04).
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
348
THE NEW ENGLAND JOURNAL OF MEDICINE
Table 2. The Length of Time from the Initiation of
Weaning to Successful Extubation in the Four
Groups.
WEANING TECHNIQUE
MEDIAN
FIRST
THIRD
QUARTILE QUARTILE
days
Intermittent mandatory ventilation
Pressure-support ventilation
Intermittent trials of spontaneous
breathing
Once-daily trial of spontaneous
breathing
5
3
11
4
3
2
2
12
6
3
1
6
Feb. 9, 1995
support ventilation was not significantly different from
that with intermittent mandatory ventilation (rate ratio,
1.38; 95 percent confidence interval, 0.68 to 2.79;
P= 0.32).
Table 4 lists outcomes for the various techniques.
More patients in the group that received intermittent
mandatory ventilation required continued ventilatory
support on the 14th day than in the groups that received once-daily trials (P=0.07) or intermittent trials
(P=0.06) of spontaneous breathing. The rates of extubation and reintubation did not significantly differ between the four groups.
DISCUSSION
required for successful weaning: age (P0.02), the duration of ventilatory support before weaning was begun
(P0.005), the time to the failure of the first trial of
spontaneous breathing (P0.001), and weaning technique (Table 3). The adjusted rate of successful weaning was higher with a once-daily trial of spontaneous
breathing than with intermittent mandatory ventilation
(rate ratio, 2.83; 95 percent confidence interval, 1.36 to
5.89; P0.006) or pressure-support ventilation (rate ratio, 2.05; 95 percent confidence interval, 1.04 to 4.04;
P0.04) but not significantly different from that with
intermittent trials of spontaneous breathing (rate ratio,
1.24; 95 percent confidence interval, 0.64 to 2.41;
P0.54). The adjusted rate of successful weaning with
intermittent trials of spontaneous breathing was higher
than that with intermittent mandatory ventilation (rate
ratio, 2.28; 95 percent confidence interval, 1.11 to 4.68;
P0.024), but it was not significantly different from
that with pressure-support ventilation (rate ratio, 1.66;
95 percent confidence interval, 0.87 to 3.16; P=0.126).
The adjusted rate of successful weaning with pressureTable 3. Rate of Successful Weaning with the Various Techniques and According to Base-Line Characteristics.*
VARIABLE
Weaning technique
Once-daily trial of spontaneous breathing vs. intermittent mandatory ventilation
Once-daily trial of spontaneous breathing vs. pressuresupport ventilation
Once-daily trial of spontaneous breathing vs. intermittent trials of spontaneous
breathing
Duration of ventilator support
before weaning begun
(1-day increments)
Time to failure of first trial of
spontaneous breathing
(10-min increments)
Age (10-yr increments)
RELATIVE RATE OF
SUCCESSFUL WEANING
(95% CONFIDENCE
INTERVAL)
P VALUE
2.83 (1.36–5.89)
0.006
2.05 (1.04–4.04)
0.04
1.24 (0.64–2.41)
0.54
0.94 (0.90–0.98)
0.005
1.15 (1.07–1.24)
0.001
0.83 (0.71–0.96)
0.02
*Proportional-hazards regression analysis was used to estimate the 95
percent confidence interval of the relative rate of successful weaning.
This study has two major findings. First, in a selected group of patients who were difficult to wean from
mechanical ventilation, the rate of successful weaning
depended on the technique employed: a once-daily trial
of spontaneous breathing led to extubation about three
times more quickly than intermittent mandatory ventilation and about twice as quickly as pressure-support
ventilation. There was no significant difference in the
rate of success between a once-daily trial and multiple
daily trials of spontaneous breathing or between intermittent mandatory ventilation and pressure-support
ventilation. Second, ventilator support was discontinued without any special weaning technique in two
thirds of an unselected group of patients, and only a
small proportion required reintubation within 48 hours.
Intermittent Mandatory Ventilation
Several advantages have been claimed for intermittent mandatory ventilation as a weaning technique: it
is supposed to prevent a patient from “fighting” the
ventilator, reduce respiratory-muscle fatigue, and expedite weaning.4,13 However, there are few data to support
these claims.14 Intermittent mandatory ventilation is
usually delivered in a synchronized manner with demand-valve circuitry, which increases the work of
breathing.14,15 The intermittent nature of assistance
also poses a problem. It was previously assumed that
the degree of respiratory-muscle rest was proportional
to the level of machine assistance. However, recent evidence indicates that respiratory-sensor output does
not adjust to breath-to-breath changes in respiratory
load,16,17 and intermittent mandatory ventilation may
therefore contribute to the development of respiratorymuscle fatigue or prevent recovery from it.
Studies of the efficacy of intermittent mandatory
ventilation in weaning have serious limitations. Schachter et al.6 compared it with conventional ventilation and
noted no difference between the two techniques in the
duration of ventilator support. Their study suffers from
a retrospective design, nonuniform study groups, and
inadequate description of the protocol. Hastings et al.7
compared trials of spontaneous breathing with intermittent mandatory ventilation at a fixed rate (4 breaths
per minute) in patients in stable condition after cardiac
surgery. The length of time to extubation was similar in
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Vol. 332
No. 6
COMPARISON OF METHODS OF WEANING PATIENTS FROM MECHANICAL VENTILATION
Table 4. Outcomes in Patients Who Were Difficult to Wean from
Mechanical Ventilation.*
WEANING TECHNIQUE
SUCCESSFUL
WEANING AND
EXTUBATION
REINTUBATION
CONTINUED MECHANICAL
VENTILATION AFTER
14 DAYS
no. of patients (%)
Intermittent mandatory
ventilation
Pressure-support ventilation
Intermittent trials of spontaneous breathing
Once-daily trial of spontaneous breathing
20 (69.0)
4 (13.8)
5 (17.2)
23 (62.2)
27 (81.8)
7 (18.9)
5 (15.2)
4 (10.8)
1 (3.0)
22 (71.0)
7 (22.6)
1 (3.2)
*The percentages do not total 100 percent in the groups that received pressure-support ventilation and a once-daily trial of spontaneous breathing because one patient died in each group
and weaning was interrupted because of an intercurrent illness in two patients in the pressuresupport group.
the two groups — approximately 2.6 hours. Their study
provides little insight, however, because 24 hours had
already elapsed since the operation and the patients
had good pulmonary function; thus, little difficulty in
weaning was anticipated. In patients in stable condition
who received ventilator support for 3.6 days, Tomlinson
et al.8 found that the duration of weaning was similar
with spontaneous-breathing trials and intermittent
mandatory ventilation — approximately 5.6 hours.
This study was weighted toward patients who received
short-term ventilatory support, and two thirds of those
weaned within 2 hours were patients who received ventilatory support for less than 72 hours postoperatively.
In contrast, we studied difficult-to-wean patients who
had received mechanical ventilation for 6.54.5 days.
Although most patients could theoretically have met
the extubation criteria within 24 hours of study entry,
17 percent were receiving ventilatory support after 14
days. Weaning took longer than in either of the trials
of spontaneous breathing.7,8 Despite the use of randomization, the patients in the group assigned to intermittent mandatory ventilation had received ventilation for
a shorter time than the patients in the other groups.
This actually resulted in a bias in their favor, since
weaning was accomplished more rapidly in patients receiving short-term support.
Pressure-Support Ventilation
Pressure-support ventilation is commonly used to
counteract the work of breathing imposed by endotracheal tubes and ventilator circuits. Theoretically, this
should help with weaning, because a patient who is
comfortable at the compensatory level of pressure support should be able to sustain ventilation after extubation. However, the level of pressure support necessary
to eliminate the work imposed by endotracheal tubes
and ventilator circuits varies considerably (from 3 to 14
cm of water)18,19; thus, any prediction of a patient’s ability to sustain ventilation after extubation is likely to be
misleading.
Brochard et al.20 recently reported that the duration
of weaning was significantly shorter with pressure sup-
349
port (5.73.7 days) than with intermittent mandatory
ventilation (9.98.2 days) or trials of spontaneous
breathing (8.58.3 days). In contrast, we found that
weaning with pressure-support ventilation took longer
than weaning with a once-daily trial of spontaneous
breathing and was not superior to weaning with intermittent mandatory ventilation. We suspect that the apparent superiority of pressure support in the study by
Brochard et al. was due to the constrained manner in
which they used other techniques. Patients had to tolerate an intermittent mandatory ventilation rate of
4 breaths per minute for at least 24 hours before being extubated. This poses a considerable ventilatory
challenge and is not the usual approach to this technique.3,4,14,21 In contrast, we extubated patients when
they tolerated a ventilator rate of five breaths per
minute for two hours. In the study by Brochard et al.,
physicians could request up to three trials of spontaneous breathing over a 24-hour period, each lasting
2 hours, before deciding to extubate a patient. Again,
this is a considerable ventilatory challenge — especially
in patients who have already had difficulty in weaning.
We consider the findings of their study and ours to be
complementary. Both show that the pace of weaning
depends on the manner in which a technique is employed. When intermittent mandatory ventilation and
spontaneous-breathing trials are used in a constrained
manner, weaning is slower than with pressure-support
ventilation.20 Weaning is expedited when a trial of
spontaneous breathing is attempted once a day. In both
studies, the results pertain to specific regimens for each
weaning technique and cannot be extrapolated to other
regimens using these techniques.
Trials of Spontaneous Breathing
Some physicians gradually increase the duration of
spontaneous-breathing trials while reinstituting mechanical ventilation between trials. Other physicians go
directly from offering a high level of ventilatory assistance to a trial of spontaneous breathing, and if the trial
is successful, extubate the patient without any further
weaning. In the present study, two thirds of the patients
initially enrolled were extubated after their first trial of
spontaneous breathing. A once-daily trial of spontaneous breathing also allowed speedier weaning than approaches offering partial ventilatory support. This approach simplifies management, since a patient’s ability
to breathe spontaneously without ventilatory support
needs to be assessed only once a day. In contrast, with
intermittent mandatory ventilation and pressure-support ventilation, ventilator settings must be adjusted repeatedly and each adjustment is usually followed by an
arterial-blood gas measurement.
An implied goal of the various weaning techniques is
to recondition respiratory muscles that may have been
weakened during the period of mechanical ventilation.
Theoretically, a once-daily trial of spontaneous breathing and a prolonged period of rest may be the most effective method of eliciting adaptive changes.22,23 This
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
350
THE NEW ENGLAND JOURNAL OF MEDICINE
approach meets the three principal requirements of a
conditioning program: overload, specificity, and reversibility.22 During the trial, patients breathe against an elevated intrinsic load, thus satisfying the overload requirement. Specificity is also satisfied, in that the trial
is an endurance stimulus and the desired objective is
enhanced endurance. Finally, the use of a daily trial
prevents regression of the adaptive changes. It must be
emphasized that this reasoning is based on indirect evidence and that the effect of different weaning techniques on respiratory-muscle reconditioning has not
been investigated.
We are indebted to Amal Jubran, M.D., and Franco Laghi, M.D.,
for their careful review of the manuscript; to Victor Abraira and William Henderson, Ph.D., for review of the statistical analysis; and to
Alejandro Fernandez for artwork.
APPENDIX
The other members of the Spanish Lung Failure Collaborative
Group are as follows: F. del Nogal and A. Algora (Hospital Severo
Ochoa, Leganés); E. Palazón and M. Cerón (Hospital Universitario
de Murcia, Murcia); J. Ibañez and J.M. Raurich (Hospital Son Dureta, Palma de Mallorca); J. Gudín and J. Cebrián (Hospital La Fé, Valencia); G. González and J.A. Gómez Rubi (Hospital Virgen de la Arrixaca, Murcia); F. Iturbe (Hospital Arnau de Vilanova, Lleida);
A. Vazquez (Hospital del Mar, Barcelona); P. Saura (Hospital Parc
Tauli, Sabadell); J. Gener (Hospital Germans Trias i Pujol, Badalona); D. Fontaneda (Complejo Hospitalario de León, León); V. Sagredo (Hospital General de Segovia, Segovia); and M.J. Prieto (Hospital
del Río Ortega, Valladolid) — all in Spain.
REFERENCES
1. Tobin MJ. Mechanical ventilation. N Engl J Med 1994;330:1056-61.
2. Esteban A, Alía I, Ibañez J, Benito S, Tobin MJ, Spanish Lung Failure Collaborative Group. Modes of mechanical ventilation and weaning: a national
survey of Spanish hospitals. Chest 1994;106:1188-93.
3. Tobin MJ, Alex CG. Discontinuation of mechanical ventilation. In: Tobin
MJ, ed. Principles and practice of mechanical ventilation. New York:
McGraw-Hill, 1994:1177-206.
4. Downs JB, Klein EF Jr, Desautels D, Modell JH, Kirby RR. Intermittent
mandatory ventilation: a new approach to weaning patients from mechanical
ventilators. Chest 1973;64:331-5.
Feb. 9, 1995
5. Brochard L. Pressure support ventilation. In: Tobin MJ, ed. Principles and
practice of mechanical ventilation. New York: McGraw-Hill, 1994:239-57.
6. Schachter EN, Tucker D, Beck GJ. Does intermittent mandatory ventilation
accelerate weaning? JAMA 1981;246:1210-4.
7. Hastings PR, Bushnell LS, Skillman JJ, Weintraub RM, Hedley-Whyte J.
Cardiorespiratory dynamics during weaning with IMV versus spontaneous
ventilation in good-risk cardiac-surgery patients. Anesthesiology 1980;53:
429-31.
8. Tomlinson JR, Miller KS, Lorch DG, Smith L, Reines HD, Sahn SA. A prospective comparison of IMV and T-piece weaning from mechanical ventilation. Chest 1989;96:348-52.
9. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity
of disease classification system. Crit Care Med 1985;13:818-29.
10. Altman DG, Doré CJ. Randomisation and baseline comparisons in clinical
trials. Lancet 1990;335:149-53.
11. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81.
12. Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972;34:187220.
13. Venus B, Smith RA, Mathru M. National survey of methods and criteria
used for weaning from mechanical ventilation. Crit Care Med 1987;15:5303.
14. Sassoon CSH. Intermittent mandatory ventilation. In: Tobin MJ, ed. Principles and practice of mechanical ventilation. New York: McGraw-Hill, 1994:
221-37.
15. Gibney RTN, Wilson RS, Pontoppidan H. Comparison of work of breathing
on high gas flow and demand valve continuous positive airway pressure systems. Chest 1982;82:692-5.
16. Marini JJ, Smith TC, Lamb VJ. External work output and force generation
during synchronized intermittent mandatory ventilation: effect of machine
assistance on breathing effort. Am Rev Respir Dis 1988;138:1169-79.
17. Imsand C, Feihl F, Perret C, Fitting JW. Regulation of inspiratory neuromuscular output during synchronized intermittent mechanical ventilation. Anesthesiology 1994;80:13-22.
18. Brochard L, Rua F, Lorino H, Lemaire F, Harf A. Inspiratory pressure support compensates for the additional work of breathing caused by the endotracheal tube. Anesthesiology 1991;75:739-45.
19. Nathan SD, Ishaaya AM, Koerner SK, Belman MJ. Prediction of minimal
pressure support during weaning from mechanical ventilation. Chest 1993;
103:1215-9.
20. Brochard L, Rauss A, Benito S, et al. Comparison of three methods of gradual withdrawing from ventilatory support during weaning from mechanical
ventilation. Am J Respir Crit Care Med 1994;150:896-903.
21. Luce JM, Pierson DJ, Hudson LD. Intermittent mandatory ventilation. Chest
1981;79:678-85.
22. Faulkner JA. Structural and functional adaptations of skeletal muscle. In:
Roussos C, Macklem PT, eds. The thorax. Part B. Vol. 29 of Lung biology
in health and disease. New York: Marcel Dekker, 1985:1329-51.
23. Rochester DF. Does respiratory muscle rest relieve fatigue or incipient fatigue? Am Rev Respir Dis 1988;138:516-7.
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Copyright, 1995, by the Massachusetts Medical Society
Volume 332
FEBRUARY 9, 1995
Number 6
A COMPARISON OF FOUR METHODS OF WEANING PATIENTS FROM MECHANICAL
VENTILATION
ANDRÉS ESTEBAN, M.D., PH.D., FERNANDO FRUTOS, M.D., MARTIN J. TOBIN, M.D., INMACULADA ALÍA, M.D.,
JOSÉ F. SOLSONA, M.D., INMACULADA VALVERDÚ, M.D., RAFAEL FERNÁNDEZ, M.D.,
MIGUEL A. DE LA CAL, M.D., SALVADOR BENITO, M.D., PH.D., ROSER TOMÁS, M.D.,
DEMETRIO CARRIEDO, M.D., SANTIAGO MACÍAS, M.D., AND JESÚS BLANCO, M.D.,
FOR THE S PANISH L UNG F AILURE C OLLABORATIVE G ROUP *
Abstract Background. Weaning patients from mechanical ventilation is an important problem in intensive care
units. Weaning is usually conducted in an empirical manner, and a standardized approach has not been developed.
Methods. We carried out a prospective, randomized,
multicenter study involving 546 patients who had received
mechanical ventilation for a mean (SD) of 7.56.1 days
and who were considered by their physicians to be ready
for weaning. One hundred thirty patients had respiratory
distress during a two-hour trial of spontaneous breathing.
These patients were randomly assigned to undergo one
of four weaning techniques: intermittent mandatory ventilation, in which the ventilator rate was initially set at a
mean (SD) of 10.02.2 breaths per minute and then
decreased, if possible, at least twice a day, usually by 2 to
4 breaths per minute (29 patients); pressure-support ventilation, in which pressure support was initially set at
18.06.1 cm of water and then reduced, if possible, by
2 to 4 cm of water at least twice a day (37 patients);
intermittent trials of spontaneous breathing, conducted
two or more times a day if possible (33 patients); or a
once-daily trial of spontaneous breathing (31 patients).
Standardized protocols were followed for each technique.
Results. The median duration of weaning was 5 days
for intermittent mandatory ventilation (first quartile, 3 days;
third quartile, 11 days), 4 days for pressure-support ventilation (2 and 12 days, respectively), 3 days for intermittent (multiple) trials of spontaneous breathing (2 and
6 days, respectively), and 3 days for a once-daily trial of
spontaneous breathing (1 and 6 days, respectively). After
adjustment for other covariates, the rate of successful
weaning was higher with a once-daily trial of spontaneous
breathing than with intermittent mandatory ventilation
(rate ratio, 2.83; 95 percent confidence interval, 1.36 to
5.89; P0.006) or pressure-support ventilation (rate ratio,
2.05; 95 percent confidence interval, 1.04 to 4.04;
P0.04). There was no significant difference in the rate
of success between once-daily trials and multiple trials of
spontaneous breathing.
Conclusions. A once-daily trial of spontaneous breathing led to extubation about three times more quickly than
intermittent mandatory ventilation and about twice as
quickly as pressure-support ventilation. Multiple daily trials of spontaneous breathing were equally successful.
(N Engl J Med 1995;332:345-50.)
A
weaning, it is surprising that the process continues to
be managed empirically and that a standardized approach has not been developed.
Weaning techniques differ considerably from one another.3 Traditionally, intermittent trials of spontaneous
breathing, conducted one or more times a day, have
been used. Intermittent mandatory ventilation was
introduced amid claims that it was superior to the traditional weaning approach. It allows the patient to
breathe spontaneously between ventilator-delivered
breaths4; thus, weaning can be considered to begin with
the institution of mechanical ventilation. In the 1980s,
pressure-support ventilation became available5; it provides a titratable pressure boost to every inspiratory effort, and weaning is accomplished by gradually decreasing the level of the pressure boost.
Efficacy studies of weaning techniques can be faulted
for having a retrospective design, inappropriate study
LTHOUGH often lifesaving, mechanical ventilation
causes numerous life-threatening complications,1
making it important to discontinue ventilator support
at the earliest possible time. More than 40 percent of
the time that a patient receives mechanical ventilation
is spent trying to wean the patient from the ventilator.2
Considering the proportion of staff time devoted to
From the Hospital Universitario de Getafe, Madrid (A.E., F.F., I.A., M.A.C.);
Loyola University, Chicago, and Hines Veterans Affairs Hospital, Hines, Ill.
(M.J.T.); Hospital del Mar (J.F.S.) and Hospital Santa Creu i Sant Pau (I.V., S.B.),
Barcelona, Spain; Hospital Parc Tauli, Sabadell, Spain (R.F.); Hospital Germans
Trias i Pujol, Badalona, Spain (R.T.); Complejo Hospitalario de León, León,
Spain (D.C.); Hospital General de Segovia, Segovia, Spain (S.M.); and Hospital
del Río Ortega, Valladolid, Spain (J.B.). Address reprint requests to Dr. Esteban
at the Servicio de Cuidados Intensivos, Hospital Universitario de Getafe, Ctra. de
Toledo km 12’500, 28905 Getafe, Madrid, Spain.
Supported in part by a grant from the Veterans Affairs Research Service.
*The other members of the Spanish Lung Failure Collaborative Group are listed in the Appendix.
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
346
THE NEW ENGLAND JOURNAL OF MEDICINE
populations, and poorly standardized protocols; in addition, most were conducted before the use of pressuresupport ventilation became widespread.6-8 Accordingly,
we performed a prospective, randomized study involving patients who were deemed ready to discontinue mechanical ventilation. In a subgroup of patients who
were difficult to wean we compared the length of time
required for weaning with the use of four techniques:
intermittent mandatory ventilation, pressure-support
ventilation, intermittent trials of spontaneous breathing
conducted several times a day, and a once-daily trial of
spontaneous breathing.
METHODS
Patients
The study was conducted between October 1992 and October
1993 in the medical–surgical intensive care units of 14 teaching hospitals in Spain. The study population consisted of 546 patients (378
men and 168 women), with a mean (SD) age of 58.218.4 years.
All received mechanical ventilation for more than 24 hours because
of acute respiratory failure. The following underlying conditions were
present: chronic obstructive pulmonary disease with acute respiratory failure in 128 patients, acute lung injury in 319, neurologic or neuromuscular disorders in 85, and miscellaneous causes in 14. The
acute lung injury was a result of surgery in 74 patients, infection in
73, heart failure in 69, multiple trauma in 51, adult respiratory distress syndrome in 23, and other pulmonary causes in 29. On admission to the intensive care unit, the patients had a mean score of
18.77.0 on the Acute Physiology and Chronic Health Evaluation
(APACHE II) scale.9 Until the first attempt was made to discontinue
ventilator support, all patients received assist–control ventilation.
The patients received mechanical ventilation for a mean of 7.56.1
days before weaning was started. No hospital contributed more than
10 percent of the study population.
To be enrolled in the study the patients had to have an improvement in or resolution of the underlying cause of acute respiratory failure; adequate gas exchange, as indicated by a ratio of the partial
pressure of arterial oxygen (PaO2) to the fraction of inspired oxygen
(FiO2) above 200 with a positive end-expiratory pressure of 5 cm
of water; a core temperature below 38ºC; a hemoglobin level above
10 g per deciliter; and no further need for vasoactive and sedative
agents. In addition, the attending physician had to agree that the patient was in stable condition and ready to be weaned from the ventilator. Patients with a tracheostomy were excluded. The study was approved by the ethics committees of the hospitals, and the patients
provided informed consent.
Protocol
Feb. 9, 1995
extubation, the patients received supplemental oxygen by face mask.
If a patient had signs of poor tolerance at any time during the trial,
assist–control ventilation was reinstituted. For the purpose of the
study, these patients were designated as being difficult to wean from
mechanical ventilation.
Even if there were no signs of distress by the end of this trial, extubation could be postponed for a maximum of 24 hours if the primary physician thought that a patient might not be able to clear secretions or protect the airway against aspiration. Patients continued
to breathe spontaneously through the T-tube circuit. If they met criteria for poor tolerance, mechanical ventilation was reinstituted.
These patients were not included in the weaning-protocol group.
Patients who were designated as being difficult to wean from mechanical ventilation were stratified according to center and randomly
assigned with the use of a random-number table10 to be weaned in
one of four ways: intermittent mandatory ventilation, pressure-support ventilation, intermittent trials of spontaneous breathing, and a
once-daily trial of spontaneous breathing. The patients were assigned
to the groups in a blinded fashion with the use of opaque, sealed,
numbered envelopes, which were opened only when a patient did not
successfully complete the two-hour trial of spontaneous breathing.
All adjustments for each weaning technique were made by the primary physician.
Intermittent Mandatory Ventilation
In the group that received intermittent mandatory ventilation, the
ventilator rate was initially set at half the frequency used during assist–control ventilation; this initial rate was 10.02.2 breaths per
minute, and mechanical breaths were synchronized with inspiratory
effort. We attempted to decrease the ventilator rate, usually by two
to four breaths per minute, at least twice a day. The ventilator rate
was decreased more rapidly if tolerated by the patient, as reflected by
clinical assessment and blood gas monitoring. Patients who tolerated
a ventilator rate of five breaths per minute for two hours without
signs of distress were extubated. A continuous positive airway pressure of 5 cm of water was permitted.
Pressure-Support Ventilation
In the group that received pressure-support ventilation, pressure
was titrated to achieve a frequency of 25 breaths per minute. Pressure support was initially set at 18.06.1 cm of water, and we attempted to reduce this level of support by 2 to 4 cm of water at least
twice a day. The pace was increased if the patient did not have signs
of distress (the same criteria were applied as in the initial trial of
spontaneous breathing, except that a respiratory frequency of 25
breaths per minute was required). Patients who tolerated pressure
support at a setting of 5 cm of water for two hours with no apparent
ill effects were extubated. A continuous positive airway pressure of
5 cm of water was permitted.
Intermittent Trials of Spontaneous Breathing
After patients were enrolled in the study, assist–control ventilation
was stopped and the patients breathed spontaneously for three minutes through a T-tube circuit, with the FiO2 set at the same level
(0.380.05) as that used during mechanical ventilation. Tidal volume and respiratory frequency were measured with a spirometer during this period. Maximal inspiratory pressure was measured three
times in succession, and the most negative value was selected. Patients who met at least two of the following criteria underwent a trial
of spontaneous breathing lasting up to two hours: maximal inspiratory pressure below 20 cm of water, tidal volume above 5 ml per
kilogram of body weight, and a respiratory frequency of less than 35
breaths per minute. Weaning was considered to have begun with the
onset of this trial. During this trial, patients received humidified oxygen-enriched gas through a T-tube circuit. The primary physician
terminated the trial if a patient had any of the following signs of distress: a respiratory frequency of more than 35 breaths per minute, arterial oxygen saturation below 90 percent, heart rate above 140 beats
per minute or a sustained increase or decrease in the heart rate of
more than 20 percent, systolic blood pressure above 180 mm Hg or
below 90 mm Hg, agitation, diaphoresis, or anxiety. Patients who had
none of these features at the end of the trial were extubated. After
Patients assigned to intermittent trials of spontaneous breathing
were disconnected from the ventilator and allowed to breathe spontaneously through either a T-tube circuit or a continuous-flow circuit
designed to provide a continuous positive airway pressure of 5 cm
of water. The duration of the trials was gradually increased, and they
were attempted at least twice a day. Between the trials, assist–control ventilation was provided for at least one hour. Patients able to
breathe on their own for at least two hours without signs of distress
were extubated.
Once-Daily Trial of Spontaneous Breathing
Patients assigned to a once-daily trial of spontaneous breathing
were disconnected from the ventilator and allowed to breathe spontaneously through a T-tube circuit for up to two hours each day. If
signs of intolerance developed, assist–control ventilation was reinstituted for 24 hours, at which time another trial was attempted. Patients who tolerated a two-hour trial without signs of distress were
extubated.
For all four methods, weaning was considered to have failed if
reintubation was necessary within 48 hours after extubation or if
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Vol. 332
No. 6
COMPARISON OF METHODS OF WEANING PATIENTS FROM MECHANICAL VENTILATION
extubation was not possible after 14 days of
weaning. Weaning was considered successful
if extubation was achieved within the 14-day
period and reintubation was not required
within 48 hours of extubation.
Statistical Analysis
347
Table 1. Characteristics of the Study Population at Base Line.*
CHARACTERISTIC
INTERMITTENT
MANDATORY
VENTILATION
(N 29)
PRESSURESUPPORT
VENTILATION
(N 37)
INTERMITTENT
SPONTANEOUSBREATHING TRIALS
(N 33)
ONCE-DAILY
SPONTANEOUSBREATHING TRIAL
(N 31)
Age — yr
64.213.3
59.916.4
59.116.4
65.014.3
APACHE II score
20.87.0
18.97.6
20.16.8
18.36.6
The chi-square test was used to compare
Chronic obstructive pulmonary
8 (27.6)
18 (48.6)
12 (36.4)
14 (45.2)
categorical data, and the Kruskal–Wallis test
disease —no. (%)
was used to compare continuous variables
Acute lung injury — no. (%)
19 (65.5)
17 (45.9)
18 (54.5)
14 (45.2)
among the groups. The Kaplan–Meier methNeurologic disorder — no. (%)
2 (6.9)
2 (5.4)
3 (9.1)
3 (9.7)
od was used to determine the probability of
the success of a particular method of weaning
243.557.9
242.360.3
223.661.8
229.265.6
Ratio of PaO2 to FiO2
over time.11 The relative probability of sucMaximal inspiratory pressure
25.911.9
30.716.7
31.418.7
30.813.4
cess over time was examined by a Cox pro— cm of water
12
portional-hazards model. Base-line covariTidal volume — ml/kg
5.30.9
6.61.7
5.21.8
7.42.1
ates included in the model were the weaning
Respiratory frequency — breaths/
28.45.4
26.86.4
28.95.4
29.98.4
technique, age, APACHE II score, ratio of
min
PaO2 to FiO2, maximal inspiratory pressure,
Duration of ventilator support be6.54.5†
10.88.6
11.57.4
8.45.3
spontaneous respiratory frequency, spontanefore weaning begun — days
ous tidal volume per kilogram, duration of
Time to failure of 1st spontaneous- 48.533.2
52.334.6
46.523.6
52.532.7
previous ventilator support, and the length of
breathing trial — min
time to the failure of the initial trial of spon*Plus–minus values are means SD.
taneous breathing. Backward elimination was
used to reduce the model to the subgroup of
†P 0.037 for the comparison with the other three groups.
factors that made statistically significant contributions to variation in the time required
for successful weaning. Data were censored on 2 patients who died
ration of ventilatory support before weaning was beduring the study, 2 patients in whom weaning was interrupted begun, which was shorter in the patients who received incause of intercurrent illness, 23 patients who required reintubation
termittent mandatory ventilation than in the other
within 48 hours of extubation, and 11 patients who were still receivgroups (Table 1).
ing ventilator support on day 14. We calculated that 31 patients were
Kaplan–Meier plots of the probability of successful
needed in each group to detect at a power of 80 percent a difference
in weaning time between groups of two days, with a two-tailed alpha
weaning with the use of each technique are shown
error of 0.05. Data are presented as means SD, medians, or proporin Figure 1, and the associated median times to suctions, as appropriate.
Of the 546 study patients, 416 (76.2 percent) successfully underwent a two-hour trial of spontaneous
breathing, and 372 (89.4 percent) of them were immediately extubated. Of these 372 patients, 58 (15.6 percent) required reintubation within 48 hours. Extubation was postponed for 24 hours in 44 patients,
primarily because of concern about their ability to
maintain clear airways. These patients breathed
through a T-tube circuit for up to 24 hours, but 16 (36.4
percent) required reinstitution of mechanical ventilation during this period. The remaining 28 (63.6 percent) were extubated within this 24-hour period, and
only 2 required reintubation within the subsequent 48
hours.
One hundred thirty patients (23.8 percent) had signs
of poor tolerance during the initial trial of spontaneous
breathing, which lasted a mean (SD) of 50.131.2
minutes (range, 5 to 110). These patients were randomly assigned to intermittent mandatory ventilation (29
patients), pressure-support ventilation (37), intermittent trials of spontaneous breathing (33) involving the
use of a T-tube (27) or continuous positive airway pressure (6) interspersed with assist–control ventilation, or
a once-daily trial of spontaneous breathing alternating
with assist–control ventilation (31). The groups were
similar with respect to the patients’ characteristics, the
indications for mechanical ventilation, and respiratory
function; the only significant difference was in the du-
cessful extubation are listed (with first and third quartiles) in Table 2. Cox proportional-hazards regression
analysis revealed four factors that predicted the time
1.0
Probability of Successful Weaning
RESULTS
0.9
0.8
0.7
0.6
0.5
Intermittent trials
0.4
0.3
Once – daily trial
0.2
Pressure – support ventilation
0.1
Intermittent mandatory
ventilation
0.0
2
4
6
8
10
12
14
Duration of Weaning (days)
Figure 1. Kaplan–Meier Curves of the Probability of Successful
Weaning with Intermittent Mandatory Ventilation, Pressure-Support Ventilation, Intermittent Trials of Spontaneous Breathing,
and a Once-Daily Trial of Spontaneous Breathing.
After adjustment for base-line characteristics in a Cox proportional-hazards model, the rate of successful weaning with a
once-daily trial of spontaneous breathing was 2.83 times higher
than that with intermittent mandatory ventilation (P0.006) and
2.05 times higher than that with pressure-support ventilation
(P0.04).
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
348
THE NEW ENGLAND JOURNAL OF MEDICINE
Table 2. The Length of Time from the Initiation of
Weaning to Successful Extubation in the Four
Groups.
WEANING TECHNIQUE
MEDIAN
FIRST
THIRD
QUARTILE QUARTILE
days
Intermittent mandatory ventilation
Pressure-support ventilation
Intermittent trials of spontaneous
breathing
Once-daily trial of spontaneous
breathing
5
3
11
4
3
2
2
12
6
3
1
6
Feb. 9, 1995
support ventilation was not significantly different from
that with intermittent mandatory ventilation (rate ratio,
1.38; 95 percent confidence interval, 0.68 to 2.79;
P= 0.32).
Table 4 lists outcomes for the various techniques.
More patients in the group that received intermittent
mandatory ventilation required continued ventilatory
support on the 14th day than in the groups that received once-daily trials (P=0.07) or intermittent trials
(P=0.06) of spontaneous breathing. The rates of extubation and reintubation did not significantly differ between the four groups.
DISCUSSION
required for successful weaning: age (P0.02), the duration of ventilatory support before weaning was begun
(P0.005), the time to the failure of the first trial of
spontaneous breathing (P0.001), and weaning technique (Table 3). The adjusted rate of successful weaning was higher with a once-daily trial of spontaneous
breathing than with intermittent mandatory ventilation
(rate ratio, 2.83; 95 percent confidence interval, 1.36 to
5.89; P0.006) or pressure-support ventilation (rate ratio, 2.05; 95 percent confidence interval, 1.04 to 4.04;
P0.04) but not significantly different from that with
intermittent trials of spontaneous breathing (rate ratio,
1.24; 95 percent confidence interval, 0.64 to 2.41;
P0.54). The adjusted rate of successful weaning with
intermittent trials of spontaneous breathing was higher
than that with intermittent mandatory ventilation (rate
ratio, 2.28; 95 percent confidence interval, 1.11 to 4.68;
P0.024), but it was not significantly different from
that with pressure-support ventilation (rate ratio, 1.66;
95 percent confidence interval, 0.87 to 3.16; P=0.126).
The adjusted rate of successful weaning with pressureTable 3. Rate of Successful Weaning with the Various Techniques and According to Base-Line Characteristics.*
VARIABLE
Weaning technique
Once-daily trial of spontaneous breathing vs. intermittent mandatory ventilation
Once-daily trial of spontaneous breathing vs. pressuresupport ventilation
Once-daily trial of spontaneous breathing vs. intermittent trials of spontaneous
breathing
Duration of ventilator support
before weaning begun
(1-day increments)
Time to failure of first trial of
spontaneous breathing
(10-min increments)
Age (10-yr increments)
RELATIVE RATE OF
SUCCESSFUL WEANING
(95% CONFIDENCE
INTERVAL)
P VALUE
2.83 (1.36–5.89)
0.006
2.05 (1.04–4.04)
0.04
1.24 (0.64–2.41)
0.54
0.94 (0.90–0.98)
0.005
1.15 (1.07–1.24)
0.001
0.83 (0.71–0.96)
0.02
*Proportional-hazards regression analysis was used to estimate the 95
percent confidence interval of the relative rate of successful weaning.
This study has two major findings. First, in a selected group of patients who were difficult to wean from
mechanical ventilation, the rate of successful weaning
depended on the technique employed: a once-daily trial
of spontaneous breathing led to extubation about three
times more quickly than intermittent mandatory ventilation and about twice as quickly as pressure-support
ventilation. There was no significant difference in the
rate of success between a once-daily trial and multiple
daily trials of spontaneous breathing or between intermittent mandatory ventilation and pressure-support
ventilation. Second, ventilator support was discontinued without any special weaning technique in two
thirds of an unselected group of patients, and only a
small proportion required reintubation within 48 hours.
Intermittent Mandatory Ventilation
Several advantages have been claimed for intermittent mandatory ventilation as a weaning technique: it
is supposed to prevent a patient from “fighting” the
ventilator, reduce respiratory-muscle fatigue, and expedite weaning.4,13 However, there are few data to support
these claims.14 Intermittent mandatory ventilation is
usually delivered in a synchronized manner with demand-valve circuitry, which increases the work of
breathing.14,15 The intermittent nature of assistance
also poses a problem. It was previously assumed that
the degree of respiratory-muscle rest was proportional
to the level of machine assistance. However, recent evidence indicates that respiratory-sensor output does
not adjust to breath-to-breath changes in respiratory
load,16,17 and intermittent mandatory ventilation may
therefore contribute to the development of respiratorymuscle fatigue or prevent recovery from it.
Studies of the efficacy of intermittent mandatory
ventilation in weaning have serious limitations. Schachter et al.6 compared it with conventional ventilation and
noted no difference between the two techniques in the
duration of ventilator support. Their study suffers from
a retrospective design, nonuniform study groups, and
inadequate description of the protocol. Hastings et al.7
compared trials of spontaneous breathing with intermittent mandatory ventilation at a fixed rate (4 breaths
per minute) in patients in stable condition after cardiac
surgery. The length of time to extubation was similar in
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Vol. 332
No. 6
COMPARISON OF METHODS OF WEANING PATIENTS FROM MECHANICAL VENTILATION
Table 4. Outcomes in Patients Who Were Difficult to Wean from
Mechanical Ventilation.*
WEANING TECHNIQUE
SUCCESSFUL
WEANING AND
EXTUBATION
REINTUBATION
CONTINUED MECHANICAL
VENTILATION AFTER
14 DAYS
no. of patients (%)
Intermittent mandatory
ventilation
Pressure-support ventilation
Intermittent trials of spontaneous breathing
Once-daily trial of spontaneous breathing
20 (69.0)
4 (13.8)
5 (17.2)
23 (62.2)
27 (81.8)
7 (18.9)
5 (15.2)
4 (10.8)
1 (3.0)
22 (71.0)
7 (22.6)
1 (3.2)
*The percentages do not total 100 percent in the groups that received pressure-support ventilation and a once-daily trial of spontaneous breathing because one patient died in each group
and weaning was interrupted because of an intercurrent illness in two patients in the pressuresupport group.
the two groups — approximately 2.6 hours. Their study
provides little insight, however, because 24 hours had
already elapsed since the operation and the patients
had good pulmonary function; thus, little difficulty in
weaning was anticipated. In patients in stable condition
who received ventilator support for 3.6 days, Tomlinson
et al.8 found that the duration of weaning was similar
with spontaneous-breathing trials and intermittent
mandatory ventilation — approximately 5.6 hours.
This study was weighted toward patients who received
short-term ventilatory support, and two thirds of those
weaned within 2 hours were patients who received ventilatory support for less than 72 hours postoperatively.
In contrast, we studied difficult-to-wean patients who
had received mechanical ventilation for 6.54.5 days.
Although most patients could theoretically have met
the extubation criteria within 24 hours of study entry,
17 percent were receiving ventilatory support after 14
days. Weaning took longer than in either of the trials
of spontaneous breathing.7,8 Despite the use of randomization, the patients in the group assigned to intermittent mandatory ventilation had received ventilation for
a shorter time than the patients in the other groups.
This actually resulted in a bias in their favor, since
weaning was accomplished more rapidly in patients receiving short-term support.
Pressure-Support Ventilation
Pressure-support ventilation is commonly used to
counteract the work of breathing imposed by endotracheal tubes and ventilator circuits. Theoretically, this
should help with weaning, because a patient who is
comfortable at the compensatory level of pressure support should be able to sustain ventilation after extubation. However, the level of pressure support necessary
to eliminate the work imposed by endotracheal tubes
and ventilator circuits varies considerably (from 3 to 14
cm of water)18,19; thus, any prediction of a patient’s ability to sustain ventilation after extubation is likely to be
misleading.
Brochard et al.20 recently reported that the duration
of weaning was significantly shorter with pressure sup-
349
port (5.73.7 days) than with intermittent mandatory
ventilation (9.98.2 days) or trials of spontaneous
breathing (8.58.3 days). In contrast, we found that
weaning with pressure-support ventilation took longer
than weaning with a once-daily trial of spontaneous
breathing and was not superior to weaning with intermittent mandatory ventilation. We suspect that the apparent superiority of pressure support in the study by
Brochard et al. was due to the constrained manner in
which they used other techniques. Patients had to tolerate an intermittent mandatory ventilation rate of
4 breaths per minute for at least 24 hours before being extubated. This poses a considerable ventilatory
challenge and is not the usual approach to this technique.3,4,14,21 In contrast, we extubated patients when
they tolerated a ventilator rate of five breaths per
minute for two hours. In the study by Brochard et al.,
physicians could request up to three trials of spontaneous breathing over a 24-hour period, each lasting
2 hours, before deciding to extubate a patient. Again,
this is a considerable ventilatory challenge — especially
in patients who have already had difficulty in weaning.
We consider the findings of their study and ours to be
complementary. Both show that the pace of weaning
depends on the manner in which a technique is employed. When intermittent mandatory ventilation and
spontaneous-breathing trials are used in a constrained
manner, weaning is slower than with pressure-support
ventilation.20 Weaning is expedited when a trial of
spontaneous breathing is attempted once a day. In both
studies, the results pertain to specific regimens for each
weaning technique and cannot be extrapolated to other
regimens using these techniques.
Trials of Spontaneous Breathing
Some physicians gradually increase the duration of
spontaneous-breathing trials while reinstituting mechanical ventilation between trials. Other physicians go
directly from offering a high level of ventilatory assistance to a trial of spontaneous breathing, and if the trial
is successful, extubate the patient without any further
weaning. In the present study, two thirds of the patients
initially enrolled were extubated after their first trial of
spontaneous breathing. A once-daily trial of spontaneous breathing also allowed speedier weaning than approaches offering partial ventilatory support. This approach simplifies management, since a patient’s ability
to breathe spontaneously without ventilatory support
needs to be assessed only once a day. In contrast, with
intermittent mandatory ventilation and pressure-support ventilation, ventilator settings must be adjusted repeatedly and each adjustment is usually followed by an
arterial-blood gas measurement.
An implied goal of the various weaning techniques is
to recondition respiratory muscles that may have been
weakened during the period of mechanical ventilation.
Theoretically, a once-daily trial of spontaneous breathing and a prolonged period of rest may be the most effective method of eliciting adaptive changes.22,23 This
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
350
THE NEW ENGLAND JOURNAL OF MEDICINE
approach meets the three principal requirements of a
conditioning program: overload, specificity, and reversibility.22 During the trial, patients breathe against an elevated intrinsic load, thus satisfying the overload requirement. Specificity is also satisfied, in that the trial
is an endurance stimulus and the desired objective is
enhanced endurance. Finally, the use of a daily trial
prevents regression of the adaptive changes. It must be
emphasized that this reasoning is based on indirect evidence and that the effect of different weaning techniques on respiratory-muscle reconditioning has not
been investigated.
We are indebted to Amal Jubran, M.D., and Franco Laghi, M.D.,
for their careful review of the manuscript; to Victor Abraira and William Henderson, Ph.D., for review of the statistical analysis; and to
Alejandro Fernandez for artwork.
APPENDIX
The other members of the Spanish Lung Failure Collaborative
Group are as follows: F. del Nogal and A. Algora (Hospital Severo
Ochoa, Leganés); E. Palazón and M. Cerón (Hospital Universitario
de Murcia, Murcia); J. Ibañez and J.M. Raurich (Hospital Son Dureta, Palma de Mallorca); J. Gudín and J. Cebrián (Hospital La Fé, Valencia); G. González and J.A. Gómez Rubi (Hospital Virgen de la Arrixaca, Murcia); F. Iturbe (Hospital Arnau de Vilanova, Lleida);
A. Vazquez (Hospital del Mar, Barcelona); P. Saura (Hospital Parc
Tauli, Sabadell); J. Gener (Hospital Germans Trias i Pujol, Badalona); D. Fontaneda (Complejo Hospitalario de León, León); V. Sagredo (Hospital General de Segovia, Segovia); and M.J. Prieto (Hospital
del Río Ortega, Valladolid) — all in Spain.
REFERENCES
1. Tobin MJ. Mechanical ventilation. N Engl J Med 1994;330:1056-61.
2. Esteban A, Alía I, Ibañez J, Benito S, Tobin MJ, Spanish Lung Failure Collaborative Group. Modes of mechanical ventilation and weaning: a national
survey of Spanish hospitals. Chest 1994;106:1188-93.
3. Tobin MJ, Alex CG. Discontinuation of mechanical ventilation. In: Tobin
MJ, ed. Principles and practice of mechanical ventilation. New York:
McGraw-Hill, 1994:1177-206.
4. Downs JB, Klein EF Jr, Desautels D, Modell JH, Kirby RR. Intermittent
mandatory ventilation: a new approach to weaning patients from mechanical
ventilators. Chest 1973;64:331-5.
Feb. 9, 1995
5. Brochard L. Pressure support ventilation. In: Tobin MJ, ed. Principles and
practice of mechanical ventilation. New York: McGraw-Hill, 1994:239-57.
6. Schachter EN, Tucker D, Beck GJ. Does intermittent mandatory ventilation
accelerate weaning? JAMA 1981;246:1210-4.
7. Hastings PR, Bushnell LS, Skillman JJ, Weintraub RM, Hedley-Whyte J.
Cardiorespiratory dynamics during weaning with IMV versus spontaneous
ventilation in good-risk cardiac-surgery patients. Anesthesiology 1980;53:
429-31.
8. Tomlinson JR, Miller KS, Lorch DG, Smith L, Reines HD, Sahn SA. A prospective comparison of IMV and T-piece weaning from mechanical ventilation. Chest 1989;96:348-52.
9. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity
of disease classification system. Crit Care Med 1985;13:818-29.
10. Altman DG, Doré CJ. Randomisation and baseline comparisons in clinical
trials. Lancet 1990;335:149-53.
11. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81.
12. Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972;34:187220.
13. Venus B, Smith RA, Mathru M. National survey of methods and criteria
used for weaning from mechanical ventilation. Crit Care Med 1987;15:5303.
14. Sassoon CSH. Intermittent mandatory ventilation. In: Tobin MJ, ed. Principles and practice of mechanical ventilation. New York: McGraw-Hill, 1994:
221-37.
15. Gibney RTN, Wilson RS, Pontoppidan H. Comparison of work of breathing
on high gas flow and demand valve continuous positive airway pressure systems. Chest 1982;82:692-5.
16. Marini JJ, Smith TC, Lamb VJ. External work output and force generation
during synchronized intermittent mandatory ventilation: effect of machine
assistance on breathing effort. Am Rev Respir Dis 1988;138:1169-79.
17. Imsand C, Feihl F, Perret C, Fitting JW. Regulation of inspiratory neuromuscular output during synchronized intermittent mechanical ventilation. Anesthesiology 1994;80:13-22.
18. Brochard L, Rua F, Lorino H, Lemaire F, Harf A. Inspiratory pressure support compensates for the additional work of breathing caused by the endotracheal tube. Anesthesiology 1991;75:739-45.
19. Nathan SD, Ishaaya AM, Koerner SK, Belman MJ. Prediction of minimal
pressure support during weaning from mechanical ventilation. Chest 1993;
103:1215-9.
20. Brochard L, Rauss A, Benito S, et al. Comparison of three methods of gradual withdrawing from ventilatory support during weaning from mechanical
ventilation. Am J Respir Crit Care Med 1994;150:896-903.
21. Luce JM, Pierson DJ, Hudson LD. Intermittent mandatory ventilation. Chest
1981;79:678-85.
22. Faulkner JA. Structural and functional adaptations of skeletal muscle. In:
Roussos C, Macklem PT, eds. The thorax. Part B. Vol. 29 of Lung biology
in health and disease. New York: Marcel Dekker, 1985:1329-51.
23. Rochester DF. Does respiratory muscle rest relieve fatigue or incipient fatigue? Am Rev Respir Dis 1988;138:516-7.
The New England Journal of Medicine
Downloaded from nejm.org on January 4, 2011. For personal use only. No other uses without permission.
Copyright © 1995 Massachusetts Medical Society. All rights reserved.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
