Introduction
The duration of weaning from mechanical ventilation (MV) represents a large proportion of the overall ventilation period
In October 2001, the French intensive care society (Société de Réanimation de Langue Francaise) organized a consensus conference to develop recommendations for MV weaning in the ICU
Methods
Study design
This prospective study was conducted in the 12-bed ICU of an 800-bed teaching hospital from January 2002 to February 2003. We employed a matched control group (1:1 matching) of patients identified from a historical database.
Selection of patients and controls
Screened patients were those who required more than 48 hours of MV and who satisfied eligibility criteria (see below) for a spontaneous breathing trial (SBT). Exclusion criteria were tracheostomy before ICU admission or within the first 48 hours of ICU management, and age under 18 years.
A computer-generated list of potential control individuals was obtained from a historical database of patients attending the same ICU (977 patients from 1999 to 2001). Controls were selected based on matching in terms of the following criteria: age (±5 years), sex, Simplified Acute Physiology Score (SAPS) II (±5 points; calculated within the first 24 hours of ICU admission), and admission diagnosis. The list of potential controls was reviewed for the best possible match, giving a ranking priority to SAPS II, followed by age, sex and then diagnosis.
Study protocol
The study protocol was written according to the final recommendations of the Société de Réanimation de Langue Francaise's consensus conference on weaning. Patient eligibility for the weaning procedure was identified by daily screening by nurses. Screening was deemed to start immediately after ICU admission. The eligibility criteria for a SBT were the following: fractional inspired oxygen <50%; positive end-expiratory pressure <5 cmH2O; no vasopressor infusion; no sedative agent infusion; and response to simple orders. Physician approval for initiation of SBT was not required. A planned SBT duration of 90 min was employed, and SBTs were always performed using a T-piece. The SBT was terminated before 90 min had elapsed and considered a failure if any of the following criteria were satisfied: pulse oximetry <90%, a respiratory rate >35 breaths/min, a heart rate or a systolic arterial pressure variation >20%, or occurrence of patient agitation. All of these criteria for failure were specifically recorded. A SBT was considered to be successful when the patient could breathe spontaneously for 90 min. Physicians were asked to approve discontinuation of MV following a successful SBT. Extubation was therefore performed if cough was subjectively considered efficient, and if a leak test was considered positive (inspiratory and/or expiratory air leaks after cuff deflation). If the SBT was not well tolerated, then the failure criteria were specifically recorded and the patient returned to their prior ventilator settings and mode. Such patients were then subjected to screening the following day (Fig.
Mechanical ventilation weaning protocol
Mechanical ventilation weaning protocol. Daily nurse screening identified patients eligible for weaning. A spontaneous breathing trial was considered to be successful when the patient could breathe spontaneously for 90 min without clinical intolerance. For such patients, physicians were then asked to approve discontinuation of mechanical ventilation. If the spontaneous breathing trial was not tolerated, then the patient was returned to their prior ventilator settings and screened the day after. FiO2, fractional inspired oxygen; PEEP, positive end-expiratory pressure; SpO2, pulse oximetry.
Pre-protocol weaning management (historical controls)
Before establishment of the nurse's protocol-directed weaning procedure, routine weaning from MV was 'physician directed', as in most European ICUs. Physician's individual preferences determined the mode of weaning (volume assisted controlled, pressure support ventilation, or T-piece). Weaning criteria were not monitored daily. The decision to extubate after a successful weaning procedure was also physician directed.
Definitions
All definitions were selected
Outcomes
The overall duration of MV and the ICU LOS were considered primary outcomes. Secondary outcomes were the overall incidence of ventilator-associated pneumonia, unsuccessful extubation rates and ICU mortality.
Statistical analysis
Categorical variables were expressed as percentage and continuous variables as mean ± standard deviation.
Results
Physiological variables
During the prospective study period 392 patients were admitted to our ICU, of whom 384 patients required ventilatory support. A total of 297 patients were mechanically ventilated through an endotracheal tube, of whom 204 required MV for longer than 48 hours. Among these 204 eligible patients, 100 were excluded from analysis because they died prior to initiation of any weaning procedure (85 patients) or because they were tracheostomized before admission or within the first 48-hour period (15 patients; Fig.
Case selection
Case selection. ICU, intensive care unit; NIV, noninvasive ventilation.
Baseline characteristics in the study patients
Physiological parameters
Protocol-directed weaning group (cases)
Physician-directed weaning group (controls)
Patients (
104
104
Male sex (
62 (60)
62 (60)
Age (years; mean ± SD)
57 ± 18
56 ± 18
SAPS II (mean ± SD)
49 ± 18
49 ± 18
Admission diagnosis (
Medicine
54 (52)
63 (61)
Surgery
25 (24)
20 (19)
Neurosurgery
19 (18)
16 (15)
COPD
6 (6)
5 (5)
Matching was successful for all parameters, and patients' physiological parameters were similar between groups. Admission diagnoses were similarly distributed between groups. COPD, chronic pulmonary obstructive disease; SAPS, Simplified Acute Physiology Score; SD, standard deviation.
Duration of mechanical ventilation and intensive care unit length of stay
All patients were under volume-assisted controlled or pressure support ventilation before entering the weaning procedure. The overall MV duration was 16.6 ± 13 days within cases, and 22.5 ± 21 days within controls (
Kaplan–Meier curves of the risk for remaining mechanically ventilated in protocol-directed (cases) versus physician-directed weaning groups (controls)
Kaplan–Meier curves of the risk for remaining mechanically ventilated in protocol-directed (cases) versus physician-directed weaning groups (controls). The protocol-directed weaning procedure allowed reduction in the overall duration of mechanical ventilation, whatever the patient's diagnosis. The overall mechanical ventilation duration was 16.6 ± 13 days in cases and 22.5 ± 21 days in controls (
Outcome comparison between the study groups, according to admission diagnosis
Outcomes
Protocol-directed weaning group (cases)
Physician-directed weaning group (controls)
Mechanical ventilation duration (days)
Medicine
15.3 ± 13
23 ± 22.6
0.02
Surgery
19.1 ± 15.2
17.4 ± 13.7
0.72
Neurosurgery
18.9 ± 12.8
26.3 ± 23.1
0.29
COPD
14.1 ± 14.6
24.9 ± 24
0.33
Combined
16.6 ± 13
22.5 ± 21
0.02
ICU length of stay (days)
Medicine
20.1 ± 13.7
27.7 ± 23
0.02
Surgery
25 ± 17
23.3 ± 15.4
0.72
Neurosurgery
23 ± 13
34 ± 24
0.09
COPD
20.4 ± 14.6
28.5 ± 25
0.47
Combined
21.6 ± 14.3
27.6 ± 21.7
0.02
Values are expressed as mean ± standard deviation. COPD, chronic pulmonary obstructive disease; ICU, intensive care unit.
Outcomes
No significant differences in unsuccessful MV discontinuation rates were observed between groups (31% for cases versus 35% for controls;
Comparison of complications between study groups
Complications
Protocol-directed weaning group
Physician-directed weaning group
Unsuccessful MV discontinuation
33 (31)
37 (35)
0.81
Reintubation within 48 hours
22 (21)
19 (18)
0.33
NIV for postextubation respiratory distress
22 (21)
26 (25)
0.15
Ventilator-associated pneumonia
21 (20)
33 (31)
0.12
ICU mortality
7 (7)
5 (5)
0.92
Values are expressed as number (%). No significant differences between groups were observed. ICU, intensive care unit; MV, mechanical ventilation; NIV, noninvasive ventilation.
Discussion
The present study was designed to determine the clinical benefit of a nurse's protocol-directed weaning procedure in a broad range of ICU patients (including both medical and surgical patients) who had not been disconnected from the ventilator after 48 hours of MV. Our findings demonstrate the effectiveness of such a procedure performed on a routine basis. The nurse's protocol-directed weaning procedure reduced the duration of MV and the overall ICU LOS in patients who remained dependent on MV after a 48-hour period, without any increase in adverse events or in rate of unsuccessful extubation. Easily implemented, this protocol-directed weaning procedure synthesizes an approach that incorporates the following: daily screening by nursing staff and a single daily 90 min SBT with T-piece trial in selected patients. The criteria used in this procedure to screen patients for readiness for the weaning trial were very simple clinical criteria.
Some authors have demonstrated a key role for daily screening using predetermined criteria during the weaning process
The mean MV duration and ICU LOS found in this study differ significantly from those of previous studies
The unsuccessful extubation rate also appears rather high in our patients as compared with previous studies (31% within cases versus 35% within controls)
Apart from the rather high mean SAPS II levels, the overall mortality rate in our study appears rather low in both groups (5%). However, in accordance with the analysis protocol we excluded patients who died before the predetermined weaning criteria were satisfied, and so this finding cannot be considered to be an adequate reflection of overall ICU mortality rate.
Case series with historical controls have advantages over randomized trials when the effects of routine daily procedures are studied. Indeed, it is easier for a broad range of nursing staff to institute a protocol throughout an entire ICU population than it is to limit it to specific monitored patients. Moreover, there is no risk for crossover effect, by which the staff may modify their behaviour toward control patients and thus blur the distinction between groups.
On the other hand the nonrandomized design of our study is a major limitation. First, only randomization can ensure comparability between cases and controls. Even if cases and controls were matched for age, sex, SAPS II and diagnosis on admission, selection bias cannot be fully excluded and might account for part of our results. Second, evolution of medical care during the study period might have influenced our results
Conclusion
This study demonstrates that a routine and simply applied nurse's protocol-directed weaning procedure was safe and promoted major clinical and outcome benefits for patients requiring MV over a 48 hour period. These benefits were obtained without increasing the average number of complications or the rate of unsuccessful extubation.
Key messages
• A simply applied nurse's protocol-directed weaning procedure was safe and promoted major clinical and outcome benefits for patients.
• Protocol-directed weaning procedure should be used in all ICUs.
Abbreviations
ICU = intensive care unit; LOS = length of stay; MV = mechanical ventilation; SAPS = Simplified Acute Physiology Score; SBT = spontaneous breathing trial.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
JMT drafted the manuscript and participated in the design of the study. GP conceived the study and helped to draft the manuscript. GLG performed the statistical analysis. CGG collected data. AR colected data. JMB participated in the design of the study. EL participated in its design and coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.