Services on Demand
Journal
Article
Indicators
- Cited by SciELO
Related links
- Similars in SciELO
Share
Revista americana de medicina respiratoria
On-line version ISSN 1852-236X
Rev. am. med. respir. vol.16 no.2 CABA June 2016
ORIGINAL ARTICLE
Characteristics Of Patients Weaning From Invasive Mechanical Ventilation. A Multicenter Study
Authors: María Lucía Giménez1, Gabriel Alejandro Verde2, Iris Gloria Salvati3, Walter Ariel Tozzi1, 2, Adriano Javier Cura2, Silvina Borello3, Paola Bustamante1, Sacha Alexis Virgilio2, Marco Bezzi3
1 Hospital de Clínicas José de San Martín, Ciudad Autónoma de Buenos Aires, Argentina.
2 Hospital General de Agudos Parmenio Piñero, Ciudad Autónoma de Buenos Aires, Argentina.
3 Hospital General de Agudos Donación Francisco Santojanni, Ciudad Autónoma de Buenos Aires,
Argentina.
Correspondence to: María Lucía Giménez. Lic. Kinesióloga Fisiatra Dirección postal: San Luis 3139 8 A CP 1186 CABA. Argentina. e-mail: gimenezmarialucia@hotmail.com Teléfono: 1131391682
Received: 12.23.2015
Accepted: 03.18.2016
Abstract
The weaning process includes the release from the ventilatory support and endotracheal
tube. It is classified into simple, difficult and prolonged, according to its difficulty and
duration.
The purpose was to describe the epidemiological characteristics of patients successfully
weaned from invasive mechanical ventilation and establish associations between the
different types of weaning and the variables influencing the evolution of these characteristics associated with mortality.
We conducted a multicenter, prospective, longitudinal, analytical cohort study in three
intensive care units of the Autonomous City of Buenos Aires, Argentina. We included
patients who required invasive mechanical ventilation for more than 12 hours and were
successfully weaned from it.
The variables to be analyzed were: type of weaning, amount of days the patients received
invasive mechanical ventilation, extubation failure and length of stay and mortality in the
intensive care unit.
The prevalence of simple, difficult or prolonged weaning was 52.2% (95/182), 25.8%
(47/182) and 22% (40/182), respectively. The average of days the patients received
invasive mechanical ventilation increased to 3.5 every time the category changed (B
Coefficient: 3.5; SE [standard error] = 0.6). Patients with extubation failure presented
a higher risk of prolonged weaning (OR [odds ratio] = 23; CI [confidence interval] = 95%: 3.55-149.45). No association was found between mortality and type of weaning
(OR = 0.68; 95% CI: 0.31-1.51).
In conclusion, the type of weaning was not associated with mortality in the intensive care
unit. The extubation failure, tracheostomy and presence of delirium were associated with
a larger amount of days receiving invasive mechanical ventilation.
Key words: Epidemiology; Invasive mechanical ventilation; Mortality; Weaning
Introduction
Mechanical ventilation (MV) is one of the therapeutic pillars of intensive care medicine1. The
generalized use of invasive mechanical ventilation
(IMV) was first considered after the sudden decrease in mortality during the poliomyelitis epidemic of 1952. From that moment until now, MV has become
the most common procedure for the management
of critically ill patients2. Golingher et al predict an
80% increase in the incidence of IMV for 2026, in
comparison with the prevalence obtained in 20041.
The weaning process is essential to the care
of critically ill patients. It includes the process of releasing the patient from the ventilatory support
and endotracheal tube, and usually represents
between 40 and 50% of the duration of the IMV3, 4.
In 2007, the International Consensus Conference on Weaning suggested a new classification
based on the difficulty and duration of the weaning
period, considering it as simple, difficult and prolonged3.
This new classification is exclusively based on
expert opinion, and until now only few studies
have used it3-6.
In 2010, Funk et al reported an increase in
hospital mortality within patients with prolonged
weaning (32%), as compared with patients with
simple weaning (13%). On the other hand, no associations have been established when comparing
the mortality rate in the intensive care unit (ICU)
between simple and difficult weaning5, 6.
There are few published studies that address
this topic with the new classification4-6. No published data are known about patients weaning from
IMV in our country.
The purpose of this study was to describe the
epidemiological characteristics of patients who successfully weaned from IMV in three intensive care
units (ICUs) of the Autonomous City of Buenos
Aires (Ciudad Autónoma de Buenos Aires, CABA).
To a lesser extent, associations were established
between the types of weaning according to the new
classification and the variables that influenced
their evolution in relation to mortality.
Materials and Methods
We conducted a multicenter, prospective, longitudinal, analytical cohort study using daily registration
of patients admitted to adult ICUs of three CABA
hospitals during the period between August 1,
2013 and January 31, 2014.
We included patients older than 18 years old
who had been admitted to the ICUs of the three
participating hospitals and had required IMV
for more than 12 hours, meeting the criteria for
successful IMV weaning. Such criteria included:
success in the spontaneous breathing trial (SBT),
without requiring reintubation, and/or restoration
of mechanical ventilation within the first 48 hours
after extubation3.
The hospitals that participated in the study
were: Hospital General de Agudos “Parmenio
Piñero, Hospital de Clínicas “José de San Martín and Hospital General de Agudos “Donación
Francisco Santojanni". They have polyvalent,
medical-surgical intensive care units, on-staff
physicians and nurses and 24-hour physician and
physiotherapist on call. One of the intensive care
units participating in this study belongs to a hospital that is a neurosurgical and trauma referral
center.
For the epidemiological analysis, data from the
following variables were defined and collected:
* Prevalence: The relationship between the
amount of patients that were included and the
total of patients requiring IMV for more than
12 hours.
* Birth data and personal history: Name and
surname, age, gender and Acute Physiology And
Chronic Health Evaluation II (APACHEII) score
at ICU admission.
The following conditions were considered as
indications for IMV:7-9
– Worsened chronic respiratory insufficiency
(CRI) (chronic obstructive pulmonary disease
[COPD], asthma or other).
– Acute respiratory insufficiency (ARI) (acute
respiratory distress syndrome [ARDS], post-surgery [PS], congestive heart failure, aspiration,
pneumonia, sepsis, multiple trauma, cardiac
arrest or other).
– Coma (metabolic, intoxication, stroke, traumatic brain injury or other).
– Neuromuscular disease.
The following intercurrent conditions associated
with IMV were recorded:
– Barotrauma: Presence of pneumoperitoneum,
pneumomediastinum, subcutaneous emphysema, pneumothorax and pneumopericardium,
associated with the implementation of IMV7-9.
– Delirium: Evaluated on a daily basis, whenever
the patient presented a value between < 3 and> 4, according to the Richmond Agitation and
Sedation Scale (RASS). The Chilean validation
of the Confusion Assesment Method Intensive
Care Unit (CAM- ICU)10 was used. Patients
with history of mental illness documented in
the medical record were excluded.11
– Intensive care unit-acquired weakness (ICUAW):
Patients with more than 72 hours of IMV who
were capable of responding to 5 instructions
(eye opening, mouth opening, sticking out their
tongue, squeezing their hand, eye tracking test)
were assessed once a week using the Medical Research Council (MRC) scale. ≤ 48 point-scores12 were considered positive for ICUAW.
– ARDS: According to the criteria of the American-European Consensus Conference regarding
the ARDS13.
– Ventilator-associated pneumonia (VAP): Assessed through the Clinical Pulmonary Infection
Score (CPIS), whenever it was suspected. ≥ 6
point- scores were considered compatible with
VAP14.
Weaning: Patients were classified into three
groups. Simple: Those who started weaning until
they achieved a successful extubation having undergone only one SBT. Difficult: Those requiring
up to three SBTs or up to 7 days after the first
SBT was conducted in order to achieve a successful
extubation. Prolonged: Those who failed more
than three SBT attempts or required more than
7 days after the first SBT was conducted in order
to achieve weaning3.
Regarding the chosen weaning method, the
following was considered: Gradual reduction of
ventilatory support, SBT (with T-piece, continuous
positive airway pressure [CPAP] of 5 cmH2O,
PSV ≤ 7 cmH2O with or without positive endexpiratory pressure [PEEP]) or a combination of
both methods.
Weaning failure: Failure of the SBT or the need
to reintubate and/or restore the patient´s ventilation within the first 48 hours after extubation.3
Tracheostomy (TQT): IMV rate and day in which
it was performed. The following temporary variables and mortality values were recorded:
Length of stay in the ICU: The number of days
was counted from the admission day (day 0) up to
the day the patient was dismissed from the unit.
Hospital length of stay: From hospital admission
(day 0) up to the day the patient was dismissed
or referred to another institution or up to the
patient´s death.
Total IMV days: Counted since it began to be
used at our ICUs (day 0) up to successful weaning,
referral to another institution or the patient´s
death.
Weaning days: Counted from day 0 (zero) as the
day in which the team of professionals decided to
begin with the process until the day in which the
patient was successfully weaned and extubated.5
Also the duration of IMV before weaning was
counted, from the moment it was first provided
at the ICUs (day 0) until the day the professionals decided to begin with the weaning process.
The time of IMV devoted to weaning was calculated by relating the weaning days with total
IMV days.
In order to calculate mortality at the ICU and
the hospital, patients who died both at the unit
and the hospital were taken into account.
The following variables were analyzed 28 days
after admission to the ICU: Total IMV days, ICU
mortality and length of stay.
Follow-up of patients was conducted from ICU
admission until January 31, 2014, hospital discharge, referral or death.
Patients dismissed from the ICU who were readmitted to the unit requiring IMV were considered
as new patients.
During 4 weeks before patient inclusion, records
were implemented in order to prepare the work
team.
According to the statistical analysis that was
conducted, the variables presenting a normal distribution were informed as mean ± SD (standard
deviation), whereas those presenting an asymmetrical distribution were expressed as median with
minimum-maximum range.
An analysis was conducted comparing the 3
weaning groups. For the categorical variable analysis, the Pearson´s Chi-Squared Test was used, as
appropriate. The APACHEII of this group of patients was compared with the Analysis of Variance
(ANOVA) Test. The variable “total IMV days" was
analyzed with the Median Test.
A multiple linear regression analysis was performed between the variables “type of weaning"
and “total IMV days".
A stepwise multinomial logistic regression
analysis was conducted, in order to analyze the
type of weaning variable in relation to all those
variables that obtained p < 0.10. Also, the confusion and interaction factors of each variable were
assessed.
A value of p ≤ 0.05 was considered a statistically
significant difference. The program SPSS®, v17
was used.
The Ethics Committees of participating hospitals approved the performance of the study and
considered it did not imply any risk for the patient
or changes in the patient´s evolution. They also
believed that due to the design and purpose of the
study, the informed consent of participants was
not necessary.
Results
57.6% (182/316) of patients finished the weaning process successfully (Figure 1), representing 44.9% of the total IMV days. The main demographic variables are described in Table 1.
Figure 1. Flow Chart
TABLE 1. Epideiological Characteristics of Patients with IMV
IMV: invasive mechanical ventilation; APACHEII: Acute Physiology And
Chronic Health Evaluation II; SD: standard deviation; COPD: chronic obstructive pulmonary disease; ARDS: acute respiratory distress; PS: postsurgery; S: stroke; TBI: traumatic brain injury; ND: neuromuscular disease.
The most prevalent indication for IMV was ARI,
with 72.5% (132/182). PS was the main cause, with
30.2% (55/182), followed by sepsis/septic shock
and pneumonia with 8.8% (16/182) each (Table 1).
As regards intercurrences registered during the
study period, we found there was 20.8% (38/182)
of mechanical ventilation associated pneumonia
(MVAP), with late onset in 25 patients. 62% developed delirium, whereas 29.2% presented ICUAW.
85.2% (155/182) of patients were able to weave
from IMV by means of SBT as the only method,
1.6% (3/182) through gradual reduction of the
ventilatory support and 13.2% (24/182) through
the combination of both methods. The prevalence
of simple, difficult and prolonged weaning was 52.2% (95/182), 25.8% (47/182) and 22% (40/182),
respectively.
Table 2 describes the epidemiological characteristics and main variables related to the type
of weaning.
TABLE 2. Epidemiological Characteristics According to the Weaning Classifcation
MinR-MaxR: minimum range - maximum range; APACHE II: Acute Physiology And Chronic Health
Evaluation II; SD: standard deviation; IMV: invasive mechanical ventilation; IQR: interquartile range;
CRI: chronic respiratory insufficiency; ARI: acute respiratory insufficiency; MVAP: mechanical ventilation associated pneumonia; DAUCI: intensive care unit-acquired weakness: polyneuropathy in
critically ill patients; TQT: tracheostomy.∞ p value of intergroup.*Simple weaning vs. difficult weaning
p = 0.54.≈ Difficult weaning vs. prolonged weaning p = 0.08. ∞Simple weaning vs. prolonged weaning
p = 0.17.
A multiple linear regression analysis was performed between the variables “type of weaning" and“total IMV days" and we found that the average of
days increased to 3.5 every time the weaning category changed (B Coefficient: 3.5; SE: 0.6. p ≤ 0.01).
A multinomial logistic regression analysis was
performed using the simple weaning as a reference
category. We did not find any statistical association
between age and type of weaning (OR = 1.01; 95%
CI: 0.99-1.04), or between the type of weaning and
mortality variables at the ICU (OR = 0.68; 95%
CI: 0.31-1.51). On the other hand, a statistically
significant association was found between the TQT
and the type of weaning (OR = 7.06; p = 0.04), that
is to say, there is a greater risk every time it goes
from simple to difficult weaning or from difficult to prolonged. Another statistical association that
was obtained suggests that the risk of presenting
difficult weaning increases every time the delirium
event is present (OR = 3.2; 95% CI: 1.15-8.92).
The extubation failure variable presented a
statistically significant association when it was
compared with prolonged weaning (OR = 23; 95%
CI: 3.55-149.45). Also, every time the extubation
failure variable was present, an increase of 6.32
days was found in the average of IMV days (B Coefficient: 6.32; SE 1.28. p ≤ 0.01). Finally, regarding the extubation failure associated with mortality, it was estimated that the mortality odds at the
ICU in subjects who failed extubation was 17.32 times higher in comparison with those subjects
who did not fail (OR: 17.32; 95% CI: 3.85-77.86).
Discussion
The prevalence of simple and difficult weaning was
similar to the one reported by Funk et al,5 whereas
the prevalence of prolonged weaning was higher
than the one reported in the literature5, 6.
Age and APACHEII values found at the sample
are similar among the three groups. However, the
severity score reported at the prolonged weaning
group was higher than the values published by
Sellares in 20104. When comparing median age
with the values reported in the literature, it was
observed that the statistical estimates that were
used differ from those used in this work, thus this
comparison is difficult4-6.
In contrast with the data reported by various
authors5, 6, a statistically significant difference was
obtained for the reintubation variable between the
types of weaning, and a statistically significant
association was found between reintubation and
prolonged weaning.
Regarding the TQT rate in the prolonged
weaning group, the results of this study contrast
with the information reported by Peñuelas et al6,
but coincide with the data published by Funk et al5.
The study of Frutos Vivar et al15 conducted in
a cohort of tracheomized patients, established an
OR = 8.77 (95% CI: 6.76-11.37) for TQT every time
the reintubation event was present. This finding
could be related to the comparisons found in this
study, where reintubated patients have greater
possibilities to undergo prolonged weaning, and a
greater risk of requiring a TQT.
This is the first study to analyze ICUAW and
delirium in patients with IMV according to the new
weaning classification. Although we couldn´t find
any statistical association between these variables
regarding the type of weaning, we did find differences among the three groups.
Significant differences regarding mortality
have been established in this study between the
simple and difficult weaning groups as compared
to the prolonged weaning group. The reason for
the highest mortality rate found in the prolonged
weaning group could be the median weaning days
(10.5 days, minimum-maximum range: 7-15). In
relation to this, Peñuelas et al6 report that the
probability to die is kept constant regardless of the type of weaning until the seventh day after
it began, which is the moment the mortality risk
increases. In contrast to various authors5, 6 and
despite the differences found in this study between
the groups as regards mortality, we were not able
to establish a statistical association.
The mortality rate that was found is lower than
the aforementioned one according to the APACHEII value obtained in each weaning group. This
finding could be related to the fact that such score
is an objectivation of the severity at ICU admission, but maybe it does not take into account the difficulty of the weaning process or the evolution
of the patient during his stay at the unit.
In this study we were able to establish an association between extubation failure and mortality at
the ICU. These results are related to the findings
of Sellares et al,4 which establish that the risk of
dying at the ICU increases every time the patient
fails extubation with an OR = 5.34 (95% CI:
3.14-9.09). To sum up, we may conclude that this
study allowed us to know the characteristics and
evolution of patients who underwent IMV for more
than 12 hours and were able to wean successfully in three ICUs of the Autonomous City of Buenos
Aires, Argentina.
We haven´t found any association between
mortality and type of weaning. The reasons for
starting to use IMV do not impact on the type
of weaning to be developed, whereas the risk of
being tracheomized increases every time the type
of weaning changes category. Extubation failure,
TQT and delirium are associated with an increase
in IMV days.
So, it is necessary to extend this study with
the purpose of increasing the number of analyzed
patients and establishing associations between
ICUAW and delirium among the different weaning
subgroups, given the fact that, as of today, this information has not been reported in the literature.
Acknowledgement:
Physiatry, full-time residency and part-time residency
service of the Hospital General de Agudos “Parmenio
Piñero.
Physiatry, full-time residency and part-time residency
service of the Hospital General de Agudos “Donación
Francisco Santojanni.
Physiatry and residency service of the Hospital de Clínicas “José de San Martín.
Licentiate Mariana Celiz Alonso, for her statistical
advice.
Special acknowledgement to licentiates Luciano Sebastián Pipolo, Villarruel Matías, María Paula Pedace, Luis
Ignacio Garegnani and Norberto Tiribelli for their cooperation with the study.
Conflicts of Interest: Nothing to declare.
1. Goligher E, Ferguson ND. Mechanical ventilation: epidemiological insights into current practices. Curr OpinCrit Care. 2009; 15(1): 44-51.
2. Frutos-Vivar F, Ferguson ND, Esteban A. Mechanical ventilation: quo vadis? Intensive Care Med. 2009; 35(5): 775-8.
3. Boles JM, BionJ, Connors A, Herridge M, Marsh B, Melot C, et al. Weaning from mechanical ventilation. Eur Respir J. 2007;29 (5): 1033-56.
4. Sellares J, Ferrer M, Cano E, Loureiro H, Valencia M, Torres A. Predictors of prolonged weaning and survival during ventilation weaning in a respiratory ICU. Intensive Care Med. 2011; 37(5): 775-84.
5. Funk GC, Anders S, Breyer MK, Burghuber OC, Edelmann G, Heindl W, et al. Incidence and outcome of weaning from mechanical ventilation according to new categories. Eur Respir J. 2010; 35(1):8 8-94.
6. Peñuelas O, Frutos-Vivar F, Fernandez C, Anzueto A,Epstein SK, Apezteguia C, et al. Characteristics and outcomes of ventilated patients according to time to liberation from mechanical ventilation. Am J Respir Crit Care Med. 2011; 184(4): 430-7.
7. Esteban A, Anzueto A, Frutos F, Alia I, Brochard L, Stewart TE, et al. Characteristics and outcomes in adult patient sreceiving mechanical ventilation: a 28- day international study. JAMA.2002; 287(3): 345-55.
8. Esteban A, Ferguson ND, Meade MO, Frutos-Vivar F, Apezteguia C, Brochard L, et al. Evolution of mechanical ventilation in response to clinical research. Am J Respir Crit Care Med. 2008;177(2):170-7.
9. Esteban A, Frutos-Vivar F, Muriel A, Ferguson ND, Penuelas O, Abraira V, et al. Evolution of mortality over time in patients receiving mechanical ventilation. Am J Respir Crit Care Med. 2013; 188(2): 220-30.
10. Tobar E, Romero C, Galleguillos T, Fuentes P, Cornejo R, Lira MT, et al. Confusion Assessment Method for diagnosing delirium in ICU patients (CAM-ICU): cultural adaptation and validation of the Spanish version. Med Intensiva. 2010; 34(1): 4-13.
11. Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, et al. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit(CAM-ICU). Crit Care Med. 2001; 29(7): 1370-9.
12. De Jonghe B, Sharshar T, Lefaucheur JP, Authier FJ, Durand-Zaleski I, Boussarsar M, et al. Paresis acquired in the intensive care unit: a prospective multicenter study. JAMA. 2002; 288(22): 2859-67.
13. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med. 1994;1 49(3 Pt 1): 818-24.
14. Luna CM, Blanzaco D, Niederman MS, Matarucco W, Baredes NC, Desmery P, et al. Resolution of ventilatorassociated pneumonia: prospective evaluation of the clinical pulmonary infection score as an early clinical predictor of outcome. Crit Care Med.2003; 31(3): 676-82.
15. Frutos-Vivar F, Esteban A, Apezteguia C, Anzueto A, Nightingale P, Gonzalez M, et al. Outcome of mechanically ventilated patients who require a tracheostomy. Crit Care Med. 2005; 33(2): 290-8.
1. Goligher E, Ferguson ND. Mechanical ventilation: epidemiological insights into current practices. Curr Opin Crit Care. 2009; 15(1): 44-51. [ Links ]
2. Frutos-Vivar F, Ferguson ND, Esteban A. Mechanical ventilation: quo vadis? Intensive Care Med. 2009; 35(5): 775-8. [ Links ]
3. Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, et al. Weaning from mechanical ventilation. Eur Respir J. 2007; 29(5): 1033-56. [ Links ]
4. Sellares J, Ferrer M, Cano E, Loureiro H, Valencia M, Torres A. Pedictors of prolonged weaning and survival during ventilation weaning in a respiratory ICU. Intensive Care Med. 2011; 37(5): 775-84. [ Links ]
5. Funk GC, Anders S, Breyer MK, Burghuber OC, Edelmann G, Heindl W, et al. Incidence and outcome of weaning from mechanical ventilation according to new categories. Eur Respir J. 2010; 35(1): 88-94. [ Links ]
6. Peñuelas O, Frutos-Vivar F, Fernandez C, Anzueto A, Epstein SK, Apezteguia C, et al. Characteristics and outcomes of ventilated patients according to time to liberation from mechanical ventilation. Am J Respir Crit Care Med. 2011; 184(4): 430-7. [ Links ]
7. Esteban A, Anzueto A, Frutos F, Alia I, Brochard L, Stewart TE, et al. Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28- day international study. JAMA. 2002; 287(3): 345-55. [ Links ]
8. Esteban A, Ferguson ND, Meade MO, Frutos-Vivar F, Apezteguia C, Brochard L, et al. Evolution of mechanical ventilation in response to clinical research. Am J Respir Crit Care Med. 2008; 177(2): 170-7. [ Links ]
9. Esteban A, Frutos-Vivar F, Muriel A, Ferguson ND, Penuelas O, Abraira V, et al. Evolution of mortality over time in patients receiving mechanical ventilation. Am J Respir Crit Care Med. 2013; 188(2): 220-30. [ Links ]
10. Tobar E, Romero C, Galleguillos T, Fuentes P, Cornejo R, Lira MT, et al. Confusion Assessment Method for diagnosing delirium in ICU patients (CAM-ICU): cultural adaptation and validation of the Spanish version. Med Intensiva. 2010; 34(1): 4-13. [ Links ]
11. Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, et al. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med. 2001; 29(7): 1370-9. [ Links ]
12. De Jonghe B, Sharshar T, Lefaucheur JP, Authier FJ, Durand-Zaleski I, Boussarsar M, et al. Paresis acquired in the intensive care unit: a prospective multicenter study. JAMA. 2002; 288(22): 2859-67. [ Links ]
13. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med. 1994; 149(3 Pt 1): 818-24. [ Links ]
14. Luna CM, Blanzaco D, Niederman MS, Matarucco W, Baredes NC, Desmery P, et al. Resolution of ventilatorassociated pneumonia: prospective evaluation of the clinical pulmonary infection score as an early clinical predictor of outcome. Crit Care Med. 2003; 31(3): 676-82. [ Links ]
15. Frutos-Vivar F, Esteban A, Apezteguia C, Anzueto A, Nightingale P, Gonzalez M, et al. Outcome of mechanically ventilated patients who require a tracheostomy. Crit Care Med. 2005; 33(2): 290-8. [ Links ]