Six-month survival of patients with acute lung injury: Prospective
cohort study*
Murat Yilmaz, MD; Remzi Iscimen, MD; Mark T. Keegan, MD; Nicholas E. Vlahakis, MD;
Bekele Afessa, MD; Rolf D. Hubmayr, MD; Ognjen Gajic, MD, MSC
A
cute lung injury (ALI) is a com-
mon critical care syndrome that
carries significant morbidity and
mortality (1). Recent advances
in invasive (2, 3) and noninvasive (4) me-
chanical ventilation, sepsis resuscitation (5,
6), glucose control (7), preventive strate-
gies (8), and intensive care unit (ICU) struc-
ture (9) have had a profound impact on the
management of patients with ALI. Both
short- and long-term outcome studies con-
ducted thus far were performed before the
full implementation of these changes (10 –
13), most often within the context of clin-
ical trials enrolling the minority of patients
with ALI (10, 14, 15). To standardize care
and facilitate the implementation of evi-
dence-based therapies, we have introduced
several hospital-based protocols, including
low tidal volume ventilation protocol, stan-
dardized weaning and sedation protocols,
glucose control protocol, and restrictive
transfusion protocol. In this prospective co-
hort study, we aimed to determine the
prognostic factors for hospital and 6-month
survival of patients with ALI admitted to
the ICU after the full implementation of
recent evidence-based critical care inter-
ventions. In addition to addressing the se-
verity of pulmonary and nonpulmonary or-
gan dysfunction, we specifically assessed
the impact of the following potentially im-
portant prognostic factors: a) underlying
comorbidities; b) presence or absence of
specific clinical trial (Acute Respiratory
Distress Syndrome Network [ARDSNet])
exclusion criteria (2, 15); and c) admission
source (transfer admissions from the an-
other hospital or another ward in the same
hospital and patients admitted from within
of outside local community (Olmsted
County)] (16).
MATERIALS AND METHODS
The institutional review board approved
the study protocol. Daily screening identified
consecutive patients with ALI who were ad-
mitted to three ICUs (medical, surgical, and
mixed medical-surgical) in the two Mayo
Clinic Hospitals from October 2005 to May
2006. Patients who did not give research au-
thorization and those with a do-not-resusci-
tate order on admission were excluded. ALI
was defined as acute onset of hypoxemia: PaO
2
/
Objective: Both short- and long-term outcome studies in acute
lung injury (ALI) performed thus far were conducted before the
implementation of recent advances in mechanical ventilation and
supportive care and/or in the context of clinical trials with re-
stricted inclusion criteria. We sought to determine the outcome of
consecutive acute lung injury patients after the implementation of
these interventions.
Design: Prospective cohort study.
Setting: Three intensive care units of two tertiary care hospitals.
Patients: Patients with acute lung injury treated from October
2005 to May 2006, excluding those with no research authorization
or do-not-resuscitate order.
Interventions: None.
Measurements and Main Results: The investigators collected
detailed information about comorbidities, severity of pulmonary and
nonpulmonary organ failures, complications, respiratory support,
and other interventions. The main outcome measure was mortality 6
months after the onset of acute lung injury. From 142 patients
enrolled over a 6-month period, 24 (17%) died in the intensive care
unit, 38 (27%) in the hospital, and 55 (39%) by the end of the 6-month
follow-up. Median (interquartile range) intensive care unit length of
stay, duration of mechanical ventilation, and number of day 28
ventilator-free days were 7.1 (3.6 –11.3), 5.7 (2.6 –10.3), and, 19.0
(0 –24.2) days. Multiple logistic regression analysis identified under-
lying Charlson comorbidity score (odds ratio 3.11, 95% confidence
interval 2.01–5.05) for each point increase, transfer admission from
the floor or outside hospital (odds ratio 3.75, 95% confidence interval
1.41–10.99), day 3 cardiovascular failure (odds ratio 3.30, 95% con-
fidence interval 1.19 –9.92), and day 3 PaO
2
/FIO
2
(odds ratio 0.94, 95%
confidence interval 0.88 – 0.99) as significant predictors of 6-month
mortality.
Conclusions: With the implementation of recent advances in
mechanical ventilation and supportive care, premorbid condition
is the most important determinant of acute lung injury survival.
(Crit Care Med 2007; 35:D1–D10)
KEY WORDS: acute lung injury; intensive care units; risk factors
prognosis; cohort; outcome assessment
*See also p. 2441.
From the Department of Internal Medicine, Division
of Pulmonary and Critical Care Medicine (MY, RI, NEV,
BA, RDH, OG), Department of Anesthesiology, Division
of Critical Care Medicine (MTK), and Mayo Epidemiol-
ogy and Translational Research in Intensive Care
(M.E.T.R.I.C.) (MTK, NEV, BA, RDH, OG), Mayo Clinic,
Rochester, MN; the Department of Anesthesiology and
Intensive Care, Akdeniz University, Medical Faculty,
Antalya, Turkey (MY); and Department of Anaesthesi-
ology and Reanimation, Uludag University School of
Medicine, Bursa, Turkey (RI).
Supported, in part, by grant K23 HL087843-01A
from the National Heart, Lung, and Blood Institute.
The authors have not disclosed any potential con-
flicts of interest.
Address requests for reprints to: Ognjen Gajic, MD,
Mayo Clinic, 200 First Street SW, Rochester, MN
55905; E-mail: gajjc.ognjen@mayo.edu
Copyright © 2007 by the Society of Critical Care
Medicine and Lippincott Williams & Wilkins
DOI: 10.1097/01.CCM.0000284505.96481.24
D1 Crit Care Med 2007 Vol. 35, No. 10