Journal of Critical Care (2011) xx, xxxxxx

The impact of coagulation parameters on the outcomes

of patients with severe community-acquired pneumonia
requiring intensive care unit admission,
Jorge I.F. Salluh MD, PhD a,b,, Ligia S.C.F. Rabello MD a , Maira M. Rosolem MD a ,
Mrcio Soares MD, PhD a , Fernando A. Bozza MD, PhD c,d , Juan Carlos R. Verdeal MD e ,
Gustavo W. Mello MD e , Hugo C. Castro Faria Neto MD, PhD b ,
Jos Roberto Lapa e Silva MD, PhD f , Patrcia T. Bozza MD, PhD b
a

Intensive Care Unit and Postgraduate Program, Instituto Nacional de Cncer, Rio de Janeiro, Brazil 20230-130
Laboratory of Immunopharmacology, Instituto Oswaldo Cruz, FIOCRUZ, Rio de Janeiro, Brazil 21040-360
c
Intensive Care Unit, Instituto de Pesquisa Clnica Evandro Chagas, FIOCRUZ, Rio de Janeiro, Brazil 21040-360
d
D'or Institute of Research and Education 22281-100
e
Intensive Care Unit, Critical Care Department, Hospital Barra D'or, Rio de Janeiro, Brazil 22775-002
f
Pulmonary Diseases Department, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil 21941-913
b

Keywords:
D-Dimer;

Coagulation;
Community-acquired
pneumonia;
Sepsis;
Outcomes

Abstract
Introduction: Coagulation abnormalities are frequent in patients with severe infections. However, the
predictive value of D-dimer and of the presence of associated coagulation derangements in severe
community-acquired pneumonia (CAP) remains to be thoroughly evaluated. The aim of this study was
to investigate the predictive value of coagulation parameters in patients with severe CAP admitted to the
intensive care unit.
Methods: D-Dimer, antithrombin, International Society of Thrombosis and Hemostasis score, clinical
variables, Sequential Organ Failure Assessment (SOFA), The Acute Physiology and Chronic Health
Evaluation II (APACHE II) and the CURB-65 score were measured in the first 24 hours. Results are

Abbreviations: APACHE II, The Acute Physiology and Chronic Health Evaluation II; ATS, American Thoracic Society; AT, antithrombin;
AUROC, area under the receiver operating curve; CAP, community-acquired pneumonia; CRP, C-reactive protein; ICU, intensive care unit; ICULOS, intensive care unit length of stay; IQR, interquartile range; ISTH, International Society of Thrombosis and Hemostasis; MV, mechanical
ventilation; PSI, Pneumonia Severity Index; PT, prothrombin time; RCT, randomized clinical trial; SOFA, Sequential Organ Failure Assessment.

Conflicts of interest: The authors report no competing interests.

This study was performed at the Intensive Care Unit, Hospital Barra D'or, Rio de Janeiro, Brazil. This study is original and was not submitted to another
primary scientific journal.
Corresponding author. Instituto Nacional de Cncer; Centro de Tratamento Intensivo, 10 Andar, Praa. Cruz Vermelha, 23; 20230-130 Rio de Janeiro,
Brazil; Tel.: +55 21 2506 6120; fax: +55 21 2294 8620.
E-mail addresses: jorgesalluh@yahoo.com.br, jsalluh@inca.gov.br (J.I.F. Salluh), ligiarabello@yahoo.com.br (L.S.C.F. Rabello),
mairarosalem@gmail.com (M.M. Rosolem), marciosoaresms@yahoo.com.br (M. Soares), bozza.fernando@gmail.com (F.A. Bozza), jverdeal@gmail.com
(J.C.R. Verdeal), gustavowmello@yahoo.com.br (G.W. Mello), hcastro@ioc.fiocruz.br (H.C. Castro Faria Neto), jrlapa.ntg@terra.com.br (J.R. Lapa e Silva),
pbozza@gmail.com (P.T. Bozza).

0883-9441/$ see front matter 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcrc.2011.02.001

J.I.F. Salluh et al.

shown as median (25%-75% interquartile range). The main outcome measure was hospital mortality.
Results: Ninety patients with severe CAP admitted to the intensive care unit were evaluated. Overall
hospital mortality was 15.5%. D-Dimer levels in nonsurvivors were higher than those in survivors. In the
univariate analysis, D-dimer, SOFA, and APACHE II scores were predictors of death. The
discriminative ability of D-dimer (area under receiver operating curve = 0.75 [95% confidence interval,
0.64-0.83]; best cutoff for D-dimer was 1798 ng/mL) for in-hospital mortality was comparable with
APACHE II and SOFA and better than C-reactive protein. Moreover, the addition of D-dimer to
APACHE II or SOFA score increased the discriminative ability of both scores (area under the receiver
operating curve = 0.82 [0.72-0.89] and 0.84 [0.75-0.91], respectively).
Conclusions: D-Dimer levels are good predictors of outcome in severe CAP and may augment the
predictive ability of scoring systems as APACHE II and SOFA.
2011 Elsevier Inc. All rights reserved.

1. Introduction

2. Materials and methods

Community-acquired pneumonia (CAP) is the most

common infectious disease requiring admission to intensive
care units (ICUs) [1]. Mortality is elevated, and hospital
length of stay is long especially in the elderly and in those
requiring advanced life support measures [2]. In this
setting, the use of broad-spectrum antimicrobials and
usual supportive measures are not enough to decrease
mortality [3], possibly leaving room for the use of
adjunctive therapies. However, early and precise risk
stratification is needed to allow adequate patient selection.
Severity of illness has been used as entry criteria in clinical
trials of adjuvant therapies in sepsis [4]. Although general
ICU scoring systems and Pneumonia Severity Index (PSI)
[5] and CURB-65 [6] are usually used to predict prognosis
in severe CAP, its accuracy for predicting outcomes has
been recently challenged [2,7], and some authors consider
that they may actually be inadequate for enrolling patients
in interventional trials [8]. In such complex scenario,
biomarkers such as coagulation parameters have been
implicated with the outcomes and are surrogates of disease
severity in patients with sepsis [8-14]. To date, several
studies using natural anticoagulants to treat sepsis have
shown conflicting results regarding mortality [4-15].
Although there is strong biological rationale for the use
of anticoagulants in sepsis, a better patient selection should
be done, including subgroup more homogeneous of septic
patients [15]. Subgroup analysis of the placebo-controlled
OPTIMIST (Optimized phase 3 tifacogin in multicenter
international sepsis) trial of tifacogin in severe sepsis
revealed a trend toward benefit in patients with procalcitonin levels of 2 ng/mL or greater as well as in those with
high baseline markers of activated coagulation [16].
However, despite this evidence, the impact of coagulation
markers on clinical outcomes was not thoroughly evaluated
in patients with severe CAP requiring ICU admission
[11,14,17,18]. We hypothesize that coagulation abnormalities are frequent and associated with worse outcomes and
may provide additional prognostic information as compared
with usually used scoring systems.

2.1. Design and setting

This single-center cohort study was performed at the ICU
of Hospital Barra D'Or, Rio de Janeiro, Brazil. Hospital
Barra D'Or is a 160-bed tertiary hospital, and the medicalsurgical ICU is a 24-bed unit.
The institutional review board of Hospital Barra D'Or
approved the study (institutional review board approval
number-099/04), and informed consent was obtained from
the participants or their representatives. The study did not
interfere with patient-management decisions.

2.2. Selection of participants, data collection,

and definitions
Patients with CAP requiring ICU admission were
consecutively included between August 2005 and June
2007. The attending physician evaluated the need for ICU
admission, and severity was assessed by the presence of
acute organ dysfunctions and the CURB-65 index. Patients
with severe immunosuppression, those with previous
coagulopathy, or those using systemic anticoagulants were
excluded (Fig. 1). Demographic, clinical, laboratory, and
outcome data were prospectively collected. The Acute
Physiology and Chronic Health Evaluation (APACHE) II
score [19] and Sequential Organ Failure Assessment (SOFA)
scores [20] were routinely calculated at the ICU. Severe CAP
was diagnosed according to the American Thoracic Society
criteria [21], and the CURB-65 index was used to evaluate its
severity [6]. We also assessed the International Society of
Thrombosis and Hemostasis (ISTH) score [22] on the first
day of ICU. Sepsis was diagnosed using the criteria of Bone
and coworkers [23]. Patients were followed until death or
hospital discharge.
C-reactive protein (CRP), leukocyte count, antithrombin
activity, and D-dimer (Coagulation Analyzer, Dade Behring,
Munich, Germany) were measured on the first day of ICU
admission. The decision to treat patients with anticoagulants,
antithrombin or activated protein C was at discretion of the

Coagulation in severe pneumonia

3
data on the cohort as well as a comparison of survivors and
nonsurvivors. As expected, age, APACHE II, and SOFA
scores were higher in nonsurvivors. However, we did not
observe differences between the 2 groups for CRP levels at
ICU admission.

3.2. Coagulation abnormalities in severe CAP

Fig. 1

Flowchart of patient inclusion in the study.

assistant intensivist. The primary end point was in-hospital

death from any cause.

2.3. Data presentation and statistical analysis

Standard descriptive statistics were used. Continuous
variables were reported as mean SD or median (25%-75%
interquartile range [IQR]). Univariate analysis was used to
identify factors associated with hospital mortality. The area
under the receiver operating characteristic (AUROC) curve
was used to evaluate the ability to discriminate between
patients who survived and those who died [24]. An optimal
cutoff was obtained considering the combination of highest
sensitivity and specificity. Two-tailed P values less than .05
were considered statistically significant. The SPSS 13.0
software package (Chicago, Ill) and Prism 3.0 (GraphPad,
California, USA) were used for statistical analysis.

3. Results
3.1. Characteristics of the study population
We screened a total of 99 patients and 90 (90.9%) fulfilled
entry criteria and were enrolled in the study (Fig. 1). The
main characteristics of the study population are presented in
Table 1. Overall, 84 (93%) patients had relevant comorbidities [25]. All patients were directly admitted to the ICU from
the emergency department. Antimicrobial therapy was
prescribed in accordance with the American Thoracic
Society guidelines [21] in all 90 patients, and treatment
failures leading to changes in therapy occurred in 5 (5.5%)
patients. Activated drotrecogin-alfa was used in 9 (10%)
patients. No patient received antithrombin or full-dose
heparin during the study period. A total of 14 (15.5%)
patients died during hospitalization. Table 1 shows detailed

Coagulation abnormalities were present in most

patients. D-Dimer (1340 ng/dL [IQR, 25%-75%; 824.52163 ng/dL)] was elevated in most patients, D-dimer levels
were higher than 500 ng/dL in 86 patients (95.5%), and all
patients had D-dimer levels higher than 250 ng/dL.
Antithrombin activity (75% [57.5%-93.5%]) and was
below 70% in 36 patients (40%). The median platelet
count was 258.000/mm3 (184-425.5 103/mm), and it was
below 150.000/mm3 in only 14 patients (15.5%). Fibrinogen levels were normal in most patients (569 mg/dL
[441-738 mg/dL]), and it was higher than 500 mg/dL in 53
(58.8%) patients. Activity prothrombin time median was
13.7 seconds (12.7-15.55 seconds), and it was longer than
15 seconds in 28 (31.1%) patients. Despite the severity of
clinical presentation, no patient fulfilled criteria for overt
DIC on day 1 of ICU stay. A detailed description of the
coagulation parameters is provided in Table 1 of the
electronic supplementary material.

3.3. D-Dimer, Antithrombin levels, ISTH score,

and hospital mortality
Neither individual changes in most coagulation parameters (antithrombin, fibrinogen levels, platelet count, and
prothrombin time) nor the ISTH score on day 1 was
associated with worse outcomes (Table 1).
However, significantly higher baseline D-dimer levels
were observed in nonsurvivors (2271 g/dL [955.8-3831 g/
dL] vs 1234 g/dL [817-2020 g/dL], P = .002; Supplementary Figure 1, ESM). We performed a comparison of survival
using Kaplan-Meier survival curves, among patients with Ddimer levels below and above the calculated optimal cutoff
values obtained by the best compromise between sensitivity
and specificity (1798 g/dL). Patients with D-dimer levels
above this cutoff had significantly lower survival rates
compared with patients with levels below the cutoff (33.3%
vs 6.7%, P = .018; Fig. 2).
When we described how D-dimer performed in different
CURB score levels (2-3 vs 4-5), we did not observe
significantly different D-dimer levels (1242 [809-2066] vs
1450 [787-2680], P = .35). Other variables such as SOFA
at admission, the need for mechanical ventilation (MV),
and in-hospital mortality were evaluated. Regarding these
variables, SOFA at admission was higher in the CURB 4-5
group (P = .01), and the need for MV was also higher
(20.8% vs 50%, P = .004); however, no differences
regarding hospital mortality were observed (P = .24).

J.I.F. Salluh et al.

Table 1

Comparisons between survivors and nonsurvivors of severe CAP

Age (y)
Sex: male
APACHE II score (points)
CURB-65 (points)
PaO2/Fio2
SOFA score (points)
Positive blood cultures
Invasive MV
Septic shock at ICU admission
CRP (mg/dL)
Antithrombin (%)
D-Dimer (ng/mL)
ISTH score
ICU-LOS (d)
Hospital LOS (d)

All patients
(n = 90)

Nonsurvivors, n = 14
(15.5%)

Survivors, n = 76
(84.5%)

73.5
40
14
3
263
4
14
31
30
12.6
75
1340
2
7
12

77.5 (68-83.5)
4 (28.5%)
16 (13.7-23.5)
4 (3. -5)
182 (157-306)
7 (3.5-11.5)
5 (35.7%)
13 (92.8%)
9 (64.2%)
16.6 (5.1-25.9)
73 (46-95)
2271 (955.8-3831)
2 (2-3.25)
18 (4-53)
25.5 (4-55.5)

72.5 (54.7-83)
36 (47.3%)
13.5 (11-16)
3 (3-4)
266 (243-310)
3 (2-5)
9 (11.8%)
18 (23.6%)
21 (26.7%)
12.5 (5.2-20.5)
75 (60-91.2)
1234 (817-2020)
2 (2-3)
6 (4-9.75)
11.5 (8-17)

.24
.24
.007
.1
.013
.002
.038
b.0001
.012
.63
.49
.03
.81
.01
.12

(57.7-83)
(44.4%)
(12-16.2)
(3-4)
(235-310)
(2-6)
(15.5%)
(33.4%)
(33.3%)
(5.4-21.8)
(57.5-93.5)
(817-2209)
(2-3)
(4-11)
(8-19.25)

Results are expressed as median (25%-75% ) IQR or n (%). LOS, length of stay.

Moreover, we did perform a correlation analysis

between SOFA, APACHE II, and CRP with D-dimer levels
but could not find any significant correlations between
SOFA (r = 0.11, P = .28), APACHE II (r = 0.09, P = .4),
or CRP levels (r = 0.17, P = .09).
We have used the ROC curve analysis to evaluate the
ability to predict death from severe CAP (ROC curve is
shown in Supplementary Figure 2). Tables 2 and 3 summarize
AUROC curve and likelihood ratios for various cutoff points.
An analysis comparing the performance of different ROC
curves was performed. The AUROC of total D-dimer to
predict death was comparable with that of the APACHE II
(P = .878) and SOFA scores (P = .818). In multivariate
analysis, a base model including APACHE II score or the
SOFA score and coagulation parameters with AUROC above
0.7 was created. Age (in years) and D-dimer were forced (each

variable was forced in separately) into the final model. Only

D-dimer was selected both in the model with APACHE II and

in the model containing SOFA score. Adjusted (for age and

APACHE II) odds ratio for D-dimer was 1.07 (95%
confidence interval [CI], 1.01-1.13).
Both models, including D-dimer and APACHE II or the
SOFA score, had good discrimination (APACHE II and Ddimer: AUROC = 0.821 [95% CI, 0.727-0.894; P b .001];
SOFA and D-dimer: AUROC = 0.843 [95% CI = 0.7500.912]; P b .001). Areas under the ROC of both models were
significantly higher as compared with those of APACHE II
score, SOFA, or D-dimer alone (Table 3).

4. Discussion
In the present study, we prospectively evaluated the
frequency and impact on outcomes of early coagulation
Table 2 Baseline D-dimer levels, SOFA, and APACHE II
score thresholds to predict mortality: sensitivity, specificity,
and likelihood ratios at various cutoff levels
Cutoff

Sensitivity
(%)

Likelihood
ratio

10.5
65.8
98.7

1.04
2.3
10.86

8
77.3
98.7

1.09
3.05
17.3

27.6
61.4
98.9

1.38
1.87
10.86

D-Dimer

Fig. 2 Survival curve of patients stratified according to D-dinner

levels. P value was obtained by log-rank test.

(g/dL)
609
92.8
1798
78.5
4642
14.3
SOFA (points)
1.5
100
5.5
69.2
12
23.1
APACHE II (points)
11.5
100
14.5
71.4
27.5
14.3

Specificity
(%)

Coagulation in severe pneumonia

Table 3 Area under the AUROC for mortality prediction in

severe CAP
Variable

AUROC (95% CI)

SOFA
APACHE II
D-Dimer
Leukocyte count
CRP
ISTH score

0.77 (0.62-0.92)
0.73 (0.59 to 0.86)
0.71 (0.52 to 0.89)
0.55 (0.38-0.71)
0.54 (0.35-0.72)
0.52 (0.32 to 0.71)

.001
.006
.01
.56
.63
.83

abnormalities in 90 patients with severe CAP requiring

ICU admission. We could demonstrate that coagulation
abnormalities were frequent and that baseline D-dimer
levels were significantly higher in nonsurvivors as
compared with survivors. In addition, D-dimer levels at
ICU admission were good predictors of outcomes as
compared with usually used severity scores (APACHE II
and SOFA) and performed better than laboratory markers
as CRP.
To date, only few studies evaluated the coagulation
parameters in patients with CAP, yet most patients included
in these studies had moderate-to-severe CAP, and only a
limited percentage required ICU admission [12,13,25]. In
accordance to our findings, Milbrandt et al [18] observed in a
recent large cohort study that coagulation abnormalities were
frequent in patients with CAP requiring hospitalization.
These authors also observed that D-dimer levels were
correlated to disease severity and survival. In a single-center
study evaluating 302 adult patients admitted to emergency
department with CAP, Querol-Ribelles et al [12] examined
the relationship between D-dimer levels and outcomes. In
accordance to our findings, higher D-dimer levels were
correlated with worse clinical outcomes. However, in the
same study, the addition of D-dimer to PSI did not increase
the predictive value of PSI alone. This finding may be
ascribed to the wide range of disease severity in the study
population that included patients with mild, moderate, and
severe CAP. Likewise, Shilon et al [13] performed a study
evaluating patients with CAP at the emergency department
and also observed the association between D-dimer levels at
admission and severity of CAP.
However, the patient population in the previously
mentioned studies was composed mostly of nonICU
patients, not only resulting in a more heterogeneous
population but also precluding the analysis of its impact
on ICU scoring systems. In the present study, we
demonstrated that coagulation parameters obtained at
ICU admission are valuable for outcome prediction, and
interestingly, they may increase the predictive ability of
traditionally used scoring systems such as the APACHE II
and SOFA scores. Interestingly, D-dimer levels at
presentation were superior to CRP levels in predicting
outcomes. Studies have recently demonstrated that in
severe CAP, CRP levels play a role in predicting

outcomes especially when assessed in a dynamic or

sequential way [26].
Despite its use in recent RCT in severe sepsis [4,27],
the role of scoring systems as entry criteria for clinical
trials has been challenged [28]. The complex nature of the
scores makes it difficult to use them at the bedside, and
there is considerable interobserver variability in score
calculation [28,29]. Moreover, a potential delay for drug
administration could happen because most scores are
calculated 24 hours after ICU admission. In addition, the
use of biomarkers with relevant roles in the pathogenesis
of organ failure and mortality has been proposed [14,18].
An ideal biomarker should be easily available (eg, in the
blood or urine), be present in high concentrations in the
early phase of the disease, and help to identify high-risk
groups [30,31]. Also, the pathways in which the biomarker
is involved should be an appropriate target for adjuvant
sepsis therapies [14,31]. In our study, D-dimer was
evaluated in patients with severe CAP requiring ICU
admission and not only fulfilled all the above premises but
also presented a better performance when coupled to
severity of illness and organ failure scores. This has proved
a relatively easy approach to improve the performance of a
marker and may help to reduce the limitations of scoring
systems [10,32]. In this context, an ideal panel of tests for
determining diagnosis of severe CAP should reflect 3
components of the systemic response: infection, inflammation, and hemostasis. Perhaps, the routine inclusion of
additional tests for patients with suspect severe CAP such
as CRP, cytokines, and procalcitonin as well as coagulation markers (eg, D-dimer, prothrombin fragment 1.2, or
thrombin-antithrombin complexes) may help to identify
patients who require additional supportive care to address
these disorders, either through admission to ICU or
through initiation of additional treatment, as corticosteroids
and activated protein C [14]. This approach should be
tested in future studies.
We acknowledge that the present study has some
limitations. First, it is an observational single-center study.
In addition, the limited number of enrolled patients
precluded an extensive analysis of subgroups. Therefore,
the lack of statistical significant differences in the
comparison of the AUROC of D-dimer and APACHE II
and SOFA score may be ascribed to the relatively small
number of patients. Particular coagulation parameters as
protein C and plasminogen activator inhibitor-1 (PAI-1) or
the interaction between coagulation parameters and
cytokines was not evaluated and should be the subject
of a future study.
In conclusion, our study demonstrates that coagulation
derangements are frequently present in patients with severe
CAP at ICU admission. Moreover, these coagulation
changes may be associated with disease severity and worse
outcomes. High baseline D-dimer levels were good predictors of mortality, as compared with usually used
laboratory tests (CRP and leukocyte count) and may

6
augment the prognostic performance of clinical scores
(APACHE II, SOFA scores).
Supplementary materials related to this article can be
found online at doi:10.1016/j.jcrc.2011.02.001.

Acknowledgments
This work was supported in part by scholarships from
CNPq (Conselho Nacional de Desenvolvimento Cientfico e
Tecnolgico), FAPERJ (Fundao de Amparo Pesquisa do
Estado do Rio de Janeiro), and PRONEX. Financial support
was provided by CNPq and PRONEX (Programa de Apoio a
Ncleos de Excelncia).
Authors' contributions:
Study concept and design: Jorge Salluh, Fernando Bozza,
and Patricia Bozza. Acquisition of data: Jorge Salluh, Gustavo
Mello, and Juan Carlos Verdeal. Analysis and interpretation of
data: Jorge Salluh, Ligia Rabello, Maira Rosolem, Marcio
Soares, and Fernando Bozza. Drafting of the manuscript: Marcio
Soares, Ligia Rabello, Jorge Salluh, and Maira Rosolem.
Critical revision of the manuscript for important intellectual
content: Jorge Salluh, Jos R. Lapa e Silva, Patricia Bozza,
Hugo Castro Faria Neto, Maira Rosolem, and Ligia Rabello.
Statistical expertise: Jorge Salluh and Mrcio Soares. Study
supervision: Patricia Bozza, Hugo Castro Faria Neto, and Jorge
Salluh. Final approval of the version to be published: all authors.

References
[1] Niederman MS. Recent advances in community-acquired pneumonia:
inpatient and outpatient. Chest 2007;131:1205-15.
[2] Rello J, Rodriguez A. Severity of illness assessment for managing
community-acquired pneumonia. Intensive Care Med 2007;33:2043-4.
[3] Rodriguez A, Lisboa T, Blot S, et al. Mortality in ICU patients with
bacterial community-acquired pneumonia: when antibiotics are not
enough. Intensive Care Med 2009;35:430-8.
[4] Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of
recombinant human activated protein C for severe sepsis. N Engl J
Med 2001;344:699-709.
[5] Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify lowrisk patients with community-acquired pneumonia. N Engl J Med
1997;336:243-50.
[6] Lim WS, van der Eerden MM, Laing R, et al. Defining community
acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax 2003;58:377-82.
[7] Buising KL, Thursky KA, Black JF, et al. A prospective comparison of
severity scores for identifying patients with severe community
acquired pneumonia: reconsidering what is meant by severe
pneumonia. Thorax 2006;61:419-24.
[8] Marshall JC, Vincent JL, Fink MP, et al. Measures, markers, and
mediators: toward a staging system for clinical sepsis. A report of the
Fifth Toronto Sepsis Roundtable, Toronto, Ontario, Canada, October
25-26, 2000. Crit Care Med 2003;31:1560-7.
[9] Bozza FA, Bozza PT, Castro Faria Neto HC. Beyond sepsis pathophysiology with cytokines: what is their value as biomarkers for
disease severity? Mem Inst Oswaldo Cruz 2005;100:217-21. [Epub 2005
Jun 2014].
[10] Bozza FA, Salluh JI, Japiassu AM, et al. Cytokine profiles as
markers of disease severity in sepsis: a multiplex analysis. Crit Care
2007;11:R49.

J.I.F. Salluh et al.

[11] Marshall JC. Biomarkers of sepsis. Curr Infect Dis Rep 2006;8:351-7.
[12] Querol-Ribelles JM, Tenias JM, Grau E, et al. Plasma D-dimer levels
correlate with outcomes in patients with community-acquired
pneumonia. Chest 2004;126:1087-92.
[13] Shilon Y, Shitrit AB, Rudensky B, et al. A rapid quantitative D-dimer
assay at admission correlates with the severity of community acquired
pneumonia. Blood Coagul Fibrinolysis 2003;14:745-8.
[14] Mira JP, Max A, Burgel PR. The role of biomarkers in communityacquired pneumonia: predicting mortality and response to adjunctive
therapy. Crit Care 2008;12(Suppl 6):S5.
[15] Abraham E, Reinhart K, Opal S, et al. Efficacy and safety of tifacogin
(recombinant tissue factor pathway inhibitor) in severe sepsis: a
randomized controlled trial. JAMA 2003;290:238-47.
[16] Laterre PF, Opal SM, Abraham E, et al. A clinical evaluation
committee assessment of recombinant human tissue factor pathway
inhibitor (tifacogin) in patients with severe community-acquired
pneumonia. Crit Care 2009;13:R36.
[17] Povoa P. Serum markers in community-acquired pneumonia and
ventilator-associated pneumonia. Curr Opin Infect Dis 2008;21:
157-62.
[18] Marshall JC, Reinhart K. Biomarkers of sepsis. Crit Care Med
2009;37:2290-8.
[19] Knaus WA, Draper EA, Wagner DP, et al. APACHE II: a severity of
disease classification system. Crit Care Med 1985;13:818-29.
[20] Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related
Organ Failure Assessment) score to describe organ dysfunction/
failure. On behalf of the Working Group on Sepsis-Related Problems
of the European Society of Intensive Care Medicine. Intensive Care
Med 1996;22:707-10.
[21] Niederman MS, Mandell LA, Anzueto A, et al. Guidelines for the
management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention. Am
J Respir Crit Care Med 2001;163:1730-54.
[22] Bakhtiari K, Meijers JC, de Jonge E, et al. Prospective validation of the
International Society of Thrombosis and Haemostasis scoring system
for disseminated intravascular coagulation. Crit Care Med 2004;32:
2416-21.
[23] Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ
failure and guidelines for the use of innovative therapies in sepsis. The
ACCP/SCCM Consensus Conference Committee. American College
of Chest Physicians/Society of Critical Care Medicine. Chest
1992;101:1644-55.
[24] Hanley JA, McNeil BJ. The meaning and use of the area under a receiver
operating characteristic (ROC) curve. Radiology 1982;143:29-36.
[25] Milbrandt EB, Reade MC, Lee M, et al. Prevalence and significance of
coagulation abnormalities in community-acquired pneumonia. Mol
Med 2009;15:438-45.
[26] Coelho L, Povoa P, Almeida E, et al. Usefulness of C-reactive protein
in monitoring the severe community-acquired pneumonia clinical
course. Crit Care 2007;11:R92.
[27] Abraham E, Laterre PF, Garg R, et al. Drotrecogin alfa (activated) for
adults with severe sepsis and a low risk of death. N Engl J Med
2005;353:1332-41.
[28] Vincent JL, Opal SM, Marshall JC. Ten reasons why we should NOT
use severity scores as entry criteria for clinical trials or in our treatment
decisions. Crit Care Med 2010;38:283-7.
[29] Vieira SR, Puybasset L, Lu Q, et al. A scanographic assessment of
pulmonary morphology in acute lung injury. Significance of the lower
inflection point detected on the lung pressure-volume curve. Am J
Respir Crit Care Med 1999;159:1612-23.
[30] Salluh JI, Bozza PT. Biomarkers of sepsis: lost in translation? Crit Care
Med 2008;36:2192-4.
[31] Salluh JI, Povoa P. Biomarkers as end points in clinical trials of severe
sepsis: a garden of forking paths. Crit Care Med 2010;38:1749-51.
[32] Oberholzer A, Souza SM, Tschoeke SK, et al. Plasma cytokine
measurements augment prognostic scores as indicators of outcome in
patients with severe sepsis. Shock 2005;23:488-93.