Vol.2, No.2, 82-89 (2010)
doi:10.4236/health.2010.22014
SciRes
Copyright © 2010 Openly accessible at http://www.scirp.org/journal/HEALTH/
Health
Pseudomonas aeruginosa ventilator assoc iated
pneumonia: improved outcomes with earlier follow-up
Elpis Giantsou1,2, N ikolaos L iratzopoulo s1, Eleni Efraimidou 1, Konstanti nos I. Manolas1, J.
Duncan You ng2
1Intensive Care Unit, Demokritus University of Thrace, Greece
2Adult intensive Care Unit, John Radcliffe Hospital, Oxford, United Kingdom; elpisgiantsou@yahoo.com
Received 27 October 2009; revised 16 December 2009; accepted 17 December 2009.
ABSTRACT
It is not clear what is the appropriate timing to
follow-up patients with ventilator-associated
pneumonia (VAP) and Clinical Pulmonary Infe-
ction Score >6 between days 3-5 of an appro-
priate antibiotic treatment. We studied 122
patients with Pseudomonas aeruginosa VAP. A
follow-up respiratory sample was collected on
days three or five ( “day-three” and “day-five”
group ) and treatment was modified 48h later.
Molecular typing identified super-infections or
persistence. For serial data another respiratory
sample was collected, on day three from the
“day-five” group and on day five from the
“day-three” group. Sixty patients, in the “day-
three” group compared to 62 in the “day-five”
group, had reduced fourteen-day mortality
( 18.3% and 38.7%; p=0.01 ) and fewer days in
intensive care unit (17.2 ± 4.3 compared to 27.3
± 4.7, p<0.05 ). Eighteen patients of the “day-
five” group were diagnosed with super-infec
tion and 22 with persistence on day five, of
whom 14 and 19 had b een havi ng these patterns
since day three. For patients with Pseudomonas
aeruginosa VAP and Clinical Pulmonary Infe-
ction Score >6, improved fourteen-day mortality
and shorter duration of stay in health-care
facilities were observed with earlier follow-up.
Keywords: Ventilator-Associated Pneumonia;
Clinical Pulmonary Infection Score; Pseudomonas
Aeruginosa
1. INTRODUCTION
In addition to its value in diagnosing ventilator-associa-
ted pneumonia (VAP), the Clinical Pulmonary Infection
Score (CPIS) has other uses [1]. Initial values of CPIS
may guide the duration of antibiotic therapy for patients
with VAP [2], while serial measurements may identify
patients as potentially un-responsive to antibiotic therapy,
when the value of the score is above six from day three
to day five of antibiotic treatment [3]. Further invest-
tigation of patients identified as potentially
un-responsive should ensure that the administered an-
timicrobials are appropriate and that extra-pulmonary
infections and non-infectious conditions are not involved
[4,5]. When the lung remains the suspected focus of in-
fection, the next step should involve a follow-up respi-
ratory sample to investigate treatment failure [6]. How-
ever, it is not clear from the data available [3,7] what is
the appropriate timing to collect a follow-up respiratory
sample from patients with VAP, in whom CPIS remains
>6, between days three to five of antibiotic treatment.
Should the follow-up respiratory sample be collected on
day three or a little later?
The objective was to evaluate an earlier compared
with a later timing to retrieve respiratory pathogens and
determine treatment failure, for patients with initially
appropriately treated Pseudomonas aeruginosa (P. A er-
uginosa) VAP, in whom the simplified CPIS [3]
remained >6 between days 3-5 of antibiotic therapy.
2. MATERIALS AND METHODS
The study was conducted at the University Hospital of
Thrace, during a 48-month period. Patients were entered
into the study if they met all the following: clinical sus-
picion for VAP [1]; two identical positive solely for P.
aeruginosa quantitative cultures, one of tracheal aspirate
and one of bronchoalveolar lavage (BAL) (thresholds of
>106 and >104 colony–forming units/ml, respectively);
and simplified CPIS>6 [3] between days 3-5 of treat-
ment. Patients were excluded if they had received solid
organ or bone marrow transplant or had evidence of
rapid deterioration within 72hr of treatment [8]. The
diagnostic and therapeutic approach is presented in Fig-
ure 1. Initial antibiotic treatment for P. aeruginosa with
daily infusion of Amikacin (20mg/kg per day) combined
with 6h bolus administration of Piperacillin- Tazobactam
4.5gm) was instigated within 6h of bronchoscopy. (
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
83
Figure 1. Schematic presentation of the diagnostic and therapeutic approach. VAP, ventilator-associated pneumonia; BAL,
bronchoalveolar-lavage; P.aeruginosa, Pseudomonas aeruginosa; CPIS, Clinical Pulmonary Infection Score.
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
84
Figure 2. Schematic representation of the approach used for serial microbiology and molecular typing.
The follow-up BAL was collected by the endoscopy
group on days three or five as determined by the avail-
ability of service and patients shall be referred to as the
“day-three” and the “day-five” group respectively. The
follow-up BAL cultures and sensitivity tests results re-
turned to the Intensive Care Unit (ICU) 48hr later and
were used to modify the antimicrobials. The duration of
treatment was 14 days [9].
Pulsed gel electrophoresis was applied to the BAL
samples at study entry and follow-up, to identify
whether P. aeruginosa isolated at follow-up was super-
infection or persistence of the initial isolate. The DNA
fragment patterns were interpreted as genetically indis-
tinguishable, closely related or unrelated [10,11]. Iso-
lates genetically unrelated to those grown at study entry
were considered super-infections, whereas isolates ge-
netically indistinguishable or closely related, were con-
sidered persistence.
For serial microbiologic and molecular typing data
two independent investigators performed BAL on day
three at the “day-five” group and on day five at the
“day-three” group (Figure 2). For cost reasons, no
sensitivity testing was undertaken for these samples,
when P. aeruginosa was isolated, because it was
previously tested from the same source. If growth, other
than P. aeruginosa, was isolated, then patients were
excluded and sensitivity was tested. The outcomes
evaluated were mortality, SAPS II [12] and SOFA [13]
on day 14, mortality on day 28, mortality in ICU and
hospital, duration of mechanical ventilation and duration
of stay in ICU and hospital after VAP. CPIS and organ
failures [13] were assessed on day 14.
3. STATISTICS
Data were expressed as mean ± standard deviation (SD)
or as percentages of total. Continuous data were com-
pared using Student’s t-test. The chi-square test with
Yates correction for proportions was used for categorical
variables. All tests were two sided. Significance was
accepted for p<.05. Data were analyzed using SPSS 11
(SPSS, Chicago, IL).
4. RESULTS
The admission and study entry characteristics of 122
patients with P.aeruginosa VAP appear in Table 1. The
CPIS score for the “day-three” and the “day-five” group
respectively it was 7.33 ( ± 0.47 ) vs 7.32 ( ± 0.59 ),
p=0.9 on day three and 7.08 ( ± 0.27 ) vs 7.1 ( 0.42 ) ,
p=0.46 on day five .
Super-infection and persistence rates of P. aeruginosa
as revealed by the follow-up BAL are presented in Table
2 and schematically in Figure 3. For both study groups,
the strains of P. aeruginosa identified at follow-up as
super-infection or persistence at significant counts were
resistant to Piperacillin-Tazobactam and sensitive to
Meropenem and Amikacin which replaced the initial
combination of Piperacillin-Tazobactam with Amikacin.
P. aeruginosa strains persistent at insignificant counts at
follow-up remained sensitive to the initial antibiotics
which remained unchanged.
Significantly lower mortality, SAPS II and SOFA
were noted on day 14 for the “day-three” group (Table
3). Schematic presentation of mortality and length of
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
85
Table 1. Admission and study entry characteristics.
Characteristics
“day-three”
group
n=60
“day-five”
group
n=62
P value
On admission
Age, mean ( SD )
Men, n (% )
SAPS II, mean, SD
SOFA, mean, SD
Admission, n (% )
Medical
Emergency surgery
Elective surgery
Reason for MV, n ( % )
Status asthmaticus
COPD
CAP
Drug overdose
Abdominal Surgery
Other surgery
CHF
Neurological emergency
On study entry
MV before VAP, mean ( SD ), d
Antibiotics before VAP, n(%)
SAPS II, mean ( SD )
SOFA, mean ( SD )
CPIS, mean ( SD )
55.4 ( 11.9 )
39 ( 65 )
43.3 ( 5.8 )
6.7 ( 2.2 )
33 ( 55 )
14 ( 23 )
13 ( 22 )
7 ( 12 )
11 ( 18 )
9 ( 15 )
9 ( 15 )
4 ( 7 )
5 ( 8 )
5 ( 8 )
10 ( 17 )
6.9 ( 1.2 )
47 ( 78 )
45.3 ( 5.8 )
6.8 ( 2.1 )
7.68 ( 0.85 )
55.6 ( 13 )
40 ( 64 )
45.7 ( 6 )
6.5 ( 2.1 )
31 ( 50 )
16 ( 26 )
15 ( 24 )
8 ( 13 )
10 ( 16 )
7 ( 11 )
8 ( 13 )
5 ( 8 )
6 ( 10 )
6 ( 10 )
12 ( 19 )
7 (1.2 )
46 ( 74 )
46 ( 6.3 )
6.7 ( 2.3 )
7.6 ( 0.7 )
0.2
0.9
0.7
0.8
0.8
0.9
0.9
0.8
0.1
0.7
0.8
Abbreviations: SAPS, Simplified Acute Physiology Score; SOFA, Sequential Organ Failure Assessment; MV, Mechanical Ventilation; COPD,
Chronic Obstructive Pulmonary Disease; CAP, Community Acquired Pneumonia; CHF, Congestive Heart Failure; VAP, Ventilator-associated
pneumonia.
0
10
20
30
40
50
60
70
Day-3
group Day-5
group
Persista n ce at insign ific an t
c ounts on follow-up
Persis tenc e a t s ig nific a nt counts
on follow-up
Super-infe c tions on follow-up
Figure 3. Schematic presentation of P.aeruginosa patterns on follow-up.
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
86
Table 2. P.aeruginosa patterns on follow-up.
“Day- three” group on follow-upa, n=60 Patients
n ( % ) Final antibiotics
Super-infectionb
Persistencec at significant counts
Persistencec at insignificant counts
15 ( 25 )
18 ( 30 )
27 ( 45 )
M / A
M / A
PT / A
“Day- five” group on follow-upa, n=62
Super-infectionb
Persistencec at significant counts
Persistencec at insignificant counts
18 ( 29 )
22 ( 35.5 )
22 ( 35.5 )
M / A
M / A
PT / A
Abbreviations: M, Meropenem; A, Amikacin; PT,Piperacillin-Tazobactam.
aThe follow-up BAL was performed on day three of therapy for the “day-three” group and on day five of therapy for the “day-five” group.
bThese Pseudomonas aeruginosa strains were identified as super-infections, because they were genetically unrelated to the clone isolated at
study entry.
cThese Pseudomonas aeruginosa strains were identified as persistence, because they were closely related or indistinguishable to the clone
isolated at study entry.
Table 3. Study outcomes.
End Point
“day-three”
group
( n=60 )
“day-five”
group
( n=62 )
p value
Mortality on day 14, n ( % )
SAPSb on day 14
SOFAb on day 14
Mortality on day 28, n (% )
Mortality in ICU, n ( % )
Mortality in hospital, ( n % )
MV after VAP, d
ICU stay after VAP, d
Hospital stay after VAP, d
CPIS on day-14
Οrgan failure on day- 14 b, n (% )
Cardiovascular
Renal
Central nervous
Hepatic
Coagulation
11 ( 18.3 )
42.1 ( 5.8 )
5.9 ( 1.6 )
17 ( 28.3 )
16 ( 26.6 )
18 ( 30 )
14.3 (2 )
17.2 ( 4.3 )
23.1 ( 3.7 )
4.2 ( 1.7 )
19 ( 38.7 )
17 ( 34.6 )
10 ( 20.4 )
2 ( 4 )
2 ( 4 )
24 ( 38.7 )
49.5 ( 9.7 )
8.1 ( 2 )
39 ( 62.9 )
40 ( 64.5 )
42 ( 67.7 )
22.7 ( 2.6 )
27.3 ( 4.7 )
35.5 ( 4.5 )
4.4 ( 1.5 )
24 ( 63.1 )
23 ( 60.5 )
14 ( 36.8 )
4 ( 10.5 )
4 ( 10.5 )
0.01
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.59
Abbreviations: SAPS, Simplified Acute Physiology Score; SOFA, Sequential Organ Failure Assessment Score; MV, mechanical ventilation; CPIS,
Clinical Pulmonary Infection Score bFor patients alive on day 14. Values are expressed as mean ( SD ) unless otherwise indicated. Organ failures
may not sum up to hundred as some patients may have >1.
stay in health care facilities appear in Figur e 4.
Eighteen patients of the “day-five” group were diag-
nosed with super-infection with a new strain of P.
aeruginosa at follow-up on day five. Of them 14 had
been having super-infection and required treatment
change, since day three, as it was diagnosed by the in-
dependent investigators on day three (Ta b le 4 ). Five of
the 14 (35.7%) P. aeruginosa strains that were respo-
nsible for super-infections in the” day-five” group, were
present at insignificant concentrations in the BAL that
was performed initially to diagnose VAP. Similarly, for
the “day-three” group six of the 15 (40%) superinfec-
tions on day 3 were due to overgrowth of P. aeruginosa,
that was present in insignificant concentrations in the
BAL performed initially to diagnose VAP. The evolution
over time of superinfection is presented schematically in
Figure 5 for both groups.
Twenty two patients of the “day-five” group were
diagnosed at follow-up on day five with persistence at
significant counts, of the initially isolated P. aeruginosa
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
87
Mortalit y on day 14
Morta lity on d a y 2 8
Mort ality in ICU
Morta lity in ho spital
MV afte r VAP
ICU sta y a fter V AP
Hospit al stay after VAP
0
5
10
15
20
25
30
35
40
45
Da y- thre e g roup
Da y- five g roup
Figure 4. Schematic presentation of mortality and length of stay in health care facilities
Table 4. Super-infection and persistence over time.
Patients, n ( % )
“day-three”
groupa
( n=60 )
“day-five”
groupb
( n=62 )
p
Super-infections on
Day three of therapy
Day five of therapy
Persistence at significant counts on
Day three of therapy
Day five of therapy
15 ( 25 )
20 ( 33.3 )
18 ( 30 )
22 ( 36.6 )
14 ( 22.5 )
18 ( 29 )
19 ( 30.6 )
22 ( 35.4 )
0.7
0.8
0.9
0.8
aFor the “day-three” group super-infections and persistence at significant counts on day three of therapy were derived from the follow-up BAL
and on day five of therapy from the BAL collected by the independent investigators.
bFor the “day-five” group super-infections and persistence at significant counts on day three of therapy were derived from the BAL collected
by the independent investigators and on day five of therapy from the follow-up BAL.
(Table 4).Of them 19 had been having this pattern and
required treatment adjustment, since day three. Fourteen-day
mortality for patients of the “day-five” group, who had
super-infection or persistence at significant counts since
day three and in whom treatment was adjusted on day
seven, was 8 of 14 (57.1%) and 11 of 19 (58%)
respectively. By contrast, fourteen-day mortality for
patients of the “day-three” group, who had super-
infection or persistence at significant counts on day three
and in whom treatment was adjusted on day five was 3 of
15 (20%) and 4 of 18 (22%) respectively. The evolution
over time of persistence at significant counts is presented
schematically in Figure 5 for both groups.
5. DISCUSSION
In this study, improved fourteen-day mortality, severity
scores and duration of stay in ICU and hospital were
observed with earlier follow-up and re-institution of an
appropriate antibiotic regimen for patients with P.
aeruginosa VAP, who were initially appropriately treated
and in whom CPIS remained >6 six between days 3-5 of
treatment.
Few data are available, on clinical outcomes with an
earlier or a later recognition of treatment failure for pa-
tients with initially appropriately treated VAP. Our study
showed mortality of 18.3% for patients re-evaluated on
day three and 38.7% for patients re-evaluated on day five.
Montravers et al. reported 35% mortality, for patients,
who on clinical suspicion for VAP were treated appropri-
ately and on day three were microbiologically re-evaluated,
to identify and treat super-infections and persistence [7].
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
88
Figure 5. Schematic presentation of super-infection and persistence at significant counts over time
The American Thoracic Society guidelines suggest
that the earliest time point to re-assess the antibiotic
regimen is day three of treatment [9]. Although, BAL
data can optimize antibiotic therapy in VAP, this is
translated to improved clinical outcomes when the anti-
biotics administered on clinical suspicion for VAP are
adequate and BAL data become available in a timely
manner [14-17]. In our data, both groups had received an
appropriate antibiotic therapy on clinical suspicion for
VAP, which by day-three became inappropriate due to
super-infections and persistence at significant counts.
We observed improved outcomes with early recognition
and treatment of non-response to therapy. Appropriate
antibiotic therapy administered in a timely manner is
suggested to be one of the primary determinants of hos-
pital outcome [18] and our data reinforce this view.
Our study has limitations. First, we used the availabil-
ity of service to determine assignment and not randomi-
zation because we could not ensure endoscopy support
on the assigned day. This would introduce bias if the
availability correlated with aspects of clinical care. Sec-
ond, the study was necessarily un-blinded, but bias
should not have occurred from this as the primary out-
come was mortality. Third, it was conducted within a
single ICU and a relatively large number of patients
were excluded. Therefore the results cannot necessarily
be extended to other populations. Fourth, patients in the
“day-five” group had CPIS >6 from day three of therapy
and treatment were modified on day seven. However, the
CPIS provided no evidence for deterioration because it
was continuously dropping. Finally, our study was not
specifically designed to test whether an earlier follow-up
is superior to a later one. To answer this question we
need a double blind randomized trial.
Although clearly important, an accurate diagnostic
technique and an appropriate initial empirical therapy
may not be sufficient to reduce mortality in patients with
P.aeruginosa VAP [19,20]. In our data, improved morta-
lity and shorter duration of stay in health–care facilities
were observed with earlier follow-up for patients with
P.aeruginosa VAP. This finding suggest that clinicians
should have low threshold to re-sample early and if ne-
cessary to revise therapy for patients with P. aeruginosa
VAP, who failed to reduce CPIS to values below six
between days three to five of antibiotic treatment.
REFERENCES
[1] Niederman, M.S. (2005) The clinical diagnosis of venti-
lator-associated pneumonia. Respiratory Care, 50,
788-796.
[2] Singh. N., Rogers, P., Atwood, C.W., Wagener, M.M., Yu,
V.L. (2000) Short-course empiric antibiotic therapy for
patients with pulmonary infiltrates in the intensive care
E. Giantsou et al. / HEALTH 2 (2010) 82-89
SciRes Copyright © 2010 Openl y accessible at http://www.scirp.org/journal/HEALTH/
89
unit. Am J Respir Crit Care Med, 162, 505-511.
[3] Luna, L.M., Blanzaco, M., Daniel, D., Niederman, M.S.,
Matarucco, W., Baredes, N.C., Desmery, P., Palizas, F.,
Menga, G., Rios, F., Apezteguia, C. (2003) Resolution of
ventilator-associated pneumonia, Prospective evaluation
of the clinical pulmonary infection score as an early
clinical predictor of outcome. Crit Care Med, 31,
676-682.
[4] Ioanas, M., Ewig, S., Torres, A. (2003) Treatment failures
in patients with ventilator-associated pneumonia. Infect
Dis Clin N Am,17, 753-771.
[5] Wunderink, R.G. (1995) Ventilator-associated pneumonia,
Failure to respond to antibiotic therapy. Clin Chest Med,
16, 173-193.
[6] Menendez, R., Perpina, R., Torres (2005) A. Evaluation
of non-resolving and progressive pneumonia. Semin Res-
pir Infect, 18, 103-111.
[7] Montravers, P., Fagon, J.Y., Chastre, J., Lesco, M.,
Domret, M.C., Trouillet, J.L., Gibert, C. (1993) Fol-
low-up protected specimen brushes to assess treatment in
nosokomial pneumonia. Am Rev Respir Dis, 147, 38-44.
[8] Fagon, J. Y., Chastre, J., Wolff, M., Gervais, C.,
Parer-Aubas, S., Stephan, F., Similowski, T., Mercat, A.,
Diehl, J.L., Sollet, J.P., Tenaillon, A. (2000) Invasive and
Non-invasive strategies for management of suspected
ventilator-associated pneumonia. Ann Intern Med, 132,
621-30.
[9] American Thoracic Society. (2005) Infectious Diseases
Society of America, guidelines for the management of
adults with hospital-acquired, ventilator-associated and
healthcare associated pneumonia. Am J Respir Crit Care
Med, 171, 388-416.
[10] Tenover, F.C., Robert, D.A., Goering, R.V., Mickelsen,
P.A., Murray, B.E., Persing, D.H., Swaminathan, B.
(1995) Interpreting chromosomal DNA restriction frag-
ments produced by pulsed field gel electrophoresis, Cri-
teria for bactrerial strain typing. Clin Microbiol, 33(9),
2233-2239.
[11] Wassenaar, T.M. (2003) Molecular typing of pathogens.
Berl Munch Tierarz tl Wochens chr , 11 6, 447-453.
[12] Le-Gall, J.R., Lemeshow, S., Saulnier, F. (1993) A new
Simplified Acute Physiology Score Based on a European/
North American study. JAMA, 270, 2957-2963.
[13] Vincent, J.L., Moreno, R., Takala, J., Willats, S., De
Mendonca, A., Bruining, H., Reinhart, C.K. Suter PM
and Thijs LG. (1996) The SOFA (Sepsis-related Organ
Failure Assessment) score to describe organ dysfunc-
tion/failure. On behalf of the Working Group on Sep-
sis-Related Problems of the European Society of Inten-
sive Care Medicine. Intensive Care Med, 22, 707-10.
[14] Luna, CM., Vujacich, P., Niedermann, M.S., Vay, C.,
Gherardi, C., Matera, J., Jolly, E.C. (1997) Impact of
BAL data on the therapy and outcome of ventila-
tor-associated pneumonia. Chest, 111 , 676-685.
[15] Luna, C.M., Aruj, P., Niederman, M.S., Garzon, J., Violi,
D., Prignoni, A., Rios, F., Baquero, S. (2008) Appropri-
ateness and delay to initiate therapy in ventila-
tor-associated pneumonia. Eur Respir J, 27, 158-64.
[16] Giantsou, E., Liratzopoulos, N., Efraimidou, E.,
Panopoulou, M., Alepopoulou, E., Kartali-Ktenidou, S.,
Minopoulos, G., Manolas, K.I. (2007) De-escalation
therapy rates are significantly higher by bronchoalveolar
lavage than by tracheal aspirates. Intensive Care Med, 33,
1533-1540.
[17] Ioanas, M., Ewig, S., Torres, A. (2003) Treatment failures
in patients with ventilator-associated pneumonia. Infect
Dis Clin N Am, 17, 753-771.
[18] Kollef, M.H. (2007) Moving towards real-time anti-mi-
crobial management of ventilator-associated pneumonia.
Clin Infect Dis, 44, 388-390.
[19] Prince, A.S. (2002) Biofilms, antimicrobial resistance
and airway infection. N Engl J Med, 347, 847-885.
[20] Dominguez, A.A., Arango, M.V., Torres, A. (2005)
Treatment failure in patients with ventilator-associated
pneumonia. Semin Respir Crit Care Med, 27, 104-12.