ORIGINAL ARTICLE INFECTIOUS DISEASES

Pneumonia treated in the internal medicine department: focus on

healthcare-associated pneumonia

M. Giannella1, B. Pinilla2, J. A. Capdevila3, J. Martı́nez Alarcón4, P. Muñoz1,5, J. López Álvarez6 and E. Bouza1,5 on behalf of the
Estudio de Neumonı́a En Medicina Interna (ENEMI) study Group from the Sociedad Española de Medicina Interna (SEMI)*
1) Department of Clinical Microbiology and Infectious Diseases and 2) Department of Internal Medicine, Hospital General Universitario Gregorio Marañón, Madrid,
3) Department of Internal Medicine, Hospital de Mataró, Mataró, 4) Department of Clinical Microbiology, Hospital General Universitario de Ciudad Real, Ciudad
Real, 5) CIBER de Enfermedades Respiratorias (CIBERES), Palma de Mallorca, and 6) Department of Internal Medicine, Hospital Universitario Prı́ncipe de Asturias,
Madrid, Spain

Abstract

Patients with pneumonia treated in the internal medicine department (IMD) are often at risk of healthcare-associated pneumonia
(HCAP). The importance of HCAP is controversial. We invited physicians from 72 IMDs to report on all patients with pneumonia hos-
pitalized in their department during 2 weeks (one each in January and June 2010) to compare HCAP with community-acquired pneumo-
nia (CAP) and hospital-acquired pneumonia (HAP). We analysed 1002 episodes of pneumonia: 58.9% were CAP, 30.6% were HCAP
and 10.4% were HAP. A comparison between CAP, HCAP and HAP showed that HCAP patients were older (77, 83 and 80.5 years;
p < 0.001), had poorer functional status (Barthel 100, 30 and 65; p < 0.001) and had more risk factors for aspiration pneumonia (18, 50
and 34%; p < 0.001). The frequency of testing to establish an aetiological diagnosis was lower among HCAP patients (87, 72 and 79;
p < 0.001), as was adherence to the therapeutic recommendations of guidelines (70, 23 and 56%; p < 0.001). In-hospital mortality
increased progressively between CAP, HCAP and HAP (8, 19 and 27%; p < 0.001). Streptococcus pneumoniae was the main pathogen in
CAP and HCAP. Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus (MRSA) caused 17 and 12.3% of HCAP. In
patients with a confirmed aetiological diagnosis, the independent risk factors for pneumonia due do difficult-to-treat microorganisms
(Enterobacteriaceae, P. aeruginosa or MRSA) were HCAP, chronic obstructive pulmonary diseases and higher Port Severity Index. Our
data confirm the importance of maintaining high awareness of HCAP among patients treated in IMDs, because of the different aetiolo-
gies, therapy requirements and prognosis of this population.

Keywords: Community-acquired pneumonia, healthcare-associated pneumonia, hospital-acquired pneumonia, hospitalized pneumonia

patient, internal medicine
Original Submission: 13 September 2011; Revised Submission: 2 December 2011; Accepted: 17 December 2011
Editor: M. Paul
Article published online: 26 December 2011
Clin Microbiol Infect 2012; 18: 786–794
10.1111/j.1469-0691.2011.03757.x
coined to identify those patients who acquired pneumonia
Corresponding author: M. Giannella MD, Department of Clinical
outside the hospital but were institutionalized or had fre-
Microbiology and Infectious Diseases, Hospital General Universitario
Gregorio Marañón, Doctor Esquerdo 46, 28007 Madrid, Spain quent contact with the healthcare system for the manage-
E-mail: maddalena.giannella@libero.it ment of their chronic underlying diseases [2]. These patients
*Members of the ENEMI Study Group are listed in Appendix 1. are at risk for infections caused by nosocomial pathogens
and have a poorer outcome than patients with CAP [3–6]. It
is a matter of debate whether the poorer outcome of HCAP
Introduction
is related to the higher prevalence of nosocomial pathogens
or to patients’ underlying conditions and whether broad-
Healthcare-associated pneumonia (HCAP) is an intermediate spectrum empirical antibiotic therapy is necessary [1,7–9].
category between community-acquired pneumonia (CAP) The number of hospitalizations for pneumonia has been
and hospital-acquired pneumonia (HAP) [1]. The term was increasing in recent years, mainly among elderly patients and

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases
CMI Giannella et al. Pneumonia in internal medicine 787

those with multiple co-morbidities [10,11]. Several studies admission in patients who did not fulfill HCAP criteria were
show that these patients are frequently managed in the inter- considered CAP. Patients were classified as having HCAP if
nal medicine departments (IMDs) [12–19]. Only one study they acquired pneumonia outside the hospital but fulfilled
describes the frequency of HCAP among patients with pneu- any of the following criteria [2,20]: prior hospitalization
monia treated in the IMDs. The authors compared CAP, (‡2 days) or surgery in the past 180 days; residence in a
HCAP and HAP and showed that the baseline conditions nursing home or long-term care facility; attending hospital
and outcome of HCAP patients were more similar to those regularly because of chronic underlying diseases; intravenous
of patients with HAP than to those of patients with CAP therapy, wound care or specialized nursing care at home in
[20]. However, no data were provided on the aetiological the past 30 days; chemotherapy in the past 30 days; and hae-
pattern of the three groups. Therefore, the question of modialysis. The criteria for HAP were pneumonia after being
whether the outcome of HCAP patients treated in IMDs is hospitalized for more than 48 h or within 10 days of leaving
associated only with the underlying conditions or with the the hospital [20]. All patients were questioned about the
higher frequency of nosocomial pathogens is open to debate. presence of any HCAP or HAP criteria at admission.
We performed a large multicentre prospective study to
compare the underlying conditions, aetiological patterns, Measurements
therapeutic management and outcome of patients with CAP, Before enrolling cases in the study, participating centres
HCAP and HAP treated in the IMDs. We also investigated were asked to report data collected for the year 2009,
the factors associated with poor outcome. including number of patients admitted to the IMD, mean days
of stay, and number of patients discharged with a diagnosis
of pneumonia. This was performed to double check the
Methods
accuracy of the incidence and other clinical data obtained
during the study weeks. We also collected the number of
Study design admissions to the IMD and that of patients newly diagnosed
We performed a multicentre prospective study in which the with pneumonia.
study population was based on a list provided by the Study Demographic data and data on underlying conditions
Group for Infectious Diseases of the Spanish Society of included age, sex, Charlson co-momorbity index, Barthel
Internal Medicine (SEMI). We invited the directors of 412 index, and the presence of risk factors for aspiration pneu-
IMDs in 360 hospitals to participate by mail and email. monia such as altered consciousness, altered gag reflex, dys-
Data were collected during two study weeks (one each in phagia, severe periodontal disease or putrid sputum [4].
January and June 2010) through a web site (http://ene- We recorded the Pneumonia Severity Index (PSI) and the
mi2010.com), where the study protocol was available. Access CURB-65 (confusion, urea nitrogen, respiratory rate, blood
was restricted to participating members by means of a per- pressure, and age 65 years or older) severity score in CAP
sonal username and password. and HCAP patients.
During each study week, the coordinating investigator We recorded collection data and results for the follow-
contacted the participating members by phone and email to ing samples: blood culture, sputum, bronchial aspirate,
monitor inclusion of all eligible cases. The completeness and bronchoalveolar lavage, pulmonary biopsy, pneumococcal
consistency of the protocols were systematically reviewed and Legionella urinary antigen, serology and nasopharyngeal
before data were entered into the database. The study was swab for detection of the influenza virus. We defined
approved by our local ethics committee. pneumonia due to difficult-to-treat (DTT) microorganisms
as those episodes caused by Enterobacteriaceae, Pseudomo-
Patients nas aeruginosa or methicillin-resistant Staphylococcus aureus
We included all adults (>16 years) treated for pneumonia in (MRSA).
the IMD during the study weeks. Pneumonia was defined by Data on therapeutic management included administration
radiological evidence of a new, or progression of a previous, of empirical antibiotic treatment, time from diagnosis to first
pulmonary infiltrate plus at least two of the following crite- antibiotic dose, route of administration, adherence to IDSA
ria: fever >38C, cough, purulent sputum, dyspnoea or >20 guidelines for the management of CAP [21] and ATS guide-
breaths per minute, pleuritic chest pain, and a leucocyte lines for the management of HAP and HCAP [2]. In patients
count of >10 000/mm3 or <4000/mm3 [20]. with positive cultures, we also evaluated the adequacy of
Patients were classified as having CAP, HCAP or HAP. All empirical therapy according to the sensitivity tests of the
community-acquired episodes diagnosed within 48 h of causative agent.

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
788 Clinical Microbiology and Infection, Volume 18 Number 8, August 2012 CMI

Outcomes were evaluated using the following: time to clini- 412 IMDs belonging to 360 hospitals were invited to participate
cal stability [22]; need for intubation and days of mechanical
ventilation; development of a complication such as empyema,
72 IMDs belonging to 66 hospitals agreed to participate
septic shock and multi-organ failure; persistent bacteraemia
and/or emergence of antibiotic resistance in the causative
organism; in-hospital death; and length of hospitalization from Potential eligible patients (n = 1043)
the diagnosis of pneumonia.
Patients with no data available
(n = 12)
Statistical analysis
Categorical variables are presented as absolute numbers and
their relative frequencies. Quantitative variables are presented Patients enrolled (n = 1031)
as mean and standard deviation (SD) if normally distributed or
as median and interquartile range (IQR) if non-normally dis- Patients who did not fulfil the
criteria for pneumonia (n = 29)
tributed. We compared categorical variables between three
epidemiological groups (CAP, HCAP and HAP) using the
Pearson chi-squared and Fisher exact tests, while the non- Patients analyzed (n = 1002)
parametric Mann–Whitney and Kruskal–Wallis tests were
used to compare quantitative variables. A post hoc Bonferroni
correction was performed for pairwise comparisons. Stepwise First study week Second study week
logistic regression models were used in the multivariate analy- (January 2010) (June 2010)
(n = 685) (n = 317)
sis to analyse risk factors for pneumonia due to DTT micro-
organisms and for in-hospital mortality. Variables with p < 0.1
Incident cases* Incident cases*
in the univariate analysis were included in the multivariate (n = 612) (n = 276)
models. Differences were considered to be significant for
p < 0.05. We performed a receiver operating characteristic
(ROC) analysis to determine the sensitivity and specificity of FIG. 1. Study flow diagram. *Incident cases refer to patients newly
the model for in-hospital mortality. The analysis was carried diagnosed with pneumonia during the study weeks.
out using SPSS 13.0 (SPSS, Chicago, IL, USA).
Comparison between CAP, HCAP and HAP
Community-acquired pneumonia was diagnosed in 58.9% of
Results
patients, HCAP in 30.6%, and HAP in 10.5%. The most com-
mon inclusion criteria for HCAP patients are shown in Table 1.
Hospital characteristics Table 2 shows the comparison of the three pneumonia
The study flow diagram is shown in Fig. 1. Seventy-two IMDs groups. HCAP patients were older and had more co-morbid-
belonging to 66 Spanish hospitals agreed to participate. Of ities than CAP patients. They also had a poorer functional
these, 47% were teaching institutions and 94% public institu- status and more frequent cerebrovascular disease, dementia
tions, with only one private facility and three mixed (public- and risk factors for aspiration pneumonia than HAP patients.
private) centres. The overall catchment population was Cough and pleuritic chest pain were more common among
16 041 888 inhabitants (range, 35 868–1 100 000 inhabitants patients with CAP, while dyspnoea and altered mental status
per hospital). The total number of beds in the institutions
TABLE 1. Inclusion criteria for the HCAP group
was 28 896 (range, 82–1559 beds).
n = 307 (%)a

Incidence of pneumonia Hospitalization in the past 180 days 170 (55.4)

Residence in a nursing home or long-term care facility 169 (55)
A retrospective survey of the year 2009 showed an inci- Attending hospital regularly because of chronic underlying diseases 169 (55)
dence of pneumonia in IMDs of 117 episodes per 1000 Wound care or specialized nursing care in the past 30 days 49 (16)
Intravenous therapy at home in the past 30 days 38 (12.4)
admissions. The incidence of pneumonia during the study Surgery in the past 180 days 19 (6.2)
Chemotherapy in the past 30 days 11 (3.6)
weeks was 111 episodes per 1000 admissions, ranging from Undergoing haemodialysis 0
141 episodes per 1000 admissions in the first week (January HCAP, healthcare-associated pneumonia.
a
The numbers and percentages add up to more than the total, as many patients
2010) to 75.3 episodes per 1000 admissions in the second presented more than one HCAP criterion.
week (June 2010).

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
CMI Giannella et al. Pneumonia in internal medicine 789

TABLE 2. Comparison of underly-

CAP, n = 591 (%) HCAP, n = 307 (%) HAP, n = 104 (%) p
ing conditions, clinical manifesta-
tions, therapeutic management Demographic data
Age, years (median, IQR) 77, 65–84 83, 76–89 80.5, 73–85.7 <0.001a,b
and outcome of patients cared for Male sex 371 (62.8) 177 (57.7) 68 (65.4) 0.22

in internal medicine departments Co-morbid conditions

COPD 221 (37.4) 125 (40.7) 49 (47) 0.15
Congestive heart failure 126 (21.3) 88 (28.7) 38 (36.5) 0.001b
Diabetes mellitus 149 (25.2) 94 (30.6) 33 (31.7) 0.13
Cerebrovascular disease 106 (17.9) 115 (37.5) 24 (23) <0.001a,c
Dementia 98 (16.6) 150 (48.9) 28 (26.9) <0.001a,c
Cancer 77 (13) 69 (22.5) 27 (26) <0.001a,b
Chronic renal failure 78 (13.2) 66 (21.5) 20 (19.2) 0.004a
Chronic liver failure 49 (8.3) 30 (9.8) 13 (12.5) 0.35
Autoimmune diseases 21 (3.6) 15 (4.9) 5 (4.8) 0.57
HIV infection 22 (3.7) 4 (1.3) 4 (3.8) 0.09
Charlson score (median, IQR) 5, 4–7 7, 6–10 7, 5–10 <0.001a,b
Barthel index (median, IQR) 100, 60–100 30, 0–80 65, 20–100 <0.001a,b,c
RF for aspiration pneumonia 108 (18.3) 154 (50.2) 35 (33.7) <0.001a,b,c

Manifestations of pneumonia
Fever 349 (59) 157 (51) 66 (63.5) 0.03
Cough 480 (81.2) 222 (72.3) 67 (64.4) <0.001a,b
Purulent sputum 342 (57.9) 150 (48.9) 54 (51.9) 0.03
Dyspnoea 438 (74) 253 (82.4) 92 (88.5) <0.001a,b
Pleuritic chest pain 158 (26.7) 33 (10.7) 13 (12.5) <0.001a,b
Altered mental status 115 (19.5) 142 (46.3) 45 (43.3) <0.001a,b
Leukocytosis 434 (73.4) 199 (64.8) 73 (70.2) 0.03a
Leukopenia 24 (4.1) 16 (5.2) 5 (4.8) 0.73
Multilobar infiltrates 150 (25.4) 85 (27.7) 39 (37.5) 0.04
Cavitation 7 (1.2) 5 (1.6) 0 0.46
Interstitial infiltrate 62 (10.5) 39 (12.7) 10 (9.6) 0.55
Pleural effusion 120 (20.3) 54 (17.6) 27 (26) 0.17

Therapeutic management
Empirical therapy 588 (99.5) 302 (98.4) 102 (98) 0.10
Administration within 6 h 420 (71) 211 (68.7) 74 (71.2) 0.79
Adherence to guidelinesd 413 (69.9) 70 (22.8) 58 (55.8) <0.001a,b,c
Adequacy according to sensitivity 54/68 (79.4) 29/45 (64.4) 15/20 (75) 0.27
Days of therapy (median, IQR) 11, 9–14 11, 9–15 12, 9–15 0.79

Outcome
>1 week to clinical stability 89 (15) 75 (24.4) 28 (27) <0.001a,b
Empyema 17 (2.9) 3 (1) 0 0.05
Septic shock 46 (7.8) 45 (14.7) 24 (23.1) 0.001a,b
Need for intubation 20 (3.4) 10 (3.3) 9 (8.7) 0.05
Multi-organ failure 33 (5.6) 34 (11) 21 (20.2) <0.001a,b
Persistent bacteraemia 1 (0.2) 4 (1.3) 2 (1.9) 0.02
Emergence of MDR 4 (0.7) 11 (3.6) 3 (2.9) <0.004a
In-hospital mortality 46 (7.8) 58 (18.9) 28 (26.9) <0.001a,b
Days of hospital staye (median, IQR) 8, 5–13 9, 6–14 11, 6–16 0.002b

CAP, community-acquired pneumonia; HCAP, healthcare-associated pneumonia; HAP, hospital-acquired pneumonia;

SD, standard deviation; COPD, chronic obstructive pulmonary disease; IQR, interquartile range; RF, risk factors;
MDR, multidrug resistance.
a
P < 0.01 for comparison between CAP and HCAP.
b
P < 0.01 for comparison between CAP and HAP.
c
P < 0.01 for comparison between HCAP and HAP.
d
Adherence to guidelines was evaluated according to the IDSA 2007 guidelines for patients with CAP and according
to the ATS 2005 guidelines for those with HCAP and HAP.
e
Days of hospital stay were computed from the day pneumonia was diagnosed.

were seen more frequently among HCAP and HAP patients. (IQR) after diagnosis was, respectively, 8 (5–13), 9 (6–14)
In CAP patients, adherence to IDSA guidelines on empirical and 11 (6–16) days (p = 0.002), and in-hospital mortality was
antibiotic treatment was 69.9%, while in HCAP and HAP 7.8, 18.9 and 26.9% (p < 0.001).
patients adherence to ATS guidelines was 22.8 and 55.8%, The median PSI and CURB-65 scores of HCAP patients
respectively (p < 0.001). Among patients with positive cul- were higher than those of CAP patients (Table 3).
tures, the adequacy of empirical therapy according to sensi-
tivity tests was lower in HCAP cases, but differences were Use of microbiological resources
not significant. Complications such as septic shock, An aetiological diagnosis (one or more samples sent to the
multi-organ failure, persistent bacteraemia and emergence of microbiology laboratory) was attempted in 86.8, 72 and
multidrug resistance in the causative pathogens were more 78.8% of CAP, HCAP and HAP patients, respectively
common among HCAP and HAP patients than among CAP (p < 0.001). Overall, 34.4% of patients were already on anti-
patients. In CAP, HCAP and HAP, the median length of stay microbial therapy at the time of sample collection.

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
790 Clinical Microbiology and Infection, Volume 18 Number 8, August 2012 CMI

TABLE 3. Comparison of the PSI and CURB-65 scores TABLE 4. Distribution of causative agents by pneumonia

Total, CAP, HCAP, group

n = 898 (%) n = 591 (%) n = 307 (%) p
CAP, HCAP, HAP,
n = 148 (%) n = 65 (%) n = 21 (%) P
PSI
Median, IQR 123, 92–152 112.5, 83–139 142.5, 116–169 <0.001
Class I or II 104 (11.6) 92 (15.5) 12 (4) Virus
Class III 109 (12) 95 (16) 14 (5) Influenza A (H1N1)v 2 (1.4) 0 0 1
Class IV 306 (34) 213 (36) 93 (30.3)
Class V 375 (42) 188 (32) 187 (61) Gram-positive
S. pneumoniae 94 (63.5) 23 (38.5) 3 (14.3) <0.001a,b
CURB-65 MRSA 1 (0.7) 8 (12.3) 2 (9.5) 0.003a
Median, IQR 2, 1–3 2, 1–3 3, 2–3 <0.001 MSSA 1 (0.7) 1 (1.5) 2 (9.5) 0.03
Class 0 100 (11.2) 86 (14.5) 14 (4.6)
Class 1 178 (20.2) 142 (24) 36 (11.8) Gram-negative
Class 2 306 (34.2) 206 (35) 100 (32.5) Enterobacteriaceae 17 (11.5) 8 (12.3) 5 (23.8) 0.26
Class 3 190 (21.2) 105 (18) 85 (27.5) P. aeruginosa 5 (3.4) 11 (16.9) 6 (28.6) <0.001a,b
Class 4 or 5 120 (13.2) 49 (8) 71 (23.3) L. pneumophila 11 (7.4) 2 (3.1) 0 0.34
H. influenzae 5 (3.4) 3 (4.6) 0 0.73
CAP, community-acquired pneumonia; HCAP, healthcare-associated pneumonia; A. baumannii 0 0 2 (9.5) 0.008
PSI, Port severity index; CURB-65, confusion-urea-respiratory rate-blood pres-
sure and age 65 years or older; IQR, interquartile range. Other
M. tuberculosis 5 (3.4) 2 (3.1) 0 1
P. jiroveci 1 (0.7) 0 1 (4.8) 0.24
Polymicrobiald 3 (2) 3 (4.6) 0 0.41
Othere 3 (2) 2 (3.1) 0 0.77
Positive culture 68 (45.9) 45 (69.2) 20 (95.2) <0.001a,b
An aetiological diagnosis was reached in 234 of the 816 Positive UAT 76 (51.4) 19 (29.2) 1 (4.8) <0.001a,b
patients (28.7%) from whom one or more microbiological CAP, community-acquired pneumonia; HCAP, healthcare-associated pneumonia;
samples were collected. Of these patients, 133 (56.8%) had HAP, hospital-acquired pneumonia; MRSA, methicillin-resistant Staphylococcus
aureus; MSSA, methicillin-sensitive S. aureus; UAT, urinary antigen tests for Legio-
positive cultures; in the remaining cases, the diagnosis was nella and S. pneumoniae.
a
P < 0.01 for comparison between CAP and HCAP.
made by urinary antigen tests in 96 (41%), serology in three b
P < 0.01 for comparison between CAP and HAP.
c
P < 0.01 for comparison between HCAP and HAP.
(1.3%) and influenza virus test in two (0.8%). d
Polymicrobial included three cases of S. pneumoniae coinfection with influenza
A (H1N1)v virus, MRSA and M. tuberculosis each, two cases of MSSA coinfection
with Enterobacteriaceae, and one case of P. aeruginosa and Nocardia spp. coin-
Aetiology fection.
e
Other included one case of varicella zoster virus infection, one of C. burnetii,
The distribution of causative agents by pneumonia group is one of C. pneumoniae, one of B. fragilis, and one case of empyema with isolation
shown in Table 4. Streptococcus pneumoniae was the main of Peptostreptococcus spp. in pleural fluid.

pathogen in CAP and HCAP cases. However, P. aeruginosa

and MRSA were isolated in 17 and 12.3% of HCAP episodes,
respectively. P. aeruginosa and Enterobacteriaceae were the TABLE 5. Multivariate analysis of risk factors for in-hospital
main pathogens in HAP cases. mortality
For the analysis of the risk factors for pneumonia due to
Adjusted odds
DTT microorganisms (Table A1 see appendix), patients with ratio (95% CI) p

detection of viral pneumonia (3), M. tuberculosis (7), P. jiroveci Age 1.01 (0.99–1.03) 0.24
(2) and C. burnetii (1) were excluded. Multivariate analysis Charlson co-morbidity index 1.13 (1.05–1.22) <0.001
Barthel index 0.99 (0.98–1.01) 0.39
showed that history of chronic obstructive pulmonary dis- Risk factors for aspiration pneumonia 0.87 (0.48–1.57) 0.66
HCAP 1.01 (0.59–1.72) 0.96
ease (COPD) (OR, 2.42; 95% CI, 1.18–4.97; p = 0.02), HAP 2.27 (1.26–4.10) 0.006
Dyspnoea 2.59 (1.26–5.33) 0.009
HCAP (OR, 3.48; 95% CI, 1.69–7.15; p = 0.001) and higher Altered mental status 4.24 (2.55–7.04) <0.001
PSI (OR, 1.01; 95% CI, 1.00–1.02; p = 0.001) were indepen- Multilobar infiltrates 2.96 (1.94–4.52) <0.001
Diagnostic testing for aetiological diagnosis 0.59 (0.37–0.94) 0.03
dent risk factors for pneumonia due to DTT microorgan- Adherence to guidelines 0.67 (0.42–1.08) 0.10
isms. Receiver operating characteristic (ROC) analysis shows: area under curve
(AUC), 0.86; 95% confidence interval (CI), 0.82–0.89. Sensitivity 80%, specificity
77%.
Risk factors for in-hospital mortality CI, confidence interval; HCAP, healthcare-associated pneumonia; HAP, hospital-
acquired pneumonia.
The risk factors for in-hospital mortality according to the
univariate analysis are shown in Table A2 (see Appendix).
Multivariate analysis showed that higher Charlson co-morbid-
Discussion
ity index, HAP, altered mental status, dyspnoea and multilo-
bar infiltrates were independently associated with in-hospital
mortality, while performance of diagnostic testing was a pro- We showed that pneumonia generates a considerable
tective factor (Table 5). The model had a high predictive workload for IMDs and that HCAP represents one-third of
capacity: sensitivity 80% and specificity 77%. cases of pneumonia cared for by internists. HCAP patients

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
CMI Giannella et al. Pneumonia in internal medicine 791

were older and had a poorer functional status than CAP portion of P. aeruginosa, MRSA and Enterobacteriaceae in
and HAP patients. The prevalence of difficult-to-treat that group. Furthermore, HCAP was one of the independent
microorganisms among HCAP patients was high, and the risk factors for DTT microorganisms (P. aeruginosa, MRSA or
infections they caused were frequently inadequately treated Enterobacteriaceae).
at admission. A progressive increase in poor outcome was Over the last decade, several national and international
observed between CAP, HCAP and HAP. Higher Charlson guidelines have been developed to improve the therapeutic
co-morbidity index, severe pneumonia manifestations and management of pneumonia. In CAP patients, adherence to
HAP were independently associated with poor outcome, empirical therapy guidelines ranged from 33 to 80%
while performing a microbiological work-up improved sur- [14,20,32]. The only data available for HCAP are those
vival. reported by Venditti et al. [20], who showed 26.7% adher-
Pneumonia is an important cause of hospitalization, and ence to the 2005 ATS guidelines. We confirmed this finding
24–75% of patients hospitalized for pneumonia are treated in our study (23%). These results could indicate low aware-
by internists [12–19,23]. Our incidence of 11.1 episodes per ness of the potential of DTT microorganisms to cause
100 admitted patients represents a high workload and is HCAP.
almost identical to that obtained by Venditti et al. [20]. In Our study is subject to a series of limitations. First, it is
Spain and Italy, variations between wintertime figures (12, an observational study of two cohorts of patients prospec-
14%) and summertime figures (5, 7%) are similar. tively included during 1 week each. The small inclusion
The type of acquisition should be assessed, because the period could have affected incidence. However, the choice
importance of HCAP is not well known. One-third of the of the winter and summer periods minimized this limita-
patients in our study fulfilled the criteria for HCAP. Our tion. Indeed the incidence obtained by the retrospective
data confirm the importance of maintaining a high awareness survey of the year 2009 was similar to that found during
of HCAP due to the different underlying conditions, therapy the study period. Second, heterogeneity of the participating
requirements and prognosis of this population [1,3–5, hospitals means that microbiology work-ups and therapeu-
8,9,20,24,25]. tic management vary. Diagnostic testing (83, 81.5, 73.4%;
Diagnostic testing is mandatory for CAP patients who p = 0.05), initiation of antibiotics within 6 h (77, 67, 65%;
require hospitalization, as it is for those with HCAP or HAP p = 0.003) and adherence to guidelines (61, 51, 54%;
for epidemiological reasons and because of the risk of multi- p = 0.02) were more frequent in hospitals with <500 beds
drug-resistant pathogens and the potential for de-escalation than in those with 500 to 1000 or >1000 beds, respec-
[1,2]. Furthermore, the microbiological study has been asso- tively. Finally, the number of patients with a known aetiol-
ciated with reduced mortality in patients hospitalized with ogy was low, thus preventing analysis of the impact of
CAP [26]. We were able to confirm this finding, as the aetiology on outcome.
microbiological work-up proved to be an independent pro- In conclusion, our data confirm the importance of main-
tective factor for in-hospital mortality in our study. taining high awareness of HCAP, given the different aetiolo-
The pathogens involved in CAP and HAP in our study gies, therapeutic needs and prognosis of this condition
population were as expected and reflect findings reported in among patients treated in IMDs.
the literature [27–29], with the exception of Enterobacteria-
ceae among CAP patients. COPD, higher median age and
Acknowledgements
Charlson’s score were associated with Enterobacteriaceae in
our CAP patients (data not shown). The aetiology of HCAP
varies, and some centres find pathogens more similar to We would like to thank Thomas O’Boyle for his help with
those causing CAP [4,9,30], while others report microorgan- the preparation of the manuscript. We are grateful to Cristi-
isms more similar to those causing HAP [3,5,6,31]. In previ- na Fernandez and Luis Alcalá for their help with the statisti-
ous studies the prevalence of multidrug-resistant cal analysis.
microorganisms varied with the study population and was
lower when no immunosuppressed patients attended by
Transparency Declaration
emergency departments were included [4,9,30] and higher if
hospitalized patients with any underlying conditions were
analysed [3,5,6,31]. We were able to confirm this finding. This study does not present any conflict of interests for the
Despite the fact that S. pneumoniae was the main causative authors. This study was partially supported by ‘Fondo de
agent in our HCAP population (38.5%), we found a high pro- Investigación Sanitaria’ (FIS) PSI 09/01257, by ‘RED Española

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
792 Clinical Microbiology and Infection, Volume 18 Number 8, August 2012 CMI

de Investigación en Patologı́a Infecciosa’ (REIPI) (RD06/0008/ López González-Cobos, Eduardo Oliveros Acebes, Marı́a Jes-
1025) and by the Programa de Centros de Investigación Bio- ús Granda Martı́n, Belén Mora Hernández; Hospital Morales
médica en Red (CIBER) de Enfermedades Respiratorias Meseguer, Murcia: Faustino Herrero Huerta, Gandia Herrero
CB06/06/0058 Marina; Hospital Universitario de Canarias, San Cristóbal de
La Laguna: Francisco Santolaria Fernández, José Juan Viña
Rodrı́guez; Hospital Universitario San Joan de Reus (URV),
Appendix 1
Reus: Antoni Castro Salomó, Simona Iftimie; Hospital de Par-
la, Parla: Gonzalo Garcı́a de Casadola, José Luis Pérez Quero;
Members of the ENEMI Study Group Consorci Sanitari de l’Anoia, Igualada: Pere Comas Casanova,
Hospital Universitario Prı́ncipe de Asturias, Alcalá de Hen- Juan Valencia Molina; Hospital Universitario Reina Sofı́a, Cór-
ares: Joaquı́n López Álvarez, Eduardo Montero Ruiz; Hospital doba: José López Miranda, Rafael Martı́nez Fernández; Hospi-
Montecelo, Pontevedra: José Marı́a de Lis Muñoz, Carmen tal Militar Central Gómez ulla, Madrid: Carmelo Perea Perea,
Cano Narrillos; Complejo Hospitalario Arquitecto Marcide- Carmen González Hernández, Paloma Lucena Calvet; Hospi-
Novoa Santos, Área Sanitária de Ferrol, A Coruña: Pascual tal Universitai de Bellvitge: Ramón Pujol, Carlos Ferre Losa;
Sesma Sánchez, Hortensia Álvarez Dı́az; Hospital General Hospital Comarcal de Vinarós, Vinarós: Francisco Cabades
Virgen de la Luz, Cuenca: Marı́a Paloma Geijo Martı́nez; Hos- O-Callaghan, Manuel Arnal Babiloni; Hospital de Sant Jaume,
pital de Leganes, Leganes: Juan José Jusdado Ruiz-Capilla, Olot: José Bisbe Company; Hospital de Barbastro, Barbastro:
Daniel Ferreiro López; Hospital Miguel Servet, Zaragoza: Juan Jesús Javier González Igual, Elena Castellar Otin; Hospital Co-
Manuel Garcı́a Lechuz; Hospital Universitario Puerta de Hier- marcal de Laredo, Laredo: Miguel Carrascosa Porras, José
ro, Madrid: Valentı́n Cuervas-Mons Martı́nez, Pedro Durán Antonio Saiz-Quevedo Garcı́a; Complejo Hospitalario de So-
del Campo, Alberto Roldán Montaud; Hospital Nuestra Señ- ria, Hospital Santa Bárbara: Valentı́n Del Villar Sordo, Mario
ora de Aranzazu, San Sebastián: Francisco Javier Vivanco Del Valle Sánchez; Hospital Ramón y Cajal, Madrid: José Pe-
Martı́nez, Amaya Egoluz Pinedo, Katixa Leizaola; Althaia Hos- rales Rodrı́guez, Sergio Diz Fariña; Hospital Santa Bárbara,
pital de Manresa, Manresa: Rafael Pérez Vidal, Omar El Bout- Puertollano: Carlos Pereda Ugarte, Filomena Ceres Alabau;
rouki; Hospital Comarcal de la Axarquia de Vélez-Málaga: Complejo Hospitalario Virgen de la Victoria, Málaga: Pedro
Carlos Mª de San Román Terán, Sonsoles Fernández Sepúlve- González Santos, Juan Luis Carrillo Linares; Complejo Hospi-
da, Emilio Santı́n Piñero; Hospital General de Albacete, Al- talario la Mancha Centro, Alcazar de San Juan: José Barbera
bacete: Javier Solera Santos, Mª Elena de Tomas Labat; Farre; Hospital de L¢Esperit Sant, Santa Coloma de Grama-
Hospital del S.V.S Vega Baja, Orihuela: Juan Custardoy Olav- net: Miguel Torres Salinas, Luis Miguel Sarmiento Méndez,
arrieta, José Manuel Murcia Zaragoza; Hospital de la Marina Marı́a Sánchez Torres; Hospital Municipal de Badalona, Bada-
Baixa, Villajoyosa: Francisco Pasquau Liaño, Concepción Am- lona: Joaquı́n Vila Planas, Jordi Grau Amorós; Hospital Santa
ador Prous; Hospital de la Santa Creu i Sant Pau, Barcelona: Caterina, Salt: Joan Colomer Paires, Cristina Soler Ferrer;
Jordi Casademont Pou, Jordi Martı́n Marcuello; Consorci Hospital Santiago Apóstol, Miranda de Ebro: Agustı́n Zam-
Hospitalari del Parc Taulı́, Sabadell: Joaquı́n Oristrell Salva, arrón Moreno, Laura Hurtado Carrillo; Complexo Hospitala-
Susana Herranz Martı́nez; Hospital General Universitario de rio de Ourense: Mª Dolores Dı́az López, Ricardo Fernánder
Murcia, Murcia: José Soriano Palao, Francisco de Asis Sarabia Rodrı́guez; Hospital Clı́nico San Carlos, Madrid: Elpidio Calvo
Marco; Hospital Santos Reyes, Aranda de Duero: Pedro Can- Manuel, Cesar Augusto Henrı́quez Camacho, Arturo Fernán-
celo Suárez, Pedro Luis Álvarez Álvarez; Hospital de Fuenlab- dez Cruz, Cristina Gómez Sánchez Biezma, Baltasar Orejas
rada, Fuenlabrada: Antonio Zapatero Gaviria, Juan Hinojosa González, Gema Fresco Navacerrada, Vanesa López Peláez;
Mena-Bernal; Hospital de Mataró, Consorci Sanitari del Hospital Francesc de Borja, Gandı́a: Salvador Bellver, Carlos
Meresme, Mataró: Peter Reth Fuster, Sandra Milena Bacca Tornero Estebanez; Hospital Universitario de Valladolid,
Camacho; Hospital Universitario de la Princesa, Madrid: Car- Valladolid: Antonio Jimeno Carruez, Luis Ángel Sánchez Mu-
men Suárez Fernández, Laura Prosper Ramos, Carmen Sáez ñoz; Hospital de Tortosa Verge de la Cinta, Tortosa: Domin-
Bejar; Hospital Universitario La Fe, Valencia: Eva Calabuig; go Bofill Montoro, Mercé Cardona Ribera; Hospital Virgen de
Hospital Mutua de Terrasa, Terrasa: Javier Garau, Esther Cal- la Salud, Toledo: Andrés Tutor Martı́nez, Marı́a Antonia Sep-
bo, Vanesa Pascual; Hospital General Universitario Gregorio úlveda Berrocal; Hospital Tres Culturas, Toledo: Julio Gon-
Marañón, Madrid: Luis Audibert, Eleonora Bunsow, Jesús zález Moraleja; Hospital General de Vic, Vic: Joan Brugues
Garcı́a Castaño, Elena Trigo Esteban, Marı́a Gómez Antúnez, Terradellas, Esteban Reynaga Sosa; Hospital Dr. Peset, Valen-
Naiara Aldezábal Polo, José Santiago Filgueira Rubio, Olga cia: Arturo Artero Mora, Francesc Puchades Gimeno; Hospi-
López Berastegui, Miguel Ángel Artacho Rodrı́guez, Cristina tal Universitario Infanta Sofia, San Sebastian de los Reyes:

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
CMI Giannella et al. Pneumonia in internal medicine 793

Jorge Francisco Gómez Cerezo, Rafael del Castillo Cantero, Table A2. Risk factors for in-hospital mortality
Lourdes Mancebo Aragoneses; Hospital de Alcañiz, Alcañiz:
Eduardo Aguilar Cortés, Marı́a Ruiz Mariscal, Carmen Piqu- Non-survivors, Survivors, p
n = 132 (%) n = 870 (%)
eras Serrano; Hospital San Rafael, Madrid: Juan L. Rodrı́guez Demographic
Calderón, Pablo Robles Ruiz; Hospital Reina Sofı́a, Tudela: Age, mean (SD) (years) 82.8 (11.3) 74 (16.5) <0.001
Male sex 81 (61.4) 535 (61.5) 1
Ángel Samperiz Legarre; Hospital Cantoblanco-La Paz,
Co-morbid conditions
Madrid: Raquel Carrillo Gómez, Arturo Noguerado Asensio; COPD 50 (37.9) 345 (39.7) 0.77
Congestive heart failure 36 (26.3) 216 (24.8) 0.6
Hospital Nuestra Señora del Prado, Talavera de la Reina: Fer- Diabetes mellitus 15 (11.4) 240 (27.6) 1
nando Marcos Sánchez, Elena Nuñez Cuerda; Hospital de Cerebrovascular disease 42 (31.8) 202 (23.2) 0.04
Dementia 69 (52.3) 206 (23.7) <0.001
Cruces, Barakaldo: Victoria Egurbide Arberas, Ramiro de la Cancer 32 (24.2) 141 (16.2) 0.02
Chronic renal failure 36 (26.3) 128 (14.7) 0.001
Prieta López; Hospital General de Requena, Requena: Pilar Chronic liver failure 15 (11.4) 77 (8.9) 0.33
Román Sánchez; Hospital Comarcal Valdeorras, Barco de Val- Autoimmune diseases 6 (4.5) 35 (4) 0.81
HIV infection 4 (3) 26 (3) 1
deorras: Josep Masferrer Serra, José Felipe Morales Martı́n; Charlson score (median, IQR) 8, 7–10 6, 4–8 <0.001
Barthel index (median, IQR) 20, 0–65 90, 35–100 <0.001
Hospital de Hellı́n, Hellı́n: José Luis Beato Pérez; Hospital Co- RF for aspiration pneumonia 75 (56.8) 222 (25.5) <0.001
marcal de la Selva, Blanes: Rita Massa Puig, Nuria Costa Ro- Pneumonia group
sell; Hospital Comarcal del Noroeste de la Región, Caravaca CAP 47 (35.6) 546 (62.8) <0.001
HCAP 53 (43.2) 248 (28.5) 0.001
de la Cruz: Miguel de Paco Moya, Juan Ramón Sánchez Llin- HAP 28 (21.2) 76 (8.7) <0.001

ares; Parc Sanitari Sant Joan de Déu, Hospital General, Sant Pneumonia manifestations and risk stratification
Fever 67 (50.8) 505 (58) 0.13
Boi de Llobregat: Francisco José Castro Bohórquez. Cough 95 (72) 674 (77.5) 0.18
Purulent sputum 63 (47.7) 483 (55.5) 0.11
Dyspnoea 122 (92.4) 661 (76) <0.001
Pleuritic chest pain 13 (9.8) 191 (22) 0.001
Appendix 2 Altered mental status 93 (70.5) 209 (24) <0.001
Leukocytosis 87 (65.9) 619 (71) 0.22
Leukopenia 8 (6.1) 37 (4.3) 0.36
Multilobar infiltrates 69 (52.3) 205 (23.6) <0.001
Table A1. Risk factors for pneumonia due to difficult- Cavitation 2 (1.5) 10 (1.1) 0.66
Interstitial infiltrate 19 (14.4) 92 (10.6) 0.23
to-treat (DTT) microorganisms Pleural effusion 32 (24.2) 169 (19.4) 0.20
PSI class (median, IQR) IV, IV–IV III, II–IV <0.001
CURB-65 class (median, IQR) 3, 2–4 2, 1–3 <0.001
Severe IDSA/ATS score 99 (75) 249 (28.6) <0.001
DDT Non-DDT p
microorganisms microorganisms
Aetiology
n = 68 (%) n = 153 (%)
Diagnostic testing 86 (65.2) 730 (83.9) <0.001
Demographic
S. pneumoniae 13/30 (43.3) 109/204 (53.4) 0.33
Age, (median, IQR) (years) 79, 74–86 75, 57–84 <0.001
Enterobacteriaceae 7/30 (23.3) 23/204 (11.3) 0.08
Male sex 50 (73.5) 89 (58.2) 0.03
P. aeruginosa 3/30 (10) 19/204 (9.3) 1
L. pneumophila 0 13/204 (6.4) 0.38
Co-morbid conditions
MSSA 1/30 (3.3) 3/204 (1.5) 0.42
COPD 39 (57.4) 53 (34.6) 0.002
MRSA 2/30 (6.7) 9/204 (4.4) 0.64
Congestive heart failure 28 (41.2) 27 (17.6) <0.001
Polymicrobial 2/30 (6.7) 4/204 (2) 0.17
Diabetes mellitus 20 (29.4) 30 (19.6) 0.12
Cerebrovascular disease 20 (29.4) 26 (17) 0.05
Therapeutic management
Cancer 14 (20.6) 18 (11.8) 0.09
Empirical therapy 128 (97) 864 (99.3) 0.03
Chronic renal failure 14 (20.6) 21 (13.7) 0.23
Administration within 6 h 101 (76.5) 604 (69.4) 0.10
Chronic liver failure 7 (10.3) 16 (10.5) 1
Adherence to guidelinesa 51 (38.6) 492 (56.6) <0.001
HIV infection 0 6 (3.9) 0.18
Adequacy according to sensitivity 15/21 (71.4) 83/112 (74.1) 0.79
Charlson score (median, IQR) 7, 6–9 5, 3–7 <0.001
Barthel index (median, IQR) 75, 20–100 95, 50–100 0.02
Complications
RF for aspiration pneumonia 17 (25) 31 (20.3) 0.48
Empyema 3 (2.3) 17 (2) 0.74
Pneumonia group
Septic shock 47 (35.6) 27 (3.1) <0.001
CAP 22 (32.4) 116 (75.8) <0.001
Need for intubation 17 (12.9) 22 (2.5) <0.001
HCAP 31 (45.6) 32 (20.9) <0.001
Multi-organ failure 73 (55.3) 15 (1.7) <0.001
HAP 15 (22) 5 (3.3) <0.001
Persistent bacteraemia 6 (4.5) 1 (0.1) <0.001
Emergence of MDR 6 (4.5) 12 (1.4) 0.06
Risk stratification of
pneumonia
PSI (median, IQR) 145, 121–167 119, 81–143 <0.001
CURB-65 (median, IQR) 2, 2–4 2, 1–3 0.001

SD, standard deviation; COPD, chronic obstructive pulmonary dis-

ease; IQR, interquartile range; RF, risk factors; CAP, community-
IQR, interquartile range; COPD, chronic obstructive pulmonary dis- acquired pneumonia; HCAP, healthcare-associated pneumonia; HAP,
ease; RF, risk factors; CAP, community-acquired pneumonia; HCAP, hospital-acquired pneumonia; PSI, Port severity index; CURB-65, confu-
healthcare-associated pneumonia; HAP, hospital-acquired pneumonia; sion-urea-respiratory rate-blood pressure and over 65 years old;
PSI, Port severity index; CURB-65, confusion-urea-respiratory rate- MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicil-
blood pressure and age 65 years or older. lin-sensitive Staphylococcus aureus; MDR multidrug resistance.

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794
794 Clinical Microbiology and Infection, Volume 18 Number 8, August 2012 CMI

a
Adherence to guidelines was evaluated according to the IDSA 2007 16. Giannelli G, Minchella L, Lopalco PL, Mastromarco G, Antonaci S.
guidelines for patients with CAP and according to the ATS 2005 Treatment of community-acquired pneumonia: a descriptive study in
an Apulian department of internal medicine. Med Sci Monit 2005; 119:
guidelines for HCAP and HAP patients. CR434–CR439.
17. Artom A, Artom P, Rattenni S, Castello C, Lo Pinto G. MIPS: multi-
center Italian pneumonia study, Results of an observational, prospec-
References tive and multicenter study on the clinical approach to community-
acquired pneumonia. Infez Med 2004; 123: 181–185.
18. Picazo JJ, Perez-Cecilia E, Herreras A. Respiratory infections in adults
1. Kollef MH, Morrow LE, Baughman RP et al. Health care-associated hospitalized in internal medicine and pneumology departments. DIRA
pneumonia (HCAP): a critical appraisal to improve identification, (Adult Respiratory Infection Day) study. Enferm Infecc Microbiol Clin
management, and outcomes – proceedings of the HCAP Summit. Clin 2003; 21: 180–187.
Infect Dis 2008; 46 (Suppl 4): S296–S334; quiz 5–8. 19. Medrano Gonzalez F, Solis Garcia del Pozo J, Gomariz Garcia S, Sol-
2. American Thoracic Society, Infectious Diseases Society of America era Santos J. Community-acquired pneumonia requiring hospital
Guidelines for the management of adults with hospital-acquired, ven- admission. A comparison of the clinical management, resource use
tilator-associated, and healthcare-associated pneumonia. Am J Respir and prognosis by different medical specialties. Rev Clin Esp 2001; 201:
Crit Care Med 2005; 171: 388–416. 65–68.
3. Kollef MH, Shorr A, Tabak YP, Gupta V, Liu LZ, Johannes RS. Epide- 20. Venditti M, Falcone M, Corrao S, Licata G, Serra P. Outcomes of
miology and outcomes of health-care-associated pneumonia: results patients hospitalized with community-acquired, health care-associ-
from a large US database of culture-positive pneumonia. Chest 2005; ated, and hospital-acquired pneumonia. Ann Intern Med 2009; 150:
128: 3854–3862. 19–26.
4. Carratala J, Mykietiuk A, Fernandez-Sabe N et al. Health care-associ- 21. Mandell LA, Wunderink RG, Anzueto A et al. Infectious Diseases
ated pneumonia requiring hospital admission: epidemiology, antibiotic Society of America/American Thoracic Society consensus guidelines
therapy, and clinical outcomes. Arch Intern Med 2007; 167: 1393– on the management of community-acquired pneumonia in adults. Clin
1399. Infect Dis 2007; 44 (Suppl 2): S27–S72.
5. Micek ST, Kollef KE, Reichley RM, Roubinian N, Kollef MH. Health 22. Halm EA, Fine MJ, Marrie TJ et al. Time to clinical stability in patients
care-associated pneumonia and community-acquired pneumonia: a hospitalized with community-acquired pneumonia: implications for
single-center experience. Antimicrob Agents Chemother 2007; 51: practice guidelines. JAMA 1998; 279: 1452–1457.
3568–3573. 23. Capelastegui A, Espana PP, Quintana JM et al. [Patients hospitalized
6. Shindo Y, Sato S, Maruyama E et al. Health-care-associated pneumo- with community-acquired pneumonia: a comparative study of out-
nia among hospitalized patients in a Japanese community hospital. comes by medical specialty area]. Arch Bronconeumol 2005; 41: 300–306.
Chest 2009; 135: 633–640. 24. Abrahamian FM, Deblieux PM, Emerman CL et al. Health care-associ-
7. Ewig S, Welte T, Chastre J, Torres A. Rethinking the concepts of ated pneumonia: identification and initial management in the ED. Am J
community-acquired and health-care-associated pneumonia. Lancet Emerg Med 2008; 26 (Suppl 6): 1–11.
Infect Dis 2010; 10: 279–287. 25. Schlueter M, James C, Dominguez A, Tsu L, Seymann G. Practice pat-
8. Brito V, Niederman MS. Healthcare-associated pneumonia is a heter- terns for antibiotic de-escalation in culture-negative healthcare-associ-
ogeneous disease, and all patients do not need the same broad-spec- ated pneumonia. Infection 2010; 38: 357–362.
trum antibiotic therapy as complex nosocomial pneumonia. Curr Opin 26. Garau J, Baquero F, Perez-Trallero E et al. Factors impacting on
Infect Dis 2009; 22: 316–325. length of stay and mortality of community-acquired pneumonia. Clin
9. Chalmers JD, Taylor JK, Singanayagam A et al. Epidemiology, antibi- Microbiol Infect 2008; 14: 322–329.
otic therapy, and clinical outcomes in health care-associated pneumo- 27. Almirall J, Boixeda R, Bolibar I et al. Differences in the etiology of
nia: a UK cohort study. Clin Infect Dis 2011; 53: 107–113. community-acquired pneumonia according to site of care: a popula-
10. Curns AT, Holman RC, Sejvar JJ, Owings MF, Schonberger LB. Infec- tion-based study. Respir Med 2007; 101: 2168–2175.
tious disease hospitalizations among older adults in the United States 28. Johnstone J, Majumdar SR, Fox JD, Marrie TJ. Viral infection in adults
from 1990 through 2002. Arch Intern Med 2005; 165: 2514–2520. hospitalized with community-acquired pneumonia: prevalence, patho-
11. Fry AM, Shay DK, Holman RC, Curns AT, Anderson LJ. Trends in gens, and presentation. Chest 2008; 134: 1141–1148.
hospitalizations for pneumonia among persons aged 65 years or older 29. Ferrer M, Liapikou A, Valencia M et al. Validation of the American
in the United States, 1988–2002. JAMA 2005; 294: 2712–2719. Thoracic Society-Infectious Diseases Society of America guidelines
12. Martinez D, Alvarez Rodriguez V, Ortiz de Zarate MM et al. Manage- for hospital-acquired pneumonia in the intensive care unit. Clin Infect
ment in the emergency room of patients requiring hospital treatment Dis 2011; 50: 945–952.
of community-acquired pneumonia. Rev Esp Quimioter 2009; 22: 4–9. 30. Garcia-Vidal C, Viasus D, Roset A et al. Low incidence of multidrug-
13. Zubillaga Garmendia G, Sanchez Haya E, Benavente Claveras J et al. resistant organisms in patients with healthcare-associated pneumonia
Pneumonia above 80 years, admitted to the hospital. An Med Interna requiring hospitalization. Clin Microbiol Infect 2011; 17: 1659–1665.
2008; 25: 117–121. 31. Grenier C, Pepin J, Nault V et al. Impact of guideline-consistent
14. Blasi F, Iori I, Bulfoni A, Corrao S, Costantino S, Legnani D. Can therapy on outcome of patients with healthcare-associated and
CAP guideline adherence improve patient outcome in internal medi- community-acquired pneumonia. J Antimicrob Chemother 2011; 66:
cine departments? Eur Respir J 2008; 32: 902–910. 1617–1624.
15. Etzion O, Novack V, Avnon L et al. Characteristics of low-risk 32. Menendez R, Torres A, Zalacain R et al. Guidelines for the treatment
patients hospitalized with community-acquired pneumonia. Eur J Intern of community-acquired pneumonia: predictors of adherence and out-
Med 2007; 18: 209–214. come. Am J Respir Crit Care Med 2005; 172: 757–762.

ª2011 The Authors

Clinical Microbiology and Infection ª2011 European Society of Clinical Microbiology and Infectious Diseases, CMI, 18, 786–794