Angrill et al. BMC Pulmonary Medicine (2020) 20:83 https://doi.org/10.1186/s12890-020-1115-0 RESEARCH ARTICLE Open Access Determinants of empirical antipseudomonal antibiotic prescription for adults with pneumonia in the emergency department Nuria Angrill1, Miguel Gallego1,2* , Juli Font3, Jordi Vallés2,4, Anisi Morón5, Eduard Monsó1,2 and Jordi Rello2,6,7 Abstract Background: Antipseudomonal antibiotics should be restricted to patients at risk of Pseudomonas aeruginosa infection. However, the indications in different guidelines on community-acquired pneumonia (CAP) are discordant. Our objectives were to assess the prevalence of antipseudomonal antibiotic prescriptions and to identify determinants of empirical antibiotic choices in the emergency department. Methods: Observational, retrospective, one-year cohort study in hospitalized adults with pneumonia. Antibiotic choices and clinical and demographic data were recorded on a standardized form. Antibiotics with antipseudomonal activity were classified into two groups: a) β-lactam antipseudomonals (β-APS), including carbapenems, piperacillin / tazobactam or cefepime (in monotherapy or combination) and b) monotherapy with antipseudomonal quinolones. Results: Data were recorded from 549 adults with pneumonia, with Pseudomonas aeruginosa being isolated in only nine (1.6%). Most (85%) prescriptions were compliant with SEPAR guidelines and 207 (37%) patients received antibiotics with antipseudomonal activity (14% β-APS and 23% levofloxacin). The use of β-APS was independently associated with ICU admission (OR 8.16 95% CI 3.69–18.06) and prior hospitalization (OR 6.76 95% CI 3.02–15.15), while levofloxacin was associated with negative pneumococcal urine antigen tests (OR 3.41 95% CI 1.70–6.85) but negatively associated with ICU admission (OR 0.26 95% CI 0.08–0.86). None of these factors were associated with P aeruginosa episodes. In univariate analysis, prior P aeruginosa infection/colonization (2/9 vs 6/372, p = 0.013), severe COPD (3/9 vs 26/372, p = 0.024), multilobar involvement (7/9 vs 119/372, p = 0.007) and prior antibiotic (6/9 vs 109/372, p = 0.025) were significantly associated with P aeruginosa episodes. Conclusions: Antipseudomonal prescriptions were common, in spite of the very low incidence of Pseudomonas aeruginosa. The rationale for prescription was influenced by both severity-of-illness and pneumococcal urine antigen test (levofloxacin) and prior hospitalization and ICU admission (β-APS). However, these factors were not associated with P aeruginosa episodes. Only prior P aeruginosa infection/colonization and severe COPD seem to be reliable indicators in clinical practice. Keywords: Pneumonia, P aeruginosa, Antipseudomonal antibiotics, Levofloxacin, Guidelines * Correspondence: mgallego@tauli.cat 1 Department of Respiratory Medicine, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain 2 CIBER de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 2 of 9 Background Study population The clinical practice guidelines recommend that antimicrobial All patients ≥18 were evaluated for study inclusion. Patients treatment for patients with community-pneumonia (CAP) were identified by primary ICD-9 codes for pneumonia should be prescribed empirically after evaluation of severity at (480.0–483.99, 485–487) and by respiratory failure or sepsis, presentation and pre-existing comorbid disease, and should be and had to meet the study definition of pneumonia prior to stratified based on prognostic risk scores [1–4]. However, the inclusion. Hospital admission was considered when the pa- recommendations are inconsistent and compliance is low; the tient remained in the ED more than 24 h before discharge or choice of antibiotic treatment at the bedside is a complex was transferred to the ward or the ICU. The following cases process influenced by several factors that have not yet been were excluded: hospital-acquired pneumonia (diagnosed definitively characterized. Factors proposed in the literature in- more than 48 h after hospital admission), witnessed aspir- clude previous hospitalization, previous antibiotic exposure ation pneumonia, solid organ or bone marrow transplant- and underlying diseases, but there are others that are ation, and alternative diagnoses (tuberculosis, cardiac failure dependent on hospital organization, local patterns of resist- or organizing pneumonia). ance, background speciality and cost-effectiveness [5–7]. Compliance of antibiotic treatment with the guidelines has Definitions been associated with improved outcomes, but mortality rates Pneumonia was defined as the presence of a new alveo- rise in patients at risk of multi-resistant microorganisms such lar opacity on chest radiography plus one or more of the as P aeruginosa [8, 9]. However, antipseudomonal antibiotics following: fever (temperature > 38 °C) or hypothermia are considered as broad spectrum and, due to their potential (temperature < 35 °C), cough with or without sputum for causing collateral damage [10, 11], their administration production, pleuritic chest pain and altered breath should be restricted. sounds on auscultation. Pneumonia due to P aeruginosa occurs in several distinct Chronic obstructive pulmonary disease (COPD) was syndromes such as community-acquired pneumonia (CAP) defined as a postbronchodilator FEV1/FVC ratio below and bacteremia in neutropenic hosts, and in intubated patients 0.7 in accordance with the Global Initiative for Chronic [12] In CAP, coverage of P aeruginosa is controversial due to Obstructive Lung Disease (GOLD) criteria in a patient the different rates of prevalence reported in the literature [13– with a smoking habit of more than 10 pack-years [21]. 18]. Factors such as structural lung diseases (especially bron- Immunocompromised group (ICP) included human im- chiectasis), repeated exacerbations of severe chronic obstruct- munodeficiency virus infection (HIV), neoplastic disease ive pulmonary disease (COPD), chronic oral steroid treated with chemotherapy or radiation therapy in the administration, alcoholism and frequent (> 4 courses per year) previous 3 months, hematological malignancy, anesple- or recent antibiotic therapy have been associated with P aeru- nia, and immunosuppressive therapy including chronic ginosa isolates [1, 2, 4, 19]. However, a large retrospective corticosteroid therapy (> 8 mg/d methylprednisolone or study reported that a significant proportion of patients with equivalent for more than 30 days) and non-steroid im- CAP due to P aeruginosa did not present any of these condi- munosuppressive therapy. tions [17]. Although not routinely included in most CAP Health Care-Associated Pneumonia (HCAP) was de- guidelines, a certain degree of immunocompromise is frequent fined according to the criteria reported by the 2005 IDSA/ in patients with community-acquired pneumonia, with neu- ATS Guidelines [22]. Episodes that did not meet the cri- tropenia being a common comorbid condition [20]. teria for HCAP or ICP were classified as CAP [23]. The aim of this study was to assess the prevalence of the use of antibiotics with antipseudomonal activity and Data collection the determinants of empirical choice, in patients The following parameters were recorded in the ED: age, sex, attended for pneumonia in the emergency department. smoking status, alcohol habits, and drug consumption, co- We hypothesizied was that antipseudomonal antibiotics morbidities, antibiotic treatment in the three months prior might be prescribed more frequently than needed, and to ED visit, clinical symptoms, clinical signs, arterial blood that determinants of prescription differ between quino- gas measurements, chest radiograph findings, laboratory pa- lones and other choices. rameters, diagnostic procedures, and empiric antibiotic ther- apy. Previous functional status was evaluated by Barthel scale Methods obtained through hospital databases and the Shared Medical Retrospective, observational cohort study of all consecu- Record of Catalonia (the CatSalut program), and the Charl- tive patients attended in the emergency department (ED) son index was calculated based on data stored in the hospital at Parc Tauli Hospital (682 beds and a reference popula- database [24]. All patients were classified according to the tion of 391,460 inhabitants), between September 2010 pneumonia severity index (PSI). Intensive care unit (ICU) ad- and September 2011. The protocol was approved by the mission, patient treatment restrictions (do-not-resuscitate or- Sabadell Hospital Ethics Committee (2014/538). ders) and mortality were also recorded. Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 3 of 9 Microbiological evaluation were compared using ANOVA or non-parametric tests The hospital protocol included the performance of two when distribution was non-normal. Variables that might blood cultures, respiratory secretions (when feasible) and have influenced the treatment decision were selected for nasopharyngeal swabs (in suspected cases of Influenza analysis: prior antibiotic exposure, prior hospitalization, virus in epidemic periods). Likewise, pleural fluid was nursing home residence, comorbidities such as COPD, analyzed in case of pleural effusion. Samples were ob- bronchiectasis, diabetes mellitus, chronic liver disease, tained for bacterial culture before starting antibiotic chronic heart failure, and criteria of immunosuppression. therapy in the ED. Likewise, urine samples for S pneu- Likewise, episode severity, measured by the need for moniae and Legionella pneumophila antigen detection ICU admission, radiological extent and urine antigen de- were routinely obtained before antibiotic therapy unless tection of S. pneumoniae were also considered. urgent antibiotic therapy was mandatory, and tested Multivariate analysis using stepwise logistic regression using the Binax NOW immunochromatography method was performed with prescription of empirical β-APS as a (Alere BinaxNOW, Streptococcus pneumoniae Antigen dependent variable. Variables showing a univariate asso- Card; Alere Inc., Waltham, MA). We considered the ciation (p < 0.05) were included in the model as covari- presence of positive urine antigens as evidence of a bac- ates. The same analysis was run for levofloxacin as terial infection. Serological tests were not used routinely dependent variable. Hosmer–Lemeshow goodness-of-fit to detect atypical organisms. In patients in whom micro- test was used for model calibration. biological results were negative, or who did not undergo Results were expressed as crude and adjusted odds ra- microbiological tests, the etiology was considered un- tios (OR) with 95% confidence intervals (95% CI). The known. Bronchoscopic samples such as tracheobronchial level of significance was set at 0.05 (two-tailed). aspirate or bronchoalveolar lavage from the lower air- ways were obtained in intubated patients. Results Cohort description Antimicrobial therapy A total of 704 consecutive episodes of pneumonia in Information was obtained on empirical antibiotic therapy adults (669 patients) were recorded during the study given within the first 24 h of admission. The antibiotic regi- period, 549 of whom required hospitalization. Pneumo- men chosen by the attending physician was initiated in the nia was classified as CAP in 295/549 (53.7%), as ICP in ED in accordance with the main national reference guide- 125 (22.8%) and as HCAP in 129 (23.5%). The main lines [4]. https://www.archbronconeumol.org/es-linkresolver- demographic and clinical characteristics of hospitalized normativas-el-diagnostico-el-tratamiento-13074594 patients are shown in Table 1. The mean age of the Antipseudomonal β-lactam (β-APS) was considered as population was 71.2 (SD 16.4) years; 64.9% were men, any regimen that included an intravenous antipseudo- mean Charlson comorbidity index was 2.5 (SD 2.3) and monal antibiotic: carbapenems (meropenem, imipenem), mean Barthel scale 85.6 (SD 23). HCAP group had worse antipseudomonal cephalosporin (ceftazidime, cefepime) functional status and more treatment restrictions than or β-Lactam/ β-lactamase inhibitor (piperacillin–tazo- the other groups, while the Charlson index and 30-day bactam) in monotherapy or in combination. mortality were highest in the ICP group. Antipseudomonal quinolone was defined when levo- floxacin was administered in monotherapy. Antibiotic regimens and microbiology Prior antibiotic therapy was defined as the exposure to For initial treatment in the ED, 305 (55.6%) patients antibiotics above 48 h during the 3 months prior to were prescribed one antibiotic and 244 (44.4%) were hospitalization. prescribed a combination regimen. Eighty-five percent Prior P. aeruginosa infection/colonization was defined of prescriptions complied with the SEPAR guidelines as confirmed infection/colonization within the year prior (for the purpose of study, β-APS use was considered to hospitalization, available from patients’ records. non-concordant). Empirical antimicrobial regimens are summarized in Table 2. The most frequently pre- Statistical analysis scribed antibiotics in monotherapy were levofloxacin The SPSS statistical package version 21.0 (SPSS; Chi- in 129 (42.3%) and amoxicillin-clavulanate in 108 cago, Illinois) was used for the statistical analysis. Results (35.4%). Ceftriaxone plus azithromycin (in 190, 34.2%) for categorical variables were expressed as absolute and was the most common combination. In 78 patients relative frequencies, and continuous variables were (14%) antipseudomonal β-lactam (β-APS) was pre- expressed as mean values and standard deviation (SD). scribed in monotherapy or in combination. The em- Differences in demographic and clinical characteristics pirical use of β-APS differed between groups and was between groups were assessed using the Chi-squared test significantly more frequent in ICP and HCAP (CAP, and Fisher’s exact test for categorical variables. Means 6.4%; ICP, 26.4%; HCAP, 19.4%; p < 0.001). Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 4 of 9 Table 1 Epidemiological and clinical characteristics of hospitalized patients Variable Total n = 549 CAP n = 295 ICP n = 125 HCAP n = 129 Age, mean ± SD 71.2 ± 16.4 67.9 ± 17.4 68.8 ± 15 81.1 ± 10.6 Gender (male) 356 (64.8) 181 (61.4) 93 (74.4) 82 (63.6) Current/former smoker 328 (59.7) 173 (59.6) 87 (69.6) 68 (52.7) Charlson index, mean ± SD 2.5 ± 2.3 1.6 ± 1.7 4.5 ± 2.6 2.4 ± 1.8 Barthel scale, mean ± SD 85.5 ± 23 91.4 ± 18.1 87.5 ± 18.2 70.5 ± 29.5 Diabetes mellitus 164 (29.9) 83 (28.1) 33 (26.4) 48 (37.2) COPD severity GOLD I-III 95(17.3) 56 (19.0) 17 (15.6) 22 (17.1) GOLD IV 38(7.1) 17 (5.8) 7 (5–6) 15 (11.6) Bronchiectasis 24 (4.4) 11 (3.7) 6 (4.8) 7 (5.4) Chronic heart disease 107 (19.5) 49 (16.6) 19 (15.2) 39 (30.2) Dementia 86 (15.7) 33 (11.2) 13 (10.4) 40 (31) Cerebrovascular disease 77 (14) 40 (13.6) 9 (7.2) 28 (21.7) Chronic renal failure 81 (14.8) 34 (11.5) 17 (13.6) 30 (23.3) Chronic liver disease 44 (8) 13 (4.4) 23 (18.4) 8 (6.2) PSI risk class I-III 181 (33.0) 141 (47.8) 22 (17.6) 18 (14.0) IV 212 (38.6) 103 (34.9) 49 (39.2) 60 (46.5) V 156 (28.4) 51 (17.3) 54 (43.2) 51 (39.5) ICU admission 59 (10.7) 40 (13.6) 16 (12.8) 3 (2.3) 30-day- mortality 52 (9.5) 15 (5.1) 22 (17.6) 15 (11.6) Do-not-resuscitate orders 86 (15.7) 24 (8.1) 27 (21.6) 35 (27.1) Prior antibiotic therapy 167 (30.4) 37 (12.5) 48 (38.4) 82((63.6) Prior hospitalization 128 (23.3) – 45 (36.0) 83 (64.3) Nursing home residence 51 (9.3) – 7 (5.6) 44 (33.8) SD Standard deviation, COPD Chronic obstructive pulmonary disease, GOLD Global initiative for chronic obstructive lung disease, PSI Pneumonia severity index, ICU Intensive care unit, ICP Immunocompromised group We collected blood cultures in 340 episodes, and 145 Determinants of empirical antipseudomonal use valid respiratory specimens were retrieved. An etiological The main factors determining selection of antipseudo- diagnosis was made in 184 patients (33.5%): 50 (40%) in monal antibiotics are summarized in Table 3. Two pat- the ICP group, 103 (35%) in the CAP group and 31 terns of utilization were documented. The first was (24%) in the HCAP group (p = 0.02). Streptococcus pneu- based on monotherapy with levofloxacin, which was not moniae was the most frequently isolated pathogen (124/ administered with antipseudomonal intention and was 549, 22.6%): the pneumococcal urine antigen test was restricted to patients with low severity-of-illness. Only positive in 111/389 (28.5%) of patients evaluated. Other three patients in the ICU received quinolones; even in microorganisms had a lower frequency: Haemophilus non-ICU patients, these agents were mainly prescribed influenzae 16 (2.9%), Moraxella catarrhalis 11 (2%) and in patients with PSI I-III and limited radiological extent. Enterobacteriaceae 10 (1.8%). Pseudomonas aeruginosa In contrast, negative pneumococcal antigen was a key was isolated in 9/549 (1.6%): three cases in the CAP factor in the prescription of levofloxacin. The other doc- group, two in the ICP group and four in the HCAP umented pattern was based on the prescription of β-APS group. Moderate-severe COPD was present in five and in monotherapy (or in combination), used for antipseu- nine patients respectively and a previous isolate of domonal coverage. Site of care was the most important Pseudomonas aeruginosa was documented in two of determinant of β-APS prescription: in fact, β-APS was them. Bacteremia occurred in only one patient, with the initial treatment in 22/59 (35.6%) patients in the ICU chronic dialysis. and in 56/490 (11.4%) in ward-admitted patients (p < The distribution of microorganisms across the groups 0.0001). This association was confirmed in the multivari- is shown in Fig. 1. ate analysis (OR: 8.16; 95% CI: 3.69–18.06; p < 0.0001). Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 5 of 9 Table 2 Description of initial antibiotic therapy for 549 inpatients Initial therapy Total n = 549 CAP n = 295 ICP n = 125 HCAP n = 129 Monotherapy 305 (55.6) 156 (52.9) 64 (51.2) 85 (65.9) Amoxicillin-clavulanate 108 (35.4) 56 (35.9) 14 (21.9) 38 (44.7) Levofloxacin 129 (42.3) 76 (48.7) 24 (37.5) 29 (34.1) Ceftriaxone 22 (7.2) 13 (8.3) 7 (10.9) 2 (2.4) β-APS 45 (14.8) 11 (7.1) 18 (28.1) 16 (18.8) Piperacillin-tazobactam 26 (8.5) 3 (1.9) 12 (18.8) 11 (12.9) Cefepime 3 (1.0) 2 (1.3) 1 (1.6) 0 (0.0) Meropenem 16 (5.2) 6 (3.8) 5 (7.8) 5 (5.9) Others 1 (0.3) Combination 244 (44.4) 139 (47.1) 61 (48.8) 44 (34.1) Ceftriaxone + macrolide 190 (77.9) 119 (85.6) 41 (67.2) 30 (68.2) β-APS + quinolone or aminoglycoside or macrolide 33 (13.5) 9 (6.5) 15 (24.6) 9 (20.5) Others 21 (8.6) 11 (7.9) 5 (8.2) 5 (11.4) β-APS Antipseudomonal β-lactam Prior hospitalization was another significant deter- population), which are classical risk factors for P aeru- minant of β-APS use in community-acquired pneumo- ginosa infection. nia (OR: 6.76; 95% CI: 3.01–15.14; p < 0.0001). This Among the 381 episodes with blood cultures and/ factor was part of the HCAP definition and was present or respiratory samples available we tested the deter- in up to 36% of the ICP population. In addition, an as- minants of antibiotic use as risk factors for P aerugi- sociation was found between β-APS prescription and nosa recovery. Only multilobar involvement (7/9 vs severe COPD and immunosuppression (ICP 119/372, p = 0.007), prior antibiotic (6/9 vs 109/372, Fig. 1 Distribution of pathogens Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 6 of 9 Table 3 Analysis of factors associated with empirical antipseudomonal use in pneumonia. Univariate and multivariate logistic regression analysis Quinolones Beta-APS Univariate Multivariate Univariate Multivariate OR CI 95% p OR CI 95% p OR CI 95% p OR CI 95% p value value value value Male sex 0.781 (0.520- 1.173) 0.234 1.971 (1.127- 3.446) 0.017 1.316 (0.660- 0.436 2.625) ICU admission 0.155 (0.048- 0.503) 0.002 0.256 (0.076- 0.028 4.608 (2.538- 8.368) < 8.166 (3.691- < 0.863) 0.001 18.065) 0.0001 PSI risk class I-III vs IV/V 0.577 (0.384- 0.866) 0.008 0.595 (0.390- 0.016 3.440 (1.770- 6.688) < 1.817 (0.861- 0.117 0.907) 0.001 3.834) Multilobar involvement 0.457 (0.281- 0.744) 0.002 0.531 (0.319- 0.015 1.676 (1.021- 2.752) 0.041 1.355 (0.732- 0.333 0.885) 2.507) Comorbidities Liver disease 0.304 (0.107- 0.867) 0.026 0.413 (0.141- 0.108 2.839 (1.412- 5.705) 0.003 1.808 (0.767- 0.175 1.214) 4.261) Bronchiectasis 0.851 (0.311 -2.326) 0.753 1.629 (0.590- 4.499) 0.346 Diabetes 1.179 (0.771 -1.802) 0.446 0.979 (0.579 -1.655) 0.936 Chronic renal disease 1.354 (0.798 -2.299) 0.261 0.829 (0.407 -1685) 0.604 Chronic heart disease 1.197 (0.737 -1.943) 0.468 1.078 (0.594 -1.954) 0.806 COPD No COPD Reference Reference category category GOLD I-III 1.061 (0.630-1.789) 0.823 0.987 (0.505-1.931) 0.971 0.790 (0.362-1.723) 0.553 GOLD IV 1.146 (0.539-2.436) 0.720 3.415 (1.653-7.055) 0.001 2.993 (1.230-7.285) 0.016 Prior hospitalization 0.836 (0.518 -1.351) 0.464 5.579 (3.370- 9.237) < 6.760 (3.017- < 0.001 15.147) 0.0001 Nursing home residence 0.580 (0.265-1.267) 0.172 0.488 (0.171-1.394) 0.180 Prior antibiotic 0.897 (0.581 -1.385) 0.624 2.680 (1.645- 4.366) < 1.058 (0.479- 0.890 0.001 2.366) Immunocompromise 0.722 (0.439- 1.186) 0.199 3.021 (1.826- 4.999) < 2.064 (1.135- 0.018 0.001 3.756) Negative pneumoccocal 3.768 (1.903- 7.459) < 3.408 (1.696- 0.001 0.754 (0.428- 1.326) 0.327 antigen 0.001 6.848) SD standard deviation, ICU intensive care unit, PSI pneumonia severity index, COPD chronic obstructive pulmonary disease, GOLD global initiative for chronic obstructive lung disease p = 0.025) and severe COPD (3/9 vs 26/372, p = contrast, the prescription of levofloxacin in monotherapy 0.024) presented significant associations. When prior was mainly associated with the presence of negative S P aeruginosa infection/colonization was included in pneumoniae urine antigen and less severe disease. the model, it was also associated with P aeruginosa Few studies carried out in the ED have focused on the pneumonia (2/9 vs 6/372, p = 0.013). determinants of antibiotic use in pneumonia, even though rates of inappropriate antibiotic treatment (for Discussion whatever reason) are known to be around 50% [25]. Im- In this study we report the drivers of antibiotic prescrip- proving adherence to guidelines has been shown to raise tion for pneumonia in the ED. Nearly 37% of hospital- the rates of appropriate treatment [26, 27]. In our study ized patients received antibiotics with potential adherence to guidelines was high, around 85%, but the antipseudomonal activity; although 14% of prescriptions etiological findings show that the use of broad spectrum were intended to provide P aeruginosa coverage, this antibiotics could be reduced. pathogen was documented in only 1.6% of episodes. Therefore, although the indication for respiratory Severity-of-illness, identified as the need for intensive fluoroquinolones was to provide coverage against poten- care unit admission, and prior hospitalization were the tial S. pneumoniae strains with reduced betalactam sus- main factors in the decision to prescribe β-APS. In ceptibility, Enterobacteriaceae and particularly atypical Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 7 of 9 organisms, in agreement with the SEPAR guidelines [4], In severe episodes of CAP requiring vasopressors or the potential of levofloxacin as an antipseudomonal anti- mechanical ventilation, beta-lactam/macrolide combina- biotic should be borne in mind. tions were preferred in view of the evidence of reduced The specific determinants of the choice of levofloxacin mortality due to the immunomodulatory effects of over other options have not been studied in depth. In macrolides [36, 37]. Moreover, in agreement with recent our study, apart from its being restricted to ward use, recommendations [38] antimicrobial de-escalation is a the most intriguing finding was its association with uni- common practice in the ICU, once the identity of organ- lobar pneumonia, suggesting that radiological spread is ism has been confirmed by laboratory tests. interpreted as a sign of severity, as reported elsewhere In our study, site of care was a more decisive factor than [28, 29]. Another interesting finding was the association high PSI; 35.6% of ICU patients received β-APS compared with negative pneumococcal urine antigen tests. The with 11.4% of ward-admitted patients (p < 0.0001). presence of a positive pneumococcal urine antigen test Another key determinant of prescription in our study is indicative of S pneumoniae etiology, but therapy is was prior hospitalization. This variable is inherent to the based on the result in fewer than 15% of cases [30]. Al- HCAP definition [22] in an attempt to identify patients at though outcomes with fluoroquinolones and β-lactams risk of infection by multidrug-resistant microorganisms seem to be similar, it is less clear whether a β lactam/ (MDR), but it has been associated with an unjustified in- macrolide combination is superior to fluoroquinolone crease in antipseudomonal prescription in patients with monotherapy; the findings are conflicting and the data pneumonia, despite the lack of any increase in nosocomial in patients with severe disease are insufficient to draw pathogens [39]. Nevertheless, several recent scores conclusions [31]. Thus, attending physicians should be attempting to refine risk factors of MDR systematically in- aware of this susceptibility pattern and should bear in clude this factor [20, 40]. In a study performed with 935 mind the recent EMA warning to restrict indications of hospitalized patients which included immunosuppressed fluoroquinolones due to potentially severe adverse patients, prior hospitalization was the main risk factor for events, due to the increased risk of cardiac events and MDR [20]. Although we used the traditional definition aneurysmal rupture in the elderly; indeed, safer treat- with a cut-off of 48 h, longer hospital stay has been associ- ments such as β-lactams are recommended [31]. ated with an increased risk of MDR infection [41], a factor In the case of the coverage of P aeruginosa, the clinical that was not analysed in our study. practice guidelines range substantially, from a non-specific Other factors related to the use of β-APS were immunosup- recommendation [3] to a recommendation in critical pa- pression and severity of COPD, which are well-known risk tients with specific risk factors [1, 4]. In general, and in factors for P aeruginosa infection. In our study severe COPD keeping with our results, overtreatment has been reported patients accounted for 60% of those infected; in the worldwide in several studies worldwide, reaching nearly 40% of pa- study by Restrepo et al (42) COPD was the main comorbidity, tients in some cases [11, 32–34]. However, when evaluat- but immunosuppression was present in only two patients. ing severe episodes, the failure to cover P aeruginosa is Determinants of P aeruginosa coverage used in the ED associated with increased mortality. For instance, in a were not useful in predicting P aeruginosa infection in our multicentre study of 529 severely ill CAP admitted to the study. However, prior P aeruginosa infection/colonization to- ICU, despite correct treatment according to IDSA guide- gether with severe COPD seem to be good predictors, al- lines in 15 (75%) of 20 cases of Pseudomonas aeruginosa though multivariate analysis was not carried out due to the infection, antimicrobial treatment at admission was inad- low number of P aeruginosa episodes. These results are in equate. Chronic obstructive pulmonary disease, malig- agreement with Restrepo et al [42] and with the 2019 ATS/ nancy, previous antibiotic exposure, and radiographic IDSA guidelines for community-acquired pneumonia [43] findings demonstrating rapid spread of disease were asso- recommendations. ciated with P aeruginosa pneumonia [9]. The most consistent risk factor for PA is prior Few studies have evaluated aetiology according to colonization. Unfortunately, this information is not always severity-of-illness at ED presentation; Cilloniz and col- available at the ED of a referral center because not all CAP leagues [35] identified Gram-negative enteric bacteria and patients have prior isolates. In any case, our data suggest that P aeruginosa as being more frequent in higher-risk groups potential prior PA respiratory colonization reported in com- than in low-risk groups, strengthening the argument that puterized clinical records should be checked in the ED. coverage of Gram-negative organisms is needed for these The main limitation of this study is its single-centre retro- patients. However, in their study, only 5% of patients ad- spective design. Furthermore, the generalizability of the find- mitted to a respiratory ICU presented P aeruginosa. ings is limited as this is a single-centre cohort from 2010 to In our study, antipseudomonal coverage was provided 2011 that may no longer reflect current clinical practices or in 22/59 of cases admitted to ICU, but in fact none of pathogen distribution; for example, moxifloxacin is not ap- these patients were infected by P aeruginosa. proved for systemic use in Spain, a circumstance that may Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 8 of 9 limit the generalization of our findings. However, the single- Autònoma de Barcelona, Sabadell, Spain. 4Critical Care Center, Parc Taulí centre design allowed us to avoid confounding factors such Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain. 5Department of as differences in protocols or bias in patient selection. Pharmacy, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain. 6Vall Conclusions d’Hebron Institute of Research (VHIR), Barcelona, Spain. 7Research Department, CHU Nîmes, Université Montpellier-Nîmes, Nîmes, France. Antipseudomonal prescription was common, in spite of the very low incidence of Pseudomonas aeruginosa. Received: 17 November 2019 Accepted: 13 March 2020 Levofloxacin was prescribed to patients with low severity and negative pneumococcal urine antigen tests. Anti- biotic prescription intended to provide P aeruginosa References 1. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, coverage was guided mainly by severity-of-illness, prior et al. Infectious Diseases Society of America/American Thoracic Society hospitalization and, less frequently, by specific risk fac- consensus guidelines on the management of community-acquired tors. However, these factors were not associated with P pneumonia in adults. Clin Infect Dis. 2007;44:S27e72. 2. Woodhead M, Blasi F, Ewig S, Garau J, Huchon G, Leven M, et al. Guidelines aeruginosa episodes. Only prior P aeruginosa infection/ for the management of adult lower respiratory tract infections full version. colonization and severe COPD seem to be reliable in Clin Microbiol Infect. 2011;17:E1e59. clinical practice. 3. Lim WS, Baudouin SV, George RC, Hill AT, Jamieson C, Le Jeune I, et al. BTS guidelines for the management of community acquired pneumonia in adults: update 2009. Thorax. 2009;64:iii1e55. Abbreviations 4. Alfageme I, Aspa J, Bello S, Blanquer J, Blanquer R, Borderías L, et al. ATS: American Thoracic Society; CAP: Community-acquired pneumonia; Guidelines for the diagnosis and management of community-acquired HCAP: Healthcare-associated pneumonia; ICP: Immunocompromise; pneumonia. Spanish Society of Pulmonology and Thoracic Surgery (SEPAR). IDSA: Infectious Diseases Society of America; COPD: Chronic obstructive Arch Bronconeumol. 2005;41(5):272–89. pulmonary disease; GOLD: Global initiative for chronic obstructive disease; 5. Prina E, Ranzani OT, Torres A. Community-acquired pneumonia. Lancet. PSI: Pneumonia severity index; EMA: European Medicines Agency 2015;386:1097–108. 6. Musher DM, Thorner AR. Community-acquired pneumonia. N Engl J Med. Acknowledgments 2014;371:1619–28. We thank Michael Maudsley for providing an outline for this manuscript and 7. Reyes Calzada S, Martinez Tomas R, Cremades Romero MJ, Martinez for support in editing and journal styling. Moragon E, Soler Cataluna JJ, Menendez Villanueva R. Empiric treatment in hospitalized community-acquired pneumonia. Impact on mortality, length Authors’ contributions of stay and re-admission. Respir Med. 2007;101:1909–15. NA, MG, JF, JV, AM, EM and JR jointly designed the study. NA and MG reviewed 8. Kollef MH, Micek ST. Patients hospitalized with pneumonia: determining the each patient’s chart with the hospital’s program for storing digital images (RAIM need for broad-spectrum antibiotic therapy. Clin Infect Dis. 2012;54(4):479–82. Server) to confirm the diagnosis of pneumonia based on clinical presentation and 9. Bodí M, Rodríguez A, Solé-Violán J, Gilavert MC, Garnacho J, Blanquer J, chest radiograph. NA, JF and AM collected the data. MG and JR performed the et al. Community-acquired pneumonia intensive care units (CAPUCI) study statistical analyses. NA drafted the manuscript and all authors contributed to the final investigators. Antibiotic prescription for community-acquired pneumonia in manuscript. The authors read and approved the final version of the manuscript. the intensive care unit: impact of adherence to Infectious Diseases Society of America guidelines on survival. Clin Infect Dis. 2005;41(12):1709–16. Funding 10. Singanayagam A, Chalmers JD. Severity assessment scores to guide Supported in part by Centro de Investigacion Biomedica en Red (CIBERES), empirical use of antibiotics in community acquired pneumonia. Lancet Instituto de Salud Carlos III, Madrid, Spain. The funding agency had no role Respir Med. 2013;1(8):653–62. in the study design; collection, analysis, or interpretation of data; drafting of 11. Webb BJ, Sorensen J, Jephson A, Mecham I, Dean NC. Broad-spectrum the manuscript; or decision to submit for publication. antibiotic use and poor outcomes in community-onset pneumonia: a cohort study. Eur Respir J. 2019;54(1):1900057. Availability of data and materials 12. Respipedia. Respiratory infections. Pseudomonas aeruginosa infections. The datasets used in this study are available from the corresponding author Www.Ers-education.Org/home/browse-all-content.Aspx?Idparent=207389. upon reasonable request. Date last Accessed: 31 Jan 2019. 13. Arancibia F, Bauer TT, Ewig S, Mensa J, Gonzalez J, Niederman MS, et al. Ethics approval and consent to participate Community-acquired pneumonia due to gram-negative bacteria and Pseudomonas The study protocol was approved by the Hospital Ethics Committee on aeruginosa: incidence, risk, and prognosis. Arch Intern Med. 2002;162(16):1849–58. Human Research (No. 2014/538). The need for informed consent was waived 14. Kang CI, Song JH, Oh WS, Ko KS, Chung DR. Peck KR; Asian network for because patient information was anonymized and de-identified prior to surveillance of resistant pathogens (ANSORP) study group. Clinical analysis. outcomes and risk factors of community-acquired pneumonia caused by gram-negative bacilli. Eur J Clin Microbiol Infect Dis. 2008;27(8):657–61. Consent for publication 15. von Baum H, Welte T, Marre R, Suttorp N, Ewig S, CAPNETZ study group. Not applicable. Community-acquired pneumonia through Enterobacteriaceae and Pseudomonas aeruginosa: diagnosis, incidence and predictors. Eur Respir J. 2010;35(3):598–605. Competing interests 16. Falguera M, Carratala J, Ruiz-Gonzalez A, Garcia-Vidal C, Gazquez I, Dorca J, JR has received research grants and consulting fees from BAYER and et al. Risk factors and outcome of community-acquired pneumonia due to PARATEK. The pharmaceutical industry did not contribute to the funding of gram-negative bacilli. Respirology. 2009;14:105–11. this study. The other authors declare that they have no competing interests. 17. Sibila O, Laserna E, Maselli DJ, Fernandez JF, Mortensen EM, Anzueto A, et al. Risk factors and antibiotic therapy in P aeruginosa community- Author details acquired pneumonia. Respirology. 2015;20:660–6. 1 Department of Respiratory Medicine, Parc Taulí Hospital Universitari, Institut 18. Cilloniz C, Gabarrus A, Ferrer M, Puig de la Bellacasa J, Rinaudo M, Mensa J, d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, et al. Community-acquired pneumonia due to multidrug- and non- Sabadell, Spain. 2CIBER de Enfermedades Respiratorias (CIBERES), Instituto de multidrug-resistant pseudomonas aeruginosa. Chest. 2016;150:415–25. Salud Carlos III, Madrid, Spain. 3Emergency Department, Parc Taulí Hospital 19. Niederman MS, Mandell LA, Anzueto A, Bass JB, Broughton WA, Campbell Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat GD, et al. American Thoracic Society. Guidelines for the management of Angrill et al. BMC Pulmonary Medicine (2020) 20:83 Page 9 of 9 adults with community-acquired pneumonia. Diagnosis, assessment of 39. Jones BE, Jones MM, Huttner B, Stoddard G, Brown KA, Stevens VW, et al. Trends in severity, antimicrobial therapy, and prevention. Am J Respir Crit Care Med. antibiotic use and nosocomial pathogens in hospitalized veterans with pneumonia 2001;163(7):1730–54. at 128 medical centers, 2006-2010. Clin Infect Dis. 2015;61(9):1403–10. 20. Aliberti S, Di Pasquale M, Zanaboni AM, Cosentini R, Brambilla AM, Seghezzi 40. Prina E, Ranzani OT, Polverino E, Cillóniz C, Ferrer M, Fernandez L, et al. Risk S, et al. Stratifying risk factors for multidrug-resistant pathogens in factors associated with potentially antibiotic-resistant pathogens in hospitalized patients coming from the community with pneumonia. Clin community-acquired pneumonia. Ann Am Thorac Soc. 2015;12(2):153–60. Infect Dis. 2012;54(4):470–8. 41. Gross AE, Van Schooneveld TC, Olsen KM, Rupp ME, Bui TH, Forsung E, et al. 21. Rabe KF, Hurd S, Anzueto A, Barnes PJ, Buist SA, Calverley P, et al. Global Epidemiology and predictors of multidrug-resistant community-acquired initiative for chronic obstructive lung disease. Global strategy for the diagnosis, and health care-associated pneumonia. Antimicrob Agents Chemother. management, and prevention of chronic obstructive pulmonary disease: GOLD 2014;58(9):5262–8. executive summary. Am J Respir Crit Care Med. 2007;176(6):532–55. 42. Restrepo MI, Babu BL, Reyes LF, Chalmers JD, Soni NJ, Sibila O, et al. Burden 22. American Thoracic Society; Infectious Diseases Society of America. and risk factors for Pseudomonas aeruginosa community-acquired Guidelines for the management of adults with hospital-acquired, ventilator- pneumonia: a multinational point prevalence study of hospitalised patients. associated, and healthcare-associated pneumonia. Am J Respir Crit Care Eur Respir J. 2018;52:1701190. Med. 2005;171(4):388–416. 43. Metlay JP, Waterer GW, Long AC, Anzueto A, Brozek J, Crothers K, et al. Diagnosis 23. Vallés J, Martin-Loeches I, Torres A, Diaz E, Seijas I, López MJ, et al. and treatment of adults with community-acquired pneumonia. An official clinical Epidemiology, antibiotic therapy and clinical outcomes of healthcare- practice guideline of the American Thoracic Society and Infectious Diseases Society associated pneumonia in critically ill patients: a Spanish cohort study. of America. Am J Respir Crit Care Med. 2019;200(7):e45–67. Intensive Care Med. 2014;40(4):572–81. 24. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in J Chronic Dis. 1987;40(5):373–83. published maps and institutional affiliations. 25. Oltra Hostalet F, Núñez-Núñez M, Portillo Cano MDM, Navarro Bustos C, Rodríguez- Baño J, Retamar Gentil P. Analysis of quality antimicrobial agent use in the emergency department of a tertiary care hospital. Emergencias. 2018;30(5):297–302. 26. Pifarre R, Falguera M, Vicente-de-Vera C, Nogues A. Characteristics of community-acquired pneumonia in patients with chronic obstructive pulmonary disease. Respir Med. 2007;101(10):2139–44. 27. Faine BA, Mohr N, Dietrich J, Meadow L, Harland KK, Chrischilles E. Antimicrobial therapy for pneumonia in the emergency department: the impact of clinical pharmacists on appropriateness. West J Emerg Med. 2017; 18(5):856–63. https://doi.org/10.5811/westjem.2017.5.33901. 28. Liapikou A, Cillóniz C, Gabarrús A, Amaro R, De la Bellacasa JP, Mensa J, et al. Multilobar bilateral and unilateral chest radiograph involvement: implications for prognosis in hospitalised community-acquired pneumonia. Eur Respir J. 2016;48(1):257–61. 29. Lisboa T, Blot S, Waterer GW, Canalis E, de Mendoza D, Rodriguez A, et al. Radiologic progression of pulmonary infiltrates predicts a worse prognosis in severe community-acquired pneumonia than bacteremia. Chest. 2009;135(1):165–72. 30. West DM, McCauley LM, Sorensen JS, Jephson AR, Dean NC. Pneumococcal urinary antigen test use in diagnosis and treatment of pneumonia in seven Utah hospitals. ERJ Open Res. 2016;19:2(4). 31. Wunderink RG, Waterer G. Advances in the causes and management of community acquired pneumonia in adults. BMJ. 2017;358:j2471. 32. Jones BE, Brown KA, Jones MM, Huttner BD, Greene T, Sauer BC, et al. Variation in empiric coverage versus detection of methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa in hospitalizations for community-onset pneumonia across 128 US veterans affairs medical centers. Infect Control Hosp Epidemiol. 2017;38(8):937–44. 33. Chalmers JD, Rother C, Salih W, Ewig S. Healthcare-associated pneumonia does not accurately identify potentially resistant pathogens: a systematic review and meta-analysispublished correction appears in. Clin Infect Dis. 2015;60(7):1143. 34. Metersky ML, Frei CR, Mortensen EM. Predictors of pseudomonas and methicillin-resistant Staphylococcus aureus in hospitalized patients with healthcare-associated pneumonia. Respirology. 2016;21(1):157–63. 35. Cilloniz C, Ewig S, Polverino E, Marcos MA, Esquinas C, Gabarrus A, et al. Microbial aetiology of community-acquired pneumonia and its relation to severity. Thorax. 2011;66:340–6. 36. Martin-Loeches I, Lisboa T, Rodriguez A, Putensen C, Annane D, Garnacho- Montero J, et al. Combination antibiotic therapy with macrolides improves survival in intubated patients with community-acquired pneumonia. Intensive Care Med. 2010;36(4):612–20. 37. Restrepo MI, Mortensen EM, Waterer GW, Wunderink RG, Coalson JJ, Anzueto A. Impact of macrolide therapy on mortality for patients with severe sepsis due to pneumonia. Eur Respir J. 2009;33(1):153–9. 38. Tabah A, Bassetti M, Kollef MH, Zahar JR, Paiva JA, Timsit JF, et al. Antimicrobial de-escalation in critically ill patients: a position statement from a task force of the European Society of Intensive Care Medicine (ESICM) and European Society of Clinical Microbiology and Infectious Diseases (ESCMID) critically ill patients study group (ESGCIP). Intensive Care Med. 2019;28:1–21.