Stratifying Risk Factors for Multidrug-Resistant Pathogens in Hospitalized Patients Coming From the Community With Pneumonia

Abstract

(See the Editorial Commentary by Kollef, on pages 479–82.)

Background. Not all risk factors for acquiring multidrug-resistant (MDR) organisms are equivalent in predicting pneumonia caused by resistant pathogens in the community. We evaluated risk factors for acquiring MDR bacteria in patients coming from the community who were hospitalized with pneumonia. Our evaluation was based on actual infection with a resistant pathogen and clinical outcome during hospitalization.

Methods. An observational, prospective study was conducted on consecutive patients coming from the community who were hospitalized with pneumonia. Data on admission and during hospitalization were collected. Logistic regression models were used to evaluate risk factors for acquiring MDR bacteria independently associated with the actual presence of a resistant pathogen and in-hospital mortality.

Results. Among the 935 patients enrolled in the study, 473 (51%) had at least 1 risk factor for acquiring MDR bacteria on admission. Of all risk factors, hospitalization in the preceding 90 days (odds ratio [OR], 4.87 95% confidence interval {CI}, 1.90–12.4]; P = .001) and residency in a nursing home (OR, 3.55 [95% CI, 1.12–11.24]; P = .031) were independent predictors for an actual infection with a resistant pathogen. A score able to predict pneumonia caused by a resistant pathogen was computed, including comorbidities and risk factors for MDR. Hospitalization in the preceding 90 days and residency in a nursing home were also independent predictors for in-hospital mortality.

Conclusions. Risk factors for acquiring MDR bacteria should be weighted differently, and a probabilistic approach to identifying resistant pathogens among patients coming from the community with pneumonia should be embraced.

Pneumonia caused by multidrug-resistant (MDR) pathogens traditionally has been confined to the hospital setting. In view of the diffusion of healthcare delivery and technology outside the hospital, resistant pathogens have extended beyond the confines of the inpatient setting. The rapid emergence of MDR bacteria that cause pneumonia in the community has created the need to identify risk factors for acquiring resistant pathogens by evaluating the contacts patients have with the healthcare environment as well as the patient’s characteristics [1].

Pneumonia that occurs in outpatients who have been in contact with the healthcare system is termed healthcare-associated pneumonia (HCAP) [2]. Prior hospitalization, residency in a nursing home, going to hemodialysis centers, and receiving domiciliary care are some of the risk factors for acquiring the resistant pathogens included in the HCAP classification. Beyond these risk factors, immunosuppression, severe underlying diseases, and the patient’s functional status also have been recognized as conditions that could lead to acquisition of MDR pathogens [3]. The American Thoracic Society/Infectious Diseases Society of America (ATS/IDSA) guidelines recognize that pneumonia in patients with these risk factors who come from the community shares bacteriologic features with nosocomial pneumonia and thus should be treated with broad-spectrum empiric antibiotic therapy [2].

It has been demonstrated that a large group of patients with risk factors for acquiring MDR bacteria may present with differences in terms of epidemiology, impact on the actual acquisition of a resistant pathogen, and response to therapy [4, 5]. Recently, concern has been raised because all risk factors for MDR acquisition are classified within the same category and a definition of different subpopulations of patients is needed [6, 7].

The aim of our study was to evaluate risk factors for acquiring MDR bacteria among patients coming from the community who were hospitalized with pneumonia on the basis of infection with a resistant pathogen and patient clinical outcome during hospitalization. A second purpose was to develop a risk-scoring tool that could be used to identify subjects who come from the community to the hospital with pneumonia caused by resistant organisms.

MATERIALS AND METHODS

Study Design and Study Patients

This was an observational, prospective study of consecutive patients coming from the community who were admitted to the Policlinico Hospital, Milan, Italy, with a diagnosis of pneumonia between April 2008 and April 2010. The Institutional Review Board of the Policlinico Hospital approved the study. Patients ≥18 years of age who satisfied the criteria for pneumonia were included in the study. Patients who were hospitalized in the previous 15 days were excluded. The patient enrollment process is detailed in the online supplement. The following data were recorded: demographics; past medical history; severity of symptoms on admission; pneumonia severity index (PSI) and CURB-65 score (confusion, urea nitrogen, respiratory rate, blood pressure, ≥65 years of age); physical, laboratory, and radiological findings on admission; microbiological data; empiric antibiotic therapy; and in-hospital mortality [8, 9].

Study Definitions

Pneumonia was defined as the presence of a new pulmonary infiltrate on chest radiograph at the time of hospitalization associated with ≥1 of the following: (1) new or increased cough with/without sputum production; (2) fever (≥37.8°C) or hypothermia (<35.6°C); or (3) abnormal white blood cell count (either leukocytosis or leukopenia), or C-reactive protein values above the local upper limit. Severe community-acquired pneumonia (CAP) was defined according to the latest ATS guidelines [10]. Severe sepsis was defined as sepsis plus ≥1 signs of organ hypoperfusion or organ dysfunction, as previously reported [11]. Length of stay was calculated as the number of days from the date of admission to the date of discharge.

Risk Factors for Acquiring MDR Pathogens

The following risk factors for acquiring MDR pathogens were recorded among the study population according to the ATS/IDSA guidelines: hospitalization for ≥2 days in the preceding 90 days, residency in a nursing home or extended-care facility, home infusion therapy (including antibiotics), home wound care, chronic dialysis within 30 days, family member with an MDR pathogen, antimicrobial therapy in the preceding 90 days, and immunosuppression [2]. Severe immunosuppression was defined by the presence of ≥1 of the following factors: active hematologic malignancy, transplantation, immunosuppressive therapy, chemotherapy, and radiotherapy. Mild-to-moderate immunosuppression was defined by the presence of ≥1 of the following factors: chronic systemic steroid therapy (prednisone ≥25 mg/day), active solid malignancy, splenectomy, and autoimmune disease.

Microbiological Analysis and Empiric Antibiotic Therapy

Microbiological examinations were performed on sputum, urine, and blood during the first 24 hours after admission and according to standards of practice. Pleural puncture, tracheobronchial aspirates, and bronchoalveolar lavage fluid, when available, were also collected and cultured. Identification of microorganisms and susceptibility testing were performed according to standard methods [12]. Microbiological results were reviewed, and a microbiological cause was assigned independently by 2 of the investigators (M. D. P. and S. S.). The etiology was considered definite if 1 of the following criteria was met: positive blood culture in the absence of an apparent extrapulmonary focus; positive bacterial culture of pleural fluid; positive urinary antigen for Legionella pneumophila (Binax Now, Trinity Biotech); positive urinary antigen for Streptococcus pneumoniae (Binax Now, Emergo Europe); a bacterial yield in cultures of valid sputum (>25 polymorphonuclear cells and <10 epithelial cells per power field, total magnification ×100) of ≥10⁶ colony-forming units (CFU)/mL, tracheobronchial aspirates of ≥10⁵ CFU/mL, bronchoalveolar lavage fluid of ≥10⁴ CFU/mL, and protected specimen brush cultures of ≥10³ CFU/mL; and occurrence of seroconversion (a 4-fold rise in immunoglobulin G [IgG] titers for Chlamydophila pneumoniae [1:512] and L. pneumophila or a rise in immunoglobulin M [IgM] titers for C. pneumoniae [1:32] and Mycoplasma pneumoniae [any titer]). When ≥2 microbiological causes were present, the patient was considered to have a polymicrobial infection. Patients for whom no microbiological tests were performed and patients with negative microbiological results were considered to have disease of an unknown etiology.

Methicillin-resistant Staphylococcus aureus (MRSA); Pseudomonas aeruginosa resistant to antipseudomonal penicillins, cephalosporins, carbapenems, and quinolones; Stenotrophomonas maltophilia; vancomycin-resistant Enterococcus; Acinetobacter baumanii; extended spectrum β-lactamase (ESBL)–producing Enterobacteriaceae; and other nonfermenting gram-negative bacilli were considered to be MDR pathogens.

Empiric antibiotic therapy was administered as soon as the diagnosis of pneumonia was reached in the emergency department. The empiric antibiotic treatment was evaluated for compliance with the European Respiratory Society guidelines [13].

Study Groups and End Points

Two study groups were identified among the study population according to the presence of risk factors for acquiring MDR bacteria: group A, patients without risk factors, and group B, patients with ≥1 risk factors for resistant pathogen. The microbiological end point was the actual isolation of an MDR pathogen. The clinical end point was the in-hospital mortality.

Statistical Analysis

All data were statistically analyzed using SPSS (version 18.0) for Mac. Descriptive statistics were reported at baseline, with continuous data expressed as a mean ± SD and categorical data expressed as counts. Patient characteristics were compared between group A and group B: all continuous explanatory variables were presented as means, with differences between the 2 groups compared by means of independent t tests. Categorical explanatory variables were summarized as frequencies and percentages, with differences between the 2 groups analyzed using the χ² test and Fisher exact test when appropriate. Risk factors for acquiring MDR bacteria independently associated with the actual presence of a resistant pathogen were evaluated by a logistic regression model, using the forward method. Goodness of fit was explored based on the Hosmer-Lemeshow test. Interactions between terms within the logistic model were also tested. Based on the logistic regression findings, a predictive additive scoring tool was developed to identify the presence of MDR pathogens. Coefficients from the logistic regression were converted to whole integers and 0.5 points were defined for the presence of ≥1 risk factors not included in the final regression model. Risk classes (low vs high) were defined by the inspection of the prevalence of MDR pathogens given the different score values. The predictive value of the scoring tool was explored for correctly indicating the presence of MDR pathogens via a receiver-operating characteristic (ROC) curve. Independent predictors for in-hospital mortality were evaluated in the entire study population by a logistic regression model, using the forward method. All tests were 2-tailed, and a P value <.05 was considered statistically significant.

RESULTS

A total of 935 consecutive patients with pneumonia were enrolled during the study period (54% males, mean ± SD age: 76 ± 15 years). Demographics; severity of disease; and clinical, laboratory, and radiological findings on admission of the study population are summarized in Table 1. Within the study population, 473 patients (51%) had ≥1 risk factors for acquiring MDR bacteria on admission (Table 2). A total of 271 patients (29%) had 1 risk factor, 129 (14%) had 2 risk factors, 54 (6%) had 3 risk factors, and 19 (2%) had ≥4 risk factors. The most common associations among risk factors were previous hospitalization plus previous antimicrobial therapy (10%) and previous hospitalization plus immunosuppression (8%).

Table 1.

Demographics; Severity of Disease; and Clinical, Laboratory, and Radiological Findings on Admission of the Study Population, According to the Presence of Risk Factors for Multidrug-Resistant Organisms (Group A: Absence of Risk Factors; Group B: Presence of ≥1 Risk Factors)

Characteristic	Study Population	Group A	Group B	P Valuea
No. (%)	935 (100)	462 (49)	473 (51)
Male, no. (%)	504 (54)	225 (49)	279 (59)	.001
Age, years, mean ± SD	76 ± 15	76 ± 16	76 ± 13	.627
Comorbidities, no. (%)
Congestive heart failure	264 (28)	117 (25)	147(31)	.050
COPD	270 (29)	121 (26)	149 (32)	.071
Diabetes mellitus	140 (15)	64 (14)	76 (16)	.360
Cerebrovascular disease	40 (4.3)	19 (10)	21 (11)	.517
Chronic renal failureb	147 (16)	62 (13)	85 (18)	.059
Liver disease	53 (5.7)	16 (3.5)	37 (8)	.004
Severity on admission, no. (%)
PSI risk class IV–V	711 (76)	309 (67)	402 (85)	<.001
CURB-65 score 3, 4, and 5	352 (38)	147 (32)	205 (43)	<.001
Altered mental status	255 (27)	110 (24)	145 (31)	.019
Severe CAP	348 (37)	135 (29)	213 (45)	<.001
Need of ventilatory support	99 (11)	51 (11)	48 (10)	.750
Need of blood pressure support	101 (11)	48 (10)	53 (11)	.200
Severe sepsis	243 (26)	96 (21)	147 (31)	<.001
Physical findings on admission, no. (%)
Hypotensionc	180 (19)	71 (16)	109 (23)	.003
Heart rate, beats/min, mean ± SD	97 ± 22	96 ± 21	98 ± 22	.264
Alteration of gas exchanged	416 (45)	207 (48)	209 (48)	.609
SpO2 %, mean ± SD	92 ± 6	93 ± 6	92 ± 6	.074
Laboratory values, mean ± SD
Arterial pH	7.44 ± 0.08	7.44 ± 0.08	7.44 ± 0.08	.591
PaCO2, mm Hg	36 ± 11	36 ± 11	36 ± 12	.681
PaO2, mm Hg	64 ± 20	65 ± 20	63 ± 20	.270
WBC count, cells/L−1	12 374 ± 7134	12 089 ± 5227	12 651 ± 8594	.229
Platelet count, cells/L−1	232 487 ± 115 443	231 257 ± 91 022	233 701 ± 135 386	.748
Hemoglobin, g/dL	12.6 ± 2	13 ± 1.8	12.1 ± 2.2	<.001
Hematocrit, %	39 ± 13	37 ± 6	41 ± 16	.001
Urea, mg/dL	62 ± 44	57 ± 39	66 ± 49	.003
Creatinine, mg/dL	1.4 ± 1.6	1.2 ± 0.7	1.5 ± 2.1	.013
Sodium, mEq/L	137 ± 6	137 ± 6	137 ± 6	.368
C-reactive protein, g/dL	13 ± 12	13 ± 13	13 ± 12	.524
Chest radiographic findings, no. (%)
Multilobar involvement	332 (36)	148 (44)	184 (48)	.296
Pleural effusion	316 (34)	143 (31)	173 (37)	.072

Abbreviations: CAP, community-acquired pneumonia; COPD, chronic obstructive pulmonary disease; CURB-65, confusion, urea nitrogen, respiratory rate, blood pressure, 65 years of age and older; PaCO₂, arterial partial pressure of carbon dioxide; PaO₂, partial pressure of oxygen in arterial blood; PSI, pneumonia severity index; SD, standard deviation; SpO₂, oxygen saturation; WBC, white blood cell.

a

Difference between group A and group B.

b

Chronic renal failure defined as creatinine >1.2 mg/dL.

c

Hypotension defined as systolic blood pressure <90 mm Hg or diastolic blood pressure <60 mm Hg.

d

Alteration of gas exchange defined as PaO₂ <60 mm Hg, PaO₂/fraction of inspired oxygen <300, or O₂ saturation <90%.

Table 2.

Risk Factors for Multidrug-Resistant Pathogens Among the Study Population

Risk Factor for MDR	Prevalence, No. (%)
Risk factors for HCAP	284 (30)
Hospitalization for ≥2 days in the preceding 90 days	200 (21)
Residency in a nursing home or extended-care facility	66 (7)
Home infusion therapy (including antibiotics)	39 (4)
Home wound care	36 (4)
Chronic dialysis within 30 days	8 (0.9)
Family member with MDR pathogen	0 (0)
Immunosuppression	267 (29)
Severe immunosuppressiona	135 (15)
Mild-to-moderate immunosuppressionb	132 (14)
Other
Antimicrobial therapy in preceding 90 days	155 (17)

Abbreviations: HCAP, healthcare-associated pneumonia; MDR, multidrug resistant.

a

Severe immunosuppression defined as hematologic malignancy, transplantation, immunosuppressive therapy, chemotherapy, radiotherapy.

b

Mild-to-moderate immunosuppression defined as chronic systemic steroid therapy (prednisone ≥25 mg/day), solid malignancy, splenectomy, autoimmune diseases.

Demographics; comorbidities; severity of disease; and physical, laboratory, and radiological findings on admission of patients with and without risk factors are depicted in Table 1. Patients in group B showed a more severe disease on admission compared with group A, with a higher prevalence of severe sepsis. S. pneumoniae was the most common pathogen isolated in both study groups. A higher prevalence of MDR bacteria was found in patients in group B compared with those in group A (6.1% vs 0.9%, respectively; P < .001) (Table 3). Among patients with an isolated resistant pathogen, 7 had bacteremia on admission: 3 due to ESBL-positive Escherichia coli, 2 due to MRSA, 1 due to Providenciastuartii, and 1 due to Proteus species. A combined etiology of resistant pathogens was identified in 3 patients among those in group B: MRSA plus S. maltophilia in 2 cases and ESBL-positive E. coli plus MRSA in 1 case.

Table 3.

Microbiological Findings and Empiric Antibiotic Therapy of the Study Population and According to the Presence of Risk Factors for Multidrug-Resistant Pathogens (Group A: Absence of Risk Factors; Group B: Presence of ≥1 Risk Factors)

Characteristic	Study Population	Group A	Group B
No. (%)	935 (100)	462 (49)	473 (51)
Microbiological finding, no. (%)
Blood culture performed	500 (53)	244 (53)	256 (54)
Patients with isolated bacteria	170 (18)	73 (16)	97 (21)
Streptococcus pneumoniae	63 (37)	27 (37)	36 (37)
Methicillin-susceptible Staphyloccocus aureus	21 (12)	9 (12)	12 (12)
Methicillin-resistant S. aureus	16 (9.4)	2 (2.7)	14 (14)
Legionella pneumophila	26 (15)	14 (19)	12 (12)
Escherichia coli ESBL+	5 (2.9)	1 (1.4)	4 (4.1)
E. coli ESBL–	10 (5.9)	3 (4.1)	7 (7.2)
Pseudomonas aeruginosa MDR+	7 (4.1)	0	7 (7.2)
P. aeruginosa MDR–	5 (2.9)	1 (1.4)	4 (4.1)
Klebsiella pneumoniae ESBL+	0	0	0
K. pneumoniae ESBL–	13 (7.6)	6 (8.2)	7 (7.2)
Haemophilus influenzae	6 (3.5)	5 (6.8)	1 (1)
Mycoplasma pneumoniae	5 (2.9)	3 (4.1)	2 (2.1)
Chlamydophila pneumoniae	4 (2.4)	4 (5.5)	0
Bordetella bronchiseptica	1 (0.6)	0	1 (1)
Stenotrophomonas maltophilia	2 (1.2)	0	2 (2.1)
Enterococcus MDR+	1 (0.6)	0	1 (1)
Enterococcus MDR–	2 (1.2)	1 (1.4)	1 (1)
Proteus mirabilis ESBL+	2 (1.2)	1 (1.4)	1 (1)
Providencia stuartii ESBL+	1 (0.6)	0	1 (1)
Acinetobacter baumanii	1 (0.6)	0	1 (1)
Patients with ≥1 MDR organisms	33 (3.3)	4 (0.9)	29 (6.1)
Polymicrobial infection	17 (1.8)	4 (0.9)	13 (2.8)
Initial empiric antibiotic treatment, no. (%)
Ceftriaxone	434 (46)	278 (60)	156 (33)
Azithromycin	384 (41)	250 (54)	134 (28)
Levofloxacin	288 (31)	116 (25)	172 (36)
Piperacillin/tazobactam	188 (20)	54 (12)	134 (28)
Vancomycin	22 (2)	3 (0.7)	19 (4)
Ampicillin/sulbactam	29 (3)	18 (4)	11 (2)
Metronidazole	25 (3)	12 (3)	13 (3)
Ceftazidime	16 (2)	4 (1)	12 (2)
Imipenem	16 (2)	5 (1)	11 (2)
Amikacin	13 (1)	2 (0.4)	11 (2)
Clarythromycin	13 (1)	8 (2)	5 (1)
Ciprofloxacin	8 (1)	1 (0.2)	7 (2)
Others	16 (2)	9 (2)	7 (2)
Compliant with ERS guidelines	672 (72)	370 (80)	302 (64)

Abbreviations: ERS, European Respiratory Society; ESBL, extended-spectrum β-lactamase; MDR, multidrug resistant.

A multivariable logistic regression model was performed for the 170 patients who had a bacterium isolated. Among all risk factors for acquiring MDR bacteria, hospitalization in the preceding 90 days (odds ratio [OR], 4.87 [95% confidence interval {CI}, 1.90–12.4]; P = .001) and residency in a nursing home or extended-care facility (OR, 3.55 [95% CI, 1.12–11.24]; P = .031) were independent predictors for an actual infection with a resistant pathogen, after adjustment for sex, age, and comorbidities such as diabetes, cerebrovascular disease, chronic renal failure (OR, 3.90 [95% CI, 1.35–11.99]; P = .014), and chronic obstructive pulmonary disease (Hosmer-Lemeshow test, 0.829; Nagelkerke R², 0.299; omnibus χ² test, 0.021).

A score for predicting the risk of infection with resistant bacteria, including factors related to contact with the healthcare environment as well as patients’ comorbidities, was computed (Table 4). The scores ranged from 0 to 12.5. Based on visual inspection, patients were grouped into low-risk and high-risk classes as a function of their overall score (Figure 1). Among patients with a score ≤0.5 on entry, the prevalence of a resistant bacteria was 8% (95% CI, 2%–13%), compared with 38% (95% CI, 25%–50%) in those with a score of ≥3 (P < .001). Figure 2 depicts the ROC curve for the score. The area under the ROC curve is 0.79 (95% CI, .71–.87). A score >0.5 was associated with the best balance between sensitivity (0.75) and specificity (0.71).

Table 4.

Scoring System to Evaluate the Presence of Multidrug-Resistant Pathogens in Patients With Pneumonia From the Community Who are Hospitalized

Variable	Score
No risk factors for MDR pathogen (including comorbidities)	0
≥1 of the following: cerebrovascular disease, diabetes, COPD, antimicrobial therapy in preceding 90 days, immunosuppression, home wound care, home infusion therapy (including antibiotics)	0.5
Residence in a nursing home or extended-care facility	3
Hospitalization for ≥2 days in the preceding 90 days	4
Chronic renal failure	5

Abbreviations: COPD, chronic obstructive pulmonary disease; MDR, multidrug-resistant.

Figure 1.

Prevalence of multidrug-resistant bacteria in patients with an isolated pathogen, according to the stratification derived from the score (low-risk and high-risk classes).

Figure 2.

Receiver-operating characteristic curve of the score.

Data on the initial antimicrobial treatment are given in Table 3. Of the 33 patients who had ≥1 MDR organisms, the pathogen was susceptible to the empiric antibiotic therapy in 22 subjects (66%). Mean ± SD length of stay in the hospital was 15 ± 11 days for the entire study population: patients in group B had a longer hospital stay compared with those in group A (15.3 ± 12.3 vs 13.8 ± 8.9 days, respectively; P = .037). In-hospital mortality for the entire study population was 16% (n = 161). Among patients in group A (n = 48), mortality was 10%; among those in group B (n = 104), mortality was 22% (P < .001). Mortality was 48% for patients coming from a nursing home and 26% for patients who were hospitalized in the preceding 90 days.

For the entire study population, a multivariable logistic regression model was used to analyze all risk factors for acquiring MDR bacteria. Hospitalization in the preceding 90 days (OR, 1.63 [95% CI, 1.04–2.54]; P = .034) and residency in a nursing home or extended-care facility (OR, 2.83 [95% CI, 1.54–5.2]; P = .001) were found to be independent predictors for in-hospital mortality, after adjustment for age, sex, PSI, severe CAP, severe sepsis on admission, and appropriate antibiotic treatment (Nagelkerke R², 0.234; omnibus χ² test, 0.024) (Table 5).

Table 5.

Independent Predictors for In-Hospital Mortality in the Study Population

Variable	OR (95% CI)	P Value
Hospitalization for ≥2 days in the preceding 90 days	1.63 (1.04–2.54)	.034
Residency in a nursing home or extended-care facility	2.83 (1.54–5.21)	.001
Pneumonia severity index	2.19 (1.58–3.03)	<.001
Severe CAP	2.52 (1.61–3.93)	<.001

Abbreviations: CAP, community-acquired pneumonia; CI, confidence interval; OR, odds ratio.

DISCUSSION

Our study shows that more than half of the patients who were admitted to the hospital from the community because of an episode of pneumonia had risk factors for MDR. Of those patients, hospitalization in the preceding 90 days and residency in a nursing home or extended-care facility were independently associated with an actual infection with a resistant pathogen, as well as in-hospital mortality. A simple score performed on admission to the hospital that included risk factors and comorbidities was used to stratify patients into different classes based on the probability of having MDR pneumonia.

Our results argue in favor of an individual evaluation of each patientin order to develop a targeted approach when selecting empiric antibiotic therapy for those patients with CAP. Among known risk factors for acquiring MDR bacteria, hospitalization in the preceding 90 days and residency in a nursing home or extended-care facility were independently associated with pneumonia caused by a resistant pathogen. These findings are in accordance with recent experiences that showed both risk factors to be related to infection with MDR bacteria in different populations, including patients with acute respiratory failure and those admitted to an intensive care unit [6, 7, 14–16]. Although our findings were observed only among patients with an isolated pathogen, previous hospitalization and nursing home residency were also found to be significantly associated with in-hospital mortality among the entire cohort. The double impact of these 2 risk factors on both microbiological and clinical outcomes emphasizes their roles.

There are possible 2 reasons for the impact of previous hospitalization and nursing home residency on resistant pathogen infection. First, the impact could be related to exposure to an extensive antibiotic coverage in these settings that leads to a selecting pressure for resistance. Second, the persistence of MDR pathogens in different wards and transmission between healthcare workers and patients is increasing, and effective healthcare policies are needed to reduce these pathogens [17].

In our population, immunosuppression did not seem to convey an increased risk for infection with a resistant pathogen. This finding is intriguing, although in accordance with recent literature [7]. Immunosuppression needs to be evaluated as the expression of different disorders, and its association with MDR is based on the type of disease leading to immunosuppression, its severity, and the effectiveness of the treatment chosen. Based on our results, contact with the healthcare system seems to play a more important role in the acquisition of a resistant pathogen than immunosuppression in patients with pneumonia.

We also found chronic renal failure to be an independent risk factor for MDR infection in our cohort. This association has been previously demonstrated in patients infected by Mycobacterium tuberculosis and other bacteria [18, 19]. We can also speculate that chronic renal failure represents a window on a patient’s functional status. Our finding could reinforce the hypothesis that functional impairment is a crucial determinant of the risk for acquiring drug-resistant pathogens, as recently suggested [3].

As an alternative to a large classification that includes different risk factors, we suggest a probabilistic approach in assuming the presence of MDR-causing pneumonia. The model we developed has the advantage of taking into consideration both the number/type of comorbidities and risk factors and the interaction among them. Recent literature indicates a shift toward this approach [7, 16]. A score to predict MDR pneumonia will allow physicians to develop both diagnostic and treatment protocols. On the one hand, a more rigorous and invasive microbiological workup could be indicated for those patients in the high-risk class. On the other hand, the administration of appropriate empiric antibiotic therapy could be optimized, thus minimizing the unnecessary use of broad-spectrum antibiotics in patients in the low-risk class.

Our study had some limitations. We found a low prevalence of MDR pathogens in our cohort; this finding could be related to the single-center design of the study as well as our healthcare organization and internal policies. Our results should thus be interpreted with caution because different causative organisms or rates of antibiotic resistance may be encountered in other countries. Furthermore, we were not able to directly detect our patients’ functional status, and some characteristics of the study population may have limited the ability to identify disease severity as one of the risk factors for MDR infection. In further multicenter studies enrolling a large number of patients, more variables could be evaluated and included in robust prediction models to identify the relationship among different risk factors for MDR organisms. Finally, although our scoring tool appears to perform quite well in identifying patients coming from the community with pneumonia caused by resistant microorganisms, we should acknowledge the absence of a validation of the model in an independent group of patients.

The strength and novelty of our prospective study rely on a specific analysis of all risk factors for acquiring MDR bacteria (including immunosuppression and comorbidities) in a large population of consecutive patients coming from the community with pneumonia and who were hospitalized in different wards. Furthermore, the analysis of risk factors for acquisition of MDR bacteria has been weighted based on both microbiological and clinical outcomes.

Pneumonia caused by an MDR pathogen acquired in the community depends on both patient comorbidities/functional status and previous contact with the healthcare system. However, a different weight of risk factors for MDR should be acknowledged because previous hospitalization and nursing home residency are the main factors leading to both resistant pathogen acquisition and mortality. We suggest that a probabilistic approach to identifying resistant pathogens among patients coming from the community with pneumonia should integrate previous classifications.

Notes

Acknowledgments.

We thank all the patients and clinicians in the Policlinico Hospital in Milan, Italy, for their cooperation and participation in the study.

Potential conflicts of interest.

All authors: No reported conflicts.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.