Risk Factors for Community-Onset Pneumonia Caused by Levofloxacin-Nonsusceptible Streptococcus pneumoniae

Abstract

Background:

Fluoroquinolones are antibiotics commonly used in the treatment of infections caused by Streptococcus pneumoniae. However, rates of fluoroquinolone resistance are increasing with their frequent use. We designed this study to verify current fluoroquinolone resistance rates and risk factors for community-onset pneumococcal pneumonia.

Methods:

A retrospective case–control study was conducted in a tertiary referral hospital. The study population comprised patients admitted for pneumococcal pneumonia between January 2011 and May 2017. The case group included community-onset pneumonia caused by levofloxacin-nonsusceptible S. pneumoniae. The control group consisted of two patients with levofloxacin-susceptible S. pneumoniae who were admitted around the same time as each case.

Results:

A total of 198 pneumococcal pneumonia cases were identified during the study period. Twenty-five levofloxacin-resistant S. pneumoniae cases and 3 levofloxacin-intermediate S. pneumoniae cases were included in the case group (nonsusceptibility rate = 14.1%). Multivariate analysis showed that healthcare-associated factors (odds ratio [OR] 4.78, 95% confidence interval [CI] 1.39–16.43, p = 0.013), bronchopulmonary disease (OR 3.79, 95% CI 1.07–13.40, p = 0.039), cerebrovascular disease (OR 6.08, 95% CI 1.24–29.75, p = 0.026), and exposure to fluoroquinolones within the previous 3 months (OR 5.89, 95% CI 1.21–28.68, p = 0.028) were associated with nonsusceptibility to levofloxacin.

Conclusion:

Independent risk factors for levofloxacin-nonsusceptible pneumococcal pneumonia were recent hospitalization, bronchopulmonary disease, cerebrovascular disease, and prior antibiotic use within 3 months. Careful selection of empirical antibiotics is thus needed in at-risk patients. Similarly, efforts to prevent the interpersonal spread of drug-resistant pathogens in long-term care facilities and to restrict unnecessary fluoroquinolone prescriptions are important.

Introduction

Pneumonia remains a major cause of death worldwide, especially in patients younger than 5 years and older than 75 years.¹ Streptococcus pneumoniae is the most common cause of community-onset bacterial pneumonia.² Empirical treatment options for community-onset pneumococcal pneumonia have changed according to antimicrobial resistance trends. Since the emergence of the first antibiotic-resistant S. pneumoniae strain in the 1960s, resistance rates to various antibiotics (including penicillin, macrolides, and aminoglycosides) have increased globally.³ The resistance rate of S. pneumoniae to penicillin ranges from 30% to 70% in many countries.^4,5 Newer fluoroquinolones (such as levofloxacin and moxifloxacin) have replaced prior antibiotics used to treat community-onset pneumonia, and they have enhanced activity against Gram-positive pathogens.⁶ However, rates of fluoroquinolone-resistant S. pneumoniae have also increased, especially in Asian region such as Korea and Hong Kong.^7–9 Previous studies have shown that risk factors for colonization or infection with levofloxacin-resistant S. pneumoniae are prior exposure to fluoroquinolones, chronic obstructive lung disease, cerebrovascular disease, residence in a nursing home or public shelter, human immunodeficiency virus infection, and heavy smoking.^10–12 We designed this study to evaluate the current epidemiology of levofloxacin-resistant S. pneumoniae and to better understand the risk factors for acquiring fluoroquinolone resistance in Korea.

Materials and Methods

Setting and patients

A retrospective case–control study was conducted by reviewing adult patients admitted to the Samsung Medical Center, a tertiary care university hospital in the Republic of Korea, for the treatment of community-onset pneumococcal pneumonia between January 2011 and May 2017. All case patients were microbiologically confirmed via the identification of S. pneumoniae in clinical specimens. The case group included pneumonia caused by levofloxacin-nonsusceptible S. pneumoniae. The control group in the 1:2 case–control study comprised two patients who were hospitalized around the same time as case patients, for pneumonia caused by levofloxacin-susceptible S. pneumoniae. Institutional Review Board approvals occurred.

Data collection and definitions

The following data were obtained from electronic medical records during the study period: baseline demographic data, underlying disease, microbiological information, and history of exposure to antibiotics before hospitalization. “Healthcare-associated factor” was used to describe already-ill nursing-home residents, patients in long-term care, patients undergoing a same-day procedure, patients receiving home- or hospital-based intravenous therapy, and patients undergoing dialysis. Microbiological information included laboratory identification and antimicrobial susceptibility tests from blood and respiratory specimens in the study population. Susceptibility tests were interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines.¹³ For levofloxacin, MIC ≤2 was interpreted as sensitive, 4 as intermediate, and ≥8 as resistant.

Statistical analyses

Continuous data were described as mean ± standard deviation for normally distributed data, and as median and interquartile range for skewed data. The Pearson χ² test and Fisher's exact test were used for categorical variables, and the Student's t-test and Mann–Whitney U test were used for continuous variables when comparing two groups. The Cox regression was used for risk factor analysis. All analyses were two-tailed, and a p-value <0.05 was considered statistically significant.

Results

A total of 198 pneumococcal pneumonia patients were identified during the study period from January 2011 to May 2017. Twenty-five S. pneumoniae were resistant to levofloxacin and 3 were intermediate from the study population. The rate of levofloxacin-nonsusceptible S. pneumoniae was 14.1% (1.5% intermediate, 12.6% resistant). Clinical characteristics between the case and the control groups are shown in Table 1. Age and sex distribution were not different between the two groups. There were more bacteremic patients in the control group. More than half of the patients in the case group had healthcare-associated factors. In the case group, 15 patients had underlying bronchopulmonary disease, which was significantly more than the control group. Tracheostomy tubes were placed in six patients with cerebrovascular disease (among nine) in the case group. The median number of comorbidities was higher in the case group than in the control group. There were no differences in the other underlying diseases between the two groups. Previous exposure to any antibiotic before hospitalization was significantly different between the two groups. Eighteen patients (64.3%) in the case group had a history of antibiotic use in the 3 months before hospitalization.

Table 1.

Clinical Characteristics and Factors Associated with Community-Onset Pneumonia Caused by Levofloxacin-Nonsusceptible Streptococcus pneumoniae

	Levofloxacin-nonsusceptible (n = 28)	Levofloxacin-susceptible (n = 56)	OR (95% CI)	p
Advanced age (≥65 years)	19 (67.9%)	33 (58.9%)	1.471 (0.566–3.824)	0.428
Male sex	25 (89.3%)	42 (75.0%)	2.778 (0.726–10.626)	0.136
Bacteremia	5 (17.9%)	23 (41.1%)	0.312 (0.103–0.941)	0.039
Healthcare-associated factors	22 (78.6%)	22 (39.3%)	6.111 (2.133–17.505)	0.001
Underlying disease
Bronchopulmonary	15 (53.6%)	15 (26.8%)	3.154 (1.220–8.150)	0.018
Diabetes mellitus	6 (21.4%)	13 (23.2%)	0.818 (0.276–2.425)	0.717
Cerebrovascular	9 (32.1%)	4 (7.1%)	6.158 (1.696–22.363)	0.006
Nonhematologic malignancy	11 (39.3%)	23 (41.1%)	0.863 (0.342–2.175)	0.754
Hematologic malignancy	2 (7.1%)	8 (14.3%)	0.641 (0.121–3.402)	0.602
Chronic renal disease	1 (3.6%)	4 (7.1%)	0.481 (0.051–0.481)	0.523
Chronic liver disease	1 (3.6%)	2 (3.6%)	1.000 (0.087–11.525)	1.000
Congestive heart failure	2 (7.1%)	0		0.999
Solid organ transplantation	0	2 (3.6%)		0.999
Neutropenia	1 (3.6%)	1 (1.8%)	2.037 (0.123–33.829)	0.620
Steroid	6 (21.4%)	4 (7.1%)	3.545 (0.910–13.811)	0.068
Tracheostomy	6 (21.4%)	0		0.999
Previous vaccination	4 (14.3%)	5 (8.9%)	2.167 (0.499–9.403)	0.302
Numbers of comorbidities	2 (1–3)	1 (1–2)
Antibiotic use within 3 months	18 (64.3%)	14 (25.0%)	5.400 (2.024–14.410)	0.001
Quinolone	12 (42.9%)	3 (5.4%)	13.250 (3.323–52.835)	<0.001
Penicillin	8 (28.6%)	4 (7.1%)	5.200 (1.408–19.201)	0.013
Cephalosporin	9 (32.1%)	7 (12.5%)	3.316 (1.081–10.171)	0.036

Potential risk factors for contracting levofloxacin-nonsusceptible S. pneumoniae are also shown in Table 1. In a univariate analysis, the following variables were associated with levofloxacin-nonsusceptible S. pneumoniae: healthcare-associated factors (odds ratio [OR] 6.11, 95% confidence interval [CI] 2.13–17.51, p = 0.001), bronchopulmonary disease (OR 3.15, 95% CI 1.22–8.15, p = 0.018), and cerebrovascular disease (OR 6.16, 95% CI 1.70–22.36, p = 0.006). Exposure to antibiotics within 3 months before hospitalization were also related to the acquisition of levofloxacin-nonsusceptibility. The results according to the types of antibiotics were as follows: any antibiotic (OR 5.40, 95% CI 2.02–14.41, p = 0.001), fluoroquinolone (OR 13.25, 95% CI 3.32–52.84, p < 0.001), penicillin (OR 5.20, 95% CI 1.41–19.20, p = 0.013), and cephalosporin (OR 3.32, 95% CI 1.08–10.17, p = 0.036). Other variables related to immunologic factors (such as diabetes mellitus, malignancy, solid organ transplantation, and steroid use) were not significantly associated with nonsusceptibility. In a multivariate analysis, healthcare-associated factors (OR 4.78, 95% CI 1.39–16.43, p = 0.013), bronchopulmonary disease (OR 3.79, 95% CI 1.07–13.40, p = 0.039), cerebrovascular disease (OR 6.08, 95% CI 1.24–29.75, p = 0.026), and prior exposure to fluoroquinolone within 3 months (OR 5.89, 95% CI 1.21–28.68, p = 0.028) were all related to levofloxacin-nonsusceptible S. pneumoniae (Table 2). Previous exposure to penicillin (OR 1.32, 95% CI 0.28–6.36, p = 0.728) or cephalosporin (OR 2.48, 95% CI 0.64–9.60, p = 0.190) within 3 months were not significantly related to levofloxacin-nonsusceptible S. pneumoniae. Similarly, exposure to fluoroquinolones at least 6 months before hospitalization was not associated with the acquisition of levofloxacin nonsusceptibility (OR 3.14, 95% CI 0.63–15.72, p = 0.163).

Table 2.

Independent Risk Factors Associated with Pneumonia Caused by Levofloxacin-Nonsusceptible Streptococcus pneumoniae

	Adjusted OR (95% CI)	p
Healthcare-associated factors	4.78 (1.39–16.43)	0.013
Bronchopulmonary disease	3.79 (1.07–13.40)	0.039
Cerebrovascular disease	6.08 (1.24–29.75)	0.026
Fluoroquinolone treatment within the prior 3 months	5.89 (1.21–26.68)	0.028

The median follow-up duration for the whole study population was 540 days. The overall 90-day mortality rate was 11.9%. The mortality rate of the levofloxacin-nonsusceptible group (17.9%, 5 of 28) was higher than that of the susceptible group (8.9%, 5 of 56), but without significant difference. Six patients in the case group and three patients in the control group were excluded in the outcome analysis because of follow-up loss within 90 days.

Discussion

An increased rate of resistance to levofloxacin is a current trend caused by its frequent use. It is troublesome because S. pneumoniae is the most common pathogen in community-onset bacterial pneumonia. Identifying risk factors for contracting levofloxacin-nonsusceptible S. pneumoniae is critical, both for effective patient treatment and for prudent antibiotic stewardship. This study has demonstrated that risk factors for contracting levofloxacin-nonsusceptible S. pneumoniae include bronchopulmonarydisease, healthcare-associated factors, cerebrovascular disease, and recent exposure (within 3 months) to fluoroquinolones.

The relationship between chronic obstructive pulmonary disease and levofloxacin-resistant S. pneumoniae has been described in previous studies.^14–16 One theory is that a relatively high concentration of S. pneumoniae in sputum may contribute to resistance against levofloxacin after fluoroquinolone therapy. In this study, 9 of 15 patients with bronchopulmonary disease (including 1 patient in the control group) had been treated with fluoroquinolones in the 3 months before hospitalization. Thus, although previous exposure to fluoroquinolones cannot be ignored, underlying bronchopulmonary disease may also promote the acquisition of resistance to fluoroquinolones. Another hypothesis is that multidrug-resistant pathogens are acquired through nosocomial spread.^10,15 In our study, 11 of 15 bronchopulmonary patients in the case group and 5 of 15 bronchopulmonary patients in the control group had healthcare-associated factors. Half of the patients with both bronchopulmonary disease and healthcare-associated factors had no previous exposure to fluoroquinolones. This supports the possibility that levofloxacin-resistant S. pneumoniae is spread interpersonally in healthcare settings.

Healthcare-associated factors were independently associated with levofloxacin-nonsusceptible S. pneumoniae in this study. The relationship between drug-resistant pathogens and healthcare-associated factors is well known.^17–19 This is due, in part, to the increased chance of exposure to antibiotics and to person-to-person transmission in nosocomial settings. Underlying cerebrovascular disease was another risk factor for the acquisition of nonsusceptibility to levofloxacin. Eight of the nine case patients with cerebrovascular disease shared a healthcare-associated factor, and one had a prior exposure to fluoroquinolones within 3 months. Cerebrovascular disease may often occur alongside healthcare-associated factors, but further study is needed to identify a clear correlation. The presence of a tracheostomy collar was notable in that all S. pneumoniae isolates from patients with tracheostomy tubes were nonsusceptible to levofloxacin, despite there being no significant association between tracheostomy and the acquisition of resistance. This result is similar to that of a previous study.²⁰ Although all patients with tracheostomy tubes had healthcare-associated factors, a direct, causal relationship has yet to be established.

Previous exposure to fluoroquinolones is a known risk factor for developing levofloxacin-nonsusceptibility; this occurs through selective pressure for mutations in the quinolone resistance-determining regions of the gyrA, parC, or parE genes.²¹ In this study, fluoroquinolone exposure within 3 months was associated with nonsusceptibility, while exposure within 6 months was not. This is notable because there is no defined time frame for loss of resistance. Further study is needed to establish when and how acquired resistance disappears, with the relevant molecular and genetic characterizations.

The rate of S. pneumoniae nonsusceptibility to levofloxacin was 14.1% in this study. This is a higher rate than previous studies. In a prior nested case–control study of Korea, Taiwan, and Hong Kong, the nonsusceptibility rate was 4.7%.¹² In a recent study of China, the rate of nonsusceptibility to levofloxacin was 1.0%.²² In North America, the rate of S. pneumoniae resistance to levofloxacin remains <2.0%.^23,24 Although high resistance rates may be caused by selection bias in tertiary referral hospitals, risk factor analysis performed in regions with high resistance are reliable.

In conclusion, this study revealed a high rate of nonsusceptibility to levofloxacin in S. pneumoniae. Potential risk factors include healthcare-associated factors, pre-existing bronchopulmonary disease, cerebrovascular disease, and exposure to fluoroquinolones within 3 months before hospitalization. Careful selection of empirical antibiotics is thus needed in at-risk patients. Moreover, efforts to prevent the interpersonal spread of drug-resistant pathogens in long-term care facilities and to restrict unnecessary fluoroquinolone use are important.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

O'Brien

K.L.

, Wolfson

L.J.

, Watt

J.P.

, Henkle

, Deloria-Knoll

, McCall

, Lee

, Mulholland

, Levine

O.S.

, and Cherian

. 2009. Burden of disease caused by Streptococcus pneumoniae in children younger than 5 years: global estimates. Lancet (London, England)., 374:893–902.

Rozenbaum

M.H.

, Pechlivanoglou

, van der Werf

T.S.

, Lo-Ten-Foe

J.R.

, Postma

M.J.

, and Hak

. 2013. The role of Streptococcus pneumoniae in community-acquired pneumonia among adults in Europe: a meta-analysis. Eur. J. Clin. Microbiol. Infect. Dis., 32:305–316.

Appelbaum

P.C.

1987. World-wide development of antibiotic resistance in pneumococci. Eur. J. Clin. Microbiol., 6:367–377.

Mamishi

, Moradkhani

, Mahmoudi

, Hosseinpour-Sadeghi

, and Pourakbari

. 2014. Penicillin-resistant trend of Streptococcus pneumoniae in Asia: a systematic review. Iran J. Microbiol., 6:198–210.

El Moujaber

, Osman

, Rafei

, Dabboussi

, and Hamze

. 2017. Molecular mechanisms and epidemiology of resistance in Streptococcus pneumoniae in the Middle East region. J. Med. Microbiol., 66:847–858.

Odenholt

, and Cars

. 2006. Pharmacodynamics of moxifloxacin and levofloxacin against Streptococcus pneumoniae, Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli: simulation of human plasma concentrations after intravenous dosage in an in vitro kinetic model. J. Antimicrob. Chemother., 58:960–965.

Lee

, Kim

S.H.

, Park

, and Bae

. 2013. High prevalence of multiresistance in levofloxacin-nonsusceptible Streptococcus pneumoniae isolates in Korea. Diagn. Microbiol. Infect. Dis., 76:227–231.

Kim

S.H.

, Bae

I.K.

, Park

, Lee

, Kim

N.Y.

, Song

S.A.

, Kim

H.R.

, Jeon

G.W.

, Urm

S.H.

, and Shin

J.H.

. 2016. Serotype distribution and antimicrobial resistance of Streptococcus pneumoniae isolates causing invasive and noninvasive pneumococcal diseases in Korea from 2008 to 2014. BioMed. Res. Int., 2016:6950482.

P.L.

, Cheng

V.C.

, and Chow

K.H.

. 2009. Decreasing prevalence of levofloxacin-resistant Streptococcus pneumoniae in Hong Kong, 2001 to 2007. J. Antimicrob. Chemother., 63:836–838.

10.

P.L.

, Tse

W.S.

, Tsang

K.W.

, Kwok

T.K.

, Ng

T.K.

, Cheng

V.C.

, and Chan

R.M.

. 2001. Risk factors for acquisition of levofloxacin-resistant Streptococcus pneumoniae: a case-control study. Clin. Infect. Dis., 32:701–707.

11.

Isea-Pena

M.C.

, Sanz-Moreno

J.C.

, Esteban

, Fernandez-Roblas

, and Fernandez-Guerrero

M.L.

. 2013. Risk factors and clinical significance of invasive infections caused by levofloxacin-resistant Streptococcus pneumoniae. Infection, 41:935–939.

12.

Kang

C.I.

, Song

J.H.

, Kim

S.H.

, Chung

D.R.

, Peck

K.R.

, So

T.M.

, Hsueh

P.R.

, and Group

A.S.

. 2014. Risk factors for levofloxacin-nonsusceptible Streptococcus pneumoniae in community-acquired pneumococcal pneumonia: a nested case-control study. Eur. J. Clin. Microbiol. Infect. Dis., 33:55–59.

13.

Clinical and Laboratory Standards Institute. 2017. Performance standards for antimicrobial susceptibility testing, document M100-S27. Clinical Laboratory Standards Institute,, Wayne, PA.

14.

Davies

B.I.

, and Maesen

F.P.

. 1999. Clinical effectiveness of levofloxacin in patients with acute purulent exacerbations of chronic bronchitis: the relationship with in-vitro activity. J. Antimicrob. Chemother. 43 Suppl C:83–90.

15.

de Galan

B.E.

, van Tilburg

P.M.

, Sluijter

, Mol

S.J.

, de Groot

, Hermans

P.W.

, and Jansz

A.R.

. 1999. Hospital-related outbreak of infection with multidrug-resistant Streptococcus pneumoniae in the Netherlands. J. Hosp. Infect., 42:185–192.

16.

Perez-Trallero

, Marimon

J.M.

, Larruskain

, Alonso

, and Ercibengoa

. 2011. Antimicrobial susceptibilities and serotypes of Streptococcus pneumoniae isolates from elderly patients with pneumonia and acute exacerbation of chronic obstructive pulmonary disease. Antimicrob. Agents Chemother., 55:2729–2734.

17.

Aliberti

, Di Pasquale

, Zanaboni

A.M.

, Cosentini

, Brambilla

A.M.

, Seghezzi

, Tarsia

, Mantero

, and Blasi

. 2012. Stratifying risk factors for multidrug-resistant pathogens in hospitalized patients coming from the community with pneumonia. Clin. Infect. Dis., 54:470–478.

18.

Shindo

, Ito

, Kobayashi

, Ando

, Ichikawa

, Shiraki

, Goto

, Fukui

, Iwaki

, Okumura

, Yamaguchi

, Yagi

, Tanikawa

, Sugino

, Shindoh

, Ogasawara

, Nomura

, Saka

, Yamamoto

, Taniguchi

, Suzuki

, Saito

, Kawamura

, and Hasegawa

. 2013. Risk factors for drug-resistant pathogens in community-acquired and healthcare-associated pneumonia. Am. J. Respir. Crit. Care Med., 188:985–995.

19.

Shorr

A.F.

, Zilberberg

M.D.

, Micek

S.T.

, and Kollef

M.H.

. 2008. Prediction of infection due to antibiotic-resistant bacteria by select risk factors for health care-associated pneumonia. Arch. Intern. Med., 168:2205–2210.

20.

Manning

, Lautenbach

, Tolomeo

, and Han

J.H.

. 2015. Risk factors for infection with Escherichia coli in nursing home residents colonized with fluoroquinolone-resistant E. coli. Infect. Control Hosp. Epidemiol. 36:575–577.

21.

Perez-Trallero

, Marimon

J.M.

, Gonzalez

, Ercibengoa

, and Larruskain

. 2005. In vivo development of high-level fluoroquinolone resistance in Streptococcus pneumoniae in chronic obstructive pulmonary disease. Clin. Infect. Dis., 41:560–564.

22.

, Zhu

, Wang

, Morrissey

, Wang

, and Torumkuney

. 2016. Results from the Survey of Antibiotic Resistance (SOAR) 2009–2011 and 2013–2014 in China. J. Antimicrob. Chemother. 71 Suppl 1:i33–i43.

23.

Karlowsky

J.A.

, Thornsberry

, Critchley

I.A.

, Jones

M.E.

, Evangelista

A.T.

, Noel

G.J.

, and Sahm

D.F.

. 2003. Susceptibilities to levofloxacin in Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis clinical isolates from children: results from 2000–2001 and 2001–2002 TRUST studies in the United States. Antimicrob. Agents Chemother., 47:1790–1797.

24.

Patel

S.N.

, McGeer

, Melano

, Tyrrell

G.J.

, Green

, Pillai

D.R.

, and Low

D.E.

. 2011. Susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. Antimicrob. Agents Chemother., 55:3703–3708.