Figures
Abstract
Stenotrophomonas maltophilia is an important nosocomial bacterial pathogen, as is Pseudomonas aeruginosa. Differentiation of these bacteria as bacteremic agents is critical in the clinical setting and to define a therapeutic strategy; however, the associated factors and prognosis for S. maltophilia bacteremia have not been fully evaluated to adequately characterize these factors. We first conducted a matched case-control study to clarify these questions. A total of 30 case patients with S. maltophilia bacteremia were compared with 30 control patients with P. aeruginosa bacteremia between January 2005 and August 2014, according to matching criteria based on underlying disease, age, and gender. The 30-day mortality rate for the case patients (53.3%) was significantly higher than that of the control group (30.0%) (P = 0.047, using the log-rank test). Conditional logistic regression analysis showed that the predisposing factors specific for the detection of S. maltophilia bacteremia were indwelling artificial products other than a central venous catheter, ICU stay, and previous use of anti-MRSA drugs. The high severity of illness was associated with mortality in both case and control patients. Interestingly, inappropriate antimicrobial treatment was an additional independent risk factor for mortality in only the case patients with S. maltophilia bacteremia (odds ratio = 13.64, P = 0.048). Monotherapy with fluoroquinolones inactive against the S. maltophilia isolates was mainly responsible for the inappropriate treatment. These results suggest that more precise prediction and more appropriate treatment might improve the prognosis of patients with S. maltophilia bacteremia.
Citation: Sumida K, Chong Y, Miyake N, Akahoshi T, Yasuda M, Shimono N, et al. (2015) Risk Factors Associated with Stenotrophomonas maltophilia Bacteremia: A Matched Case-Control Study. PLoS ONE 10(7): e0133731. https://doi.org/10.1371/journal.pone.0133731
Editor: Zaccaria Ricci, Bambino Gesù Children's Hospital, ITALY
Received: March 8, 2015; Accepted: June 3, 2015; Published: July 24, 2015
Copyright: © 2015 Sumida et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Stenotrophomonas maltophilia has emerged as an important pathogen that induces nosocomial infections [1]. S. maltophilia is a non-fermentative, gram-negative bacillus as well as Pseudomonas aeruginosa and causes severe infectious diseases, particularly bacteremia in the hospital setting, similar to those caused by P. aeruginosa [2,3]. However, the therapeutic strategy for the management of S. maltophilia is very different from that applied to P. aeruginosa, because S. maltophilia is intrinsically resistant to lots of antibiotics, including β-lactams and aminoglycosides [1]. Therefore, in a clinical setting that predisposes patients to S. maltophilia bacteremia, the differentiation between these two causative agents is critical.
A limited number of case-control studies on S. maltophilia bacteremia have been reported [4–10]. Various risk factors and mortality rates were obtained using different control groups [4–8]. Moreover, few studies have investigated P. aeruginosa bacteremia in control groups [9,10]. Adequate matching eliminates the influence of potentially confounding factors for underlying diseases and, in our study, it helped to more practically characterize the predisposing factors for S. maltophilia bacteremia and its prognosis in clinical settings similar to those that favor P. aeruginosa bacteremia. Therefore, we first conducted a matched case-control study using control patients with P. aeruginosa bacteremia. The risk factors for mortality of patients with bacteremia were also evaluated.
Methods
Patients and Ethics Statement
This matched, retrospective, case-control study was conducted at the Kyushu University Hospital, a 1,275-bed tertiary-care hospital in Fukuoka, Japan. Data were collected between January 2005 and August 2014 from electric medical records. First, patients who had positive blood culture for S. maltophilia or P. aeruginosa were selected. The Research Ethics Committee of Kyushu University Hospital approved this study under protocol No. 26–288 and exempted the need for obtaining informed consent from each patient.
Enrollment and Matching Criteria
The analysis of bacteremia data indicated bacteremia due to S. maltophilia in 32 patients and due to P. aeruginosa in 122 patients. No blood samples were simultaneously positive for S. maltophilia and P. aeruginosa. In 2 patients, S. maltophilia was isolated from a single blood culture, and in subsequent blood culture, bacteremia was cleared without any treatment. These patients were excluded because of possible sample contamination and the remaining 30 patients with S. maltophilia bacteremia were enrolled as the case group. Control patients with P. aeruginosa bacteremia were matched to case patients on a 1:1 ratio using a stepwise procedure in the order of underlying disease, age, and gender, to ensure the best match. We focused on matching case and control patients with the same underlying primary disease. Subsequently, control patients with similar age (±10 years) and the same gender were chosen as case patients. If this criterion was not satisfied, control patients with an age similar to that of case patients were prioritized. Gender could not be matched in 3 patient pairs. All case and control patient pairs were matched with the same primary disease, as shown in Table 1. Patients with non-hematological diseases comprised 3 with cholangiocarcinoma, 1 with hepatocellular carcinoma, 4 with liver cirrhosis, 2 with dilated cardiomyopathy, and 4 with other diseases. Patients with the other non-hematological diseases included 2 with acute pancreatitis, 1 with acute pneumonia, and 1 with multiple trauma.
Variables and Definitions
Clinical data were collected from medical records to evaluate the risk factors and mortality rates. The variables associated with the detection of S. maltophilia or P. aeruginosa bacteremia included indwelling central venous catheter (CVC) or other artificial products, neutropenia, persistent neutropenia, prolonged hospitalization >30 days, ICU stay, history of chemotherapy and/or transplantation within 30 days, and previous antimicrobial treatment within 30 days. Artificial products other than a CVC primarily included endotracheal tube, drainage tube, urethral catheter and intravascular catheter other than a CVC. In many patients, more than two artificial products, such as an endotracheal tube and a urethral catheter, were simultaneously implanted such that we could not separately evaluate those factors by statistical analysis. Therefore, “artificial products” were used as a variable in the logistic analysis. Neutropenia was defined as the absolute neutrophil count of < 100 cells/mm3 at the onset of bacteremia [9]. Persistent neutropenia was defined as clinical episodes in which neutrophil counts <100/mm3 persisted for more than 2 weeks before the onset of bacteremia. The variables associated with the mortality of patients with S. maltophilia or P. aeruginosa bacteremia included neutropenia, polymicrobial bacteremia, septic shock [11], Sequential Organ Failure Assessment (SOFA) score [12], history of chemotherapy and/or transplantation within 30 days, and inappropriate antimicrobial treatment. Appropriate antimicrobial treatment was defined as a regimen that included 1 or more antimicrobial agent to which the isolate was susceptible in vitro [6], while those who received inappropriate treatment were indicated as all patients who were not included in the definition of appropriate treatment.
Microbiological Analysis
After bacterial isolation using an automated blood culture system, the species were identified using the Vitek system (bioMerieux Japan Ltd., Tokyo, Japan). Antibiotic susceptibility was determined using the breakpoints standardized by the Clinical and Laboratory Standards Institute (CLSI) [13].
Statistical Analysis
Gender and age were compared between the case and control groups using the chi-square test and the Mann-Whitney U test, respectively. Univariate and multivariate conditional logistic regression analyses were performed to extract risk factors for the detection of S. maltophilia bacteremia. The variables with a P < 0.2 on univariate analysis were included into the multivariate model. Univariate and multivariate logistic regression analyses were performed to assess the risk factors for mortality. Multivariate analysis was performed using a stepwise logistic regression model with a p-value cut-off point of 0.2. The log-rank test was used to compare the survival curves obtained using the Kaplan-Meier method for patients with S. maltophilia and P. aeruginosa bacteremia. P < 0.05 was considered to be statistically significant. All statistical analyses were performed using the JMP Pro software, version 11 (SAS Institute, Inc., Cary, NC, USA).
Results
A total of 30 case patients with S. maltophilia bacteremia were matched with 30 control patients with P. aeruginosa bacteremia for underlying disease, age, and gender (Table 1). Approximately 50% of the patients with S. maltophilia bacteremia had hematological diseases. The 30-day mortality rate for the case and control patients was 53.3% (16/30 patients) and 30% (9/30), respectively, and this difference was statistically significant (P = 0.047). Therefore, the attributable mortality was 23.3%.
Case patients were compared with control patients considering the predisposing risk factors for the detection of S. maltophilia bacteremia using conditional univariate logistic analysis (Table 2). The frequency of CVC use was similar between the two groups whereas the other artificial products were more likely to be implanted in the case patients compared with the control patients (P = 0.016). In addition, multiple artificial products other than a CVC were present in 11 case patients and in 3 control patients. Moreover, in 7 of the 11 case patients, an endotracheal tube and a urethral catheter were used simultaneously. Thus, there was no specific trend in use of the artificial products. The mean duration of hospitalization before the onset of bacteremia was similar between the case and control patients. The case patients stayed in the ICU at the onset of bacteremia longer than the control patients (P = 0.038). Most of the case and control patients received antimicrobial therapy within 30 days of the onset of bacteremia. The use of carbapenems or antipseudomonal cephalosporins was not associated with the emergence of S. maltophilia bacteremia, while the frequency of patients who had taken anti-MRSA drugs was significantly higher among the case group compared with the control group. The multivariate analysis did not extract any independent risk factors for the detection of S. maltophilia bacteremia when compared with the matched control group with P. aeruginosa bacteremia.
Univariate logistic analysis was performed to assess the risk factors for mortality in the patients with S. maltophilia or P. aeruginosa bacteremia (Table 3). Severe neutropenia and chemotherapy or transplantation were not associated with the mortality of patients with S. maltophilia bacteremia. High severity score was a factor significantly associated with the mortality in the patients with S. maltophilia bacteremia as well as in those with P. aeruginosa bacteremia (P = 0.006 and 0.01, respectively). Inappropriate antimicrobial therapy was an additional risk factor for the mortality in patients with S. maltophilia bacteremia (P = 0.045). The results of the multivariate analysis also indicated that inappropriate antimicrobial therapy was an independent risk factor for mortality in patients with S. maltophilia bacteremia, as shown in Table 4 (odds ratio = 13.64, P = 0.048). The inappropriate antimicrobial treatment in the 7 non-surviving patients with S. maltophilia bacteremia included 3 patients who had not taken any agents recommended for S. maltophilia and 4 patients in whom monotherapy with fluoroquinolones was inactive against the isolated S. maltophilia. Specifically, patients with S. maltophilia bacteremia had monotherapy with fluoroquinolones in 6 of the 16 non-survivors and 3 of the 14 survivors. Among the 6 non-survivors, 4 patients received only fluoroquinolones, to which S. maltophilia were not susceptible.
Discussion
It is possible to assume that the clinical settings are similar between the hospitalized patients with S. maltophilia and P. aeruginosa bacteremia. Therefore, to adequately characterize the associated factors and mortality rates for S. maltophilia bacteremia, matching patients with S. maltophilia bacteremia with those with P. aeruginosa bacteremia is important. The clinical setting used in the present matched case-control study was similar for both patient groups.
The number of case-control studies for S. maltophilia bacteremia is limited. Moreover, since the control patients involved not only those with P. aeruginosa bacteremia, but also those with Escherichia coli bacteremia or those without bacteremia, the extraction of different risk factors for S. maltophilia bacteremia was inevitable. The presence of CVC was not associated with the occurrence of S. maltophilia bacteremia in this study. In previous studies, CVC was a factor associated with E. coli bacteremia [4,8], but not with P. aeruginosa bacteremia [9,10]. These findings clearly indicate that the background of control patients influences data extraction on the risk factors for the case patients. Interestingly, indwelling artificial products other than a CVC and ICU stay were associated with S. maltophilia bacteremia, although these factors were not independent. In fact, all of the ICU patients placed not only a CVC, but also multiple artificial products. ICU stay and the implantation of multiple artificial products more likely induce the emergence of S. maltophilia bacteremia even when compared with P. aeruginosa bacteremia.
Previous treatment with antipseudomonal broad-spectrum antibiotics has been reported to be a risk factor for the detection of bacteremic S. maltophilia [6,8,10]. Our results indicated that the use of carbapenems or antipseudomonal cephalosporins was not a factor that predisposed S. maltophilia bacteremia. This result may be attributable to having organized patients’ medical history using the matching process conducted in our study. Thus, it may be difficult to distinguish bacteremia due to S. maltophilia from bacteremia due to P. aeruginosa based on the previous use of these agents. On the other hand, the use of anti-MRSA agents was associated with S. maltophilia bacteremia in univariate analysis, which is consistent with the results of other studies [8,10]. It is difficult to adequately interpret this finding because anti-MRSA agents can induce the colonization of both of P. aeruginosa and S. maltophilia. Further studies are needed to elucidate this issue.
It has been reported that the severity of illness is associated with poor outcome among patients with S. maltophilia bacteremia [9,10,14,15], which was corroborated by the results of the present study. Patients with hematological malignancies accounted for the majority of those with S. maltophilia bacteremia, with a frequency ranging between 40% and 60% [6,14,16,17], and this rate was approximately 50% in our hospital. The presence of neutropenia and/or transplantation, which is often associated with hematological malignancies, has been reported to be a risk factor for the mortality of patients with S. maltophilia bacteremia [6,9,14,17,18], but these factors were not significant in this study. Rather, inappropriate antimicrobial treatment was associated with the mortality of patients with S. maltophilia bacteremia. The results of the univariate analysis indicated that inappropriate therapy led to poor outcome following S. maltophilia bacteremia [6,9,16,19]. In this study, this parameter was extracted as an independent factor in the multivariate analysis.
Recent studies indicated that fluoroquinolones are a feasible alternative to trimethoprim-sulfamethoxazole (TMP-SMX) for the treatment of S. maltophilia bacteremia [20,21]. In our study, approximately 70% of the patients were treated with quinolones. However, the rapid acquisition of resistance to quinolones has been a cause for concern [1,22]. In this study, monotherapy with quinolones inactive against the isolated S. maltophilia had a negative impact on prognosis. The effectiveness of quinolone monotherapy against S. maltophilia bacteremia should be carefully assessed. Even though TMP-SMX plays a central role in the treatment of S. maltophilia infection [1,2], only 20% of the patients in this study were treated with that agent. In addition, the use rate of TMP-SMX was low regardless of the prophylactic treatment prior to the onset of bacteremia. The use of TMP-SMX may need to be better evaluated in our hospital.
The most important limitation of the present study involved the sample size, which especially, might be small to adequately assess the prognostic factors for S. maltophilia bacteremia. Another limitation of this study was data collection at a single institution. The choice of therapeutic agents for S. maltophilia bacteremia may tend to be different at each institution. Thus, inappropriate treatment could not be a risk factor for mortality in other hospitals. The two groups do not match according to SOFA score. It may be considered a limitation of this study because it is not possible to assess that the attributable mortality was not influenced by the severity of the underlying disease.
This is the first case-control study of patients with S. maltophilia bacteremia matched with a control group with P. aeruginosa bacteremia. The multivariate analysis applied to the matched pairs did not extract any independent factors for the detection of S. maltophilia bacteremia. This result indicates that we cannot easily diagnose the emergence of either S. maltophilia or P. aeruginosa in cases of bacteremia. However, the univariate analysis possibly indicates that further analysis may help extract data on the use of artificial products or ICU stay as independent factors for S. maltophilia bacteremia. Moreover, this study suggested that the high mortality due to S. maltophilia bacteremia was not attributable to neutropenia or transplantation in patients with hematological malignancies, but rather to inappropriate antimicrobial treatment. Therefore, the prognosis of patients with S. maltophilia bacteremia may be improved, at least to the same level as that of those with P. aeruginosa bacteremia, through adequate antimicrobial therapy.
Author Contributions
Conceived and designed the experiments: KS YC. Performed the experiments: KS YC. Analyzed the data: KS YC. Wrote the paper: KS YC. Manuscript editing: NM TA MY NS SS YM KA.
References
- 1. Looney WJ, Narita M, Mühlemann K (2009) Stenotrophomonas maltophilia: an emerging opportunist human pathogen. Lancet Infect Dis 9: 312–323. pmid:19393961
- 2. Safdar A, Rolston KV (2007) Stenotrophomonas maltophilia: changing spectrum of a serious bacterial pathogen in patients with cancer. Clin Infect Dis 45: 1602–1609. pmid:18190323
- 3. Fihman V, Le Monnier A, Corvec S, Jaureguy F, Tankovic J, Jacquier H, et al. (2012) Stenotrophomonas maltophilia—the most worrisome threat among unusual non-fermentative gram-negative bacilli from hospitalized patients: a prospective multicenter study. J Infect 64: 391–398. pmid:22245400
- 4. Victor MA, Arpi M, Bruun B, Jønsson V, Hansen MM (1994) Xanthomonas maltophilia bacteremia in immunocompromised hematological patients. Scand J Infect Dis 26: 163–170. pmid:8036472
- 5. Labarca JA, Leber AL, Kern VL, Territo MC, Brankovic LE, Bruckner DA, et al. (2000) Outbreak of Stenotrophomonas maltophilia bacteremia in allogenic bone marrow transplant patients: role of severe neutropenia and mucositis. Clin Infect Dis 30: 195–197. pmid:10619754
- 6. Senol E, DesJardin J, Stark PC, Barefoot L, Snydman DR (2002) Attributable mortality of Stenotrophomonas maltophilia bacteremia. Clin Infect Dis 34: 1653–1656. pmid:12032905
- 7. Apisarnthanarak A, Mayfield JL, Garison T, McLendon PM, DiPersio JF, Fraser VJ, et al. (2003) Risk factors for Stenotrophomonas maltophilia bacteremia in oncology patients: a case-control study. Infect Control Hosp Epidemiol 24: 269–274. pmid:12725356
- 8. Metan G, Hayran M, Hascelik G, Uzun O (2006) Which patient is a candidate for empirical therapy against Stenotrophomonas maltophilia bacteraemia? An analysis of associated risk factors in a tertiary care hospital. Scand J Infect Dis 38: 527–531. pmid:16798705
- 9. Micozzi A, Venditti M, Monaco M, Friedrich A, Taglietti F, Santilli S, et al. (2000) Bacteremia due to Stenotrophomonas maltophilia in patients with hematologic malignancies. Clin Infect Dis 31: 705–711. pmid:11017819
- 10. Hotta G, Matsumura Y, Kato K, Nakano S, Yunoki T, Yamamoto M, et al. (2014) Risk factors and outcomes of Stenotrophomonas maltophilia bacteraemia: a comparison with bacteraemia caused by Pseudomonas aeruginosa and Acinetobacter species. PLoS ONE 9: e112208. pmid:25375244
- 11. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. (2013) Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Vol. 41. pp. 580–637. pmid:23353941
- 12. Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL (2001) Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 286: 1754–1758. pmid:11594901
- 13.
Clinical and Laboratory Standards Institute: Performance standards for antimicrobial susceptibility testing; twenty-second informational supplement M100-S22. CLSI, Wayne, PA, USA, 2012
- 14. Muder RR, Harris AP, Muller S, Edmond M, Chow JW, Papadakis K, et al. (1996) Bacteremia due to Stenotrophomonas (Xanthomonas) maltophilia: a prospective, multicenter study of 91 episodes. Clin Infect Dis 22: 508–512. pmid:8852971
- 15. Paez JIG, Tengan FM, Barone AA, Levin AS, Costa SF (2008) Factors associated with mortality in patients with bloodstream infection and pneumonia due to Stenotrophomonas maltophilia. Eur J Clin Microbiol Infect Dis 27: 901–906. pmid:18483755
- 16. Friedman ND, Korman TM, Fairley CK, Franklin JC, Spelman DW (2002) Bacteraemia due to Stenotrophomonas maltophilia: An Analysis of 45 Episodes. Journal of Infection 45: 47–53. pmid:12217732
- 17. Araoka H, Baba M, Yoneyama A (2010) Risk factors for mortality among patients with Stenotrophomonas maltophilia bacteremia in Tokyo, Japan, 1996–2009. Eur J Clin Microbiol Infect Dis 29: 605–608. pmid:20177726
- 18. Garazi M, Singer C, Tai J, Ginocchio CC (2012) Bloodstream infections caused by Stenotrophomonas maltophilia: a seven-year review. J Hosp Infect 81: 114–118. pmid:22494851
- 19. Metan G, Uzun O (2005) Impact of initial antimicrobial therapy in patients with bloodstream infections caused by Stenotrophomonas maltophilia. Antimicrob Agents Chemother 49: 3980–3981. pmid:16127088
- 20. Cho SY, Kang C-I, Kim J, Ha YE, Chung DR, Lee NY, et al. (2014) Can levofloxacin be a useful alternative to trimethoprim-sulfamethoxazole for treating Stenotrophomonas maltophilia bacteremia? Antimicrob Agents Chemother 58: 581–583. pmid:24126583
- 21. Wang YL, Scipione MR, Dubrovskaya Y, Papadopoulos J (2014) Monotherapy with fluoroquinolone or trimethoprim-sulfamethoxazole for treatment of Stenotrophomonas maltophilia infections. Antimicrob Agents Chemother 58: 176–182. pmid:24145530
- 22. Ba BB, Feghali H, Arpin C, Saux M-C, Quentin C (2004) Activities of ciprofloxacin and moxifloxacin against Stenotrophomonas maltophilia and emergence of resistant mutants in an in vitro pharmacokinetic-pharmacodynamic model. Antimicrob Agents Chemother 48: 946–953. pmid:14982788