Eur J Clin Microbiol Infect Dis (2015) 34:277–286 DOI 10.1007/s10096-014-2233-5
ARTICLE
Infection with Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae in cancer patients M. P. Freire & L. C. Pierrotti & H. H. C. Filho & K. Y. Ibrahim & A. S. G. K. Magri & P. R. Bonazzi & L. Hajar & M. P. E. Diz & J. Pereira & P. M. Hoff & E. Abdala
Received: 7 May 2014 / Accepted: 15 August 2014 / Published online: 30 August 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae (KPC-Kp) is an emergent pathogen in healthcare-associated infections (HAIs). The aim of this study was to describe HAIs due to KPC-Kp, as well as identify mortality risk factors in cancer patients. In patients diagnosed with HAIs due to KPC-Kp between January 2009 and July 2013, we evaluated only the first infection episode of each patient, analyzing mortality separately for patients treated for ≥48 h with at least one antimicrobial agent proven to display in vitro activity against KPC-Kp. We evaluated variables related to the malignancy, the severity and characteristics of the HAI, and the antimicrobial therapy. We identified 83 HAIs due to KPC-Kp. The 30-day mortality was 57.8 % for all infections and 72.7 % for bacteremic infections. Of the 83 patients, 60 patients received ≥48 h of appropriate treatment and 44
(53 %) developed bacteremia. Ten patients (12 %) were neutropenic at HAI diagnosis and 33 (39.8 %) had infection at the tumor site. The most common HAI was urinary tract infection, seen in 26 patients (31.3 %), followed by primary bloodstream infection, seen in 24 patients (28.9 %). Forty-four patients (73.3 %) received combination antimicrobial therapy, most often including polymyxin (68.3 %). Risk factors for 30-day mortality are high sequential organ failure assessment (SOFA) score, need for intensive care stay at diagnosis of infection, and acute kidney injury; the removal of invasive devices related to infection and treatment with effective antibiotics for KPC-Kp are protective factors. In cancer patients, high mortality is associated with HAI due to KPC-Kp and mortality risk factors are more often related to acute infection than to the underlying disease.
M. P. Freire : L. C. Pierrotti (*) : K. Y. Ibrahim : A. S. G. K. Magri : P. R. Bonazzi : E. Abdala Hospital Control Infection and Infectious Diseases Service, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil e-mail: [email protected]
M. P. E. Diz : P. M. Hoff Service of Clinical Oncology, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
H. H. C. Filho : P. M. Hoff Division of Molecular Biology, Central Laboratory, Laboratório de Investigação Médica 03 (LIM-03, Laboratory for Medical Research 03), Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-903, Brazil L. Hajar Intensive Care Service, Instituto do Câncer do Estado de São Paulo, Hospital Control Infection and Infectious Diseases Service, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
J. Pereira Service of Hematology, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
E. Abdala Department of Infectious and Parasitic Diseases, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-903, Brazil
278
Eur J Clin Microbiol Infect Dis (2015) 34:277–286
Introduction
Methods
Bacteria of the family Enterobacteriaceae are major agents of healthcare-associated infections (HAIs), especially in cancer patients [1]. Carbapenems are drugs with great activity for these agents and are commonly used in the treatment of those infections. However, resistance to the carbapenem class of antibiotics has emerged in recent years. In Enterobacteriaceae, carbapenem resistance can be caused by the overexpression of class C (AmpC-type) or class A (TEM- or SHV-type) extended-spectrum betalactamases—such overexpression having been associated with changes in the membrane (outer membrane permeability or upregulation of efflux systems)—or by the production of enzymes that specifically hydrolyze carbapenems (carbapenemases), the latter being the more common of the two mechanisms [2]. The carbapenemases identified in Enterobacteriaceae are metallo-beta-lactamases, extended-spectrum oxacillinases, and beta-lactamases inhibited by clavulanic acid. Within that last group, one that poses a significant threat is Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae (KPCKp) [2], which was first identified in the United States in the 1990s [3]. Since then, KPC-Kp has spread throughout most of the world [4]. The risk factors for HAIs due to KPC-Kp are related to invasive devices procedures, antibiotic use, severity of patients, state of immunosuppression, and patients submitted to transplantation [5–9]. Mortality associated with HAIs due to KPC-Kp ranges from 28 to 68 %. Among patients with such infections, a delay in the introduction of appropriate therapy and more prolonged hospital stays have been reported [5, 10, 11], as has a risk of death during hospitalization up to four times greater than that observed in patients with other types of HAIs. The risk factors associated with mortality from HAIs due to KPC-Kp are a high APACHE score, inappropriate initial antimicrobial therapy, advanced age, and shock [5–7, 10, 12–14]. There are little data on the incidence and outcomes of HAIs due to KPC-Kp in immunocompromised patients. The mortality rate in solid organ transplant recipients with such infections ranges from 42 to 47 % [15, 16]. In a study of 18 cases of bloodstream infection with carbapenem-resistant Enterobacteriaceae in patients with hematologic malignancies, the 14-day mortality was 53 % [17]. Therefore, it is important to know the details of such infections in immunocompromised patients, especially among those with cancer. The aim of this study was to describe HAIs due to KPC-Kp in patients with solid tumors and hematologic malignancies. In addition, we attempted to identify the risk factors associated with mortality in such cases.
This was a retrospective study including all patients diagnosed with an HAI due to KPC-Kp, as confirmed by polymerase chain reaction, acquired during hospitalization at the São Paulo State Cancer Institute between January 2009 and July 2013. In patients with consecutive HAIs, only the first incidence was considered. The criteria used in order to identify and classify HAIs were those outlined by the (United States) National Healthcare Safety Network (NHSN) [18]. During the study period, samples for surveillance cultures were collected from all patients who were admitted as transfers from other hospitals, as well as on a weekly basis from all intensive care unit (ICU) patients. Samples were obtained from the perineum and rectum, and were transported in Stuart’s medium. Susceptibility testing, in order to identify carbapenem-resistant strains of K. pneumoniae, was performed by automated broth microdilution (VITEK; bioMérieux, Marcy-l’Étoile, France). The minimum inhibitory concentrations (MICs) were classified according to the breakpoints established by the Clinical and Laboratory Standards Institute (CLSI) [19]. We reclassified all strains following the breakpoints established in 2012. Isolates determined to be ertapenem-resistant (according to the breakpoints at the time of the case identification) were sent to the molecular biology laboratory for polymerase chain reaction testing for the bla KPC gene. Tigecycline susceptibility was interpreted according to the breakpoints approved by the (United States) Food and Drug Administration (FDA) [20]. Resistance to polymyxin was initially identified by automated broth microdilution (VITEK; bioMérieux), strains with an MIC>2 mg/L were confirmed by the Etest with polymyxin B strips, and the breakpoints used were those defined by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) [21]. In the 30-day mortality analysis, we included only patients who received appropriate treatment. We defined appropriate treatment as having been treated for≥48 h with at least one antimicrobial agent that displayed in vitro activity against KPC-Kp [13]. The variables included in the 30-day mortality analysis were as follows: age, gender, type of tumor, Eastern Cooperative Oncology Group (ECOG) performance status score, progression of the cancer disease, neutropenia at HAI diagnosis, chemotherapy in the last 15 days before HAI diagnosis, site of infection, dialysis, bacteremia, time from HAI diagnosis (first positive culture) to the initiation of appropriate treatment, combination therapy, sequential organ failure assessment (SOFA) score at HAI diagnosis, lactate and albumin at HAI diagnosis, ICU admission, catheter-associated bloodstream infection (resulting in early catheter removal) or deep infection. Renal function was monitored by determining the serum levels of creatinine. Renal toxicity was defined as acute
Eur J Clin Microbiol Infect Dis (2015) 34:277–286
kidney injury defined by the Acute Kidney Injury Network (AKIN) classification; we considered any of the three stages as acute kidney injury [22]. In the statistical analysis, we investigated dichotomous variables using the Chi-square test and Fisher’s exact test, as indicated, whereas we investigated continuous variables using the Mann–Whitney test. Multivariate analysis was performed by stepwise logistic regression, in which we initially included all variables with p
ARTICLE
Infection with Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae in cancer patients M. P. Freire & L. C. Pierrotti & H. H. C. Filho & K. Y. Ibrahim & A. S. G. K. Magri & P. R. Bonazzi & L. Hajar & M. P. E. Diz & J. Pereira & P. M. Hoff & E. Abdala
Received: 7 May 2014 / Accepted: 15 August 2014 / Published online: 30 August 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae (KPC-Kp) is an emergent pathogen in healthcare-associated infections (HAIs). The aim of this study was to describe HAIs due to KPC-Kp, as well as identify mortality risk factors in cancer patients. In patients diagnosed with HAIs due to KPC-Kp between January 2009 and July 2013, we evaluated only the first infection episode of each patient, analyzing mortality separately for patients treated for ≥48 h with at least one antimicrobial agent proven to display in vitro activity against KPC-Kp. We evaluated variables related to the malignancy, the severity and characteristics of the HAI, and the antimicrobial therapy. We identified 83 HAIs due to KPC-Kp. The 30-day mortality was 57.8 % for all infections and 72.7 % for bacteremic infections. Of the 83 patients, 60 patients received ≥48 h of appropriate treatment and 44
(53 %) developed bacteremia. Ten patients (12 %) were neutropenic at HAI diagnosis and 33 (39.8 %) had infection at the tumor site. The most common HAI was urinary tract infection, seen in 26 patients (31.3 %), followed by primary bloodstream infection, seen in 24 patients (28.9 %). Forty-four patients (73.3 %) received combination antimicrobial therapy, most often including polymyxin (68.3 %). Risk factors for 30-day mortality are high sequential organ failure assessment (SOFA) score, need for intensive care stay at diagnosis of infection, and acute kidney injury; the removal of invasive devices related to infection and treatment with effective antibiotics for KPC-Kp are protective factors. In cancer patients, high mortality is associated with HAI due to KPC-Kp and mortality risk factors are more often related to acute infection than to the underlying disease.
M. P. Freire : L. C. Pierrotti (*) : K. Y. Ibrahim : A. S. G. K. Magri : P. R. Bonazzi : E. Abdala Hospital Control Infection and Infectious Diseases Service, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil e-mail: [email protected]
M. P. E. Diz : P. M. Hoff Service of Clinical Oncology, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
H. H. C. Filho : P. M. Hoff Division of Molecular Biology, Central Laboratory, Laboratório de Investigação Médica 03 (LIM-03, Laboratory for Medical Research 03), Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-903, Brazil L. Hajar Intensive Care Service, Instituto do Câncer do Estado de São Paulo, Hospital Control Infection and Infectious Diseases Service, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
J. Pereira Service of Hematology, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-000, Brazil
E. Abdala Department of Infectious and Parasitic Diseases, Faculdade de Medicina da Universidade de São Paulo, São Paulo 01246-903, Brazil
278
Eur J Clin Microbiol Infect Dis (2015) 34:277–286
Introduction
Methods
Bacteria of the family Enterobacteriaceae are major agents of healthcare-associated infections (HAIs), especially in cancer patients [1]. Carbapenems are drugs with great activity for these agents and are commonly used in the treatment of those infections. However, resistance to the carbapenem class of antibiotics has emerged in recent years. In Enterobacteriaceae, carbapenem resistance can be caused by the overexpression of class C (AmpC-type) or class A (TEM- or SHV-type) extended-spectrum betalactamases—such overexpression having been associated with changes in the membrane (outer membrane permeability or upregulation of efflux systems)—or by the production of enzymes that specifically hydrolyze carbapenems (carbapenemases), the latter being the more common of the two mechanisms [2]. The carbapenemases identified in Enterobacteriaceae are metallo-beta-lactamases, extended-spectrum oxacillinases, and beta-lactamases inhibited by clavulanic acid. Within that last group, one that poses a significant threat is Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae (KPCKp) [2], which was first identified in the United States in the 1990s [3]. Since then, KPC-Kp has spread throughout most of the world [4]. The risk factors for HAIs due to KPC-Kp are related to invasive devices procedures, antibiotic use, severity of patients, state of immunosuppression, and patients submitted to transplantation [5–9]. Mortality associated with HAIs due to KPC-Kp ranges from 28 to 68 %. Among patients with such infections, a delay in the introduction of appropriate therapy and more prolonged hospital stays have been reported [5, 10, 11], as has a risk of death during hospitalization up to four times greater than that observed in patients with other types of HAIs. The risk factors associated with mortality from HAIs due to KPC-Kp are a high APACHE score, inappropriate initial antimicrobial therapy, advanced age, and shock [5–7, 10, 12–14]. There are little data on the incidence and outcomes of HAIs due to KPC-Kp in immunocompromised patients. The mortality rate in solid organ transplant recipients with such infections ranges from 42 to 47 % [15, 16]. In a study of 18 cases of bloodstream infection with carbapenem-resistant Enterobacteriaceae in patients with hematologic malignancies, the 14-day mortality was 53 % [17]. Therefore, it is important to know the details of such infections in immunocompromised patients, especially among those with cancer. The aim of this study was to describe HAIs due to KPC-Kp in patients with solid tumors and hematologic malignancies. In addition, we attempted to identify the risk factors associated with mortality in such cases.
This was a retrospective study including all patients diagnosed with an HAI due to KPC-Kp, as confirmed by polymerase chain reaction, acquired during hospitalization at the São Paulo State Cancer Institute between January 2009 and July 2013. In patients with consecutive HAIs, only the first incidence was considered. The criteria used in order to identify and classify HAIs were those outlined by the (United States) National Healthcare Safety Network (NHSN) [18]. During the study period, samples for surveillance cultures were collected from all patients who were admitted as transfers from other hospitals, as well as on a weekly basis from all intensive care unit (ICU) patients. Samples were obtained from the perineum and rectum, and were transported in Stuart’s medium. Susceptibility testing, in order to identify carbapenem-resistant strains of K. pneumoniae, was performed by automated broth microdilution (VITEK; bioMérieux, Marcy-l’Étoile, France). The minimum inhibitory concentrations (MICs) were classified according to the breakpoints established by the Clinical and Laboratory Standards Institute (CLSI) [19]. We reclassified all strains following the breakpoints established in 2012. Isolates determined to be ertapenem-resistant (according to the breakpoints at the time of the case identification) were sent to the molecular biology laboratory for polymerase chain reaction testing for the bla KPC gene. Tigecycline susceptibility was interpreted according to the breakpoints approved by the (United States) Food and Drug Administration (FDA) [20]. Resistance to polymyxin was initially identified by automated broth microdilution (VITEK; bioMérieux), strains with an MIC>2 mg/L were confirmed by the Etest with polymyxin B strips, and the breakpoints used were those defined by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) [21]. In the 30-day mortality analysis, we included only patients who received appropriate treatment. We defined appropriate treatment as having been treated for≥48 h with at least one antimicrobial agent that displayed in vitro activity against KPC-Kp [13]. The variables included in the 30-day mortality analysis were as follows: age, gender, type of tumor, Eastern Cooperative Oncology Group (ECOG) performance status score, progression of the cancer disease, neutropenia at HAI diagnosis, chemotherapy in the last 15 days before HAI diagnosis, site of infection, dialysis, bacteremia, time from HAI diagnosis (first positive culture) to the initiation of appropriate treatment, combination therapy, sequential organ failure assessment (SOFA) score at HAI diagnosis, lactate and albumin at HAI diagnosis, ICU admission, catheter-associated bloodstream infection (resulting in early catheter removal) or deep infection. Renal function was monitored by determining the serum levels of creatinine. Renal toxicity was defined as acute
Eur J Clin Microbiol Infect Dis (2015) 34:277–286
kidney injury defined by the Acute Kidney Injury Network (AKIN) classification; we considered any of the three stages as acute kidney injury [22]. In the statistical analysis, we investigated dichotomous variables using the Chi-square test and Fisher’s exact test, as indicated, whereas we investigated continuous variables using the Mann–Whitney test. Multivariate analysis was performed by stepwise logistic regression, in which we initially included all variables with p
