Outcomes and Predictive Factors Associated with Adequacy of Antimicrobial Therapy in Patients with Central Line-Associated Bloodstream Infection

Background Central venous catheters are significant risk factors for bloodstream infection (BSI), which are directly associated with increased morbidity and mortality. Methods This study was a retrospective cohort study for the time period of July 2011–June 2014 in patients with central line-associated bloodstream infection (CLABSI) to determine the microbiological profile and antimicrobial adequacy of patients with CLABSI in a tertiary hospital. Results One hundred and twenty-one CLABSI cases were identified. Ninety-two percent (n = 111) of patients had monomicrobial BSI. Gram-negative bacteria were the most prevalent (49%, n = 63), with Klebsiella spp. predominating (30%, n = 19). Among the Gram-positive bacteria (n = 43, 33%), coagulase-negative staphylococci was the major pathogen (58%, n = 25), and all isolates were methicillin resistant. Antimicrobial therapy was assessed as adequate in 81% (n = 98) of cases. In-hospital mortality was 36% (n = 43 cases). Conclusion Our CLABSI patients had a high mortality, although antimicrobial therapy was appropriate. Gram-negative bacteria were responsible for almost half of the cases and there was a high rate of bacteria resistance to extended-spectrum antibiotics.

To improve antimicrobial therapy for BSI, it is necessary to understand the microorganism profile as well as the antimicrobial susceptibility of these microorganisms (6)(7)(8). In a recent publication from our hospital, which analyzed the impact of the implementation of a sepsis protocol, 60% of our patients developed severe sepsis and septic shock outside of the Emergency Department, which increases the probability of nosocomial infection with antibiotic-resistant microorganisms (9,10). There are serious concerns about the rapid increase of multidrug resistant bacteria, mainly the Gram-negative microorganisms (10).
Thus, the objectives of this study were to determine the microbiological profile and antimicrobial adequacy of patients with CLABSI in a tertiary hospital.

MaTerials anD MeThODs
This study was conducted in a 629-bed tertiary care (Hospital Israelita Albert Einstein), private hospital in São Paulo, Brazil, with approximately 194,000 patient-days yearly. This study was approved by the Institutional Review Board (CAAE # 40993115.8.0000.0071) with a waiver of informed consent.
We performed a retrospective cohort study for the time period of July 2011-June 2014, in patients with central line-associated bloodstream infection.

Definitions
Central lines were defined using the definition of the National Healthcare Safety Network (NHSN) (11). The NHSN methodology for CLABSI employs active surveillance with standardized definitions by infection preventionists. The definition of CLABSI was an NHSN defined laboratory-confirmed bloodstream infection that was not secondary to an infection at another body site (11).

antimicrobial Therapy
As in our previous study (10), antimicrobial therapy was considered appropriate if the bacterial species identified in blood culture was susceptible to at least one of the antibiotics administered within 24 h after the collection of culture. If the isolated microorganism was not susceptible by in vitro testing to the antibiotic(s) used, the therapy was considered inadequate. The microbiology laboratory has an alert system to notify physicians of patients with positive blood culture and their Gram stain results. Also the proper antimicrobial dose, proper interval administration, monitoring of drug levels when appropriate, and avoidance of unwanted drug interactions were evaluated (6) by the pharmaceutical hospital in our medical routine.

Microbiological Methods
All samples were identified by manual or automated methods and confirmed using the Vitek 2 system (bioMerieux Vitek, Inc., Hazelwook, MO, USA). The same organism with identical antimicrobial susceptibility profiles from the same or different anatomic sites in the same patient was considered a single isolate. Antimicrobial susceptibility testing was performed by an automated method or by disk diffusion as described by the Clinical and Laboratory Standards Institute (13).

statistical analysis
Continuous variables were described by mean and SD or median and interquartile range (IQR). Categorical variables were described as absolute numbers and percentages. Logistic regression models were used to investigate variables associated with death or adequate antimicrobial therapy. Variables with P < 0.1 in the univariate analysis were included in the multivariate models, which were submitted to stepwise method to select the best subset of predictors to each outcome. Results were presented as odds ratios and 95% confidence intervals. The level of significance was set at 0.05 and analyses were performed with statistical package R (14).

resUlTs
A total of 121 CLABSI cases were identified from July 2011 to June 2014. The mean age of the cases was 67 years (SD = 18), and the median length of hospital stay was 43 days (exact median cannot be calculated as two patients remain hospitalized). Of the cases, 56% (n = 68) were male. ESKAPE group pathogens accounted for 21% (n = 25) of the cases. Polymicrobial CLABSIs accounted for 9% (n = 11) of the cases. Antimicrobial therapy was assessed as adequate in 81% (n = 98) of cases. Cardiac dysfunction occurred in 63% (n = 76) and respiratory dysfunction occurred in 52% (n = 63). In-hospital mortality was 36% (n = 43 cases) ( Table 1).

DiscUssiOn
Our findings identified the pathogen profile in patients with CLABSI. Regardless of the hospital unit where CLABSI occurred, Gram-negative bacteria, were the most common pathogens, especially Klebsiella spp. (30% of all Gram-negative pathogens), and nearly all of which (95%) were resistant to the third-and fourth-generation cephalosporins. This differs from international studies, including a meta-analysis that assessed 18 studies carried out in different intensive care units, which showed that Gram-positive bacteria are the most prevalent pathogens (15). In the Brazilian Surveillance and Control of Pathogens of Epidemiological Importance Project, the spectrum of pathogens was similar to the present study (16). Fisman et al. demonstrated that geography has a direct influence on the development of BSI caused by Gram-negative bacteria (17).
In our earlier investigation, evaluating patients with severe sepsis and septic shock, the bacterial profile was similar, with a predominance of Gram-negative organisms; however, there was a greater prevalence of E. coli (10). There was a great difference in the proportion of BSI caused by fungi, only 1.5% of cases in the previous study (10), compared to 18% in the current study with 65% in-hospital mortality.
The most common microorganisms responsible for CLABSI change over time. For example, the best practices in ICU care have reduced the proportion of CLABSI due to Gram-positive bacteria (15,18,19). The higher prevalence of Klebsiella spp. was also identified in a previous study (16). Also previous findings descri bed by Correa et al. warned off the increased resistance of K. pneumoniae (20) and reinforced by the European Antimicro bial Resistance Surveillance System in its latest report of hospitals (2011-2012), which was analyzed in 29 European countries (21).
Heart failure was shown to be an independent risk factor for death. In addition, the presence of liver, renal, and hematological dysfunctions significantly increased the risk of death. Moreover, concomitant toxicity of antimicrobial treatment of multidrug resistant microorganisms, particularly with prolonged duration of therapy, adds to the morbidity of the systemic inflammatory response induced by the pathogen (6,18).
Antimicrobial therapy in a previous study was an independent predictor for lower mortality in patients with severe sepsis and septic shock (10). In our study, inadequacy of therapy showed a trend toward increased mortality. However, the characteristics of these patients show us that appropriate initial antimicrobial treatment of these cases portend a greater chance of error, when polymicrobial infection or ESKAPE pathogens are possible.
The limitations of this study include performance in a single center, which may not allow generalization to other hospitals. Although some information was retrospectively obtained from medical records, the patients were followed prospectively by the hospital epidemiology unit.
In conclusion, our CLABSI patients had a high mortality, although antimicrobial therapy was appropriate. Gram-negative bacteria were responsible for almost half of the cases and there was a high rate of bacteria resistance to extended-spectrum antibiotics.

eThics sTaTeMenT
This study was approved by the Institutional Review Board of Albert Einstein Institute of Research (CAAE # 40993115. 8.0000.0071) with a waiver of informed consent. aUThOr cOnTriBUTiOns PY, AM, TB, and EV participated in the data collected and data analysis. PY, AM, OS, and ME participated in the design and coordination. PY, AM, EV, and ME helped to draft the manuscript and to provide critical review of the manuscript. All authors read and approved the final manuscript.

acKnOWleDgMenTs
We gratefully acknowledge the patients from Hospital Israelita Albert Einstein.