Abstract
Identifying patients at risk for bloodstream infection (BSI) due to Acinetobacter baumannii-Acinetobacter calcoaceticus complex (ABC) and providing early appropriate therapy are critical for improving patient outcomes. A retrospective matched case-control study was conducted to investigate the risk factors for BSI due to ABC in patients admitted to the Detroit Medical Center (DMC) between January 2006 and April 2009. The cases were patients with BSI due to ABC; the controls were patients not infected with ABC. Potential risk factors were collected 30 days prior to the ABC-positive culture date for the cases and 30 days prior to admission for the controls. A total of 245 case patients were matched with 245 control patients. Independent risk factors associated with BSI due to ABC included a Charlson's comorbidity score of ≥3 (odds ratio [OR], 2.34; P = 0.001), a direct admission from another health care facility (OR, 4.63; P < 0.0001), a prior hospitalization (OR, 3.11; P < 0.0001), the presence of an indwelling central venous line (OR, 2.75; P = 0.011), the receipt of total parenteral nutrition (OR, 21.2; P < 0.0001), the prior receipt of β-lactams (OR, 3.58; P < 0.0001), the prior receipt of carbapenems (OR, 3.18; P = 0.006), and the prior receipt of chemotherapy (OR, 15.42; P < 0.0001). The median time from the ABC-positive culture date to the initiation of the appropriate antimicrobial therapy was 2 days (interquartile range [IQR], 1 to 3 days). The in-hospital mortality rate was significantly higher among case patients than among control patients (OR, 3.40; P < 0.0001). BSIs due to ABC are more common among critically ill and debilitated institutionalized patients, who are heavily exposed to health care settings and invasive devices.
INTRODUCTION
During the 1970s, Acinetobacter baumannii emerged as a prominent nosocomial pathogen (1), increasingly implicated in bloodstream infections (BSI), ventilator-associated pneumonia, meningitis, and surgical site infections (2). During the past 2 decades, A. baumannii strains have evolved as important multidrug-resistant (MDR) bacteria, thriving in hospital environments worldwide and remaining viable on equipment and surfaces for weeks (3). Of particular importance is the ability of A. baumannii to cause BSI in critically ill patients admitted to intensive care units (ICUs) (4–6). A. baumannii infection also occurs in neonatal ICUs. One study reported A. baumannii as the causative agent in 6.2% of cases of septicemia resulting in an overall attributable mortality rate of 20% (7). Higher morbidity and mortality rates have been reported in patients infected with MDR A. baumannii strains as opposed to those infected with non-MDR A. baumannii strains (8).
Previously published studies investigating the risk factors for BSI due to A. baumannii have been limited by several methodological problems, including small sample sizes and suboptimal methodology for case and control group selection (9–12). The risk factors for A. baumannii bacteremia reported in these studies included recent invasive procedures, the use of broad-spectrum antimicrobials, prolonged ICU stays, certain underlying diseases or conditions, hospital size, and urinary catheterization (13, 14). The mortality rates reported among patients with A. baumannii bacteremia are in excess of 20%, with risk factors predictive of mortality being immunosuppression, shock, recent surgical intervention, central line placement, urinary catheterization, nasogastric tube placement, and mechanical ventilation (15). Given the predilection of A. baumannii for critically ill patients, some outbreaks have required closure of ICUs for decontamination, with costs nearing $350,000 (U.S. dollars [USD]) per outbreak (16). Overall, annual health care expenditures for A. baumannii infections have been estimated to range between $7.4 million and $26.1 million in the United States (17).
The objective of this study was to analyze the risk factors and outcomes of patients with BSI specifically due to Acinetobacter baumannii-Acinetobacter calcoaceticus complex (ABC) through a matched analysis with uninfected control patients. The current study included a large number of cases and matched controls, thereby allowing accurate and reliable assessment of the relationships between the potential risk factors and BSI due to ABC. More importantly, selection of cases was highly refined by including patients with pure ABC infections and excluding patients with ABC colonization or polymicrobial infections.
MATERIALS AND METHODS
Study design and setting.
We conducted a retrospective matched case-control study to investigate the risk factors predisposing for BSI due to ABC in patients admitted to the Detroit Medical Center (DMC) from January 2006 to April 2009. The DMC consists of eight hospitals located in the greater metropolitan Detroit area containing a total of >2,200 beds. The institutional review boards at the DMC and Wayne State University approved the study prior to its initiation.
Patient population.
The study population was identified by querying the DMC Clinical Microbiology Laboratory electronic database for positive ABC blood culture results. Cases were defined based on the systemic inflammatory response syndrome (SIRS) criteria of infection (18) and positive ABC blood cultures. Cases with positive ABC cultures but no SIRS criteria and those with polymicrobial BSIs were excluded from the study. Controls were matched to case patients in a 1:1 ratio based on the admitting hospital, the admitting unit, and the at-risk period (the time from admission to the positive culture date for cases and the total duration of hospitalization for controls). The at-risk time had to be at least as long for the control as for its matched case.
Variables and definitions.
Patient charts were reviewed for potential risk factors and pertinent clinical data, including patient demographics, comorbidities, prior antibiotic exposure, activities of daily living (ADL), ICU stays, surgeries, mechanical ventilation, central venous catheters (CVC), urinary catheters, and inpatient mortality. Risk factor data were collected for the 30 days prior to the ABC-positive culture date for the cases and 30 days prior to discharge for the controls. Prior antibiotic exposure in cases was defined as the administration of an antibiotic from 30 days prior to the culture date up to 1 day prior to the culture date. In controls, prior antibiotic exposure was defined as the administration of any antibiotic from 30 days prior to discharge up to the day of discharge. Appropriate antimicrobial therapy was defined as therapy with an agent with in vitro activity against the infecting pathogen.
ICU admissions were captured within 30 days of the ABC-positive culture date for the cases and until discharge for the controls. Emergency visits and in-hospital readmissions within 30 days of discharge were captured for all cases and controls. In-hospital mortality rates and mortality rates within 6 months of the primary admission date were captured for both cases and controls.
Microbiology.
The DMC has a single centralized clinical microbiology laboratory, which processes ∼500,000 samples annually. Bacteria are identified to the species level, and susceptibilities are determined for predefined antimicrobials on the basis of an automated broth microdilution system (MicroScan; Siemens AG, Germany) and in accordance with CLSI criteria (19).
Statistical analysis.
All analyses were performed using SAS software, version 9.3 (SAS, Cary, NC). The Student t test and Mann-Whitney U test (Wilcoxon rank-sum test) were used to analyze continuous variables. Both the chi-square and Fisher's exact tests were used for the bivariate analyses of the discrete variables. Variables with a P value of <0.10 in the bivariate analyses were included as candidates for the multivariate model. Multivariate analysis, using logistic regression, was performed to adjust for confounders and to select for variables in the final model. All significance tests were 2-sided and used a 5% level of significance.
RESULTS
Patient characteristics and hospital-related exposures.
All cases in this study were selected from a recently published cohort analysis of patients with BSI due to carbapenem- and ampicillin-sulbactam-resistant (CASR) A. baumannii (20). In the present study, 245 of these cases with BSI due to ABC were matched to 245 controls not infected with ABC. Forty-five cases (18%) had BSI due to ampicillin-sulbactam-resistant ABC (MIC, ≥16/8 μg/ml), 26 (11%) had BSI due to carbapenem-resistant ABC (MIC, ≥8 μg/ml), and 62 (25%) had BSI due to CASR ABC. The mean age of the case patients was 54 ± 24 years, 47% were males, and 73% were African-Americans. The mean age of the control patients was 54 ± 24 years, 52% were males, and 81% were African-Americans (Table 1).
TABLE 1.
Demographics, comorbidities, and hospital-related exposures prior to positive culture
| Variable | Results for: |
ORa | 95% CIb | P | |
|---|---|---|---|---|---|
| Cases (n = 245) | Controls (n = 245) | ||||
| Demographics | |||||
| Age (mean ± SD) (yr) | 55 ± 24 | 54 ± 24 | 0.71 | ||
| African-American ethnicity (no. [%]) | 179 (73) | 199 (81) | 0.63 | 0.41–0.96 | 0.031 |
| Male gender (no. [%]) | 116 (47) | 128 (52) | 0.82 | 0.58–1.17 | 0.28 |
| Severity of illness scores (no. [%]) | |||||
| Charlson's score of ≥3 | 165 (67) | 101 (41) | 2.94 | 2.03–4.25 | <0.0001 |
| Charlson's score (median [IQR]) | 4 (2–8) | 2 (0–4) | <0.0001 | ||
| Comorbid conditions (no. [%])c | |||||
| Prior hospitalization | 98 (40) | 38 (15.5) | 3.63 | 2.36–5.58 | <0.0001 |
| Directly admitted from another health care facility | 99 (40) | 23 (9) | 6.51 | 3.95–10.74 | <0.0001 |
| Any dependence for ADLd | 121 (49) | 118 (48) | 1.05 | 0.74–1.50 | 0.78 |
| Dependence for ≥3 ADL | 99 (40) | 92 (37) | 1.13 | 0.78–1.62 | 0.52 |
| Prior use of chemotherapy | 39 (16) | 3 (1.2) | 15 | 4.65–50.14 | <0.0001 |
| Low albumin level | 27 (11) | 82 (34) | 0.24 | 0.15–0.39 | <0.0001 |
| Neutropenia | 8 (3) | 5 (2) | 1.62 | 0.52–5.02 | 0.40 |
| Prior receipt of TPN | 68 (28) | 16 (7) | 5.50 | 3.07–9.77 | <0.0001 |
| Hospital exposures prior to culture (no. [%])c | |||||
| Central line placement | 51 (21) | 31 (13) | 1.81 | 1.11–2.95 | 0.015 |
| Use of Foley catheter | 21 (8.5) | 51 (21) | 0.33 | 0.19–0.60 | 0.0001 |
| ICU admission | 98 (40) | 88 (36) | 1.20 | 0.82–1.71 | 0.35 |
| Intubation | 83 (34) | 75 (31) | 1.16 | 0.79–1.70 | 0.44 |
| Prior surgeries | 51 (21) | 75 (31) | 0.60 | 0.40–0.90 | 0.013 |
OR, odds ratio.
CI, confidence interval.
Potential risk factors were assessed for all patients within the 30 days prior to the ABC-positive culture date.
ADL, activities of daily living.
In the present study, multiple statistically significant (P < 0.05) risk factors were identified in the bivariate analysis (Table 1). The case patients with BSI due to ABC were more likely to be directly admitted from another health care facility (including an acute care hospital, a long-term care facility, or a rehabilitation center) than the control patients (odds ratio [OR], 6.51; 95% confidence interval [CI], 3.95 to 10.74; P < 0.0001). The cases had higher Charlson's comorbidity scores (score, ≥3) than the controls (OR, 2.94; 95% CI, 2.03 to 4.25; P < 0.0001). The top three comorbid conditions more commonly present among the cases than the controls were peripheral vascular disease (OR, 13.00; 95% CI, 5.46 to 30.56; P < 0.0001), diabetes with end-organ damage (OR, 11.26; 95% CI, 4.74 to 26.74; P < 0.0001), and peptic ulcer disease (OR, 8.54; 95% CI, 3.29 to 22.12; P < 0.0001). Forty percent of the patients with BSI due to ABC had a history of prior hospitalization compared to 15.5% of the controls (OR, 3.63; 95% CI, 2.36 to 5.58; P < 0.0001). A greater proportion of the cases were found to have had prior recent use of chemotherapy than of the controls (16% versus 1.2%, OR, 15.00; 95% CI, 4.65 to 50.14; P < 0.0001). Twenty-eight percent of the case patients with BSI due to ABC received total parenteral nutrition (TPN) compared to 7% of the control patients (OR, 5.50; 95% CI, 3.07 to 9.77; P < 0.0001). A central venous catheter was indwelling in 21% of the cases compared to 13% of the controls (OR, 1.81; 95% CI, 1.11 to 2.95; P = 0.015).
Prior antimicrobial exposure.
There was significantly more prior antimicrobial exposure among cases than among controls (Table 2). Aminoglycosides (OR, 1.75; 95% CI, 1.12 to 2.74; P = 0.014), β-lactams (including penicillins and cephalosporins but not carbapenems) (OR, 2.50; 95% CI, 1.74 to 3.60; P < 0.0001), and carbapenems (OR, 2.60; 95% CI, 1.52 to 4.45; P = 0.0004) were more commonly administered to the case patients than to the control patients. Individual antimicrobials, with significant Gram-negative activity, more commonly used in the cases than in the controls were aztreonam (OR, 4.52; 95% CI, 1.27 to 16.07; P = 0.011), cefepime (OR, 2.50; 95% CI, 1.70 to 3.64; P < 0.0001), colistin (OR, 2.43; 95% CI, 1.13 to 5.23; P = 0.019), meropenem (OR, 2.82; 95% CI, 1.60 to 4.97; P = 0.0002), and tobramycin (OR, 2.32; 95% CI, 1.30 to 4.14; P = 0.0036).
TABLE 2.
Exposure to antimicrobials prior to positive culture
| Antimicrobial | Exposure (no. [%]) for: |
ORa | 95% CIb | P | |
|---|---|---|---|---|---|
| Cases (n = 245) | Controls (n = 245) | ||||
| Prior antibiotics by class | |||||
| Aminoglycosides | 61 (25) | 39 (16) | 1.75 | 1.12–2.74 | 0.014 |
| β-Lactamsc | 143 (58) | 88 (36) | 2.50 | 1.74–3.60 | <0.0001 |
| Carbapenems | 50 (20) | 22 (9) | 2.60 | 1.52–4.45 | 0.0004 |
| Fluoroquinolones | 32 (13) | 25 (10) | 1.32 | 0.76–2.30 | 0.32 |
| Prior antimicrobials | |||||
| Amikacin | 7 (3) | 3 (1.2) | 2.37 | 0.61–9.28 | 0.20 |
| Amphotericin | 1 (0.41) | 1 (0.41) | 1.00 | 0.06–16 | 1.00 |
| Ampicillin-sulbactam | 34 (14) | 22 (9) | 1.63 | 0.92–2.88 | 0.088 |
| Aztreonam | 13 (5) | 3 (1.2) | 4.52 | 1.27–16.07 | 0.011 |
| Cefepime | 116 (47) | 65 (26.5) | 2.50 | 1.70–3.64 | <0.0001 |
| Colistimethate | 23 (9) | 10 (4) | 2.43 | 1.13–5.23 | 0.019 |
| Ertapenem | 1 (0.41) | 1 (0.41) | 1.00 | 0.06–16.08 | 1.0 |
| Gentamicin | 22 (9) | 20 (8) | 1.11 | 0.59–2.09 | 0.75 |
| Imipenem-cilastatin | 3 (1.2) | 2 (0.8) | 1.51 | 0.25–9.09 | 0.65 |
| Linezolid | 19 (8) | 10 (4) | 1.97 | 0.90–4.34 | 0.085 |
| Meropenem | 47 (19) | 19 (8) | 2.82 | 1.60–4.97 | 0.0002 |
| Piperacillin-tazobactam | 25 (10) | 27 (11) | 0.92 | 0.52–1.63 | 0.77 |
| Rifampin | 10 (4) | 8 (3) | 1.26 | 0.49–3.25 | 0.63 |
| Tigecycline | 23 (9) | 12 (5) | 2.01 | 0.98–4.14 | 0.054 |
| Tobramycin | 40 (16) | 19 (8) | 2.32 | 1.30–4.14 | 0.0036 |
OR, odds ratio.
CI, confidence interval.
β-Lactams did not include carbapenems and included only antimicrobials with significant effects on Gram-negative organisms (ampicillin-sulbactam, cefepime, piperacillin, and piperacillin-tazobactam).
Independent predictors of BSI due to ABC.
In multivariate analysis, factors independently associated with the patients with BSI due to ABC included a Charlson's comorbidity score of ≥3 (OR, 2.34; 95% CI, 1.40 to 3.92; P = 0.001), a direct admission from another health care facility (including long-term care) (OR, 4.63; 95% CI, 2.38 to 9.20; P < 0.0001), a prior recent hospitalization (OR, 3.11; 95% CI, 1.70 to 5.67; P < 0.0001), the presence of an indwelling central venous catheter (OR, 2.75; 95% CI, 1.26 to 6.00; P = 0.011), receipt of TPN (OR, 21.20; 95% CI, 7.83 to 57.37; P < 0.0001), the prior use of β-lactams (OR, 3.58; 95% CI, 2.10 to 6.13; P < 0.0001) or carbapenems (OR, 3.18; 95% CI, 1.39 to 7.27; P = 0.006), and chemotherapy (OR, 15.42; 95% CI, 3.87 to 61.40; P < 0.0001) (Table 3). In contrast, exposures independently associated with a decreased risk for BSI due to ABC included African-American ethnicity (OR, 0.36; 95% CI, 0.19 to 0.68; P = 0.001), the presence of an indwelling urinary catheter (OR, 0.24; 95% CI 0.09 to 0.60; P = 0.045), and a low albumin level (OR, 0.06; 95% CI, 0.03 to 0.13; P < 0.0001). The model was adjusted for the confounding effect of recent prior surgery (OR, 0.51; 95% CI, 0.25 to 1.00; P = 0.051).
TABLE 3.
Independent predictors of BSI due to ABC
| Variable | ORa | 95% CIb | P |
|---|---|---|---|
| African-American ethnicity | 0.36 | 0.19–0.86 | 0.001 |
| Charlson's comorbidity score of ≥3 | 2.34 | 1.40–3.92 | 0.001 |
| Direct admission from another health care facility | 4.63 | 2.38–9.20 | <0.0001 |
| Prior hospitalization | 3.11 | 1.70–5.67 | <0.0001 |
| Central line placement | 2.75 | 1.26–6.00 | 0.011 |
| Use of Foley catheter | 0.24 | 0.09–0.60 | 0.002 |
| Prior surgeries | 0.51 | 0.25–1.00 | 0.051 |
| Low albumin level | 0.06 | 0.03–0.13 | <0.0001 |
| Prior receipt of TPN | 21.20 | 7.83–57.37 | <0.0001 |
| Prior use of β-lactams | 3.58 | 2.10–6.13 | <0.0001 |
| Prior use of carbapenems | 3.18 | 1.39–7.27 | 0.006 |
| Prior use of chemotherapy | 15.42 | 3.87–61.40 | <0.0001 |
OR, odds ratio.
CI, confidence interval.
Clinical outcomes of cases and controls.
The median duration of hospitalization was 14 days in the case patients (interquartile range [IQR], 8 to 27 days) and 8 days in control patients (IQR, 3 to 21 days) (P < 0.0001). Among the cases, the median duration from the time of the culture to the time of initiation of the appropriate antimicrobial therapy was 2 days (IQR, 1 to 3 days). Compared to the control patients, the case patients with BSI due to ABC were more likely to be admitted to the ICU during the period after culture prior to discharge (OR, 34.00; 95% CI, 4.60 to 252.00; P < 0.0001), to have more frequent emergency room visits (OR, 55.00; 95% CI, 7.50 to 401.80; P < 0.0001), and to be readmitted to the hospital within 30 days of discharge (OR, 81.00; 95% CI, 11.11 to 589; P < 0.0001) (Table 4).
TABLE 4.
Clinical outcomes in the cases compared to those in the controls
| Variable | Results for: |
ORa | 95% CIb | P | |
|---|---|---|---|---|---|
| Cases (n = 245) | Controls (n = 245) | ||||
| Duration of hospitalization (days) | |||||
| Mean ± SD | 22 ± 23 | 17 ± 23 | <0.0001 | ||
| Median (IQR) | 14 (8–27) | 8 (3–21) | <0.0001 | ||
| ICU admissions, readmissions, and ERc visits (no. [%]) | |||||
| ICU admission after positive culture | 30 (12) | 1 (0.41) | 34 | 4.60–252.00 | <0.0001 |
| Urgent care visitsd | 45 (18) | 1 (0.41) | 55 | 7.50–401.80 | <0.0001 |
| 30-day readmissiond | 61 (25) | 1 (0.41) | 81 | 11.11–589 | <0.0001 |
| Mortality | |||||
| In-hospital mortality | 47 (19) | 16 (6.5) | 3.40 | 1.87–6.18 | <0.0001 |
| Overall 6-mo mortality | 55 (22.4) | 18 (7) | 3.65 | 2.07–6.43 | <0.0001 |
OR, odds ratio.
CI, confidence interval.
ER, emergency room.
Urgent care visits and readmissions were assessed within 30 days after discharge.
The in-hospital mortality (OR, 3.40; 95% CI, 1.87 to 6.18; P < 0.0001) and overall 6-month mortality (OR, 3.65; 95% CI, 2.07 to 6.43; P < 0.0001) were higher among the cases than among the controls.
DISCUSSION
To our knowledge, the present study is the largest case-control analysis pertaining to risk factors for BSI due to ABC. Previously published studies analyzing risk factors for A. baumannii infections were limited by several methodological issues, including small sample size, the inclusion of colonized patients, and inappropriate consideration of the at-risk period for matching cases to controls (9–12, 21, 22). In the present study, the large sample size and the matching based primarily on the at-risk period have enabled us to reliably identify risk factors for BSI due to ABC. Moreover, all of the cases in this study were appropriately defined, as the diagnosis of patients was made based on both the clinical picture and a positive culture for ABC organisms isolated from blood samples. In particular, exclusion of subjects deemed to be colonized (and not infected) and patients with polymicrobial infections was a particular strength of this study. This study included cases of ABC with different degrees of antimicrobial resistance. This diversity of susceptibility patterns adds generalizability to the results.
The present study demonstrated that TPN is the strongest independent predictor of BSI due to ABC. Yin et al. previously reported that the administration of TPN for ≥2 weeks was associated with a higher incidence of nosocomial A. baumannii bacteremia (23). TPN has also been implicated in the nosocomial acquisition of bacteremia due to carbapenem-resistant A. baumannii (24, 25). Surprisingly, the multivariate analysis demonstrated that a low albumin level was associated with a decreased risk for BSI due to ABC. This contradicts results of previous studies showing that low albumin levels were significant predictors for acquisition of bacterial infections originating in the bloodstream (26, 27). It has been proposed that a low albumin level may result in significant intestinal edema and capillary leakage, therefore, facilitating the translocation of microorganisms from the intestinal lumen to the bloodstream (27). Perhaps, the ABC species at our facilities might have had less tendency to colonize the intestinal tract than the species reported in other studies.
The multivariate analysis identified receipt of chemotherapy as a strong predictor of BSI due to ABC. This finding was not unexpected, owing to the fact that chemotherapy compromises the host immune system, thus making patients highly vulnerable for opportunistic nosocomial infections such as ABC. Fukuta et al. reported that acquisition of MDR A. baumannii infections in cancer patients was not directly related to immunosuppressive therapy and might be attributed to the severity of the underlying illness and the extensive health care exposures associated with their primary condition (28).
In our study, admission from another health care facility, a Charlson's comorbidity score of ≥3, prior hospitalizations, and the prior receipt of carbapenems and β-lactams were all identified as independent predictors of BSI due to ABC. Most of these factors have been reported by other studies as risk factors for A. baumannii infections (22, 29–31). This emphasizes the opportunistic nature of ABC infections and reflects the tendency of ABC to infect critically ill patients who have impaired natural host defenses. In addition, the current study demonstrated that the patients who had indwelling central venous catheters had an almost 3-fold increase in the risk of acquiring BSI due to ABC compared to that of the controls. This is concordant with results of other studies which investigated the effect of central venous catheter insertion on the incidence of A. baumannii acquisition (29, 30, 32). However, in prior studies, invasive procedures (surgical procedures, endotracheal intubation, and venous and arterial catheter insertion) were combined together as a single variable, and the analyses were performed in ICU settings, which limited the ability to determine the specific risk associated with central venous catheters (29, 30, 32). Because central venous catheters were independently associated with BSI due to ABC and Foley catheters were not, the urinary tract was probably not a common source of ABC infection and contaminated central venous catheters were probably a relatively common infection source.
Being retrospective, this study might have had a selection bias. However, the matched case-control study design helped limit this type of bias. In addition, although the readmission and mortality rates were significantly higher among the cases than among the controls, these rates could have been attributed to the comorbidities and severity of illness among the cases as opposed to the controls. Moreover, since practices for procedures such as CVC placement and receipt of TPN might differ from one health care facility to another, the potential effect of these practices on acquisition of BSI due to ABC in our study might have been over- or underestimated. Previously, our group described the risk factors for and outcomes associated with BSI due to CASR AB compared to those for more susceptible strains of AB (20). In an attempt to make the findings in this report more generalizable, we chose to focus on BSI due to ABC in general, rather than analyzing different susceptibility phenotypes individually.
In conclusion, this large study demonstrated that acquisition of BSI due to A. baumannii-calcoaceticus complex is more common in critically ill patients previously exposed to health care settings. In particular, chemotherapy, TPN administration, and central line placement were strong independent risk factors for BSI due to ABC. These findings stress the importance of central venous catheter care and infection control bundles in preventing BSI due to ABC. Because ABC infection affects some of the sickest, highest risk hospitalized populations and is associated with significant morbidity and mortality (20), attention to optimal infection prevention practices is particularly important in these populations.
Footnotes
Published ahead of print 2 June 2014
REFERENCES
- 1.Towner KJ. 2009. Acinetobacter: an old friend, but a new enemy. J. Hosp. Infect. 73:355–363. 10.1016/j.jhin.2009.03.032 [DOI] [PubMed] [Google Scholar]
- 2.Smolyakov R, Borer A, Riesenberg K, Schlaeffer F, Alkan M, Porath A, Rimar D, Almog Y, Gilad J. 2003. Nosocomial multi-drug resistant Acinetobacter baumannii bloodstream infection: risk factors and outcome with ampicillin-sulbactam treatment. J. Hosp. Infect. 54:32–38. 10.1016/S0195-6701(03)00046-X [DOI] [PubMed] [Google Scholar]
- 3.Moultrie D, Hawker J, Cole S. 2011. Factors associated with multidrug-resistant Acinetobacter transmission: an integrative review of the literature. AORN J. 94:27–36. 10.1016/j.aorn.2010.12.026 [DOI] [PubMed] [Google Scholar]
- 4.Jang TN, Lee SH, Huang CH, Lee CL, Chen WY. 2009. Risk factors and impact of nosocomial Acinetobacter baumannii bloodstream infections in the adult intensive care unit: a case-control study. J. Hosp. Infect. 73:143–150. 10.1016/j.jhin.2009.06.007 [DOI] [PubMed] [Google Scholar]
- 5.Cisneros JM, Rodriguez-Bano J. 2002. Nosocomial bacteremia due to Acinetobacter baumannii: epidemiology, clinical features and treatment. Clin. Microbiol. Infect. 8:687–693. 10.1046/j.1469-0691.2002.00487.x [DOI] [PubMed] [Google Scholar]
- 6.Cetin ES, Durmaz R, Tetik T, Otlu B, Kaya S, Caliskan A. 2009. Epidemiologic characterization of nosocomial Acinetobacter baumannii infections in a Turkish university hospital by pulsed-field gel electrophoresis. Am. J. Infect. Control 37:56–64. 10.1016/j.ajic.2008.01.010 [DOI] [PubMed] [Google Scholar]
- 7.De AS, Rathi MR, Mathur MM. 2013. Mortality audit of neonatal sepsis secondary to Acinetobacter. J. Glob. Infect. Dis. 5:3–7. 10.4103/0974-777X.107165 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Neonakis IK, Spandidos DA, Petinaki E. 2011. Confronting multidrug-resistant Acinetobacter baumannii: a review. Int. J. Antimicrob. Agents 37:102–109. 10.1016/j.ijantimicag.2010.10.014 [DOI] [PubMed] [Google Scholar]
- 9.Corbella X, Montero A, Pujol M, Dominguez MA, Ayats J, Argerich MJ, Garrigosa F, Ariza J, Gudiol F. 2000. Emergence and rapid spread of carbapenem resistance during a large and sustained hospital outbreak of multiresistant Acinetobacter baumannii. J. Clin. Microbiol. 38:4086–4095 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fierobe L, Lucet JC, Decre D, Muller-Serieys C, Deleuze A, Joly-Guillou ML, Mantz J, Desmonts JM. 2001. An outbreak of imipenem-resistant Acinetobacter baumannii in critically ill surgical patients. Infect. Control Hosp. Epidemiol. 22:35–40. 10.1086/501822 [DOI] [PubMed] [Google Scholar]
- 11.Harris AD, Karchmer TB, Carmeli Y, Samore MH. 2001. Methodological principles of case-control studies that analyzed risk factors for antibiotic resistance: a systematic review. Clin. Infect. Dis. 32:1055–1061. 10.1086/319600 [DOI] [PubMed] [Google Scholar]
- 12.Paterson DL. 2002. Looking for risk factors for the acquisition of antibiotic resistance: a 21st-century approach. Clin. Infect. Dis. 34:1564–1567. 10.1086/340532 [DOI] [PubMed] [Google Scholar]
- 13.Wroblewska MM, Towner KJ, Marchel H, Luczak M. 2007. Emergence and spread of carbapenem-resistant strains of Acinetobacter baumannii in a tertiary-care hospital in Poland. Clin. Microbiol. Infect. 13:490–496. 10.1111/j.1469-0691.2007.01694.x [DOI] [PubMed] [Google Scholar]
- 14.Chen CH, Wu SS, Huang CC. 2013. Two case reports of gastroendoscopy-associated Acinetobacter baumannii bacteremia. World J. Gastroenterol. 19:2835–2840. 10.3748/wjg.v19.i18.2835 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chen HP, Chen TL, Lai CH, Fung CP, Wong WW, Yu KW, Liu CY. 2005. Predictors of mortality in Acinetobacter baumannii bacteremia. J. Microbiol. Immunol. Infect. 38:127–136 [PubMed] [Google Scholar]
- 16.Garlantézec R, Bourigault C, Boles JM, Prat G, Baron R, Tonnelier JM, Cosse M, Lefevre M, Jourdain S, Lelay G, Daniel L, Virmaux M, Le Du I, Tande D, Renault A, Lejeune B. 2011. Cost-analysis of an intensive care unit closure due to an imipenem-resistant oxa-23 Acinetobacter baumannii outbreak. J. Hosp. Infect. 77:174–175. 10.1016/j.jhin.2010.09.027 [DOI] [PubMed] [Google Scholar]
- 17.Lee BY, McGlone SM, Doi Y, Bailey RR, Harrison LH. 2010. Economic impact of Acinetobacter baumannii infection in the intensive care unit. Infect. Control Hosp. Epidemiol. 31:1087-1089. 10.1086/656378 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ. 1992. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 101:1644–1655 [DOI] [PubMed] [Google Scholar]
- 19.Clinical Laboratory Standards Institute. 2009. Performance standards for antimicrobial susceptibility testing; 19th informational supplement. CLSI document M100-S19. Clinical and Laboratory Standards Institute, Wayne, PA [Google Scholar]
- 20.Chopra T, Marchaim D, Awali RA, Krishna A, Johnson P, Tansek R, Chaudary K, Lephart P, Slim J, Hothi J, Ahmed H, Pogue JM, Zhao JJ, Kaye KS. 2013. Epidemiology of bloodstream infections caused by Acinetobacter baumannii and impact of drug resistance to both carbapenems and ampicillin-sulbactam on clinical outcomes. Antimicrob. Agents Chemother. 57:6270–6275. 10.1128/AAC.01520-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Koeleman JG, Parlevliet GA, Dijkshoorn L, Savelkoul PH, Vandenbroucke-Grauls CM. 1997. Nosocomial outbreak of multi-resistant Acinetobacter baumannii on a surgical ward: epidemiology and risk factors for acquisition. J. Hosp. Infect. 37:113–123. 10.1016/S0195-6701(97)90181-X [DOI] [PubMed] [Google Scholar]
- 22.Lortholary O, Fagon JY, Hoi AB, Slama MA, Pierre J, Giral P, Rosenzweig R, Gutmann L, Safar M, Acar J. 1995. Nosocomial acquisition of multiresistant Acinetobacter baumannii: risk factors and prognosis. Clin. Infect. Dis. 20:790–796. 10.1093/clinids/20.4.790 [DOI] [PubMed] [Google Scholar]
- 23.Yin T, Chiang MC, Liaw JJ, Kuo SC, Chen TL, Katherine Wang KW. 2012. Clinical characteristics of Acinetobacter baumannii complex bacteremia in patients receiving total parenteral nutrition. J. Chin. Med. Assoc. 75:102–108. 10.1016/j.jcma.2011.12.015 [DOI] [PubMed] [Google Scholar]
- 24.Huang ST, Chiang MC, Kuo SC, Lee YT, Chiang TH, Yang SP, Ti Y, Chen TL, Fung CP. 2012. Risk factors and clinical outcomes of patients with carbapenem-resistant Acinetobacter baumannii bacteremia. J. Microbiol. Immunol. Infect. 45:356–362. 10.1016/j.jmii.2011.12.009 [DOI] [PubMed] [Google Scholar]
- 25.Huang L, Chen TL, Lee YT, Lee MH, Kuo SC, Yu KW, Dou HY, Fung CP. 30 April 2013. Risk factors for imipenem-nonsusceptible Acinetobacter nosocomialis bloodstream infection. J. Microbiol. Immunol. Infect. 10.1016/j.jmii.2013.02.002 [DOI] [PubMed] [Google Scholar]
- 26.Li C, Wen TF, Mi K, Wang C, Yan LN, Li B. 2012. Analysis of infections in the first 3-month after living donor liver transplantation. World J. Gastroenterol. 18:1975–1980. 10.3748/wjg.v18.i16.1975 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Avkan-Oguz V, Ozkardesler S, Unek T, Ozbilgin M, Akan M, Firuzan E, Kose H, Astarcioglu I, Karademir S. 2013. Risk factors for early bacterial infections in liver transplantation. Transplant. Proc. 45:993–997. 10.1016/j.transproceed.2013.02.067 [DOI] [PubMed] [Google Scholar]
- 28.Fukuta Y, Muder RR, Agha ME, Clarke LG, Wagener MM, Hensler AM, Doi Y. 2013. Risk factors for acquisition of multidrug-resistant Acinetobacter baumannii among cancer patients. Am. J. Infect. Control 41:1249-1252. 10.1016/j.ajic.2013.04.003 [DOI] [PubMed] [Google Scholar]
- 29.Mulin B, Talon D, Viel JF, Vincent C, Leprat R, Thouverez M, Michel-Briand Y. 1995. Risk factors for nosocomial colonization with multiresistant Acinetobacter baumannii. Eur. J. Clin. Microbiol. Infect. Dis. 14:569–576. 10.1007/BF01690727 [DOI] [PubMed] [Google Scholar]
- 30.Scerpella EG, Wanger AR, Armitige L, Anderlini P, Ericsson CD. 1995. Nosocomial outbreak caused by a multiresistant clone of Acinetobacter baumannii: results of the case-control and molecular epidemiologic investigations. Infect. Control Hosp. Epidemiol. 16:92–97. 10.2307/30140949 [DOI] [PubMed] [Google Scholar]
- 31.Peacock JE, Jr, Sorrell L, Sottile FD, Price LE, Rutala WA. 1988. Nosocomial respiratory tract colonization and infection with aminoglycoside-resistant Acinetobacter calcoaceticus var anitratus: epidemiologic characteristics and clinical significance. Infect. Control Hosp. Epidemiol. 9:302–308 [DOI] [PubMed] [Google Scholar]
- 32.Levin AS, Mendes CM, Sinto SI, Sader HS, Scarpitta CR, Rodrigues E, Sauaia N, Boulos M. 1996. An outbreak of multiresistant Acinetobacter baumannii in a university hospital in Sao Paulo, Brazil. Infect. Control Hosp. Epidemiol. 17:366–368. 10.2307/30141136 [DOI] [PubMed] [Google Scholar]