Abstract
Introduction
During recent decades, the rates of multidrug resistance, including resistance to carbapenems, have increased dramatically among Acinetobacter species. Tigecycline has activity against multidrug-resistant Acinetobacter spp, including carbapenem-resistant isolates. However, reports of tigecycline-resistant Acinetobacter spp are emerging from different parts of the world. The purpose of this study was to evaluate potential risk factors associated with tigecycline non-susceptible Acinetobacter bacteremia.
Methods
The medical records of 152 patients with Acinetobacter bacteremia attending Samsung Medical Center between January 2010 and December 2014 were reviewed. Non-susceptibility to tigecycline was defined as a minimum inhibitory concentration (MIC) of tigecycline ≥4 μg/ml. Cases were patients with tigecycline non-susceptible Acinetobacter bacteremia and controls were those with tigecycline-susceptible Acinetobacter bacteremia.
Results
Of the 152 patients included in the study, 61 (40.1%) had tigecycline non-susceptible Acinetobacter bacteremia (case group). These patients were compared to 91 patients with tigecycline-susceptible Acinetobacter bacteremia (control group). The case group showed high resistance to other antibiotics (>90%) except colistin (6.6%) and minocycline (9.8%) when compared to the control group, which exhibited relatively low resistance to other antibiotics (<50%). Multivariate analysis showed that recent exposure to corticosteroids (minimum 20 mg per day for more than 5 days within 2 weeks) (adjusted odds ratio (OR) 2.887, 95% confidence interval (CI) 1.170–7.126) and carbapenems (within 2 weeks) (adjusted OR 4.437, 95% CI 1.970–9.991) were significantly associated with tigecycline non-susceptible Acinetobacter bacteremia. Although prior exposure to tigecycline was more common in the case group than in the control group (9.8%, 6/61 vs. 2.2%, 2/91; p = 0.046), this variable was found not to be a significant factor associated with tigecycline non-susceptibility after adjustment for other variables (adjusted OR 1.884, 95% CI 0.298–11.920; p = 0.501).
Conclusions
These data suggest that tigecycline non-susceptible Acinetobacter spp have emerged and disseminated in the hospital in association with a recent exposure to carbapenems and an immunosuppressed state. This indicates that the rational use of antibiotics through a comprehensive antimicrobial stewardship program, especially in immunosuppressed patients, may be essential in limiting the emergence and spread of multidrug-resistant organisms such as tigecycline-resistant Acinetobacter spp, which are difficult to treat.
Keywords
Introduction
Acinetobacter baumannii, the most important representative species of the Acinetobacter genus, has emerged globally as a major cause of healthcare-associated infection. It can cause various infections, including bacteremia, pneumonia, meningitis, catheter-related bloodstream infections, intra-abdominal infections, urinary tract infections, and skin and soft tissue infections. The organism commonly infects immunosuppressed patients who are critically ill and has been associated with an increase in mortality of between 8% and 40% (
Fournier and Richet, 2006
, Maragakis and Perl, 2008
, Peleg et al., 2008
).Acinetobacter spp can survive for long periods in the hospital environment and on the surface of healthcare workers’ hands. These ubiquitous characteristics and its environmental resilience are thought to allow the clonal spread of isolates and make Acinetobacter spp easy to transmit from person to person. Another concern associated with Acinetobacter spp is their ability to rapidly acquire resistance determinants, leading to multidrug resistance. This phenomenon makes it difficult to treat Acinetobacter infections in the current antibiotic era (
Fournier and Richet, 2006
, Maragakis and Perl, 2008
, Peleg et al., 2008
, Wagenvoort et al., 2002
, Higgins et al., 2010
, Bonomo and Szabo, 2006
).Acinetobacter spp have shown resistance to beta-lactam antibiotics, fluoroquinolones, and aminoglycosides since the early 1990s. Carbapenems have been considered the drug of choice for the treatment of these resistant organisms (
Falagas et al., 2006
). However, Acinetobacter infections caused by carbapenem-resistant strains are increasing, leading to the need for new therapeutic options.Tigecycline is a glycylcycline with in vitro activity against multidrug-resistant (MDR) Acinetobacter spp including carbapenem-resistant isolates (
Seifert et al., 2006
, Ni et al., 2016
). This agent provides a therapeutic option for the treatment of these organisms. However, reports of tigecycline-resistant MDR Acinetobacter spp are also increasing in different parts of the world (Spiliopoulou et al., 2015
, Al-Sweih et al., 2011
, Reid et al., 2007
, Ruzin et al., 2007
, Navon-Venezia et al., 2007
, Sun et al., 2013
). Emerging tigecycline resistance, a potential problem of clinical significance, appears to be of great concern globally. Although there have been numerous reports of risk factor analyses for extensively drug resistant (XDR) Acinetobacter spp infections or acquisition in which there is only susceptibility to colistin and tigecycline (Moghnieh et al., 2016
, Park et al., 2010
, Paterson, 2002
, Ng et al., 2014
, Sunenshine et al., 2007
, Lortholary et al., 1995
, Wisplinghoff et al., 2000
, Tacconelli et al., 2008
), risk factors for tigecycline non-susceptible Acinetobacter spp infection have not been elucidated. This study was conducted to evaluate the potential risk factors associated with tigecycline non-susceptible Acinetobacter bacteremia.Materials and methods
Study design and data collection
An electronic search of the microbiology laboratory database was performed to identify all blood cultures positive for Acinetobacter spp from patients admitted to Samsung Medical Center in Seoul, Korea between January 2010 and December 2014. This is a 1950-bed tertiary care teaching hospital. The case group included adult patients (>18 years) who had isolation of tigecycline non-susceptible Acinetobacter spp. The control group included adult patients with tigecycline-susceptible Acinetobacter spp bacteremia. Only the first isolate from each patient was included in this study.
Data were collected from administrative, pharmaceutical, and laboratory computerized databases maintained by the medical information team of Samsung Medical Center. Clinical records were reviewed and analyzed for potential risk factors. The following information was included: age, sex, the presence of underlying disease or comorbid conditions, source of bacteremia, polymicrobial infection, duration of intensive care unit (ICU) stay prior to blood culture, surgical procedures within the past 30 days, length of hospital stay prior to blood culture, and number of admissions to the hospital in the prior year. The severity of illness was estimated using the Pitt bacteremia score (
Paterson et al., 2004
), white blood cell count (WBC), and C-reactive protein (CRP) level on the date of index culture. Exposure to antimicrobial agents, cytotoxic drugs, or corticosteroids within 14 days prior to the isolation of Acinetobacter spp was also examined.Definitions
The source of bacteremia was assessed by the study investigators based on the medical records, clinical signs and symptoms, imaging data, and microbiological culture of other specimens according to the National Healthcare Safety Network (NHSN) criteria (
Horan et al., 2008
). The source of bacteremia was categorized as primary bacteremia, respiratory infection, meningitis, cardiovascular infection including catheter-related bloodstream infection and infective endocarditis, skin and soft tissue infection, intra-abdominal infection, urinary tract infection, or unexplained neutropenic fever (Peleg et al., 2008
, Horan et al., 2008
, Freifeld et al., 2011
).The severity of underlying disease was estimated using Charlson’s weighted index of comorbidity (WIC). The severity of illness in bacteremia was assessed using the Pitt bacteremia score, which has been validated in several previous studies (
Rhee et al., 2009
, - Rhee J.Y.
- Kwon K.T.
- Ki H.K.
- Shin S.Y.
- Jung D.S.
- Chung D.R.
- et al.
Scoring systems for prediction of mortality in patients with intensive care unit-acquired sepsis: a comparison of the Pitt bacteremia score and the Acute Physiology and Chronic Health Evaluation II scoring systems.
Shock. 2009; 31: 146-150
Al-Hasan et al., 2013
). Polymicrobial infection was defined as the isolation of more than one microorganism during a single bacteremic episode, excluding possible contamination (Garcia-Garmendia et al., 2001
). Prior antibiotic exposure was defined as at least 24 h of therapy administered within the 14 days prior to the positive blood culture. Cytotoxic drugs were defined as all chemotherapy drugs except for targeted agents. The administration of prednisolone 20 mg daily or equivalent dosage for more than 5 days was defined as exposure to steroids.In vitro antimicrobial susceptibility testing
Preliminary species identification of Acinetobacter spp isolates was performed using the VITEK-2 system. In vitro susceptibility testing was then performed by determining the minimum inhibitory concentration (MIC) of the following 11 antibiotics using broth microdilution methods: tigecycline, imipenem, meropenem, doripenem, colistin, ciprofloxacin, cefepime, ceftazidime, amikacin, piperacillin–tazobactam, and ampicillin–sulbactam. The results were interpreted according to the criteria recommended in the Clinical and Laboratory Standards Institute guidelines (CLSI) (
Clinical and Laboratory Standards Institute, 2014
). Interpretative criteria for tigecycline MICs were defined based on the United States Food and Drug Administration breakpoint criteria (susceptibility at ≤2 μg/ml, intermediate at 4 μg/ml, and resistant at 8 μg/ml). Non-susceptibility to tigecycline included both the resistant and intermediate categories.Statistical analysis
In the analysis of clinical variables, the Mann–Whitney U-test was used for the comparison of continuous variables, and the chi-square test or Fisher’s exact test was used for the comparison of categorical variables. Charlson’s WIC, the number of hospital admissions in the past year, and variables with statistical significance in the univariate analysis were included in the multivariate analysis. Variables with similar meanings included only one variable, even if the p-value was less than 0.05. Logistic regression was used to identify independent predictors of tigecycline non-susceptible Acinetobacter bacteremia. All biologically possible variables with p-values of <0.05 in the univariate analysis were incorporated into a model using a backward approach. For all calculations, a two-tailed p-value of <0.05 was considered statistically significant. IBM SPSS Statistics version 20.0 for Windows (IBM Corp., Armonk, NY, USA) was used for all statistical analyses.
Results
Study population and baseline characteristics
For the selected study period, 152 patients with Acinetobacter spp bacteremia were identified. Of the 152 isolates, 149 were identified as Acinetobacter baumannii and three were identified as Acinetobacter lwoffii. In 61 cases, the Acinetobacter spp was non-susceptible to tigecycline (case group); the remaining 91 patients had Acinetobacter spp that were susceptible to tigecycline (control group).
The demographic and clinical characteristics of the 152 patients are given in Table 1. There were no significant differences in median age, sex, or polymicrobial infection. Liver disease was more frequently seen in case patients (26.2% vs. 7.7%, respectively; p = 0.002), whereas solid cancer was more prevalent in control patients (51.6% vs. 19.7%, respectively; p< 0.001). For this reason, the mean Charlson WIC, in which solid cancer is assigned a score of six, was higher in the control group than in the case group (3.77 ± 2.44 vs. 2.67 ± 2.28, respectively; p< 0.002).
Table 1Demographic and clinical characteristics of patients in the case and control groups.
| Case (n = 61) | Control (n = 91) | p-Value | OR (95% CI) | |
|---|---|---|---|---|
| Demographic data | ||||
| Age, years | 59 (48–68) | 64 (52–74) | 0.060 | NA |
| Male | 41 (67.2%) | 53 (58.2%) | 0.264 | 0.680 (0.345–1.340) |
| Coexisting condition | ||||
| Hypertension | 13 (21.3%) | 19 (20.9%) | 0.949 | 1.026 (0.464–2.271) |
| Diabetes mellitus | 12 (19.7%) | 10 (11%) | 0.136 | 1.994 (0.798–4.933) |
| Cardiovascular disease | 2 (3.3%) | 8 (8.8%) | 0.179 | 0.352 (0.072–1.716) |
| Respiratory disease | 9 (14.8%) | 6 (6.6%) | 0.098 | 2.542 (0.825–7.286) |
| Liver disease | 16 (26.2%) | 7 (7.7%) | 0.002 | 3.373(1.092–11.134) |
| Neurological disease | 5 (8.2%) | 8 (8.8%) | 0.898 | 0.926 (0.288–2.978) |
| Connective tissue disease | 0 (0%) | 1 (1.1%) | 0.411 | 0.989 (0.968–1.011) |
| Solid cancer | 12 (19.7%) | 47 (51.6%) | <0.001 | 0.229 (0.108–0.487) |
| Hematological malignancy | 16 (26.2%) | 16 (16.5%) | 0.144 | 1.801 (0.813–3.989) |
| Solid organ transplantation | 7 (11.5%) | 5 (5.5%) | 0.180 | 2.230 (0.674–7.381) |
| Charlson’s WIC | 2.67 ± 2.28 | 3.77 ± 2.44 | 0.002 | NA |
| Source of infection | ||||
| Primary bacteremia | 0 (0%) | 3 (3.3%) | 0.212 | 0.967 (0.931–1.004) |
| Meningitis | 0 (0%) | 1 (1.1%) | 0.599 | 0.989 (0.968–1.011) |
| Cardiovascular | 3 (4.9%) | 15 (16.5%) | 0.031 | 0.262 (0.072–0.948) |
| Skin and soft tissue infection | 6 (9.8%) | 0 (0%) | 0.002 | 1.109 (1.021–1.205) |
| Intra-abdominal infection | 19 (31.1%) | 39 (42.9%) | 0.145 | 0.603 (0.305–1.194) |
| Pneumonia | 30 (49.2%) | 22 (24.2%) | 0.001 | 3.305 (1.516–5.078) |
| Urinary tract infection | 0 (0%) | 6 (6.6%) | 0.043 | 0.934 (0.884–0.986) |
| Neutropenic fever, unexplained | 3 (4.9%) | 5 (5.5%) | 0.593 | 0.890 (0.205–3.868) |
| Polymicrobial infection | 13 (21.3%) | 20 (22%) | 0.922 | 0.961 (0.437–2.115) |
| Origin of infection | ||||
| Community-acquired | 0 (0%) | 2 (2.2%) | 0.357 | 0.978 (0.948–1.009) |
| Healthcare-associated | 0 (0%) | 21 (23.1%) | <0.001 | NA |
| Hospital-acquired | 61 (100%) | 66 (72.5%) | <0.001 | NA |
OR, odds ratio; CI, confidence interval; NA, not applicable; WIC, weighted index of comorbidity.
a Data are expressed as the number (%) of patients, mean ± standard deviation, or median (interquartile range).
The results of in vitro antimicrobial susceptibility testing of the Acinetobacter isolates to 11 types of antibiotic are shown in Table 2. The case group showed high levels of resistance to other antibiotics (>90%) except for colistin (6.6%) and minocycline (9.8%) compared to the control group, which exhibited relatively low levels of resistance to other antibiotics (<50%) (Table 3).
Table 2Antimicrobial susceptibility results for 152 Acinetobacter species clinical isolates.
| Antimicrobial agent | MIC (mg/l) | Number (%) of non-susceptible isolates | ||
|---|---|---|---|---|
| 50% | 90% | Range | ||
| Tigecycline | 2 | 8 | 0.06–64 | 61 (40.1) |
| Minocycline | 0.12 | 4 | 0.06–64 | 8 (5.3) |
| Ampicillin–sulbactam | 64/32 | >64/32 | 0.06/0.03–64/32 | 87 (57.2) |
| Piperacillin–tazobactam | >256/4 | >256/4 | 0.25/4–256/4 | 93 (61.2) |
| Ciprofloxacin | 64 | >64 | 0.06–64 | 94 (61.8) |
| Cefepime | 64 | >128 | 0.12–128 | 96 (63.2) |
| Ceftazidime | >64 | >64 | 0.06–64 | 96 (63.2) |
| Imipenem | >64 | >64 | 0.06–64 | 84 (55.3) |
| Meropenem | >64 | >64 | 0.06–64 | 84 (55.3) |
| Doripenem | >16 | >16 | 0.015–16 | 84 (55.3) |
| Colistin | 0.5 | 2 | 0.06–64 | 13 (8.6) |
| Amikacin | 64 | >128 | 0.12–128 | 81 (53.3) |
MIC, minimum inhibitory concentration.
a 50% and 90%: MIC at which 50% and 90%, respectively, of the isolates tested are inhibited.
Table 3Comparison of antimicrobial resistance rates of Acinetobacter species isolates between the case and control groups.
| Antimicrobial agents | Case (n = 61) | Control (n = 91) | p-Value |
|---|---|---|---|
| Carbapenem | 55 (90.2%) | 29 (31.9%) | <0.001 |
| Colistin | 4 (6.6%) | 9 (9.9%) | 0.471 |
| Minocycline | 6 (9.8%) | 2 (2.2%) | 0.061 |
| Ciprofloxacin | 58 (95.1%) | 36 (39.6%) | <0.001 |
| Cefepime | 57 (93.4%) | 39 (42.9%) | <0.001 |
| Ceftazidime | 57 (93.4%) | 39 (42.9%) | <0.001 |
| Amikacin | 55 (90.2%) | 26 (28.6%) | <0.001 |
| Piperacillin–tazobactam | 58 (95.1%) | 35 (38.5%) | <0.001 |
| Ampicillin–sulbactam | 56 (91.8%) | 31 (34.1%) | <0.001 |
On univariate logistic analysis, patients with hospital-acquired Acinetobacter spp bacteremia (100% vs. 72.5%, respectively; p< 0.001) and those with the respiratory tract as the source of bacteremia (49.2% vs. 24.2%, respectively; p = 0.001) were significantly more likely to have tigecycline non-susceptible Acinetobacter bacteremia.
Clinical presentation
Severity variables and variables associated with hospital exposures present within 48 h prior to episodes of bacteremia are shown in Table 4. Based on univariate analysis, the following variables were risk factors for tigecycline non-susceptible Acinetobacter bacteremia: Pitt bacteremia score (3 vs. 2, respectively; p< 0.006), ICU stay at risk (60.7% vs. 28.6%, respectively; p< 0.001), hospital days before blood culture (20 vs. 9, respectively; p = 0.005), recent surgery (27.8% vs. 14.3%, respectively; p = 0.039), and high-dose steroids within the 14 days prior to blood culture (45.9% vs. 13.2%, respectively; p< 0.001). The median time spent in the ICU before the onset of Acinetobacter bacteremia was 4 days in the case group and 0 days in the control group (p< 0.001).
Table 4Severity variables and hospital exposures between the case and control groups.
| Case (n = 61) | Control (n = 91) | p-Value | OR (95% CI) | |
|---|---|---|---|---|
| Severity variables at presentation | ||||
| WBC, × 109/l | 3.910 (1.100–12.270) | 8.670 (5.210–13.450) | 0.026 | NA |
| CRP, mg/dl | 9.51 (3.95–15.93) | 6.67 (3.45–16.34) | 0.294 | NA |
| Neutropenia | 13 (21.3%) | 12 (13.2%) | 0.185 | 1.783 (0.752–4.225) |
| Pitt bacteremia score | 3 (1–6) | 2 (0–3) | 0.006 | NA |
| Hospital exposures | ||||
| ICU stays at risk | 37 (60.7%) | 26 (28.6%) | <0.001 | 3.854 (1.941–7.654) |
| Hospital days before culture | 20 (7.5–41.5) | 9 (1–25) | 0.005 | NA |
| ICU stay in days before culture | 4 (0–13.5) | 0 (0–4) | <0.001 | NA |
| Number of admissions in the past year | 2.369 (1–3) | 1 (1–3) | 0.886 | NA |
| Recent surgery | 17 (27.9%) | 13 (14.3%) | 0.039 | 2.318 (1.030–5.217) |
| Recent exposure to cytotoxic drugs | 16 (26.2%) | 30 (33%) | 0.375 | 0.723 (0.352–1.483) |
| Recent exposure to steroids | 28 (45.9%) | 12 (13.2%) | <0.001 | 5.586 (2.538–12.294) |
| Exposures to antibiotics | 58 (95.1%) | 53 (58.2%) | <0.001 | 13.862 (4.040–47.566) |
| Penicillin | 30 (49.2%) | 26 (28.6%) | 0.010 | 2.419 (1.229–4.763) |
| Cephalosporins | 25 (41%) | 21 (23.1%) | 0.018 | 2.315 (1.143–4.689) |
| Quinolones | 19 (31.1%) | 15 (16.5%) | 0.033 | 2.292 (1.056–4.974) |
| Carbapenems | 39 (63.9%) | 18 (19.8%) | <0.001 | 7.189 (3.450–14.982) |
| Antifungal agents | 13 (21.3%) | 8 (8.8%) | 0.028 | 2.810 (1.087–7.264) |
| Tigecycline | 6 (9.8%) | 2 (2.2%) | 0.046 | 4.855 (0.946–24.907) |
OR, odds ratio; CI, confidence interval; WBC, white blood cell count; NA, not applicable; CRP, C-reactive protein; ICU, intensive care unit.
a Data are expressed as the number (%) of patients, or median (interquartile range).
The prior use of antimicrobial agents was categorized into six groups: penicillin, cephalosporins, quinolones, carbapenems, tigecycline, and antifungal agents. There were statistically significant differences between the case and control groups for all classes of antimicrobial agent. Carbapenems showed the highest odds ratio (OR) among the six classes of antimicrobial agent (adjusted OR 7.189, 95% confidence interval (CI) 3.450–14.982; p< 0.001). For this reason, prior exposure to carbapenems was included in the multivariate analysis.
Multivariate analysis for risk factors
By multivariate logistic analysis (Table 5), prior exposure to carbapenems (adjusted OR 4.437, 95% CI 1.970–9.991; p< 0.001) and prior exposure to steroids (adjusted OR 2.887, 95% CI 1.170–7.126; p< 0.020) were independently associated with tigecycline non-susceptible Acinetobacter bacteremia. Although prior exposure to tigecycline was more common in the case group than in the control group (9.8%, 6/61 vs. 2.2%, 2/91; p = 0.046), this variable was found not to be a significant factor associated with tigecycline non-susceptibility after adjustment for other variables (adjusted OR 1.884, 95% CI 0.298–11.920; p = 0.501).
Table 5Multivariable analysis of risk factors for tigecycline non-susceptible Acinetobacter bacteremia.
| Variables | p-Value | Adjusted OR (95% CI) |
|---|---|---|
| Charlson’s WIC | 0.690 | 0.965 (0.809–1.150) |
| Number of admissions in the past year | 0.775 | 0.978 (0.841–1.138) |
| ICU stays at risk | 0.224 | 1.675 (0.730–3.843) |
| Hospital days before culture | 0.984 | 1.000 (0.995–1.005) |
| Recent surgery | 0.552 | 1.349 (0.503–3.617) |
| Recent exposure to steroids | 0.021 | 2.887 (1.170–7.126) |
| Prior exposure to carbapenems | <0.001 | 4.437 (1.970–9.991) |
| Prior exposure to tigecycline | 0.501 | 1.884 (0.298–11.920) |
OR, odds ratio; CI, confidence interval; WIC, weighted index of comorbidity; ICU, intensive care unit.
Discussion
The recent wide dissemination of XDR Acinetobacter spp, which are only susceptible to colistin and tigecycline (
Falagas et al., 2006
), has resulted in the treatment of these infections becoming a serious clinical challenge, and the evidence suggests that hospital-acquired Acinetobacter infections prolong the length of hospital stay and increase healthcare costs (Peleg et al., 2008
, Sunenshine et al., 2007
, Giammanco et al., 2017
, Stefani and Dowzicky, 2016
). For this reason, the treatment of XDR Acinetobacter spp is challenging, and new therapeutic options are required. Tigecycline is regarded as one of the therapeutic options for infections caused by XDR Acinetobacter spp; however, tigecycline resistance is increasingly reported (Al-Sweih et al., 2011
, Navon-Venezia et al., 2007
, Sun et al., 2013
). According to a previous study involving 250 isolates of Acinetobacter spp, 13.6% were resistant to tigecycline and 88.4% were MDR (Al-Sweih et al., 2011
). Rates of non-susceptibility of Acinetobacter spp to tigecycline of >10% have been reported, especially in Asia (Seifert et al., 2006
, Sun et al., 2013
, Park et al., 2009
).Risk factors for MDR Acinetobacter spp or XDR Acinetobacter spp acquisition or infection in various clinical settings have been analyzed in previous studies. An extended hospital stay, severity of the illness, mechanical ventilation, ICU stay, presence of a central venous catheter, invasive procedures including surgery, and previous antibiotic therapy were found to be risk factors (
Moghnieh et al., 2016
, Park et al., 2010
, Ng et al., 2014
, Tacconelli et al., 2008
, Chan et al., 2014
). However, there have been no studies analyzing risk factors that have focused only on tigecycline non-susceptible Acinetobacter bacteremia. Thus, this study was conducted to determine the risk factors for tigecycline non-susceptible Acinetobacter bacteremia.In the current study, those with hospital-acquired Acinetobacter spp bacteremia and those with the respiratory tract as the source of bacteremia were significantly more likely to have tigecycline non-susceptible Acinetobacter bacteremia. Although it is not easy to distinguish upper airway colonization from true pneumonia in many clinical settings, ventilator-associated pneumonia (VAP) or hospital-acquired pneumonia (HAP) due to MDR Acinetobacter spp is likely to occur in patients with upper airway colonization by Acinetobacter spp. As bacterial isolates colonizing the airway are more likely to be antimicrobial-resistant, patients with the respiratory tract as the source of bacteremia are significantly more likely to have tigecycline non-susceptible Acinetobacter bacteremia (
Brotfain et al., 2016
).Although it was attempted to estimate the severity of illness using the Pitt bacteremia score, WBC, and CRP level, only the Pitt bacteremia score was identified as a risk factor for tigecycline non-susceptible Acinetobacter bacteremia on univariate analysis. Simple clinical markers that reflect the host inflammatory response are the WBC with differential and CRP. An elevated WBC is classically associated with bacterial infection; however, values for leukocytosis as a predictor of infection have not been standardized (
Seigel et al., 2012
). The CRP level may be increased significantly in infected patients across the different clinical settings, however it is non-specific (Matson et al., 1991
, Povoa et al., 2005
). This study did not demonstrate an association between leukocytosis or high CRP levels and disease severity, because only the laboratory results at the time of bacteremia were used, rather than the serial changes in WBC and CRP levels, which may represent more significant prognostic markers.This study demonstrated that previous exposure to carbapenems and steroids were independently associated with the development of tigecycline non-susceptible Acinetobacter bacteremia through multivariate analysis. Interestingly, prior exposure to tigecycline was found not to be a significant risk factor associated with tigecycline non-susceptibility. Although the exact mechanisms of tigecycline resistance have not been fully determined, reduced susceptibility to tigecycline appears to be mediated by overexpression of the active efflux pump. However, a correlation between the overexpression of the efflux pump and exposure to tigecycline has still not been proven (
Reid et al., 2007
, He et al., 2015
, Gordon and Wareham, 2009
, Peleg et al., 2007
). A previous in vitro experiment demonstrated that a tigecycline-susceptible clinical strain induced a rapid rise in the MIC of tigecycline after tigecycline exposure (Gordon and Wareham, 2009
, Peleg et al., 2007
). In contrast, other clinical reports have shown that among 56 patients with tigecycline non-susceptible Acinetobacter infections, only six had a history of exposure to tigecycline (Sun et al., 2010
, Deng et al., 2014
). This phenomenon demonstrates that selective pressure caused by other antibiotics might lead to non-susceptibility to tigecycline (Deng et al., 2014
, Kuo et al., 2012
, Hsueh et al., 2005
).Another important feature was that more patients with tigecycline non-susceptible Acinetobacter infections had been treated with carbapenems within 1 month before Acinetobacter spp isolation than those with tigecycline-susceptible Acinetobacter infections. This finding is in accordance with other studies showing that the use of carbapenems might induce resistance to carbapenems, as well as many other antibiotics including tigecycline (
Kuo et al., 2012
). The present study results suggest that carbapenems might be potent inducers of multidrug resistance in Acinetobacter spp, and also emphasize that monitoring the use of broad-spectrum antibiotics such as carbapenems or tigecycline is required to reduce the nosocomial spread of tigecycline non-susceptible Acinetobacter spp.This study has several limitations. First, it was a single-center study, therefore the findings may not be generalizable to other settings. Second, the retrospective design limited the validity and reliability of comparisons. Case and control patients were not matched based on underlying disease and source of bacteremia. Third, it was not possible to assess the infection control strategy, which could be an important risk factor. The survival of Acinetobacter spp in the environment (e.g. respiratory equipment, hospital bed rails, and healthcare workers’ hands) could contribute to its transmission from patient to patient or healthcare worker to patient (
Fournier and Richet, 2006
, Ho et al., 2010
). Because of this, an infection control strategy, such as patient isolation or environmental decontamination, could also have affected the mode and rate of spread of tigecycline non-susceptible Acinetobacter spp. However, it was not possible to determine whether an infection control strategy was implemented. Fourth, resistance mechanisms and the molecular epidemiology of tigecycline non-susceptible Acinetobacter spp were not identified in this study. Finally, the number of patients with exposure to tigecycline was relatively small and the statistical power of the risk factor analyses might be limited.In conclusion, the study data showed that recent exposure to carbapenems and corticosteroids were independent risk factors for acquiring tigecycline non-susceptible Acinetobacter bacteremia. Prior exposure to tigecycline was not identified as a significant factor associated with tigecycline non-susceptibility after adjustment for other variables. These results indicate that the rational use of antibiotics through a comprehensive antimicrobial stewardship program, especially in immunosuppressed patients, may be essential in limiting the emergence and spread of MDR organisms such as tigecycline-resistant Acinetobacter spp, which are difficult to treat.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of interest
No conflict of interest to declare.
Acknowledgements
We thank Jinseob Kim for the statistical advice.
References
- Predictive scoring model of mortality in Gram-negative bloodstream infection.Clin Microbiol Infect. 2013; 19: 948-954
- Emergence of tigecycline and colistin resistance in Acinetobacter species isolated from patients in Kuwait hospitals.J Chemother. 2011; 23: 13-16
- Mechanisms of multidrug resistance in Acinetobacter species and Pseudomonas aeruginosa.Clin Infect Dis. 2006; 43: S49-S56
- Multidrug Resistance Acinetobacter Bacteremia Secondary to Ventilator-Associated Pneumonia: Risk Factors and Outcome.J Intensive Care Med. 2016;
- Risk factors for healthcare-associated extensively drug-resistant Acinetobacter baumannii infections: a case-control study.PLoS One. 2014; 9: e85973
- Performance standards for antimicrobial susceptibility testing. Twenty-fourth international supplement. Document M100-S24.CLSI, Wayne, PA2014
- Molecular epidemiology and mechanisms of tigecycline resistance in clinical isolates of Acinetobacter baumannii from a Chinese university hospital.Antimicrob Agents Chemother. 2014; 58: 297-303
- The diversity of definitions of multidrug-resistant (MDR) and pandrug-resistant (PDR) Acinetobacter baumannii and Pseudomonas aeruginosa.J Med Microbiol. 2006; 55: 1619-1629
- The epidemiology and control of Acinetobacter baumannii in health care facilities.Clin Infect Dis. 2006; 42: 692-699
- Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the infectious diseases society of america.Clin Infect Dis. 2011; 52: e56-e93
- Risk factors for Acinetobacter baumannii nosocomial bacteremia in critically ill patients: a cohort study.Clin Infect Dis. 2001; 33: 939-946
- Global Assessment of the Activity of Tigecycline against Multidrug-Resistant Gram-Negative Pathogensbetween 2004 and 2014 as Part of the Tigecycline Evaluation and Surveillance Trial.mSphere. 2017; 2
- A review of clinical and microbiological outcomes following treatment of infections involving multidrug-resistant Acinetobacter baumannii with tigecycline.J Antimicrob Chemother. 2009; 63: 775-780
- Tigecycline susceptibility and the role of efflux pumps in tigecycline resistance in KPC-producing Klebsiella pneumoniae.PLoS One. 2015; 10: e0119064
- Global spread of carbapenem-resistant Acinetobacter baumannii.J Antimicrob Chemother. 2010; 65: 233-238
- Epidemiology and clonality of multidrug-resistant Acinetobacter baumannii from a healthcare region in Hong Kong.J Hosp Infect. 2010; 74: 358-364
- CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting.Am J Infect Control. 2008; 36: 309-332
- Relationships between antimicrobial use and antimicrobial resistance in Gram-negative bacteria causing nosocomial infections from 1991-2003 at a university hospital in Taiwan.Int J Antimicrob Agents. 2005; 26: 463-472
- Imipenem: a potent inducer of multidrug resistance in Acinetobacter baumannii.Int J Antimicrob Agents. 2012; 39: 33-38
- Nosocomial acquisition of multiresistant Acinetobacter baumannii: risk factors and prognosis.Clin Infect Dis. 1995; 20: 790-796
- Acinetobacter baumannii: epidemiology, antimicrobial resistance, and treatment options.Clin Infect Dis. 2008; 46: 1254-1263
- C-reactive protein as a diagnostic test of sepsis in the critically ill.Anaesth Intensive Care. 1991; 19: 182-186
- Extensively drug-resistant Acinetobacter baumannii in a Lebanese intensive care unit: risk factors for acquisition and determination of a colonization score.J Hosp Infect. 2016; 92: 47-53
- High tigecycline resistance in multidrug-resistant Acinetobacter baumannii.J Antimicrob Chemother. 2007; 59: 772-774
- A multicenter case-case control study for risk factors and outcomes of extensively drug-resistant Acinetobacter baumannii bacteremia.Infect Control Hosp Epidemiol. 2014; 35: 49-55
- Tigecycline treatment experience against multidrug-resistant Acinetobacter baumannii infections: a systematic review and meta-analysis.Int J Antimicrob Agents. 2016; 47: 107-116
- In vitro activity of tigecycline against colistin-resistant Acinetobacter spp: isolates from Korea.Int J Antimicrob Agents. 2009; 33: 289-290
- Extensively drug-resistant Acinetobacter baumannii: risk factors for acquisition and prevalent OXA-type carbapenemases–a multicentre study.Int J Antimicrob Agents. 2010; 36: 430-435
- Antibiotic therapy for Klebsiella pneumoniae bacteremia: implications of production of extended-spectrum beta-lactamases.Clin Infect Dis. 2004; 39: 31-37
- Looking for risk factors for the acquisition of antibiotic resistance: a 21st-century approach.Clin Infect Dis. 2002; 34: 1564-1567
- Tigecycline Efflux as a Mechanism for Nonsusceptibility in Acinetobacter baumannii.Antimicrob Agents Chemother. 2007; 51: 2065-2069
- Acinetobacter baumannii: emergence of a successful pathogen.Clin Microbiol Rev. 2008; 21: 538-582
- C-reactive protein as a marker of infection in critically ill patients.Clin Microbiol Infect. 2005; 11: 101-108
- Rapid development of Acinetobacter baumannii resistance to tigecycline.Pharmacotherapy. 2007; 27: 1198-1201
- Scoring systems for prediction of mortality in patients with intensive care unit-acquired sepsis: a comparison of the Pitt bacteremia score and the Acute Physiology and Chronic Health Evaluation II scoring systems.Shock. 2009; 31: 146-150
- AdeABC multidrug efflux pump is associated with decreased susceptibility to tigecycline in Acinetobacter calcoaceticus-Acinetobacter baumannii complex.J Antimicrob Chemother. 2007; 59: 1001-1004
- Comparative in vitro activities of tigecycline and 11 other antimicrobial agents against 215 epidemiologically defined multidrug-resistant Acinetobacter baumannii isolates.J Antimicrob Chemother. 2006; 58: 1099-1100
- Inadequacy of temperature and white blood cell count in predicting bacteremia in patients with suspected infection.J Emerg Med. 2012; 42: 254-259
- In vitro activity of tigecycline and colistin against A. baumannii clinical bloodstream isolates during an 8-year period.J Chemother. 2015; 27: 266-270
- Assessment of the Activity of Tigecycline against Gram-Positive and Gram-Negative Organisms Collected from Italybetween 2012 and 2014, as Part of the Tigecycline Evaluation and Surveillance Trial (T.E.S.T.).Pharmaceuticals (Basel). 2016; 9
- Overexpression of the adeB gene in clinical isolates of tigecycline-nonsusceptible Acinetobacter baumannii without insertion mutations in adeRS.Antimicrob Agents Chemother. 2010; 54: 4934-4938
- The emergence of clinical resistance to tigecycline.Int J Antimicrob Agents. 2013; 41: 110-116
- Multidrug-resistant Acinetobacter infection mortality rate and length of hospitalization.Emerg Infect Dis. 2007; 13: 97-103
- Prediction models to identify hospitalized patients at risk of being colonized or infected with multidrug-resistant Acinetobacter baumannii calcoaceticus complex.J Antimicrob Chemother. 2008; 62: 1130-1137
- Epidemic Acinetobacter baumannii strain with MRSA-like behaviour carried by healthcare staff.Eur J Clin Microbiol Infect Dis. 2002; 21: 326-327
- Nosocomial bloodstream infections caused by Acinetobacter species in United States hospitals: clinical features, molecular epidemiology, and antimicrobial susceptibility.Clin Infect Dis. 2000; 31: 690-697
Article info
Publication history
Published online: July 01, 2017
Accepted:
June 26,
2017
Received in revised form:
June 2,
2017
Received:
February 13,
2017
Corresponding Editor: Eskild Petersen, Aarhus, DenmarkIdentification
Copyright
© 2017 The Author(s). Published by Elsevier Ltd on behalf of International Society for Infectious Diseases.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
