Abstract
Background:
Multi-drug resistant (MDR) Acinetobacter is a growing concern and has been identified as a serious threat by the Centers of Disease Control and Prevention. However there is little information on MDR Acinetobacter in individuals with spinal cord injuries and disorders (SCI/D). Therefore, the objective of this study was to identify risk factors for, and assess outcomes of MDR Acinetobacter in Veterans with SCI/D.
Methods:
This was a retrospective cohort study from January 1, 2012 to December 31, 2013 using national Veterans Affairs (VA) medical encounter and microbiology data.
Results:
A total of 773 Acinetobacter cultures were identified in 571 patients, of which 58.9% were MDR. Inpatient culture, male gender, sputum and other specimen type, receipt of antibiotics within 90 days before culture date, and pressure ulcers were identified as independent predictors of MDR Acinetobacter. Highest odds of MDR Acinetobacter was seen with previous antibiotic use (OR=7.27; 95% CI: 2.59-20.54). 30-day mortality was 5.3% in this study. MDR, previous mechanical ventilation 90 days before the culture, and cancer were all independent risk factors for 30 day mortality.
Discussion:
There should be increased efforts to highlight the importance of antimicrobial stewardship in order to improve infection control to help limit spread of Acinetobacter in healthcare settings.
Background:
Multi-drug resistant gram-negative organisms (MDRGNOs) have been an increasing concern in both healthcare and community settings.1 Infections with MDRGNOs are associated with increased mortality, prolonged hospital length of stay, and higher healthcare costs.2 The Centers of Disease Control and Prevention (CDC) estimate that at least 23,000 people die each year as a direct result of an antibiotic-resistant infection.1 Among MDRGNOs, multi-drug resistant (MDR) Acinetobacter is of particular concern due to this organism’s frequent role in severe hospital-acquired infections and nosocomial outbreaks.3;4 Approximately 63% of all Acinetobacter infections are considered multi-drug resistant (MDR), and MDR Acinetobacter has been highlighted as a ‘Serious Threat’ by the CDC.1 Furthermore, approximately 2% of all healthcare-associated infections reported to the National Healthcare Safety Network (NHSN) in 2009-2010 were caused by Acinetobacter baumannii; of these, 68% of A. baumannii central-line associated bloodstream infections (CLABSI) and 78% of A. baumannii catheter associated urinary tract infections (CAUTI) were reported to be MDR.4
Individuals with spinal cord injuries and disorders (SCI/D) are at increased risk for infections due to frequent contact with the healthcare system and use of invasive devises such as vascular and urinary catheters.5;6 The increased and repeated use of antibiotics to treat infections in SCI/D may also increase the risk for MDR infections.6 MDR infections have ranged around 60% in those with SCI/D.7 Risk factors for MDR organisms in persons with SCI/D include catheterization, pressure ulcers, and previous antibiotic exposure.5;6;8;9 Acinetobacter infections range from 0.7-12.6% in persons with SCI/D9;10, and Acinetobacter baumannii has been found to be more frequent in those with SCI/D versus those without SCI/D.7 Existing literature on Acinetobacter infections in patients with SCI/D is scarce, and are limited to urine cultures and urinary tract infections,9;10 are embedded within studies focusing on multiple organisms,5;7-10 and do not focus on outcomes.5;7-10
While several studies have focused on risk factors and outcomes for MDR Acinetobacter infections, no studies have specifically evaluated this organism in SCI/D. Similar to other MDRGNOs, previous antibiotic use and specifically broad spectrum antibiotics such as carbapenems and fluoroquinolones are risk factors for MDR Acinetobacter in the general population.11;12 Other notable risk factors are mechanical ventilation,13;14 pressure ulcers,11;15 and male gender.16 MDR Acinetobacter infections are also associated with higher rates of mortality and longer hospital stays.16;17
Therefore the objective of this study was to describe the burden of, identify risk factors for, and assess outcomes of MDR Acinetobacter in Veterans with SCI/D.
Methods:
Study Design, Setting, & Population
This was a retrospective cohort study from January 1, 2012 to December 31, 2013 using national Veterans Affairs (VA) medical encounter and microbiology data. Data were included for 110 VA facilities. The sample included all Veterans with SCI/D seen at a VA medical center during the study period. We excluded Veterans with multiple sclerosis and amyotrophic lateral sclerosis as individuals with stable non-progressive spinal cord neurological deficits are the focus of the VA SCI/D system of care. Veterans with SCI/D were included in the sample if they had a positive culture for Acinetobacter spp. within the study period.
Only those cultures with associated antibiotic sensitivity testing performed were included in this study. Cultures were identified as MDR if they were resistant to at least 1 agent in at least 3 or more antimicrobial categories tested, as defined by Magiorokos et al.3 These antimicrobial categories were aminoglycosides, antipseudomonal carbapenems, antipseudomonal fluoroquinolones, antipseudomonal penicillins + β-lactamase inhibitors, extended-spectrum cephalosporins, folate pathway inhibitors, penicillins + β-lactamase inhibitors, polymixins, and tetracyclines.5 All isolates not meeting the MDR criteria were categorized as non-MDR. Duplicate cultures, defined as Acinetobacter cultures with the same antibiotic susceptibility profile, were excluded if the duplicate culture occurred within 30 days of a previous culture.18 This study was approved by the Edward Hines Jr VA Hospital Institutional Review Board.
Data Sources & Definitions
Patient demographics, healthcare utilization, facility characteristics, microbiology, and pharmacy data were collected from the VA Corporate Data Warehouse (CDW). The CDW is a national repository that includes clinical and administrative data from the Veterans Health Administration (VHA) and is updated daily. Mortality data was obtained from the VA Vital Status File which contains dates of death combined from the Veterans Benefits Administration (VBA) Beneficiary Identification and Records Locator System (BIRLS) Death file, the VA Medicare Vital Status File, and the Social Security Administration Death Master File. SCI/D characteristics such as duration and level of injury were obtained from the VA SCD Registry, a national database containing information on spinal cord characteristics derived from patient registries.
Demographic data such as age, sex, race/ethnicity, and comorbidities were identified 365 days prior to and during the visit or admission where the culture was identified. Specimen type was categorized into four categories including urine, blood, sputum and other. Other specimen type included wound, tissue, body fluid, and bone cultures. Comorbidities were identified using ICD-9 codes of conditions from the Deyo-Charlson comorbidity index, including pressure ulcer which is a common condition in patients with SCI/D.19 30-day and 1-year mortality was defined as death of the patient within 30 days and 1 year from the date of the culture collected. 30-day hospital readmission was defined as patient readmission within 30-days of the positive culture date and was only applicable for those patients with an initial culture collected in an inpatient setting.
The VA SCI/D system of care includes 24 regional SCI Centers, called ‘hubs’, which provide comprehensive care delivered by interdisciplinary teams. These ‘hub’ SCI/D centers are connected to ‘spoke’ facilities that provide community-based care for SCI/D patients. Data was analyzed to assess whether there were differences in MDR Acinetobacter in ‘hub’ versus ‘spoke’ facilities. Region was defined using the U.S. Census Bureau regions. San Juan, Puerto Rico, and Manilla were grouped into the ‘South’ region.20
Statistical Analysis
Bivariate analysis using chi-square tests were conducted and unadjusted odds ratios (OR) and 95% confidence intervals (95% CIs) were calculated to identify potential risk factors and outcomes of MDR Acinetobacter. A multivariable regression model, cluster adjusted by patient to account for multiple cultures per patient, was utilized to identify independent predictors. Variables that were significant in the unadjusted results were included in the multivariable model. Bivariate analysis was also utilized to identify risk factors related to 30-day mortality. Only the first culture of Acinetobacter per patient during the study period was included for mortality analysis. Poisson regression was used to identify independent risk factors for 30-day mortality and the most parsimonious model was selected. Incidence rate rations (IRRs) and 95% CIs were reported for the Poisson regression. A p-value ≤ 0.05 was considered statically significant. All statistical analyses were performed using STATA software version 14.1 (Stata Corp LP).
Results:
There were a total of 978 Acinetobacter cultures identified during the study timeframe. After excluding duplicates (n=205) there were 773 Acinetobacter cultures identified in 571 individuals with SCI/D. Overall, of a total 27,904 MDRGNO cultures among 8,691 patients were identified, of which 2.8% of cultures were Acinetobacter and 6.6% of SCI/D patients were found to be colonized or infected with Acinetobacter. Over two thirds (66.8%) of all Acinetobacter cultures were obtained from patients seen at SCI specialty care centers, with inpatient cultures accounting for 60.4% of all isolates. This is comparable to the overall MDRGNO cultures where 63.0% were obtained from SCI centers.
The average age and duration of injury of the cohort was 61.3 years and 18.4 years respectively, and 48.1% had tetraplegia. Over half (58.9%, n=455) of the 773 Acinetobacter cultures, were MDR. Highest rates of MDR were seen in the 50-64 age group (61.9%) compared to those younger than 50 (46.8%) and those greater than 65 (58.7%). Over two thirds of all inpatient (71.7%) and long term care (LTC) (65.1%) cultures were MDR respectively, while only 32.5% of outpatient cultures were MDR. Cultures at SCI centers had a higher proportion of MDR (62.6%) compared to non-SCI VA facilities (51.4%). Furthermore, significant variation was seen by U.S. geographic region where the highest proportion of MDR Acinetobacter was seen in the Northeast (72.1%) while the lowest was in the Midwest (53.5%).
The bivariate analysis showed a number of factors that were associated with MDR Acinetobacter. Older age, male gender, and Northeast region were more likely to be associated with MDR Acinetobacter (Table 1). MDR Acinetobacter was not associated with any of the SCI characteristics examined except for a protective effect for MDR in those with duration of injury 11-20 years and 21+ years compared to those injured for 0-10 years. MDR Acinetobacter was more likely to be found in an inpatient culture versus an outpatient culture (OR=5.27; 95% CI: 3.77-7.37) and at an SCI center (OR=1.58; 95% CI: 1.17-2.15). Compared to urine specimens, blood specimens (OR= 2.71; 95% CI: 1.16-6.32), sputum specimens (OR= 3.83; 95% CI: 2.30-6.40) and other specimen types (OR= 3.08; 95% CI: 2.17-4.38) were more likely to be MDR. Comorbidities that were associated with multi-drug resistance were chronic heart failure, cerebrovascular accident, myocardial infarction, and pressure ulcer. Hospitalization, surgery, mechanical ventilation, and ICU stay in the 90 days prior to positive culture were associated with increased odds of multi-drug resistance. Antibiotic use in the previous 90 days was also associated with an increased odds of MDR (OR=3.04; 2.20-4.19). Most antibiotic exposures in the previous 90 days also had increased odds of MDR with the exception of 1st and 2nd generation cephalosporins. All bivariate results are displayed in Table 1.
Table 1.
Bivariate analysis of risk factors for MDR Acinetobacter
| Characteristic | MDR (n=455) | Non-MDR (n=318) |
Unadjusted OR (95% CI) |
p-value |
|---|---|---|---|---|
| Demographics & SCI Characteristics | ||||
| Age (years) | ||||
| 18-49 | 44 (9.7%) | 50 (15.7%) | Reference | |
| 50-64 | 239 (52.5%) | 147 (46.2%) | 1.85 (1.17-2.91) | p=0.008 |
| 65+ | 172 (37.8%) | 121 (38.1%) | 1.62 (1.01-2.58) | p=0.043 |
| Gender | ||||
| Male | 452 (99.3%) | 310 (97.5%) | Reference | |
| Female | 3 (0.7%) | 8 (2.5%) | 0.26 (0.07-0.98) | p=0.032 |
| Region | ||||
| Northeast | 62 (13.6%) | 24 (7.5%) | Reference | |
| Midwest | 85 (18.7%) | 74 (23.3%) | 0.44 (0.25-0.78) | p=0.005 |
| South | 257 (56.5%) | 176 (55.4%) | 0.56 (0.34-0.94) | p=0.028 |
| West | 51 (11.2%) | 44 (13.8%) | 0.45 (0.24-0.83) | p=0.011 |
| Extent | ||||
| Incomplete | 159 (34.9%) | 131 (41.2%) | Reference | |
| Complete | 229 (50.3%) | 145 (45.6%) | 1.30 (0.95-1.78) | p=0.097 |
| Missing | 67 (14.7%) | 42 (13.2%) | 1.31 (0.84-2.06) | p=0.234 |
| Level | ||||
| Tetraplegia | 221 (48.6%) | 151 (47.5%) | Reference | |
| Paraplegia | 189 (41.5%) | 141 (44.3%) | 0.92 (0.68-1.24) | p=0.567 |
| Missing | 45 (9.9%) | 26 (8.2%) | 1.18 (0.70-2.00) | p=0.532 |
| Onset of Injury | ||||
| Non-Traumatic | 97 (21.35) | 64 (20.1%) | Reference | |
| Traumatic | 279 (61.35) | 222 (69.8%) | 0.83 (0.58-1.19) | p=0.31 |
| Missing | 79 (17.4%) | 32 (10.1%) | 1.63 (0.97-2.73) | p=0.064 |
| Duration of Injury | ||||
| 0-10 years | 189 (41.5%) | 102 (32.1%) | Reference | |
| 11-20 years | 54 (11.9%) | 51 (16.0%) | 0.57 (0.36-0.90) | p=0.015 |
| 21+ years | 168 (36.9%) | 128 (40.3%) | 0.71 (0.51-0.99) | p=0.042 |
| Missing | 44 (9.7%) | 37 (11.6%) | 0.64 (0.39-1.06) | p=0.082 |
| Clinical Characteristics | ||||
| Source | ||||
| Outpatient | 79 (17.4%) | 164 (51.6%) | Reference | |
| Inpatient | 335 (73.6%) | 132 (41.5%) | 5.27 (3.77-7.37) | p<0.0001 |
| Long Term Care | 41 (9.0%) | 22 (6.9%) | 3.87 (2.16-6.93) | P<0.0001 |
| SCI Center | ||||
| No | 132 (29.0%) | 125 (39.3%) | Reference | |
| Yes | 323 (71.0%) | 193 (60.7%) | 1.58 (1.17-2.15) | p=0.003 |
| Specimen Type | ||||
| Urine | 200 (44%) | 228 (71.7%) | Reference | |
| Blood | 19 (4.2%) | 8 (2.5%) | 2.71 (1.16-6.32) | p=0.017 |
| Sputum | 74 (16.3%) | 22 (6.9%) | 3.83 (2.30-6.40) | p<0.0001 |
| Other | 162 (35.6%) | 60 (18.9%) | 3.08 (2.17-4.38) | p<0.0001 |
| Charlson Comorbidity index, median (SD) | 2.6 (1.7) | 1.8 (1.9) | p<0.0001 | |
| Pressure Ulcer | ||||
| No | 153 (33.6%) | 225 (70.8%) | Reference | |
| Yes | 302 (66.4%) | 93 (29.2%) | 4.78 (3.50-6.51) | p<0.0001 |
| Healthcare exposures in previous 90 days | ||||
| Long Term Care | ||||
| No | 442 (97.1%) | 311 (97.8%) | Reference | |
| Yes | 13 (2.9%) | 7 (2.2%) | 1.31 (0.52-3.31) | p=0.572 |
| ICU Stay | ||||
| No | 358 (78.7%) | 293 (92.1%) | Reference | |
| Yes | 97 (21.3%) | 25 (7.9%) | 3.18 (1.99-5.06) | p<0.0001 |
| Previous Hospitalization | ||||
| No | 231 (50.8%) | 224 (70.4%) | Reference | |
| Yes | 224 (49.2%) | 94 (29.6%) | 2.31 (1.71-3.13) | p<0.0001 |
| Surgery | ||||
| No | 361 (79.3%) | 279 (87.7%) | Reference | |
| Yes | 94 (20.7%) | 39 (12.3%) | 1.86 (1.24-2.79) | p=0.002 |
| Mechanical Ventilation | ||||
| No | 390 (85.7%) | 301 (94.7%) | Reference | |
| Yes | 65 (14.3%) | 17 (5.3%) | 2.95 (1.69-5.14) | p=0.0001 |
| Genitourinary procedure | ||||
| No | 432 (94.9%) | 305 (95.9%) | Reference | |
| Yes | 23 (5.1%) | 13 (4.1%) | 1.25 (0.62-2.50) | p=0.530 |
| Antibiotic exposure in previous 90 days | ||||
| Any Antibiotic | ||||
| No | 50 (11.0%) | 101 (31.8%) | Reference | |
| Yes | 405 (89.0%) | 217 (68.2%) | 3.77 (2.59-5.50) | p<0.0001 |
| Chronic Steroids | ||||
| No | 450 (98.9%) | 315 (99.1%) | Reference | |
| Yes | 5 (1.1%) | 3 (0.9%) | 1.17 (0.28-4.92) | p=0.834 |
| Extended Spectrum Penicillins | ||||
| No | 277 (60.9%) | 271 (85.2%) | Reference | |
| Yes | 178 (39.1%) | 47 (14.8%) | 3.71 (2.58-5.32) | p<0.0001 |
| 1st & 2nd Generation Cephalosporins | ||||
| No | 383(84.2%) | 280 (88.1%) | Reference | |
| Yes | 72 (15.8%) | 38 (11.9%) | 1.39 (0.91-2.11) | p=0.129 |
| 3rd & 4th Generation Cephalosporins | ||||
| No | 284 (62.4%) | 268 (84.3%) | Reference | |
| Yes | 171 (37.6%) | 50 (15.7%) | 3.23 (2.26-4.61) | p<0.0001 |
| Carbapenems | ||||
| No | 305 (67.0%) | 295 (92.8%) | Reference | |
| Yes | 150 (33.0%) | 23 (7.2%) | 6.31 (3.95-10.06) | p<0.0001 |
| Tetracyclines | ||||
| No | 413 (90.8%) | 309 (97.2%) | Reference | |
| Yes | 42 (9.2%) | 9 (2.8%) | 3.49 (1.67-7.28) | p=0.0004 |
| Aminoglycosides | ||||
| No | 3613 (79.3%) | 304 (95.6%) | Reference | |
| Yes | 94 (20.7%) | 14 (4.4%) | 5.65 (3.16-10.12) | p<0.0001 |
| Quinolones | ||||
| No | 231 (50.8%) | 221 (69.5%) | Reference | |
| Yes | 224 (49.2%) | 97 (30.5%) | 2.21 (1.63-2.99) | p<0.001 |
| Vancomycin | ||||
| No | 217 (47.7%) | 255 (80.2%) | Reference | |
| Yes | 238 (52.3%) | 63 (19.8%) | 4.44 (2.19-5.38) | p<0.0001 |
The multivariable regression analysis showed that inpatient culture, male gender, sputum and other specimen type, receipt of antibiotics within 90 days before culture date, and pressure ulcers were independent predictors of MDR Acinetobacter. The highest odds of MDR Acinetobacter was seen with previous antibiotic use (OR=7.27; 95% CI: 2.59-20.54). A sputum specimen had a 5.6 times increased odds for MDR (OR=5.59; 95% CI: 1.39-22.48) while other specimen types had 7.66 times increased odds for MDR (OR=7.66; 95% CI: 2.53-23.20). Although female gender was significantly protective in the bivariate results, it was dropped from the final multivariable model as the data were too sparse with only 9 females included in the study. Results of the multivariable regression can be seen in Table 2.
Table 2.
Multivariable regression analysis of risk factors for MDR Acinetobacter
| Variables | Adjusted Odds Ratio (95% CI), p-value |
|---|---|
| Source (Reference=Outpatient) | |
| Inpatient | 8.45 (2.82-25.93), p<0.0001 |
| Long Term Care | 1.84 (0.33-10.22), p=0.485 |
| Specimen Type (Reference=Urine) | |
| Blood | 1.62 (0.21-12.77), p=0.647 |
| Sputum | 5.06 (1.279-22.68), p=0.022 |
| Other | 7.47 (2.46-22.68), p<0.0001 |
| Any antibiotic in previous 90 days | 9.38 (3.08-29.57), p<0.001 |
| Pressure Ulcer | 5.69 (2.02-16.14), p<0.0001 |
When evaluating outcomes based on MDR, there were a total of 41 deaths (5.3%) within 30 days of the culture and 176 deaths (22.8%) within 1 year. In the bivariate analysis, mortality was greater in those with MDR Acinetobacter for 30-day mortality and 1-year mortality. MDR did not increase 30-day readmission in inpatients. For the mortality analysis, cultures were limited to first culture per patient with a total of 571 patients in the cohort. There were 25 deaths within 30-days of first culture. Average age of the cohort was 61.2 with a mean Charlson score of 2.2. Bivariate analysis showed that, source, specimen type, multi-drug resistance, cancer, pressure ulcers, previous hospitalization, previous mechanical ventilation, and previous exposure to 3rd & 4th generation cephalosporins, carbapenems, quinolones, colimethates, and vancomycins. A polymicrobial culture was found to be protective for mortality. Unadjusted odds ratios are displayed in Table 3. The multivariate Poisson model showed that MDR, previous mechanical ventilation 90 days before the culture, and cancer are all independent risk factors for 30 day mortality. Cancer was the largest risk factor (IRR: 6.41, 95% CI: 1.45-28.40). Those with MDR cultures had 4.4 times the risk of 30 day mortality (IRR: 4.4, 95% CI: 1.48-13.10). Results of the multivariate regression for 30-day mortality are also displayed in Table 3.
Table 3.
Bivariate and multivariable regression analysis of risk factors of 30-day Mortality
| Characteristic | Died (n=25) | Alive (n=546) |
Unadjusted OR (95% CI), p-value |
IRR (95% CI), p-value |
|---|---|---|---|---|
| MDR | ||||
| No | 4 (16.0%) | 247 (50.2%) | Reference | Reference |
| Yes | 21 (84.0%) | 272 (49.8%) | 5.23 (1.79-15.61). p=0.003 | 4.40 (1.48-13.10), p=0.008 |
| Duration of Injury | ||||
| 0-10 years | 7 (28.0%) | 200 (36.6%) | Reference | |
| 11-20 years | 2 (8.0%) | 81 (14.9%) | 0.71 (0.14-3.47) | |
| 21+ years | 11 (44.0%) | 207 (37.9%) | 1.52 (0.58-3.99) | |
| Missing | 5 (20.0%) | 58 (10.6%) | 2.46 (0.75-8.05) | |
| Source | ||||
| Outpatient | 1 (4.0%) | 196 (35.9%) | Reference | |
| Inpatient | 21 (84.0%) | 309 (56.6%) | 13.32 (1.78-99.82), p=0.12 | |
| LTC | 3 (12.0%) | 41 (7.5%) | 14.34 (1.46-141.35), p<0.001 | |
| Specimen | ||||
| Urine | 6 (24.0%) | 313 (57.3%) | Reference | |
| Blood | 4 (16.0%) | 18 (3.3%) | 11.59 (3.00-44.78), p<0.001 | |
| Sputum | 8 (32.0%) | 51 (9.4%) | 8.18 (2.73-24.56), p<0.001 | |
| Other | 7 (28.0%) | 164 (30.0%) | 2.23 (0.74-6.73), p=0.156 | |
| Cancer | ||||
| No | 23 (92.0%) | 541 (99.1%) | Reference | Reference |
| Yes | 2 (8.0%) | 5 (0.92%) | 9.41 (1.73-51.10), p=0.009 | 6.41 (1.45-28.4), p=0.014 |
| Pressure Ulcer | ||||
| No | 8 (32.0%) | 288 (52.7%) | Reference | |
| Yes | 17 (68.0%) | 258 (47.3%) | 2.37 (1.01-5.59), p=0.048) | |
| Previous Hospitalization | ||||
| No | 10 (40.0%) | 342 (62.6%) | Reference | |
| Yes | 15 (60.0%) | 204 (37.4%) | 2.51 (1.11-5.70), p=0.027 | |
| Mechanical Ventilator Use | ||||
| No | 16 (64.0%) | 500 (91.6%) | Reference | Reference |
| Yes | 9 (36.0%) | 46 (8.4%) | 6.11 (2.56-14.60), p<0.001 | 3.57 (1.53-8.32), p<0.001 |
| 3rd & 4th Generation Cephalosporins | ||||
| No | 13 (52.0%) | 417 (76.4%) | Reference | |
| Yes | 12 (48.0%) | 129 (23.6%) | 2.98 (1.33-6.70), p=0.008 | |
| Carbapenems | ||||
| No | 14 (56.0%) | 467 (85.5%) | Reference | |
| Yes | 11 (44.0%) | 79 (14.5%) | 4.64 (2.04-10.60), p<0.001 | |
| Quinolones | ||||
| No | 9 (36.0%) | 348 (63.7%) | Reference | |
| Yes | 16 (64.0%) | 198 (36.3%) | 3.12 (1.36-7.20), p=0.007 | |
| Colismethate | ||||
| No | 24 (96.0%) | 545 (99.8%) | Reference | |
| Yes | 1 (4.0%) | 1 (0.2%) | 22.71 (1.37-374.1), p=0.029 | |
| Vancomycin | ||||
| No | 8 (32.0%) | 366 (67.0%) | Reference | |
| Yes | 17 (68.0%) | 180 (33.0%) | 4.32 (1.83-10.20), p=0.001 | |
| Polymicrobial culture | ||||
| No | 22 (88.0%) | 316 (57.9%) | Reference | |
| Yes | 3 (12.0%) | 230 (42.1%) | 0.19 (0.05-.63), p=0.007 |
Note: Previous hospitalization, mechanical ventilator use, and antibiotics were assessed 90 days before Acinetobacter culture.
Discussion:
The goal of this study was to identify risk factors for MDR Acinetobacter and assess outcomes in Veterans with SCI/D. Our results showed that over half the A. baumannii cultures in the study period were MDR. This is similar to the CDCs national rate of 63%.1 Considering that 6.6% of all SCI/D patients with cultures had Acinetobacter in our study, it is important to note that prevalence of MDR Acinetobacter is prominent in Veterans with SCI/D.
Our study also showed that antibiotic exposure in the previous 90 days was the greatest risk factor for MDR Acinetobacter, with nearly 7-fold increased odds for MDR Acinetobacter compared to patients with non-MDR isolates. Previous antibiotic exposure has been identified in prior studies as a significant risk factor for MDR Acinetobacter. 13;12;15 Specifically, previous receipt of carbapenems,11;12 quinolones, 12;15 and metronidazole12 have all been found to increase risk. Although our multivariable analysis focused on any type of antibiotic exposure, results from the bivariate analysis showed that receipt of quinolones and carbapenems, specifically, increased odds of MDR in this population.
Furthermore, this study identified that presence of a pressure ulcer, upon admission or in the year prior to the culture, as well as an inpatient culture increased the odds of MDR. Pressure ulcers have been identified as a risk factor for MDROs in general patient populations, both in acute care and long-term care settings.11;15 Our study found that persons with SCI/D with a pressure ulcer had nearly 6 times the odds of MDR Acinetobacter. Our findings were twice the rate of that identified by Lim et al. (OR=3.69 95%CI: 1.06-12.86) at a long-term care facility.11 Due to neurologic impairments and immobility, patients with SCI/D are at much higher risk of pressure ulcers compared to general patient populations, and patients with SCI/D and pressure ulcers also have frequent and prolonged contact with the healthcare system. This likely contributes substantially to the increased risk of MDROs in patients with SCI/D. Our study confirms this association for MDR Acinetobacter, in particular, and further supports the importance of appropriate pressure ulcer prevention and care in this population to minimize risk of MDRO colonization and spread.
We also found that, compared to urine cultures, sputum and other specimen cultures were more likely to be MDR. Although bloodstream and urinary tract infections are common, Acinetobacter is also frequency cultured from the respiratory tract, wounds, and intra-abdominal infections where it may be a true pathogen or a colonizer.4;21;22 High rates of MDR are often observed from Acinetobacter isolates obtained from these sites.22 Few prior studies have examined the importance of specimen type in determining the likelihood of MDR in Acinetobacter isolates in patients with SCI/D, as most prior studies of risks for MDROs in patients with SCI/D have been limited to the urinary tract.9;10
The 30-day mortality for this study was 4.4%, when limiting to the first Acinetobacter culture which is lower than other studies reporting 30-day mortality between 30% and 44.8%.12;23 This difference in 30-day mortality may be due to our restriction to first Acinetobacter specimen per patient within the study period for mortality. Additionally Gulen et al. and Guo et al. included only blood stream isolates in their study compared to our study that also included urine and wound isolates and these patients are more likely to be colonized and not infected. We found that those with MDR Acinetobacter had higher odds for 30 day mortality and 1 year mortality compared to those with non-MDR Acinetobacter. This is consistent with Fitzpatrick et al. showing that MDR Acinetobacter results in worse outcomes17 and MDR is a risk factor for mortality.2;23
Other factors identified as a risk factor for mortality were cancer and mechanical ventilation. This is not surprising as infections are a leading cause of mortality among cancer patients as they are immunocompromised.14 Cancer was found to be the greatest risk factor for 30-day mortality in our study. Cancer has been identified as a risk factor for MDR Aceinetobacter14 and has been noted as a concern.24 This is concerning in SCI/D as multiple comorbidities increase risk of infection in this population.
Mechanical ventilation was also an independent risk factor for 30-day mortality in Veterans with SCI/D and use of mechanical ventilation has been identified as a risk factor in other studies.13;23 Acinetobacter baumannii is commonly detected in those with healthcare associated and ventilator associated pneumonia and could be why mechanical ventilation was a risk factor for mortality.25 Guo et al. and Park et al. both identified mechanical ventilation as current use in patients.13;23 However, our study defined mechanical ventilation as use up to 90 days prior to the culture. Collectively these findings suggest both current and previous mechanical ventilator use present risk for mortality from Acinetobacter infections.
Our study has some important limitations. First, we did not distinguish between infection and colonization and did not collect data on antibiotic treatments prescribed due to Acinetobacter. Not specifying between colonization and infection may have contributed to the lower mortality rate reported in this study if many of the patients included were colonized rather than infected. Second, our study employed a retrospective cohort design, which did not allow us to separate factors associated with both non-MDR cases and uninfected individuals. Third, we did not capture any non-VA data such as cultures performed outside of the VA system; this may introduce bias as there may be unique characteristics associated with this group of patients. Finally, because our study used exclusively VA data with predominantly male subjects, these results may not be generalizable to other populations with SCI/D.
An advantage of this study is the benefit of data from the Veterans Health Administration (VHA), the largest integrated health care system in the U.S. as well as the largest provider of care of patients with SCI/D. Our study is the largest known study of MDR Acinetobacter for patients with SCI/D.
Risk factors identified for both MDR and 30-day mortality in our study were similar to those found in the general population. However, for Veterans with SCI/D these risk factors are an added concern as they are often in contact with healthcare systems, exposed to antibiotics, and frequently use invasive devices. Antibiotic exposure is strong risk factor that is potentially modifiable. Therefore, there should be increased efforts to highlight the importance of antimicrobial stewardship in order to decrease multi-drug resistant infection and improve infection control to help limit spread of Acinetobacter in healthcare settings.
Acknowledgements:
This work was supported by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Rehabilitation Research and Development Service Merit Review Award (grant no. B-1583-P). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government.
References:
- 1).Centers for Disease Control. Antibiotic resistance threats in the United States, 2013. Last accessed 8 Aug 2016.
- 2).Vardakas KZ, Rafailidis PI, Konstantelias AA, Falagas ME. Predictors of mortality in patients with infections due to multi-drug gram negative bacteria: The study, the patient, the bug or drug? J Infect 2013; 66: 401–14. [DOI] [PubMed] [Google Scholar]
- 3).Magiorakos AP, Srinivasan A, Carey RB, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbial Infect 2012; 18(3):268–81. [DOI] [PubMed] [Google Scholar]
- 4).Sievert DM, Ricks P, Edwards JR, Schneider A, Patel J, Srinivasan A, Kallen A, Limbago B, Fridkin S, National Healthcare Safety Network (NHSN( Team and Participating NHSN Facilites. Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2009–2010. Infect Control Hosp Epidemiol 2013; 34(1):1–14. [DOI] [PubMed] [Google Scholar]
- 5).Evans CT, LaVela SL, Weaver FM, Priebe M, Sandford P, Niemiec S, Miskevics S, Parada JP. Epidemiology of hospital-acquired infections in veterans with spinal cord injury and disorder. Infect Control Hosp Epidemiol 2008; 29(3):234–42. [DOI] [PubMed] [Google Scholar]
- 6).Fitzpatrick MA, Suda KJ, Safdar N, Goldstein B, Jones MM, Poggensee L, Ramanathan S, LeWan R, Evans CT. Unique risks and clinical outcomes associated with extended-spectrum beta-lactamase Enterobacteriaceae in Veterans with spinal cord injury/disorder: a case-case-control study. Infect Control Hosp Epidemiol 2016; 37(7):768–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7).Suda KJ, Patel UC, Sabzwari R, Cao L, Ramanathan S, Hill JN, Evans CT. Bacterial susceptibility patterns in patients with spinal cord injury and disorder (SCI/D): an opportunity for customized stewardship tools. Spinal Cord 2016; 1–9. [DOI] [PubMed] [Google Scholar]
- 8).Mylotte JM, Kahler L, Graham R, Young L, Goodnough S. Prospective surveillance for antibiotic-resistant organisms in patients with spinal cord injury admitted to an acute rehabilitation unit. Am J Infect Control 2000; 28; 291–7. [DOI] [PubMed] [Google Scholar]
- 9).Togan T, Azap OK, Durukan E, Arslan H. The prevalence, etiologic agents, and risk factors for urinary tract infection among spinal cord injury patients. Jundishapur J Microbiol 2014; 7(1):e8905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10).Yoon SB, Lee BS, Lee KD, Hwang SI, Lee HJ, Han ZA. Comparison of bacterial strains and antibiotic susceptibilities in urinary isolates of spinal cord injury patients from the community and the hospital. Spinal Cord 2014; 52: 298–301. [DOI] [PubMed] [Google Scholar]
- 11).Lim CJ, Cheng AC, Kennon J, Spelman D, Hale D, Melicann G, Sidjabat HE, Paterson DL, Kong DCM, Peleg AY. Prevalence of multi-resistant organisms and risk factors for carriage in long-term facilities: a nested case-control study. J Antimicrob Chemother 2014; 69: 1972–80. [DOI] [PubMed] [Google Scholar]
- 12).Gulen TA, Guner R, Celikbilek N, Keske S, Tasyran M. Clinical importance and cost of bacteremia caused by nosocomial multi drug resistant acinetobacter baumanii. Int J Infect dis 2015; 38:32–5. [DOI] [PubMed] [Google Scholar]
- 13).Nseir S, Blazejewski C, Lubert R, Wallet F, Courcol R, Durocher A. Risk of acquiring multidrug-resistant Gram-negative bacilli from prior room occupants in the intensive care unit. Clin Microbiol Infect 2011; 17: 1201–08. [DOI] [PubMed] [Google Scholar]
- 14).Fukata Y, Muder RR, Agha ME, Clarke LG, Wagener MM, Hensler AM, Doi Y. Risk factors for acquisition of multidrug-resistant Acinetobacter baumanii among cancer patients. Am J Infect Contr 2013: 41: 1249–52 [DOI] [PubMed] [Google Scholar]
- 15).Cheng VCC, Chen JHK, So SYC, Wong SCY, Yan MK, Chau PH, Lee WM, To KKW, Chan JFW, Hung IFN, Ho PL, Yuen KY. Use of fluoroquinolones is the single most important risk factor for the high bacterial load in patients with nasal and gastrointestinal colonization by multidrug-resistant Acinetobacter baumannii. Eur J Clin Microbiol Infect Dis 2015; 34: 2359–66. [DOI] [PubMed] [Google Scholar]
- 16).Wareham DW, Bean DC, Khanna P, Hennessy EM, Krahe D, Ely A, Miller E. Bloodstream infection due to Acinetobacter spp: epidemiology, risk factors and impact of multi-drug resistance. Eur J Clin Microbiol Infec Dis 2008; 27:607–12. [DOI] [PubMed] [Google Scholar]
- 17).Fitzpatrick MA, Ozer E, Bolon MK, Hauser AR. Influence of ACB complex genospecies on clinical outcomes in a U.S. hospital with high rates of multidrug resistance. J Infect 2015; 70(2): 144–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18).CLSI. Analysis and Presentation of Cumulative Antimicrobial Susceptibility Test Data, 3rd edn, Guideline M39-A3. Clinical and Laboratory Standards Institute: Wayn, PA, USA, 2009. [Google Scholar]
- 19).Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373–383. [DOI] [PubMed] [Google Scholar]
- 20).U.S. Census Bureau. Census Regions and Divisions of the United States. Available from: URL: http://www.census.gov/geo/maps-data/maps/reference.html. Last accessed 9 Feb 2017. [Google Scholar]
- 21).Manchanda V, Sanchaita S, Singh NP. Multidrug resistant Acinetobacter. J Glob Infect Dis 2010;2(30): 291–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22).Lob SH, Hoban DJ, Sahm DF, Badal RE. Regional differences and trends in antimicrobial susceptibility of Acinetobacter baumannii. Int J Antimicrob Agents. 2016. April;47(4):317–23 [DOI] [PubMed] [Google Scholar]
- 23).Guo N, Xue W, Tang D, Ding J, Zhao B. Risk factors and outcomes of hospitalized patients with blood infections caused by multidrug-resistant Acinetobacter baumannii complex in a hospital of Northern China. Am J Infect Control 2016; 44: e37–39. [DOI] [PubMed] [Google Scholar]
- 24).Friere MP, de Oliveira Garcia D, Garcia CP, Campagnari Bueno MF, Cmargo CH, Kono Magri AS, Francisco GR, Reghini R, Vieira MF, Ibrahim KY, Rossi F. Hajjar L, Levin AS, Hoff PM, Pierrotti LC, Abdala E. Bloodstream infection caused by extensively drug-resistant Acinetobacter baumannii in cancer patients: high mortality associated with delayed treatment rather than with the degree of neutropenia. Clin Microbiol Infec 2016; 22(4): 352–8. [DOI] [PubMed] [Google Scholar]
- 25).Qi F, Zhang GX, Liang ZX, Wang RT, Yang Z, Chen LA, Cui JC. Healthcare-associated pneumonia: clinical features and retrospective analysis over 10 years. Chin Med J (Engl) 2015; 128(20): 2707–13. [DOI] [PMC free article] [PubMed] [Google Scholar]