Different Clinical Characteristics and Impacts of Carbapenem-Resistance on Outcomes Between Acinetobacter Baumannii and Pseudomonas Aeruginosa Bacteraemia: A Multicentre Prospective Observational Study

Background: Carbapenem-resistance (CR) causes poor clinical outcomes and has limited treatment options. We aim to evaluate the differences in clinical characteristics and impact of carbapenem-resistance on outcomes between Acinetobacter baumannii (ABA) and Pseudomonas aeruginosa (PAE) bacteraemia. Methods: We prospectively identied all patients with ABA and PAE bacteraemia in 10 hospitals over a 1-year period and collected their detailed clinical information. Treatment failure was dened as all-cause 30-day mortality, persistent bacteraemia, or recurrence within 30 days. Results: We included 304 ABA and 241 PAE bacteraemia cases. CR was detected in 216 ABA (71%) and 55 PAE (23%). Treatment failure was signicantly higher in CR-ABA than in CR-PAE (60.6% vs. 34.5%, P = 0.001). Multivariate analyses were stratied by patient data according to CR and the appropriateness of empirical therapy. For ABA patients, severe sepsis or septic shock and high Pitt bacteraemia score were independent risk factors for treatment failure in the inappropriate empirical antibiotics group. Pneumonia was a signicant risk factor in the appropriate group. For PAE patients, hospital-acquired infection and high Pitt bacteraemia score were independent risk factors for treatment failure in both groups. CR was an independent risk factor in ABA for treatment failure in both the inappropriate (adjusted odds ratio [aOR]: 6.17, 95% condence interval [CI]: 1.13-33.75, P = 0.036) and appropriate empirical treatment groups (aOR: 4.15, 95% CI: 1.16-14.84, P = 0.029), but not for patients with PAE bacteraemia. Conclusions: We demonstrated signicant differences in the clinical characteristics and impact of CR on the clinical outcomes between ABA and PAE bacteraemia. Our ndings suggest that different approaches may be needed to treat ABA and PAE bacteraemia. factors associated with treatment failure in ABA and PAE bacteraemia. Carbapenem resistance was associated with increased risk of treatment failure in ABA, but not in PAE bacteraemia. Because carbapenem resistance is signicantly associated with treatment failure in the appropriate empirical antibiotics group with ABA, this study suggests that appropriate empirical antibiotics cannot improve the clinical outcomes in CR-ABA. Hence, we suggest that it would be important to reduce the occurrence of CR-ABA through infection control in high-risk groups such as patients in intensive care units.


Background
Antimicrobial resistance is a major threat to public health. Recent reports from the Centers for Disease Control and Prevention concluded that the threat of antibiotic-resistance was greater than previously understood, and listed carbapenem resistance in non-fermenting gram negative bacteria, such as Acinetobacter baumannii (ABA) and Pseudomonas aeruginosa (PAE), as urgent threats (1). There are limited treatment options for carbapenem resistance; due to its severity and poor clinical outcomes, it is an emerging public health concern (2-4). Therefore, to further improve treatment for infectious diseases that show resistance to antibiotics, identifying their clinical features is critical.
Both carbapenem-resistant ABA (CR-ABA) and carbapenem-resistant PAE (CR-PAE) are important causes of healthcare-associated infections (5). Further, it is known that antibiotic susceptibility and clinical features vary between them. However, there are no integrated observational clinical studies assessing the clinical impacts of carbapenem resistance in ABA and PAE. Due to the limited treatment options for CR-ABA and CR-PAE, it is important to elucidate the impact of carbapenem resistance on the clinical outcomes and risk factors for treatment failure. Therefore, we aimed to evaluate the differences in the clinical characteristics and impacts of carbapenem resistance on clinical outcomes between ABA and PAE bacteraemia.

Patient population and clinical data
All patients with ABA or PAE bacteraemia were identi ed prospectively from clinical microbiology laboratories in 10 hospitals over a 1-year period from 1 September 2017 to 31 August 2018. We prospectively collected patients' demographic and microbiological data, underlying medical conditions, source of bacteraemia, clinical severity, antibiotic therapy, and clinical outcomes. This study was approved by the Institutional Review Board (IRB) of Seoul National University Bundang Hospital (IRB no. B-1804-463-105) and the IRBs of each participating hospital. The study was conducted in accordance with the principles outlined in the 1964 Declaration of Helsinki and its later amendments. The need for informed consent was waived by the IRBs.

De nitions
Mixed bacteraemia was de ned as the isolation of other organisms except common skin ora, such as coagulase-negative staphylococci (6), from the rst blood culture that yielded PAE or ABA. Bacteraemia detected in patients who were hospitalised for over 48 hours was classi ed as a hospital-acquired infection (HAI). Healthcare-associated infection was de ned as ful lling at least one of the following criteria: (1) hospitalisation for 2 or more days in the previous 12 months; (2) residence in a nursing home in the previous 12 months; (3) treatment with parenteral antibiotics within 30 days; and (4) renal replacement therapy within 30 days. Use of immunosuppressant was de ned as the use of any anticancer chemotherapy, corticosteroids more than 20 mg of prednisolone equivalent over 1 week, or any other immunosuppressive agent within the past 30 days.
The severity of illness was assessed using the Pitt bacteraemia score (7), and the severity of underlying disease was assessed using Charlson's weighted index of comorbidity (WIC) (8). Empirical antimicrobial therapy was de ned as initial treatment used more than 24 hours after the sampling of the rst positive blood culture. Antibiotic therapy was considered inappropriate if the isolate was not susceptible to the treatment regimen in vitro.
Treatment failure was de ned as ful lling any of the following events: (1) all-cause 30-day mortality; (2) persistent bacteraemia de ned as positive blood cultures after ≥7 days of appropriate antimicrobial therapy; (3) recurrent bacteraemia within 30 days after discontinuation of appropriate antimicrobial therapy (9).

Data management and statistical analysis
The chi-squared test or Fisher's exact test were used to compare categorical variables, and the Student ttest or Mann Whitney U test were used to compare continuous variables. A stepwise multiple logistic regression was used to identify independent risk factors for treatment failure. Risk factors with a P value of <0.10 in univariate analysis and other variables of clinical signi cance were included in multivariate analysis. To avoid multicollinearity, severe sepsis or septic shock and Pitt bacteraemia score were analysed in a separate multivariate model. All underlying diseases and Charlson's WIC were analysed in a separate multivariate model for the same reason. P values <0.05 were considered statistically signi cant. The Hosmer-Lemeshow test was used to assess the goodness of t for logistic regression models. All statistical analyses were performed using IBM SPSS Statistics, version 26.0 (IBM Corp., Armonk, NY, USA).
The clinical characteristics of PAE bacteraemia according to carbapenem susceptibility are described in Table S2. The CR-PAE group showed signi cantly higher percentages of HAI (61.8% vs. 45.2%, P = 0.030) and underlying liver disease (25.5% vs. 14.0%, P = 0.044) than the CS-PAE group. The CR-PAE group showed signi cantly higher percentages of inappropriate empirical therapy than the CS-PAE group (52.7% vs. 26.3%, P < 0.001). There was no signi cant difference in treatment failure between the CR-PAE and CS-PAE groups (34.5% vs. 24.7%, P = 0.150).
Comparisons of clinical characteristics of the CR-ABA and CR-PAE groups are shown in Table 1. The CR-ABA group showed signi cantly higher percentages of severe sepsis or septic shock than the CR-PAE group. Inappropriate empirical therapy was more frequent in the CR-ABA group than the CR-PAE group.
All-cause 30-day mortality and treatment failure were signi cantly higher in the CR-ABA group than the CR-PAE group. Antibiotic susceptibility of carbapenem-resistant isolates Antibiotic susceptibility of carbapenem-resistant isolates is described in Table 2. Not all results of susceptibility tests were available for carbapenem-resistant isolates. For all antibiotics except colistin and trimethoprim/sulfamethoxazole, the percentages of non-susceptible isolates were signi cantly different between the CR-ABA and CR-PAE groups. The percentages of non-susceptible isolates, except for tigecycline and minocycline, were signi cantly higher in the CR-ABA group than the CR-PAE group.

Risk factors for treatment failure
The results of univariate analysis between risk factors and treatment failure in ABA are listed in Table S3. We strati ed the patients' data according to carbapenem resistance and the appropriateness of empirical therapy in the multivariate analyses of risk factors for treatment failure, because of the signi cant interaction between those factors. Table 3 shows the risk factors for treatment failure in ABA identi ed using multivariate analysis according to the appropriateness of empirical therapy. While severe sepsis or septic shock and high Pitt bacteraemia score were independent risk factors for treatment failure in the inappropriate empirical antibiotics group, pneumonia, and Charlson's WIC ≥ 3 were independent risk factors for treatment failure in the appropriate empirical antibiotics group. The results of univariate analysis between risk factors and treatment failure in PAE bacteraemia are listed in Table S5. Signi cant risk factors for treatment failure in PAE identi ed by multivariate analysis are listed by the appropriateness of empirical therapy in Table 4. HAI, severe sepsis or septic shock, and high Pitt score were independent risk factors for treatment failure in both the inappropriate and appropriate empirical antibiotics groups. Urinary tract infection was associated with decreased treatment failure in the appropriate antibiotics group.

Discussion
Resistance to carbapenem is an important concern in both ABA and PAE bacteraemia. Here, we found different clinical characteristics and impacts of carbapenem-resistance on outcomes in ABA and PAE bacteraemia.
Carbapenem resistance was an independent risk factor for treatment failure in ABA bacteraemia regardless of the appropriateness of empirical antibiotics. This nding agrees with previous studies, which report that carbapenem resistance is signi cantly associated with mortality in ABA bacteraemia (10,11). A systematic review showed that carbapenem resistance may increase the risk of mortality of ABA infection (12). In contrast, we found that carbapenem resistance was not an independent risk factor for treatment failure in PAE bacteraemia regardless of the appropriateness of empirical antibiotics. This is in agreement with a previous study showing that carbapenem resistance was not an independent risk factor for mortality in PAE bacteraemia (13). This difference in the impact of carbapenem-resistance on ABA and PAE may be due to available treatment options, different sites of infection, and differences of clinical severity.
We found different susceptibilities and clinical characteristics in ABA and PAE bacteraemia. In ABA bacteraemia, 71.1% (216/304) of the cases were carbapenem resistant, while only 22.8% (55/241) of the cases in PAE bacteraemia were carbapenem resistant. The higher percentage of carbapenem resistance in ABA compared with PAE bacteraemia is consistent with previous results, which showed that the proportion of carbapenem resistance was 92.1% in ABA strains and the proportions of resistance were 19.5% to imipenem and 18.1% to meropenem in PAE strains (14). We found that CR-ABA have resistance to almost all antibiotics; however, CR-PAE is susceptible to some antibiotics such as cipro oxacin and amikacin. The difference in antibiotic susceptibility might explain the difference in the proportion of inappropriate empirical therapies between the groups. However, the results of multivariate analysis using strati cation to the appropriateness of empirical therapy showed that carbapenem resistance is an independent risk factor for treatment failure in ABA, even if the empirical therapy was appropriate. This might be related to the virulence of ABA. A previous study suggested that high bacterial cytotoxicity signi cantly affected the mortality in ABA ventilator-associated pneumonia (15).
Treatment failure was not signi cantly different between CR-PAE and CS-PAE, whereas it was signi cantly different between CR-ABA and CS-ABA. Moreover, a previous study found a signi cant difference in hospital mortality between CR-ABA and CS-ABA, but not between CR-PAE and CS-PAE (16).
Compared to the CR-PAE group, the CR-ABA group showed a signi cantly higher percentage of treatment failure (60.6% vs. 34.5%). The different clinical severity and antibiotics susceptibility between CR-ABA and CR-PAE might contribute to the differences in treatment failure. Clinicians should consider the higher rates of treatment failure when carbapenem resistance is detected in ABA.
In ABA bacteraemia, we found independent risk factors, including pneumonia, Charlson's WIC ≥ 3, and carbapenem resistance in the appropriate empirical antibiotics group. We speculate that if pneumonia was the source of bacteraemia, or if carbapenem resistance was detected in ABA bacteraemia, the risk of treatment failure would be high, even if appropriate empirical antibiotics were administered. One study identi ed independent risk factors for mortality, including carbapenem resistance, neutropenia, and prolonged intensive care unit stay in ABA bacteraemia (10). Another study showed that septic shock, carbapenem resistance, pneumonia, and inappropriate de nite antimicrobial therapy are independent risk factors for 14-day mortality (17).
Independent risk factors for treatment failure in PAE bacteraemia included HAI, severe sepsis or septic shock, and high Pitt bacteraemia score, regardless of the appropriateness of empirical antibiotics. One retrospective study reported the presence of septic shock and high Acute Physiologic Assessment and Chronic Health Evaluation II score as independent risk factors for 14-day mortality (13). Another study showed that recent hospitalisation, corticosteroid treatment, Charlson WIC, non-urinary source, and high Sequential Organ Failure Assessment score are signi cant risk factors for 30-day mortality in PAE bacteraemia (18).
By additional analysis according to carbapenem resistance, we found that inappropriate empirical antibiotics were independently associated with treatment failure in the CS-PAE group. Appropriate empirical antibiotics, including antipseudomonal coverage, may be important in the management of patients with risk of Pseudomonas infection. The independent risk factors for treatment failure in the CR-PAE group were HAI, use of immunosuppressants, and high Pitt bacteraemia score. This is consistent with a previous study reporting that high Pitt bacteraemia score is a predictor of death in CR-PAE bacteraemia (19).
In both the CR-ABA and CS-ABA groups, inappropriate empirical antibiotics was not an independent factor for treatment failure. Furthermore, a previous study found that 28-day mortality was not independently associated with appropriate empirical antimicrobial therapy in invasive Acinetobacter infection (20). Although studies found that appropriate antimicrobial therapy decreased 14-day mortality (21,22), these de ned appropriate antimicrobial therapy as antibiotics administration within 48 hours after the onset of bacteraemia and did not address the appropriateness of empirical antibiotics.
This study has several limitations. First, being an observational study, it was prone to bias and unmeasured confounding factors. Second, no microbiologic investigation other than antibiotic susceptibility tests was conducted on the strains. As the effect of carbapenemase on clinical outcomes is not clear, further studies on the effect of the microbiological characteristics on the outcome in ABA and PAE are needed. Lastly, while colistin was the only drug available in Korea for pandrug-resistant ABA and di cult-to-treat PAE at the time of the study, several new drugs, such as ceftolozane-tazobactam, ceftazidime-avibactam, imipenem-relebactam, and ce derocol were available in other countries (23,24).
This difference in availability of drugs may have contributed to the differences in prognosis.

Conclusions
In conclusion, we identi ed clinical characteristics and risk factors associated with treatment failure in ABA and PAE bacteraemia. Carbapenem resistance was associated with increased risk of treatment failure in ABA, but not in PAE bacteraemia. Because carbapenem resistance is signi cantly associated with treatment failure in the appropriate empirical antibiotics group with ABA, this study suggests that appropriate empirical antibiotics cannot improve the clinical outcomes in CR-ABA. Hence, we suggest that it would be important to reduce the occurrence of CR-ABA through infection control in high-risk groups such as patients in intensive care units.

Declarations
Ethics approval and consent to participate