Clinical, microbiological characteristics and predictors of mortality in patients with carbapenemase-producing Enterobacterales bloodstream infections: a multicentre study

doi:10.21203/rs.3.rs-2757860/v1

Download PDF

Research Article

Clinical, microbiological characteristics and predictors of mortality in patients with carbapenemase-producing Enterobacterales bloodstream infections: a multicentre study

https://doi.org/10.21203/rs.3.rs-2757860/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objectives

To investigate the microbiological characteristics and clinical outcomes of patients with bloodstream infections (BSI) due to carbapenemase-producing Enterobacterales (CPE).

Methods

A multicentre retrospective service evaluation of patients with BSIs due to CPE admitted to six UK hospitals was conducted between 2011 and 2021. Multivariate analysis was used to identify factors predicting 30-day case fatality rate (CFR).

Results

There were 84 episodes of CPE-BSIs, 37 (44%) due to OXA-48, 35 (42%) to metallo-betalactamases (MBL) and 12 (14%) to KPC. 63% of patients were male with a median age of 64 years. Common organisms included Klebsiella spp. (61%), Escherichia coli (20%) and Enterobacter spp. (13%). Urinary devices were more often involved in OXA-48 BSIs (12/37; 32%) compared to infections caused by MBL and KPC (4/35; 11% and 1/12; 8%; p 0.046). In contrast, central venous catheters were more frequently present in KPC-BSIs (10/12; 92%) compared with OXA-48 and MBL (11/37; 30% and 20/35; 57%; p 0.002). Effective definitive antimicrobials were received by 72/84 (86%) patients – either empirically or following microbiology results – comprising monotherapy (32/72; 44%) or combination therapy (40/72; 56%). Overall, 30-day case fatality rate (CFR) was 38%. Sepsis or septic shock was associated with death [OR 3.81 (CI 1.19–12.14), p 0.024].

Conclusion

CFR of CPE-BSI remains relatively high. We identified risks from urinary devices and intravenous catheters, but further larger studies are needed to characterise and compare the clinical and microbiological profile of BSIs due to different types of carbapenemases.

carbapenemase-producing Enterobacterales

CPE

OXA-48

KPC

MBL

NDM

bloodstream infection

bacteraemia

Bloodstream infections (BSIs) caused by carbapenemase-producing Enterobacterales (CPE) are a public health concern associated with a 75% increase in in-hospital mortality (1, 2), prolonged hospitalisation and increased healthcare costs (3). The most common carbapenemases are oxacillinase-48 (OXA-48), Klebsiella pneumoniae carbapenemase (KPC), Guiana extended-spectrum beta-lactamase (GES) and the metallo-beta-lactamases (MBL) known as New Delhi metallo-beta-lactamase (NDM), Verona integron-mediated metallo-beta-lactamase (VIM), and active-on-imipenem carbapenemase (IMP); these have considerable geographical variation. Indeed, the distribution of CPE is variable across the UK, with 11% of referred isolates coming from the Southeast of England and 38% of isolates being from London (3). In the UK, the prevalence of CPE bloodstream infections has progressively risen in the last decade, accounting for 10.4% of the total reported CPE infections under surveillance in 2020 (3).

Clinical management and antibiotic choices in patients with CPE bloodstream infection are not yet standardised. The European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and the Infectious Diseases Society of America (IDSA) recently published guidelines with recommendations on the treatment of resistant Gram-negative infections,including infections caused by CPE (4, 5). However, there are limited options available for the treatment of some CPE infections, particularly those harbouring MBL, where therapeutic failure is not uncommon (6). Furthermore, studies of CPE-BSIs are scarce in the literature and most focus on one specific class of carbapenemase, and often do not report detailed antimicrobial susceptibility data. For this reason, studies providing detailed analysis of CPE-BSI sensitivity data combined with clinical outcomes across regions with high incidence may have the potential to improve clinical management for these patients in the short-term.

We aimed to investigate the microbiological characteristics, antimicrobial and clinical outcomes and to identify factors associated with 30-day case fatality rate (CFR) in patients with a CPE-BSI presenting to hospitals in London, UK over a decade.

Study design and participants

A multicentre retrospective service evaluation of patients with BSIs due to CPE admitted to four university hospitals and two district general hospitals in London was conducted from 1^st November 2011 to 31^st October 2021. Adult patients (>18 years old) with monomicrobial BSIs due to CPE were included. Samples were excluded if: (i) more than one organism grew in the blood culture (polymicrobial sample); or (ii) a second or subsequent repeat sample was positive for the same CPE during the same hospital admission. Each patient had only one episode/admission with CPE BSI.

Study outcomes

The main outcome was descriptive: demographics, microbiological characteristics, antimicrobial and clinical outcomes of patients with bloodstream infections due to CPE were compared between the three classes of carbapenemase identified (OXA-48, KPC and MBL). Note that some isolates harboured more than one resistance mechanism, and for the purpose of this study, MBL took precedence over KPC, and KPC over OXA-48 - recognising the typically broader resistance patterns and clinical significance of each carbapenemase. A subsequent anlysis was undertaken to identify risk factors associated with 30-day CFR.

Data collection and demographic variable definitions

Cases were identified by searching laboratory information management systems, or other available databases, at individual hospitals. Presence of CPE genes was confirmed by nucleic acid amplification assays (NAATs) or lateral flow immunochromatographic methods (https://www.eucast.org/resistance_mechanisms) in UKAS-accredited diagnostic laboratories. Demographic, clinical, microbiological and treatment data were extracted from the medical records by investigators at each centre and entered into a secure data collection tool. Further laboratory results were collected at each site from laboratory information management systems. All collected data from individual sites were pseudonymised and collated into a single secure data collection sheet for analysis.

The demographic and baseline variables collected were age, gender, previous colonisation by CPE, underlying medical conditions and Charlson comorbidity index for adults (7). Immunosuppression was defined as the presence of diabetes mellitus, neutropenia, HIV infection, or receipt of steroids (at any dose) or other immunosuppressive agents in the 30 days prior to bacteraemia. Nosocomial acquisition was defined as onset of symptoms >48 hours after hospital admission or within 48 hours of a previous hospital discharge. Severity of illness at the time of onset of infection was assessed by the presence of severe sepsis or septic shock, defined as sepsis with hypotension (SBP <90 mmHg) or requiring vasopressors with elevated lactate (>2 mmol/L). Invasive procedures included mechanical ventilation, insertion of central venous catheter, urinary devices, intra-abdominal or prosthetic devices. Source of infection was determined as pneumonia, urinary tract infection, intra-abdominal, line-related infection, bone and joint, using the definitions of the Centers for Disease Control and Prevention (8). Early source control was determined as an intervention that contributed to achieving control of the infection (e.g. drainage of a collection, indwelling device removal) that occurred within 72 hours of the positive blood culture. Infection onset was defined as the date and time that the index CPE blood culture was collected.

Microbiological data

Bacterial identification and antibiotic susceptibilities were performed and reported by UKAS-accredited laboratories according to the protocols extant at the time, with additional identification of resistance mechanisms and antimicrobial susceptibility by the reference laboratory where required (Public Health England, Colindale, UK). Isolate sensitivity tests reported as intermediate were deemed not sensitive. In instances of antimicrobial susceptibility discrepancies between the local and the reference laboratories, the result from the reference laboratory took precedence.

Antimicrobial outcomes

Antibiotic therapy was defined as effective if the isolate was susceptible in vitro to that antibiotic. We defined effective combination therapy as a regimen including two or more in vitro active antimicrobial, and monotherapy as including only a single active drug. Effective antimicrobial treatment was categorized as i) effective empirical treatment which was administered before antimicrobial susceptibility results were available, and ii) effective definitive treatment, which was administered following susceptibility results. Early effective therapy was defined as the receipt of at least one in vitro active antibiotic within 48 hours of blood culture collection.

Clinical outcomes

Case fatality was defined as all-cause death within 7, 30 or 90 days after the positive blood culture. Length of stay was time to discharge from hospital after positive blood culture.

Statistical analysis

For descriptive analysis, categorical variables were presented as absolute numbers and their relative percentages; continuous variables were summarized as mean ± standard deviation (SD) or median and interquartile range (IQR), according to their distribution. Baseline characteristics and treatment outcomes were compared between the groups of patients according to carbapenemase gene. ANOVA tests were used to examine differences across these groups. Multinomial logistic regression was used to build a model to explore which factors might influence 30-day mortality. Variables with a p-value of <0.1 on the univariate analysis, were included in the model. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated. All p-values were two-tailed and a p-value of <0.05 was considered statistically significant. The statistical analysis was performed using IBM SPSS v21.0.

Ethical approval

The study was initally approved by St. George’s Hospitals NHS Foundation Trust as a service evaluation, and approval was also obtained at each participating centre according to local requirements. According to the UK Policy Framework for Health and Social Care Research v3.2 October 2017, service evaluations do not require formal ethical review. Informed consent was not obtained as included patients were not subjected to extra procedures or questions. No personal information was stored in the study database. Samples were collected as part of standard care, which was unchanged during the service evaluation.

Between 1^st November 2011 and 31^st October 2022, 89 patients with BSI due to CPE were identified. Of these, 5 patients were excluded due to age <18 years (n = 2) and missing key data (n = 3). The remaining 84 patients were included in the main analysis, representing 84 distinct hospital admissions (see Figure 1). The median Charlson score of patients in this cohort was 5 (range 0-14).

Baseline characteristics

Of 84 episodes of BSIs caused by CPE, 37 (44%) were due to OXA-48, 35 (42%) to MBL, and 12 (14%) to KPC. 7 isolates had 2 resistance mechanisms: 4 had both MBL and OXA-48, and a further 3 had KPC and OXA-48. Among the 35 MBL isolates, New Delhi β-lactamase (NDM) accounted for 28, IMP β-lactamase (IMP) accounted for 6 cases, and VIM β-lactamase (VIM) was detected in 1 case (Fig. 1). 53/84 (63%) of the patients were male, and the median age of all patients was 64 years (range: 18-95 years). Causative organisms included Klebsiella spp. (51/84; 61%), E. coli (20/84; 24%), Enterobacter spp. (11/84; 13%), Serratia marcescens (1/84; 1%) and Proteus mirabillis (1/84; 1%). Nosocomial acquisition was identified in 59 (70%) cases. Prior colonisation was reported in 25 (30%) cases, although it is worth noting that screening policies and their implementation vary between hospitals. The common sources of bacteraemia included urine (21/84; 25%), endovascular (18/84; 21%), intra-abdominal/biliary (20/84; 24%), respiratory (8/84; 10%), and skin and soft tissue (3/84; 4%); the source was unclear or not identified in 14 (17%) cases. Invasive devices were present in 64 patients (76%) at the time of the BSI. Underlying diseases notably included haematological malignancy (29; 35%), diabetes mellitus (26; 31%), chronic liver disease (20; 24%), chronic kidney disease (18; 21%) and organ transplantation (5; 6%). Severe sepsis or septic shock was present in 49 (59%).

Baseline characteristics across groups: OXA-48, KPC and MBL

Klebsiella spp. accounted for 11/12 (92%) of KPC isolates, as expected, 23/37 (62%) of OXA-48 isolates, and 17/35 (49%) of MBL isolates. In contrast, Enterobacter spp. harboured MBL almost exclusively, with this resistance mechanism being identified in 10/11 (91%) of all Enterobacter isolates. Urinary tract infection was identified as the main source of BSI in 15 (41%) of OXA-48 cases - significantly more commonly than in KPC and MBL (1; 8% and 5; 14%)(p = 0.012). Consistent with this, urinary devices were more frequently present in OXA-48-related BSIs (12/37; 32%) compared to KPC and MBL (1/11; 9% and 4/28; 14%)(p = 0.046). Central venous catheters and peripherally inserted central catheters, were the commonest invasive devices identified in this series, and were over-represented in KPC BSIs (10/11; 90%) compared with MBL and OXA-48 (20/35; 57% and 11/37; 30%)(p = 0.002) (Table 1).

Microbiological characteristics

None of the isolates was sensitive to either piperacillin-tazobactam or ertapenem, and only 8 were sensitive to ceftriaxone – these were all OXA-48-producers. This is notable as these agents are common first-line agents for treating sepsis. Meropenem was reported as susceptible in 31/84 (37%) of all CPE isolates, including OXA-48 (27/37; 73%) and MBL (4/35; 11%). Ciprofloxacin was susceptible in 25/75 (33%), including OXA-48 (14/36; 40%), MBL (10/31; 33%) and KPC (1/10; 10%). Amikacin was susceptible in 54/84 (64%), including OXA-48 (30/37; 81%), MBL (22/35; 63%) and KPC (2/12; 17%). Fosfomycin was susceptible in 56/71 (79%), for OXA-48 (26/34; 77%), MBL (25/30; 83%) and KPC (5/7; 71%). Tigecycline was susceptible in 41/71 (58%), for OXA-48 (19/31; 61%), MBL (19/28; 68%) and KPC (3/12; 25%). Colistin was susceptible in 68/71 (96%) of all CPE-BSI. Ceftazidime-avibactam was susceptible in 24/38 (63%), for OXA-48 (16/18; 89%), MBL (3/10; 30%) and KPC (5/10; 50%). Ceftolozane-tazobactam was susceptible for OXA-48 (6/10; 60%), MBL (2/5; 40%) and KPC (0/1; 0%) and cefiderocol was susceptible in 2/2 (100%) of MBL isolates (see Figure 2). Antimicrobial susceptibility testing profile for each type of carbapenemase and microorganism is displayed in Figure 3.

Antimicrobial outcomes

Empiric antimicrobials were effective in vitro in 16/84 (19%) cases, and effective definitive treatment was received by 72/84 (86%) patients, of which monotherapy was prescribed in 32/72 (44%) and combination therapy in 40/72; (56%). OXA-48 infections were most likely to have effective definitive treatment with monotherapy (24/37; 67%), compared to KPC and MBL (2/12; 22% and 6/27; 22%, respectively)(p <0.001). Almost 3/4 of our patients received effective definitive antimicrobial treatment within three days of blood culture positivity. The most frequent antimicrobials used in monotherapy were ciprofloxacin, meropenem, aminoglycosides and ceftazidime-avibactam. The most common antibiotics used in combination therapy included colistin, aminoglycosides, meropenem, tigecycline, and ceftazidime-avibactam. The specific antimicrobials administered, time to start treatment and duration of antimicrobials are shown in Table 2.

Clinical outcomes

Thirty-seven (44%) cases were admitted to intensive care following BSI onset. The median length of hospital stay was 15 days (range: 0-241 days). 7-day CFR after BSI onset was 21%, rising considerably to 38% by day 30, and 44% by day 90. No mortality differences were found across different CPE groups (see Table 2).

In the univariate analysis for 30-day CFR, we found that severe sepsis or septic shock [OR 3.87 (CI 1.42-10.57), p = 0.008] was associated with higher mortality and in vitro antimicrobial resistance to meropenem [OR 0.32 (CI 0.12-0.88), p = 0.024] was found to be associated with 30-day mortality. However, in the multivariate analysis, only severe sepsis or septic shock [OR 3.81 (CI 1.19-12.14), p = 0.024] was associated with higher mortality.

Our study involved a series of 84 patients with CPE-BSI, including 37 patients with OXA-48, 35 patients with MBL and 12 with KPC presenting over a 10-year period. More than half of the patients presented with severe sepsis or septic shock. Amikacin was effictive in vitro in 64% of the total CPE isolates and meropenem in nearly 3/4 of the OXA-48 isolates. Almost 3/4 of our patients received effective definitive antimicrobial treatment within three days of blood culture positivity. Monotherapy was more often administered in patients with OXA-48-BSI, whereas combination therapy was more frequently administered in patients with KPC and MBL-BSI. This reflects the fact that preferred agents with reliable activity were more often available in OXA-48 cases. We did not identify an association between mortality at any time-point and class of carbapenemase, showing that extent of resistance per se is not associated with mortality in our cohort. The overall 30-day CFR was 38% and only severe sepsis or septic shock was associated with 30-day mortality.

The patient characteristics were similar to the ones reported in other studies (9-13). Severe sepsis or septic shock occurred in 59% of cases (14) (15). Although, in our KPC group, severe sepsis was much higher than in other KPC-BSI series (11). The average Charlson comorbidity score in this cohort was higher than the one reported elsewhere (14, 16). Urinary source and a urinary device in place were more frequently found in OXA-48-BSI compared to KPC and MBL (17). We had anticipated a greater proportion of patients having a urinary source, but notably 17% of our cohort had no source clearly identifiable on detailed review of their notes, and some of these are likely to have been urinary.

Klebsiella spp. was the most prevalent organism in patients with KPC, followed by OXA-48 and MBL-BSI; this is similar to other series (15). Antimicrobial susceptibility results of CPE showed moderate susceptibility to aminoglycosides, specifically for amikacin (64%). Although this is similar to the susceptibilities found in a previous study (18), it gives cause for concern because aminoglycosides are often depended upon to provide reliable Gram-negative bacteriocidal activity.Meropenem was susceptible in vitro in 73% of the OXA-48 clinical isolates. Accordingly, meropenem was the backbone of a great part of the combination treatment regimens.

Almost three-quarters of our patients received effective definitive antimicrobial treatment within three days of blood culture positivity. This compares favorably with othe reports, where fewer than half of the patients were initiated on early effective antibiotics (11).Almost half of the patients who received effective definitive treatment were administered monotherapy and the other half received combination therapy with at least one antibiotic being active in vitro. Monotherapy was more often administered in patients with OXA-48-BSI, whereas combination therapy was more frequently administered in patients with KPC and MBL-BSI. This finding may be related to the fact that OXA-48 blood culture isolates had higher percentages of antimicrobial susceptibility compared to KPC and MBL. This offers a greater choice of preferred antimicrobial options agents with more reliable activity in vivo.

The overall 30-day CFR of 38% was very similar to that reported in 2017 (14). However, mortality for KPC-BSI was lower compared to another series (11). According to the results of the multivariate analysis, the major predictor of 30-day mortality was severe sepsis or septic shock, as seen previously (14, 19). While the univariate analysis showed blood culture isolates that were in vitro susceptible to meropenem were associated with a lower mortality, this association was not shown apparent in the multivariate analysis. Earlier microbiological studies have shown a correlation between carbapenem minimum inhibitory concentration and clinical outcome(20) and this is biologically plausible given the frequent use of carbapenems in this setting. Other factors associated with adverse outcomes identified in other studies include co-morbidities and/or underlying disease (21), treatment monotherapy (9, 11, 19) non-urinary source (17) and delayed effective antimicrobial therapy (14, 22). The relatively small cohort in our study may have contributed to these not being identified as significant in our multivariate analysis. Neither combination treatment nor different treatment regimens showed any impact on mortality in this study (11, 23). Although dose and toxicity were not explored in the present study, new less toxic antibiotics are becoming available with optimized dosing regimens, which may lead to better outcomes than current options.

The main limitations of this study include its retrospective, observational nature. The sample size of each group was relatively small. The multivariate analysis may be affected by the low number of cases. Data on new antimicrobial treatment options such as ceftazidime-avibactam, ceftolozane-tazobactam and cefiderocol are scarce, as this study captured data from 2011 when such treatment options were not available in all centres. Antibiotic sensitivity testing was undertaken in several laboratories and over the 10-year period of the study changes in EUCAST guidance means there is variation in the testing methods and classification of sensitivity.

The case-mix in this series reflects the hospitals’ catchment areas, with a diverse population. Several of the hospitals also offer specialist haemato-oncology services. Whilst rates of CPE infection are highest in London, rates are expected to rise nationally with increased antibiotic use and the global movement of people.

In conclusion, knowledge of the clinical and microbiological characteristics among the three main CPE classes responsible for BSIs is important for management. The high mortality rate associated with CPE-BSIs underscores the need for improving the diagnosis and management in these patients. No single antimicrobial class had reliable activity against CPE isolates – highlighting the importance of rapid antimicrobial susceptibility testing. This includes prompt testing of second- and third-line agents, ideally performed locally. Further larger studies should be aimed at characterising and comparing the clinical and microbiological profile of BSIs due to different types of carbapenemases to establish optimal treatment choices.

Authors’ contributions

VAV and TP led the study, including design, analysis and write-up of the manuscript. SF, JE, DS, KD, LH, SP, SM-J and BD collected clinical and microbiological data. MA, MM and MD reviewed the manuscript. All authors approved the final manuscript.

Competing interests

The authors declare no competing interests

Declarations of interests: none

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Acknowledgements

We would like to thank each of the medical microbiology departments who participated in the study at the following sites: St George's University Hospitals NHS Foundation Trust, King's College Hospital NHS Foundation Trust, University College London Hospitals NHS Foundation Trust, Barts Health NHS Trust, Epsom and St. Helier University Hospitals NHS Trust, and Lewisham and Greenwich NHS Trust.

Gutiérrez-Gutiérrez B, Pérez-Galera S, Salamanca E, de Cueto M, Calbo E, Almirante B, et al. A Multinational, Preregistered Cohort Study of β-Lactam/β-Lactamase Inhibitor Combinations for Treatment of Bloodstream Infections Due to Extended-Spectrum-β-Lactamase-Producing Enterobacteriaceae. Antimicrobial Agents and Chemotherapy. 2016;60(7):4159-69.
Tamma PD, Goodman KE, Harris AD, Tekle T, Roberts A, Taiwo A, et al. Comparing the Outcomes of Patients With Carbapenemase-Producing and Non-Carbapenemase-Producing Carbapenem-Resistant Enterobacteriaceae Bacteremia. Clin Infect Dis. 2017;64(3):257-64.
Agency UHS. English surveillance programme for antimicrobial utilisation and resistance (ESPAUR). Report 2020 to 2021. 2021.
Tamma PD AS, Bonomo RA, Mathers AJ, van Duin D, Clancy CJ. IDSA Guidance on the Treatment of Antimicrobial-Resistant Gram-Negative Infections: Version 1.0. 2022.
Paul M, Carrara E, Retamar P, Tängdén T, Bitterman R, Bonomo RA, et al. European Society of Clinical Microbiology and Infectious Diseases (ESCMID) guidelines for the treatment of infections caused by multidrug-resistant Gram-negative bacilli (endorsed by European society of intensive care medicine). Clin Microbiol Infect. 2022;28(4):521-47.
Falcone M, Tiseo G, Antonelli A, Giordano C, Di Pilato V, Bertolucci P, et al. Clinical Features and Outcomes of Bloodstream Infections Caused by New Delhi Metallo-β-Lactamase-Producing Enterobacterales During a Regional Outbreak. Open forum infectious diseases. 2020;7(2):ofaa011.
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. Journal of Chronic Diseases. 1987;40(5):373-83.
Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. American journal of infection control. 2008;36(5):309-32.
Qureshi ZA, Paterson DL, Potoski BA, Kilayko MC, Sandovsky G, Sordillo E, et al. Treatment outcome of bacteremia due to KPC-producing Klebsiella pneumoniae: superiority of combination antimicrobial regimens. Antimicrob Agents Chemother. 2012;56(4):2108-13.
Gasink LB, Edelstein PH, Lautenbach E, Synnestvedt M, Fishman NO. Risk factors and clinical impact of Klebsiella pneumoniae carbapenemase-producing K. pneumoniae. Infect Control Hosp Epidemiol. 2009;30(12):1180-5.
Zarkotou O, Pournaras S, Tselioti P, Dragoumanos V, Pitiriga V, Ranellou K, et al. Predictors of mortality in patients with bloodstream infections caused by KPC-producing Klebsiella pneumoniae and impact of appropriate antimicrobial treatment. Clin Microbiol Infect. 2011;17(12):1798-803.
Wang Q, Zhang Y, Yao X, Xian H, Liu Y, Li H, et al. Risk factors and clinical outcomes for carbapenem-resistant Enterobacteriaceae nosocomial infections. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 2016;35(10):1679-89.
Souli M, Galani I, Antoniadou A, Papadomichelakis E, Poulakou G, Panagea T, et al. An outbreak of infection due to beta-Lactamase Klebsiella pneumoniae Carbapenemase 2-producing K. pneumoniae in a Greek University Hospital: molecular characterization, epidemiology, and outcomes. Clin Infect Dis. 2010;50(3):364-73.
Gutiérrez-Gutiérrez B, Salamanca E, de Cueto M, Hsueh PR, Viale P, Paño-Pardo JR, et al. Effect of appropriate combination therapy on mortality of patients with bloodstream infections due to carbapenemase-producing Enterobacteriaceae (INCREMENT): a retrospective cohort study. The Lancet Infectious diseases. 2017;17(7):726-34.
Jorgensen SCJ, Trinh TD, Zasowski EJ, Lagnf AM, Bhatia S, Melvin SM, et al. Evaluation of the INCREMENT-CPE, Pitt Bacteremia and qPitt Scores in Patients with Carbapenem-Resistant Enterobacteriaceae Infections Treated with Ceftazidime-Avibactam. Infectious diseases and therapy. 2020;9(2):291-304.
Balkan, II, Aygun G, Aydin S, Mutcali SI, Kara Z, Kuskucu M, et al. Blood stream infections due to OXA-48-like carbapenemase-producing Enterobacteriaceae: treatment and survival. International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases. 2014;26:51-6.
Corbella L, Fernández-Ruiz M, Ruiz-Ruigómez M, Rodríguez-Goncer I, Silva JT, Hernández-Jiménez P, et al. Prognostic factors of OXA-48 carbapenemase-producing Klebsiella pneumoniae infection in a tertiary-care Spanish hospital: A retrospective single-center cohort study. International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases. 2022;119:59-68.
Park JW, Lee H, Park SY, Kim TH. Epidemiological, clinical, and microbiological characteristics of carbapenemase-producing Enterobacteriaceae bloodstream infection in the Republic of Korea. Antimicrobial resistance and infection control. 2019;8:48.
Daikos GL, Tsaousi S, Tzouvelekis LS, Anyfantis I, Psichogiou M, Argyropoulou A, et al. Carbapenemase-producing Klebsiella pneumoniae bloodstream infections: lowering mortality by antibiotic combination schemes and the role of carbapenems. Antimicrob Agents Chemother. 2014;58(4):2322-8.
Patel TS, Nagel JL. Clinical outcomes of Enterobacteriaceae infections stratified by carbapenem MICs. J Clin Microbiol. 2015;53(1):201-5.
Tumbarello M, Viale P, Viscoli C, Trecarichi EM, Tumietto F, Marchese A, et al. Predictors of mortality in bloodstream infections caused by Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: importance of combination therapy. Clin Infect Dis. 2012;55(7):943-50.
Falcone M, Bassetti M, Tiseo G, Giordano C, Nencini E, Russo A, et al. Time to appropriate antibiotic therapy is a predictor of outcome in patients with bloodstream infection caused by KPC-producing Klebsiella pneumoniae. Critical Care. 2020;24(1):29-.
Gomez-Simmonds A, Nelson B, Eiras DP, Loo A, Jenkins SG, Whittier S, et al. Combination Regimens for Treatment of Carbapenem-Resistant Klebsiella pneumoniae Bloodstream Infections. Antimicrob Agents Chemother. 2016;60(6):3601-7.

Table 1. Baseline patient characteristics.

Total

N=84

OXA-48

N=37

KPC

N=12

MBL

N=35

p-value

Age (years); median (range)

64 (18-95)

67 (26-95)

50 (18-86)

61 (25-95)

0.074

Male sex

53 (63%)

21 (57%)

9 (75%)

23 (66%)

0.489

Nosocomial acquisition

Known prior colonisation

59 (70%)

25 (30%)

22 (60%)

7 (19%)

10 (83%)

6 (50%)

27 (77%)

12 (36%)

0.150

0.093

Charlson score; median (range)

Chronic kidney disease

Chronic liver disease

Diabetes mellitus

Haematology malignancy

Organ transplantation

5 (0-14)

18 (21%)

20 (24%)

26 (31%)

29 (35%)

5 (6%)

5 (0-14)

6 (17%)

10 (29%)

15 (43%)

12 (34%)

2 (6%)

5 (2-12)

4 (36%)

5 (46%)

2 (18%)

5 (0-12)

8 (24%)

5 (15%)

9 (27%)

14 (42%)

1 (3%)

0.958

0.448

0.134

0.209

0.348

0.220

Severe sepsis or septic shock

49 (59%)

19 (53%)

10 (83%)

20 (57%)

0.173

Enterobacterales

Klebsiella spp.^a

Escherichia coli

Enterobacter spp.^b

Serratia marcescens

Proteus mirabillis

51 (61%)

20 (24%)

11 (13%)

1 (1%)

23 (63%)

12 (32%)

1 (3%)

0 (0%)

11 (92%)

1 (8%)

0 (0%)

17 (49%)

7 (20%)

10 (29%)

0 (0%)

1 (3%)

0.029

0.189

0.001

0.535

0.502

Source

Urine

Endovascular

Intra-abdominal

Pneumonia

Skin and soft tissues

Unknown

21 (25%)

18 (21%)

20 (24%)

8 (10%)

3 (4%)

14 (17%)

15 (41%)

7 (19%)

10 (27%)

2 (5%)

1 (3%)

2 (5%)

1 (8%)

3 (25%)

2 (17%)

1 (8%)

3 (25%)

5 (14%)

8 (23%)

4 (11%)

1 (3%)

9 (26%)

0.012

0.877

0.760

0.462

0.639

0.049

Invasive devices

Type of invasive devices

Urinary device

Central line

Intra-abdominal catheter

Prosthetic joint

Mechanical ventilation

66 (79%)

17 (26%)

41 (62%)

5 (8%)

2 (3%)

1 (2%)

27 (73%)

12 (32%)

11 (30%)

1 (3%)

2 (5%)

1 (3%)

11 (92%)

1 (8%)

10 (92%)

0 (0%)

28 (80%)

4 (11%)

20 (57%)

4 (11%)

0 (0%)

0.403

0.046

0.002

0.194

0.280

^aK. pneumoniae (n = 45), K. oxytoca (n = 2), K. aerogenes (n = 1). ^b E. cloacae (n = 9), E. hormaechei (n = 2)

Table 2. Antimicrobial and clinical outcomes.

Antimicrobial and clinical outcomes

Total

N=84

OXA-48

N=37

KPC

N=12

MBL

N=35

p-value

Effective empirical antibiotic

16 (19%)

8 (23%)

1 (9%)

6 (19%)

0.473

Time to effective treatment; median (range)

Early effective treatment ≤ 3 days

Duration of treatment; median (range)

1 (0-6)

56 (73%)

9 (0-30)

1 (0-6)

29 (83%)

7 (0-30)

2 (0-4)

7 (64%)

8.5 (1-14)

2 (0-6)

20 (65%)

10 (1-17)

0.090

0.085

0.949

Effective definitive antibiotic

Monotherapy

Meropenem

Aminoglycosides

Ciprofloxacin

Ceftazidime/avibactam

Colistin

72 (86%)

32 (44%)

8 (25%)

5 (16%)

16 (50%)

4 (13%)

1 (3%)

36 (97%)

24 (67%)

8 (33%)

3 (13%)

10 (42%)

3 (13%)

0 (0%)

9 (75%)

2 (22%)

0 (0%)

1 (50%)

27 (77%)

6 (22%)

2 (33%)

6 (22%)

0 (0%)

0.022

<0.001

0.041

0.675

0.185

0.163

0.022

Combination therapy* Meropenem Ciprofloxacin Fosfomycin Tigecycline Aminoglycosides Colistin Ceftazidime/avibactam Cefiderocol	40 (56%) 15 (38%) 3 (8%) 2 (5%) 11 (28%) 18 (45%) 22 (55%) 7 (18%) 1 (3%)	12 (33%) 6 (50%) 0 (0%) 0 (0%) 3 (25%) 7 (58%) 6 (50%) 2 (17%) 0 (0%)	7 (78%) 2 (29%) 0 (0%) 0 (0%) 2 (29%) 2 (29%) 5 (71%) 3 (43%) 0 (0%)	21 (78%) 7 (33%) 3 (14%) 2 (10%) 6 (29%) 9 (43%) 11 (52%) 2 (10%) 1 (5%)	0.004 0.425 0.226 0.386 0.939 0.569 0.569 0.131 0.633
Admitted to ICU (post-bacteraemia episode)	37 (44%)	15 (41%)	8 (67%)	14 (40%)	0.137
Length of stay (days); median (range)	15 (0-241)	15 (0-241)	11 (0-117)	15 (1-76)	0.235
7-day mortality	18 (21%)	8 (22%)	4 (33%)	6 (17%)	0.498
30-day mortality	32 (38%)	13 (35%)	6 (50%)	13 (37%)	0.663
90-day mortality	37 (44%)	14 (38%)	7 (58%)	16 (46%)	0.440

*Antibiotics listed under combination treatment were part of the combined treatment. The most common combination treatments were meropenem plus aminoglycosides (10/40, 25%), tigecycline plus colistin (8/40; 20%), meropenem plus colistin (4/40; 10%) and ceftazidime/avibactam plus colistin (5/40; 13%).

Table 3. Univariate and multivariate analysis for 30-day Case fatality rate of patients with BSI due to CPE.

Age >=60 years (vs <60)

Male (vs female)

Nosocomial acquisition

Known prior colonisation

Source (urine vs others)

Invasive devices

Severe sepsis or septic shock

Charlson score >2

CKD (vs no)

Liver disease (vs no)

Onco-haematological malignancy

Organ transplant (vs no)

E. coli (vs others)

Klebsiella spp. (vs others)

Enterobacter spp. (vs others)

OXA-48 (vs others)

MBL (vs others)

KPC (vs others)

Meropenem susceptible

Effective empirical antibiotic

Effective definitive antibiotic

Early effective antibiotic (≤72h)

Combination therapy (vs mono)

Meropenem-based therapy

Ciprofloxacin-based therapy

Fosfomycin-based therapy

Tigecycline-based therapy

Aminoglycoside-based therapy

Colistin-based therapy

Ceftazidime/avibactam-based

Early source control (≤72h)

Survived

N=52

26 (50%)

30 (58%)

32 (62%)

13 (25%)

16 (31%)

38 (73%)

24 (46%)

40 (77%)

11 (21%)

10 (19%)

21 (40%)

4 (8%)

14 (27%)

29 (56%)

7 (13%)

24 (46%)

21 (40%)

6 (12%)

24 (46%)

10 (19%)

46 (88%)

34 (65%)

24 (46%)

7 (13%)

15 (29%)

1 (2%)

8 (15%)

12 (23%)

14 (27%)

8 (15%)

29 (56%)

Died

N=32

18 (56%)

22 (69%)

26 (81%)

12 (38%)

5 (16%)

25 (78%)

30 (94%)

7 (22%)

10 (32%)

8 (25%)

1 (3%)

5 (16%)

22 (69%)

4 (13%)

13 (41%)

6 (19%)

7 (22%)

6 (19%)

26 (96%)

24 (92%)

16 (59%)

9 (53%)

4 (15%)

1(4%)

3 (11%)

11 (41%)

9 (33%)

3 (11%)

11 (38%)

Univariate

OR (95% CI), P-value

1.24 (0.51-3.00), 0.640

1.54 (0.61-3.91), 0.364

2.57 (0.89-7.38), 0.079

1.75 (0.68-4.55), 0.248

0.41 (0.13-1.25), 0.115

1.22 (0.43-3.49), 0.708

3.87 (1.42-10.57), 0.008

4.13 (0.85-20.01), 0.079

1.02 (0.35-2.97), 0.974

1.86 (0.67-5.16), 0.231

0.48 (0.18-1.26), 0.136

0.39 (0.04-3.55), 0.395

0.49 (0.16-1.52), 0.217

1.67 (0.66-4.23), 0.281

0.90 (0.24-3.35), 0.873

0.77 (0.32-1.89), 0.566

0.98 (0.40-2.40), 0.960

1.73 (0.51-5.92), 0.382

0.32 (0.12-0.88), 0.024

0.96 (0.31-2.97), 0.943

1.69 (0.17-17.15), 0.655

0.41 (0.14-1.22), 0.100

1.52 (0.59-3.92), 0.391

1.72 (0.64-4.61), 0.282

0.39 (0.12-1.34), 0.136

1.85 (0.11-30.74), 0.669

0.64 (0.16-2.65), 0.539

2.12 (0.76-5.80), 0.143

1.25 (0.45-3.44), 0.666

0.64 (0.16-2.65), 0.539

0.40 (0.16- 1.03), 0.058

Multivariate

OR (95% CI), P-value

2.60 (0.70-9.67), 0.154

3.81 (1.19-12.14), 0.024

3.94 (0.69-22.51), 0.124

0.36 (0.11-1.21), 0.098

0.28 (0.07-1.09), 0.096

0.41 (0.14-1.22), 0.110

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Clinical, microbiological characteristics and predictors of mortality in patients with carbapenemase-producing Enterobacterales bloodstream infections: a multicentre study

Status:

Version 1

Abstract

Objectives

Methods

Results

Conclusion

Figures

Introduction

Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Status:

Version 1