Empirical antibiotics of non-carbapenems for ESBL-producing E. coli and K. pneumoniae bacteremia in children : a retrospective medical record review

doi:10.21203/rs.3.rs-1618099/v1

Download PDF

Research Article

Empirical antibiotics of non-carbapenems for ESBL-producing E. coli and K. pneumoniae bacteremia in children : a retrospective medical record review

https://doi.org/10.21203/rs.3.rs-1618099/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: The efficacy of non-carbapenems as an empirical antibiotic for extended-spectrum β-lactamases (ESBL)-producing Escherichia coli and Klebsiella pneumoniae bacteremia in children remains controversial. We compared clinical and microbial outcomes according to the types of empirical antibiotics for treating pediatric patients with ESBL-producing E. coli and K. pneumoniae bacteremia.

Methods: Data from pediatric patients aged ≤18 years who were hospitalized with monomicrobial ESBL-producing E. coli or K. pneumoniae bacteremia between January 2014 and May 2019 were analyzed retrospectively. The impact of empirical therapy was assessed as 30-day all-cause mortality and 2-day microbiological outcomes evaluated by sterility of blood cultures collected on day 2 after empirical antibiotic administration. Logistic regression analysis was used to control for the effects of confounding variables.

Results: A total of 53 patients with bacteremia caused by ESBL-producing E. coli (n=29) and K. pneumoniae (n=24) were included in this study; the median age was 3.6 years, and all had underlying comorbidities. As empirical antibiotics, 27 patients were treated with meropenem, and non-carbapenem agents were administered to 26 patients; 84.6% (22/26) were converted to carbapenem antibiotics as the definitive antibiotic by day 2 after empirical antibiotic administration. Overall, the 30-day all-cause mortality of ESBL-producing E. coli and K. pneumoniae bacteremia was 17.0% (9/53). After adjustment, there was no statistically significant association of use of a non-carbapenem agent as an empirical antibiotic with microbiological failure at day 2 and 30-day all-cause mortality [adjusted odds ratio (OR), 1.0; 95% confidence interval (CI), 0.22–4.88, and adjusted OR, 0.1; 95% Cl, 0.01–1.56].

Conclusions: The empirical use of non-carbapenems might not be a risk factor for mortality and early microbiological outcomes in pediatric patients with ESBL-producing E. coli and K. pneumoniae BSI if early transition to appropriate antimicrobial therapy was possible.

ESBL producers

bacteremia

empirical antibiotics

carbapenem

children

Enterobacteriaceae are important opportunistic pathogens and common microflora, but can cause severe diseases including meningitis, pneumonia, pyelonephritis, intra-abdominal infection, and septicemia(1). Antibiotic resistant has been increased all over the world that is considered a public health threat, and the emergence of multidrug-resistant (MDR) bacterial pathogens of various origins including humans and animals, which have increased the need for screening of MDR strains to select appropriate antibiotics(2). According to estimates of the global number of infections caused by antibiotic-resistant Escherichia coli and Klebsiella pneumoniae in 2014 was high: 50·1 million serious third-generation cephalosporin-resistant infections and 3·1 million serious carbapenem-resistant infections(3). One of the main mechanisms for resistance to 3rd generation cephalosporins (CSs) is the production of β-lactamases including extended-spectrum β-lactamases (ESBLs), which are plasmid-mediated β-lactamases that can hydrolyze most β-lactams such as penicillin, aztreonam, and extended-spectrum CSs (4).

Bloodstream infections (BSIs) caused by extended-spectrum β-lactam-resistant E. coli and K. pneumoniae are increasing public health concerns worldwide with significant morbidity and mortality in children (5–7). A meta-analysis of pediatric BSIs from 1996 to 2013 showed the prevalence of ESBL-producing Enterobacteriaceae accounted for 9% of BSIs, with an annual increase of 3.2%, and a higher prevalence of 11% was observed in neonates (8). However, the prevalence of ESBL producers among E. coli and K. pneumoniae has already reported to be high as 50% at a tertiary children’s hospital in Korea (9). BSIs caused by ESBL-producing Enterobacteriaceae are associated with poor outcomes including increased mortality and long hospital stays in adults and children (5–7). In particular, in neonates infected with ESBL-producing Enterobacteriaceae, the all-cause mortality is reported to be as high as 36% (8, 10–12).

Carbapenem antibiotics have been recommended as the drugs of choice for serious infections such as BSIs or pneumonia caused by ESBL-producing Enterobacteriaceae. However, along with the increased incidence of ESBL-producing Gram-negative infections worldwide, increased use of carbapenems has exerted selection pressure on carbapenem resistance and the spread of carbapenemases (13). As a carbapenem-sparing option for infections caused by ESBL producers, β-lactam/β-lactamase inhibitors (BLBLIs) can be considered as an alternative option. Although ESBLs are defined as β-lactamases that are inhibited by clavulanate in vitro, there are concerns that in vitro susceptibility to β-lactamase inhibitors might not translate into clinical efficacy, especially in high inoculum infections (13). Furthermore, findings from a recent randomized clinical trial including hospitalized adult patients with ceftriaxone-resistant E. coli or K. pneumoniae BSIs did not support the use of piperacillin-tazobactam as a definitive treatment (14).

However, whether the use of carbapenem antibiotics as an empirical regimen is associated with better outcomes than the use of non-carbapenem antibiotics such as BLBLIs remains controversial (15). Few studies have examined the efficacy of BLBLIs as an empirical regimen against BSIs caused by ESBL producers (15–17). Furthermore, relevant data on empirical regimens for ESBL producers in children are lacking.

In this study, we aimed to compare the clinical and microbiological outcomes of pediatric patients with bacteremia caused by ESBL-producing E. coli and K. pneumoniae who were given carbapenem and non-carbapenem antibiotics as empirical regimens. The impact of initial empirical therapy on clinical and microbiological outcomes was assessed as 30-day all-cause mortality and the sterility of blood cultures collected on day 2 after initial empirical antibiotic administration, respectively.

Study population

This is a single-center, retrospective, observational study comparing the outcomes of patients with ESBL-producing E. coli or K. pneumoniae bacteremia who were treated with different empirical antibiotics. Pediatric patients aged ≤ 18 years who were hospitalized with monomicrobial ESBL-producing E. coli or K. pneumoniae bacteremia at Asan Medical Center Children’s Hospital between January 2014 to May 2019 were included. Clinical and demographic data were abstracted from electronic medical records including the clinical diagnosis, underlying medical conditions, antibiotics administered, prior intensive care unit (ICU) state, presentation with septic shock, need for a higher level of respiratory support, and all-cause 30-day mortality. Only the first isolate was included during a single clinical episode occurring within a 4-week period, and duplicates from the same patient were excluded. The exclusion criteria were as follows: 1) polymicrobial BSI, 2) BSI caused by E. coli or K. pneumoniae without an ESBL phenotype, 3) BSI caused by carbapenem-resistant isolates, and 4) patients who did not receive a carbapenem as the definitive therapy even though ESBL production was confirmed.

The primary focus of BSIs were classified as follows; central-line associated BSIs (laboratory-confirmed BSI in which an eligible BSI organism was identified and an eligible central line present on the date of event or the day before) (18), urinary tract infection (UTI), intra-abdominal infection, pneumonia, or none. Our institute performs cerebrospinal fluid (CSF) culture as well as blood culture in infants younger than 3 months of age who are suspected of having clinical sepsis before initiating empirical antibiotics. However, if the feasibility of a lumbar puncture is controversial, such as in extremely low birth weight infants or patients with unstable vital signs, empirical antibiotics may be administered immediately after obtaining blood cultures without performing a lumbar puncture; in the case of infants aged < 3 months with BSIs caused by ESBL-producing E. coli or K. pneumoniae, the dosage and duration of meropenem as a definitive therapy is adjusted on the premise that meningitis is present.

Definitions

Initial presentation with severe infection was defined as the need for a higher level of respiratory support or presentation with shock. Shock was defined as the need to use new inotropic or to increase the dose of an inotropic to maintain blood pressure. The need for a higher level of respiratory support was defined as whether respiratory support at a higher stage was required at least 3 days before the blood culture test was performed and was classified as none (room air), low-flow system (nasal cannula, simple mask), high-flow system (high-flow nasal cannula, continuous positive airway pressure), and mechanical ventilation. Empirical therapy was defined as appropriate if started within 24 hours of initial blood culture collection and if the isolate was in vitro susceptible. Definitive antibiotics referred to the antibiotic therapy that was administered subsequent to receipt of antibiotic susceptibility test results. Microbiological failure at day 2 was defined as failure of sterility of blood cultures collected on or before calendar day 2 after empirical antibiotic administration.

Antibiotics and outcomes

Each empirical and definitive antibiotic was administered according to the ‘Pediatric and Neonatal Dosage Handbook’(19), and each dose was infused over 30 minutes. Although carbapenems are recommended as definitive antibiotics for invasive infections caused by ESBL-producing Enterobacteriaceae, some patients diagnosed with UTI +/- bacteremia who show improvement in clinical symptoms and bacterial sterilization may maintain the initial non-carbapenem antibiotic as the definitive therapy. Among carbapenems, meropenem was used when meningitis or meningitis could not be excluded.

Clinical outcome of patients with ESBL-producing E. coli or K. pneumoniae BSI was assessed as 30-day all-cause mortality. The microbiological outcome at 2 days was evaluated by the sterility of blood cultures collected on day 2 after empirical antibiotic administration.

Isolation, identification, and antimicrobial susceptibility test of E. coli and K. pneumoniae

Aseptically collected blood was inoculated into BACTEC PEDS PLUS/F and BACTEC LYTIC Anaerobic/F Bactec Plus (BACTEC 9240, BD, Sparks, MD, USA). Species identification and antimicrobial susceptibilities were performed using the MicroScan WalkAway 96 plus system and Neg Combo Panel Type 72 (Beckman Coulter, Brea, CA, USA). ESBL producers were determined according to the Clinical and Laboratory Standards Institute M100 (20), which measure the minimum inhibitory concentration around both cefotaxime and ceftazidime disks with or without clavulanate for E. coli, K. pneumoniae, K. oxytoca, and Proteus mirabilis.

Statistical analysis

To compare the baseline characteristics between the two empirical treatment groups, the Mann-Whitney U test or Chi square test was used. In identifying the independent risk factors for 30-day all-cause mortality and microbiological outcome at day 2, logistic regression analysis was used to control for the effects of confounding variables. P values < 0.05 were considered statistically significant. Statistical analyses were performed using SPSS version 20.0 (IBM Software, Chicago, IL, USA).

Ethical approval and consent to participate

This study was approved by the Institutional Review Board of Asan Medical Center with a waiver of informed consent due to its retrospective, de-identified data collection and analysis (IRB No. 2019 − 0617). All methods were carried out in accordance with relevant guidelines and regulations.

Demographic and clinical characteristics

During the study period over 5 consecutive years, a total of 197 isolates of E. coli (n = 113) and K. pneumoniae (n = 84) were detected from blood cultures at Asan Medical Center Children`s Hospital (Fig. 1). A total of 134 isolates were excluded from this analysis for the following reasons: isolates obtained from the patients aged ≥ 19 years, non-ESBL producers, repetitive culture positivity, or multiple bacteria from a blood culture obtained simultaneously. In addition, 10 isolates were also excluded because 5 isolates were obtained from patients with urosepsis who showed improvement in clinical symptoms and bacterial sterilization while maintained on the initial non-carbapenems as the definitive therapy, 4 were carbapenem-resistant organisms, and one patient did not have a follow-up blood culture.

Finally, a total of 53 (26.9%) patients with BSIs caused by ESBL-producing E. coli (n = 29) and K. pneumoniae (n = 24) were included in this study. Baseline demographic and clinical data according to empirical therapy are presented in Table 1. The median age was 3.6 years (range, 0–18.8 years), and all of the patients had underlying comorbidities such as hemato-oncologic diseases (n = 26) and prematurity with a gestational age < 37 weeks (n = 11). The most common source of bacteremia was a central vascular catheter (n = 38), followed by intra-abdominal infection (n = 6), and UTI (n = 4). Overall, 39.6% (21/53) of BSIs occurred in ICU patients, and 32.1% (17/53) of patients presented with severe infections at the onset of bacteremia including septic shock (26.4%; 14/53) and/or the requirement for higher respiratory support (22.6%; 12/53).

Table 1

Baseline Characteristics of the Study Population.
Characteristics	Empirical antibiotics			P-value
Characteristics	Non-carbapenem (N = 26)	Carbapenem (N = 27)	Total (N = 53)	P-value
Median age, years (range)	3.2 (0- 18.7)	3.8 (0- 18.8)	3.6 (0 -18.8)	0.60
Sex (M:F)	17:9	14:13	31:22	0.32
Strain (E. coli: K. pneumoniae)	17:9	12:15	29:24	0.13
Presence of underlying diseases	26 (100%)	27 (100%)	53 (100%)
Hemato-oncologic disease	13 (50%)	13 (48.1%)	26 (49.1%)	0.90
Preterm infant (GA < 37 wks)	3 (11.5%)	8 (29.6%)	11 (20.8%)	0.11
Congenital heart disease	3 (11.5%)	4 (14.8%)	7 (13.2%)	0.73
Vesicoureteral reflux	2 (7.7%)	0 (0%)	2 (3.8%)	0.14
Others	5† (19.2%)	2^‡ (7.4%)	7 (13.2%)	0.20
Primary focus of bacteremia
None	2 (7.7%)	0 (0%)	2 (3.8%)	0.14
CLABSI	16 (61.5%)	22 (81.5%)	38 (71.7%)	0.11
Intra-abdominal infection	1 (3.8%)	5 (18.5%)	6 (11.3%)	0.09
NEC	1 (3.8%)	4 (14.8%)	5 (9.4%)	0.17
Peritonitis	0 (0%)	1 (3.7%)	1 (1.9%)	0.32
Urinary tract infection	4 (15.4%)	0 (0%)	4 (7.5%)	0.03
Pneumonia	2 (7.7%)	0 (0%)	1 (1.9%)	0.30
Meningitis	1 (3.8%)	0 (0%)	1 (1.9%)	0.30
ICU stay prior to infection	8 (30.8%)	13 (48.1%)	21 (39.6%)	0.20
Initial presentation with severe infection	4 (15.4%)	13 (48.1%)	17 (32.1%)	0.01
Presentation with shock	3 (11.5%)	11 (40.7%)	14 (26.4%)	0.02
Need for higher respiratory support	3 (11.5%)	9 (33.3%)	12 (22.6%)	0.06

Abbreviations: blood stream infection (BSI), central-line-associated blood stream infection (CLABSI), gestational age (GA), intensive care unit (ICU), necrotizing enterocolitis (NEC)

^† includes ornithine transcarbamylase deficiency (n=1), Alagille syndrome (n=1), bronchopulmonary dysplasia (n=1), chronic kidney disease (n=1), and meningomyelocele (n=1).

^‡ includes DiGeorge syndrome (n=1) and liver transplantation state (n=1).

Baseline characteristics according to initial empirical antibiotics

Overall, the empirical antibiotic groups were balanced with respect to the baseline characteristics, although more patients in the non-carbapenem group than in the carbapenem group had vesicoureteral reflux and a diagnosis of UTI (p=0.14 and p=0.03, respectively) (Table 1). More patients in the carbapenem group than in the non-carbapenem group had a history of prematurity and were diagnosed with intra-abdominal infection (p=0.11 and p=0.09, respectively). The incidence of initial presentation with septic shock was significantly higher in the carbapenem group than in the non-carbapenem group (41% vs 12%; p=0.02).

Empirical antibiotics and outcomes

As empirical antibiotics, 27 patients were treated with carbapenems, all of which were meropenem, and non-carbapenem agents were administered to 26 patients. Non-carbapenem antibiotics included cefepime (n=14), cefotaxime (n=4), piperacillin/tazobactam (n=6), and amikacin (n=2), all of which were applied with or without glycopeptides at the discretion of the physicians (Table 2). In the non-carbapenem group, only 23.1% (6/26) were appropriate empirical antibiotics, and 84.6% (22/26) were converted to a carbapenem as the definitive antibiotic by day 2 after empirical antibiotic administration; another 2 cases were converted by day 3, and one was converted by day 4. One patient who experienced fatality in the non-carbapenem group was initially administered cefepime that was in vitro resistant and died within one day before switching to carbapenems. Among 3 patients who were maintained non-carbapenem antibiotics for 3 or 4 days after initial blood culture collection, microbiological failure at day 2 occurred in 2 patients; one was treated with an in vitro susceptible empirical antibiotic of piperacillin/tazobactam for 4 days, and the other was treated with an in vitro resistant cefepime for 3 days; however, there were no fatal cases.

Table 2

Types and duration of administration of non-carbapenems as empirical antibiotic
Non-carbapenem antibiotics	Number of cases (total n = 26)
Types
Amikacin	2 (7.7%)
Cefepime	14 (53.8%)
Cefotaxime	4 (15.4%)
Piperacillin/Tazobactam	6 (23.1%)
Duration of administration (calendar days)
1	16 (61.5%)
2	7 (26.9%)
3	2 (7.7%)
4	1 (3.8%)

Overall, the 30-day all-cause mortality of ESBL-producing E. coli and K. pneumoniae bacteremia was 17.0% (9/53); the 30-day all-cause mortality was 29.6% (8/27) in those receiving a carbapenem antibiotic and 3.8% (1/26) in those receiving a non-carbapenem antibiotic (Table 3). Microbiological failure by day 2 occurred in 7 of 26 patients (26.9%) in the non-carbapenem group compared with 11 of 27 (40.7%) patients in the carbapenem group (p=0.29).

Table 3

Clinical Outcomes of the Study Population.
Outcomes	Empirical antibiotics			P-value
Outcomes	Non-carbapenem^† (N = 26)	Carbapenem (N = 27)	Total (N = 53)	P-value
30-day all-cause mortality	1 (3.8%)	8 (29.6%)	9 (17.0%)	0.01
Microbiological failure at day 2	7 (26.9%)	11 (40.7%)	18 (34.0%)	0.29
^† In the non-carbapenem group, of in vitro susceptible appropriate empirical antibiotics (n = 6), there were no fatal cases and 3 cases with microbiological failure at day 2. In the non-carbapenem group, of in vitro resistant inappropriate empirical antibiotics (n = 20), there was one fatal case and 4 cases with microbiological failure at day 2.

Impact of empirical antibiotics on microbiological and clinical outcomes

In the univariate analysis, statistically significant risk factors associated with microbiological failure at day 2 were being a preterm infant [odds ratio (OR), 42.5; 95% confidence interval (CI), 4.73–381.72], having an intra-abdominal infection (OR, 13.1; 95% CI, 1.39–122.85), and having a severe clinical presentation (OR, 5.0; 95% CI, 1.44–17.37) (Table 4). After adjusting for confounding factors including empirical use of non-carbapenems and variables with p-value < 0.05 in the univariate analysis, only being a preterm infant was associated with 2-day microbiological failure (adjusted OR, 30.6; 95% CI, 2.43–386.19). However, there was no statistically significant association of use of a non-carbapenem agent as an empirical antibiotic with microbiological failure at day 2.

Table 4

Factors Associated with Persistent Bacteremia by Day 2 and 30-day All-cause Mortality.
Variables †	Microbiological failure at day 2		30-day all-cause mortality
Variables †	OR (95% CI)	Adjusted OR^‡ (95% CI)	OR (95% CI)	Adjusted OR^‡ (95% CI)
Empirical use of non-carbapenems	0.5 (0.17–1.71)	1.0 (0.22–4.88)	0.1 (0.01–0.83)	0.1 (0.01–1.56)
Preterm infant	42.5 (4.73–381.72)	30.6 (2.43–386.19)	15.6 (2.94–82.84)	7.6 (1.04–56.07)
Intra-abdominal infection	13.1 (1.39–122.85)	1.8 (0.08–42.72)	2.9 (0.44–18.68)	N.A.
Urinary tract infection	0.6 (0.06–6.31)	N.A.	N.A.	N.A.
Initial presentation with clinically severe infection	5.0 (1.44–17.37)	1.2 (0.18–8.03)	11.9 (2.13–66.62)	3.2 (0.40–25.25)
Abbreviations: confidence interval (CI), Not applicable (N.A.), Odds ratio (OR)
^† Univariate analysis was performed using the variables of initial empirical antibiotics, preterm infants, clinical diagnosis and clinical severity at presentation with p-values < 0.1 in Table 1. However, the factor ‘urinary tract infection’ as an infection focus was not included in 30-day all-cause mortality because none of the cases were fatal.
^‡ The multivariate analysis was performed using the following variables: empirical use of non-carbapenems and those with p-value < 0.05 in the univariate analysis.

With respect to the 30-day all-cause mortality in patients with BSIs caused by ESBL-producing E. coli and K. pneumoniae, being a preterm infant was the only independent risk factor (adjusted OR, 7.6; 95% CI, 1.04–56.07) after adjustment for receiving empirical therapy with non-carbapenem, being a preterm infant, and having a severe clinical presentation.

Our retrospective analysis of pediatric BSIs caused by ESBL-producing E. coli and K. pneumoniae with a 30-day mortality of 17% suggested that administration of non-carbapenem antibiotics as an initial empirical antibiotic was not a significant risk factor for 30-day mortality or microbiological failure at day 2 if early transition to a definitive carbapenem regimen was possible when susceptibility was proven. Being a preterm infant was the only significant risk factor for 30-day all-cause mortality of BSIs caused by ESBL-producing E. coli and K. pneumoniae in this study.

Based on the data from our institute from 2009 to 2018, the resistance rates of E. coli and K. pneumoniae to 3^rd and/or 4^th generation CSs were as high as 40–50%, and their resistance rates to piperacillin/tazobactam were approximately £20% (9, 21). If the antimicrobial resistance rate to a certain antibiotic in a hospital or community is higher than 10–20%, clinicians should consider using a different antibiotic or combination therapy as an empirical regimen (22-24). Considering the high ESBL-positive rate of Enterobacteriaceae at our institution, if the strain is reported as a Gram-negative strain in the blood culture reporting system, the treatment is first changed to a carbapenem; then, de-escalation is performed if necessary. In cases of septic shock, carbapenems tend to be used as empirical antibiotics. In suspected CNS infections, meropenem is the empirical choice of antibiotic to achieve CNS penetration because the usual doses of tazobactam have been considered to be insufficient for appropriate treatment of CNS infections (25), although fewConsidering the high ESBL-positive rate of Gram-negative strains at this medical institution, if the G(-) strain is reported to the blood culture notification system, the antibiotic susceptibility result of the strain is reported and the treatment is changed to carbapenem. In the present medical gram Considering that the high prevalence of ESBL negative strains blood culture system reported in the G (-) when the reported isolates and treatment to become a cover antibiotic susceptibility results of isolates reported changes penem. Although studies have reported the successful treatment of patients with CNS infections with piperacillin/tazobactam (26). So, carbapenems could be easily used as the initial regimen, especially in high-risk patients. However, considering the positive correlation between antibiotic burden and antibiotic resistance (27-29), administration of broad-spectrum antibiotics should be considered carefully.

Carbapenems have been considered the best choice for definitive therapy in serious infections caused by ESBL producers (30). However, there is controversy regarding appropriate initial empirical antibiotics in both pediatric and adult patients infected with ESBL producers. Inappropriate initial empirical antibiotic treatment has been associated with increased mortality in patients with bacteremia caused by ESBL-producing Enterobacteriaceae (31, 32). In contrast, one multicenter study including approximately 10% of pediatric patients aged <18 years suggested that inappropriate empirical therapy within <24 hours was not associated with increased mortality (33). Our single center study suggested that there is no significant advantage in choosing a carbapenem as an initial antimicrobial agent in terms of the microbiological failure rate by day 2 and 30-day mortality. However, this could not be applied in general. The microbiological laboratory of our institute operates the blood culture system 24 hours a day, year-round; when an automatic signal is reported from the BACTEC blood culture system, laboratory personnel perform Gram staining and report the results to an electronic medical record in real time. Therefore, in the case of Gram-negative bacteria isolated from a blood culture, it usually takes less than 24 hours to report the automatic positive signal with the Gram-staining result for a corresponding one. Therefore, most patients with BSIs caused by ESBL-producing Enterobacteriaceae treated with non-carbapenems as initial empirical antibiotics can be switched to a carbapenem as the definitive therapy within two calendar days, if necessary; furthermore, more than half of them can be switched within one calendar day. These conditions might explain why non-carbapenems as empirical antibiotics were not associated with an increased mortality rate or microbiological failure rate in our study.

A recent case-control study including 110 pediatric patients showed that neonates were more likely than older children to develop bacteremia caused by ESBL producers (6). Moreover, one study about risk factors for infection with ESBL-producing Enterobacteriaceae during an outbreak in a neonatal ICU (NICU) found that infections with ESBL producers were associated with prematurity (34). During our study period, an outbreak of ESBL-producing K. pneumoniae bacteremia among extremely low birth weight infants in the NICU resulted in four fatal cases associated with necrotizing enterocolitis (NEC) that were all initially treated with carbapenems and died within 2 days. Furthermore, two more fatal cases were observed in preterm infants as sporadic cases. Considering the high prevalence of bacteremia caused by ESBL-producing Enterobacteriaceae and high fatality rates regardless of types of empirical antibiotics in neonates (8, 35-38), delicate care and appropriate management for premature infants infected with ESBL producers are necessary to minimize morbidity and mortality.

There are limitations in assessing the causal relationship of some phenomena because this was a retrospective observational study. The more severe the initial symptoms and signs were, the more likely the physician was to administer carbapenems from the beginning, especially for premature infants with septic shock. Most of the fatal cases in this study were associated with premature infants. Even though we performed a multivariate logistic analysis to adjust for possible confounding factors, the sample size is just too small to draw definite conclusions about the effect of empirical antibiotic on microbiological failure or mortality while taking into account other risk factors. Furthermore, it was not possible to analyze the statistical difference between the 30-day mortality according to the appropriateness of empirical antibiotics used in the non-carbapenem group. However, no fatal cases were observed among patients who received appropriate empirical non-carbapenem antibiotics. In addition, we did not identify virulent factors, antibiotic resistance mechanism including specific ESBL genes, and host immune responses, which could be an important way to explain the outcomes of the ESBL producing G(-) bacteremic patients (39), because our study was a retrospective study, and PCR test for ESBL genes was not routinely performed at our institute. Although our result cannot be extrapolated to the general pediatric population, these findings suggest that it is not always necessary to select a carbapenem as an initial empirical antibiotic for ESBL-producing E. coli and K. pneumoniae bacteremia in children if the appropriate antibiotics can be administered without delay with the support of a culture reporting system. Laboratory-based antimicrobial resistance surveillance systems can incorporate epidemiological data in determining appropriate empirical antibiotics in high endemic setting while balancing between selecting appropriate empirical antibiotics and minimizing antibiotic pressure.

In conclusion, if early transition to carbapenems as a definitive antimicrobial therapy is possible, empirical use of non-carbapenems may be considered an appropriate option in pediatric bacteremia caused by ESBL-producing E. coli and K. pneumoniae. A large multicenter study should be conducted to identify appropriate carbapenem-sparing options for medically fragile pediatric patients and to minimize the spread of MDR pathogens such as carbapenem-resistant bacteria.

Ethical approval and consent to participate

All methods in this study were carried out in accordance with relevant guidelines and regulations. This study was approved by the Institutional Review Board of Asan Medical Center with a waiver of informed consent due to its retrospective, de-identified data collection and analysis (IRB No. 2019-0617).

Consent for publication

Not applicable.

Availability of data and materials

All data generated or analysed during this study are included in this published article and its supplementary information files.

Competing interests

The authors declare that they have no competing interests.

Funding

Not applicable.

Authors' contributions

SR as first writing author analyzed and interpreted the patient data and clinical outcomes. HJ checked the grammar and helped with data collection. MN provided relevant data and microbiological advice. JN, corresponding author, supervised and reviewed all aspects of the overall writing, including statistical analysis, conclusion and discussion. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect. 2012;18(3):268–81.
Makharita RR, El-Kholy I, Hetta HF, Abdelaziz MH, Hagagy FI, Ahmed AA, et al. Antibiogram and genetic characterization of carbapenem-resistant gram-negative pathogens incriminated in healthcare-associated infections. Infection and drug resistance. 2020;13:3991.
Temkin E, Fallach N, Almagor J, Gladstone BP, Tacconelli E, Carmeli Y, et al. Estimating the number of infections caused by antibiotic-resistant Escherichia coli and Klebsiella pneumoniae in 2014: a modelling study. The Lancet Global Health. 2018;6(9):e969-e79.
Jacoby GA. EXTENDED-SPECTRUM β-LACTAMASES AND OTHER ENZYMES PROVIDING RESISTANCE TO OXYIMINO-β-LACTAMS. Infectious Disease Clinics of North America. 1997;11(4):875–87.
Blomberg B, Jureen R, Manji KP, Tamim BS, Mwakagile DSM, Urassa WK, et al. High Rate of Fatal Cases of Pediatric Septicemia Caused by Gram-Negative Bacteria with Extended-Spectrum Beta-Lactamases in Dar es Salaam, Tanzania. Journal of Clinical Microbiology. 2005;43(2):745–9.
Ndir A, Diop A, Faye PM, Cissé MF, Ndoye B, Astagneau P. Epidemiology and Burden of Bloodstream Infections Caused by Extended-Spectrum Beta-Lactamase Producing Enterobacteriaceae in a Pediatric Hospital in Senegal. PLOS ONE. 2016;11(2):e0143729.
Kim Y-K, Pai H, Lee H-J, Park S-E, Choi E-H, Kim J, et al. Bloodstream infections by extended-spectrum β-lactamase-producing Escherichia coli and Klebsiella pneumoniae in children: epidemiology and clinical outcome. Antimicrobial agents and chemotherapy. 2002;46(5):1481–91.
Flokas ME, Karanika S, Alevizakos M, Mylonakis E. Prevalence of ESBL-Producing Enterobacteriaceae in Pediatric Bloodstream Infections: A Systematic Review and Meta-Analysis. PloS one. 2017;12(1):e0171216-e.
Kim S, Yoo R, Lee J. The Impact of the Antibiotic Burden on the Selection of its Resistance among Gram Negative Bacteria Isolated from Children. Pediatr Infect Vaccine. 2015;22(3):178–85.
Ballot DE, Nana T, Sriruttan C, Cooper PA. Bacterial bloodstream infections in neonates in a developing country. ISRN Pediatr. 2012;2012:508512.
Kayange N, Kamugisha E, Mwizamholya DL, Jeremiah S, Mshana SE. Predictors of positive blood culture and deaths among neonates with suspected neonatal sepsis in a tertiary hospital, Mwanza-Tanzania. BMC Pediatr. 2010;10:39.
Rao YK, Midha T, Garg A, Garg J, Dwivedi G, Singh N, et al. Neonatal septicemia in north india due to extended spectrum beta lactamase (ESBL) producing gram negative bacteria.
Harris PN, Tambyah PA, Paterson DL. β-lactam and β-lactamase inhibitor combinations in the treatment of extended-spectrum β-lactamase producing Enterobacteriaceae: time for a reappraisal in the era of few antibiotic options? Lancet Infect Dis. 2015;15(4):475–85.
Harris PNA, Tambyah PA, Lye DC, Mo Y, Lee TH, Yilmaz M, et al. Effect of Piperacillin-Tazobactam vs Meropenem on 30-Day Mortality for Patients With E coli or Klebsiella pneumoniae Bloodstream Infection and Ceftriaxone Resistance: A Randomized Clinical Trial. Jama. 2018;320(10):984–94.
Peralta G, Lamelo M, Alvarez-García P, Velasco M, Delgado A, Horcajada JP, et al. Impact of empirical treatment in extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella spp. bacteremia. A multicentric cohort study. BMC infectious diseases. 2012;12:245-.
Chaubey VP, Pitout JD, Dalton B, Ross T, Church DL, Gregson DB, et al. Clinical outcome of empiric antimicrobial therapy of bacteremia due to extended-spectrum beta-lactamase producing Escherichia coli and Klebsiella pneumoniae. BMC Res Notes. 2010;3:116.
Rodríguez-Baño J, Navarro MD, Retamar P, Picón E, Pascual Á, the Extended-Spectrum Beta-Lactamases–Red Española de Investigación en Patología Infecciosa/Grupo de Estudio de Infección Hospitalaria G. β-Lactam/β-Lactam Inhibitor Combinations for the Treatment of Bacteremia Due to Extended-Spectrum β-Lactamase–Producing Escherichia coli: A Post Hoc Analysis of Prospective Cohorts. Clinical Infectious Diseases. 2012;54(2):167–74.
NHSN. NHSN Patient safeyy component Manual 2019; criteria of IDSA 2009. NHSN. 2019.
Taketomo CK, Hodding JH, Kraus D. Pediatric and neonatal dosage handbook: Hudson (OH): Lexi Comp; 2010.
CLSI. Performance standards for Antimicrobial Susceptibility testing. 29th ed.CLSI supplement M100.Wayne, PA: Clinical and Laboratory Standards Institute; 2019.
Park S, Lee E, So HJ, Yoo RN, Lee J. The Impact of Antibiotic Burden on the Selective Resistance of Gram Negative Bacteria in Children. Pediatric Infection and Vaccine. 2021;28(2):82–91.
Beardsley JR, Williamson JC, Johnson JW, Ohl CA, Karchmer TB, Bowton DL. Using Local Microbiologic Data To Develop Institution-Specific Guidelines for the Treatment of Hospital-Acquired Pneumonia. Chest. 2006;130(3):787–93.
Fridkin SK, Edwards JR, Tenover FC, Gaynes RP, McGowan JE, Jr. Antimicrobial Resistance Prevalence Rates in Hospital Antibiograms Reflect Prevalence Rates among Pathogens Associated with Hospital-Acquired Infections. Clinical Infectious Diseases. 2001;33(3):324–9.
Kalil AC, Metersky ML, Klompas M, Muscedere J, Sweeney DA, Palmer LB, et al. Management of Adults With Hospital-acquired and Ventilator-associated Pneumonia: 2016 Clinical Practice Guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clinical Infectious Diseases. 2016;63(5):e61-e111.
Nau R, Kinzig-Schippers M, Sörgel F, Schinschke S, Rössing R, Müller C, et al. Kinetics of piperacillin and tazobactam in ventricular cerebrospinal fluid of hydrocephalic patients. Antimicrobial agents and chemotherapy. 1997;41(5):987–91.
Javad A, Vahid P, Kazem G, Alireza S, Gholamali G, Qulamreza K. Piperacillin/tazobactam in treatment of brain abscess. Scandinavian journal of infectious diseases. 2006;38(3):224–6.
Willemsen I, Bogaers-Hofman D, Winters M, Kluytmans J. Correlation between antibiotic use and resistance in a hospital: Temporary and ward-specific observations. Infection. 2009;37(5):432.
Livermore DM, James D, Reacher M, Graham C, Nichols T, Stephens P, et al. Trends in fluoroquinolone (ciprofloxacin) resistance in enterobacteriaceae from bacteremias, England and Wales, 1990–1999. Emerg Infect Dis. 2002;8(5):473–8.
Ryu S, Klein EY, Chun BC. Temporal association between antibiotic use and resistance in Klebsiella pneumoniae at a tertiary care hospital. Antimicrobial Resistance & Infection Control. 2018;7(1):83.
Morrissey I, Hackel M, Badal R, Bouchillon S, Hawser S, Biedenbach D. A Review of Ten Years of the Study for Monitoring Antimicrobial Resistance Trends (SMART) from 2002 to 2011. Pharmaceuticals (Basel, Switzerland). 2013;6(11):1335–46.
Kang C-I, Kim S-H, Park WB, Lee K-D, Kim H-B, Kim E-C, et al. Bloodstream infections caused by antibiotic-resistant gram-negative bacilli: risk factors for mortality and impact of inappropriate initial antimicrobial therapy on outcome. Antimicrobial agents and chemotherapy. 2005;49(2):760–6.
Tumbarello M, Sanguinetti M, Montuori E, Trecarichi EM, Posteraro B, Fiori B, et al. Predictors of mortality in patients with bloodstream infections caused by extended-spectrum-β-lactamase-producing Enterobacteriaceae: importance of inadequate initial antimicrobial treatment. Antimicrobial agents and chemotherapy. 2007;51(6):1987–94.
Frakking FNJ, Rottier WC, Dorigo-Zetsma JW, van Hattem JM, van Hees BC, Kluytmans JAJW, et al. Appropriateness of empirical treatment and outcome in bacteremia caused by extended-spectrum-β-lactamase-producing bacteria. Antimicrobial agents and chemotherapy. 2013;57(7):3092–9.
Linkin DR, Fishman NO, Patel JB, Merrill JD, Lautenbach E. Risk factors for extended-spectrum beta-lactamase-producing Enterobacteriaceae in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2004;25(9):781–3.
Jain A, Roy I, Gupta MK, Kumar M, Agarwal SK. Prevalence of extended-spectrum beta-lactamase-producing Gram-negative bacteria in septicaemic neonates in a tertiary care hospital. J Med Microbiol. 2003;52(Pt 5):421–5.
Vijayakanthi N, Bahl D, Kaur N, Maria A, Dubey NK. Frequency and characteristics of infections caused by extended-spectrum beta-lactamase-producing organisms in neonates: a prospective cohort study. Biomed Res Int. 2013;2013:756209.
Sehgal R, Gaind R, Chellani H, Agarwal P. Extended-spectrum beta lactamase-producing gram-negative bacteria: clinical profile and outcome in a neonatal intensive care unit. Ann Trop Paediatr. 2007;27(1):45–54.
Stapleton PJ, Murphy M, McCallion N, Brennan M, Cunney R, Drew RJ. Outbreaks of extended spectrum beta-lactamase-producing Enterobacteriaceae in neonatal intensive care units: a systematic review. Arch Dis Child Fetal Neonatal Ed. 2016;101(1):F72-8.
Wang J, Peng C, Li H, Zhang P, Liu X. The impact of microplastic-microbe interactions on animal health and biogeochemical cycles: A mini-review. Science of The Total Environment. 2021:145697.

No competing interests reported.

rawdata.xlsx

Download PDF

Editorial decision: Major revision
06 Jun, 2022
Reviews received at journal
02 Jun, 2022
Reviewers agreed at journal
22 May, 2022
Reviewers invited by journal
20 May, 2022
Editor assigned by journal
20 May, 2022
Editor invited by journal
20 May, 2022
Submission checks completed at journal
20 May, 2022
First submitted to journal
02 May, 2022

You are reading this latest preprint version

Empirical antibiotics of non-carbapenems for ESBL-producing E. coli and K. pneumoniae bacteremia in children : a retrospective medical record review

Status:

Version 1

Abstract

Figures

Introduction

Materials And Method

Study population

Definitions

Antibiotics and outcomes

Statistical analysis

Results

Demographic and clinical characteristics

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1