Abstract
Background
The impact of rectal colonization with carbapenemase-producing Enterobacterales (CPE) on subsequent infection and mortality in febrile neutropenia (FN) remains unclear, as does the role of surveillance cultures in guiding empirical therapy. This study assessed the role of colonization and/or infection by OXA-48-producing Enterobacterales (OXA-48-E) on 15-day mortality in hematologic patients with FN.
Method
This single-center retrospective cohort study included hematologic patients receiving chemotherapy, hospitalized for FN, and screened for CPE rectal colonization at admission and weekly. Univariate and multivariate analyses were performed to identify predictors of OXA-48-E infection and 15-day mortality. FN episodes were the analysis unit.
Results
Eighty-seven patients experienced 170 FN episodes (48 in OXA-48-E carriers, 122 in noncarriers). OXA-48-E infection occurred in 15/48 (31%) carrier episodes and in none of the noncarriers, yielding a 100% negative predictive value for recent negative rectal swabs. Other OXA-48-E infection predictors included acute leukemia (aOR 6.71; 95% CI: 1.62–27.77; P = .009), recent carbapenem exposure (aOR 3.37; 95% CI: 1.07-10.61; P = .038) and sepsis/septic shock (aOR 4.26; 95% CI: 1.25–14.48; P = .020).
OXA-48-E infection episodes showed lower rates of active antibiotic therapy within 24 hours (7/15 [47%] vs 128/145 [88%]; P < .001) and higher 15-day mortality (4/15 [27%] vs 5/145 [3%]; P = .005). Independent predictors of mortality were sepsis/septic shock (aOR 18.16; 95% CI: 3.24–101.85; P < .001) and no active antibiotic therapy within 24 hours (aOR 21.68; 95% CI: 2.08–225.63; P = .010).
Conclusions
Rectal colonization strongly predicts OXA-48-E infection in hematologic patients with FN. Furthermore, lack of active antibiotic therapy within the first 24 hours of FN was independently associated with increased mortality.
Keywords: febrile neutropenia, hematologic patients, mortality, OXA-48-producing Enterobacterales, rectal colonization
Rectal colonization strongly predicts OXA-48-producing Enterobacterales infection in hematologic patients with febrile neutropenia. Delayed active therapy was independently associated with higher mortality. Empirical OXA-48 coverage may be considered among carriers with severe presentation, with early de-escalation if infection is excluded.
INTRODUCTION
Patients with hematologic malignancies undergoing cytotoxic chemotherapy frequently develop prolonged neutropenia, increasing their susceptibility to severe infections. Among these, multidrug-resistant Gram-negative bacilli (MDR-GNB) infections pose a growing clinical challenge [1, 2]. Of particular concern is the rising prevalence of carbapenem-resistant Enterobacterales (CRE), which, in many regions, is mainly driven by the production of carbapenemase enzymes [3]. Class D carbapenemases—particularly OXA-48 and OXA-48-like subgroups—are among the most prevalent worldwide [4], with notably high rates reported in Spain and other Mediterranean countries [5–8].
In patients with chemotherapy-induced mucositis, primary bacteremia is a frequent complication during febrile neutropenia (FN). Thus, prior rectal colonization with MDR-GNB may play an important role in the development of subsequent infections by the same microorganisms. However, the true incidence of infections due to the colonizing strain during postchemotherapy FN, and its potential association with mortality, is not well established. Only a few observational studies have addressed this question, mostly in patients colonized with extended-spectrum β-lactamase (ESBL)-producing Enterobacterales, but methodological heterogeneity and conflicting findings limit their applicability [9–19].
Whether surveillance cultures should be used to guide empirical antibiotic therapy in hematologic patients with FN is still a matter of debate [20]. This knowledge gap is clinically relevant, as standard empirical regimens for FN recommended by international guidelines, such as cefepime, piperacillin/tazobactam, or meropenem [21], are not active against many carbapenemase-producing Enterobacterales (CPE). Indeed, the management of empirical antibiotic treatment in patients colonized with resistant bacteria was a topic of discussion at the 10th European Conference on Infections in Leukaemia (ECIL) [22].
If CPE colonization is ultimately shown to significantly increase the risk of subsequent infection, this information should be incorporated into empirical treatment decision-making. In this context, developing antimicrobial stewardship (AMS) strategies will be essential to balancing optimal therapy with the prudent use of newly available antibiotics. Moreover, since not all carbapenemases may exhibit the same clinical behavior, resistance mechanism-stratified studies are needed to define the most appropriate clinical management for hematologic patients colonized with specific types of CPE.
Considering that OXA-48 is the most prevalent carbapenemase in our region, this study aimed to identify predictive factors of OXA-48-producing Enterobacterales (OXA-48-E) infection in hematologic patients with chemotherapy-induced FN, including prior rectal colonization as a potential independent risk factor. Additionally, it sought to assess 15-day mortality following the FN episode and associated predictors.
METHODS
Design and Patients
This single-center retrospective cohort study included all patients treated at Valme University Hospital between October 2022 and September 2025 who fulfilled the following criteria: (1) adults with hematologic malignancies receiving chemotherapy, (2) development of FN requiring hospitalization, and (3) rectal swab surveillance upon admission and weekly thereafter. Two patients colonized with non-OXA-48 CPE were excluded.
Surveillance Colonization Cultures and Diagnostic Procedures
Since September 2022, routine epidemiological surveillance cultures have been performed for all patients admitted to the Hematology ward, as part of standard clinical practice. Those with an initial negative rectal swab underwent weekly repeat swabs throughout their hospital stay.
Rectal Surveillance Specimens
Rectal surveillance swabs were collected using FecalSwab™ (Copan, Italy), which contains a modified Cary–Blair medium specifically designed for the recovery, preservation and detection of enteric pathogens. Samples were directly inoculated onto chromogenic selective media for the detection of ESBL-producing Enterobacterales (chromID® ESBL Agar, bioMérieux, France) and CPE (chromID® CARBA, Becton Dickinson, USA). Blood agar plates (Becton Dickinson, USA) were used as growth controls.
When growth was observed on carbapenemase-selective plates, isolates were screened using a colorimetric assay (β-CARBA™ test, Bio-Rad, USA). Species identification was subsequently performed by matrix-assisted laser desorption/ionization time-Of-flight mass spectrometry (MALDI-TOF MS; Bruker, Germany). If the screening test was positive, first-line characterization of the carbapenemase type was carried out by immunochromatography, using the NG-Test® CARBA 5 (Labymed S.A.) assay, which detects KPC, NDM, VIM, IMP, and OXA-48-like enzymes. Subsequent molecular confirmation was performed using the Xpert® Carba-R assay (GenXpert System, Cepheid, USA), a multiplex real-time polymerase chain reaction test targeting the most prevalent carbapenemase genes.
Antimicrobial susceptibility testing of recovered isolates was performed using the MicroScan WalkAway Plus system (Beckman Coulter, USA), and results were interpreted according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) clinical breakpoints.
Data on the colonizing OXA-48-E isolates and their antimicrobial susceptibility profiles are provided in Supplementary Table 1.
Clinical Specimens
Clinical specimens were processed according to standard microbiological procedures for organism identification and antimicrobial susceptibility testing. Specifically, in the case of blood culture specimens, samples were processed using the BD BACTEC™ FX automated blood culture system (Becton Dickinson, USA). Upon positivity, bottles were subcultured onto blood agar plates or Schaedler agar (Becton Dickinson, USA) for anaerobic cultures, and a direct Gram stain was performed.
Species identification was performed by MALDI-TOF (Bruker, Germany). Antimicrobial susceptibility testing was carried out using the VITEK® 2 automated system (bioMérieux, France), and results were interpreted according to the EUCAST clinical breakpoints.
When Gram-negative bacilli were detected in blood specimens, rapid phenotypic screening for β-lactam resistance mechanisms was performed directly from the positive blood culture using the β-LACTA™ Test (Bio-Rad, USA).
For other clinical specimens, phenotypic screening for ESBL, AmpC, and carbapenemase production was performed using the double-disk synergy test with clavulanic acid, cloxacillin-based synergy testing, and combination disk testing using boronic acid, cloxacillin, dipicolinic acid, and temocillin, respectively, according to the EUCAST guidelines.
Positive results prompted further evaluation for carbapenemase production from the recovered isolate following the same procedures described for rectal surveillance specimens.
Definitions
FN was defined as absolute neutrophil count < 500/mm3 with either a single body temperature ≥ 38.3°C, a sustained ≥ 38.0°C for ≥1 hour, or 2 readings ≥ 38.0°C within 12 hours [23]. Distinct episodes of FN in the same patient during the same hospitalization were considered as independent events, provided there was no evidence of an uncontrolled infectious focus and the patient remained afebrile for at least 96 hours between episodes, without antibiotics during this interval. OXA-48-E colonization status was defined for each episode using the most recent surveillance result available before or at FN onset. For patients whose colonization status changed over follow-up, FN episodes prior to becoming colonized were classified as noncolonized, whereas subsequent episodes were classified as colonized.
Severe FN was defined as presentation with sepsis or septic shock, according to classic [24] and Sepsis-3 criteria [25]. The Giannella Risk Score (GRS) was calculated based on its original description [26]. Microbiologically documented infection was defined as FN accompanied by positive blood cultures and/or microbiological cultures obtained aseptically from clinically suspected sites of infection, excluding cases of contamination. The source of infection was determined from clinical and laboratory data using classic US Centers for Disease Control and Prevention criteria [27]. Endogenous source of bloodstream infection (BSI) was considered in neutropenic patients with absent or mild gastrointestinal symptoms, in whom gut translocation was suspected.
Active antibiotic therapy encompassed the administration of at least 1 active drug against the isolated pathogen, based on susceptibility testing. Monotherapy with meropenem was considered nonactive therapy for OXA-48-E, in line with current therapeutic guidelines for infections caused by these CPE [28, 29].
Data Collection
FN episodes were identified retrospectively through electronic clinical records from the Hematology Unit, based on diagnoses coded as FN at admission or discharge. Additionally, the Department of Microbiology's database was reviewed to ascertain whose patients were colonized with MDR-GNB.
Collected variables included demographics, comorbidities, and history of hospitalization; prior exposure to antibiotics, corticosteroids, and/or chemotherapy in the 3 months before the FN episode; and MDR-GNB colonization or infection in the prior 6 months. All variables related to the FN episode were also recorded, including the suspected infection site, clinical severity, minimum neutrophil count, and duration of neutropenia, as well as antimicrobial treatment administered and microbiological findings.
Patient Management
As a retrospective study, patient management was determined by the Hematology Unit at the discretion of the attending medical team and in coordination with the infectious diseases consultant. Fluoroquinolone prophylaxis was prescribed by the treating hematologist in accordance with the guidelines in force [30].
Rectal surveillance swabs were processed in the Microbiology laboratory on the day of collection. New cases of rectal colonization with CPE were typically reported within 24–48 hours and communicated via daily multidisciplinary meeting with Infectious Diseases and Preventive Medicine. In parallel, the treating hematologist was notified, and an electronic alert was entered in the patient's medical record.
Empirical antibiotic regimens for high-risk FN in hematologic patients proposed by our local antimicrobial guidelines are available in the Supplementary Data. In confirmed OXA-48-E infection, targeted therapy recommended was ceftazidime/avibactam, based on the susceptibility profile of the predominant circulating clone in our center. However, final treatment decisions were made by the attending hematologist.
Statistical Analysis
Descriptive analyses encompassed the characterization of the study population, including an assessment of the baseline prevalence of rectal colonization with OXA-48-E. Continuous variables were expressed as medians (quartile 1–quartile 3 [Q1–Q3]) and categorical variables as frequencies (percentages).
The primary outcome variable was the occurrence of microbiologically documented OXA-48-E infection within 15 days of FN onset. Secondary outcome included 15-day mortality following the FN episode.
Univariate and multivariate analyses were performed to identify factors associated with OXA-48-E infection, including baseline colonization as a potential independent risk factor. Continuous variables were compared by means of the Student's t test or the Mann–Whitney U test, depending on the normality tests. The chi-square and the Fisher's tests were used for comparisons between categorical variables. Variables with P ≤ .1 in univariate analyses were entered in logistic regression multivariate analysis. Differences were considered statistically significant for P ≤ .05. aOR and the respective 95% CIs were calculated.
The same statistical analysis approach was performed for 15-day mortality outcome variable. For the primary mortality analyses, FN episodes with a final diagnosis of nonbacterial etiology, such as viral or fungal infections, as well as those due to Clostridioides difficile, were excluded, giving the risk of introducing bias in the assessment of the impact of active empirical antibiotic therapy on mortality.
Analyses were carried out using SPSS statistical software package release 24.0 version (IBM Corporation, Somers, NY, USA) and Stata SE 17.0 version (StataCorp LP, College Station, Texas, USA).
RESULTS
Features of the Study Population
During the study period, 170 FN episodes occurred in 87 hematologic patients. Of these patients, 18 (21%) were colonized with OXA-48-E at the moment of FN (n = 11 Enterobacter cloacae, n = 6 Klebsiella pneumoniae, n = 1 Escherichia coli). Baseline characteristics of these patients are summarized in Table 1. As more than 1 episode of FN might occur in a single patient, there were 48 episodes of FN among the 18 patients with a previous colonization, whereas 122 episodes occurred in the 69 noncolonized patients. Among patients with more than 1 FN episode, the median time from 1 episode to the next was 46 (28–74) days. Five patients were noncolonized at their first FN episode and became colonized later during follow-up, whereas only 1 patient was initially colonized and subsequently became noncolonized.
Table 1.
Main Baseline Characteristics of the Study Population (N = 87)
| Characteristic, No. (%) | All Patients | OXA-48-E-Colonized Patients (n = 18) | OXA-48-E-Noncolonized Patients (n = 69) | P Valuea |
|---|---|---|---|---|
| Age (y)b | 65 (55–72) | 70 (53–71) | 65 (55–73) | .564 |
| Male sex | 64 (74) | 15 (83) | 49 (71) | .378 |
| Diabetes mellitus | 16 (18) | 3 (17) | 13 (19) | .593 |
| Age-adjusted Charlson Comorbidity Indexb | 5 (3–6) | 5 (4–7) | 5 (3–6) | .584 |
| Underlying hematological disease | .059c | |||
| Acute leukemia | 31 (36) | 9 (50) | 22 (32) | |
| Non-Hodgkin lymphoma | 30 (34) | 6 (33) | 24 (35) | |
| Multiple myeloma | 9 (10) | 1 (6) | 8 (12) | |
| Myelodysplastic syndrome | 6 (7) | 0 (0) | 6 (9) | |
| Chronic myeloid leukemia | 2 (2) | 1 (6) | 1 (1) | |
| Others | 9 (10) | 1 (6) | 8 (12) | |
| Hospitalization within the previous 3 m | 32 (37) | 9 (50) | 23 (33) | .192 |
| Antibiotics within the previous 3 m | 43 (49) | 15 (83) | 28 (41) | .001 |
| Corticosteroid treatmentd,e | 35 (40) | 13 (72) | 22 (32) | .002 |
| Induction chemotherapye | 31 (36) | 11 (61) | 20 (29) | .011 |
| Autologous HSCT within the previous 3 m | 10 (11) | 3 (17) | 7 (10) | .561 |
| Current use of fluoroquinolone prophylaxis at the time of FN | 20 (23) | 7 (39) | 13 (19) | .112 |
| Venous catheter placement in the previous monthf | 134 (79) | 46 (96) | 88 (72) | <.001 |
| Bladder catheterization in the previous monthf | 41 (24) | 19 (40) | 22 (18) | .003 |
| Time (d) from 1st positive colonization study to FN episodeb,g | 33 (2–78) | 33 (2–78) | - | - |
| Minimum neutrophil count (× 109/L)b,f | 5 (0–93) | 0 (0–30) | 15 (0–100) | .011 |
| Sepsis or septic shockf | 41 (24) | 11 (23) | 30 (25) | .818 |
These baseline characteristics were determined at the time of each patient's first FN episode, except where otherwise indicated by the superscript numbers.
Abbreviations: FN, febrile neutropenia; HSCT, hematopoietic stem-cell transplantation; OXA-48-E, OXA-48-producing Enterobacterales.
aIn bold P < .05.
bMedian (Q1–Q3).
cMantel–Haenszel test for linear trend.
dCorticosteroid treatment: prednisone at a dose of 10 mg per day or its equivalent for more than 3 wks.
eActive or within the previous 3 m.
fOut of 170 FN episodes, 48 in OXA-48-E-colonized patients and 122 in noncolonized ones.
gOut of 48 FN episodes in OXA-48-E-colonized patients.
Incidence, Etiology, and Management of Documented Infections
Microbiologically documented infections were identified in 87 (51%) cases. The distribution of pathogens by baseline colonization status is shown in Supplementary Table 2.
Most common infection sites included unknown/endogenous source (n = 23, 26%), nonbiliary abdominal (n = 21, 24%), and respiratory tract (n = 17, 20%), followed by catheter-related (n = 12, 14%), urinary (n = 7, 8%), and others (n = 7, 8%). Bacteremia occurred in 63 (72%) out of the 87 documented infections.
Empirical antibiotic treatment for the 48 FN episodes in OXA-48-E-colonized patients is detailed in Table 2. Ceftazidime/avibactam was used empirically in 19 (40%) episodes, while cefiderocol was administered in 2 (4%) cases involving patients colonized with OXA-48-E strains previously known to be resistant to ceftazidime/avibactam. Among 15 confirmed OXA-48-E infections, targeted antibiotic therapy included ceftazidime/avibactam in 14 (93%) cases, while the remaining episode was a septic shock in a patient colonized with a ceftazidime/avibactam-resistant strain that was treated with cefiderocol. The median duration of targeted treatment was 13 (3–14) days. Clinical, microbiological, and management data of these 15 cases are summarized in Supplementary Table 3.
Table 2.
Empirical Antibiotic Treatment Administered for All FN Episodes, Stratified by Baseline Colonization Status (N = 170)
| Antibiotic Regimens, No. (%) | Non-OXA-48-E-colonized Patients (n = 122) | OXA-48-E-Colonized Patients (n = 48) |
|---|---|---|
| Cefepime or piperacillin/tazobactam | 69 (57) | 12 (25) |
| Cefepime or piperacillin/tazobactam plus amikacin | 10 (8) | 6 (12.5) |
| Cefepime or piperacillin/tazobactam plus amikacin → meropenema | 13 (11) | 3 (6) |
| Cefepime or piperacillin/tazobactam plus amikacin → ceftazidime/avibactama | … | 6 (12.5) |
| Meropenem | 9 (7) | 3 (6) |
| Meropenem plus amikacin | 15 (12) | 3 (6) |
| Meropenem plus amikacin → ceftazidime/avibactama | … | 2 (4) |
| Ceftazidime/avibactam → cefepime or piperacillin/tazobactam plus amikacinb | … | 3 (6) |
| Ceftazidime/avibactam → meropenem plus amikacinb | … | 1 (2) |
| Ceftazidime/avibactam | … | 6 (12.5) |
| Ceftazidime/avibactam → cefiderocola | … | 1 (2) |
| Cefiderocol → Meropenem plus amikacinb | … | 1 (2) |
| Others | 6 (5) | 1 (2) |
| Association of a glycopeptide empirically to the initial regimen | 68 (56) | 36 (75) |
Abbreviations: FN, febrile neutropenia; OXA-48-E, OXA-48-producing Enterobacterales.
aEmpirical escalation approach at the discretion of the attending medical team.
bEmpirical de-escalation approach at the discretion of the attending medical team.
Clinical Predictors of OXA-48-Producing Enterobacterales Infection
Fifteen (17%) out of the 87 microbiologically documented infections were due to OXA-48-E, all of which occurred in patients previously colonized (Supplementary Table 2). In 12 (80%) cases, colonization had been identified before the onset of OXA-48-E infection, while in the remaining 3 (20%) cases, colonization and infection were detected simultaneously. Notably, no cases of OXA-48-E infection were observed in patients with either a prior or concomitant negative rectal swab. Thus, a recent negative rectal swab—performed within a median of 5 (2–10) days before the onset of FN—showed a 100% negative predictive value (NPV) for OXA-48-E infection.
Giannella Risk Score did not predict OXA-48-E infection among colonized patients. Thus, among the 48 FN episodes in OXA-48-E-colonized patients, 13 (32.5%) OXA-48-E infections occurred in the 40 cases with GRS < 7, whereas 2 (25%) OXA-48-E infection was confirmed among the 8 cases with GRS ≥7 (P = 1.0).
Table 3 summarizes univariate and multivariate predictors of confirmed OXA-48-E infection. Since no OXA-48-E infections occurred in noncolonized patients, it was not possible to develop a multivariate model including OXA-48-E colonization as a covariate. Instead, a model including baseline characteristics was assessed, assuming unknown colonization status. Thus, after multivariate analysis adjusted for age, sex, corticosteroid treatment, and receipt of fluoroquinolone prophylaxis, the independent predictors of OXA-48-E infection were acute leukemia (AL) (aOR 6.71; 95% CI: 1.62–27.77; P = .009), prior carbapenem use within the last 3 months (aOR 3.37; 95% CI: 1.07–10.61; P = .038) and sepsis or septic shock (aOR 4.26; 95% CI: 1.25–14.48; P = .020).
Table 3.
Univariate and Multivariate Analyses of the Predictors of Confirmed OXA-48-E Infection (N = 170)
| Factor | N/N (%) | P Bivariatea | Adjusted Odds Ratio (95% CI) | P Multivariatea |
|---|---|---|---|---|
| Age (y)b | - | .201 | 1.06 (0.99–1.14) | .093 |
| Sex | ||||
| Women | 4/51 (8) | … | ||
| Men | 11/119 (9) | 1.000 | 0.79 (0.17–3.58) | .756 |
| Underlying acute leukemia | ||||
| No | 3/90 (3) | … | ||
| Yes | 12/80 (15) | .007 | 6.71 (1.62–27.77) | .009 |
| Hospitalization within the previous 3 m | ||||
| No | 4/69 (6) | … | ||
| Yes | 11/101 (11) | .250 | - | - |
| Carbapenem use within the previous 3 m | ||||
| No | 6/122 (5) | … | ||
| Yes | 9/48 (19) | .013 | 3.37 (1.07–10.61) | .038 |
| Corticosteroid treatmentc,d | ||||
| No | 6/110 (5) | … | ||
| Yes | 9/60 (15) | .036 | 3.32 (0.89–12.35) | .073 |
| Induction chemotherapyd | ||||
| No | 8/113 (7) | … | ||
| Yes | 7/57 (12) | .259 | - | - |
| Current use of fluoroquinolone prophylaxise | ||||
| No | 8/125 (6) | … | ||
| Yes | 7/45 (16) | .073 | 2.69 (0.74–9.81) | .135 |
| Baseline OXA-48-E colonization | ||||
| No | 0/122 (0) | … | ||
| Yes | 15/48 (31) | <.001f | - | - |
| Sepsis or septic shock | ||||
| No | 8/129 (6) | … | ||
| Yes | 7/41 (17) | .053 | 4.26 (1.25–14.48) | .020 |
Abbreviations: CI, confidence interval; OXA-48-E, OXA-48-producing Enterobacterales.
aIn bold P < .05.
bConsidered as a continuous variable.
cCorticosteroid treatment: prednisone at a dose of 10 mg per day or its equivalent for more than 3 wks.
dActive or within the previous 3 m.
eAt the time of febrile neutropenia.
fSince no OXA-48-E infections occurred in noncolonized patients, it was not possible to develop a multivariate model including OXA-48-E colonization as a covariate.
Independent Predictors of 15-Day Mortality
For these analyses, we excluded the 10 episodes of documented infections caused by C. difficile and nonbacterial etiologies, yielding 160 evaluable FN episodes. Patients with OXA-48-E infection were less frequently treated with an active antibiotic therapy within the first 24 hours of FN (7/15 [47%] vs 128/145 [88%] in the remaining cohort; P < .001).
Overall, 9 (6%) patients died within the first 15 days: 4 (27%) among those with confirmed OXA-48-E infection and 5 (3%) in the remaining cohort (P = .005). Table 4 summarizes univariate analyses performed to identify risk factors for 15-day mortality.
Table 4.
Univariate Analysis of the Predictors of 15-Day Mortality (N = 160)
| Factor | N/N (%) | P Bivariatea |
|---|---|---|
| Age (y)b | … | .470 |
| Sex | ||
| Women | 3/50 (6) | |
| Men | 6/110 (5) | 1.000 |
| Underlying acute leukemia | ||
| No | 5/82 (6) | |
| Yes | 4/78 (5) | 1.000 |
| Age-adjusted Charlson Comorbidity Indexb | … | .973 |
| Corticosteroid treatmentc,d | ||
| No | 5/103 (5) | |
| Yes | 4/57 (7) | .722 |
| Current use of fluoroquinolone prophylaxis at the time of FN | ||
| No | 6/115 (5) | |
| Yes | 3/45 (7) | .712 |
| Baseline OXA-48-E colonization | ||
| No | 4/116 (3) | |
| Yese | 5/44 (11) | .116 |
| More than 1 episode of FN | ||
| No | 3/82 (4) | |
| Yes | 6/78 (8) | .319 |
| Duration of neutropenia ≥ 7 df,g | ||
| No | 4/109 (4) | |
| Yes | 5/47 (11) | .130 |
| Respiratory focus | ||
| No | 5/135 (4) | |
| Yes | 4/25 (16) | .034 |
| Sepsis or septic shock | ||
| No | 2/122 (2) | |
| Yes | 7/38 (18) | <.001 |
| Confirmed OXA-48-E infection | ||
| No | 5/145 (3) | |
| Yes | 4/15 (27) | .005 |
| Receiving active antibiotic therapy within the first 24 h of FN | ||
| Yes | 1/52 (2) | |
| Undetermined-absence of microbiological isolation | 3/83 (4) | |
| No | 5/25 (20) | .005h |
Abbreviations: FN, febrile neutropenia; OXA-48-E, OXA-48-producing Enterobacterales.
aIn bold P < .05.
bConsidered as a continuous variable.
cCorticosteroid treatment: prednisone at a dose of 10 mg per day or its equivalent for more than 3 wks.
dActive or within the previous 3 m.
eIncluding the 3 cases in which colonization and infection were concomitantly detected.
fDuration (days) of neutropenia prior to the onset of the FN episode.
gAvailable in 156 patients.
hMantel–Haenszel test for linear trend.
After multivariate analysis adjusted for age, respiratory focus, and OXA-48-E infection, the independent predictors of 15-day mortality were sepsis or septic shock (aOR 18.16; 95% CI: 3.24–101.85; P < .001) and not receiving active antibiotic therapy within the first 24 hours of FN (aOR 21.68; 95% CI: 2.08–225.63; P = .010) (Table 5, model A). Similarly, in a model restricted to baseline variables—ie, the first 24 hours after FN diagnosis (including prior colonization with OXA-48-E as a covariate but excluding confirmed OXA-48-E infection)—the same factors were identified as independent predictors of 15-day mortality, whereas OXA-48-E colonization was not (Table 5, model B).
Table 5.
Independent Predictors of 15-Day Mortality (Model A) and Independent Predictors of 15-Day Mortality Restricted to Baseline Variables (Model B)
| Variables | Adjusted Odds Ratio (95% CI) | P Valuea |
|---|---|---|
| Model (A) | ||
| Age (y)b | 1.02 (0.93–1.11) | .731 |
| Respiratory focus | 3.32 (0.52–20.99) | .203 |
| Sepsis or septic shock | 18.16 (3.24–101.85) | <.001 |
| Confirmed OXA-48-E infection | 8.76 (0.83–92.0) | .074 |
| Receiving active antibiotic therapy within the first 24 h of FN | ||
| Yes | Reference category | … |
| Undetermined-absence of microbiological isolation | 5.23 (0.48–57.51) | .176 |
| No | 21.68 (2.08–225.63) | .010 |
| Model (B) | ||
| Age (y)b | 1.01 (0.93–1.10) | .739 |
| Baseline OXA-48-E colonizationc | 3.65 (0.68–19.65) | .131 |
| Respiratory focus | 2.72 (0.46–16.08) | .270 |
| Sepsis or septic shock | 14.57 (2.35–90.47) | .004 |
| Receiving active antibiotic therapy within the first 24 h of FN | ||
| Yes | Reference category | … |
| Undetermined-absence of microbiological isolation | 5.0 (0.43–58.84) | .201 |
| No | 14.80 (1.33–164.88) | .028 |
Abbreviations: FN, febrile neutropenia; OXA-48-E, OXA-48-producing Enterobacterales.
aIn bold P < .05.
bConsidered as a continuous variable.
cIncluding the 3 cases in which colonization and infection were concomitantly detected.
We performed a secondary analysis in the overall set of 170 FN episodes, including those due to confirmed C. difficile, viral, and fungal infections. Ten deaths (6%) occurred within 15 days of FN onset. Table 6 summarizes the univariate and multivariate analyses of 15-day mortality in the overall cohort. In the adjusted multivariate model, respiratory focus (aOR 5.99; 95% CI 1.22–29.36; P = .027), sepsis or septic shock (aOR 11.06; 95% CI 2.09–58.5; P = .005), and OXA-48-E infection (aOR 8.28; 95% CI 1.44–47.77; P = .018) were independently associated with 15-day mortality. Nevertheless, an interaction between confirmed OXA-48-E infection and not receiving active empirical antibiotic therapy within the first 24 hours was observed. After adding the interaction term to the model, OXA-48-E infection alone was no longer independently associated with 15-day mortality.
Table 6.
Univariate and Multivariate Analyses of the Predictors of 15-Day Mortality: Model Including Less Frequent Microbiological Etiologies (N = 170)
| Factor | N/N (%) | P Bivariatea | Adjusted Odds Ratio (95% CI) | P Multivariatea |
|---|---|---|---|---|
| Age (y)b | … | .229 | 1.03 (0.94–1.13) | .498 |
| Sex | ||||
| Women | 3/51 (6) | … | ||
| Men | 7/119 (6) | 1.000 | … | … |
| Underlying acute leukemia | ||||
| No | 6/90 (7) | … | ||
| Yes | 4/80 (5) | .751 | … | … |
| Age-adjusted Charlson Comorbidity Indexb | … | .854 | … | … |
| Corticosteroid treatmentc,d | ||||
| No | 6/110 (5) | … | ||
| Yes | 4/60 (7) | .743 | … | … |
| Current use of fluoroquinolone prophylaxise | ||||
| No | 7/125 (6) | … | ||
| Yes | 3/45 (7) | .725 | … | … |
| Baseline OXA-48-E colonization | ||||
| No | 5/122 (4) | … | ||
| Yesf | 5/48 (10) | .147 | … | … |
| More than 1 episode of FN | ||||
| No | 3/87 (3) | … | ||
| Yes | 7/83 (8) | .204 | … | … |
| Duration of neutropenia ≥ 7 dg,h | ||||
| No | 5/117 (4) | … | ||
| Yes | 5/48 (10) | .157 | … | … |
| Respiratory focus | ||||
| No | 5/139 (4) | … | ||
| Yes | 5/31 (16) | .019 | 5.99 (1.22–29.36) | .027 |
| Sepsis or septic shock | ||||
| No | 2/129 (2) | … | ||
| Yes | 8/41 (20) | <.001 | 11.06 (2.09–58.5) | .005 |
| Confirmed OXA-48-E infection | ||||
| No | 6/155 (4) | … | ||
| Yes | 4/15 (27) | .006 | 8.28 (1.44–47.77) | .018 |
| Receiving an active antibiotic therapyi | ||||
| Yes | 2/54 (4) | Reference category | … | |
| Undetermined-absence of microbiological isolation | 3/83 (4) | 4.54 (0.49–41.63) | .181 | |
| No | 5/33 (15) | .051j | 4.45 (0.60–33.23) | .146 |
Abbreviations: FN, febrile neutropenia; OXA-48-E, OXA-48-producing Enterobacterales.
aIn bold P < .05.
bConsidered as a continuous variable.
cCorticosteroid treatment: prednisone at a dose of 10 mg per day or its equivalent for more than 3 wks.
dActive or within the previous 3 m.
eAt the time of FN.
fIncluding the 3 cases in which colonization and infection were concomitantly detected.
gDuration (days) of neutropenia prior to the onset of the FN episode.
hAvailable in 165 patients.
iWithin the first 24 h of FN.
jMantel–Haenszel test for linear trend.
DISCUSSION
Our study shows that prior rectal colonization is the strongest predictor of OXA-48-E infection in hematologic patients with FN, as in our cohort, all these infections occurred exclusively in previously colonized individuals. Mortality was higher among colonized patients with microbiologically confirmed OXA-48-E infection, which may be partly explained by the reduced likelihood of receiving an active antibiotic early during empirical therapy. These findings should be incorporated in treatment decisions when caring for hematologic patients with FN in settings where OXA-48-E infection might be a concern.
The increasing incidence of CPE infections worldwide has complicated FN management in hematologic patients, with growing interest in the identification of risk factors for CPE infections in this population. In our study, a recent negative surveillance culture had a 100% NPV for excluding OXA-48-E infection in FN patients, which reinforces the importance of conducting surveillance cultures in high-risk units, such as Hematology. Beyond their role in infection control, surveillance cultures could also serve as a valuable tool for guiding initial decisions on empirical antibiotic therapy in FN, as patients with a recent negative surveillance culture can be safely managed with standard empirical antibiotic regimens for FN.
In our cohort, colonization and infection due to OXA-48-E were concomitantly detected in 3 FN episodes, which may limit the utility of prior screening results for guiding empirical therapy in such cases. However, this also highlights the importance of obtaining up-to-date colonization data at FN onset. In high-risk settings, performing a rectal swab at the onset of clinical suspicion—regardless of prior negative results within a routine surveillance program—and, when available, incorporating rapid molecular diagnostic techniques may provide timely information on colonization status and better inform empirical antibiotic selection. Additionally, we have identified that, in the absence of a recent colonization study, patients with underlying AL, recent exposure to carbapenems, and/or a severe clinical presentation were predictors of OXA-48-E infection during FN, which was consistent with previous literature [20].
While the NPV of surveillance cultures to exclude OXA-48-E infection was very high, not all carriers develop a subsequent infection caused by the colonizing pathogen. In our cohort, 31% of colonized patients developed an OXA-48-E infection during FN. Reported infection rates among CPE carriers in the literature vary widely [15–19], ranging from 11% [19] to 30% [17]. These discrepancies likely reflect differences in inclusion criteria, patient profiles, CPE strains, and the time elapsed from colonization. Nevertheless, all these studies consistently confirm that more than half of CPE-colonized patients will not develop a documented invasive infection caused by the colonizing strain.
There is a growing controversy on the selection of empirical antibiotic therapy for FN in hematologic patients colonized with CPE. Recently, the 10th ECIL updated guidelines have recommended empiric use of novel anti-Gram-negative beta-lactams in all colonized patients [22]. Thus, characterizing not only colonization status but also which carbapenemase is involved might contribute to further tailoring of empirical therapy. However, given that a significant proportion of colonized patients will not develop an infection due to resistant bacteria—more than a half in our study—combined with high CPE colonization prevalence in certain regions [32], ECIL recommendations raise concerns from an AMS perspective, as they could contribute to excessive use of novel beta-lactams and resistance development [33].
On the other hand, absence of early active antibiotic therapy was independently associated with increased mortality in our cohort. This finding is consistent with previous studies linking inappropriate initial empirical antibiotic treatment to worse outcomes in neutropenic patients [14, 34]. In our series, ceftazidime/avibactam was the most frequently used targeted therapy, as recommended by international guidelines [28, 29], but was administered empirically in 40% of OXA-48-E-colonized patients. The higher mortality observed in confirmed OXA-48-E infections may have been partly related to suboptimal empirical coverage during the early phase of infection, as suggested by our multivariate analysis. In the mortality subanalysis including all FN episodes, the detrimental effect of not receiving early active therapy may have been attenuated by the inclusion of etiologies for which early appropriateness of empirical therapy is likely less determinant for 15-day mortality, such as C. difficile infection or certain fungal and viral infections.
In the same vein, the TARZAN study, a descriptive analysis of CPE-BSI episodes in neutropenic onco-hematologic patients, reported 30% empirical ceftazidime/avibactam use and a 7-day infection-attributable mortality of 11% [35], compared with 6% 15-day mortality in our cohort. Although the data are not directly comparable due to some methodological differences—the TARZAN study focused exclusively on confirmed CPE-BSI cases, whereas our work included all episodes of FN—both point to the potential negative clinical impact of a nonactive empirical therapy against CRE in this high-risk population.
These data may provide a rationale for the broader empirical use of novel beta-lactams in colonized patients. Nevertheless, in order to prevent unnecessary overuse of these last-resort treatments, future research should attempt to stratify which colonized patients have a low risk of invasive infection due to the colonizing strain. Meanwhile, the elevated mortality observed in confirmed OXA-48-E infections may prompt consideration of novel beta-lactams as empirical treatment while awaiting microbiologic results, as suggested by ECIL—at least in colonized patients with severe clinical presentation. Their use in these cases should be accompanied by early de-escalation if initial cultures return negative. In fact, if OXA-48-E infection is not confirmed within the first 24–48 hours after sample collection, de-escalation to a narrower-spectrum agent could represent a potential option, as previous studies have shown that all MDR-GNB BSI episodes were positive within the first 24 hours from the beginning of incubation [36].
Given that prior colonization with OXA-48-E itself was not associated with poorer outcomes, and that nearly two-thirds of colonized patients with FN did not develop a confirmed OXA-48-E infection, the clinical impact of colonization alone may be limited. Consequently, implementing risk scores to better stratify patients could offer a more cost-effective approach than relying solely on colonization status. In this sense, GRS is a risk score for predicting KPC-producing K. pneumoniae bacteremia, which was originally validated exclusively in individuals colonized with this microorganism [26], and later proposed to assist clinical decisions in this specific setting [31]. However, the development population of GRS included a limited number of neutropenic patients and the performance GRS has not been specifically assessed in hematological patients nor in OXA-48-E rectal carriers. In our study, GRS failed to reliably predict OXA-48-E infection in colonized patients developing FN, highlighting the need for tailored risk scores in this population. Although a few recent retrospective studies have attempted to develop prediction models for CPE infections in hematologic patients [37, 38], they are limited by small sample sizes, single-center retrospective designs, and a lack of external validation. Consequently, no predictive model for CPE infections has yet proven both reliable and universally applicable in this setting.
Our study has some limitations. First, its single-center retrospective design limits the generalizability of our findings. Thus, clinical predictors of OXA-48-E infection identified might be influenced by the specific characteristics of our setting, although they align with prior data [20]. Second, the absence of molecular typing to assess genetic relatedness between surveillance and clinical isolates in patients with confirmed OXA-48-E infection precludes confirming that they were the same strain. However, susceptibility data for the colonizing isolates were available, and their phenotypic profiles closely matched those of the clinical isolates. Third, antibiotic treatments were not randomized, which may introduce potential indication bias. Fourth, because FN episodes—rather than individual patients—were used as the unit of analysis, some degree of interdependence between FN episodes within the same patient cannot be excluded. However, carefully defined criteria were applied to identify independent episodes (no uncontrolled infection and ≥96 hours afebrile without antibiotics between episodes). Consistent with this, among patients with multiple episodes, the median interval between episodes was 46 days, suggesting that most occurred in distinct clinical settings. Finally, the number of events—both OXA-48-E infections and deaths—was relatively low, potentially limiting the detection of additional prognostic factors in multivariate analyses. Nonetheless, this study represents one of the largest reported cohorts of hematologic patients colonized with OXA-48 producers. We deliberately selected 15-day mortality outcome variable to better capture infection-attributable deaths and minimize confounding from underlying hematologic disease or unrelated comorbidities. While this shorter follow-up may have missed some later events, it provides a clearer view of infection-attributable mortality burden. Finally, the lack of allogeneic hematopoietic stem-cell transplantation and cellular therapies, such as chimeric antigen receptor (CAR)-T-cell therapy at our center also represents a limitation. As a result, our sample may not fully reflect populations treated in centers that provide these immunosuppressive therapies.
In summary, prior rectal colonization emerged as the strongest predictor of OXA-48-E infection in hematologic patients with FN, since all cases in our cohort occurred exclusively in previously colonized individuals. Surveillance cultures may therefore aid empirical decision-making in FN, helping to avoid unnecessary coverage of these strains in patients with a recent negative rectal swab. Similarly, considering that delayed active antibiotic therapy was associated with increased mortality, together with the higher risk of OXA-48-E infections in carriers, empirical use of a novel beta-lactam with OXA-48 activity could be considered in this subgroup, particularly in episodes with severe FN, ie, those manifesting as sepsis or septic shock. Finally, although our data are limited, they also suggest that, in the absence of confirmed OXA-48-E infection, early de-escalation to standard regimens might be considered a safe option for colonized patients with FN. However, further studies are needed to confirm these findings.
Supplementary Material
Notes
Author contributions . N. M. had full access to all the study data and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: M. R.- F., P. R.- G., and N. M. Data collection, data analysis, and interpretation of data: all authors. Statistical analysis: M. R.- F., P. R.- G., and N. M. Drafting of the manuscript: M. R.- F. and N. M. Critical revision of the manuscript for important intellectual content: all authors. Study supervision: N. M.
Data statement . All relevant data are presented in the manuscript. More detailed data can be provided by the authors upon reasonable request.
Ethical approval . This study was designed and performed according to the Helsinki Declaration and was approved by Hospital Universitario de Valme Ethics Committee (Ref. SICEIA-2024-002821). The need to obtain written informed consent was waived due to the observational nature of the study.
Financial support . This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Contributor Information
Miguel Rodríguez-Fernández, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Pablo Rosales-García, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Pedro Martínez Pérez-Crespo, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Lucía Valiente-De-Santis, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
José Luis Cuevas-Corrales, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
José María San-Emeterio Garrido, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Ana Isabel Aller-García, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Carmen Serrano-Martino, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Marta Trigo-Rodríguez, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Rocio Herrero, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
María Dolores Valverde-Fredet, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Antonio Fernández-Pevida, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Eva León, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Nicolás Merchante, Unidad de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme. Universidad de Sevilla, Instituto de Investigaciones Biomédicas de Sevilla, IBIS (Universidad de Sevilla, Junta de Andalucía, CSIC), Seville, Spain.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. Garcia-Vidal C, Stern A, Gudiol C. Multidrug-resistant, gram-negative infections in high-risk haematologic patients: an update on epidemiology, diagnosis and treatment. Curr Opin Infect Dis 2021; 34:314–22. [DOI] [PubMed] [Google Scholar]
- 2. Trecarichi EM, Pagano L, Candoni A, et al. Current epidemiology and antimicrobial resistance data for bacterial bloodstream infections in patients with hematologic malignancies: an Italian multicentre prospective survey. Clin Microbiol Infect 2015; 21:337–43. [DOI] [PubMed] [Google Scholar]
- 3. Bonomo RA. β-Lactamases: a focus on current challenges. Cold Spring Harb Perspect Med 2017; 7:a025239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Van Duin D, Doi Y. The global epidemiology of carbapenemase-producing Enterobacteriaceae. Virulence 2017; 8:460–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Palacios-Baena ZR, Oteo J, Conejo C, et al. GEIH-GEMARA (SEIMC) and REIPI group for CPE. Comprehensive clinical and epidemiological assessment of colonisation and infection due to carbapenemase-producing Enterobacteriaceae in Spain. J Infect 2016; 72:152–60. [DOI] [PubMed] [Google Scholar]
- 6. Rivera-Izquierdo M, Láinez-Ramos-Bossini AJ, Rivera-Izquierdo C, et al. OXA-48 carbapenemase-producing Enterobacterales in Spanish hospitals: an updated comprehensive review on a rising antimicrobial resistance. Antibiotics (Basel) 2021; 10:89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Cañada-García JE, Moure Z, Sola-Campoy PJ, et al. CARB-ES-19 multicenter study of carbapenemase-producing Klebsiella pneumoniae and Escherichia coli from all Spanish provinces reveals interregional spread of high-risk clones such as ST307/OXA-48 and ST512/KPC-3. Front Microbiol 2022; 13:918362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gracia-Ahufinger I, López-González L, Vasallo FJ, et al. The CARBA-MAP study: national mapping of carbapenemases in Spain (2014-2018). Front Microbiol 2023; 14:1247804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Alevizakos M, Karanika S, Detsis M, Mylonakis E. Colonisation with extended-spectrum β-lactamase-producing Enterobacteriaceae and risk for infection among patients with solid or haematological malignancy: a systematic review and meta-analysis. Int J Antimicrob Agents 2016; 48:647–54. [DOI] [PubMed] [Google Scholar]
- 10. Arnan M, Gudiol C, Calatayud L, et al. Risk factors for, and clinical relevance of, faecal extended-spectrum β-lactamase producing Escherichia coli (ESBL-EC) carriage in neutropenic patients with haematological malignancies. Eur J Clin Microbiol Infect Dis 2011; 30:355–60. [DOI] [PubMed] [Google Scholar]
- 11. de la Court JR, Heijmans J, Huynh J, et al. Guidance of empirical antimicrobial therapy by surveillance cultures in high-risk neutropenic patients: a retrospective cohort study. Antimicrob Resist Infect Control 2022; 11:160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Liss BJ, Vehreschild JJ, Cornely OA, et al. Intestinal colonisation and blood stream infections due to vancomycin-resistant enterococci (VRE) and extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBLE) in patients with haematological and oncological malignancies. Infection 2012; 40:613–9. [DOI] [PubMed] [Google Scholar]
- 13. Vehreschild MJGT, Hamprecht A, Peterson L, et al. A multicentre cohort study on colonization and infection with ESBL-producing Enterobacteriaceae in high-risk patients with haematological malignancies. J Antimicrob Chemother 2014; 69:3387–92. [DOI] [PubMed] [Google Scholar]
- 14. Satlin MJ, Chavda KD, Baker TM, et al. Colonization with levofloxacin-resistant extended-spectrum β-lactamase-producing Enterobacteriaceae and risk of bacteremia in hematopoietic stem cell transplant recipients. Clin Infect Dis 2018; 67:1720–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Demiraslan H, Cevahir F, Berk E, Metan G, Cetin M, Alp E. Is surveillance for colonization of carbapenem-resistant gram-negative bacteria important in adult bone marrow transplantation units? Am J Infect Control 2017; 45:735–9. [DOI] [PubMed] [Google Scholar]
- 16. Cattaneo C, Di Blasi R, Skert C, et al. Bloodstream infections in haematological cancer patients colonized by multidrug-resistant bacteria. Ann Hematol 2018; 97:1717–26. [DOI] [PubMed] [Google Scholar]
- 17. Micozzi A, Gentile G, Santilli S, et al. Reduced mortality from KPC-K. pneumoniae bloodstream infection in high-risk patients with hematological malignancies colonized by KPC-K. pneumoniae. BMC Infect Dis 2021; 21:1079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Alrstom A, Alsuliman T, Daher N, Abouharb R. The impact of modifying empirical antibiotic therapy based on intestinal colonization status on clinical outcomes of febrile neutropenic patients. Infect Chemother 2021; 53:63–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Torres I, Huntley D, Tormo M, et al. Multi-body-site colonization screening cultures for predicting multi-drug resistant gram-negative and gram-positive bacteremia in hematological patients. BMC Infect Dis 2022; 22:172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lalaoui R, Javelle E, Bakour S, Ubeda C, Rolain J-M. Infections due to carbapenem-resistant bacteria in patients with hematologic malignancies. Front Microbiol 2020; 11:1422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Averbuch D, Orasch C, Cordonnier C, et al. European guidelines for empirical antibacterial therapy for febrile neutropenic patients in the era of growing resistance: summary of the 2011 4th European Conference on Infections in Leukemia. Haematologica 2013; 98:1826–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Averbuch D. Final recommendations of the ECIL 10 working group . Conference from September 19th to 21st, 2024. Available at: https://www.ecil-leukaemia.com/en/resources/resources-ecil. Accessed 11 November 2025.
- 23. Schmidt-Hieber M, Teschner D, Maschmeyer G, Schalk E. Management of febrile neutropenia in the perspective of antimicrobial de-escalation and discontinuation. Expert Rev Anti Infect Ther 2019; 17:983–95. [DOI] [PubMed] [Google Scholar]
- 24. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS international sepsis definitions conference. Crit Care Med 2003; 31:1250–6. [DOI] [PubMed] [Google Scholar]
- 25. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 2016; 315:801–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Giannella M, Trecarichi EM, De Rosa FG, et al. Risk factors for carbapenem-resistant Klebsiella pneumoniae bloodstream infection among rectal carriers: a prospective observational multicentre study. Clin Microbiol Infect 2014; 20:1357–62. [DOI] [PubMed] [Google Scholar]
- 27. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988; 16:128–40. [DOI] [PubMed] [Google Scholar]
- 28. Paul M, Carrara E, Retamar P, 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; 28:521–47. [DOI] [PubMed] [Google Scholar]
- 29. Tamma PD, Heil EL, Justo JA, Mathers AJ, Satlin MJ, Bonomo RA. Infectious Diseases Society of America 2024 guidance on the treatment of antimicrobial-resistant gram-negative infections. Clin Infect Dis 2024; 79:ciae403. [DOI] [PubMed] [Google Scholar]
- 30. Taplitz RA, Kennedy EB, Bow EJ, et al. Antimicrobial prophylaxis for adult patients with cancer-related immunosuppression: ASCO and IDSA clinical practice guideline update. J Clin Oncol 2018; 36:3043–54. [DOI] [PubMed] [Google Scholar]
- 31. Cano A, Gutiérrez-Gutiérrez B, Machuca I, et al. Risks of infection and mortality among patients colonized with Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: validation of scores and proposal for management. Clin Infect Dis 2018; 66:1204–10. [DOI] [PubMed] [Google Scholar]
- 32. Willems RPJ, van Dijk K, Vehreschild MJGT, et al. Incidence of infection with multidrug-resistant gram-negative bacteria and vancomycin-resistant enterococci in carriers: a systematic review and meta-regression analysis. Lancet Infect Dis 2023; 23:719–31. [DOI] [PubMed] [Google Scholar]
- 33. Shields RK, Potoski BA, Haidar G, et al. Clinical outcomes, drug toxicity, and emergence of ceftazidime-avibactam resistance among patients treated for carbapenem-resistant Enterobacteriaceae infections. Clin Infect Dis 2016; 63:1615–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Martinez-Nadal G, Puerta-Alcalde P, Gudiol C, et al. Inappropriate empirical antibiotic treatment in high-risk neutropenic patients with bacteremia in the era of multidrug resistance. Clin Infect Dis 2020; 70:1068–74. [DOI] [PubMed] [Google Scholar]
- 35. Sastre-Escolà E, Ntziora F, Chiurlo M, et al. Ceftazidime/avibactam for the treatment of bloodstream infection due to carbapenemase-producing Enterobacterales in onco-haematologic neutropenic patients (the TARZAN study). J Antimicrob Chemother 2025; 80:1673–81. [DOI] [PubMed] [Google Scholar]
- 36. Puerta-Alcalde P, Cardozo C, Suárez-Lledó M, et al. Current time-to-positivity of blood cultures in febrile neutropenia: a tool to be used in stewardship de-escalation strategies. Clin Microbiol Infect 2019; 25:447–53. [DOI] [PubMed] [Google Scholar]
- 37. Liu J, Zhang H, Feng D, et al. Development of a risk prediction model of subsequent bloodstream infection after carbapenem-resistant Enterobacteriaceae isolated from perianal swabs in hematological patients. Infect Drug Resist 2023; 16:1297–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Gallardo-Pizarro A, Teijón-Lumbreras C, Monzo-Gallo P, et al. Development and validation of a machine learning model for the prediction of bloodstream infections in patients with hematological malignancies and febrile neutropenia. Antibiotics (Basel) 2024; 14:13. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.