Study design and setting
Multicentre retrospective cohort study of patients hospitalised at three large teaching hospitals in northern Italy: i) Sant’Orsola Malpighi Hospital, Bologna; ii) City of Health and Sciences, Molinette Hospital, Turin; iii) San Martino Hospital, Genoa. The study period was from January 1st 2013 to December 31st 2016. Patients were identified through the records of the Microbiology Laboratory of each hospital.
Clinical charts and hospital records were reviewed, until 90 days after the index blood cultures (BCs), to gather study variables using a pre-established case report form. A local senior investigator systematically checked data for accuracy before they were recorded into a database. In addition, the numbers of patient days per year were recorded to assess the incidence of NFGN-BSI in the participating hospitals during the study period.
Ethic Committees of each centre approved the study, waiving of informed consent was obtained due to the retrospective non-interventional study design. Data were collected anonymously.
Inclusion criteria were all adult (≥18 years) patients diagnosed with NFGN-BSI, defined as one or more positive BCs obtained from a patient suspected of having infection. Patients were considered only once at the time of first episode (index BCs).
Exclusion criteria included: (i) polymicrobial BSI, defined as growth of more than one micro-organism, excluding potential contaminants (i.e. coagulase-negative staphylococci, Corynebacterium spp., Propionibacterium spp.); (ii) clinical data not available.
Variables and definitions
The primary outcome was all-cause mortality within 30 days after index BCs .
Predictor variables included: age, sex. Underlying diseases were assessed according to the Charlson comorbidity index . Immunosuppression included neutropenia (neutrophil count <500/mm3), solid organ transplantation, hematopoietic stem cell transplantation, corticosteroid therapy at a dosage higher then or equivalent to prednisone 16 mg/day ≥ 15 days, uncontrolled HIV infection (<200 CD4/mm3).
BSI was classified according to the site of acquisition into nosocomial, healthcare-associated and community acquired using Friedman’s criteria . Clinical severity at infection onset was assessed according to SOFA score and new septic shock criteria . BSI sources were established according to Centers for Disease Control and Prevention (CDC) criteria . In the absence of a recognized source, BSI was considered as primary. BSI was defined as complicated when the infection source was not fully removable.
The susceptibility pattern of isolates was classified according to Magiorakos et al. criteria  as non-MDR, MDR, XDR or PDR.
In addition, CDC surveillance definitions were used to assess susceptibility to carbapenems, extended-spectrum cephalosporins (ESC) and fluoroquinolones (FQ) (https://gis.cdc.gov/grasp/PSA/Downloads/AR-PhenotypeDefinitions.pdf).Moreover susceptibility to betalactam/betalactamase inhibitors (BL/BLI) and colistin was determined according to European Committee for Antimicrobial Susceptibility Testing (EUCAST) criteria.
Finally the new definition of “difficult to treat resistance” (DTR) was also assessed as reported elsewhere [11, 12] Empirical therapy was defined as antibiotics administered before the susceptibility report was available. It was considered appropriate when at least one in vitro active drug (according to the susceptibility pattern of the isolate) was administered within 24 h after drawing index BCs. Delayed or no active antibiotic administration within this period was considered as inappropriate empirical therapy. Definitive antibiotic therapy was defined as antibiotic treatment administered according to susceptibility results. Antibiotic regimens including more than one anti-gram-negative agents, irrespective of their in vitro activity against the BSI isolate, during more than 50% of treatment duration were defined as “combination regimen”. Antibiotic therapy including at least two drugs showing in vitro activity against the BSI isolate was labelled as “2-in vitro active combination regimen”. Duration of antibiotic treatment was defined as the number of consecutive days during which the patient received an appropriate antibiotic regimen. Source control was defined as the removal of the infection source within 7 days of index BCs, including the performance of non-surgical or surgical procedures to treat an obstructive focus or abscess at any site including, among others, the urinary tract, biliary tract and surgical site, and the removal of any device deemed as the source of BSI.
BCs were incubated using the BACTEC FX Automated Blood Culture System (Becton Dickinson, Franklin Lakes, NJ). All positive BCs were processed with Maldi Biotyper MALDI-TOF system (Bruker Daltonics, Bremen, Germany) for rapid and reliable species identification of microorganisms. Antimicrobial susceptibility testing was performed using the Vitek 2 automated system (bioMerieux, Marcy l’Etoile, France) in two hospitals (Bologna and Genova) and the MicroScan system in the remaining hospital (Torino). The minimum inhibitory concentrations (MICs) were interpreted using EUCAST clinical breakpoints for all tested antibiotics.
For the descriptive analysis, categorical variables were presented as absolute numbers and their relative frequencies. Continuous variables were presented as the mean and standard deviation if normally distributed or as the median and interquartile range (IQR) if non-normally distributed.
Risk factors for all-cause 30-day mortality were analysed by univariate and multivariate analysis. Categorical variables were compared using χ2 or Fisher exact test when appropriate. Continuous variables were compared using the Mann–Whitney U-test. Significant and clinically relevant covariates identified in univariate analysis were introduced into a multivariable Cox regression survival model after verifying for proportional hazards and collinearity. Significance was considered for p<0.05. All the analysis were performed using SPSS software.