Table 2.

Twenty key points for the management of bloodstream infection in critically ill patients

Statements
1.	The rising incidence of ESBLE is the most prominent matter of concern in community-acquired BSI
2.	The rising incidence of CPE and XDR Acinetobacter baumannii in HA-BSI is a matter of serious concern
3.	ICU-acquired BSI frequently occurs in critically ill patients, especially those with high severity indexes, immunosuppression, a surgical reason for admission, and the need for ECMO or other invasive procedures
4.	Most of ICU-acquired BSIs are related to catheter infection, intra-abdominal infections, and ventilator-associated pneumonia though no definite source is identified for a substantial proportion of cases
5.	Direct identification using Maldi-TOF or genotypic methods are accurate for bacterial identification especially for Gram-negative pathogens
6.	Genotypic methods of bacterial detection and resistance mechanisms identification are accurate. These methods may positively impact the timing and adequacy of antimicrobial therapy in ICU patients with BSI though real-life clinical studies are still needed to appraise their input precisely
7.	Choices about antimicrobials for treating critically ill patients with BSI should take into account several overlapped factors: (i) the empirical or targeted nature of the treatment; (ii) the presumed or proven origin site of the infection; (iii) the suspected or proven presence of antimicrobial resistance; (iv) immune status, and (v) the suspected or proven presence of candidemia
8.	A reasoned choice of empirical agents should be based on the suspected pathogen/s and on the estimated individual and environmental risks of MDR infection
9.	Recently approved, novel agents active against MDR organisms might be used, only if clearly, appropriate according to local epidemiology, for empirical treatment in critically ill patients
10.	In critically ill patients with BSI and increased distribution volume, loading dosages of hydrophilic antibiotics should be increased compared to dosages usually prescribed in non-critically ill patients
11.	Maintenance dosages should be adjusted according to fluctuations in the estimated renal function
12.	TDM should be routinely performed for vancomycin and aminoglycosides, and whenever feasible for polymyxins. TDM of beta-lactams may be used, especially for preventing neurotoxicity, but further research and standardization are needed for clearly delineating advantages and impact on patients’ outcomes
13.	Continuing combination therapy in BSI due to XDR Gram-negative bacteria may have an outcome benefit in the most severely ill patients with septic shock
14.	Source control including immediate removal of suspected intravascular catheters is always urgent in patients with septic shock
15.	In life-threatening surgical site infections, a “damage control” approach is the safest way to gain time and achieve stability
16.	ADE describes the initial re-evaluation of antimicrobial therapy when it targets decreasing the exposure to broad-spectrum antimicrobials. For treatment of BSI, it consists in stopping companion antibiotics or narrowing the spectrum of a pivotal antibiotic
17.	The antimicrobial regimen should be re-evaluated for its spectrum and effectiveness every day after the blood culture becomes positive and new information becomes available
18.	In ICU patients with uncomplicated BSI, duration of treatment can be matched to that of the source and the causative pathogen. In the absence of specific risk factors, a duration of when clinical stability is reached, shorter (≤ 7 days) should be proposed In the absence of specific risk factors, septic shock and if the source control is appropriate. preferred to longer antibiotic courses
19.	Specific pathogens at risk of septic metastasis or treatment failure require duration of 14 days in cases of uncomplicated infections and up to 4–8 weeks for Specific sources such as bone and joint infections, empyema, septic metastasis or sources not amenable to adequate source control
20.	Ongoing instability should not be a reason to blindly increase the duration of antimicrobial, but rather lead to investigate for insufficient source control, superinfection, drug-resistant pathogens or non-infectious causes of fever and shock. Continuing, escalating or stopping the antimicrobials accordingly should always be preceded by new microbiological specimens including blood cultures

ESBLE extended-spectrum beta-lactamase-producing Enterobacterales, BSI bloodstream infection, CPE carbapenemase-producing Enterobacterales, XDR extensively drug-resistant, ICU intensive care unit, ECMO extra-corporeal membrane oxygenation, MDR multidrug-resistant, TDM therapeutic drug monitoring, ADE antibiotic de-escalation