Table 1.

Summary of the main studies on follow up blood cultures (FUBCs) in Gram negative bacteremia.

Study: First Author, Year Design	✔Inclusion Criteria ✕Exclusion Criteria	FUBC Definition N of FUBCs/N of Patients (%)	Positive FUBC /N of FUBCs (%)	Positive ICU FUBCs/ICU Patients (%)	Results and Conclusions	Limitations
Kang, 2013 retrospective case-control [18]	✔Age ≥ 18 ✔First episode of KpB ✕Polymicrobial ✕Recurrent KpB	BC drawn after more than 2 days from the index BC 862/1068 (81)	62/862 (7.2)	NA	Risk factors for persistent KpB included in a clinical score: - unfavorable treatment response - intra-abdominal infection - high CCI - SOT Routine FUBCs not justified.	Retrospective Only one pathogen considered No ICU data No multivariate for mortality risk factors
Wiggers, 2016 retrospective cohort [19]	✔Age ≥ 18 ✔First episode of GNB ✕Recurrent GNB	BC drawn 2–7 days from the index BC 247/901 (27.4) [GN only considered]	27/247 (10.9) [GNB only considered]	NA	Increased 30 days mortality in patients undergoing FUBCs, regardless of the result Repeated BC in GN bacteremia offer low yield	Retrospective Mixed population of GP and GN No ICU data Small sample with PB (defined as positive BC 2–7 days after the index BC
Canzoneri, 2017 retrospective case-control [15]	✔Age ≥ 18 ✔One positive BC ✕Fungemia ✕Potential contaminants	BC drawn after at least 24 h from the index BC 383/500 (77) [GP + GN considered]	8/140 (6) [GNB only considered]	18/165 (10.9) [GP + GN considered]	PB more common for GPs (21%), than polymicrobial (10%), than GNs (6%) FUBCs have little utility in patients with GNB	Retrospective Mixed population of GP and GN Polymicrobial infections included Small number of patients with PB
Shi, 2019 retrospective case-control [20]	✔Age ≥ 18 ✔Bacteremic UTI ✔At least one FUBC ✕Non-urinary source of bacteremia	More than one separate BC taken more than 24 h after the index BC 306/333 (92) [GP + GN considered]	39/264 (14.8) [GN only considered]	20/55 (36.4)	Not recommended routine FUBC Predictors for positive FUBC in bacteremic UTI: malignancy, initial ICU admission, high CRP level and longer time to defervescence	Retrospective Mixed population with GP and GN Only UTIs considered
Uehara, 2019 retrospective observational [21]	✔Age ≤ 18 ✔GN B ✕Polymicrobial ✕FUBCs ≤ 24 h from the index BC ✕AT started ≥ 48 h from the index BC	BCs taken over 24 h from the index BC 99/137 (72.3)	21/99 (21.2)	NA	FUBC may still be useful in the management of GNB in children Presence of a CVC and resistance to empirical antibiotics were risks for positive FUBCs	Retrospective Small sample Possible selection bias
Giannella, 2020 retrospective cohort [22]	✔Age ≥ 18 ✔GNB ✕Polymicrobial ✕Potential contaminants ✕Death ≤ 72 h after index BC ✕Unavailable clinical data	BCs drawn between 24 h and 7 days after the index BC 278/1576 (17.6)	107/278 (38.5)	21/126 (16.6)	FUBCs drawn in more severe, high risk, antibiotic resistant and initially inappropriately treated patients In this context, FUBCs execution associated to higher rate of source control, ID consultation and lower 30-day mortality	Retrospective Single center
Maskarinec, 2020 prospectively enrolled cohort [23]	✔Age ≥ 18 ✔GNB ✕Polymicrobial ✕Death ≤ 24 h after index BC	BCs drawn from 24 h to 7 days from index BC 1164/1702 (68.4)	228/1164 (19.6)	4/41 (9.8)	FUBCs drawn in high risk patients Obtaining FUBCs associated with decreased all-cause and attributable mortality Positive FUBCs associated with increased all-cause and attributable mortality	Poor data on patients’ clinical status at FUBCs collection No data on management changes based on FUBC results
Jung, 2020 retrospective observational cohort [24]	24GNB ✕Age﹤18 ✕Death ≤ 48 h after index BC ✕Polymicrobial ✕Different species from the index BC identified by FUBC	BCs drawn 2–7 days from index BC 1276/1481 (86.2)	122/1276 (9.6)	NA	FUBCs can be avoided in most uncomplicated cases of GNB and could be considered selectively in high risk patients Two clinical scores for patients with eradicable and non-eradicable source of infection	Retrospective Not evaluated the impact of FUBCs on patient outcome
Mitaka, 2020 retrospective multicenter observational [25]	✔Age ≥ 18 ✔GNB Potential contaminants	BC draws after at least 24h of AT 306/463 (66.1)	28/306 (9.2)	18/130 (13.9)	RF for positive FUBCs: ESRD on hemodialysis, intravascular device, ESBL or carbapenemase-producing organism Higher yield of positive FUBCs in patients with ≥ 1 RF Routine FUBCs are not necessary	Retrospective Only the first FUBC analyzed 26% FUBCs not performed without standardizing the decision
Spaziante, 2020 retrospective observational study [26]	✔Age ≥ 18 ✔GNB ✔FUBC ≤ 24–72 h from the FUBC or ≤ 48 h from the beginning of AAT ✕Polymicrobial	BCs done within 48 h from the beginning of AT and then every 24–72 h after FUBCs 78/107 (73) [GNB episodes from 69 patients]	28/78 (35.9) [patients]	[All patients were in ICU]	Septic thrombus infection was the source in 14 (50%) cases of GNB-PB. A MDR isolate was in 60 BCs (76.9%) FUBCs represent a useful tool in the management of GN-PB, especially if caused by STI	Retrospective Only ICU patients from polytrauma ICU Small sample

Abbreviations: AT, antimicrobial therapy; BC, blood culture; BSI, blood stream infection; CCI, Charlson comorbidity index; CRP, C-reactive protein; CVC, central venous catheter; ESBL, extended spectrum β-lactamase; ESRD, end stage renal disease; FUBC, follow-up blood cultures; GN, Gram negative; GNB, Gram negative bacteremia; GP, Gram positive; ICU, intensive care unit; KpB, Klebsiella pneumoniae bacteremia; MDR, multidrug resistant; NA, not available/not applicable; PB, persistent bacteremia; RF, risk factor; SOT, solid organ transplant; UTI, urinary tract infection.