Safety and risk of febrile recurrence after early antibiotic discontinuation in high-risk neutropenic patients with haematological malignancies: a multicentre observational study

Abstract

Background Early antibiotic discontinuation according to the Fourth European Conference on Infections in Leukaemia (ECIL-4) recommendations is not systematically applied in high-risk neutropenic patients with haematological malignancies. Methods A retrospective multicentre observational study was conducted over 2 years to evaluate the safety of early antibiotic discontinuation for fever of unknown origin (FUO) during neutropenia after induction chemotherapy or HSCT, in comparison with a historical cohort. We used Cox proportional hazards models, censored on neutropenia resolution, to analyse factors associated with febrile recurrence. Results Among 147 included patients in the ECIL-4 cohort, mainly diagnosed with acute leukaemia (n = 104, 71%), antibiotics were discontinued during 170 post-chemotherapy neutropenic episodes. In comparison with the historical cohort of 178 episodes of neutropenia without antibiotic discontinuation, no significant differences were observed regarding febrile recurrences [71.2% (121/170) versus 71.3% (127/178), P = 0.97], admission in ICUs [6.5% (11/170) versus 11.2% (20/178), P = 0.17], septic shock [0.6% (1/170) versus 3.9% (7/178), P = 0.07] and 30 day mortality [1.4% (2/147) versus 2.7% (4/150), P = 0.084]. In the ECIL-4 cohort, the rate of bacteraemia in case of febrile recurrence was higher [27.1% (46/170) versus 11.8% (21/178), P < 0.01] and antibiotic consumption was significantly lower (15.5 versus 19.9 days, P < 0.001). After early antibiotic discontinuation according to ECIL-4 recommendations, enterocolitis was associated with febrile recurrence [HR = 2.31 (95% CI = 1.4–3.8), P < 0.001] and stage III–IV oral mucositis with bacteraemia [HR = 2.26 (95% CI = 1.22–4.2), P = 0.01]. Conclusions After an FUO episode in high-risk neutropenia, compliance with ECIL-4 recommendations for early antibiotic discontinuation appears to be safe and mucosal damage was associated with febrile recurrence and bacteraemia. Prospective interventional studies are warranted to assess this strategy in high-risk neutropenic patients.

Approach to febrile neutropenia in patients undergoing treatments for hematologic malignancies

“…There was no difference in the composite outcome of ICU admission or death within 30 days from chemotherapy initiation, while the median days of broad‐spectrum antibiotics significantly differed by 10 days 62 . Other retrospective studies implementing ECIL‐4‐based de‐escalation similarly evaluated broad‐spectrum antibiotic durations 52,54–56 and/or clinical safety parameters 52,54,64 . While more randomized controlled trials comparing different approaches to de‐escalation of broad‐spectrum antibiotics for FN without identified infection would strengthen its adoption, existing evidence supports the safety of ECIL‐4‐based de‐escalation protocols.…”

“…63 Several observational studies compare clinical outcomes in patients undergoing early de-escalation of empiric broad-spectrum antibiotics versus continuation through ANC recovery, with the six adopting an ECIL-4-based de-escalation strategy. [52][53][54][55][56]64 The largest examined 958 admissions in 596 high-risk hematologic malignancy patients in Belgium undergoing chemotherapy or HCT before and after ECIL-4 implementation in 2017. 53 They found that bacteremia occurred more frequently in the de-escalation group, but without an increase in infection-related ICU admission or mortality.…”

“…studies implementing de-escalation similarly evaluated broad-spectrum antibiotic durations 52,[54][55][56] and/or clinical safety parameters. 52,54,64 While more randomized controlled trials comparing different approaches to de-escalation of broad-spectrum antibiotics for FN without identified infection would strengthen its adoption, existing evidence supports the safety of ECIL-4-based de-escalation protocols.…”

Neutropenia and antibiotics: when, what, how and why?

“…Several of these studies demonstrated a significant reduction in BSA use, yet no differences in ICU transfers, bacteremia incidence, infection relapses, or mortality [29–32]. While two studies observed a higher bacteremia incidence in patients that followed ECIL-guidelines [33 ▪ ,34], two also showed a decreased risk of ICU [35 ▪ ] admission and death [34,35 ▪ ]. The ANTIBIOSTOP trial was a single-center, prospective, observational trial including 238 episodes of FN in 123 patients, where antibiotics were stopped after 48 h versus five days for all patients without an identifiable infection.…”

Efficacy of an antimicrobial stewardship intervention for early adaptation of antibiotic therapy in high-risk neutropenic patients

“…However, most of these studies solely investigated the effects of one form of adaptation (i.e. de-escalation or discontinuation) [ 13 , 15 , 16 , 18 – 21 ] or focused on specific presentations of FN [ 13 , 15 , 18 , 19 , 21 , 24 , 25 ] or patient profiles [ 14 – 16 , 18 , 24 ]. One randomized controlled trial has found that discontinuation of EAT after 72 h of apyrexia and clinical recovery in high-risk neutropenic patients without microbiological documentation is safe and can significantly reduce unnecessary exposure to antimicrobials [ 13 ].…”