Annane 2013.

Methods	Randomised, multicentre, single‐blind clinical trial in 8 French ICUs
Participants	Inclusion criteria: Adults admitted to a participating ICU were eligible if they, < 48 h, had SIRS, acute dysfunction of at least 1 organ, absence of indisputable clinical infection, and negative microbial cultures Exclusion criteria: Pregnancy, burns over ≥ 15% of body surface area, trauma, outpatient or inpatient cardiac arrest, post‐orthopaedic surgery status, drug‐related neutropenia, withdrawal of life‐supportive therapies or a decision to withhold them, indisputable clinical infection or antibiotic exposure ≥ 48 h during the time shortly before ICU admission Included in this study: 62/1250 screened patients were eligible for the study, of whom 31 were randomised to each arm. 4 post randomisation exclusion (4 withdrew their consent)
Interventions	Guiding antibiotic decisions in ICU patients with non‐microbiologically proven apparent severe sepsis Algorithm used in this study: In the experimental arm, both initiation and discontinuation of antibiotics were guided by a PCT‐based algorithm, applied at 6 h and on day 3 and day 5 post randomisation. Briefly, antibiotic therapy was not to be started or was to be halted when PCT was < 0.25 μg/L, was strongly discouraged when PCT was ≥ 0.25 to < 0.5 μg/L, was recommended when PCT was ≥ 0.5 to < 5 μg/L, and was strongly recommended when PCT was ≥ 5 μg/L. Owing to the fact that surgery can increase PCT levels, for 12 participants enrolled in the 48‐hour postoperative period, the respective PCT cut‐offs were < 4 μg/L, ≥ 4 to < 9 μg/L, and ≥ 9 μg/L. Investigators were strongly advised not to overrule the algorithm every day up to the study day 5. In the control arm, the decision to start or stop antibiotic therapy was at the discretion of the participant’s physician, without knowledge of the participant’s PCT concentrations.
Outcomes	proportion of participants receiving antibiotics at day 5 post randomisation death at day 5, at ICU discharge, and at hospital discharge proportion of participants started on antibiotics post randomisation duration of antibiotic exposure SOFA score at day 3 and day 5 proportion of participants with infection acquired between randomisation and day 3, day 5, and ICU discharge ICU and hospital length‐of‐stay
Notes	Funding: Research grant partly by Thermo Fisher B·R·A·H·M·S France. The sponsor had no input in study design, conduct, or reporting. Follow‐up time: Until hospital discharge or 30 days' post randomisation, whichever came first Registration: NCT01025180
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomised in a 1:1 ratio according to a computer‐generated list. Randomisation was centralised through a secured web site and performed by an independent statistician.
Allocation concealment (selection bias)	Low risk	Centralised randomisation using permutation blocks
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	The control arm remained blinded to PCT levels. Masking of antibiotic therapy was not feasible in this study.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Investigators remained blinded to PCT levels in the control arm, but no blinding of overall outcome assessment was mentioned in the study.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Follow‐up for participants on antibiotic on day 5 was: 58/62. 4 participants withdrew their consent.
Selective reporting (reporting bias)	Low risk	Outcomes correspond to study protocol. Trial registered (NCT01025180).
Other bias	Unclear risk	Moderate adherence in the PCT arm: physicians were non‐compliant with the PCT‐based algorithm in 19% of participants at 6 h, 17% on day 3, and 37% on day 5. The study was stopped prematurely owing to the low incidence of eligible patients. As a consequence, the study population is small with low statistical power.