Procalcitonin-guided duration of antibiotic treatment in children hospitalised with confirmed or suspected bacterial infection in the UK (BATCH): a pragmatic, multicentre, open-label, two-arm, individually randomised, controlled trial.

Summary

In a multicentre pragmatic RCT of 1,949 hospitalized children, a procalcitonin-guided algorithm did not shorten the duration of IV antibiotics versus usual care and was non-inferior for safety. The findings argue against routine use of procalcitonin-guided algorithms where robust pediatric stewardship is in place.

Key Findings

• No reduction in IV antibiotic duration: median 96.0 h (PCT) vs 99.7 h (usual care); HR 0.96 (95% CI 0.87–1.05).
• Safety was non-inferior: composite adverse events in 9% vs 9%; adjusted risk difference −0.81% (95% CI upper bound 1.11, below 5% margin).
• Pragmatic, multicentre RCT across 15 hospitals randomizing 1,949 children using minimisation with age and site.

Clinical Implications

Do not implement procalcitonin-guided algorithms solely to reduce IV antibiotic duration in pediatric inpatients where strong stewardship exists; maintain current stewardship-based decision-making.

Limitations

• Open-label design could influence clinician behavior despite objective primary outcome.
• Findings reflect UK settings with strong stewardship and may not generalize to regions with different practices; focus on IV duration may not capture total antibiotic exposure.

Future Directions

Evaluate procalcitonin or combined biomarker algorithms in settings with less mature stewardship, and identify subgroups (e.g., severe sepsis, immunocompromised) where biomarker guidance may add value.