A permissive pressure target in ventilated children cut vasopressor use by a third

Design

Pragmatic, unblinded, multicentre randomised trial

Population

1,900 ventilated children on vasoactive drugs for hypotension in 23 PICUs

Primary outcome

Composite of death and duration of invasive ventilation at 30 days

Effect

Probabilistic index 0.52 (0.49–0.54); vasopressor dose −37.5%

PRESSURE, published in JAMA on 28 September, was a pragmatic, unblinded trial in 23 paediatric intensive care units in three countries. 1,900 ventilated children on vasoactive drugs for hypotension were randomised to a permissive mean arterial pressure target — anything above the fifth centile for age — or to the target the treating consultant chose.

The permissive group ran a median pressure of 58 mm Hg against 61 mm Hg, and received 37.5% less vasopressor by noradrenaline-equivalent dose. The composite of death and duration of ventilation at 30 days did not differ (probabilistic index 0.52, 95% CI 0.49 to 0.54). Kidney replacement therapy and adverse events were similar; intensive care stay among survivors was slightly shorter with the permissive target.

The finding is that less was not worse. For anyone who looks after hypotensive children — in paediatric units, general intensive care units that take children, or emergency departments stabilising them — it supports tolerating a pressure at the low end of normal for age rather than escalating drugs to reach a higher number. It does not show that lower targets improve survival, and it does not apply to adults, where separate trials exist.

• In ventilated children on vasopressors, a MAP above the fifth centile for age was as safe as usual targets.
• The permissive approach used 37.5% less vasopressor without more kidney replacement or harm.
• Keep age-specific blood pressure centile charts at the bedside to set the floor.
• Look at perfusion — lactate, urine output, capillary refill — as well as the number.

Why it matters

It removes the pressure to chase a higher number with more drug in a sick child.

Don't overread it

No difference in outcomes is not the same as benefit; this shows a lower target is acceptable, not better.

The statistics, in plain English

A probabilistic index of 0.52 is the chance that a randomly chosen child in one group did better than one in the other; 0.50 means no difference. The interval 0.49 to 0.54 includes 0.50, so there was no detectable effect on the primary outcome in either direction. The slightly shorter intensive care stay was one of seven secondary outcomes and should be treated as a hint.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.