Stop checking gastric residual volumes in ventilated children

Design

pragmatic, multicentre, randomised non-inferiority trial in 24 paediatric intensive care units

Population

4700 children aged 0 to 16 receiving invasive ventilation and starting enteral feeds, median age 8 months, 42.6% female

Primary outcome

co-primary: survival and ventilator-free days at 30 days (non-inferiority), and percentage meeting energy requirements by 72 hours (superiority)

Effect

median 25 ventilator-free days in both arms, adjusted OR 0.95 (95% CI 0.86 to 1.05); energy requirements met 80.3% vs 76.8%, difference 3.2 points (1.3 to 5.2)

GASTRIC-PICU randomised 4700 children aged 0 to 16 receiving invasive ventilation and starting enteral feeds, across 23 UK paediatric intensive care units and one in Switzerland, to six-hourly gastric residual volume checks or none, with feed tolerance in the no-check arm judged on clinical signs alone. Median age was 8 months.

Not checking was non-inferior for the composite of survival and ventilator-free days at 30 days: median 25 days in both arms, adjusted odds ratio 0.95 (95% CI 0.86 to 1.05), with a per-protocol analysis in agreement. On the nutritional co-primary outcome it was superior: 80.3% of energy requirements met by 72 hours against 76.8%, an adjusted difference of 3.2 percentage points (95% CI 1.3 to 5.2).

This is a practice performed several times a day on almost every ventilated child in almost every unit, on no evidence, and its main effect has been to trigger the withholding of feeds. A 4700-patient trial is a decisive answer, and the direction is unusual for a de-implementation study: removing the test improved the outcome it was supposed to protect. Changing this needs a protocol change and nursing agreement rather than an individual decision, and the substitute has to be specified - clinical signs of intolerance, defined and documented, not nothing at all.

• Rewrite the local feeding protocol; this is not a change an individual clinician can make at the bedside
• Define what replaces the measurement: abdominal distension, vomiting, discomfort, and what each should trigger
• Audit the percentage of energy requirements met at 72 hours, which is where the gain appears
• Children needing gastric decompression for a specific clinical reason are a separate matter and are unaffected

The statistics, in plain English

Two co-primary outcomes were set with different questions: non-inferiority for the clinical composite and superiority for nutrition. The clinical odds ratio of 0.95 with an interval of 0.86 to 1.05 sits around 1.0 and stays inside the prespecified margin, and the per-protocol analysis agreeing with the intention-to-treat analysis matters more here than usual - in a non-inferiority trial, poor adherence pushes the two arms together and can manufacture a false non-inferiority, so agreement between the two analyses is the reassurance. The nutritional difference of 3.2 percentage points is statistically clear but modest in size; the case for the change rests on removing a burdensome practice that does no good, not on the calories alone.

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.