Intravenous rehydration in severe malnutrition: the harm did not appear

Design

systematic review and meta-analysis of 3 randomised controlled trials, GRADE moderate certainty

Population

484 children with severe acute malnutrition hospitalised with severe dehydration from gastroenteritis in lower- and middle-income countries; 72 with kwashiorkor

Primary outcome

in-hospital mortality

Effect

RR 0.71 (95% CI 0.46 to 1.10, I² 0%); severe hyponatraemia at 24 h RR 0.66 (0.44-0.99); no fluid overload events in either arm

Guidance has long treated the intravenous drip as dangerous in a severely malnourished child with gastroenteritis, on the reasoning that a depleted myocardium and disordered sodium handling make fluid overload likely. This review found the three randomised trials that have actually tested it — 484 children in lower- and middle-income countries, 72 with kwashiorkor, one at low risk of bias and two with some risk.

In-hospital mortality was 29% lower with intravenous rehydration, but the interval crosses no effect (RR 0.71, 95% CI 0.46 to 1.10), with no heterogeneity between trials (I² 0%). No fluid overload events were reported in either arm. Severe hyponatraemia at 24 hours was less frequent with the drip (RR 0.66, 0.44-0.99). Single-trial estimates for shock, hypernatraemia and 28-day mortality were all imprecise.

The authors state the position plainly: the effect ranges from a 54% relative reduction to a 10% relative increase in death, with fewer adverse events on the intravenous side. That is not a mandate to change first-line practice — oral and nasogastric rehydration remain appropriate for a child who can take them. It is a reason to stop treating the drip as a contraindicated act in a child with severe dehydration who cannot, and to stop delaying it while an oral attempt fails.

• Try oral or nasogastric rehydration first in a child who can tolerate it — nothing here changes that
• Do not withhold intravenous fluid from a severely malnourished child with severe dehydration who cannot drink
• Measure sodium at presentation and at 24 hours; hyponatraemia was the commoner electrolyte problem, and it was commoner with oral rehydration
• Watch for fluid overload clinically regardless — absence of events in 484 children does not establish absence of risk
• This applies to severe dehydration from gastroenteritis, not to shock, and not to the child with some dehydration

Why it matters

A long-standing caution against intravenous fluid in these children rests on physiological reasoning that three trials have not borne out.

Don't overread it

484 children across three trials is a small evidence base, and moderate certainty is not a licence to abandon oral rehydration.

The statistics, in plain English

A risk ratio of 0.71 with an interval of 0.46 to 1.10 does not exclude no effect, so this is not proof that drips save lives. What it does exclude is a large increase in death, and the absence of any fluid overload event in 484 children makes the specific feared harm look less likely than the guidance implies. I² of 0% means the three trials agreed with each other, which strengthens the estimate even though it remains imprecise.

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.