Admission albumin after a burn predicts the kidney injury, not just the death

Design

systematic review of 19 studies with meta-analysis of 9, random-effects pooling with hierarchical summary ROC curves

Population

patients admitted with burn injury, serum albumin measured on admission

Primary outcome

in-hospital mortality and acute kidney injury

Effect

mortality OR 9.51 (95% CI 3.04-29.78, I² 49.3%); acute kidney injury OR 2.83 (2.49-3.22, I² 0%)

A systematic review screened 5,587 studies, included 19 and pooled nine, to test whether serum albumin measured on admission after burn injury predicts outcome. The mortality association is the eye-catching one and the least reliable: an odds ratio of 9.51 (95% CI 3.04-29.78) with moderate heterogeneity (I² 49.3%) and an interval spanning an order of magnitude.

The acute kidney injury finding is the usable one. Admission hypoalbuminaemia carried an odds ratio of 2.83 for acute kidney injury, with a tight interval (2.49-3.22) and no heterogeneity at all (I² 0%). Evidence for length of stay, ventilation, sepsis and pulmonary infection was limited and heterogeneous. The authors are explicit that optimal cut-off values have not been established and that albumin trends within the first 24 hours have not been characterised.

What this supports is using a number already on the admission panel as a trigger for renal vigilance rather than as a therapeutic target. A low admission albumin in a burn patient is a reason to be careful with nephrotoxins, to track urine output formally rather than casually, and to involve nephrology early. It is not a reason to give albumin — no trial here tested replacing it, and correcting a prognostic marker is not the same as changing the prognosis.

• Treat a low admission albumin as a flag for renal risk, not as something to correct
• Formalise urine output monitoring and avoid avoidable nephrotoxins in these patients
• No cut-off is established; use the value in context rather than as a threshold
• The mortality odds ratio has an interval from 3 to 30 and should not be quoted as a number
• Burn units in India frequently admit late with established hypoalbuminaemia; that changes what the marker means at presentation

Why it matters

It turns a routine admission blood test into a renal risk flag in a population where kidney injury is often noticed late.

Don't overread it

This is prognostic association from observational studies — albumin marks the sick patient rather than causing the kidney injury, and giving albumin has not been shown to help.

The statistics, in plain English

An odds ratio of 2.83 with an interval of 2.49 to 3.22 and I² of 0% means the included studies agreed closely and the estimate is precise — unusual in this literature. The mortality estimate is the opposite: 3.04 to 29.78 is so wide that the only defensible statement is that the association exists and its size is unknown.

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.