The Brain Injury Guidelines rule out deterioration with about 98% sensitivity in the lowest-risk group

Design

Systematic review and random-effects meta-analysis of diagnostic accuracy, PROSPERO registered

Population

23 studies of the original Brain Injury Guidelines in traumatic brain injury, from 4,733 screened

Primary outcome

Neurological deterioration, neurosurgical intervention and death

Effect

BIG1 pooled sensitivity 98.3% (95.7 to 99.4), 98.7% (97.4 to 99.4) and 98.2% (96.6 to 99.1); specificity about 13%

The Brain Injury Guidelines (BIG) sort patients with traumatic brain injury and a CT abnormality into risk categories to guide admission, repeat imaging and neurosurgical consultation. This systematic review and meta-analysis, registered with PROSPERO, screened 4,733 studies, included 47 articles, and pooled 23 studies of the original BIG.

For the lowest-risk category, BIG1, pooled sensitivity to rule out adverse outcomes was 98.3% (95% CI 95.7 to 99.4) for neurological deterioration, 98.7% (97.4 to 99.4) for neurosurgical intervention and 98.2% (96.6 to 99.1) for death. Specificity was low, around 13% to 14% with wide intervals (1.5% to 58.7%), meaning many patients classed higher did well too.

The authors conclude that BIG may identify patients suitable for discharge without further immediate investigation or neurosurgical consultation, and may reduce unnecessary transfers. These are mostly observational validation studies. High sensitivity means few missed events, but a small number of patients in BIG1 still deteriorated, so it works alongside clinical judgement, anticoagulant status and social circumstances, and does not replace local protocol.

• Consider using BIG to support decisions about observation and transfer for CT-positive mild head injury, if your hospital has adopted it
• A BIG1 classification does not remove the need for return advice and a responsible adult at home
• Anticoagulants, intoxication and an unreliable history can change the decision whatever the category
• Many BIG2 and BIG3 patients do well too, so low specificity means higher categories are a prompt for review, not a verdict
• Discuss any change to transfer policy with the neurosurgical team

Why it matters

Transfers for neurosurgical opinion are costly and often unnecessary, and this quantifies how safely a simple tool can reduce them.

Don't overread it

The pooled data come from observational validation studies, and a sensitivity near 98% still leaves some patients in the lowest category who worsen.

The statistics, in plain English

Sensitivity of 98% means that of patients who went on to deteriorate, about 98 in 100 had been classed in a higher-risk category than BIG1. Low specificity of about 13% means most patients classed above BIG1 did not deteriorate, so BIG is good at ruling low-risk patients in for discharge but poor at identifying who will worsen.

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.