A MAP alarm at 72 mmHg was non-inferior to the Hypotension Prediction Index

Design

Single-centre, blinded, randomised non-inferiority trial

Population

143 adults having moderate- or high-risk elective non-cardiac surgery with arterial monitoring

Primary outcome

Area under MAP 65 mmHg threshold

Effect

Median 3.75 vs 4.00 mmHg·min; log mean difference 0.03 (-0.24 to 0.29), non-inferior

In this single-centre, blinded randomised trial, 143 adults having moderate- or high-risk elective non-cardiac surgery with continuous arterial monitoring were allocated to a MAP alarm at below 72 mmHg or an HPI alarm above 85.

Median area under the MAP 65 mmHg threshold was 3.75 mmHg·min with the MAP alarm and 4.00 with HPI. The log-transformed mean difference was 0.03 (95% CI -0.24 to 0.29), within the pre-set non-inferiority margin of -0.4. Hypotension and hypertension metrics, alarm duration, kidney and myocardial injury, 30-day mortality, length of stay, vasoactive drugs and fluids did not differ.

The HPI requires proprietary monitoring and correlates strongly with MAP itself. This trial shows that the benefit can be achieved with an alarm every arterial monitor already has.

The practical implication is that departments without HPI-capable monitoring are not offering inferior blood pressure care if they set an earlier MAP alarm.

• Set a MAP alarm at about 72 mmHg for patients at risk of intraoperative hypotension.
• Treat when the alarm sounds, before MAP falls below 65 mmHg.
• Reserve HPI platforms for settings where their other features justify the cost.
• Clinical outcomes were not powered; this trial measured hypotension burden.

Why it matters

It shows the earlier trigger, not the proprietary index, is what prevents intraoperative hypotension.

Don't overread it

The trial was single-centre with 143 patients and was not powered for kidney injury, myocardial injury or mortality.

The statistics, in plain English

Non-inferiority means the MAP alarm was shown to be no worse than HPI by more than a pre-agreed margin. Hypotension burden was skewed, so it was log-transformed before comparing groups; the confidence interval stayed well inside the margin.

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.