Maheshwari HPI

Hypotension Prediction Index for Prevention of Hypotension during Moderate- to High-risk Noncardiac Surgery

Patient / Population Intervention / Exposition Vergleich Endpunkt

In 214 adults having moderate- or high-risk non-cardiac surgery with invasive arterial pressure monitoring, Hypotension Prediction Index guidance with a treatment algorithm did not reduce intra-operative hypotension (time-weighted average MAP <65 mmHg) compared with haemodynamic management without index guidance.

N
214
Design
Pilot RCT
Endpunkt
Time-weighted average MAP <65 mmHg
Relevanz
2
ErgebnisMedian time-weighted average MAP <65 mmHg 0.14 vs 0.14 mmHg (median difference 0, 95% CI −0.03 to 0.04; P=0.757); no reduction in hypotension <60 or <55 mmHg either. 105 (49%) randomised to index guidance; half of alerts were not followed by treatment.
Maheshwari K, et al. Anesthesiology. 2020;133(6):1214-1222. 10.1097/ALN.0000000000003557
Diskussion & Kritik

The negative counterpart to HYPE, published the same year: with an alert threshold of 85 and a complex treatment algorithm, index guidance did not reduce hypotension at all. The authors' own explanation — half of alerts went untreated because of short warning time, a cumbersome algorithm or clinicians ignoring the alert — is the lasting lesson: a prediction is only as good as the response it triggers. A pilot trial, with a post hoc signal of benefit only in episodes where clinicians actually intervened.