Neuroanesthesia & neurocritical care
ICP wave morphology as a screening test to exclude intracranial hypertension in brain-injured patients: a non-invasive perspective
Authors' conclusion
To conclude, the P2/P1 ratio of the noninvasive ICP waveform showed in this study a high Negative Predictive Value and Likelihood Ratio in different acute neurological conditions to rule out IH.
Study context
- Population
- Adults with acute brain injury requiring invasive ICP monitoring, moderate to high probability of intracranial hypertension, 2 Brazilian centres, 2017–2020111 monitorizations
- Participants
- 69 adults
- Comparison
- Noninvasive ICP waveform P2/P1 ratio vs invasive ICP monitoring
- Outcome
- Discriminatory power of P2/P1 ratio for intracranial hypertension (AUC, NPV, LR-)
Methodological detailsretrospective analysis of a prospective study, multicentre, Recruitment August 2017 to March 2020, Measurements simultaneous 5 min after external ventricular drain (EVD) drainage closure; noninvasive monitoring ≥10 min per session
- Retrospective analysis of a prospective study
- Multicentre
- Recruitment August 2017 to March 2020
- Measurements simultaneous 5 min after EVD drainage closure; noninvasive monitoring ≥10 min per session
Performance
As reported in the abstract
Discrimination of IH
| Test or model | Cut-off | Sensitivity | Specificity | NPV | AUC |
|---|---|---|---|---|---|
| P2/P1 ratio | 1.13 | 93% | 60% | 97% | 0.83 |
P2/P1 ratio
Cut-off1.13
Sensitivity93%
Specificity60%
NPV97%
AUC0.83
- Negative likelihood ratio (LR-) for intracranial hypertension (IH) at 1.13 cut-off
- 0.11
- Approximate reduction in condition probability after a negative test
- 45%
- Mean P2/P1 ratio, no IH vs IH population (p < 0.01)
- 1.01 vs 1.32
Further findings
Mean P2/P1 ratio, no IH vs IH
- Mean P2/P1 in sample
- 1.12
- Mean P2/P1, no IH
- 1.01
- Mean P2/P1, IH (p < 0.01)
- 1.32
Original publication
ICP wave morphology as a screening test to exclude intracranial hypertension in brain-injured patients: a non-invasive perspective. J Clin Monit Comput. Published 14 February 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →