Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2025 May 21;29(1):207.
doi: 10.1186/s13054-025-05458-9.

Cerebral perfusion pressure targets after traumatic brain injury: a reappraisal

Affiliations
Observational Study

Cerebral perfusion pressure targets after traumatic brain injury: a reappraisal

Stefan Yu Bögli et al. Crit Care. .

Abstract

Introduction: Cerebral perfusion pressure (CPP) management is central to neurocritical care management after traumatic brain injury (TBI). While the Brain Trauma Foundation recommends a target of 60-70 mmHg, it is unclear whether this range reflects the lower limit or the optimal level, when viewed through the prism of cerebrovascular autoregulation. Autoregulation aims at stabilizing cerebral blood flow and can be estimated continuously using the pressure reactivity index (PRx). Personalized CPP targets can be derived from PRx including CPPopt (optimal CPP), LLA, and ULA (lower and upper limit of autoregulation). Emerging data suggests an asymmetric relationship around CPPopt, with more pronounced autoregulation deterioration with decreasing compared to increasing CPP. Based on the hypothesis that higher CPP levels may be less harmful than lower CPP levels, we aimed to reassess rigorously the prognostic value of the different CPP targets.

Material and methods: We analyzed 809 adult TBI patients admitted from 2002 to 2023 who underwent invasive intracranial pressure monitoring and had available 6-month outcomes. CPPopt, LLA, and ULA were estimated using previously published methodologies. Deviations of CPP from fixed or personalized targets were assessed describing the overall dose, the hourly dose and the percentage time spent outside of these targets and examined using Chi-squared tests, logistic regressions, heatmaps, ordinal analyses, and group-based trajectory modelling.

Results: Our data confirms an asymmetric CPP/PRx relationship with steeper increases in PRx with decreasing and only modest elevations with increasing CPP. Even small decreases below CPPopt were consistently linked to worse outcomes (OR 1.04 (CI 1.02-1.06) and 1.09 (CI 1.04-1.15) for hourly dose and percentage time spent below CPPopt, p < 0.001 and p = 0.001 respectively). The strength of association increased with further decreases in CPP away from CPPopt towards the LLA with OR of 1.11 (CI 1.07-1.14) and 1.26 (CI 1.18-1.35) for hourly dose and percentage time spent below LLA respectively (p < 0.001). Conversely, higher-than CPPopt levels generally showed no association to worse outcomes. Distinct trajectories in the relationship between CPPopt and LLA (introduced as the Lower Limit Margin) could be identified with worse outcomes in those with decreasing distance between these targets (Mortality of 18% vs. 45% for patients with increasing vs. decreasing lower limit margins, p = 0.003).

Conclusion: Our findings corroborate experimental work suggesting that TBI patients are more vulnerable to CPP reductions below as compared to elevations above personalized thresholds. The results highlight the need for individualized CPP management strategies that prioritize avoidance of lower CPP levels and suggest using CPPopt as the lower CPP limit.

Keywords: Blood pressure targets; Cerebral perfusion pressure; Cerebrovascular autoregulation; Multimodality monitoring; Traumatic brain injury.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethical approval and consent to participate: The data was accessed through the Brain Physics database that is approved by the local ethics committee (REC 23/YH/0085). Informed consent was waived by the local ethics committee. Consent to publish: Not applicable. Competing interest: ICM + is a software licensed by Cambridge Enterprise Ltd. PS has a financial interest in a part of licensing fee; the licensing fee was waived for this study.

Figures

Fig. 1
Fig. 1
The relationship between CPP and PRx. The following figures were prepared including the full patient cohort (N = 809). The relationship between CPP and PRx are displayed considering all patients (panel A) or stratified by outcome (panel B), mortality (panel C), level of ICP (panel D, considering above vs below 20 mmHg), or age (panel E). For the age group 18, all patients aged below 20 years were considered. For all the other age groups, the age described + / − 5 years were considered (i.e. for 25 years, patients aged 20 to 29 years were included)
Fig. 2
Fig. 2
The association between CPP targets and outcome: Chi-squared tests, logistic regression, heatmaps. The following figures were prepared including the full patient cohort (N = 809). Panel A displays the results of the Chi-squared testing for PRx, CPP, and the autoregulation derived measures. The results are displayed either for the differentiation of mortality (red) or outcome (yellow). Panel B displays the results of the logistic regression analyses for the different raw or autoregulation derived CPP targets considering overall dose (Dose); hourly dose (hDose); or percentage time (ptime). The derived odds ratios are displayed for every 1000 mmHg*h Dose, 10 mmHg hDose, and 5% ptime. For the personalized CPP targets, the number behind the target represents the deviation from the cutoff (e.g. LLA 5 uses the cutoff that is 5 mmHg below the LLA). The corresponding OR and p-values are shown in Supplement D. Panel C displays the heatmaps describing the associations between deviations from LLA (left), CPPopt (middle), ULA (right) and outcome. The associations to worse outcome are colored blue, while the associations to better outcome are colored yellow with the depth corresponding to the strength of association. The most distinct increase in association can be appreciated with any time spent below the LLA. Even if sufficient time is spent close, but still above, the LLA, there is an association with worse outcome. Similarly, for CPPopt, time spent below the optimum is associated with unfavorable outcome while there seems to be a distinct safe band above the CPPopt. For ULA, only prolonged deviations above the limit were weakly associated with worse outcomes. The key aspect to appreciate in these heatmaps is the white transition zone, which indicates where the association shifts from favorable to unfavorable outcomes, as well as the consistency of this association. The strength of these associations is quantified in the formal statistical analyses
Fig. 3
Fig. 3
Trajectory Modeling. The different lower limit margin (i.e. deviation between CPPopt and LLA) trajectories identified using the optimal model, which comprised four trajectories (each characterized by cubic polynomials) are displayed as a function of time. The temporal change differs by trajectory with an early increase in the lower limit margin in Group 3 and a slightly later increase in Group 1. In Group 2, the trajectory remains largely stable, while in Group 4 there is a distinct decrease over time

Comment in

References

    1. Carney N, Totten AM, O’Reilly C, et al. Guidelines for the management of severe traumatic brain injury. Neurosurgery. 2017;80(1):6–15. - PubMed
    1. Claassen JA, Thijssen DH, Panerai RB, Faraci FM. Regulation of cerebral blood flow in humans: physiology and clinical implications of autoregulation. Physiol Rev. 2021;101(4):1487–559. - PMC - PubMed
    1. Patel PM, Drummond JC, Lemkuil B. Miller’s anesthesia. Elsevier Saunders; 2014.
    1. Brassard P, Labrecque L, Smirl JD, et al. Losing the dogmatic view of cerebral autoregulation. Physiol Rep. 2021;9(15): e14982. - PMC - PubMed
    1. Czosnyka M, Smielewski P, Kirkpatrick P, Laing RJ, Menon D, Pickard JD. Continuous assessment of the cerebral vasomotor reactivity in head injury. Neurosurgery. 1997;41(1):11–9. - PubMed

Publication types

LinkOut - more resources