Skip to main content

Novel models to predict elevated intracranial pressure during intensive care and long-term neurological outcome after TBI

Introduction

Elevated intracranial pressure (ICP) episodes are associated with poor outcome and should be prevented. We developed models to predict these episodes 30 minutes in advance, and to predict long-term neurological outcome by using dynamic characteristics of continuous ICP and mean arterial pressure (MAP) monitoring.

Methods

The Brain-IT [1] dataset has records for 264 patients from 22 neuro-ICUs in 11 European countries. Logistic regression and Gaussian processes (machine learning method) were used. CRASH [2] and IMPACT [3] predictors were used together with dynamic data.

Results

Predictions of elevated ICP episodes (Figure 1) were externally validated with good calibration and discrimination (AUROC 0.87). Prediction of poor neurological outcome at 6 months (GOS 1 to 2) with static data had 0.72 AUROC; adding dynamic information increased performance to 0.9 (Table 1).

Figure 1
figure1

Elevated ICP episode.

Table 1 Model performance

Conclusion

Dynamic data in continuous MAP and ICP monitoring allows prediction of elevated ICP. Adding information of the first 24 hours of ICP and MAP to known risk factors allows accurate prediction of neurological outcome at 6 months.

References

  1. 1.

    Piper I, et al: Acta Neurochir. 2003, 145: 615-628. 10.1007/s00701-003-0066-6.

    CAS  Article  PubMed  Google Scholar 

  2. 2.

    Murray GD, et al: J Neurotrauma. 2007, 24: 329-337. 10.1089/neu.2006.0035.

    Article  PubMed  Google Scholar 

  3. 3.

    CRASH Trial Collaborators: BMJ. 2008, 336: 425-429.

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to F Guiza.

Rights and permissions

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article

Guiza, F., Depreitere, B., Piper, I. et al. Novel models to predict elevated intracranial pressure during intensive care and long-term neurological outcome after TBI. Crit Care 16, P291 (2012). https://doi.org/10.1186/cc10898

Download citation

Keywords

  • European Country
  • Machine Learning
  • Arterial Pressure
  • Emergency Medicine
  • Mean Arterial Pressure