Clinical Medicine Research

| Peer-Reviewed |

The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg

Received: Jan. 21, 2019    Accepted: Feb. 22, 2019    Published: Mar. 12, 2019
Views:       Downloads:

Share This Article

Abstract

During treatment in an intensive care unit (ICU), traumatic brain injury (TBI) patients sometimes suffer an increase in intracranial pressure (ICP). An increase beyond a currently unknown and to-be-determined threshold is very often life-threatening and requires intervention by the clinical staff. Because this threshold value is considered unknown, ‘conventional wisdom’ of practitioners argue it to be 20 mm Hg. No published studies include statistical methods that could supply a rigorous outcome for the threshold value. Here, we use a clustering algorithm (K-means clustering) to find three-dimensional clusters of the 984 triples of ICP, temperature and patient state index (PSI, a proxy for sedation level). The algorithm outputs three clusters and two gaps. One gap separates two clusters from a third and is almost planar, and perpendicular to the ICP axis (implying a threshold across all temperatures and all sedation levels); the other is perpendicular to the temperature axis, which terminates at the aforementioned gap. The first gap provides a statistically rigorous threshold of 13.625 mm Hg for ICP intervention. The second gap defines a threshold temperature (36.5°C). The gap between the two temperature regimes does not continue into Cluster 3, implying that the intervention threshold for ICP is independent of temperature.

DOI 10.11648/j.cmr.20190801.12
Published in Clinical Medicine Research ( Volume 8, Issue 1, January 2019 )
Page(s) 6-15
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Intracranial Pressure, Traumatic Brain Injury, Clustering Algorithms, Patient State Index, Akaike’s Information Criterion, ICP Intervention Threshold, K-means Clustering

References
[1] M. Smith, (2008) Monitoring intracranial pressure in traumatic brain injury. Anesthesia and Analgetics 106, 240–248.
[2] M. Czosnyka, and J. D. Pickard (2004) Monitoring and interpretation of intracranial pressure. Journal of Neurological Neurosurgery and Psychiatry 5, 813–821.
[3] N. Stocchetti, A. Colombo, F. Ortolano. et al. (2007) Time course of intracranial hypertension after traumatic brain injury. Journal of Neurotrauma 24, 1339–1346.
[4] S. L. Bratton, R. M. Chestnut, J. Ghajar, et al. (2007) VIII. Intracranial Pressure Thresholds. Journal of Neurotrauma 24, Suppl. 1, 55–58.
[5] E. Sorrentino, J. Dredler, M. Kaspowicz et al. (2012) Critical thresholds for cerebrovascular reactivity after traumatic brain injury. Neurocritical Care 16, 258–266.
[6] N. Stocchetti, M. Carbonara, G. Citero, A. Ercole, M. B. Skrifars, P. Smielewski, T. Zaerle, and D. Menon (2017) Severe traumatic brain injury: targeted management in the intensive care unit. Lancet Neurology 16, 452–464.
[7] F. G. Guiza, B. D. Depreitere, I. P. Piper, G. V. Van den Berghe, and G. M. Meyfroidt (2014) New look at 20 mm Hg ICP Threshold. Journal Critical Care 18 (Suppl 1), 458.
[8] N. Carney, A. M. Totten, C. O’Reilly et al. (2017) Guidelines for the management of severe traumatic brain injury. Fourth edition. Neurosurgery 80, 6–15.
[9] R. A. Weerakkody, M. Czosnyka, R. A. Trivedi, and P. Hutchinson P (2009) “Intracranial Pressure monitoring in head injury.” in “Head Injury: A Multidisciplinary Approach”, P. C. Whitfield, E. O. Thomas, F. Summers, M. Whyte, and P. J. Hutchinson, Eds. Cambridge, UK, Cambridge University Press.
[10] C. Lazaridis C et al. (2014) Patient-specific thresholds of intracranial pressure in severe traumatic brain injury. Journal of Neurosurgery 120, 893–900.
[11] T. G. Saul and T. B. Drucker (1982) Effects of intracranial pressure monitoring and aggressive treatment on mortality in severe head injury. Journal of Neurosurgery 56, 498–503.
[12] W. C. Fallis (2002) Monitoring urinary bladder temperature in the intensive care unit: state of the science. American Journal of Critical Care 11, 38–45.
[13] P. L. Purdon, A. Sampson, K. J. Pavone et al. (2015) Clinical Electronencephalography for Anesthesiologists — Part I: Background and basic signatures. Anesthesiology 123, 937–960.
[14] M. Oddo and L. A. Steiner (2016) Sedation and analgesia in the neurocritical care unit. Chapter 6 in, “Textbook of Neurocritical Care”, M. Smith, G. Citerio, and W. A. Kafka, Eds. Oxford, UK, Oxford University Press, pp. 65–77.
[15] D. R. Drover, H. J. Lemmens, E. T. Pierce, G. Plourde, G. Lloyd, E. Ornstein, L S. Prichep, R. J. Chabot, L. Gugino (2002) Patient state index: Titration of delivery and recovery from Propofol, Alfentanil, and Nitrous Oxide anesthesia. Anesthesiology 97, 82–89.
[16] S. Vacas, E. McInrue, M. A. Gropper, M. Maze, R. Zak, E. Lim, J. M. Leung (2016) The feasibility and utility of continuous sleep monitoring in critically ill patients using a portable electroencephalography monitor. Anesthesiology and Analgetics 123, 206–212.
[17] S. S. Wilks (1937) The large-sample distribution of the likelihood ratio for testing composite hypotheses. Philosophical Transactions of the Royal Society A 231, 60–62.
[18] J. P. Huelsenbreck and K. A. Crandall (1997) Phylogeny estimation and hypothesis testing using maximum likelihood. Annual Review of Ecological Systems 28, 437–466.
[19] H. Prossinger and F. L. Bookstein (2003) Statistical estimators of frontal sinus cross section ontogeny from very noisy data. Journal of Morphology 257, 1–8.
[20] D. J. C. MacKay Information Theory, Inference, and Learning Algorithms, 2004, Cambridge, UK, Cambridge University Press.
[21] M. Bishop (2006) “Pattern recognition and Machine Learning”, New York, NY, USA, 2006, Springer.
[22] P. J. Andrews, H. L. Sinclair, A. Rodriguez et al. (2015) Hypothermia for intracranial hypertension after traumatic brain injury. New England Journal of Medicine 373, 2403–24012.
[23] N. Stochetti and A. I. Maas (2014) Traumatic intracranial hypertension. New England Journal of Medicine 370, 2121–2130.
[24] Le Roux P (2016) “Intracranial pressure monitoring and measurement.” Chapter 15 In: “Translational Research in Traumatic Brain Injury.” D. Laskowitz, and G. Grant, Eds. Boca Raton (FL): CRC Press/Taylor and Francis Group, 2016.
[25] H. Akaike (1973) “Information theory as an extension of the maximum likelihood principle.” In: “Second International Symposium on Information Theory.” B. N. Petrov, and F. Csaki, Eds. Budapest, Hungary, Akademiai Kiado.
[26] Takeuchi K (1976) Distribution of informational statistics and a criterion of model fitting. Suri-Kagaku (Mathematical Sciences) 153, 12–18. (In Japanese)
[27] K. P. Burnham and D. R. Anderson (2002) “Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach.” 2nd edition. New York, NY, USA, Springer.
[28] H. Akaike (1981) Likelihood of a model and information criteria. Journal of Econometrics 16, 3–14.
[29] F. Guiza, I. Piper, G. Van den Berghe, G. Meyfroidt (2013) Novel methods to predict increased intracranial pressure during intensive care and long-term neurological outcome after traumatic brain injury: Development and validation in a multicenter dataset. Neurological and Clinical Care 41, 554–564.
[30] A. I. R. Maas, D. K. Memon, P. D. Adelson, et al. (2017) Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurology 16, 987–1048.
Cite This Article
  • APA Style

    Hermann Prossinger, Hubert Hetz, Alexandra Acimovic, Reinhard Berger, Karim Mostafa, et al. (2019). The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg. Clinical Medicine Research, 8(1), 6-15. https://doi.org/10.11648/j.cmr.20190801.12

    Copy | Download

    ACS Style

    Hermann Prossinger; Hubert Hetz; Alexandra Acimovic; Reinhard Berger; Karim Mostafa, et al. The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg. Clin. Med. Res. 2019, 8(1), 6-15. doi: 10.11648/j.cmr.20190801.12

    Copy | Download

    AMA Style

    Hermann Prossinger, Hubert Hetz, Alexandra Acimovic, Reinhard Berger, Karim Mostafa, et al. The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg. Clin Med Res. 2019;8(1):6-15. doi: 10.11648/j.cmr.20190801.12

    Copy | Download

  • @article{10.11648/j.cmr.20190801.12,
      author = {Hermann Prossinger and Hubert Hetz and Alexandra Acimovic and Reinhard Berger and Karim Mostafa and Alexander Grieb and Heinz Steltzer},
      title = {The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg},
      journal = {Clinical Medicine Research},
      volume = {8},
      number = {1},
      pages = {6-15},
      doi = {10.11648/j.cmr.20190801.12},
      url = {https://doi.org/10.11648/j.cmr.20190801.12},
      eprint = {https://download.sciencepg.com/pdf/10.11648.j.cmr.20190801.12},
      abstract = {During treatment in an intensive care unit (ICU), traumatic brain injury (TBI) patients sometimes suffer an increase in intracranial pressure (ICP). An increase beyond a currently unknown and to-be-determined threshold is very often life-threatening and requires intervention by the clinical staff. Because this threshold value is considered unknown, ‘conventional wisdom’ of practitioners argue it to be 20 mm Hg. No published studies include statistical methods that could supply a rigorous outcome for the threshold value. Here, we use a clustering algorithm (K-means clustering) to find three-dimensional clusters of the 984 triples of ICP, temperature and patient state index (PSI, a proxy for sedation level). The algorithm outputs three clusters and two gaps. One gap separates two clusters from a third and is almost planar, and perpendicular to the ICP axis (implying a threshold across all temperatures and all sedation levels); the other is perpendicular to the temperature axis, which terminates at the aforementioned gap. The first gap provides a statistically rigorous threshold of 13.625 mm Hg for ICP intervention. The second gap defines a threshold temperature (36.5°C). The gap between the two temperature regimes does not continue into Cluster 3, implying that the intervention threshold for ICP is independent of temperature.},
     year = {2019}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - The Intervention Threshold for Intracranial Pressure of Traumatic Brain Injury Patients Can Be Determined by Clustering Algorithms and Is Observed to Be 13 mm Hg
    AU  - Hermann Prossinger
    AU  - Hubert Hetz
    AU  - Alexandra Acimovic
    AU  - Reinhard Berger
    AU  - Karim Mostafa
    AU  - Alexander Grieb
    AU  - Heinz Steltzer
    Y1  - 2019/03/12
    PY  - 2019
    N1  - https://doi.org/10.11648/j.cmr.20190801.12
    DO  - 10.11648/j.cmr.20190801.12
    T2  - Clinical Medicine Research
    JF  - Clinical Medicine Research
    JO  - Clinical Medicine Research
    SP  - 6
    EP  - 15
    PB  - Science Publishing Group
    SN  - 2326-9057
    UR  - https://doi.org/10.11648/j.cmr.20190801.12
    AB  - During treatment in an intensive care unit (ICU), traumatic brain injury (TBI) patients sometimes suffer an increase in intracranial pressure (ICP). An increase beyond a currently unknown and to-be-determined threshold is very often life-threatening and requires intervention by the clinical staff. Because this threshold value is considered unknown, ‘conventional wisdom’ of practitioners argue it to be 20 mm Hg. No published studies include statistical methods that could supply a rigorous outcome for the threshold value. Here, we use a clustering algorithm (K-means clustering) to find three-dimensional clusters of the 984 triples of ICP, temperature and patient state index (PSI, a proxy for sedation level). The algorithm outputs three clusters and two gaps. One gap separates two clusters from a third and is almost planar, and perpendicular to the ICP axis (implying a threshold across all temperatures and all sedation levels); the other is perpendicular to the temperature axis, which terminates at the aforementioned gap. The first gap provides a statistically rigorous threshold of 13.625 mm Hg for ICP intervention. The second gap defines a threshold temperature (36.5°C). The gap between the two temperature regimes does not continue into Cluster 3, implying that the intervention threshold for ICP is independent of temperature.
    VL  - 8
    IS  - 1
    ER  - 

    Copy | Download

Author Information
  • Department of Evolutionary Anthropology, Faculty of Life Sciences, University of Vienna, Vienna, Austria

  • Department for Anaesthesiology and Intensive Medical Care, Trauma Center Vienna, Location Meidling, Vienna, Austria

  • Department for Anaesthesiology and Intensive Medical Care, Trauma Center Vienna, Location Meidling, Vienna, Austria

  • Department for Anaesthesiology and Intensive Medical Care, Trauma Center Vienna, Location Meidling, Vienna, Austria

  • Faculty of Medicine, Sigmund Freud Private University, Vienna, Austria

  • Faculty of Medicine, Sigmund Freud Private University, Vienna, Austria

  • Department for Anaesthesiology and Intensive Medical Care, Trauma Center Vienna, Location Meidling, Vienna, Austria; Faculty of Medicine, Sigmund Freud Private University, Vienna, Austria

  • Section