gms | German Medical Science

68th Annual Meeting of the German Society of Neurosurgery (DGNC)
7th Joint Meeting with the British Neurosurgical Society (SBNS)

German Society of Neurosurgery (DGNC)

14 - 17 May 2017, Magdeburg

Cerebral Physiology after Traumatic Brain Injury: Entropy of Monitoring Indices

Meeting Abstract

  • Erhard Lang - Neurosurgical Associates, Red Cross Hospital, Kassel, Deutschland
  • Michal Placek - Department of Biomedical Engineering, Wroclaw University of Technology, Wroclaw, Poland
  • Magdalena Kasprowicz - Department of Biomedical Engineering, Wroclaw University of Technology, Wroclaw, Poland
  • Phil Lewis - Department of Neurosurgery, The Alfred Hospital, Monash University, Melbourne, Australia
  • Ari Ercole - Department of Anaesthesiology, Addenbrooke's Hospital, Cambridge University, Cambridge, United Kingdom
  • Peter Smielewski - Academic Neurosurgery Unit, Addenbrooke's Hospital, Cambridge University, Cambridge, United Kingdom
  • Marek Czosnyka - Academic Neurosurgery Unit, Addenbrooke's Hospital, Cambridge University, Cambridge, United Kingdom

Deutsche Gesellschaft für Neurochirurgie. Society of British Neurological Surgeons. 68. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC), 7. Joint Meeting mit der Society of British Neurological Surgeons (SBNS). Magdeburg, 14.-17.05.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocDI.09.02

doi: 10.3205/17dgnc229, urn:nbn:de:0183-17dgnc2292

Published: June 9, 2017

© 2017 Lang et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at



Objective: Depleted entropy of mean intracranial pressure (ICP), measured by a grading tool titled “complexity index” (CI), calculated with multiscale entropy analysis, correlates with worse outcome after traumatic brain injury (TBI). We went on to study whether entropy of indices, describing ICP fluctuations, cerebral perfusion pressure (CPP), cerebral compensatory reserve, and cerebral autoregulation (CA) correlates with outcome ?

Methods: We analyzed continuous high resolution recordings from the Addenbrooke’s Hospital, Cambridge, UK, TBI-single center database in 313 patients. Arterial blood pressure (ABP) and ICP recordings were converted to 60-second averages of (a) ICP pulse amplitude (AMP); (b) cerebral perfusion pressure (CPP); (c) RAP, an index of cerebral compensatory reserve, measured as a correlation coefficient (R) between ICP amplitude (A) and mean ICP pressure (P); and (d) PRx, an index of cerebral autoregulation, measured as a moving correlation coefficient between MAP and ICP. These time series were subjected to multiscale entropy analysis, resulting in complexity indices: CI-AMP, CI-CPP; CI-RAP, and CI-PRx calculated as the sum of the first 20 scales of entropy. CIs were then compared to outcome at 6 months post injury, measured by the 5-point Glasgow Outcome Scale (GOS 1-good; GOS 5-dead).

Results: There was a 24% mortality in our series with a 43% favorable outcome (good & moderate recovery). Twenty-two patients (7%) had an average ICP>25mmHg. Both CI-AMP and CI-CPP decreased with worsening outcome (R= -0.18, p= 0.0018; and R= -0.16, p=0.0058, resp.). CI-RAP increased with worsening outcome (R=0.15, p=0.01). The CI-PRx outcome correlation was not significant. CI-RAP and CI-AMP were useful for distinction between favorable and unfavorable outcome (Kruskal Wallis test: p=00002, and p=0.001, resp.), while CI-PRx and CI-CPP were not. AMP, but not RAP was significantly higher in patients with ICP>25mmHg (p=0.023 vs. p=0.78, resp.)

Conclusion: Entropy of secondary cerebral monitoring parameters varies. Preservation of the complexity of ICP amplitude fluctuations and CPP entropy are positive prognostic TBI markers. Entropy of cerebral compensatory reserve, however, is a negative prognostic marker. Entropy of CA is unrelated to outcome. Further studies should investigate if these findings may yield early monitoring signs of deterioration for individual patients, which may lead to targeted therapy adjustments, and eventually better outcomes.