Skip to main content
2024 AANS Annual Scientific Meeting Main banner
Back

Trauma/Critical Care Rapid Fire Abstract Session

Trauma

Comparison of cerebral perfusion pressure monitor output values and raw data from high-frequency recordings

Saturday, May 4, 2024
1:53 PM – 1:55 PM CT
Location:  S405a

    Presenting Author(s)

  • CB

    Carl Benjamin Blümel, MD

    Neurosurgeon
    Department of Neurosurgery, Community Hospital Herdecke, Germany
    Herdecke, DE

    • Co-Author(s)

  • DC

    Dirk Cysarz, n/a

    Professor
    University of Witten/Herdecke, Germany, United States

  • BP

    Benjamin Petzold, n/a

    Scientific Assistant
    Department of Neurosurgery, Community Hospital Herdecke, Germany, United States

  • CR

    Christa Raak, n/a

    Scientific Assistant
    University of Witten/Herdecke, Germany, United States

  • WS

    Wolfram Scharbrodt, n/a

    Head of Neurosurgery, Professor
    Department of Neurosurgery, Community Hospital Herdecke, University of Witten/Herdecke, Germany, United States

Disclosure(s):
Carl Benjamin Blümel, MD: No financial relationships to disclose
Introduction: In intensive care, the processing of the underlying raw data is not standardized. Post-processing of the data in monitor systems by averaging and/or filtering may lead to significant deviations. The purpose of this study is to compare monitor output values and raw data for CPP and to examine the suitability of conventional values to serve as a basis for clinical decisions.

Methods: 39 patients from our ICU with ICH who had a complete 60-minute recording from 2 a.m. on the first day after probe implantation were included. ICP, and BP were collected as raw data at 100 Hz and automatically stored on a scientific server system.
Similarly, the CPP output from the Draeger Infinity Delta XL system was recorded. The starting point of each ICP and MAP signal for all 39 data sets was identified and synchronized in time. MAP and ICP were determined for each heartbeat. The area integrals of the MAP and ICP pressure curves were calculated for this purpose and each was divided by the time of the cardiac cycle (MAParea and ICParea). The CPParea was then calculated and compared with the monitor output data.

Results: The Shapiro-Wilk test showed that the data of CPParea and CPPdraeger do not follow a Gaussian normal distribution. The significance level was calculated using the Wilcoxon sign rank test and medians. There is a significant difference for each data set. CPPdraeger was higher in 24 and lower in 15 patients compared to CPParea. The difference in medians ranges from -43mmHg to +54mmHg. CPPdraeger is significantly lower in amplitude than CPParea. This may result from the use of filters and/or averaging.

Conclusion : The CPParea values calculated from raw data deviate from the output monitor data to varying degrees. The question arises to what extent the values available at the patient's bedside are suitable as a basis for therapeutic decisions. Treatment guidelines based on non-standardized data collection should be reviewed.