Estimation of Subject Specific ICP Dynamic Models Using Prospective Clinical Data Biomedicine 2005, Bologna, Italy

1
Estimation of Subject Specific ICP Dynamic Models
Using Prospective Clinical Data Biomedicine
2005, Bologna, Italy

• W. Wakeland 1,2, J. Fusion 1, B. Goldstein 3
• 1 Systems Science Ph.D. Program, Portland State
  University, Portland, Oregon, USA
• 2 Biomedical Signal Processing Laboratory,
  Department of Electrical and Computer
  Engineering, Portland State University, Portland,
  Oregon, USA
• 3 Complex Systems Laboratory, Doernbecher
  Childrens Hospital, Division of Pediatric
  Critical Care, Oregon Health Science
  University, Portland, Oregon, USA

This work was supported in part by the Thrasher
Research Fund
2
Aim

• To develop tools for improving care of children
  with severe traumatic brain injury (TBI)
• Help improve diagnosis and treatment of elevated
  intracranial pressure (ICP)
• Improve long-term outcome following severe TBI
• One potential approach
• Create subject-specific computer models of ICP
  dynamics
• Use models to evaluate therapeutic options

3
Motivation

• TBI is the leading cause of death and disability
  in children
• 150,000 pediatric brain injuries
• 7,000 deaths annually (50 of all childhood
  deaths)
• 29,000 children with new, permanent disabilities
• Death rate for severe TBI (defined as a Glasgow
  Coma Scale score lt 8) remains between 30-45 at
  major children's hospitals
• A recently published evidence-based medicine
  review reports that elevated ICP is a primary
  determinant of outcome following TBI

4
Background Intracranial Pressure (ICP)

• TBI often causes ICP to increase
• Frequently due, at least initially, to internal
  bleeding (hematoma)
• Elevated ICP is defined as gt 20 mmHg
• Persistent elevated ICP ? reduced blood flow ?
  insufficient tissue perfusion (ischemia) ?
  secondary injury ? poor outcome
• Poor outcomes often occur despite the
  availability of many treatment options
• The pathophysiology is complex and only partially
  understood

5
Background Treatment Options

• Treatment options include, among many others
• Draining cerebral spinal fluid (CSF) via a
  ventriculostomy catheter
• Raising the head-of-bed (HOB) elevation to 30? to
  promote jugular venous drainage
• Inducing mild hyperventilation

6
Background ICP Dynamic Modeling

• Many computer models of ICP have been developed
  over the past 30 years
• Models have sophisticated logic (differential
  eqns.)
• Potentially very helpful in a clinical setting
• However, clinical impact of models has been
  minimal
• Complex models are difficult to understand and
  use
• Another issue is that clinical data often lack
  the annotations needed to facilitate modeling
• Exact timing for medications, CSF drainage,
  ventilator adjustments, etc.

7
Method Research Approach

• Use an experiment protocol (next slide) to
  collect prospective clinical data
• Physiologic signals recorded continuously
• electrocardiogram, respiration, arterial blood
  pressure, ICP, oxygen saturation
• Plus annotations to indicate the precise timing
  of therapies and physiologic challenges
• Use collected data to create subject-specific
  computer models of ICP dynamics
• Use subject-specific models to predict patient
  response to treatment and challenges

8
Method Experimental Protocol

• Mild physiologic challenges
• Applied over multiple iterations to three
  subjects with severe traumatic brain injury
• Change the angle of the head of the bed (HOB)
• Randomly assigned, between 0º and 40º, in 10º
  increments, for 10 minute intervals
• Change minute ventilation (or respiration rate,
  RR)
• Clinician adjusts RR to achieve specified ETCO2
  target from -3 to -4 mmHg to 3 to 4 mmHg
  from baseline

9
Method Model Estimation
HOB and RR Challenges
Initial Parameters
Nonlinear Optimizing Algorithm
ICP DynamicModel
Estimated Parameters
Error
Predicted ICP
Error Computation
Measured ICP
10
Method Simulink ICP Dynamic Model
11
Method Model, Core Logic

• The timing for physiologic challenges is a key
  input to the model
• The state variables are the volumes of each fluid
  compartment
• Key feedback loops
• Volume ?pressure ? flow ? volume
• ? (volumes) ? ICP ? pressures ? flows ? ?
  (volumes)
• Autoregulation is modeled by changing
  arterial-to-capillary flow resistance only

12
Method Model, Impact of Challenges

• Impact of HOB angle (?) on ICP

• Impact of RR on ICP

13
Method Parameters Estimated

• Autoregulation factor
• Basal cranial volume
• CSF drainage rate
• Hematoma increase rate
• ? pressure time constant
• ETCO2 time constant
• Smooth muscle gain constant
• Systemic venous pressure

14
Results Patient 1, Session 4. A series of
changes to HOB elevation and RR
15
Results Patient 2, Session 1. A series of
changes to HOB elevation
16
Results Patient 2, Session 4. A series of
changes to RR
17
Results Patient 2, Session 7. A series of
changes to HOB elevation and RR
18
Results Summary
19
Discussion Model vs. Actual Response

• Model response to HOB changes was very similar to
  actual response (error lt 1 mmHg)
• Response to RR changes did not fully reflect the
  patients actual response in all cases
• Error gt 2 mmHg in many cases
• Revealed several model deficiencies
• Lack of systemic adaptation
• Does not capture interaction affects
• Incorrect response to RR changes

20
Discussion Model Deficiencies

• Systemic adaptation (make change return to
  baseline)
• P2S7 When HOB moved from 30º to 0º then back to
  30º, the ending in vivo ICP was lower than its
  starting point
• In the model, ICP returned to its original value
• Interaction of interventions
• ICP impact depended on whether the interventions
  were temporally clustered or dispersed
• Model did not capture these differences
• Incorrect model response to RR changes
• Changes in smooth muscle tone in the model affect
  the arterial-to-capillary blood flow resistance,
  but not directly the arterial volume

21
Discussion Summary

• Model of ICP dynamics was calibrated to replicate
  the ICP recorded from specifics patient during an
  experimental protocol
• Results demonstrated the potential for using
  clinically annotated prospective data to create
  subject-specific computer simulation models
• Future research will focus on improving the logic
  for cerebral autoregulatory mechanisms and
  physiologic adaptation