mricp_overview_v3_oc..

Noninvasive Measurement of Intracranial Pressure by MRI (MR-ICP)

Overview

Noam Alperin, PhD

Physiologic Imaging and Modeling Lab Department of RadiologyUniversity of Illinois at Chicago

Importance

Normal brain function requires regulation of cerebral blood flow (CBF) and intracranial pressure (ICP). In many neurological problems this regulation is disrupted.

Our lab is developing noninvasive method to quantify these important physiological parameters by MRI. The method could potentially become an important diagnostic test for over million patients annually in the US alone, who suffer from related neurological problems (e.g., strokes, intracranial hemorrhages head injuries, hydrocephalus, Chiari malformations, etc…)

ICP Measurement by MRIICP Measurement by MRI

The MRI-based method (MR-ICP) integrates knowledge from human neuro-physiology, principles of fluid dynamics, and dynamic MRI to non-invasively measure ICP. These are briefly described in the following slides.

Monitoring

Diagnostic test

Blood and CSF Flows through the Craniospinal Compartments

The pulsatile cerebral blood flow drives the cerebrospinal fluid (CSF) flow. During systole, arterial inflow exceeds venous outflow, which result in CSF outflow1. Intracranial compliance and pressure are calculated from measurements of these flows.

1. Alperin et al, Magn Reson, in Med. 1996

The Neurophysiology BasisIntracranial pressure and volume are exponentially related.

Thus, at low ICP, a small increase in volume (dV) will cause a small increase in pressure (dP).

At high ICP, the same small volume increase (dV) will cause a large increase in pressure (dP).

The ratio of dP/dV (elastance) is a linear function of ICP.

MR-ICP measures dP and dV using CSF and blood volumetric flow rate measurements.

Intracranial Pressure-Volume Curve

Elastance = dP/dV

MR-ICP is the noninvasive analogous of the bolus pressure-volume infusion test (Marmarou 1978)

Pulsatile cerebral blood flow causes a momentary increase in intracranial volume (dV) during systole

dV causes the pulse pressure (dP) dV/dP is intracranial compliance The inverse of compliance is a linear function of

mean ICP. dV and dP are measured by MR imaging of blood

and CSF flow2

The Principles of MR-ICP

2. Alperin et. al. Radiology. 2000; 217 (3); 877–885.

The systolic increase in intracranial volume (ICV) during each cardiac cycle is derived from momentary difference between volumes of blood, and CSF that entering and leaving the cranium during the cardiac cycle

How is dV Measured?

ICV

CSF out and in

Arterialblood in

Venousblood out

Phase contrast MRI scan with high velocity encoding provides velocity images from which blood flow to and from the cranium is calculated.

How is Volumetric Blood Flow Measured?

Velocity image

Blood Flow Dynamics

Volumetric Blood Flow Measurement

Int. Carotid art.

Vertebral art.

Jugular v.

Volumetric Flow = Velocity X Lumen area

3. Alperin N, Lee S. Magn. Reson. in Med. 49:934–944 (2003)

Lumen area is identified automatically using the Pulsatility Based Segmentation (PUBS) Method 3

Volumetric Arterial Inflow and Venous Outflow (in mL/min)

0

200

400

600

800

1000

1200

1400

0 200 400 600 800 1000

time (msec)

flo

w (

mL

/min

)

a

v

Total CBF = is the cycle-averaged arterial inflow

One cardiac cycle

Cord

CSF

How is CSF Flow Measured?

Phase contrast MRI scan with low velocity encoding provides images of CSF velocities from which CSF flow to and from the cranium is calculated.

Velocity images

Systole Diastole

Cervical CSF and Epidural Venous Flow Dynamics

CSF Volumetric Flow

-150

-100

-50

0

50

100

150

200

0 200 400 600 800 1000

time (msec)

flo

w (

mL

/min

)

csf

One cardiac cycle

Calculating Systolic ICV Change (∆ ICV)

-250

0

250

500

0 200 400 600 800 1000Time (msec)

Flo

w (

mL

/min

) Net transcranialflow

-250

0

250

500

0 200 400 600 800 1000Time (msec)

Flo

w (

mL

/min

) Net transcranialflow

∆ ICV waveform

-0.2

0

0.2

0.4

0.6

0.8

0 200 400 600 800 1000Time (msec)

volu

me

(mL)

∆ICV= 0.5 mL

ICV

A - V – CSF

∫dt

-200

0

200400

600

800

1000

1200

1400

0 200 400 600 800 1000

Time (msec)

Flo

w (m

L/m

in)

csf

a

v

The MR-ICP Provides a Patient’s Specific Cranio-Spinal Flow Dynamics

● The pressure change during the cardiac cycle is derived from CSF pressure gradient (∇ p), i.e., the pressure difference that causes the CSF to flow out from and back into the cranium.4

● The Navier-Stokes equation is used to calculate CSF pressure gradients from the CSF velocities.

ρ ( dv/dt + v • ∇ v ) − µ ∇2 v = − ∇ p

inertial force viscous losses

How is dP Measured?

4 Loth FM, Yardimici MA, Alperin N. Jour. of Biomechanical Engineering. 2001, Vol. 123, pp. 71-79.

CSF Pressure Gradient Waveform

%SD ofPTPis <7%

-0.02

-0.01

0

0.01

0.02

0.03

40 240 440 640 840 1040

Time (msec)

Pre

ssur

e G

rdie

nt (m

mH

g/cm

)

Invasive ICP Recording and Corresponding MRI Derived Pressure Gradients in Humans

¯ ]"�

¯ "�

¯ ¬"�

¯ S"�

¯ ú"�¯

À I�

À ñ�

À ˜�

À ?�

À ç� �

À Ž� �

(

�¡

ù¡ a¡

…¡

1¡

š¡ ¡

Time (ms)

Pre

ssu

re G

rad

ien

t (m

mH

g/c

m)

Low ICP

Elevated ICP

red = viscous term

-0.05

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

0.05

0.06

41

12

6

21

1

29

5

38

0

46

5

55

0

63

5

Time (ms)

Pre

ss

ure

/Le

ng

th (

mm

Hg

/cm

)

Validation of dP Measurement

HumanBaboon

Validations were done with large nonhuman primates because fluid dynamics principles are scale-dependent.

0

0.002

0.004

0.006

0.008

0.01

0.012

0 2 4 6 8 10

PTP Pressure (mmHg))

Pre

ssu

re g

rad

ien

t (m

mH

g/c

m)

Initial Validation of MR-ICP in Humans

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

4 8 12 16 20 24 28

ICP (mmHg) (EVD)

Ela

sta

nc

e I

nd

ex .

Alperin et. al. Radiology. 2000; 217 (3); 877–885.

The MR-ICP method was validated in patients with SAH who had an EVD at the time of their MRI scan.

What is Normal MR-ICP ?

0

5

10

15

20

25

1 to3

3 to5

5 to7

7 to9

9 to11

11to13

13to15

15to17

17to19

ICP (mmHg)

# o

f m

easu

rem

ents

MR-ICP in Healthy Subjects (71 measurements in 23 subjects)

Mean: 9.6 mmHg ± 31%Range: 3.5 to 17.1 mmHg

ICP in 1033 Normal Human Subjects (from Merritt and Fremont-Smith 1937)

0

30

60

90

120

150

180

2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00

Pressure (mmHg)

Num

ber o

f Cas

es

Invasive

Reduction to Practice

Total MRI scan time with the current technique is approximately 3 minutes.

A software tool is being developed to facilitate quantitation of the blood and CSF volumetric flow rates, from which total CBF, compliance, and ICP are determined within few minutes.

The automated lumen segmentation tool is based on the PUBS technique.3

3. Alperin N, Lee S. Magn. Reson. in Med. (2003)

• Noam Alperin, PhD• Aaron Lee, MS• Naresh Yallapragada, MS • Hasan Dhoondia, MS• Anusha Sivaramakrishnana

Collaborators:• Terry Lichtor, MD, PhD - Neurosurgery, Cook County• Roberta Glick, MD - Neurosurgery, Cook County• Francis Loth, PhD - Mechanical Eng., Uni. of Illinois

Physiologic Imaging and Modeling Lab, Dept. of RadiologyUniversity of Illinois at Chicago

RSNA 2003

Current and previous lab members who contributed to this work