Measuring Brain Stiffness In Chronically Shunted...
1
>1 rev’s Controls 0-1 rev’s Mark E Wagshul 1 , A. Kobets 2 , I. Rick Abbott 2 , James T. Goodrich 2 , J. Pat McAllister III 3 , and David L. Limbrick 3 1 Albert Einstein College of Medicine, 2 Department of Neurological Surgery, Albert Einstein College of Medicine/Children’s Hospital at Montefiore, Bronx NY, 3 Department of Neurological Surgery, St. Louis Children’s Hospital of Washington University, St. Louis, MO. Measuring Brain Stiffness In Chronically Shunted Hydrocephalus Patients Using MR Elastography Methods Objective 1A: Patient and control brain stiffness Figure 3: Brain Regional Differences Between Patients and Control Group: Stiffness is reduced in patients compared to controls Frontal GM : G* = 1.48± 0.14 kPa vs. 1.34 ± 0.16 kPa, p=0.00133 Occipital GM : G* = 1.28± 0.11 kPa vs. 1.11 ± 0.20 kPa, p<0.001 Parietal GM: G* = 1.30 ± 0.14 kPa vs. 1.23 ± 0.21 kPa, p=0.395 Figure 1: A. MRE vibrations are induced through spring loaded, air activated pistons, and fits inside the standard MRI head coil. It presses gently against cheekbone after the patient has been positioned in the coil. Simple 1-2 minutes to set up. B. An anatomical T1 image. C. Examples of wave amplitude maps; AP = Anterior Posterior, RL = Right Left, HF = Head Foot. Results Cont’d Conclusion & References Results Objective 1B: Regional differences in brain stiffness Introduction Objective 2A: Ventricular Size and Headache Disability Index Association with Brain Stiffness AP Wave Amplitude Ultrasound elastography used clinically but limited because of v 1) penetrate through bone (e.g. skull) v 2) high frequencies limit the penetration depth External applicators vibrate the tissue v synchronized to the MRI sequences v sensitized to microscopic tissue motion Use MRE to understand hydrocephalus Hypothesis: changes in brain elastance with hydrocephalus related to symptoms, e.g., chronic headaches [1]. Changes in elastance association with disease onset or in development overtime with shunting Causes headaches with a functioning shunt [2, 3] Goal: MRE use to investigate brain elastance role in pathophysiology and symptoms of pediatric hydrocephalus 27 shunt-dependent hydrocephalus patients (age 14-35, median age 19), shunted as infants were selected 20 healthy controls (age 8-46, median age 22) Excluded Patients: abnormally large ventricles 3T Philips MRI used MRE data acquired, inducing vibration via MRI-compatible pneumatic pistons at 30Hz, transmitted through the zygomatic arches Clinical measures: The Headache Disability Index (HDI), Hydrocephalus Outcome Questionnaire (HOQ), and other clinical data collected e.g., Beck Depression Inventory (BDI), clinical history and outcome, and shunt revision Brain elastance averaged across white and grey matter masks and within lobar regions Investigate linear associations with ventricular size, HDI and HOQ • Brain elastance is lower (softer) in chronically shunted hydrocephalus • Enlarged ventricles, as indexed by linear measures of ventricle size, are associated with softer grey and white matter • Softer brain tissue is associated with history of multiple shunt revisions Air pressure (at 30Hz) Brain motion < 100 mm Figure 2: MRE uses a cine phase contrast sequence. Multiple phase are acquired separately to reconstruct image across one vibration cycle. Typical phase sensitivity ~ 2 μm/radian. Brain stiffness was reduced in patients compared to controls Whole brain WM: G* = 1.71 ± 0.20 kPa vs. 1.93 ± 0.13 kPa, p < 0.001 Frontal GM: G* = 1.48± 0.14 kPa vs. 1.34 ± 0.16 kPa, p=0.00133 Occipital GM: G* = 1.28± 0.11 kPa vs. 1.11 ± 0.20 kPa, p<0.001 Parietal GM: G* = 1.30 ± 0.14 kPa vs. 1.23 ± 0.21 kPa, p=0.395 WM and Occipital GM stiffness was negatively correlated with ventricular size R 2 = 0.11; 0.23, p = 0.006; 0.003 respectively Weak negative correlation between parietal lobe grey matter stiffness and Headache Disability Index HDI, R 2 = 0.17, p = 0.028 Decreased stiffness with more HDI score, R 2 = 0.17, p = 0.028 Decreased brain stiffness was associated with multiple shunt revisions. C C P P Frontal Parietal Occipital Temporal Figure 2: Bar Graph of Brain Stiffness in Patients and Control Group. Brain stiffness is decreased in white matter in chronically shunted patients compared to healthy controls WM: G* = 1.71 ± 0.20 kPa vs. 1.93 ± 0.13 kPa respectively. Patients Controls G* - Stiffness (Pa) Lobar grey matter stiffness Frontal Occipital Parietal Frontal Occipital Parietal (p = 0.003) 1. Rekate, H.L., Shunt-related headaches: the slit ventricle syndromes. Childs Nerv Syst, 2008. 24(4): p. 423-30. 2. Foltz, E.L., Hydrocephalus: slit ventricles, shunt obstructions, and third ventricle shunts: a clinical study. Surgical neurology, 1993. 40(2): p. 119-124. 3. Benzel, E., et al., Slit ventricle syndrome in children: clinical presentation and treatment. Acta neurochirurgica, 1992. 117(1-2): p. 7-14. Objective 2B: Shunt revision number association to brain stiffness References: C: Parietal stiffness vs Headache Disability Index HDI. There was a negative correlation between parietal lobe grey matter stiffness and HDI (decreased stiffness with more HDI score). R 2 = 0.11 p = 0.006 R 2 = 0.17 (p = 0.028) Elastography MRI Patients (n = 28) Controls (n = 20) G* - Stiffness (Pa) Whole brain white matter stiffness p < 0.001 Figure 4: Brain Regional Stiffness Correlations to ventricular size and Headache Disability Inventory Score HDI: A & B:. Correlational Between White matter and Occipital grey matter stiffness with ventricular size represented as Total Evans Ratio and Total Occipital Horn Ratio respectively. There was a negative correlation with ventricular size (WM, Occipital R 2 = 0.11; 0.23, p = 0.006; 0.003 respectively). B. Occipital Grey Matter stiffness in the patients group with more than one revision had decreased stiffness compared to the controls and patients with less than one shunt revision. Figure 5: White Matter and Occipital Grey Matter Stiffness of Control and Patient’s Group with Different Numbers of Shunt Revisions A. White matter stiffness in the patients group with more than one revision had decreased stiffness compared to the controls and patients with less than one shunt revision. A. B B A C R 2 = 0.23 p = 0.003 A. MRE Setup B. T1 Image RL Wave Amplitude HF Wave Amplitude Phase Encode Read Encode RF Pulse Slice Select Vibration Direction (phase) = position (i.e. vibration) Spin Phase - g x 0 ∫ G(t) sin( w t) dt C. G*_Stiffness (Pa) G*_Stiffness (Pa) G*_Parietal Stiffness (Pa) Total HDI G*_White Matter Stiffness (Pa) G*_Occipital Stiffness (Pa) Total Evans Ratio Total Occipital Horn Ratio Frontal Occipital Parietal Frontal Occipital Parietal
Transcript of Measuring Brain Stiffness In Chronically Shunted...
>1 rev’sControls 0-1 rev’s
Mark E Wagshul1, A. Kobets2, I. Rick Abbott2, James T. Goodrich2, J. Pat McAllister III3, and David L. Limbrick3
1 Albert Einstein College of Medicine, 2 Department of Neurological Surgery, Albert Einstein College of Medicine/Children’s Hospital at Montefiore, Bronx NY, 3 Department of Neurological Surgery, St. Louis Children’s Hospital of Washington University, St. Louis, MO.
Measuring Brain Stiffness In Chronically Shunted Hydrocephalus Patients Using MR Elastography
MethodsObjective 1A: Patient and control brain stiffness
Figure 3: Brain Regional Differences Between Patients and Control Group: Stiffness is reduced in
patients compared to controls
Frontal GM : G* = 1.48± 0.14 kPa vs. 1.34 ± 0.16 kPa, p=0.00133
Occipital GM : G* = 1.28± 0.11 kPa vs. 1.11 ± 0.20 kPa, p<0.001
Parietal GM: G* = 1.30 ± 0.14 kPa vs. 1.23 ± 0.21 kPa, p=0.395
Figure 1: A. MRE vibrations are induced through spring loaded, air activated
pistons, and fits inside the standard MRI head coil. It presses gently against
cheekbone after the patient has been positioned in the coil. Simple 1-2 minutes
to set up.
B. An anatomical T1 image. C. Examples of wave amplitude maps; AP = Anterior
Posterior, RL = Right Left, HF = Head Foot.
Results Cont’d
Conclusion & References
Results
Objective 1B: Regional differences in brain stiffness
IntroductionObjective 2A: Ventricular Size and Headache Disability
Index Association with Brain Stiffness
AP Wave Amplitude
Ultrasound elastography used clinically but limited because ofv 1) penetrate through bone (e.g. skull)v 2) high frequencies limit the penetration depth
External applicators vibrate the tissuev synchronized to the MRI sequencesv sensitized to microscopic tissue motion
Use MRE to understand hydrocephalusHypothesis: changes in brain elastance with hydrocephalus related tosymptoms, e.g., chronic headaches [1].Changes in elastance association with disease onset or indevelopment overtime with shuntingCauses headaches with a functioning shunt [2, 3]Goal: MRE use to investigate brain elastance role in pathophysiologyand symptoms of pediatric hydrocephalus
27 shunt-dependent hydrocephalus patients (age 14-35, median age 19),
shunted as infants were selected
20 healthy controls (age 8-46, median age 22)
Excluded Patients: abnormally large ventricles
3T Philips MRI used
MRE data acquired, inducing vibration via MRI-compatible pneumatic
pistons at 30Hz, transmitted through the zygomatic arches
Clinical measures: The Headache Disability Index (HDI), Hydrocephalus
Outcome Questionnaire (HOQ), and other clinical data collected e.g., Beck
Depression Inventory (BDI), clinical history and outcome, and shunt revision
Brain elastance averaged across white and grey matter masks and within
lobar regions
Investigate linear associations with ventricular size, HDI and HOQ
• Brain elastance is lower (softer) in chronically shunted hydrocephalus
• Enlarged ventricles, as indexed by linear measures of ventricle size, are associated with softer grey and white matter
• Softer brain tissue is associated with history of multiple shunt revisions
Air pressure(at 30Hz)
Brain motion < 100 mm
Figure 2: MRE uses a cine phase contrast sequence. Multiple phase are
acquired separately to reconstruct image across one vibration cycle. Typical
phase sensitivity ~ 2 µm/radian.
Brain stiffness was reduced in patients compared to controls
Whole brain WM: G* = 1.71 ± 0.20 kPa vs. 1.93 ± 0.13 kPa, p < 0.001
Frontal GM: G* = 1.48± 0.14 kPa vs. 1.34 ± 0.16 kPa, p=0.00133
Occipital GM: G* = 1.28± 0.11 kPa vs. 1.11 ± 0.20 kPa, p<0.001
Parietal GM: G* = 1.30 ± 0.14 kPa vs. 1.23 ± 0.21 kPa, p=0.395
WM and Occipital GM stiffness was negatively correlated with ventricular size
R2 = 0.11; 0.23, p = 0.006; 0.003 respectively
Weak negative correlation between parietal lobe grey matter stiffness and
Headache Disability Index HDI, R2 = 0.17, p = 0.028
Decreased stiffness with more HDI score, R2 = 0.17, p = 0.028
Decreased brain stiffness was associated with multiple shunt revisions.
C CP
P
Frontal Parietal Occipital Temporal
Figure 2: Bar Graph of Brain Stiffness in Patients and Control Group. Brain stiffness
is decreased in white matter in chronically shunted patients compared to healthy
controls WM: G* = 1.71 ± 0.20 kPa vs. 1.93 ± 0.13 kPa respectively.
PatientsControls
G*
-Stif
fnes
s (P
a)
Lobar grey matter stiffness
Frontal Occipital Parietal
Frontal Occipital Parietal
(p = 0.003)
1. Rekate, H.L., Shunt-related headaches: the slit ventricle syndromes. Childs Nerv Syst, 2008. 24(4): p. 423-30.
2. Foltz, E.L., Hydrocephalus: slit ventricles, shunt obstructions, and third ventricle shunts: a clinical study. Surgical neurology, 1993. 40(2): p. 119-124.
3. Benzel, E., et al., Slit ventricle syndrome in children: clinical presentation and treatment. Acta neurochirurgica, 1992. 117(1-2): p. 7-14.
Objective 2B: Shunt revision number association to brain stiffness
References:
C: Parietal stiffness vs Headache
Disability Index HDI. There was a
negative correlation between
parietal lobe grey matter
stiffness and HDI (decreased
stiffness with more HDI score).
R2 = 0.11p = 0.006
R2 = 0.17(p = 0.028)
Elastography MRI
Patients(n = 28)
Controls(n = 20)
G*
-Stif
fnes
s (P
a)
Whole brain white matter stiffnessp < 0.001 Figure 4: Brain Regional Stiffness Correlations to ventricular size and Headache
Disability Inventory Score HDI:
A & B:. Correlational Between White matter and Occipital grey matter stiffness with
ventricular size represented as Total Evans Ratio and Total Occipital Horn Ratio
respectively. There was a negative correlation with ventricular size (WM, Occipital
R2 = 0.11; 0.23, p = 0.006; 0.003 respectively).
B. Occipital Grey Matter stiffness in the
patients group with more than one
revision had decreased stiffness
compared to the controls and patients
with less than one shunt revision.
Figure 5: White Matter and Occipital
Grey Matter Stiffness of Control and
Patient’s Group with Different Numbers
of Shunt Revisions
A. White matter stiffness in the patients
group with more than one revision had
decreased stiffness compared to the
controls and patients with less than one
shunt revision.
A.
B
BA
C
R2 = 0.23 p = 0.003
A. MRE Setup B. T1 Image
RL Wave Amplitude HF Wave Amplitude
Phase Encode
Read Encode
RF Pulse
Slice Select
Vibration
Direction (phase) = position (i.e. vibration)
Spin Phase - gx0 ∫ G(t) sin(wt) dt
C.
G*_
Stiff
ness
(Pa)
G*_
Stiff
ness
(Pa)
G*_
Parie
tal S
tiffn
ess
(Pa)
Total HDI
G*_
Whi
te M
atte
r St
iffne
ss (P
a)
G*_
Occi
pita
l Stif
fnes
s (P
a)
Total Evans Ratio Total Occipital Horn Ratio
Frontal Occipital Parietal Frontal Occipital Parietal
