Docstoc

Lessons from clinical hyperthermia in brain tissue.pdf

Document Sample
Lessons from clinical hyperthermia in brain tissue.pdf Powered By Docstoc
					Lessons from clinical hyperthermia in
brain tissue
Nathan McDannold



Harvard Medical School, Brigham & Women’s Hospital, Boston, MA 
FUS in the brain: getting through the skull
  •   Hemisphere array to distribute energy over large area
  •   Lower frequency US
  •   Active cooling of scalp
  •   Phased array transducer to correct for skull-induced aberrations




                                   Treatment device ExAblate 3000
                                   InSightec Ltd., Haifa, Israel
Treatment plan: Live MRI, CT, prior contrast MRI
Transcranial MRgFUS: Focusing possible


Maximum power:
Patient 1: 600 W
(conservative power
setting)
Patient 2: 800 W
(max available)
Patient 3: 600 W            Patient 2      Patient 2
(pain)

Max temperature
achieved during a 20s
sonication was
approximately 51°C




                           Patient 1       Patient 3

McDannold et al, Neurosurgery (in press)
Transcranial MRgFUS: Skull-induced heating



 Skull-induced heating measured on
 average of 3-5 images where
 temperature stabilized


 Metric 1:
 Outer 2 voxels on brain surface
 Did not take into account hotspots

 Metric 2:
 Found hottest 5% of the voxels
 Expanded by ±1 voxel to reduce noise
 Conservative, likely biased by noise




McDannold et al, Neurosurgery (in press)
Transcranial MRgFUS: Focus and brain surface ΔT




Focus/Brain Surface ΔT ratio:
 4.0 ± 1.7
 10.5 ± 5.5

For peak T of 60°C:
 Expect ~39-43°C on brain
   surface




McDannold et al, Neurosurgery (in press)
MRI-based threshold study

Objectives:
1. Demonstrate MR-temperature imaging can determine
   thresholds
2. Compare predictive value of input parameters (acoustic
   power, J) and measurements (peak T, TEM43)
3. Evaluate sensitivity of MRI to histological changes
4. Determine whether we heat-induced BBB disruption can
   occur without irreversible tissue damage
5. Use advantages of FUS heating and temperature mapping to
   improve our ability to measure thresholds


McDannold et al. 2004 Magn. Res. Med.
Difficulties in standard in vivo threshold
estimate experiments
  • Aligning heat source with temperature probe
  • Temperature probe interference with energy
    deposition and vice versa
  • Accurately correlating damage effects to probe
    location
  • With high temperatures, region of interest is at
    periphery of hot spot
        High thermal gradients
        Indirect tissue effects of adjacent tissue
        Tissue swelling
  • Expense of many repetitions
MR temperature imaging + phased array transducer

 • Tailor temperature profile
 • Map entire heating field
 • Easily treat multiple locations

                                      Simple phased
                                      array
MR temperature imaging + phased array transducer

                                    Temperature and Dose Profiles


• Can use single measurement
  at center of heated zone
• Flat temperature profile allows
  for precise temperature
  estimation
MRI Findings         CE-T1WI   T2WI




   0 to 4 h:
   CE-T1WI:    YES
   T2WI:       NO




   0 to 4 h:
   CE-T1WI:    NO
   T2WI:       YES

                                mm
MRI Findings         CE-T1WI   T2WI




   0 to 4 h:
   CE-T1WI:    YES
   T2WI:       NO




   48 h:
   CE-T1WI:    NO
   T2WI:       YES

                                      mm
MRI Findings      T1WI   CE-T1WI   T2WI   T2WI




0 to 4 h:
T1WI:       NO
CE-T1WI:    YES
T2WI:       YES




48 h:
T1WI:       YES
CE-T1WI:    YES
T2WI:       YES
                                          mm
MRI Findings      T1WI   CE-T1WI   T2WI   T2WI




0 to 4 h:
T1WI:       NO
CE-T1WI:    YES
T2WI:       YES




48 h:
T1WI:       YES
CE-T1WI:    YES
T2WI:       YES
                                                 mm
MRI Thermometry Findings




  Damage



    No
  Damage
           1                 10               100
                 Accumulated thermal dose (eq. min at 43°C)




McDannold et al. 2004 Magn. Res. Med.
MRI Thermometry Findings
                            Uncertain region:
                        12.3-40.1 eq. min at 43°C


  Damage



    No
  Damage
           1                 10               100
                 Accumulated thermal dose (eq. min at 43°C)




McDannold et al. 2004 Magn. Res. Med.
Damage
                                        LD50: 17.5 eq. min
                                        at 43°C
  No
Damage
         1              10                100
             Accumulated thermal dose (eq. min at 43°C)




     Probit regression              95% C.I.
                        Uncertain region:
                          48.0-50.8°C


Damage



 No
Damage
                   45                50       55
                           Max Temp (°C)




      Probit regression            95% C.I.
Damage

                                          LD50: 48.4°C
 No
Damage
                   45               50              55
                          Max Temp (°C)




      Probit regression           95% C.I.
Receiver operator characteristic curves*
                          1



                         0.8
True Positive Fraction




                                                                       Area under curve:
                         0.6

                                                                                            Az
                         0.4                                           Thermal Dose:        0.88
                                                                       Peak Temperature:    0.85
                         0.2                    Accumulated Dose       Acoustic Energy:     0.71
                                                Peak Temperature       Peak Acoustic Power: 0.65
                                                Applied Energy
                                                Peak Power
                          0
                               0   0.2    0.4      0.6   0.8       1
                                    False Positive Fraction

*fit using software (ROCKIT) courtesy C.
Metz, Univ. Chicago
                                                                               ** univariate z-score test
Lesion development
                                       5

                                      4.5

                                       4




             Diameter 2-5 days (mm)
                                      3.5

                                       3

                                      2.5

                                       2
                                                                CE-T1WI (2,5 days)
                                      1.5                       vs. CE-T1WI (0-4h)
                                                                T2WI (2,5 days)
                                       1                        vs. T2WI (0-4h)
                                      0.5                       T2WI (2-3 days)
                                                                vs. CE-T1WI (0-4h)
                                       0
                                            0   1        2      3        4           5
                                                    Diameter 0-4h (mm)
Histology
Histology

      4h                48h




Vykhodtseva et al, Ultras. Med. Biol. 2000
Conclusions:
 • BBB disruption always associated with damage
 • Thermal dose was the best for predicting tissue
   damage
 • Dose/temperature significantly better than
   applied ultrasound parameters
 • MRI sensitive to threshold level tissue damage
Discussion:
 • Results similar to prior thresholds for MRI
   changes in muscle (McDannold et al., Radiology
   2000):
     LD50: 17.6 vs. 17.5 min and 48.4 vs. 49.2°C
 • Threshold values also close to previous MRI
   studies (Chen et al, JMRI 2002; Vykhodtseva et
   al. 2000)
 • Consistent with prior hyperthermia studies
Summary:
 • MRTI + focused ultrasound provides a good platform for
   determining thermal damage thresholds
 • MRI is also sensitive to other subtle effects (flow/perfusion,
   edema, vascular damage, function) that may be investigated
 • Changes in nerve excitation and function may have lower
   thresholds than damage (see Dewhirst report)
 • However, current temperature sensitivity is ~0.5 - 1°C. This
   in principle can be lowered, but may be hindered by
   limitations of MRTI.
 • Ongoing clinical trials with MR-guided FUS and laser may
   provide more human data
Acknowledgements
Kullervo Hynynen        Clinical
Ferenc Jolesz             Ferenc Jolesz
                          Peter Black
                          Alex Golby
Acoustic models
  Greg Clement
  P. Jason White        MRI
  Jie Sun                 Nathan McDannold
  Chris Conner            Robert Mulkern
                          Larry Panych
                          Andy Chung
Histology
                          Vincenza Sorrentino
  Natalia Vykhodtseva
                        Support:
Hardware
                          NIH (R01EB003268, R33EB000705,
  InSightec
                          U41RR019703, P01CA067165)
  Doug Daum
  Jose Juste              CIMIT, FUSF
  Todd Fjield             InSightec

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:12/21/2012
language:English
pages:27
tongxiamy tongxiamy http://
About