ROLE OF MOISTURE CONTENT IN MECHANICAL CHARACTERIZATION OF BRAIN TISSUE HENRY W. HASLACH, JR. DEPARTMENT OF MECHANICAL ENGINEERING CENTER for ENERGETICS CONCEPTS DEVELOPMENT UNIVERSITY OF MARYLAND COLLEGE PARK, MD 20742 USA 1
INTRODUCTION A. RAT BRAIN SERVES AS AN ANIMAL MODEL FOR mtbi B. FOR A FEM, CHARACTERIZE THE MECHANICAL RESPONSE C. SEEK MECHANICAL CAUSE OF mtbi D. INCLUDE MOISTURE CONTENT OF THE TISSUE 1. QUESTION NEGLECTED BY MOST OTHER RESEARCHERS 2
ROLE OF MOISTURE A. ROLE OF MOISTURE CONTENT IS A KEY TO UNDERSTANDING THE DAMAGE CAUSED BY SHOCK WAVES 1. MOISTURE CONTRIBUTES TO THE VISCOELASTIC (TIME DEPENDENT) MECHANICALRESPONSE TO LOADS 2. SHOCK WAVES MAY MOVE INTERNAL TISSUE MOISTURE 3. CAN EFFECT GLASS TRANSITION OF TISSUE PROTEINS B. WE MEASURED MOISTURE CONTENT BY WEIGHT AS 81.5% 1. BRAIN TISSUE DENSITY IS ABOUT 104g/cc 1.04 g/cc. 2. MOISTURE CONTENT BY VOLUME IS ABOUT THE SAME AS MOISTURE CONTENT BY WEIGHT 3
EVAPORATION TEST A. HOW SOON AFTER HARVEST MUST THE TEST BE PERFORMED IF SPECIMEN IS NOT KEPT IN PBS TO REHYDRATE B. AMBIENT 30% RELATIVE HUMIDITY 25.00% 20.00% 15.00% Experiment 1: Left Hem. Experiment 2: Right Hem. Experiment 3: Left hemisphere "Experiment 4: Right hemisphere" 10.00% 00% 5.00% 0.00% 0 20 40 60 80 100 120 140 160 180 200 4
RESULTS IN THE BIOMECHANICS LITERATURE A. PROPOSED MATHEMATICAL MODELS 1. LINEAR VISCOELASTIC (SUPERPOSITION HOLDS) 2. STRESS STRAIN TESTS SHOW a. BRAIN TISSUE IS NONLINEAR VISCOELASTIC e. g. MILLER AND CHINZEI 2002 b. AD HOC NONLINEAR VISCOELASTIC BASED ON RUBBER 3. IGNORE ROLE OF MOISTURE CONTENT B. EXPERIMENTAL 1. INDENTERS FOR LOAD DEFORMATION RELATION a. CAN ONLY GIVE LINEAR VISCOELASTIC 5
2. HOPKINSON BAR FOR HIGH STRAIN RATES a. ONLY GIVES COMPRESSION AND ONLY AT CONSTANT RATES Zhang et al., 2011 6
SMALL SIZED SIZED SAMPLE TESTS A. DETERMINE STRESS STRAIN CURVES IN COMPRESSION AND IN SHEAR 1. SUCH LOADS BELIEVED RELATED TO mtbi 2. CONFINED COMPRESSION TESTS 3. TRANSLATIONAL SHEAR TESTS B. MEASURE PERMEABILITY TO ACCOUNT FOR MOISTURE ROLE C. FIT WITH NEW NONLINEAR VISCOELASTIC MODEL 7
I. CONFINED COMPRESSION TEST A. METHOD TO DETERMINE UNIAXIAL PERMEABILITY B. APPARATUS SPECIMEN IN CYLINDER 0.25 filter specimen filter Z Plunger h water flow 8
PROPERTIES TO MEASURE A. UNDER LINEARLY INCREASING COMPRESSION, HOW DOES RATE INFLUENCE LOAD RESPONSE B. IS TISSUE STRAIN HARDENING OR SOFTENING? 1. REFLECTED IN CONCAVITY OF LOAD CURVE 2. INERTIA OF INTERCELLULAR WATER OR PORE CLOSING? C. PERMEABILITY AS A FUNCTION OF STRAIN AND STRAIN RATE 1. MEASURED BY FITTING SOLUTION TO THE BIPHASIC PDE TO A STRESS RELAXATION CURVE 2. PERMEABILITY INCREASES IF STRAIN SOFTENING 9
QUASI STATIC STATIC CONFINED COMPRESSION A. STRAIN RATE IS 6.4 X 10-4 /s (16 HRS), INNER SAGITTAL SLICE B. DAMAGE APPARENT C. STRAIN SOFTENING brain40511e 5000-0.22-0.17-0.12-0.07-0.02 0-5000 stress (Pa) -10000-15000 -20000-25000 strain -30000 10
QUASI STATIC STATIC UNCONFINEDCOMPRESSION COMPRESSION A. MILLER CHINZEI: SWINE CORTEX 30 mm DIA, 10 mm HIGH AT 6.4 X 10 4 /s IS STRAIN HARDENING B. DUE TO MULTI AXIAL MOISTURE FLOW? RADIAL EXPANSION OBSERVED Miller02 brain test 4c 50 0-0.35-0.3-0.25-0.2-0.15-0.1-0.05-50 0-100 tress (Pa) st -150-200 -250-300 -350-400 strain -450 11
EVIDENCE OF DAMAGE AND MOISTURE FLOW A. STRAIN 5%, THEN STRESS RELAX REPEAT CYCLES FOR STRAIN TO 10, 15, 20, 25% B. PLATEAU AT END OF LINEARLY -40 load (g) 10-10 -20-30 brain41211a 0 0 1000 2000 3000 4000 5000 6000 INCREASING DEFORMATION REGION TO 10% C. DAMAGE IN PLATEAU REGION IS TYPICAL OF DRYING (AT PLUNGER) AND SHIFTING OF SOLID MATERIAL -20 D. STRESS RELAXATIONREGION, -30 WATER REDISTRIBUTES -40-50 -60 E. PATTERN REPEATS AT ALL STRAIN -70 LEVELS 10,15,20 15 25 % -80 load (g) -50-60 10 sample number brain32211b 0 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000-10 sample number 12
LOAD SEGMENT OF LOAD RELAXATION A. STRAIN SOFTENING IN COMPRESSION (0.001 mm/s) B. FROM 0.2 TO 0.29 GLOBAL STRAIN, LENGTH = 3mm Brain 121710g 0-10 0 100 200 300 400 500 600 700-20 -30 load (g) -40-50 -60-70 -80-90 -100 sample number (displacement) 13
SHEAR FIXTURE 14
QUASI STATIC STATIC TEST A. STRAIN RATE IS 3.2 X 10-4 /s (16 HOUR TEST) B. INNER SAGITTAL SLICE: 6x6x4 mm shear51011sq 2500 2000 1500 stress (Pa a) 1000 500 0-0.025 0 0.025 0.05 0.075 0.1 0.125 0.15 0.175 0.2 0.225-500 strain C. FAILURE AT 10% - DUE TO DEHYDRATION? 15
ARTERY TISSUE A. THE AORTA IS A LOAD BEARING MATERIAL B. SOFTENS WHEN SHEARED PARALLEL TO COLLAGEN FIBERS AORTA - CIRCUMFERENTIAL 8000 7000 6000 STRESS (Pa) 5000 4000 3000 2000 1000 0-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7-1000 STRAIN 16
LOAD RELAXATION LOAD RELAXATION A. RELAXATION ALSO OCCURS IN SHEAR B. 1mm/s RATE, 12x6x3mm, INNER SAGITTAL SLICE shear50311b 0.5 0-1000 -0.5 1000 3000 5000 7000 9000 11000 13000 15000 17000 19000 load (g) -1-1.5-2 -2.5-3 -3.5-4 sample number 17
HALF SINE DISPLACEMENT. shear51011m A. SINUSOIDAL RAMP TO INVESTIGATE VARIABLE STRAIN RATE stress 90 80 70 60 50 40 30 TOP: 0.1 mm/s, 7x6x3 mm, 10 INNER SAGITTAL SLICE 0 20 0.95 1 1.05 1.1 1.15 1.2 1.25 stretch 1. SLIP WITHIN SPECIMEN, LARGE DAMAGE AT 10 % BOTTOM: 10 mm/s, 5x6x3 mm INNER SAGITTAL SLICE stress (Pa a) shear51011i 100 90 80 70 60 50 40 30 20 10 0 0.98-10 1 1.02 1.04 1.06 1.08 1.1 1.12 1.14 1.16 stretch 18
NONLINEAR NON-EQUILIBRIUM PROCESSES A. BRAIN TISSUE MATHEMATICAL MODEL CANNOT BE AD HOC B. BASED ON NONLINEAR THERMODYNAMICS OF SOLIDS C. DYNAMIC RESPONSE ORGANIZED BY EQUILIBRIUM STATES 1. THIS IS WHY QUASI-STATIC RELATIONS NEEDED D. SUPPLEMENTS THE SECOND LAW BY A MAXIMUM DISSIPATION PRINCIPAL TO GIVE PROCESS DIRECTION 19
THERMODYNAMIC MODEL E. THERMODYNAMIC MODULUS, k, DEFINES SPEED F. LONG-TERM HYPERELASTIC STRAIN ENERGY DENSITY WHERE x IS STATE VARIABLE (e.g. STRESS) AND y IS CONTROL VARIABLE (e.g. STRAIN) G. GIVES A UNIQUE NON-EQULIBRIUM EVOLUTION EQUATION 1. CONTRAST TO CONTINUUM THERMODYNAMICS CLAUSIUS- DUHEM INEQUALITY WHICH DOES NOT DETERMINE THE RESPONSE 2. THE SMALLER k, THE MORE VISCOUS 20
POSSIBLE PRIMARY MECHANICAL CAUSE OF mtbi A. AXON STRAIN (WHITE TISSUE) DIFFUSE AXONAL INJURY (DAI) 1. DYNAMIC STRETCH OF AXON MAY INDUCE SWELLING BULBS ALONG THE AXON 2. BULBS ARE FOCUS OF DEGENERATION (SMITH AND MEANEY, 2000) 21
THRESHOLD CRITERIA FOR mtbi? A. SMALL STRAINS KNOWN TO REDUCE AXON CONDUCTIVITY B. SWELLINGS IN RAT SPINAL NERVE ROOTS ACCUMULATE AMYLOID PRECURSER PROTEIN ( APP), A MARKER FOR IMPAIRED TRANSPORT (SINGH ET AL., 2006). 1. COMPLETE IMPAIRMENT AT 9% STRAIN AT 15 mm/s C. DAI DUE TO INERTIAL LATERAL ROTATIONAL LOAD 1. 5 10% STRAIN COMPLETE IMPAIRMENT (MARGULIES AND THIBAULT, 1992) D. EVEN SMALLER STRAINS CAUSE DAMAGE 22
HYPOTHESIS A. INTERCELLULAR MOISTURE FLOW FROM PRESSURE DEFORMS AXON TRACTS 1. KEY: MECHANICAL RESPONSE OF SUBSTRUCTURES B. LEADS TO AXON STRAIN AND DAMAGE 1. SECONDARY BIOCHEMICAL CHANGES C. IN RAT, TEST HIPPOCAMPUS AND CORPUS CALLOSUM FOR AXON MOTION UNDER HIGH STRAIN RATE D. CHANGE IN OSMOTIC PRESSURE MAY CAUSE CELL DEATH 23
SOME OPEN QUESTIONS I. HOW ARE MECHANICAL PROPERTIES OF BRAIN TISSUE RELATED TO mtbi? A. THREE CAUSES OF PRIMARY MECHANICAL INJURY 1. SHOCK WAVE, 2. IMPACT, 3. INERTIAL ACCELERATION B. HOW DO LOCAL FLUID PRESSURE, STRAIN, STRAIN RATE, VARIABLE STRAIN RATE AFFECT PERMEABILITY? C. HOW IS STRAIN SOFTENING RELATED TO mtbi? D. HOW DOES PERMEABILITY RELATE TO STRESS WAVE SPEED AND AMPLITUDE IN TISSUE? E. HOW DO LOCAL INTERCELLULAR FLUID PRESSURE AND FLOW RELATE TO AXON DAMAGE? F. DOES SHOCK WAVE DEHYDRATE LOCAL REGIONS? 24
OPEN QUESTIONS II. HOW CAN A FINITE ELEMENT MODEL PREDICT mtbi? A. WHAT LEVEL OF SUBSTRUCTURES MUST BE INCLUDED IN A FEM? B. IS THERE REALLY A THRESHOLD CRITERION? 1. OR JUST A THRESHOLD AT WHICH mtbi IS IMMEDIATELY NOTICEABLE? C. FEM MUST ACCOUNT FOR INTERCELLULAR FLUID BEHAVIOR 25
RAT BRAIN SCHEMATIC From http://learn.genetics.utah.edu/content/addiction/genetics/images/brains_large.jpg 25