Characterization of pediatric ocular material properties for implementation in finite element modeling

Abusive head trauma (AHT) is a prominent cause of death and disability in children in the United States. Retinal hemorrhage (RH) is often used to diagnose AHT, but injury mechanisms and thresholds are unknown. One goal of our research is to develop a finite element (FE) model of the human infant eye to evaluate changes in retinal stress and strain during infant head trauma. However, there are no published data characterizing agedependent material properties of ocular tissues. To characterize age and strain-rate dependent properties, we tested sclera and retina from preterm, infant, and adult sheep according to two uniaxial tensile test protocols. In general, scleral strength decreased with age, whereas no age effect was found for the retina. Sclera and retina had a stiffer elastic response when tested at higher strain-rates. Anterior sclera was stiffer than posterior sclera. In preparation to collect human tissue, viable storage techniques and postmortem time frames for material testing were determined. Pediatric scleral specimens were evaluated up to 24 hours postmortem. Retinal and scleral fresh, frozen-then-thawed, and fixed specimens were also evaluated. Adult sclera maintains its integrity for 24 hours, but immature sclera softened after 10 hours postmortem. Freezing then thawing had minimal effect on the material properties of retina and sclera suggesting this may be a suitable shipping method for the pediatric ocular tissues. The mechanical data were used to determine appropriate constitutive models for the sclera and retina. The material models were implemented into a FE model of the eye and validated against experimental ocular inflation tests. Finally, a whole model was generated to represent an infant eye subjected to shaking. Vitreoretinal interaction parameters were varied to analyze the changes in retinal stress and strain. Interaction parameters minimally affected retinal stress and strain. Overall, the equatorial retina experienced the greatest stress and strain. Stress and strain increased with the addition of shaking cycles. The anterior retina experienced greater strain than the posterior region after the first cycle and for the remaining rotation sequence. With additional refinement, these models will be valuable to investigate potential injury mechanisms of RH and potentially differentiate abuse-related RH.

CHARACTERIZATION OF PEDIATRIC OCULAR MATERIAL PROPERTIES FOR IMPLEMENTATION IN FINITE ELEMENT MODELING by Jami Marie Saffioti A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Mechanical Engineering The University of Utah August 2014 Copyright © Jami Marie Saffioti 2014 All Rights Reserved The Universi ty of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Jami Marie Saffioti has been approved by the following supervisory committee members: Brittany Coats Kenneth Monson Andrew Merryweather Kenneth L. DeVries Robert Hoffman , Chair Member , Member , Member , Member 6/9/2014 Date Approved 6/9/2014 Date Approved 6/9/2014 Date Approved 6/9/2014 Date Approved 6/10/2014 Date Approved and by Tim Ameel Chair of the Department of Mechanical Engineering and by David B. Kieda, Dean of The Graduate School. ABSTRACT Abusive head trauma (AHT) is a prominent cause of death and disability in children in the United States. Retinal hemorrhage (RH) is often used to diagnose AHT, but injury mechanisms and thresholds are unknown. One goal of our research is to develop a finite element (FE) model of the human infant eye to evaluate changes in retinal stress and strain during infant head trauma. However, there are no published data characterizing age-dependent material properties of ocular tissues. To characterize age and strain-rate dependent properties, we tested sclera and retina from preterm, infant, and adult sheep according to two uniaxial tensile test protocols. In general, scleral strength decreased with age, whereas no age effect was found for the retina. Sclera and retina had a stiffer elastic response when tested at higher strain-rates. Anterior sclera was stiffer than posterior sclera. In preparation to collect human tissue, viable storage techniques and postmortem time frames for material testing were determined. Pediatric scleral specimens were evaluated up to 24 hours postmortem. Retinal and scleral fresh, frozen-then-thawed, and fixed specimens were also evaluated. Adult sclera maintains its integrity for 24 hours, but immature sclera softened after 10 hours postmortem. Freezing then thawing had minimal effect on the material properties of retina and sclera suggesting this may be a suitable shipping method for the pediatric ocular tissues. The mechanical data were used to determine appropriate constitutive models for the sclera and retina. The material models were implemented into a FE model of the eye and validated against experimental ocular inflation tests. Finally, a whole model was generated to represent an infant eye subjected to shaking. Vitreoretinal interaction parameters were varied to analyze the changes in retinal stress and strain. Interaction parameters minimally affected retinal stress and strain. Overall, the equatorial retina experienced the greatest stress and strain. Stress and strain increased with the addition of shaking cycles. The anterior retina experienced greater strain than the posterior region after the first cycle and for the remaining rotation sequence. With additional refinement, these models will be valuable to investigate potential injury mechanisms of RH and potentially differentiate abuse-related RH. iv TABLE OF CONTENTS ABSTRACT................................................................................................................................iii ACKNOWLEDGEMENTS.................................................................................................... viii INTRODUCTION......................................................................................................................1 Abusive Head Trauma............................................................................................................. 1 Computational Modeling.........................................................................................................2 Research Objective...................................................................................................................3 Chapter Structure......................................................................................................................4 References................................................................................................................................6 CHAPTERS 1. CHARACTERIZATION OF AGE, REGION, AND STRAIN DEPENDENT MATERIAL PROPERTIES OF OVINE SCLERA............................................................... 9 1.1 Abstract..............................................................................................................................9 1.2 Introduction.....................................................................................................................10 1.3 Material and Methods..................................................................................................... 12 1.3.1. Tissue Collection and Sample Preparation..........................................................12 1.3.2. Mechanical Testing................................................................................................. 15 1.3.3. Statistical Analysis................................................................................................. 17 1.4 Results..............................................................................................................................17 1.4.1 Age............................................................................................................................ 19 1.4.2 Region ........................................................................................................................ 33 1.4.3 Strain-rate..................................................................................................................41 1.5 Discussion....................................................................................................................... 47 1.6 Conclusions.....................................................................................................................50 1.7 Acknowledgements.........................................................................................................51 1.8 References....................................................................................................................... 51 2. CHARACTERIZATION OF AGE AND STRAIN-RATE DEPENDENT MATERIAL PROPERTIES OF OVINE RETINA...................................................................................... 55 2.1 Abstract........................................................................................................................... 55 2.2 Introduction.....................................................................................................................56 2.3. Materials and Methods.................................................................................................. 57 2.3.1. Tissue Sample Preparation....................................................................................57 2.3.2. Mechanical Testing................................................................................................. 59 2.3.3. Statistical Analysis................................................................................................. 61 2.4. Results............................................................................................................................ 61 2.4.1 Age............................................................................................................................ 61 2.4.2 Strain-rate..................................................................................................................63 2.5 Discussion....................................................................................................................... 64 2.6 Conclusion ........................................................................................................................ 66 2.7 Acknowledgements.........................................................................................................66 2.8 References....................................................................................................................... 67 3. CHARACTERIZING THE EFFECT OF POSTMORTEM TIME AND STORAGE CONDITION ON MECHANICAL PROPERTIES OF IMMATURE AND MATURE OVINE SCLERA AND RETINA...........................................................................................69 3.1 Abstract........................................................................................................................... 69 3.2 Introduction.....................................................................................................................70 3.3 Materials and Methods................................................................................................... 71 3.3.1 Tissue Collection and Storage................................................................................71 3.3.2 Tissue Dissection ..................................................................................................... 72 3.3.3 Mechanical Testing.................................................................................................. 74 3.3.4 Statistical Analysis .................................................................................................. 77 3.4 Results..............................................................................................................................77 3.4.1 PMT - Immature Sclera......................................................................................... 77 3.4.2 PMT - Mature Sclera..............................................................................................78 3.4.3 Storage Condition - Immature Sclera ................................................................... 85 3.4.4 Storage Condition - Mature Sclera ....................................................................... 88 3.4.5 Storage Condition - Retina..................................................................................... 92 3.5 Discussion ........................................................................................................................ 95 3.6 Conclusions.....................................................................................................................96 3.7 Acknowledgements.........................................................................................................96 3.8 References....................................................................................................................... 97 4. MATERIAL MODEL IDENTIFICATION AND VERIFICATION............................ 98 4.1 Abstract........................................................................................................................... 98 4.2 Introduction.....................................................................................................................98 4.3 Materials and Methods................................................................................................... 99 4.3.1 Inflation Device Design...........................................................................................99 4.3.2 Ocular Specimen Preparation and Inflation........................................................101 4.3.3 Three-Dimensional Digital Image Correlation.................................................. 102 4.3.4 FE Model................................................................................................................ 103 4.4 Results........................................................................................................................... 111 4.4.1 Digital Image Correlation..................................................................................... 111 4.4.2 Finite Element M odel............................................................................................115 4.5 Discussion.....................................................................................................................121 vi 4.6 Conclusions....................................................................................................................130 4.7 Acknowledgements.......................................................................................................130 4.8 References.....................................................................................................................130 5. WHOLE EYE MODEL......................................................................................................132 5.1 Abstract......................................................................................................................... 132 5.2 Introduction...................................................................................................................133 5.3 Materials and Methods................................................................................................. 134 5.3.1 Geometry and Meshing........................................................................................ 134 5.3.2 Material Definition................................................................................................ 136 5.3.3 Boundary Conditions.............................................................................................137 5.3.4 Interaction Parameters...........................................................................................138 5.4 Results........................................................................................................................... 139 5.4.1 Interaction Effects.................................................................................................. 139 5.4.2 Multiple Shaking Cycles...................................................................................... 141 5.5 Discussion.....................................................................................................................141 5.6 Conclusions....................................................................................................................144 5.7 Acknowledgements.......................................................................................................145 5.8 References.....................................................................................................................145 CONCLUSIONS AND FUTURE WORK.......................................................................... 147 Summary of Key Findings................................................................................................. 147 Sclera Material Properties..............................................................................................147 Retina Material Properties..............................................................................................148 Effect of Postmortem Time and Storage Condition................................................... 148 Eye Inflation FE Validation............................................................................................149 Whole Eye Model............................................................................................................149 Limitations and Future Work.............................................................................................150 APPENDICES A: MATLAB CODE FOR LOADING DATA....................................................................152 B: MATLAB CODE FOR SCLERAL ANALYSIS AND PLOTTING......................... 170 C: MATLAB CODE FOR RETINAL ANALYSIS AND PLOTTING.......................... 201 D: SCLERA AND RETINA DATA..................................................................................... 216 E: DATA TABLES FOR MATERIAL MODELING AND CONVERGENCE STUDY.................................................................................................................................... 228 vii ACKNOWLEDGEMENTS This dissertation work was by no means a sole effort. I must recognize key people who have contributed a great deal of support throughout this project. First, I would like to express my gratitude for the guidance provided by my graduate advisor, Brittany Coats. Thank you, Brittany, for all of your patience and devotion to my work and success. I have matured as an engineer and developed a new work ethic under your supervision. Next, I would like to recognize my fellow peers in the Pediatric Injury Biomechanics Laboratory. It has been an honor to work alongside such a brilliant and motivated research group. Not only will I remember my lab members as amazing colleagues, but I have gained lifelong friends. Thank you, MarJanna Dahl and the rest of the Albertine Laboratory for providing assistance in obtaining all of our ocular specimens. And a huge ‘thank you' to the Knights Templar Eye Foundation for their funding support. Last, but most important, I want to acknowledge my family for the abundance of love and care they have given me throughout this program. Although they may not fully understand my research, they understood my frustration at times and never stopped encouraging me to persevere and do the best that I could. I love you, Mama, Dad, Darren and Jonathan, so very much. And of course, I love you, Joe. You are my rock and have kept me focused on the end goal of all of our hard work and efforts. INTRODUCTION Abusive Head Trauma Abusive head trauma (AHT) is a leading cause of death and disability in children in the United States.3,5,617 During diagnosis of AHT, intracranial and intraocular hemorrhages are carefully considered for their consistency with the provided medical history. Retinal hemorrhage (RH), bleeding from the blood vessels in the retina, is commonly present with AHT (Figure 1). RH injuries have been reported in 78-85% of AHT cases.13,14 However, RH has also been reported in 0-20% of accidental trauma cases14 and since the underlying injury mechanism of RH is unclear, presence of RH cannot definitely discern abuse. (a) (b) Modified image from - www. alilamedicalimages. com RH Modified image from - dontshake.org. Figure 1: Ocular anatomy and abuse-related injury. (a) The sclera is the tough, outer-protective layer of the eye. The retina is the layered tissue lining the inner surface that houses the photoreceptor cells used for vision. Immediately interior to the retina is the vitreous. (b) Retinal hemorrhage is the condition in which bleeding occurs in the retina, and is commonly associated with abusive head trauma. Etiology of RH typically identified with AHT cases is widespread, multilayered RH, with bilateral formation (RH occurring in both right and left sides).2,9'1115'20 RH from nonabusive cases are typically identified as fewer in number and unilateral.1 The exact mechanism for RH is unknown. One theory, however, is that during rapid head rotations, the vitreous, lying immediately inside and firmly attached to the retina, pulls on the retina, causing retinal traction or vitreoretinal detachment. Unfortunately, the only research to date are clinical epidemiology studies and witness accounts, which do not offer any evidence regarding the biomechanics of RH. A thorough understanding of ocular mechanics and RH injury mechanisms will be invaluable to clinical diagnoses, proper legal rulings, and prevention of repeated abuse incidences. Computational Modeling Finite element (FE) modeling may be a useful tool for analyzing the mechanical responses of the pediatric eye during traumatic events. Models may shed light on injury thresholds and mechanisms of RH, and provide data that can assist clinicians in differentiating injuries from accidental and abusive head trauma. Through computational modeling we will be able to better assess retinal stress and strain experienced during kinematic loading conditions. Currently, there are two FE models of the pediatric eye in the literature. Hans et al. generated a pediatric eye model comparing the retinal force experienced from shaking to that of an impact pulse. Their results suggest that shaking alone is capable of causing retinal stresses high enough for RH.10 The other FE model by Ranganrajan et al. had a simplified ocular geometry and was used to evaluate the influence of the vitreous and 2 extraocular fat on retinal stress and stress distribution. The prescribed angular acceleration was similar to shaking. They concluded that accurately modeling the vitreous has a significant influence on the retina, and that peak stresses occurred in the posterior retina.18 In both of these studies, ocular material property data were based on adult material properties, and the potential for mechanical changes in the developing eye was neglected. The primary reason current pediatric eye models use adult ocular material property data is due to its absence in the literature. To date, there are only two studies investigating changes in ocular properties with age. Krag et al. previously characterized the age-related mechanical differences in human anterior lens capsule from donors 7 months to 98 years old, and found a decrease in strength with age.12 Curtin et al. assessed differences in the mechanical response of premature, child (4 - 6 years old) and adult human sclera and found that the adult posterior sclera is more extensible than the younger groups. An inverse age-relation was seen for the anterior and equatorial sclera.4 These studies, and other studies conducted in our lab, indicate that there are developmental changes in many ocular tissues, and pediatric ocular tissues must be mechanically characterized in order to build accurate FE models for pediatric vision research. Research Objective In this study, we set out to develop the first pediatric eye FE model that incorporates age-appropriate material property data. Vitreous has been characterized previously in our lab, so our efforts were focused on characterizing the pediatric sclera and retina. The sclera is the strong outer layer of the eye (Figure 1). Its function is to protect the eye by providing resistance to intraocular pressure, and more importantly, retinal deformation. The retina is 3 the innermost layer of the eye and houses the light-sensitive, photoreceptor cells used for vision. The biomechanical behavior of ocular tissues is likely complex with both hyperelastic and viscoelastic material responses. Careful consideration of these characteristics must be taken into account when exploring the mechanical nature of ocular tissues through experimental testing. The ultimate goal of this dissertation research was to mechanically characterize the age-dependent material properties of the sclera and retina in order to determine appropriate constitutive models to implement into a FE model of the infant eye for assessing retinal stress and strain. To achieve this goal, we performed material property testing (Chapters 1, 2), assessed storage and testing time frames for the collection of pediatric ocular specimens (Chapter 3), developed and validated a computational model of the pediatric eye (Chapter 4), and used the data to generate an overall infant eye model to investigate the influence of vitreoretinal adhesion on retinal stress and strain (Chapter 5). Combined, these studies significantly advance the state of knowledge of pediatric ocular mechanics, and lend insight into mechanical parameters influential in predicting retinal stress and strain from repetitive head trauma. Chapter Structure Chapter 1 details the collection and preparation, mechanical testing procedures, as well as data processing and analysis of ovine scleral samples. Human pediatric ocular tissues are limited, so ovine ocular tissue were selected to evaluate a potential age, strain, and strain-rate dependent response. Ovine sclera from premature, infant (3 days - 6 weeks), and adult (> 4 years) human-equivalent age groups were tested in uniaxial tension 4 according to two testing protocols. A small strain and low strain-rate test protocol was implemented to measure the scleral response to physiologic increased intraocular pressure. A large strain and high strain-rate test protocol was implemented to measure the scleral response to trauma. To evaluate possible regional effects on the material properties, tissue was tested from the anterior and posterior regions of the sclera. Chapter 2 describes the collection, testing, and analysis of ovine retinal samples. Age and strain-rate dependent material properties were evaluated in retina from immature (0- 6 weeks) and mature (> 4 years) ovine eyes. Specimens were tested according to high and low strain-rate uniaxial tension protocols. Material property testing is ideally conducted immediately postmortem to reflect the truest physiologic mechanical response for that specimen. Human ocular tissues are difficult to obtain. They may only be available 24-72 hours postmortem, and require shipping from multiple eye banks located across the country. To date, there are no known studies assessing the effect of postmortem time (PMT) on pediatric material properties. Furthermore, it is unclear what storage/shipping parameters are suitable, if any. In preparation for the collection of human ocular specimens, we sought to determine a viable time period and shipping strategy for material testing. Therefore, in Chapter 3, we characterized the effect of PMT and storage condition on the mechanical response of sclera and retina from immature and mature ovine eyes. Sclera was tested up to 24 hours postmortem, and differences were assessed among fresh, frozen then thawed, and fixed sclera and retina. These findings will guide the mechanical testing protocols when pediatric eye tissue becomes available from human donors. In Chapter 4, the material property data detailed in Chapters 1 and 2, and measured 5 from other studies conducted in our lab were used to determine age-appropriate constitutive models for the sclera, retina, and vitreous. We then generated and validated a finite element model of the infant ovine eye by predicting scleral surface strains in a simulation of experimental ocular inflation. The model's anatomical geometry, material models, meshes, and boundary conditions were defined based on ex vivo measurements, as well as data found in previous literature. This validated model progressed the design of an entire infant eye model investigating retinal stress due to rapid head rotations. In Chapter 5, a whole ovine infant eye model was generated to simulate a traumatic shaking event. Given the likely importance of vitreoretinal (VR) adhesion in evaluating the theory of VR traction as a cause of RH, VR adhesion parameters were varied and changes in distribution and magnitude of retinal stress and strain were compared. This is the first immature eye model to incorporate age-dependent mechanical properties, and serves to more closely approximate the retinal stress and strain due to repetitive head rotations compared to existing models. Additional refinement of the model will result in an advanced tool to provide insight into injury mechanisms and prediction of RH. Chapter 6 summarizes the key findings of this research, as well as limitations and suggestions for future work. References 1. Betchel, K., Stoessel, K., Leventhal, J.M., Ogle, E., Teague, S.L., Banyas, B., Allen, K., Dziura, J., Duncan, C., Characteristics that distinguish accidental from abusive injury in hospitalized young children with head trauma. Pediatrics, 2004. 114(1): p. 165-168. 2. Binnenbaum, G., Mirza-George, N., Christian, C.W., Forbes, B.J. Odds of abuse associated with retinal hemorrhages in children suspected of child abuse. Journal of the American Association for Pediatric Ophthalmology and Stribismus, 2009. 13(3): p. 268-272. 6 7 3. Brenner, R.A., Overpeck, M.D., Trumble, A.C., DerSimonian, R., Berendes, H. Deaths attributable to injuries in infants, United States, 1983-1991, Pediatrics, 1999. 103(5 Pt 1): p. 968-974. 4. Curtin, B.J., Physiopathologic Aspects of Scleral Stress-Strain. Tr. Am. Ophth. Soc., 1969. 67: p. 417-461. 5. Duhaime, A.C., Gennarelli, T.A., Thibault, L.F., Bruce, D.A., Margulies, S.S>, Wiser, R. The shaken baby syndrome: a clinical, pathological, and biomechanical study. Journal of Neurosurgery, 1987. 66(3): p. 409-415. 6. Duhaime, A.C., Christian, C.W., Rorke, L.B., Zimmerman, R.A. Nonaccidental head injury in infants-the "shaken baby syndrome". The New England Journal of Medicine, 1998. 338(25): p. 1822-1829. 7. Gilles, E.E., McGregor, M.L., Levy-Clarke, G. Retinal hemorrhage symmetry in inflicted head injury: a clue to pathogenesis? Journal of Pediatrics, 2003. 143(4): p. 494-499. 8. Girard, M.J.A., Suh, J.-K.F., Hart, R.T., Burgoyne, C.F., Downs, J.C., Effects of storage time on the mechanical properties of rabbit peripapillary sclera after enucleation. Current Eye Research, 2007. 32(5): p. 465-470. 9. Green, M.A., Lieberman, G., Milroy, C.M., Parsons, M.A. Ocular and cerebral trauma in non-accidental injury in infancy: underlying mechanisms and implications for paediatric practice. British Journal of Ophthalmology, 1996. 80(4): p. 282-287. 10. Hans, S.A., Bawab, S.Y., Woodhouse, M.L., A finite element infant eye model to investigate retinal forces in shaken baby syndrome. Graefe's Archives for Clinical and Experimental Ophthalmology, 2009. 247(4): p. 561-571. 11. Kivlin, J.D. Manifestations of the shalen baby syndrome. Curr Opin Ophthalmology, 2001. 12(3): p. 158-163. 12. Krag, S., Olsen, T., Andreassen, T. Biomechanical characteristics of the human anterior lens capsule in relation to age. Investigative Ophthalmology & Visual Science, 1997. 38(2): p.357-363. 13. Levin, A.V., Retinal Hemorrhage in Abusive Head Trauma. Pediatrics, 2010. 126(5): p. 961-970.Levin, A.V., Christian, C.W. The eye examination in the evaluation of child abuse. Pediatrics, 2010. 162(2): p. 376-380. 14. Maguire, S.W., Watts, P.O., Shaw, A.D., Holden, S., Taylor, R.H., Watkins, W.J., Mann, M.K., Tempest, V., Kemp, A.M., Retinal haemorrhages and related findings in abusive and non-abusive head trauma: a systemic review. Eye, 2012. 27(1): p. 28-36. 15. Morad, Y., Kim, Y.M., Armstrong, D.C., Huver, D., Mian, M., Levin, A.V. Correlation between retinal abnormalities and intracranial abnormalities in the shaken baby 8 syndrome. American Journal of Ophthalmology, 2002. 134(3): p. 354-359. 16. Moran, P.R. (2012) Characterization of the vitreoretinal interface and vitreous in the porcine eye as it changes with age (Master's thesis). Retrieved from http://content.lib.utah.edu/cdm/ref/collection/etd3/id/1932 17. Overpeck, M.D., Brenner, R.A., Trumble, A.C., Trifiletti, L.B., Berendes, H.W. Risk factors for infant homicide in the United States. The New England Journal of Medicine, 1998. 339(17): p. 1211-1216. 18. Rangarajan, N., Kamalakkhannan, S., Hasija, V., Shams, T., Jenny, C., Serbanescu, I., Ho, J., Rusinek, M., Levin, A., Finite element model of ocular injury in abusive head trauma. Journal of the American Association for Pediatric Ophthalmology and Stribismus, 2009. 13(4): p. 364-369. 19. Sebag, J. Ageing of the vitreous. Eye (Lond), 1987. 1(Pt 2): p. 254-262. 20. Sturm, V., Knecht, P.B., Landau, K., Menke, M.N. Rare retinal hemorrhages in translational accidental head trauma in childrens. Eye (Lond), 2009. 23(7): p. 1535­1541. CHAPTER 1 CHARACTERIZATION OF AGE, REGION, AND STRAIN-DEPENDENT MATERIAL PROPERTIES OF OVINE SCLERA 1.1 Abstract There is a paucity of infant eye material property data and as yet there are no thorough investigations characterizing the age-dependent material properties of sclera. To quantify the effect of age on the mechanical response of sclera, we tested tissue from the anterior and posterior regions of preterm, infant, and adult ovine eyes. Two strain-dependent uniaxial tensile tests were implemented to assess the mechanical response to different loading conditions. Differences were statistically tested by comparing the stress relaxation constants and material properties across age, region, and strain-rate. Young's modulus was significantly larger for preterm and infant sclera than adult sclera at high strain-rates. At low strain-rates, only the modulus of posterior sclera significantly decreased with age. The ultimate stress was also age-dependent with the adult posterior sclera having a significantly lower average ultimate stress than both the preterm and infant posterior sclera when tested at low strain-rates. Similar age-dependent trends were seen for both anterior and posterior sclera when tested at high strain-rates. Stress relaxation constants were assessed at high strain-rates and the preterm sclera experienced the highest stresses, which again decreased with age. In the region study, anterior sclera was stiffer and had higher ultimate stress than posterior sclera for all age groups tested at the low strain-rate, but only adult anterior sclera was stiffer than posterior sclera at the high strain-rate. However, at the high strain-rate, posterior sclera interestingly was stiffer than anterior sclera for the preterm and infant groups. At the high strain-rate, anterior sclera had higher stress constants than posterior sclera for all age groups. In the strain-rate study, sclera tested at the high strain-rate generally had greater elastic modulus and ultimate stress than sclera tested at the low strain-rate. The results from our region and strain-rate analyses agree with the existing literature that the anterior sclera exhibits a stiffer elastic response than posterior and that sclera is stiffer at higher strain-rates. Our trend with age, on the other hand, contrasts ophthalmic experience that adult sclera feels stiffer than pediatric. This contradiction is likely explained by the structural rigidity of sclera. The thicker adult tissue would qualitatively feel stiffer than the thinner pediatric sclera. The data herein show that there are age-related mechanical differences of ovine sclera that are age-dependent. Similar differences are likely to be found in human pediatric and adult sclera. 1.2 Introduction Finite element (FE) analysis may be a useful tool in understanding the mechanical response of the infant eye and assist in the prediction of ocular injuries from accidental or abusive head trauma. However, current FE models of the pediatric eye are based on adult material properties with little or no consideration for changes during maturation.17,26 There are limited data thoroughly characterizing the age-dependent material properties of ocular tissues. Age-related changes in the anterior lens capsule from donors 7 months to 98 years 10 old have been reported showing a decrease in strength with age.19 Curtin et al. investigated mechanical properties of human sclera which included preterm, toddler, and adult tissue.6 In this study, a static load-dependent tensile test was implemented to measure the strain response of the sclera at given stresses. However, the infant age group was not investigated and the premature tissue was only evaluated for anterior sclera. Studies in our lab indicate that there are early developmental changes in the vitreous20 which suggests that age-dependent changes in other ocular tissues must be considered. The sclera is the tough outer membrane which protects the eye and helps maintain globe shape by providing resistance to forces such as intraocular pressure. It is a major load bearing, connective tissue and is likely an essential component to most computational models of the eye. Scleral mechanics has been thoroughly characterized for the adult and elderly population with much of eye research focusing on ocular diseases such as macular degeneration and glaucoma.4,5'7'8'10'12'18'22,24'25'28'29 The posterior sclera is the thickest region, becoming noticeably thinner towards the equator of the eye and slightly thickening again near the front of the eye. Studies reporting regional differences in the mechanical properties of adult sclera have shown a stiffer anterior sclera compared to the equatorial and posterior sclera, with posterior sclera exhibiting the least stiff response. These data infer the sclera is a region-dependent material.11 Rate-dependence has been previously assessed in adult sclera and the results show that the modulus increased at higher strain-rates. 11 Direction has no significant effect on scleral mechanics and the sclera may be regarded as an isotropic material.4 Structural changes in sclera with age suggest that material properties of sclera are age-dependent, but there is a paucity of infant eye material property data in the literature to support the assertion. The scleral extracellular matrix is 11 composed predominantly of type I collagen. Elastin may guide some of the viscoelastic response of sclera, but type I collagen and glycosaminoglycans (GAGs) are said to be the most influential constituents because they act as load bearing structures and dampening mechanisms, respectively.1 GAGs are hydrophilic and thought to control tissue hydration. With age, there is a degradation of collagen and GAGs and the sclera becomes increasingly dehydrated.2 These age-related changes in the scleral extracellular matrix highlight the inadequacies of using adult material properties in infant eye computational models. For this study, we assessed region and strain-rate dependent material properties in the sclera in addition to age-related changes. Ocular specimens from premature, infant (3 days - 6 weeks), and adult (> 4 years) sheep were tested according to uniaxial tensile test protocols in which anterior and posterior regions of the sclera were subjected to either a low or high strain test. An ovine animal model was selected because its ocular anatomy closely resembles the human eye sharing common major components. The similarities in mass, geometry, and physiology in the ovine eye to the age-equivalent human eye makes this a good animal model to observe mechanical differences throughout development. The availability of animal ocular tissue allows for a more thorough evaluation of age, rate, and region dependence of material properties. 1.3 Material and Methods 1.3.1. Tissue collection and sample preparation Newborn lamb and mature sheep whole eyes were obtained immediately postmortem from nonocular studies being conducted at the University of Utah. From this group we received lamb eyes from newborns delivered prematurely (128-136 days 12 gestation) and from normal birth (~150 days gestation). Lambs were survived from 3 days up to 6 weeks and age was determined based on development rather than birth. Eyes were tested immediately (<1 hour postmortem) or stored in phosphate buffered saline (PBS) at 2°C and tested within 6 hours postmortem. Prior to testing, enucleated eyes were transferred to a petri dish for dissection. An aqueous environment of PBS was maintained throughout dissection to prevent the ocular tissues from drying out. The extraocular muscles and soft tissues were trimmed from the eye and discarded, and the optic nerve was severed at the optic nerve scleral junction. Each eye was carefully bisected sagittally into nasal and temporal halves (Figure 2a). Sclera was isolated by removing all intraocular tissues with tweezers. The resulting hemisections of sclera were placed on a cutting board. Anterior and posterior scleral samples were cut from each ocular half (Figure 2b) using a custom made dog-bone cutting die (Figure 2c). Tissue thickness was measured using an optical microscope at 1x magnification (SZX16, Olympus, Center Valley, PA). Tissue specimens were often naturally curved when cut. 13 Figure 2: Ocular dissection procedure. (a) The direction of dissection cut on an eye globe. (b) The eye was bisected and scleral samples were taken from the anterior and posterior regions of each half. (c) A custom dog bone cutting die was used to cut scleral samples. (d) Tissue samples were aligned in custom screw-driven grips. They were placed on a pair of flat tweezers. The surface tension created from the moist tissue caused the tissue to flatten (without any pressure) onto the tweezers. Note that the tweezers were not compressed at all. They merely acted as a means to transport the tissue. Three thickness measurements were taken by imaging one side of the tissue. The tissue was rotated 180 degrees to visualize the other side and an additional three thickness measurements were taken (Figure 3). The six thickness measurements were taken for each tissue sample at the center and each end of the gage length of both sides. The average of these six measurements was recorded for every sample. Width (3 mm) and gage length (6 mm) were determined by the dog-bone shape of the tissue sample. The scleral sample was aligned in custom screw-driven clamps (Figure 2d) that were fixed to a material test system (Model 5943, Instron, Norwood, MA) equipped with a 1-kN or 500-g load cell (LSB210, Futek, Irvine, CA) for high and low strain tests, respectively. 14 Figure 3: Scleral samples were placed on a thin metal sheet and turned on its side to measure tissue thickness (posterior on top, anterior on bottom). 1.3.2. Mechanical testing Scleral samples were subjected to one of two uniaxial tensile test protocols in order to quantify strain and strain-rate dependent behavior (Figure 4). A low strain protocol was implemented to characterize the mechanical response of ocular tissues under normal physiologic intraocular pressure.16 A high strain protocol was implemented to characterize the mechanical response of ocular tissues during high rate, high strain trauma. All tests were performed in an environmental bath filled with PBS at room temperature. Studies have shown significant differences in mechanical testing of sclera in different environments,4 thus we implemented the most physiologic environment we could. 1.3.2.1 Low strain-rate. Each tissue was subjected to ten cycles of preconditioning from 0 to 1% strain at a strain-rate of 0.01 s-1. Specimens were allowed to recover for 60 s and then subjected to tensile ramp to failure at 0.01 s-1. 1.3.2.2 High strain-rate. Each tissue was subjected to ten cycles of preconditioning from 0 to 5% strain at a strain-rate of 0.05 s-1. Specimens were allowed to recover for 60 s and then a stress relaxation test was performed by applying 25% strain and holding for 900 s. The tissue was allowed to recover for 60 s, and then subjected to tensile ramp to failure at 0.1 s-1. The raw load and displacement data were sampled at 10 Hz and extracted to calculate engineering stress and strain. A custom Matlab (Mathworks, Natick, MA) code was implemented for scleral analysis and plotting which can be seen in Appendices A and B. Stress was calculated by dividing the current force by the reference cross-sectional area. Strain was calculated by dividing displacement of the Instron crosshead by the original gage length. Each tissue was preloaded to approximately 0.08 N to remove any slack in 15 16 Tensile Test Protocol 100% 75% • | 50% 25% High S tra in ------- Low S tra in ------ 0.1/sec /' / / Pull-to-failure / 1 Stress-relaxation / 1 > / < < Preconditioning ; ■ . \ ./ / ' 1 / . r ------------------------ ' • / 1' 0.01/sec 20 s 6 0 s 900 s 6 0 s Time (s) Figure 4: The strain-dependent uniaxial tensile test protocol consisted of preconditioning, stress relaxation, and pull-to-failure. the tissue sample. Stress relaxation data for each specimen were fit to a two-term generalized Maxwell model [Eq.1].27 A least-squares curve-fitting technique in Matlab was used to solve for equilibrium stress (oe), intermediate stresses (01, 02), and the decay (T1, T2) constants. Instantaneous stress (oi) was defined as the sum of the equilibrium and intermediate stresses [Eq. 2]. The strain length of the toe region (stoe), elastic modulus (E), ultimate stress (oult), and ultimate strain (sult) were extracted from each pull-to-failure test. stoe was defined as the strain achieved at the end of the nonlinear elastic response during pull-to-failure. Young's modulus was defined as the slope of the linear region during pull-to- failure. The ultimate stress and strain were the maximum stress and strain achieved by the specimen. 17 1.3.3. Statistical analysis Age, region, and strain-rate were analyzed independently for this study. A one­way analysis of variance (ANOVA) was used to determine if age significantly affected tissue thickness. One-way ANOVAs were also used to determine if (1) age significantly affected the relaxation constants (n, T2, Oi, Oe, 01, 02) and material properties (stoe, E, Oult, Sult), (2) region significantly affected the relaxation constants (11, T2, Ci, Oe, 01, 02) and material properties (stoe, E, Oult, Sult), and (3) strain-rate significantly affected the material properties (stoe, E, Oult, Sult). A p-value of 0.05 was used to define significance. A Tukey's Honestly Significant Difference test with a p-value of 0.05 was used post-hoc to test for significant differences within the one-way analysis of variance. The scleral data which were analyzed in Matlab were implemented into statistical software (JMP, Cary, NC) and can be seen in Appendix D. 1.4 Results The posterior sclera was significantly thicker than anterior sclera for all age groups (p<0.05). Scleral thickness significantly increased with age (p<0.005) as the mature sclera was roughly 1.65 and 2.45 times greater than the infant and preterm sclera, respectively (Table 1). Scleral thickness was significantly different between all age groups (Figure 5). (1) °i - °e + °1 + °2 (2) 18 Table 1: Average ± standard deviation for regional scleral thickness (mm) in each age group. The number of specimens for each group is provided in parentheses. Preterm Anterior (n=32) 0.39±0.08 Posterior (n=30) 0.86±0.20 Infant Anterior (n=21) 0.54±0.16 Posterior (n=21) 1.28±0.35 Adult Anterior (n=26) 0.93±0.21 Posterior (n=26) 2.16±0.36 ■ Anterior ■ Posterior 2.5- enkc ii hT Preterm Infant Adult Figure 5: Average and standard deviation for sclera thickness of preterm, infant, and adult anterior and posterior sclera. * p<0.005 * * All scleral samples had a good overall fit to the second order Maxwell model (Figure 6). Average goodness of fits for preterm, infant, and adult specimens were 0.95, 0.98, and 0.97, respectively. In order to obtain adequate data resolution of tensile tests at low strain, the 500-g (~4.9 N) load cell was used. The ultimate force measured during the low strain-rate pull-to-failure, which immediately followed stress relaxation, occasionally exceeded the load cell limits. As a result, ultimate stress and strain were not reported for these tests. Additional pull-to-failure samples were tested to replace the removed data. 19 Figure 6: Representative curve fit of stress relaxation data to the second order Maxwell model for infant sclera. 1.4.1 Age The Young's modulus and stress constants of the preterm sclera were generally greater than the infant and adult groups. The scleral material properties at the low strain-rate can be seen in Table 2. The scleral material properties at the high strain-rate can be seen in Table 4. 1.4.1.1 Low strain-rate. The strain length of the toe region (stoe) and Young's modulus (E) of the preterm anterior sclera were generally larger than the infant and adult anterior sclera but no significant differences were found. stoe of the preterm posterior sclera was significantly longer (p<0.005) than the adult posterior sclera (Figure 7). stoe decreased with age for both anterior and posterior sclera but was only statistically significant between adult and preterm posterior sclera. The Young's modulus of the posterior sclera decreased with age and was significantly different (p<0.05) between all three age groups (Figure 8). The preterm anterior sclera generally had a higher Young's modulus and ultimate stress 20 Table 2: Average ± standard deviation of material properties for preterm, infant, and adult anterior and posterior sclera tested at low strain-rate. Similar symbols (*,f) in each row indicate groups that were significantly different than each other (p<0.05).______________ Low Strain-rate Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior Stoe 0.25 ± 0.29 0.25 ± 0.04 * 0.13 ± 0.03 0.20 ± 0.06 0.06 ± 0.03 0.13 ± 0.06 * E (MPa) 20.35 ± 20.71 17.22 ± 4.00 * 9.58 ± 4.75 9.35 ± 4.63 10.17 ± 12.52 2.49 ± 4.58 * Oult (MPa) 5.70 ±5.62 4.62 ± 1.4 * 2.09 ± 0.61 2.69 ± 1.65 t 1.81 ± 3.11 0.72 ± 1.09 *t Sult 0.46 ± 0.11 0.57 ± 0.12 0.42 ± 0.07 0.48 ± 0.16 0.53 ± 0.33 0.49 ± 0.14 than the infant and adult anterior sclera yet no statistically significant differences were found due to large variability. The adult posterior sclera had a significantly lower ultimate stress than the preterm (p<0.05) and infant (p<0. 0005) posterior sclera (Figure 9). No statistically significant differences were found for the ultimate strain of sclera tested at low strain-rate (Figure 10). The low strain-rate pull-to-failure responses for all scleral specimens can be seen in Figure 11. As mentioned earlier, not all tissues reached failure due to the limits of the low force load cell. Figure 12 includes only the scleral trials that achieved failure at the low strain-rate, and the additional pull-to-failure specimens tested. The averaged responses across age and region for trials that achieved failure can be seen in Figure 13. Stress for every age and region combination was averaged at every 0.1 mm/mm strain increment up to 1 mm/mm. Preterm sclera exhibits the stiffest response and greatest ultimate stress. Infant sclera responded similar to adult at low strain-rates. 1.4.1.2 High strain-rate. The stress relaxation constants for sclera tested at the high strain level can be seen in Table 3. All stress constants (oi, 01, 02, Oe) for the preterm posterior sclera were significantly higher than both the infant (p<0.05) and adult (p<0.001) posterior sclera. The preterm anterior sclera also experienced higher stresses than the 21 0.5 0.3 £1 £ Si £ o.i ■ Anterior ■ Posterior n=8 n=7 n=10 n=10 n=8 n=8 Preterm Infant Adult Figure 7: Average and standard deviation for toe region across age and region for sclera tested at low strain. * p<0.01 * 4 0 - ■ Anterior ■ Posterior n=8 n=7 n=10 n=10 n=8 n=8 Preterm Infant Adult Figure 8: Average and standard deviation for Young's modulus across age and region for sclera tested at low strain. * p<0.01 22 aP ■ Anterior ■ Posterior n=8 n=5 Preterm n=7 n=6 Infant n=7 n=8 Adult Figure 9: Average and standard deviation for ultimate stress across age and region for sclera tested at low strain. * p<0.005 * * 0.8 0.7 0.6 i i i i 0.5 3 CO 0.4 0.3 0.2 0.1 0.0 ■ Anterior ■ Posterior n=8 n=5 n=7 n=6 n=7 n=8 Preterm Infant Adult Figure 10: Average and standard deviation for ultimate strain across age and region for sclera tested at low strain. * p<0.05 23 Strain (nun, 111111) Figure 11: Pull-to-failure response of all included trials at low strain-rate across age and region. Strain (mm/mm) Figure 12: Pull-to-failure response at low strain-rate across age and region, excluding trials that do not fail. 24 Figure 13: Average pull-to-failure response at low strain-rate across age and region. infant and adult anterior sclera. No significant age differences were found for the anterior sclera tested at high rate (Figure 14). The preterm anterior sclera had a significantly higher immediate and long-term decay time than the adult anterior sclera (p<0.005). The immediate decay time constant of the adult posterior sclera was significantly lower than the preterm and infant posterior sclera (p<0.003). The long-term decay time of the adult posterior sclera was significantly lower than the infant posterior sclera (p<0.02) (Figure 15, Figure 16). The high strain relaxation responses can be seen in Figure 17 and the averaged responses with age and region can be seen in Figure 18. Preterm sclera experiences the highest stresses and most rapid decay rates, then infant and adult. No significant differences in the s toe of anterior sclera were found among the three ages (Figure 19). 25 Table 3: Average ± standard deviation of stress relaxation constants for preterm, infant, and adult anterior and posterior sclera tested at high strain. Similar symbols (*,f,i§) in each row indicate groups that were significantly different than each other (p<0.05).______ High Strain Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior oi (MPa) 1.78 ± 1.37 1.39 ± 0.72 *t 1.26 ± 0.65 0.61 ± 0.44 * 0.84 ± 0.61 0.14 ± 0.04 t Oi (MPa) 0.45 ± 0.39 0.33 ± 0.18 *t 0.27 ± 0.13 0.17 ± 0.11 * 0.13 ± 0.09 0.03 ± 0.01 t o2 (MPa) 0.71 ± 0.50 0.54 ± 0.25 *t 0.57 ± 0.32 0.27 ± 0.19 * 0.54 ± 0.44 0.07 ± 0.03 t o e (MPa) 0.62 ± 0.49 0.52 ± 0.31 *t 0.40 ± 0.23 0.18 ± 0.15 * 0.17 ± 0.11 0.03 ± 0.01 t T1 (sec) 309.13 ± 118.71 * t 243.93 ± 33.50 * 214.98 ± 29.51 * 237.64 ± 39.25 § 120.74 ± 90.65 t 144.43 ± 49.73 *§ t2 (sec) 12.44 ± 4.45 *t 10.60 ± 3.28 *§ 7.38 ± 2.85 * 11.56 ± 3.23 * 3.49 ± 3.54 t 6.09 ± 2.95 § n=4 n=5 n=9 n=9 n=6 n=7 Preterm Infant Adult Figure 14: Average and standard deviation for stress constants across age and region for sclera tested at high strain. The thick red lines indicate significance across all stress constants for posterior sclera. ** p<0.001 * p<0.05 T1 (sec) 26 ■ Anterior ■ Posterior * ------------1 n=4 n=5 n=8 n=9 n=6 n=6 Preterm Infant Adult Figure 15: Average and standard deviation for immediate decay time across age and region for sclera tested at high strain. * p<0.05 27 n=4 n=5 n=8 n=9 n=6 n=6 Preterm Infant Adult Figure 16: Average and standard deviation for long-term decay time across age and region for sclera tested at high strain. * p<0.05 Figure 17: Relaxation response of all scleral trials at high strain across age and region. 28 Figure 18: Averaged relaxation responses at high strain across age and region. Pull-to-failure data for sclera tested at the high strain level can be seen in Table 4. The E of the preterm anterior sclera was significantly higher than the adult anterior sclera (p<0.05). The Stoe of the preterm posterior sclera, however, was significantly shorter than the Stoe of the infant posterior sclera (p<0.03). The preterm Stoe was also smaller than adult, but a larger variation in the toe region of adult posterior sclera posterior sclera precluded significance. The E of the adult posterior sclera was significantly less (p<0.02) than both the preterm and infant posterior sclera (Figure 20). The adult anterior sclera had a significantly lower ultimate stress than both the preterm and infant anterior sclera (p<0.02). The ultimate stress of the infant posterior sclera was significantly larger (p<0.002) than the adult posterior sclera (Figure 21). No age effect was found for the ultimate strain of sclera (Figure 22). The high strain-rate, pull-to-failure responses for all scleral specimens can be seen in Figure 23 and the averaged responses with age and region can be seen in Figure 24. Stress was averaged up to 100% strain in increments of 10% strain. 29 0.35 0.25 S I £ 0.20 Si £ W 0 1 5 n=4 n=5 Preterm n=9 n=9 Infant ■ Anterior ■ Posterior n=6 n=7 Adult Figure 19: Average and standard deviation for toe region across age and region for sclera tested at high strain-rate. * p<0.05 Table 4: Average ± standard deviation of material properties for preterm, infant, and adult anterior and posterior sclera tested at high strain-rate. Similar symbols (*,f,+) in each row indicate groups that were significantly different than each other (p<0.05). High Strain-rate Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior S toe 0.27 ± 0.09 0.23 ± 0.03 * 0.26 ± 0.06 0.29 ± 0.03 * 0.24 ± 0.01 0.26 ± 0.05 E (MPa) 20.62 ± 6.30 * 21.55 ± 9.42 t 18.00 ± 5.74 19.82 ± 7.02 * 10.86 ± 5.70 * 6.36 ± 7.89 t* Oult (MPa) 4.47 ± 1.60 * 4.16 ± 1.27 * 3.74 ± 1.20 t 5.32 ± 1.58 * 1.63 ± 1.01 * t 1.68 ± 2.22 S ult 0.52 ± 0.10 0.45 ± 0.09 0.58 ± 0.13 0.59 ± 0.07 0.54 ± 0.15 0.62 ± 0.22 E (MPa) 30 30- 10- ■ Anterior ■ Posterior n=4 n=5 Preterm n=9 n=9 Infant n=6 n=7 Adult * Figure 20: Average and standard deviation for Young's modulus across age and region for sclera tested at high strain-rate. * p<0.05 31 6- PgSh D ■ Anterior ■ Posterior n=4 n=5 n=9 n=9 n=6 n=7 Preterm Infant Adult Figure 21: Average and standard deviation for ultimate stress across age and region for sclera tested at high strain-rate. * p<0.05 * * * ■ Anterior ■ Posterior n=4 n=5 n=9 n=9 n=6 n=7 Preterm Infant Adult Figure 22: Average and standard deviation for ultimate strain across age and region for sclera tested at high strain. 32 Strain (mm/mm) Figure 23: Pull-to-failure response of all included trials at high strain-rate across age and region. Figure 24: Average pull-to-failure response at high strain-rate across age and region. 33 1.4.2 Region Regional material properties can be seen in Table 5 and Table 6. 1.4.2.1 Low strain-rate. The Stoe of the preterm anterior sclera was longer than the preterm posterior sclera. The Stoe of the infant and adult posterior sclera was significantly longer (p<0.05) than the anterior sclera (Figure 25). Generally, the anterior sclera trended towards a higher Young's modulus and ultimate stress than the posterior sclera for all age groups but no statistically significant differences were found (Figure 26, Figure 27). No significant differences were found for the ultimate stress or strain of the anterior and posterior sclera of all age groups (Figure 28). Table 5: Average ± standard deviation of material properties for preterm, infant, and adult anterior and posterior sclera tested at low strain-rate. Similar symbols (*,f) in each row indicate groups that were significantly different than each other (p<0.05).______________ Low Strain-rate Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior Stoe 0.25 ± 0.29 0.25 ± 0.04 0.13 ± 0.03 * 0.20 ± 0.06 * 0.06 ± 0.03 + 0.13 ± 0.06 + E (MPa) 20.35 ± 20.71 17.22 ± 4.00 9.58 ± 4.75 9.35 ± 4.63 10.17 ± 12.52 2.49 ± 4.58 ault (MPa) 5.70 ±5.62 4.62 ± 1.4 2.09 ± 0.61 2.69 ± 1.65 1.81 ± 3.11 0.72 ± 1.09 Sult 0.46 ± 0.11 0.57 ± 0.12 0.42 ± 0.07 0.48 ± 0.16 0.53 ± 0.33 0.49 ± 0.14 Table 6: Average ± standard deviation of stress relaxation constants for preterm, infant, and adult anterior and posterior sclera tested at high strain. Similar symbols (*) in each row indicate groups that were significantly different than each other (p<0.05).______________ High Strain Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior a i (MPa) 1.78 ± 1.37 1.39 ± 0.72 1.26 ± 0.65 * 0.61 ± 0.44 * 0.84 ± 0.61 * 0.14 ± 0.04 * a! (MPa) 0.45 ± 0.39 0.33 ± 0.18 0.27 ± 0.13 * 0.17 ± 0.11 * 0.13 ± 0.09 * 0.03 ± 0.01 * a 2 (MPa) 0.71 ± 0.50 0.54 ± 0.25 0.57 ± 0.32 * 0.27 ± 0.19 * 0.54 ± 0.44 * 0.07 ± 0.03 * ae (MPa) 0.62 ± 0.49 0.52 ± 0.31 0.40 ± 0.23 * 0.18 ± 0.15 * 0.17 ± 0.11 * 0.03 ± 0.01 * ti (sec) 309.13 ± 118.71 243.93 ± 33.50 214.98 ± 29.51 237.64 ± 39.25 120.74 ± 90.65 144.43 ± 49.73 t 2 (sec) 12.44 ± 4.45 10.60 ± 3.28 7.38 ± 2.85 * 11.56 ± 3.23 * 3.49 ± 3.54 6.09 ± 2.95 34 0.5 ■ Anterior ■ Posterior 0.4- n=8 n=7 n=10 n=10 n=8 n=8 Preterm Infant Adult Figure 25: Average and standard deviation for toe region across age and region for sclera tested at low strain. * p<0.05 40- ■ Anterior ■ Posterior 30- n=8 n=7 n=10 n=10 n=8 n=8 Preterm Infant Adult Figure 26: Average and standard deviation for Young's modulus across age and region for sclera tested at low strain. 35 ■ Anterior ■ Posterior 3 4 - D 2- 0- n=8 n=5 n=7 n=6 n=7 n=8 Preterm Infant Adult Figure 27: Average and standard deviation for ultimate stress across age and region for sclera tested at low strain. 0.8 0.7 0.6 i i i i 0.5 3 CO 0.4 0.3 0.2 0.1 0.0 ■ Anterior ■ Posterior n=8 n=5 Preterm n=7 n=6 Infant n=7 n=8 Adult Figure 28: Average and standard deviation for ultimate strain across age and region for sclera tested at low strain. 1.4.2.2 High strain-rate. The pull-to-failure data measured at high strain-rate can be seen in Table 7. The preterm anterior sclera generally experienced higher stresses during relaxation than the preterm posterior sclera but no significant differences were found for any of the stress constants (oi, 01, 02, Oe). All stress constants except 01 for the infant anterior sclera were significantly higher (p<0.05) than the infant posterior sclera. All stress constants for the adult anterior sclera were significantly higher (p<0.05) than the adult posterior sclera (Figure 29). The decay time constants for the preterm anterior sclera were higher than the preterm posterior sclera but no significant differences were found (Figure 30). The decay time constants for the anterior sclera of infant and adult age groups were generally lower than the posterior sclera. The infant anterior sclera had a significantly shorter (p<0.05) long-term decay time constant than the infant posterior sclera (Figure 31). No significant differences were found for the strain length of the toe region (stoe) and Young's modulus () of the anterior and posterior sclera for all age groups (Figure 32, Figure 33). Interestingly, the preterm and infant posterior sclera were stiffer than the anterior regions when tested at the high strain-rate, while the adult anterior sclera was stiffer than the posterior region. The ultimate stress and strain of both the infant and adult anterior sclera were lower than the posterior region. The ultimate stress of the infant anterior sclera was significantly lower (p<0.05) than the infant posterior sclera (Figure 34). The ultimate stress and strain of the preterm anterior sclera were higher than the preterm posterior sclera (Figure 35). 36 37 Table 7: Average ± standard deviation of material properties for preterm, infant, and adult anterior and posterior sclera tested at high strain. Similar symbols (*) in each row indicate groups that were significantly different than each other (p<0.05).______________________ High Strain-rate Preterm Infant Adult Anterior Posterior Anterior Posterior Anterior Posterior ^toe 0.27 ± 0.09 0.23 ± 0.03 0.26 ± 0.06 0.29 ± 0.03 0.24 ± 0.01 0.26 ± 0.05 E (MPa) 20.62 ± 6.30 21.55 ± 9.42 18.00 ± 5.74 19.82 ± 7.02 10.86 ± 5.70 6.36 ± 7.89 ^ t (MPa) 4.47 ± 1.60 4.16 ± 1.27 3.74 ± 1.20 5.32 ± 1.58 1.63 ± 1.01 1.68 ± 2.22 Sult 0.52 ± 0.10 0.45 ± 0.09 0.58 ± 0.13 * 0.59 ± 0.07 * 0.54 ± 0.15 0.62 ± 0.22 ■ Anterior ■ Posterior a O a o a o a o 3.0­2.5­2.0­1.5 1.0 0.5 0.0­0.8­0.6- 0.0­1.2 1.0­0.8­0.6­0.4­0.2­0.0. 1.0 0.8­0.6­0.4­0.2­0.0- - n=4 n=5 Preterm n=9 n=9 Infant 1 n=6 n=7 Adult Figure 29: Average and standard deviation for stress constants across age and region for sclera tested at high strain. * p<0.05 38 ■ Anterior ■ Posterior n=4 n=5 n=8 n=9 n=6 n=6 Preterm Infant Adult Figure 30: Average and standard deviation for immediate decay time across age and region for sclera tested at high strain. n=4 n=5 n=8 n=9 n=6 n=6 Preterm Infant Adult Figure 31: Average and standard deviation for long-term decay time across age and region for sclera tested at high strain. * p<0.05 39 S I £ 0.20 Si £ W 015 0.05 n=4 n=5 Preterm n=9 n=9 Infant ■ Anterior ■ Posterior n=6 n=7 Adult Figure 32: Average and standard deviation for toe region across age and region for sclera tested at high strain-rate. * p<0.05 30 n=4 n=5 n=9 n=9 n=6 n=7 Preterm Infant Adult Figure 33: Average and standard deviation for Young's modulus across age and region for sclera tested at high strain-rate. 40 cd 4- Oh ■ Anterior ■ Posterior n=4 n=5 n=9 n=9 n=6 n=7 Preterm Infant Adult Figure 34: Average and standard deviation for ultimate stress across age and region for sclera tested at high strain. * p<0.05 £1 £ Si £ n=4 n=5 Preterm n=9 n=9 Infant n=6 n=7 Adult Figure 35: Average and standard deviation for ultimate strain across age and region for sclera tested at high strain. 1.4.3 Strain-rate Sclera tested at high strain-rate generally had a greater Young's modulus and experienced higher stresses (Table 8). 1.4.3.1 Preterm. The preterm sclera tested at the high strain-rate had a greater Young's modulus than sclera tested at the low strain-rate but no significant differences were found. Interestingly, the preterm sclera tested at the low strain-rate experienced higher ultimate stress than the sclera tested at the high strain-rate but no significant differences were found. Preterm posterior sclera tested at the low strain-rate generally had a longer Stoe and higher ultimate strain than posterior sclera tested at the high strain-rate. Conversely, the preterm anterior sclera tested at the low strain-rate had a shorter Stoe and lower ultimate strain that anterior sclera tested at the high strain-rate (Figure 36). 1.4.3.2 Infant. In general, all material properties of infant sclera tested at the high strain-rate were higher than sclera tested at the low strain-rate. Stoe, Young's modulus, ultimate stress, and ultimate strain for infant anterior sclera tested at high strain-rate were significantly higher than infant sclera tested at low strain-rate (p<0.05). The Stoe, Young's modulus, and ultimate stress for infant posterior sclera tested at high strain-rate were significantly higher than infant sclera tested at low strain-rate (p<0.05) (Figure 37, Figure 38). 1.4.3.3 Adult. In general, all material properties of adult sclera tested at the high strain-rate were higher than sclera tested at the low strain-rate. Stoe of the adult sclera tested at the high strain-rate was significantly longer (p<0.05) than the toe region of those tested at the low strain-rate (Figure 39). Adult sclera tested at the high strain-rate was stiffer than sclera tested at the low strain-rate but no significant differences were found (Figure 40). 41 42 Table 8: Average standard deviation of material properties for preterm, infant, and adult anterior and posterior sclera. Similar symbols (*,f) in each row indicate groups that were significantly different than each other (p<0.05)._______________ Preterm High Strain-rate Low Strain-rate Anterior Posterior Anterior Posterior Stoe 0.27 ± 0.09 0.23 ± 0.03 0.25 ± 0.29 0.25 ± 0.04 E (MPa) 20.62 ± 6.30 21.55 ± 9.42 20.35 ± 20.71 17.22 ± 4.00 Oult (MPa) 4.47 ± 1.60 4.16 ± 1.27 5.70 ±5.62 4.62 ± 1.4 Sult 0.52 ± 0.10 0.45 ± 0.09 0.46 ± 0.11 0.57 ± 0.12 Infant High Strain-rate Low Strain-rate Anterior Posterior Anterior Posterior Stoe 0.26 ± 0.06 * 0.29 ± 0.03 t 0.13 ± 0.03 * 0.20 ± 0.06 t E (MPa) 18.00 ± 5.74 * 19.82 ± 7.02 t 9.58 ± 4.75 * 9.35 ± 4.63 t Oult (MPa) 3.74 ± 1.20 * 5.32 ± 1.58 t 2.09 ± 0.61 * 2.69 ± 1.65 t Sult 0.58 ± 0.13 * 0.59 ± 0.07 0.42 ± 0.07 * 0.48 ± 0.16 Adult High Strain-rate Low Strain-rate Anterior Posterior Anterior Posterior Stoe 0.24 ± 0.01 * 0.26 ± 0.05 t 0.06 ± 0.03 * 0.13 ± 0.06 t E (MPa) 10.86 ± 5.70 6.36 ± 7.89 10.17 ± 12.52 2.49 ± 4.58 Oult (MPa) 1.63 ± 1.01 1.68 ± 2.22 1.81 ± 3.11 0.72 ± 1.09 Sult 0.54 ± 0.15 0.62 ± 0.22 0.53 ± 0.33 0.49 ± 0.14 43 ■ Anterior ■ Posterior n=8 n=5 n=4 n=5 Low S train-rate High Strain-rate Figure 36: Average and standard deviation for material properties of preterm sclera across region and strain-rate. 44 * ■ Anterior I I n=10 n=10 n=9 n=9 Low Strain-rate High Strain-rate Figure 37: Average and standard deviation for toe region and Young's modulus for infant sclera across region and strain-rate. Black line indicating significant strain-rate effect between both anterior and posterior sclera. * p<0.05 45 aP ■ Anterior ■ Posterior n=7 n=6 n=9 n=9 Low Strain-rate High Strain-rate Figure 38: Average and standard deviation for ultimate stress and strain for infant sclera across region and strain-rate. Black line indicating significant strain-rate effect between both anterior and posterior sclera. Blue line indicating significant strain-rate effect in the anterior group only. * p<0.05 * * 46 ■ Anterior ■ Posterior £| £ I £ O 0.15 CO 0.10 n=8 n=8 n=6 n=7 Low Strain-rate High Strain-rate Figure 39: Average and standard deviation for toe region of adult sclera across region and strain-rate. Black line indicating significant strain-rate effect between anterior and posterior sclera. * p<0.05 * aP aP I £ 0.4 0.2 0.0 ■ Anterior ■ Posterior n=7 n=8 n=6 n=7 Low Strain-rate High Strain-rate Figure 40: Average and standard deviation for ultimate strain of adult sclera across region and strain-rate. 1.5 Discussion Overall, the younger aged sclera had a higher Young's modulus and ultimate stress than the adult sclera. The mechanical differences with age found herein have an interesting correlation to the extracellular matrix of the developing sclera. Particularly, there is a loss in collagen and GAGs in the aging sclera.1 As aforementioned, these are the most influential constituents, acting as the load bearing structures and dampening mechanisms. This leads us to believe that the sclera should become less elastic, or stiffer with age. The sclera has been anecdotally reported to stiffen with age. Elastic modulus of the sclera from our study decreased with age. The discrepancy to this common perception is likely due to the structural rigidity of the sclera. Elastic modulus is the ratio of stress to strain and is independent of specimen. Structural rigidity is defined by the product of the Young's modulus and the moment of inertia. In our study, the scleral cross-sectional geometry can be simplified as a rectangle. Therefore, the moment of inertia is (wldth)(^ lckness ). Considering that the adult sclera thickness is 1.65 times larger than the infant, the adult anterior and posterior sclera have approximately 4.45 and 1.42 times greater structural rigidity than the infant sclera, respectively. Furthermore, studies reporting age-related "stiffening" in the sclera incorporate older age ranges than those used in our study. Our infant age was modeled with 3-day to 6-week old lambs, where the youngest groups used in other studies were 4-5 years old humans,6 6-8 month old pigs,30 or 1.5 year old monkeys.15 These immature sclera models are outside the range of our infant group as the animal models correspond to toddlers, adolescents, and even adults. We believe our animal model more appropriately represents an infant and that there are biomechanical differences in the sclera that are being 47 overlooked between the reported younger populations and our youngest age. The species-related differences also must be considered when using an animal model to characterize sclera material properties. The human sclera grows in size rapidly during the first three years and is said to have decreased cellularity and undergo a densening of the extracellular matrix.1 The sclera reaches maturity around 13-16 years1 which raises concern about the mechanical changes happening up to this age. In future studies, supplementary histology should be conducted to parallel the similarities in the ovine and human sclera to bolster the age-related changes found in our study. The infant sclera is a biphasic material and perhaps the water content trapped at this young age significantly influences the stiff response seen in our results. The stress-relaxation analysis shows that decay rate decreases with age with the preterm and infant groups exhibiting similar responses. As mentioned above, older sclera is more dehydrated making the tissue less viscous which was seen through the lower decay rates. No region-dependence was seen which can be attributed to an evenly dehydrated sclera throughout. Generally, the anterior sclera was stiffer than the posterior sclera and had higher stress values. Our regional mechanical differences coincide with structural differences within the sclera as well as published data. There is no difference between the collagen content in the anterior and posterior sclera, but there is a substantial difference in the collagen arrangement.1 The anterior sclera contains smaller, denser collagen bundles where the posterior sclera contains larger, looser collagen bundles having a "wide-angle weave".1 This agrees with our finding that the toe region of the posterior sclera was generally longer and more extensible than the toe region of the anterior sclera. There were 48 fewer regional findings for the preterm sclera which may be explained by a premature growth phase of the sclera. In embryo, the sclera grows in an anterior to posterior fashion and both regions are developed by 11 weeks gestation. During the rest of gestation, the sclera continues to thicken and the extracellular matrix densens. Perhaps there are structural changes occurring in the sclera before birth that we are unable to detect. The results for preterm anterior sclera were very variable. This may be explained by some samples not purely being cut from the anterior region and partially including equatorial (mid) sclera. During the development of our testing protocols, scleral samples were taken from the equatorial region of preterm and infant eyes. To maximize the number of samples taken from each eye, only anterior and posterior specimens were collected. Data from these few equatorial specimens suggest that differences across anterior, equatorial, and posterior regions exist and the mid-scleral region should also be explored to further understand the mechanics of the younger eye. In our study, the Young's modulus decreased with age, but was only significant in the posterior region. The preterm sclera had a higher Young's modulus than both infant and adult. The posterior sclera has a delayed growth as the anterior sclera is the first region to develop. This was seen in our significant differences only in the posterior sclera between the infant and preterm groups, while the difference between the infant and adult posterior sclera was minimal. A previous study incorporated low strain testing of human preterm, immature (4 - 6 years old) and adult sclera by subjecting specimens to a load-dependent tensile test. Weights were incrementally added to a lever arm and the stabilized displacement of the tissue was recorded. Results showed that human adult posterior sclera is more extensible than premature posterior sclera. This concurs with our findings and can 49 best be described by the relationship of stress and strain - more extensible, less stiff. Premature anterior sclera was not assessed in this work, but the results did show that adult anterior sclera was more stiff (less extensible) than the child anterior sclera. This trend does not correlate with our anterior sclera findings, but given the older age (4 - 6 years old), compared to adult there may be developmental changes. For example, as the sclera grows, one paper suggests the sclera is broken down and rebuilt. This would explain a stiffer infant eye, less stiff toddler eye, and stiff adult eye. The Young's modulus in anterior sclera compared to posterior sclera was only noticeable at low rates. There was minimal significant differences at the high rate except that the adult sclera was generally different than the younger ages. At high rates, the regional differences are not seen. Similarly, at high rates, the differences between preterm and infant sclera are not seen. Sclera tested at the higher strain-rate was generally stiffer and had higher stress values than sclera tested at the low strain-rate, which agrees with existing findings and shows that the ovine sclera is a viscoelastic material exhibiting rate-dependence under uniaxial tension. However, preterm sclera showed no significant strain-rate effects. This may be attributed to the incomplete growth of the tissue. Further analysis should focus on the developing constituents of sclera that may influence this behavior. 1.6 Conclusions Scleral elastic modulus and ultimate stress were found to decrease with age, increase with strain-rate, and be greater in the anterior region. There is a wide spread of values reported for the material property data of sclera and our results are within the bounds 50 51 of published adult sclera mechanical properties. However, there is still a gap in the literature in the quantification of developmental and mechanical changes of the sclera across a broad age range. More pediatric ocular material property research is crucial. Previous experiments have been conducted to examine the scleral strain response in a few ages across a wide range of ages (premature, 4 - 6 year old, and adults)6 but trends throughout development are still unclear. Our data are a start to characterizing the early developmental changes to sclera mechanics. These data will be used to identify an age-appropriate constitutive model for the sclera to be implemented into the first infant-specific eye FE model. 1.7 Acknowledgements We would like to thank Dr. Kurt Albertine at the University of Utah for donating ovine ocular tissue. We would also like to thank the Knights Templar Eye Foundation for sponsoring this work. 1.8 References 1. Adler's Physiology of the Eye. 11th ed. St. Louis, Missouri: Mosby; 2003. 2. Brown, C.T., Vural, M., Johnson, M., Trinkaus-Randall, V. Age-related changes in scleral hydration and sulfated glycosaminoglycans. Mechanisms of Ageing and Development, 1994. 77(): p. 97-107. 3. Cirovic, S., Bhola, R.M., Howard, I.C., Lawford, P.V., Marr, J.E., Parsons, M.A., Computer modeling of the mechanism of optic nerve injury in blunt trauma. British Journal of Ophthalmology, 2006. 90(6): p. 778-783.Chen, K., Rowley, A.P., Weliand, J.D., Elastic properties of porcine ocular posterior soft tissues. Journal of Biomedical Materials Research, 2009. 93(2): p. 634-645. 4. Chen, K., Rowley, A.P., Weliand, J.D., Elastic properties of porcine ocular posterior soft tissues. Journal of Biomedical Materials Research, 2009. 93(2): p. 634-645. 52 5. Courdillier, B., Tian, J., Alexander, S., Myers, K.M., Quigley, H.A., Nguyen, T.D., Biomechanics of the human posterior sclera: age- and glaucoma-related changes measured using inflation testing. Investigative Ophthalmology & Visual Science, 2012. 53(4): p. 1714-1728. 6. Curtin, B.J., Physiopathologic Aspects of Scleral Stress-Strain. Tr. Am. Ophth. Soc., 1969. 67: p. 417-461. 7. Downs, J.C., Suh, J-K.f., Thomas, K.A., Belleza, A.J., Hart, R.T., Burgoyne, C.F., Viscoelastic material properties of the peripapillary sclera in normal and early-glaucoma monkey eyes. Investigative Ophthalmology & Visual Science, 2005. 46(2): p. 540-546. 8. Downs, J.C., Suh, J-K.F., Thomas, K.A., Belleza, A.J., Burgoyne, C.F., Hart, R.T., Viscoelastic characterization of peripapillary sclera: material properties by quadrant in rabbit and monkey eyes. Journal of Biomechanical Engineering, 2003. 125(1): p. 124­131. 9. Downs, J.C., Burgoyne, C.F., Thomas, K.A., Thompson, H.W., Hart, R.T. Effects of strain-rate on the mechanical properties of posterior rabbit sclera. in BMES/EMBS. 1999. Atlanta, GA. 10. Eilaghu, A., Flanagan, J.G., Tertinegg, I., Simmons, C.A., Biaxial mechanical testing of human sclera. Journal of Biomechanics, 2010. 43(9): p. 1696-1701. 11. Elsheikh, A., Geraghty, B., Alhasso, D., Knappett, J., Regional variation in the biomechanical properties of the human sclera. Experimental Eye Research, 2010. 90(5): p. 624-633. 12. Fazio, M.A., Grytz, R., Bruno, L., Girard, M.J.A., Gardiner, S., Girkin, C.A., Downs, J.C., Regional variations in mechanical strain in the posterior human sclera. Investigative Ophthalmology & Visual Science, 2012. 53(9): p. 5326-5333. 13. Fazio, M.A., Grytz, R., Morris, J.S., Bruno, L., Girkin, C.A., Downs, J.C. Age-related changes in the non-linear mechanical strain response of human peripapillary sclera. in 2013 ASME Summer Bioengineering Conference. 2013. Sunriver, OR. 14. Girard, M., Suh, J-K., Bottlang, M., Burgoyne, C.F., Downs, J.C., Biomechanical changes in the sclera of monkey eyes exposed to chronic IOP elevations. Investigative Ophthalmology & Visual Science, 2011. 52(8): p. 5656-5668. 15. Girard, M., Suh, J.-K.F., Hart, R.T., Bottlang, M., Burgoyne, C.F., Downs, J.C., Scleral biomechanics in the aging monkey eye. Investigative Ophthalmology & Visual Science, 2009. 50(11): p. 5226-5237. 16. Girard, M.J.A., Downs, J.C., Burgoyne, C.F., Downs, J.C., Suh, J.K., Peripapillary and posterior scleral mechanics-part I: development of an anisotropic hyperelastic constitutive model. Journal of Biomechanical Engineering, 2009. 131(5): p. 051011. 53 17. Hans, S.A., Bawab, S.Y., Woodhouse, M.L., A finite element infant eye model to investigate retinal forces in shaken baby syndrome. Graefe's Archives for Clinical and Experimental Ophthalmology, 2009. 247(4): p. 561-571. 18. Kim, W., Argento, A., Rozsa, F.W., Mallett, K., Constitutive behavior of ocular tissues over a range of strain-rates. Journal of Biomechanical Engineering, 2012. 134(6): p. 061002. 19. Krag, S., Olsen, T., Andreassen, T. Biomechanical characteristics of the human anterior lens capsule in relation to age. Investigative Ophthalmology & Visual Science, 1997. 38(2): p.357-363. 20. Moran, P.R. (2012) Characterization of the vitreoretinal interface and vitreous in the porcine eye as it changes with age (Master's thesis). Retrieved from http://content.lib.utah.edu/cdm/ref/collection/etd3/id/1932 21. Myers, K.M., Courdillier, B., Boyce, B.L., Nguyen, T.D., The inflation response of the posterior bovine sclera. Acta Biomaterialia, 2010. 6(11): p. 4327-4335. 22. Nagase, S.Y., M., Tanaka, R., Yasui, T., Miura, M., Iwasaki, T., Goto, H., Yasuno, Y. (2013) Anisotropic Alteration of Scleral Birefringence to Uniaxial Mechanical Strain. PLoS ONE 8(3): e58716. doi:10.1371/journal.pone.0058716 23. Palko, J.R., Pan, X., Liu, J., Dynamic testing of regional viscoelastic behavior of canine sclera. Experimental Eye Research, 2011. 93(6): p. 825-832. 24. Palko, J.R., Iwabe, S., Pan, X., Agarwal, G., Komaromy, A.M., Liu, J., Biomechanical properties and correlation with collagen solubility profile in the posterior sclera of canine eyes with an ADAMTS10 mutation. Investigative Ophthalmology & Visual Science, 2013. 54(4): p. 2685-2695. 25. Pallikaris, I.G., Kymionis, G.D., Harilaos, S.G., Kounis, G.A., Tsilimbaris, M.K., Ocular rigidity in living humans. Investigative Ophthalmology & Visual Science, 2005. 46(2): p. 409-414. 26. Rangarajan, N., Kamalakkhannan, S., Hasija, V., Shams, T., Jenny, C., Serbanescu, I., Ho, J., Rusinek, M., Levin, A., Finite element model of ocular injury in abusive head trauma. Journal of the American Association for Pediatric Ophthalmology and Stribismus, 2009. 13(4): p. 364-369. 27. Rudra, R.P., A curve-fitting program to stress relaxation data. Canadian Agricultural Engineering, 1987. 29(2): p. 209-211. 28. Schultz, D.S., Lotz, J.C., Lee, S.M., Trinidad, M.L., Stewart, J.M., Structural factors that mediate scleral stiffness. Investigative Ophthalmology & Visual Science, 2008. 49(10): p. 4232-4236. 29. Siegwart, J.T., Norton, T.T., Regulation of the mechanical properties of tree shrew 54 sclera by the visual environment. Vision Research, 1998. 39(2): p. 387-407. 30. Stewart, J.M., Schultz, D.S., Lee, O., Trinidad, M.L. Exogenous collagen cross-linking reduces sclera permeability: modeling the effects of age-related cross-link accumulation. Investigative Ophthalmology & Visual Science, 2009. 50(1): p. 352­357. CHAPTER 2 CHARACTERIZATION OF AGE AND STRAIN-RATE RATE DEPENDENT MATERIAL PROPERTIES OF OVINE RETINA 2.1 Abstract Retinal hemorrhages (RH) are prominent findings in abusive head trauma; however, injury mechanisms of RH are unclear. Finite element modeling may be useful in understanding the mechanical response of the retina yet current computational models of the pediatric eye do not incorporate age appropriate material properties. There is a paucity of infant eye material property data and as yet there are no published data characterizing the age-dependent material properties of retina. To quantify the effect of age on the mechanical response of retina, we tested tissue from immature and mature ovine eyes. Two strain-dependent uniaxial tensile tests were implemented to assess the mechanical response to different loading conditions. Differences were statistically tested by comparing the material properties (stoe, E, Oult, Sult) across age and strain-rate. Mature retina had higher Young's modulus and ultimate stress than immature retina but no statistically significant differences were found between immature and mature retinal material properties. Retina tested at the high strain-rate had a greater Young's modulus and higher ultimate stress compared to retina tested at the low strain-rate. However, no statistically significant rate effects were found for the material properties of immature and mature retina. Although age did not have a significant effect on the mechanical properties of retina, the strain-rate dependence suggests that retina is sensitive to different loading conditions and may provide useful insight into understanding injury mechanisms of RH. 2.2 Introduction Finite element (FE) analysis can be used as a tool to investigate the mechanical response of the infant eye to trauma and assist in the prediction of ocular injuries from accidental or abusive head trauma. However, current FE models of the pediatric eye are based on adult material properties with little or no consideration for mechanical changes during maturation.9,12 To date, there are no published data characterizing the age-dependent material properties of retina. Our studies indicate that there are developmental changes in the vitreous and sclera, which suggests that changes in other ocular tissues should be considered.12 The retina is the light sensitive, fibrous inner layer of the eye which connects to the optic nerve and delivers visual information to the brain. The retina is a multilayered structure, and is delicate and vulnerable to deformation. Traditional tensile testing3,4,5,14,15,16 and atomic force microscopy (AFM)7,8 have been used to characterize the mechanical response of adult retina. From these studies, adult retina has been reported to be rate-dependent14, inhomogeneous, and anisotropic. Retinal samples containing vasculature were stiffer than specimens containing no vasculature.4 Retina containing a vein in the axial direction was found to be stiffer and exhibit greater stresses than retina containing a vein in the circumferential direction.4 All of the aforementioned studies were performed on adult retina. It is unknown if 56 characteristics of immature retina are similar to adult retina. Furthermore, it is unclear if there are significant mechanical changes occurring during early development. Therefore, in this study, we characterized the age and strain-rate dependent material properties of immature ovine retina. Immature (n=12) and mature (n=13) ovine retina were tested according to a uniaxial tensile test protocol to measure the mechanical response to tensile ramp to failure. Retina was tested according to either a low or high strain-rate test. These data will be used to identify an age-appropriate constitutive model for retina to implement in a FE model of the infant eye. 2.3. Materials and Methods 2.3.1. Tissue sample preparation Immature (0-6 weeks) and mature (> 4 years) sheep eyes were obtained from non­ocular studies being conducted at the University of Utah. Whole eyes were collected immediately upon death and stored in phosphate buffered saline (PBS) at ~2°C. All ocular tissues were tested within 6 hours postmortem. Prior to testing, enucleated eyes were transferred to a petri dish containing PBS. Eyes were kept in PBS throughout dissection to prevent the ocular tissues from drying out. The extraocular muscles and soft tissues were removed from the eye and discarded. The optic nerve was severed at the optic nerve scleral junction. Each eye was bisected sagittally into nasal and temporal halves (Figure 41a). The vitreous was removed from each half by gently pulling with tweezers while squirting PBS between the vitreous and retina. The retina was isolated using the same technique by squirting PBS between the retina and choroid allowing the retina to detach and fall into the petri dish of PBS. The hemisections of retina were carefully cut using a 57 58 Figure 41: Ocular dissection procedure. (a) The red dashed line indicates the dissection cut on an eye globe. (b) A dog bone cutting die was used to cut samples. (c) A paper frame was used to support the retinal samples to transfer into grips. (d) Each specimen and support frame was aligned in custom grips. The frame was cut prior to testing. custom made, dog-bone cutting die (Figure 41b). Each retinal sample was shifted onto a glass slide for support as it was lifted out of the PBS. Excessive water surrounding the sample was absorbed with a tissue so retina would not slip off the glass. The glass slide was turned on its side and tissue thickness was measured with an optical microscope at 1x magnification (SZX16, Olympus, Center Valley, PA). A minimum of three measurements was taken for each tissue, at the center and ends of the gage length. A precut paper frame was placed on the exposed surface of the retina (Figure 41c). The paper and retina were peeled away by lifting a corner of the paper as the glass offers minimal adhesion to the retina. The paper support frame and retina were placed in custom, screw-driven grips (Figure 41d). Once the retina was properly aligned in the clamps, the two sides of the paper support frame were cut (dotted red line in Figure 41c). Width (3 mm) and gage length (6 mm) were determined by the dog-bone shape. The material test system (5943, Instron, Norwood, MA) was equipped with a 500 gram load cell (LSB210, Futek, Irvine, CA). 2.3.2. Mechanical testing All specimens were subjected to uniaxial tension according to one of two protocols (Figure 42). A low strain test protocol was implemented to characterize the mechanical response of ocular tissues during physiological increased intraocular pressure.16 A high strain protocol was implemented to characterize the mechanical response of ocular tissues during high rate trauma.6 All tests were performed in a water bath filled with PBS at room temperature. 2.3.2.1 Low strain-rate. Each tissue was subjected to ten cycles of preconditioning from 0 to 1% strain at a strain-rate of 0.01 s-1. Specimens were allowed to recover for 60 s and then subjected to a tensile ramp to failure at 0.01 s-1. 2.3.2.2 High strain-rate. Each tissue was subjected to ten cycles of preconditioning from 0 to 5% strain at a strain-rate of 0.05 s-1. Specimens were allowed to recover for 60 s and then subjected to a tensile ramp to failure at 0.1 s-1. The raw load and displacement data were sampled at 10 Hz and extracted to calculate engineering stress and strain. A custom Matlab (Mathworks, Natick, MA) code was implemented for retinal analysis and plotting and can be seen in Appendices A and C. Stress was calculated by dividing the current force by the reference cross-sectional area. Strain was calculated by dividing displacement by the original gage length. Each tissue was preloaded to approximately 0.001 N to remove any slack in the tissue sample. The strain length of the toe region (stoe), elastic modulus (E), ultimate stress (oult), and ultimate strain (sult) were extracted from each pull-to-failure test. 59 60 Tensile Test Protocol High Strain-rate - ■ - ■ Low S tra in -ra te ------ 0.1/sec Pull to failure / Preconditioning o f biological tissues for fiber alignment ' > '\ 11 ^ ' 1 /' < 1 A I , i\ / , , \ ' v ' > ;\ «' > 5 % : - '' ' ' \ / \ / \ / \ / \ : \ ! '> '' ' / V V V V v' '''j' V V °« 1 % 0.01/sec 20 s Time (s) 60 s Figure 42: The strain-dependent uniaxial tensile test protocol consisted of preconditioning, a recovery phase, and pull-to-failure. 2.3.3. Statistical analysis Age and strain-rate were analyzed independently in this study. A Student's t-test with a p-value of 0.05 was used to determine if age or strain-rate significantly affected the material properties (stoe, E, Oult, Sult). The retinal data can be seen in Appendix D. 61 2.4. Results 2.4.1 Age The immature and mature retinal samples had roughly the same thickness (Table 9). Average and standard deviations for retinal material properties can be seen in Table 10. No significant age difference was found for retinal thickness (Figure 43). At both high and low rate, the mature retina had higher Young's modulus an ultimate stress. No statistical significance with age was found for the material properties at the high and low strain-rates (Figure 44-45). Table 9: Average ± standard deviations thickness of immature and mature retina Immature Mature Thickness (mm) 0.16 ± 0.03 0.18 ± 0.03 Table 10: Average ± standard deviations of the material properties for immature and mature retina tested at the high and low strain-rates.___________________________ Immature Mature Low Strain-rate (n=8) High Strain-rate (n=9) Low Strain-rate (n=8) High Strain-rate (n=8) m m Stoe ( ------ I \m m / 0.1698 ± 0.2482 0.181 ± 0.149 0.2446 ± 0.2747 0.1477 ± 0.1649 E (MPa) 0.0082 ± 0.0139 0.009 ± 0.011 0.0178 ± 0.0226 0.0211 ± 0.02 Oult (MPa) 0.0030 ± 0.0014 0.0076 ± 0.0113 0.0113 ± 0.0130 0.0261 ± 0.024 m m Sult ( - I \m m J 1.0898 ± 0.6253 1.0366 ± 0.4033 1.0294 ± 0.3356 1.76 ± 1.06 62 0.20 (n=17) (n=16) Immature Mature Figure 43: Average and standard deviation for immature and mature retinal thickness. aP E aP 3O Si g 3 CO 0.4 0.2 0.0 - 0.2 0.04 0.03 0.02 0.01 0.00 - 0.01 0.05 0.03 0.01 - 0.01 2.5 2.0 1.5 1.0 0.5 0.0 - ■ Immature ■ Mature n=8 n=8 Low Strain-rate n=9 n=8 High Strain-rate Figure 44: Average and standard deviation for retinal material properties across age and strain-rate. 63 Figure 45: Pull-to-failure response of all included trials at low and high strain-rate across age and region. 2.4.2 Strain-rate The strain length of the toe region, Young's modulus, and ultimate stress of immature retina increased at high rate. The ultimate strain of immature retina was lower at the high rate. For the mature retina, the Young's modulus, ultimate stress, and ultimate strain increased at high rate. The stoe of mature retina decreased when tested at the high strain-rate. No statistically significance strain-rate effect was found for immature and mature retina (Figure 46). 64 ■ Low Strain-rate n=8 n=9 n=8 n=8 Immature Mature Figure 46: Average and standard deviation for retinal material properties across age and strain-rate. 2.5 Discussion Retinal hemorrhages are thought to be key indicators of pediatric abusive head trauma. It is important to incorporate an age-appropriate constitutive model into finite element models of the infant eye when investigating mechanical influences on the retina. In this study, we sought to characterize the age-dependent material properties of ovine retina. The retina proved to be an extremely difficult tissue to handle and test mechanically. A number of samples were lost due to tearing or simply being damaged during preparation and handling. Future techniques to measure the biomechanics of the retina in-vitro would be helpful. We found the elastic modulus of retina to be about three orders of magnitude less than our scleral findings. This mechanical comparison stresses the importance of the scleral structure to protect the inner ocular components, specifically preventing deformation of the retina. We found the Young's modulus of all retina to be between 8-26 kPa when tested at low and high strain-rates. Our findings fall within the range of values from previous adult retina studies. Typical tensile testing resulted in Young's moduli of approximately 2-110 kPa.3,4,5'14'15'16 Atomic force microscopy resulted in Young's moduli of 0.94-3.6 kPa.7,8 Our findings agree with existing literature that the retina stiffens with increased strain-rate. The retina is a viscoelastic material and indeed experiences higher stresses when stretched at high strain-rates. This trend was seen previously as the Young's modulus of retina was reported to be 100 and 110 kPa at low and high rate, respectively.14 The average moduli in this study were smaller; however, a couple of the retinal samples had Young's moduli greater than 50 kPa. The variation may be attributed to discrepancies in vasculature between specimens. In this study, we were able to capture the retinal vasculature in several of the specimens by imaging the tissue with our microscope. Retinal samples either contained vasculature perpendicular or parallel to the direction of the load, or no visible vessel. However, the samples sizes within each age group and strain-rate were not large enough to include vasculature directionality as a variable. Anecdotal comparisons indicate that retina containing vasculature in the direction of the applied force (parallel) at low strain-rate is stiffer than retina containing vasculature perpendicular to the direction of the applied force. Future mechanical testing of the retina should compare tissue samples with different compositions and orientations of vasculature. Histology in these specimens would also be useful to better understand the extracellular matrix of ovine retinal specimens and its 65 contribution to the mechanical response. We did not find any significant age effect on the material properties of retina. The retina is a well-organized, multilayered lining of the eye which does not change drastically from birth to adulthood. The retinal layer is made up photoreceptor cells and accessory components designed specifically for vision. These do not offer any mechanical support; therefore, we would not expect to see a significant change in the material response between the infant and adult retina. 2.6 Conclusion The material properties of the retina were not significantly different between the immature and mature age groups. Anatomically, this may support that structurally and functionally, the retina should not change drastically with age unless there is a specific vision-related disease or damage occurring in an elderly eye. In accordance with the literature, the retina is a strain-rate dependent material and becomes stiffer with increased strain-rate. This was also seen for the sclera and may shed light on injury mechanisms of retinal hemorrhages. These data will be implemented as the material definitions into an age-appropriate FE model of the infant eye, and thereby increasing the accuracy of computational models investigating retinal stress and strain. 2.7 Acknowledgements We would like to thank Dr. Kurt Albertine at the University of Utah for donating ovine ocular tissue. We would also like to thank the Knights Templar Eye Foundation for sponsoring this work. 66 67 2.8 References 1. Alamouti, B., Funk, J., Retinal thickness decreases with age: an OCT study. British Journal of Ophthalmology, 2003. 87(7): p. 899-901. 2. Bottega, W.J., Bishay, P.L., On the mechanics of a detaching retina. Mathematical Medicine and Biology, 2013. 30(4): p. 287-310. 3. Chen, K., Rowley, A.P., Weliand, J.D., Elastic properties of porcine ocular posterior soft tissues. Journal of Biomedical Materials Research, 2009. 93(2): p. 634-645. 4. Chen, K., Weiland, J.D., Anisotropic and inhomogeneous mechanical characteristics of the retina. Journal of Biomechanics, 2010. 43(7): p. 1417-1421. 5. Chen, K., Weiland, J.D., Mechanical characteristics of the porcine retina in low temperatures. Retina, 2012. 32(4): p. 844-847. 6. Chen, K., Weiland, J.D., Discovery of retinal elastin and its possible role in age-related macular degeneration. Annals of Biomedical Engineering, 2013. 42(3): p. 678-684. 7. Franze, K., Francke, M., Gunter, K., Christ, A.F., Korber, N., Reichenbach, A., Guck, J., Spatial mapping of the mechanical properties of the living retina using scanning force microscopy. Soft Matter, 2011. 7(7): p. 3147-3154. 8. Grant, C.A., Twigg, P.C., Savage, M.D., Woon, W.H., Wilson, M, Estimating the mechanical properties of retinal tissue using contact angle measurements of a spreading droplet. Langmuir, 2013. 29(16): p. 5080-5084. 9. Hans, S.A., Bawab, S.Y., Woodhouse, M.L., A finite element infant eye model to investigate retinal forces in shaken baby sundrome. Graefe's Archives for Clinical and Experimental Ophthalmology, 2009. 247(4): p. 561-571. 10. Jerdan, J.A., Glaser, B.M., Retinal microvessel extracellular matrix: an immunofluorescent study. Investigative Ophthalmology & Visual Science, 1986. 27(2): p. 194-203. 11. Jones, I.L., Warner, M., Steven, J.D., Mathematical modelling of the elastic properties of retina: a determination of Young's modulus. Eye, 1992. 6(Pt 6): p. 556-559. 12. Moran, P.R. (2012) Characterization of the vitreoretinal interface and vitreous in the porcine eye as it changes with age (Master's thesis). Retrieved from http://content.lib.utah.edu/cdm/ref/collection/etd3/id/1932 13. Rangarajan, N., Kamalakkhannan, S., Hasija, V., Shams, T., Jenny, C., Serbanescu, I., Ho, J., Rusinek, M., Levin, A., Finite element model of ocular injury in abusive head trauma. Journal of the American Association for Pediatric Ophthalmology and Stribismus, 2009. 13(4): p. 364-369. 68 14. Wollensak, G., Spoerl, E., Biomechanical Characteristics of retina. Retina, 2004. 24(6): p. 967-970. 15. Wollensak, G., Spoerl, E., Grosse, G., Wirbelauer, C., Biomechanical significance of the human internal limiting lamina. Retina, 2006. 26(8): p. 965-968. 16. Wu, W., Peter, W.H., Hammer, M.E., Basic mechanical properties of retina in simple elongation. Journal of Biomechanical Engineering, 1987. 109(1): p. 65-67. CHAPTER 3 CHARACTERIZING THE EFFECT OF POSTMORTEM TIME AND STORAGE CONDITION ON MECHANICAL PROPERTIES OF IMMATURE AND MATURE OVINE SCLERA AND RETINA 3.1 Abstract Material property testing of soft biological tissues is ideally conducted just after death to reflect the most physiologic response for that species. However, human ocular tissues may only be available 24-72 hours postmortem. To date, there are no known studies evaluating the effect of postmortem time (PMT) on pediatric ocular tissues. Furthermore, it is unclear what storage parameters are suitable, if any, during shipping and transportation. To determine a viable time period for material testing, we characterized the effect of PMT on the mechanical response of immature and mature ovine sclera. To determine a shipping strategy for material testing, we characterized the effect of storage condition on the mechanical response of immature and mature ovine sclera and retina. Scleral samples were tested in uniaxial tension up to 24 hours postmortem, and differences were assessed among fresh, frozen/thawed, and fixed sclera and retina. A significant negative correlation with PMT was found for stress relaxation constants, Young's modulus, and ultimate stress for the immature sclera, with the primary change occurring after 10 hours postmortem. PMT had no significant effect on the material properties of mature sclera. In the storage condition analysis, fixed immature and mature sclera and retina were significantly stiffer than fresh tissue and had higher ultimate stresses. Freezing then thawing only had a significant effect on the ultimate stress of immature posterior sclera and ultimate strain of retina. These data suggest that immature sclera can be mechanically tested up to 10 hours postmortem and freezing sclera or retina may be a viable shipping technique for pediatric ocular tissues. Mature ovine sclera can be stored in phosphate buffered saline for up to at least 24 hours postmortem. 3.2 Introduction There is a paucity of pediatric eye material property data in the literature as obtaining human donor eyes in this age range is difficult. In order to obtain a sufficient number of specimens for testing, eye banks across the country will need to be utilized. Material property testing of any soft biological tissues is ideally conducted just after death to reflect the most physiologic mechanical response for that species. However, pediatric donor eyes may only be available 24-72 hours postmortem, and will likely need to be shipped across the country. The effect of postmortem time (PMT) on the material properties of mature rabbit sclera has been previously measured and suggests that it can be stored up to 72 hours in phosphate buffered saline (PBS).2 To date, there are no known studies evaluating the effect of PMT on pediatric ocular tissues. Furthermore, it is unclear what storage parameters are suitable, if any, during shipping and transportation. Fixation and freezing are two storage methodologies that have not been explored for sclera and retina. Fixation has only been investigated for the cornea, which becomes stiffer at higher 70 concentrations of glutaraldehyde fixation.3 To determine viable shipping and storage strategies for pediatric ocular tissues, we characterized the effect of PMT and storage condition on the mechanical response of mature and immature ovine sclera. Due to the limited availability of retinal samples, only PMT was assessed for sclera over a broad range of testing time frames. These data will provide guidance for the requirements of collecting and accurately measuring the material properties of human sclera and retina. 3.3 Materials and Methods 3.3.1 Tissue collection and storage Whole eyes were collected from newborn lambs and adult sheep immediately upon death and stored according to the desired storage condition (Table 11 and Table 12). Whole eyes for PMT evaluation were stored in a 2°C refrigerator in containers of PBS and tested up to ~24 hours postmortem. Frozen/thawed whole eyes were collected within an hour postmortem, placed in PBS, and stored in a -23°C freezer immediately. The frozen samples were kept in the freezer 24 hours then and allowed to thaw at room temperature for approximately 3 hours before testing. Fixed eyes were also collected within an hour postmortem, but stored in a 1%-formaldehyde/1.25%-glutaraldehyde mixture for a minimum of 72 hours. Table 11: Retinal sample sizes by storage condition (low strain-rate). 71 Immature Mature Storage Condition (low strain) Fresh n=8 n=8 Frozen n=3 n=2 Fixed n=11 n=5 72 Table 12: Scleral sample sizes by postmortem time (high strain-rate) and storage condition (low strain-rate).________________________________________ Immature Mature Anterior Posterior Anterior Posterior PMT (fresh, high strain) 0-6 hrs n=13 n=14 n=6 n=7 6-12 hrs n=6 n=5 n=N/A n=N/A 12-24 hrs n=11 n=10 n=7 n=6 > 24 hrs n=5 n=5 n=5 n=5 Storage Condition (low strain) Fresh n=11 n=11 n=5 n=5 Frozen n=8 n=8 n=4 n=4 Fixed n=8 n=9 n=6 n=6 3.3.2 Tissue dissection On the day of testing, enucleated eyes were transferred to a petri dish containing PBS. Eyes were kept in PBS throughout dissection to prevent the ocular tissues from drying out. The extraocular muscles and soft tissues were trimmed from the eye and discarded, and the optic nerve was severed at the optic nerve scleral junction. Each eye was bisected sagittally into nasal and temporal halves (Figure 47a,c). The vitreous was removed from each half by gently pulling with tweezers while squirting PBS between the vitreous and retina. The retina was isolated by squirting PBS between the retina and choroid allowing the retina to detach and fall into the petri dish of PBS. The hemisections of retina were carefully cut using a dog-bone cutting die (Figure 47b). Each retinal sample was shifted onto a glass slide for support as it was lifted out of the PBS. Any excessive water surrounding the sample was absorbed with a tissue so that the retina would not slip 73 Figure 47: Ocular dissection procedure. (a) The red dashed line indicates the direction of dissection cut on an eye globe. (b) A custom dog bone cutting die was used to cut scleral samples. (c) The eye is bisected sagittally leaving nasal and temporal halves from which scleral and retinal samples were taken from anterior and posterior regions. (d )A paper frame was used to support the retinal samples during transfer into grips. (e) Each tissue sample was aligned in custom screw-driven grips. off the glass slide. The glass slide was turned on its side and tissue thickness was measured with an optical microscope at 1x magnification (SZX16, Olympus, Center Valley, PA). Thickness was measured at the center and ends of the gage length. A precut paper support frame was placed on the exposed surface of the retina (Figure 47d) and both were slid away from the slide by lifting a corner of the paper. The paper support frame and retina were placed in custom made, screw-driven clamps (Figure 47e). Once the retina was properly aligned in the clamps, the two sides of the paper frame were cut (dotted red line in Figure 47d). Width (3 mm) and gage length (6 mm) were determined by the dog-bone shape. The material test system (Model 5943, Instron, Norwood, MA) was equipped with a 500 gram load cell (LSB210, Futek, Irvine, CA). Sclera was isolated by removing the choroid with tweezers. The resulting hemisections of sclera were placed on a cutting board and press-cut with the dog-bone cutting die. Anterior and posterior scleral samples were cut from each of the ocular halves (Figure 47c). Tissue thickness was measured using the optical microscope at 1x magnification and the average of three measurements were taken from the center and each end of the gage length. Width and gage length were determined by the dog-bone shape of the tissue sample. Each scleral sample was aligned in the clamps and measured with a 1 kN (Instron, Norwood, MA) or 500 gram load cell for high and low strain tests, respectively. 3.3.3 Mechanical testing Uniaxial stress relaxation and pull-to-failure tests in tension were performed on fresh sclera at low strain-rates (0.01 s-1, Figure 48). To determine the rate dependence of PMT, a high strain-rate protocol was also performed on sclera (0.1 s-1). Due to limited retinal specimens and constraints of the lower limit accuracy of the load cell, only low strain-rate pull-to-failure tests were performed on retina (Figure 49). All tests were performed in an environmental bath filled with phosphate buffered saline at room temperature. 3.3.3.1 PMT study - Sclera. Each tissue was subjected to ten cycles of preconditioning from 0 to 5% strain at a strain-rate of 0.05 s-1. Specimens were allowed to recover for 60 s and then a stress relaxation test was performed by applying 25% strain and holding for 900 s. The tissue was allowed to recover for 60 s, and then subjected to a tensile ramp to failure at 0.1 s-1. 74 75 Figure 48: The strain-dependent uniaxial tensile test protocol for the PMT and storage studies of sclera consisted of preconditioning, stress relaxation, and pull-to-failure at either a high or low strain level. 3.3.3.2 Storage study - Sclera. Each tissue was subjected to ten cycles of preconditioning from 0 to 1% strain at a strain-rate of 0.01 s-1. Specimens were allowed to recover for 60 s and then a stress relaxation test was performed by applying 1% strain and holding for 900 s. The tissue was allowed to recover for 60 s, and then subjected to a tensile ramp to failure at 0.01 s-1. 3.3.3.3 Storage study - Retina. Each tissue was subjected to ten cycles of preconditioning from 0 to 1% strain at a strain-rate of 0.01 s-1. Specimens were allowed to recover for 60 s and then subjected to a tensile ramp to failure at 0.01 s-1. The retina load response during stress relaxation tests was very low and was strongly influenced by low-frequency noise. This prohibited us from performing stress relaxation tests on retinal samples. 76 Figure 49: The strain-dependent uniaxial tensile test protocol for retinal samples consisted of preconditioning and pull-to-failure at a low strain level. The raw load and displacement data were sampled at 10 Hz and extracted to calculate engineering stress and strain. Stress relaxation data for each scleral specimen were fit to a two-term generalized Maxwell model [Eq.1].4 A least-squares curve-fitting technique was used to solve for equilibrium stress (oe), intermediate stress (01, 02), and the decay (11, T2) constants. Instantaneous stress (oi) was defined as the sum of the equilibrium and intermediate stress constants [Eq.2]. The strain length of the toe region (stoe), elastic modulus (E), ultimate stress (oult), and ultimate strain (sult) were extracted from the scleral and retinal pull-to-failure tests. Preliminary results repeatedly showed a good fit to the experimental data using this viscoelastic material model. a( t ) = o e + t=i t T; (1) 77 a i = a e + a 1 + o 2 (2) 3.3.4 Statistical analysis A no rm al b iv a ria te co rrelatio n analysis was p e rfo rm ed to ev a lu a te significant p o stm o rtem tim e changes in the stress re lax a tio n co n stan ts (oi, Oe, 0 1 , 0 2 , T1 , T2) and material prope rtie s (oult, Sult, E, stoe) o f an terio r and p o ste rio r scleral samples. A P e a rso n co rrelatio n co e ffic ien t was com p u ted to id en tify sig n ifican t co rrelatio n with PM T (p=0.95). A o n e ­way analysis o f v arian ce (ANOVA) was u sed to d eterm in e if storage co n d itio n sig n ifican tly affected th e re lax atio n co n stan ts o f sclera and m ateria l p ro p e rtie s o f sclera and retina. A D u n n e tt's te st with a p -v a lu e o f 0.05 was u sed to id en tify significant diffe ren c es b e tw e en fresh tissu e and fro z en /th aw ed and fixed tissue. A g e and reg io n we re analyzed in d ep en d en tly fo r b o th th e PM T and storage co n d itio n study. Th e scleral and re tina l d a ta can b e seen in A p p en d ix D. 3.4 Results Th e re su lts from th e co rre la tio n analyses w e re d ep icted u sing e ith e r d iag o n al or straight lines. The d iag o n a l lines do n o t rep resen t th e fit lines, b u t ra th e r representing sig n ifican t correlation. Similarly, the straight lines signify no sig n ifican t correlation. 3.4.1 PMT - Immature sclera A slig h tly n eg a tiv e co rre la tio n w ith PM T w as seen fo r the im m ed ia te and lo n g -term d e c ay co n stan ts fo r im m a tu re sclera, b u t this n eg a tiv e co rre la tio n was n o t significant (Figure 50). A sig n ifican t n eg a tiv e co rrelatio n w ith PM T was found fo r th e in stan tan eo u s 78 400 )ces 300 200 100 0. 16 14 12 )ces 10 8- 2 6- 4 2- 0 6* ° - o e _ 0-6 6-12 x Anterior, r = -0.1467 o Posterior, r = -0.1322 x Anterior, r = -0.3465 o Posterior, r = -0.2146 12-24 >24 PMT (hours) Figure 50: Statistical correlation (red and blue lines) found no significant effect of PMT on the decay time constants for immature anterior and posterior sclera. stress (oi), intermediate stress constants (01, 02), and equilibrium stress (oe) of the immature anterior and posterior sclera (Figure 51, Figure 52). No changes with PMT were seen for the stoe of the immature sclera (Figure 53), but a significant negative correlation with PMT was found for the Young's modulus (E) (Figure 54). A significant negative correlation with PMT was found for the ultimate stress (out) of the immature anterior and posterior sclera (Figure 55), but there was no correlation of ultimate strain with PMT (Figure 56). 3.4.2 PMT - Mature sclera Unlike immature sclera, no significant correlation with PMT was found for any of the stress relaxation constants or material properties of the mature anterior and posterior sclera (Figure 57-Figure 63). 79 PMT (hours) Figure 51: Statistical correlation (red and blue lines) found significant effect of PMT on the instantaneous and equilibrium stress for immature anterior and posterior sclera. PMT (hours) Figure 52: Statistical correlation (red and blue lines) found significant effect of PMT on the intermediate stresses for immature anterior and posterior sclera. 80 £1 £ Si £ 0.35 0.30- 0.20 x Anterior, r = -0.1231 o Posterior, r = 0.0206 PMT (hours) Figure 53: Statistical correlation (red and blue lines) found no significant effect of PMT on the strain length of the toe region for immature anterior and posterior sclera. 25- Ph 201 pq 15 x Anterior p<0.0001, r = -0.6825 o Posterior Figure 54: the Young 0-6 6-12 12-24 >24 PMT (hours) Statistical correlation (red and blue lines) found significant effect of PMT on s modulus for immature anterior and posterior sclera. 81 PMT (hours) Figure 55: Statistical correlation (red and blue lines) found significant effect of PMT on the ultimate stress for immature anterior and posterior sclera. £1 £ o.s 0.4- x Anterior, r = 0.2158 o Posterior, r = 0.2924 PMT (hours) Figure 56: Statistical correlation (red and blue lines) found no significant effect of PMT on the ultimate strain for immature anterior and posterior sclera. 82 ce ce 700 x Anterior, r = 0.4492 600 o Posterior, r = -0.1180 500 o 400 300 x - o# o O x 200 O x O flf V 100 O O x o 0 X * 16 o x Anterior, r = 0.4682 14 o o Posterior, r = 0.1616 12 O p 10 0 * 8' o ° X O x 6 o 0 4 o° o xx X x o 2 0 * X 0-6 12-24 >24 PMT (hours) ire 57: Statistical correlation (red and blue lines) found no significant effect of PMT e decay time constants for mature anterior and posterior sclera. a Ph o a Ph o x Anterior, r = -0.3642 o Posterior, r = 0.1695 0.5- X 0.0- ° o cPcff) Q o o ° o 0.4- x Anterior, r = -0.0709 o Posterior, r = 0.2896 0.3- * - 0.2- * * 0.1- X o 0 0 O q O cP o£>x o 0-6 12-24 >24 PMT (hours) Figure 58: Statistical correlation (red and blue lines) found no significant effect of PMT on the instantaneous and equilibrium stress for mature anterior and posterior sclera. 83 aP 0.25' 0.20 >; 0.15 D o.io 0.05 0.00 1.2 ctf 0.8 Oh S 0.6 <N D 0.4 0.2 0.0 * o.o (Pcfb x Anterior, r = -0.3033 o Posterior, r = 0.1957 * * o oft CD o ° 0 ° o x Anterior, r = -0.4504 o Posterior, r = 0.0484 x - 12-24 PMT (hours) Figure 59: Statistical correlation (red and blue lines) found no significant effect of PMT on the intermediate stresses for mature anterior and posterior sclera. 0.34 0.32- g I g 0.30* |Uj: 8 0.28- co 0.24- x Anterior, r = 0.3247 o Posterior, r = -0.0004 12-24 PMT (hours) Figure 60: Statistical correlation (red and blue lines) found no significant effect of PMT on the strain length of the toe region for mature anterior and posterior sclera. 84 aP E 15 10- x Anterior, r = 0.0337 o Posterior, r = 0.1242 PMT (hours) Figure 61: Statistical correlation (red and blue lines) found no significant effect of PMT on the Young's modulus for mature anterior and posterior sclera. aP Ou x Anterior, r = 0.2980 o Posterior, r = 0.1109 12-24 PMT (hours) Figure 62: Statistical correlation (red and blue lines) found no significant effect of PMT on the ultimate stress for mature anterior and posterior sclera. 85 x Anterior, r = 0.1524 o Posterior, r = 0.0387 & X Q x o o * x O o o (Te 12-24 >24 PMT (hours) Figure 63: Statistical correlation (red and blue lines) found no significant effect of PMT on the ultimate strain for mature anterior and posterior sclera. 3.4.3 Storage condition - Immature sclera Fixation of immature sclera significantly stiffened the tissue and increased the ultimate stress of the anterior and posterior sclera (p<0.05). The toe region of fixed immature posterior sclera was significantly shorter than fresh (<6 hours) immature posterior sclera (p<0.05). Freezing then thawing significantly decreased the ultimate stress of immature posterior sclera (p<0.05). Average and standard deviations of the immature scleral material properties for each storage condition can be found in Table 13. Figure 64 illustrates all immature scleral trials subjected to tensile ramp to failure for this study. The measured force from several scleral tests exceeded the upper limits of the load cell before failure. These specimens are removed in Figure 65. Average pull-to-failure responses for the immature scleral samples in Figure 65 are shown in Figure 66. 1.0 0.9 I £ Si £ 0.6 0.4 86 Table 13: Average +/- standard deviation for immature scleral material properties. Fresh sclera was tested within 6 hours postmortem. * p<0.05__________ Immature Sclera Anterior stoe (mm/mm) E (MPa) Oult (MPa) suit (mm/mm) Fresh 0.14 ± 0.04 12.22 ± 5.29 3.18 ± 1.61 0.43 ± 0.10 Frozen 0.11 ± 0.06 6.23 ± 2.67 1.85 ± 0.73 0.54 ± 0.16 Fixed 0.10 ± 0.03 38.85 ± 12.28 * 7.88 ± 4.63 * 0.26 ± 0.09 Posterior stoe (mm/mm) E (MPa) Oult (MPa) sult (mm/mm) Fresh 0.22 ± 0.05 12.99 ± 6.63 3.48 ± 2.27 0.49 ± 0.17 Frozen 0.25 ± 0.07 4.76 ± 3.98 0.40 ± 0.14 * 0.58 ± 0.16 Fixed 0.09 ± 0.02 31.95 ± 12.12 * 13.88 ± 0.32 * 0.32 ± 0.02 Eresh Ant (N=14) Frp:«h Post fW=1 ^ 2 1____________I____________I____________I____________I____________I____________I____________I____________I____________I____________I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Strain (mm/mm) Figure 64: All pull-to-failure responses for immature anterior and posterior sclera by storage condition. 87 0,8 1 1.2 Strain (mm/mm) Figure 65: Pull-to-failure responses for immature anterior and posterior sclera that reached tissue failure before the upper limit of the load cell. Legend on graph indicates samples sizes for storage condition. 0.4 0.5 0.6 Strain (iran/trim) Figure 66: Averaged pull-to-failure response for the immature anterior and posterior sclera that failed. 3.4.4 Storage condition - Mature sclera Similar to immature sclera, fixation of mature sclera significantly stiffened the tissue and increased the ultimate stress of the anterior and posterior sclera (p<0.05). stoe of fixed mature posterior sclera was significantly shorter than fresh mature posterior sclera (p<0.05). Freezing then thawing had no significant effect on the material properties of mature sclera (p<0.05). Average and standard deviations of the immature scleral material properties for each storage condition can be found in Table 14. Figure 67 illustrates all mature scleral trials subjected to tensile ramp to failure for this study. Scleral samples that reached the maximum limit of the load cell before failure were removed in Figure 68. Average pull-to-failure responses for the mature scleral samples in Figure 68 are shown in Figure 69. Significant storage condition effects for the immature and mature sclera can be seen in Figure 70-Figure 72. 88 Table 14: Average +/- standard deviation for mature scleral material properties. Fresh tissue was tested within 6 hours postmortem. * p<0.05________________ Mature Sclera Anterior stoe (mm/mm) E (MPa) out (MPa) Suit (mm/mm) Fresh 0.06 ± 0.03 10.17 ± 12.52 1.81 ± 3.11 0.53 ± 0.33 Frozen 0.07 ± 0.01 16.95 ± 11.75 1.74 ± 0.77 0.55 ± 0.03 Fixed 0.06 ± 0.02 34.56 ± 11.71 * 9.69 ± 4.21 * 0.27 ± 0.03 Posterior stoe (mm/mm) E (MPa) oult (MPa) sult (mm/mm) Fresh 0.13 ± 0.06 2.49 ± 4.58 0.72 ± 1.09 0.49 ± 0.14 Frozen 0.14 ± 0.03 0.99 ± 0.27 0.27 ± 0.05 0.41 ± 0.09 Fixed 0.07 ± 0.03 * 13.16 ± 4.65 * 7.23 ± 1.48 * 0.28 ± 0.09 89 Figure 67: All pull-to-failure responses for mature anterior and posterior sclera by storage condition. Figure 68: Pull-to-failure responses for mature anterior and posterior sclera that reached tissue failure before the limits of the load cell were reached. Legend on graph indicates samples sizes for storage condition. 90 Strain (mnrmm) Figure 69: Averaged pull-to-failure response for the mature anterior and posterior sclera that failed. Dip in fixed posterior average is due to offset failure strains in the two specimens averaged. Immature Mature ■ Anterior ■ Posterior Figure 70: Average and standard deviation for stoe and Young's modulus across storage condition for immature and mature anterior and posterior sclera. 91 40- aP E 10- Immature Fresh Fixed Fresh. Mature ■ Anterior ■ Posterior 1 Fixed Figure 71: Average and standard deviation for stoe and Young's modulus across storage condition for immature and mature anterior and posterior sclera. Immature Mature aP uou £1 £ Si £ Fresh Frozen Fixed Fresh Frozen Fixed Figure 72: Average and standard deviation for ultimate stress and strain across storage condition for immature and mature anterior and posterior sclera. * * 3.4.5 Storage condition - Retina Fixation of immature and mature retina significantly stiffened the tissue (p<0.05), increased the ultimate stress (p<0.05), and decreased the ultimate strain (p<0.05). Freezing then thawing immature and mature retina significantly increased the ultimate strain (p<0.05). Average and standard deviations for the immature and mature retinal material properties can be seen in Table 15 and Table 16. Significant storage condition effects for the immature and mature retina can be seen in Figure 73-Figure 76. 92 Table 15: Average +/- standard deviation for immature retinal material properties. Fresh retina tested within 6 hours. * p<0.05____________________________________________ Immature Retina stoe (mm/mm) E (MPa) oult (MPa) sult (mm/mm) Fresh (n=8) 0.170 ± 0.278 0.008 ± 0.014 0.003 ± 0.001 1.090 ± 0.625 Frozen (n=3) 0.224 ± 0.211 0.001 ± 0.0002 0.002 ± 0.0006 2.238 ± 0.510 * Fixed