Edited by: Bernardo Innocenti, Free University of Brussels, Belgium
Reviewed by: Jason Luck, Duke University, United States; Silvia Piangiani, Free University of Brussels, Belgium
This article was submitted to Biomechanics, a section of the journal Frontiers in Bioengineering and Biotechnology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
While several studies have qualitatively investigated age- and region-dependent adhesion between the vitreous and retina, no studies have directly measured the vitreoretinal strength of adhesion. In this study, we developed a rotational peel device and associated methodology to measure the maximum and steady-state peel forces between the vitreous and the retina. Vitreoretinal adhesion in the equator and posterior pole were measured in human eyes from donors ranging 30 to 79 years of age, and in sheep eyes from premature, neonatal, young lamb, and young adult sheep. In human eyes, maximum peel force in the equator (7.24 ± 4.13 mN) was greater than in the posterior pole (4.08 ± 2.03 mN). This trend was especially evident for younger eyes from donors 30 to 39 years of age. After 60 years of age, there was a significant decrease in the maximum equatorial (4.69 ± 2.52 mN,
Any disruption to the layers of the retina, particularly the separation of the photosensitive cells from the retinal pigment epithelium, can result in blindness or severe visual impairment. The most common cause of disruption is retinal detachment. Retinal detachment occurs in one of every 10,000 people (Mitry et al.,
Little is quantitatively known about adhesion at the vitreoretinal interface. Sebag (
Vitreoretinal adhesion has qualitatively been shown to vary with region. Gandorfer et al. (
No studies to date have directly measured adhesive forces at the vitreoretinal interface, but there are several studies that quantified retinal separation at the pigment epithelium (RPE). The earliest series of studies were by Zauberman and deGuillebon (Zauberman and Berman,
The limitation of these peel studies is the potential disruption of the interface between the RPE and choroid prior to testing. Sandwich specimens were physically removed from the spherical eye and laid flat for testing. The excision likely damages structures at the cut interface, and straightening the sample likely causes shear forces between the layers which may compromise the interface. To overcome these limitations, Kita et al. (
The objective of our study was to quantify vitreoretinal adhesive forces in sheep and human eyes, and evaluate how measurements change with age and region of the eye. To achieve this objective, an innovative testing device was created to overcome many of the limitations of the previous retinal adhesion methods. This device allows the retina to be peeled from the vitreous without altering the curvature of the specimens, or requiring dissection of the retina. Using this technique, quantitative vitreoretinal adhesive forces can be directly measured. These forces will be necessary to understand the biomechanics of vitreoretinal adhesion and create numerical tools for predicting retinal detachment.
The adhesive strength between the vitreous and retina was measured in sheep (
Sheep eyes were removed immediately upon sacrifice and refrigerated
Testing was performed in the equatorial and posterior pole regions for each eye. The order in which samples were tested (right/left, equator/posterior pole) was randomly selected. Extraocular tissue and the optic nerve were removed from the globe prior to all dissections. For equatorial peels, a cut through the sclera, leaving the choroid, retina, and vitreous intact, was made anteriorly from the small opening at the optic nerve head to the equatorial region, ending ~15 mm posterior to the cornea (Cut 1 in Figure
For posterior peels, an initial cut was made similar to the equatorial peels (Cut 1 in Figure
A peel test system was created to test vitreoretinal adhesion with minimal dissection and disruption to the vitreoretinal interface. This was achieved by keeping the retina and vitreous in their natural configuration, and rotating the eye as the retina was peeled away from the vitreous. The prepared eye with a window of exposed retina was placed onto a flexible membrane (Dragon Skin, Smooth-On, Macungie, PA) molded to cup the eye (Figure
For each peel test,
A plastic L-shaped tab was connected to an Instron Universal Testing Instrument (Model No. 5943, Instron, Norwood, MA) where load measurements were made with a 5 N uniaxial load cell (Model No. 2530-5N, Instron, Norwood, MA). A thin layer of cyanoacrylate adhesive was applied to the bottom surface of the tab (5.0 × 4.5 mm) and lowered until it was in contact with the retina (Figure
Following rotational peel tests, 2 mm square sections of peeled retina not directly attached to the tab were removed to characterize damage to the vitreoretinal interface. Specimens were placed in a 1% buffered formaldehyde and 1.25% glutaraldehyde solution for 24 h. For processing, samples were placed in a 0.1 M sodium cacodylate buffer rinse twice for 5 min while being agitated. Samples were then placed in a 1:1 mixture of diluted osmium tetroxide
The novel rotational peel system was validated by measuring adhesion between metal and tape (Daigger, Vernon Hills, IL) and comparing measurements in the rotational system to measurements from linear peel testing using ASTM standard D6862-11 (Figure
In any peel test, a maximum force is required to initiate a peel, and a steady-state peel force maintains the peel. In this study, these attributes were extracted through careful examination of videos of each peel test in conjunction with the force-time data.
The maximum peel force was defined as the maximum force measured prior to separation between the retina and the vitreous. The period of steady-state peel was defined as the time period of relatively constant force after the vitreous separated from the retina, leaving vitreous in contact at the base of the peel. A video showing a typical peel test is provided as
A two-way ANOVA with repeated measures was performed to evaluate the effect of region and age on the maximum and steady-state peel force in the sheep eyes and human eyes. Repeated measures were used because regional data was collected from the same eyes. To perform this test in humans, donor ages were binned into the 4, 5, 6, 7, or 8th decade of life. Tukey-Kramer
To identify significant differences in mechanisms of failure, the light microscopy images of the peeled retina were rated according to the following criteria: 0–ILM cleanly separated with no disruption or evidence of traction; 1–ILM cleanly separated, but ILM is undulated or there is evidence of traction on the ILM; 2–ILM cleanly separated, with the exception of 1–2 small localized disruptions (typically around vessels); 3–ILM torn and disrupted. A chi-square test was performed on these scores to identify significant differences between region and age. A logistic regression was also used to determine if the maximum and steady-state peel forces were predictive of failure type (i.e., 0, 1, 2, or 3). For all statistical tests,
The forces measured from the linear and rotational tape peel tests did not exhibit a definitive peak force (Figure
Average peel force measurements from the rotational tests were within 1.3% of the average peel forces measured from linear peel tests using ASTM standard D6862-11. Only one rotational peel test is plotted for clarity.
No significant differences with region were seen for either the max or steady-state peel force except for the adults which had significantly higher max peel force in the equatorial region (16.67 ± 8.46 mN) compared to the posterior region (8.46 ± 2.43 mN,
At all ages, the maximum peel forces in the equator (7.16 ± 4.08 mN) were greater than those in the posterior pole (4.08 ± 2.03 mN, Figure
Steady-state peel force in the equator of the youngest human eyes was significantly greater than the posterior pole, and significantly greater than the equatorial steady-state adhesion of all other ages. Error bars indicate standard deviation from the mean. *
Inspection of light microscopy images in the sheep revealed generally clean separation at the vitreoretinal interface (Figure
In this study, we developed a novel rotational peel test system to measure the strength of adhesion at the vitreoretinal interface. We used the device to quantify age- and region-dependent vitreoretinal adhesion in human and sheep eyes. These data are critical to improving biomechanical understanding of the vitreoretinal interface. They can be implemented into computational models to investigate posterior vitreous detachment, or to simulate traumatic ocular injury. Further, the peel test system and associated data can be used to evaluate the effectiveness of chemical compounds to assist in surgical separation of the retina and vitreous.
In both sheep and human eyes, there was an interesting distinction between the maximum force of vitreoretinal adhesion in the equator and posterior pole that was highly dependent on the maturity of the animal. In premature, neonatal, and young lamb eyes, there were no regional differences. By the time the sheep developed into young adults (~28–36 years human age equivalent), the equatorial vitreoretinal adhesion significantly increased while adhesion in the posterior pole remained unchanged from the immature sheep eyes. This may be due to posterior extension of the vitreous base and ILM thickening with age. Wang et al. investigated the anteroposterior length of the vitreous base in 58 pairs of human eyes from subjects spanning ages of 8–96 years old (Wang et al.,
Similar to the sheep eyes, the maximum peel force in the equator of the human eyes was greater than in the posterior pole. This was only significant in the youngest age group (30–39 years old), but all binned age groups younger than 70 years of age had higher average adhesion in the equator than in the posterior pole. One interesting finding from our dataset was a significant decrease in equatorial maximum peel force in eyes from human subjects > 60 years of age (Figure
Our study is the first quantitative measurement of vitreoretinal adhesion, but other groups have qualitatively evaluated adhesion at the vitreoretinal interface (Sebag,
Microscopic examination of peeled sheep retinas suggested generally clean separation at the ILM, regardless of age or region. Occasionally, the ILM was torn in immature eyes, regardless of region. In adult sheep, the ILM was never torn, but the nerve fiber layer was often stretched and occasionally disrupted locally around blood vessels. There was a single premature specimen with failure occurring in the ganglion cell layer, which is similar to Sebag. We obtained light microscopy specimens from 37% of our sheep peel tests. A more extensive microscopic analysis might correlate more closely with observations by Sebag.
In human adult peeled retinas, there was significant distinction between failure in the equator and posterior pole. The peeled retina in the equator was generally smooth with some undulation or traction-like peaks on the surface of the ILM, while many of the posterior peeled retinas exhibited complete disruption of the ILM. This was surprising given that adhesion in the posterior pole is significantly lower than the equator. We hypothesize that numerous collagen penetrations in the posterior pole create large disruptions of the ILM when pulled, and that there is less collagen penetration in the equator where adhesion is thought to be dominated by adhesive proteins acting as an extracellular glue. Failure of this “glue” would result in cleaner separation. The traction-like characteristics may be caused by sparse penetration of collagen fibrils. Interestingly, ILM disruption in the posterior pole was significantly greater in eyes from donors older than 60 years of age. Because the adhesive strength is weak at this age, we hypothesize that the disruption of the ILM is an indication of decreased retinal structural integrity. Regardless, these data show adhesive strength cannot be inferred solely from locations of damage. Mechanism of adhesion and changes in retinal structural integrity with age likely contribute to patterns of failure.
Gandorfer et al. evaluated the effect of plasmin in different regions of porcine eyes (Gandorfer et al.,
Several studies have quantitatively measured adhesion between the neurosensory layer and retina pigment epithelium (RPE) (Zauberman and Berman,
Retinal detachment or tearing can occur due to vitreoretinal traction, so adhesion at the vitreoretinal interface is likely greater than adhesion at the RPE in healthy young adults. No studies have measured RPE adhesion in sheep or in humans, making comparison of our measurements with the literature challenging. Kita and Marmor (
In standard peel tests, a thin membrane is typically peeled from a solid surface. In these studies, we peeled a thin membrane (retina) from a gel (vitreous). This resulted in deformation of both materials prior to and during peeling. The maximum peel forces measured in this study were defined as the maximum force before clear separation of the vitreous from the retina. With this definition, the maximum peel force may incorporate some retinal stretching or separation from the scored retinal edges in addition to the peeling force. During steady-state peeling, the retina also experienced some deformation, however, careful examination of video in conjunction with the peel force data provided confidence that peeling, and not retinal deformation, was the primary contributor to the steady-state measurements. A computational simulation of the peel tests is planned for a future study to separate tissue adhesion and vitreous deformation in the maximum peel force measurements.
The data collected in this study was on the low end of the load cell limit. Because of this, we had the load cell carefully calibrated at its reported lower limit of 9 mN. The uncertainty measurement at this limit was 0.002 gf, or 0.0196 mN, and decreased with decreasing load. The maximum peel force data in our study ranged from 4 to 15 mN, which is near the lower limit of the load cell. The steady state peel forces were lower than the maximum peel forces (1–3 mN), but were still on the same order of magnitude as the calibrated limit, and still two orders of magnitude larger than the uncertainty measurements. It is possible the load cell limits contributed to the variability in the steady-state measurements, but we have strong confidence in the conclusions and trends of the study.
Vitreoretinal adhesion to large blood vessels is thought to be greater than adhesion in regions without blood vessels. Our observations of ILM tearing surrounding blood vessels support this notion. Further, we observed steady-state peel forces drop after passing a blood vessel. For this study, we extracted steady-state values from regions without the blood vessels in order to maintain a consistent comparison across all ages and regions. A comprehensive and focused assessment of the effect of blood vessels on adhesion will be performed in a future study.
We developed a novel device to quantify vitreoretinal adhesion in the equator and posterior pole of human and sheep eyes. Maximum vitreoretinal adhesion in adult human eyes (30–79 years old) was greater in the equator than in the posterior pole, especially at young ages (30–39 years old). After 60 years of age, there was a significant drop in equatorial adhesion that may be protective to the retina by facilitating vitreous detachment during liquefaction. In immature (premature, neonatal, and young lamb) and mature (young adult) sheep eyes, there was no significant difference in posterior vitreoretinal adhesion, but maximum equatorial adhesion in mature eyes was 2 times greater than immature eyes. This may be caused by the extension of the vitreous base during development. These data are the first quantitative measurements of vitreoretinal adhesion, and will be useful in the development of computational models for simulating posterior vitreous detachment or ocular trauma. The methods and technology developed for this study can be used to evaluate mechanisms of adhesion, and assess the efficacy of enzymes to remove or reduce vitreoretinal adhesion for surgical interventions.
All data used in this manuscript is available via doi:
CC performed testing, sectioning, preliminary analysis and manuscript writing. JC assisted with all aspects of the data analysis and manuscript writing. BC was responsible for the study conception and design, data interpretation, and manuscript editing.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors thank the Utah Lions Eye Bank and donor families for providing human donor eyes. The authors gratefully acknowledge donated sheep eyes from Dr. Kurt Albertine and the Neonatal Chronic Lung Disease and Development Lab at the University of Utah. Also acknowledged are Keith Warner and Krishna Womack for their efforts in analyzing video recordings.
The Supplementary Material for this article can be found online at: