Human Scleral Hydraulic Conductivity: Age-Related Changes, Topographical Variation, and Potential Scleral Outflow Facility

Abstract

purpose. To measure the specific hydraulic conductivity (K) of human sclera over a range of ages, to assess topographical variation, and to provide a theoretical estimate of potential scleral outflow facility.

methods. Human donor sclera (n = 18; mean age 56.7 ± 25.9 years; range 4–89) was clamped in a modified Ussing chamber connected to a water column set at 15.7 mm Hg. Column descent was measured over 24 hours at 20°C with a digital micrometer. Scleral thickness of glutaraldehyde-fixed specimens was measured by light microscopy, taking the mean of 15 measurements per donor. Topographical variation in hydraulic conductivity (HC) was determined in an additional 10 donor eyes (mean age, 54.1 ± 26.4 years; range 12–89), comparing anterior, equatorial, and posterior sclera. The potential transscleral outflow facility was calculated by multiplying HC by total scleral surface area and adjusting water viscosity to core body temperature.

results. Mean K ± 1SD in adults (>18 years) was 5.85 ± 3.89 × 10⁻¹⁸ m². K tended to be higher in pediatric donors, but there was no statistically significant age-related change. However, when all data sets were combined (n = 28), HC showed a significant decline with age. There was no significant topographical variation in HC. The potential transscleral outflow facility was 0.33 μL · min⁻¹ · mm Hg⁻¹.

conclusions. Quantifying HC may help refine ocular pharmacotherapy, as transscleral water movement increases intraocular drug elimination and impedes transscleral drug delivery. The potential scleral outflow is two to three times higher than that which occurs in vivo; hence, medical or surgical interventions that fully exploit this pathway have considerable capacity to lower intraocular pressure.

Determining scleral hydraulic conductivity (HC) is clinically relevant for several reasons. HC influences ocular fluid dynamics and intraocular pressure (IOP), as transscleral fluid movement is a well-recognized component of uveoscleral outflow. There has been renewed interest in uveoscleral outflow with the widespread introduction of antiglaucoma medications that are thought to act on this pathway, most notably the prostaglandin analogues. ¹ Glaucoma procedures such as deep sclerectomy also rely on increasing scleral permeability, as do novel approaches such as enzymatic sclerostomy. ² Scleral HC also influences ocular pharmacokinetics: transscleral delivery of drugs into the eye with periocular injection, iontophoresis, ³ or osmotic pumps ⁴ are opposed by transscleral movement of water out of the eye. Conversely, the elimination of intraocular drugs may be increased by transscleral aqueous movement. These aspects of ocular pharmacokinetics are of particular interest because of the increasing use of intravitreal injections to treat conditions such as age-related macular degeneration, macular edema, and intraocular inflammation. Transscleral water movement may also influence the resolution of choroidal fluid collections, and aggravate ocular hypotony caused by cyclodialysis clefts. Alterations in scleral conductivity may contribute to the pathogenesis of conditions such as uveal effusion syndrome. ^{5 6 7 8 9}  

Despite its clinical relevance, the only study of human scleral HC relied on just one donor. ¹⁰ A more thorough assessment of scleral HC may increase our knowledge of normal ocular fluid dynamics, guide strategies for manipulating IOP, and assist in the development of novel strategies of ocular drug delivery. It may also improve our understanding of certain ocular diseases. 

In this study, we sought to determine human HC over a range of ages, to assess the degree of topographical variation, and to estimate potential scleral outflow facility. 

Methods

Tissue Preparation and Measurement of HC

Full-thickness scleral specimens were dissected from 10 male and 8 female human donor eyes obtained from the UK. Tissue Transplant Service (Bristol, UK), in accordance with the guidelines of the Declaration of Helsinki for research involving human tissue. Mean time from death to enucleation was 20.1 ± 12.0 (SD) hours and from enucleation to experiment was 65.1 ± 19.8 hours. The tissue, which was not frozen, was transported from the eye bank on ice, and stored at 4°C in the laboratory. Experiments were conducted at 20°C rather than body temperature, as mounting the tall vertical water column in an incubator presented practical difficulties, and an increased temperature would increase the risk of evaporative loss and infection. It is also easy to adjust for water viscosity at different temperatures. Donors ranging in ages from 4 to 89 years (mean, 56.7 ± 25.9) were selected. Those with known or visible eye disease were excluded, as were those with known systemic conditions with ocular manifestations. Squares of full-thickness sclera measuring approximately 6 by 6 mm were dissected from the pre-equatorial region 8 to 14 mm posterior to the corneoscleral limbus, avoiding areas under the rectus muscles. The episclera and choroid were carefully removed by an ophthalmic surgeon (TLJ) with microsurgical forceps, spring scissors, and a dissecting microscope. Specimens with any suggestion of damage or emissary vessels were excluded. A razor blade was used to remove 1-mm triangular sections from the corners of the square specimens, and these were used to determine scleral thickness, as described later. 

The main specimens were clamped in a modified Ussing chamber with a 4-mm aperture (Figs. 1 2) . ^{11 12} The locking screws that held the assembly together were tightened with a torque-range screwdriver (RS Components, Corby, UK), by applying the minimum force necessary to prevent leakage around the specimen (approximately 30 cN/m). Studies have demonstrated the integrity of tissue clamped in this device. ^{11 12 13} Care was taken to maintain scleral hydration throughout the experiment. Figure 3shows the degree of tissue compression from clamping in the Ussing chamber and the state of hydration in the neighboring unclamped tissue held in the central aperture. 

The Ussing chamber was placed in a humidified container, and the top of the scleral specimen was covered with 250 μL of phosphate-buffered saline (PBS; sodium chloride 120 mM, potassium chloride 2.7 mM, and phosphate buffer salts 10 mM [pH 7.4] at 25°C; Sigma-Aldrich, Poole, UK). The chamber was connected to a water column filled with degassed PBS containing penicillin 100,000 IU/L, streptomycin 100 mg/L, and amphotericin B 250 μg/L (Sigma-Aldrich). The open top of the water column was covered with a water-soaked gauze pad to reduce evaporation. The integrity of the system was checked at intervals with a water-proof plastic film instead of scleral tissue, to ensure that leaks were not causing descent of the water column. The volume of each water column was determined before the start of the experiment by filling the length of the column with mercury that was then weighed on analytical scales. This was used to calculate the volume per unit length for each column. 

Column height was set at 21.4 cm (including the thin film of PBS covering the scleral surface), equivalent to a normal IOP of 15.7 mm Hg. A typical glass tube at 21.4 cm of water contained 1.13 mL. After 1 hour, the clear tubing and chamber were inspected, and if any gas bubbles were present, the apparatus was lightly tapped so that they rose to the top of the water column. The column height was then reset and the experiment commenced. The descent of the water column was determined over 24 hours with a digital micrometer attached to a sliding rule, secured alongside the glass column. Our pilot studies, and work by others, ¹⁴ suggest that there is a roughly linear relationship between pressure and flow over the range of pressure used in this study. This range extended from 15.7 mm Hg at the start of the experiment (in all specimens) to a minimum of 13.4 mm Hg at the end of the experiment (in the scleral specimen with the greatest column descent). Lee et al. ¹⁵ have shown that hydrostatic pressures <60 mm Hg do not significantly alter tissue compression or hydration. 

Measurement of Scleral Thickness

Scleral thickness was determined in the tissue removed from the main specimen before mounting it in the Ussing chamber. Sclera was fixed for 1 hour in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer containing 10 g/L calcium chloride (final pH 7.4; Sigma-Aldrich). Samples were rinsed in sucrose, postfixed in 2% osmium tetroxide (Sigma-Aldrich) in 0.2 M sodium cacodylate buffer for at least 1 hour, dehydrated in a graded series of ethanols, and fixed in epoxy resin (Araldite CY212; Agar Scientific Ltd., Cambridge, UK). The orientation of scleral specimens was preserved, and multiple semithin (1-μm) sections were cut, transferred to plain microscope slides containing a film of distilled water, flattened with trichloroethylene vapor, and stained with toluidine blue. Three sections were randomly selected, and five scleral thickness measurements were taken at equidistant points over the length of each, with a light microscope with a calibrated, sliding-graticule (Leitz, Wetzlar, Germany). The scleral thickness of each specimen was taken as the mean of these 15 readings. Olsen et al. ¹⁶ compared the thickness of fixed human scleral sections with ultrasonic measurement of unfixed tissue. They found little difference, with fresh tissue 8% thinner than fixed tissue. In a later study, ¹⁷ they used fixed sections in preference to ultrasound. 

Topographical Variation in HC

HC experiments were repeated in an additional 10 donor eyes (mean age, 54.1 ± 26.4 years; range, 12–89) with sclera at three sites for each eye. A posterior scleral trephine was taken 2 to 6 mm from the edge of the optic disc (incorporating the foveolar region), an equatorial sample at 10 to 14 mm from the edge of the optic disc, and an anterior sample at 18 to 22 mm from the edge of the optic disc. 

Data Analysis

Calculation of Specific HC.

Specific HC (K) ¹⁸ was calculated as follows:  

\[K\ {=}\ \frac{Q\ {\cdot}\ {\mu}\ {\cdot}\ L}{A\ {\cdot}\ {\Delta}P}\]

where Q is flow, determined from the descent of the water column in meters, multiplied by the constant of the individual tube (volume per unit length, in mm²), divided by time in seconds (m³ · s⁻¹); μ is viscosity of the water at 20°C (Pa · s); L is the pathlength (scleral thickness in meters); A is the surface area of the aperture (in m²); and ΔP is the mean pressure head over the duration of the experiment (in Pascals). 

Potential Scleral Outflow Facility.

The HC measurements in the topographical study were taken in an attempt to estimate the potential flow of intraocular fluid across the sclera. Mean HC was multiplied by the reported ¹⁶ total scleral surface area of the human eye (16.3 cm²). The viscosity of water under experimental conditions (1.002 × 10⁻³ Pa · s at 20°C) was adjusted to core body temperature (0.705 × 10⁻³ Pa · s at 37°C). 

Statistical Tests.

The unpaired, two-tailed t-test was used to compare means, and the Pearson r was used for linear regression. Kolmogorov-Smirnov assumptions tests were used to confirm Gaussian distribution. P ≤ 0.05 level was considered significant. 

Results

Specific HC and Scleral Thickness

The mean specific HC (K) ± 1SD was 8.54 ± 10.45 × 10⁻¹⁸ m² with a mean scleral thickness of 509 ±115 μm. There was no clear relationship between age and K (Fig. 4) , except that a scleral specimen from a 4-year-old donor was 3.7 SD above the mean. Scleral thickness showed a possible slight reduction with age, but this trend was not significant (P = 0.122). Thicker sclera tended to have higher HC (P = 0.027; Fig. 5 ). In adults older than 18 years, the mean scleral thickness was 494 ±113 μm, and the mean K was 5.85 ± 3.89 × 10⁻¹⁸ m². 

To determine whether the enucleation time might act as a confounding variable, it was compared to K. Increasing enucleation time did not result in either an increased or decreased K (linear R ² = 0.003; P = 0.829). Similar results were obtained when K was compared to the time from death to the start of the experiment (R ² = 0.036; P = 0.451). 

The mean K in adult male eyes (mean age, 63.0 ± 20.9 years) was 5.34 ± 3.28 × 10⁻¹⁸ m². The mean K in adult female eyes (mean age, 62.1 ± 21.8 years) was 6.47 ± 4.78 × 10⁻¹⁸ m². This difference was not significant (P = 0.583). The mean scleral thickness in the male specimens was 476 ± 125 μm compared with 508 ± 101 μm in the female specimens (P = 0.690). 

Topographical Variation in HC

There was no statistically significant topographical variation in HC (Fig. 6) . 

Potential Scleral Outflow Facility

To increase sample size and help estimate potential scleral outflow facility, HC from the main data set and the topographical data set (anterior specimens) were combined (n = 28; mean age 54.3 ± 26.4 years; range, 4–89). The anterior topographic specimens were used, as they were taken from the same site as the main data set. The combined mean HC was 21.0 ± 15.6 × 10⁻¹² m · s⁻¹ · Pa⁻¹ with a significant (P = 0.015) decrease with age (Fig. 7) . When the mean HC in adults (18.0 ± 11.9 × 10⁻¹⁰ cm · s⁻¹ · Pa⁻¹) was multiplied by a scleral surface area of 16.3 cm², this gave a theoretical outflow facility of 293 × 10⁻¹⁰ cm³ · s⁻¹ · Pa⁻¹, or 0.235 μL · min⁻¹ · mm Hg⁻¹. Correction for viscosity at 37°C resulted in a potential outflow facility of 0.33 μL · min⁻¹ · mm Hg⁻¹). Given that the pressure in the suprachoroidal space is 0.8 to 3.7 mm Hg below IOP, ¹⁹ then at an IOP of 15 the suprachoroidal pressure is, on average, ∼13 mm Hg. At this pressure, the potential flow across the sclera would be 4.3 μL/min. 

Discussion

The purpose of this study was to determine the HC of human sclera over a range of ages, to look for topographical variation, and to make a theoretical estimate of potential scleral outflow facility. When HC was adjusted for the thickness of each scleral specimen, the mean adult value was 5.85 × 10⁻¹⁸ m². This adjustment for pathlength gives an intrinsic HC for scleral tissue, termed specific HC, or K. ¹⁸  

There was no clear relationship between K and age, although one sample from a 4-year-old donor had a K 3.7 SD above the mean. It is possible that this represents experimental error, although there was nothing during the course of the experiment to suggest error, with a linear descent over the 24 experiment (and the subsequent 24 hours). Further, when the main data set and topography data set were combined, there was a significant reduction in HC with age. This finding is consistent with the well-known observation that sclera is more compliant in children undergoing surgery, that uveoscleral outflow is thought to be higher in young people, and that scleral composition changes with age. ^{20 21} Boubriak et al. ²² observed increased partition coefficients in sclera from young donors, but the youngest of these was aged 37 years. They also observed that hydration levels were higher in young donors. Assuming that increased hydration expands the space between scleral collagen fibrils, then this is likely to increase HC. 

HC was relatively similar over the surface of the sclera. This lack of topographical variation was perhaps unexpected, given that scleral thickness varies from the front to the back of the eye ¹⁶ and that this variation might be expected to alter HC. However, the 6-mm trephines would have included a range of scleral thickness; there was some spread in HC levels; and as Figure 5shows, the relationship between scleral thickness and HC was not necessarily constant. 

HC results were combined with the reported scleral surface area, ¹⁶ to provide a theoretical estimate of scleral outflow facility—that is, the potential flow of water over the entire scleral surface. Assuming a suprachoroidal pressure of 13 mm Hg, scleral outflow was calculated to be 4.3 μL/min. It is difficult to draw direct comparisons with other reports, as there are few studies of human scleral HC, and experimental methods differ. Fatt and Hedbys ¹⁰ measured HC in rabbit and one human specimen, with swelling pressure rather than a direct measure of scleral water flow. They found little interspecies variation and estimated transscleral outflow to be 0.53 μL/min. This was based on the assumption that the human sclera had a surface area of 11.5 cm² and a uniform average thickness of 600 μm. They also assumed that the pressure gradient was the difference between an IOP set at 18 mm Hg and atmospheric pressure. Their result can be converted into the units used in the present study, to overcome these differences in baseline assumptions, giving a K of 1.92 × 10⁻¹⁸ m², approximately 1 SD from our mean K of 5.85 ± 3.89 × 10⁻¹⁸ m². Given normal biological variability, these results appear consistent. Our findings are also consistent with K in the cornea, ²³ estimated to be from 0.5 to 10 × 10⁻¹⁸ m². 

Dan et al. ² studied water movement through bovine sclera exposed to a pressure head of 20 mm Hg, in an attempt to quantify the effect of collagenase on scleral permeability. They did not specify surface area or scleral thickness; hence, their study did not quantify either HC or K. Rudnick et al. ¹⁴ examined transscleral diffusion of ³H-water in rabbit sclera, but they measured diffusion rather than flow. This latter point is important: There are many studies ^{3 4 24 25 26} in which the diffusion of solutes across the sclera has been investigated—particularly the diffusion of therapeutic agents such as antibiotics, ^{27 28} steroids, ^{29 30 31 32 33 34} and antiangiogenic factors. ^{35 36} However, in the present study, we did not measure diffusion, but rather the flow of water across the sclera. 

Uveoscleral outflow is thought to comprise transscleral water movement (as measured in the present study), bulk flow around the emissaria, and absorption into the choroidal vessels. The relative contribution of each is not certain. Measurement of uveoscleral outflow in humans is technically difficult, but one in vivo study suggested a uveoscleral outflow of 1.5 μL/min in younger adults, decreasing to 1.1 μL/min in older subjects. ³⁷ This level is less than the potential scleral outflow estimated in the present study (4.3 μL/min). The apparent discrepancy is unsurprising, as the flow in vivo would only match the potential flow across the sclera if the aqueous had unimpeded access to the entire scleral surface. There are, however, barriers to flow as evidenced by the fact that suprachoroidal pressure is lower than IOP. ¹⁹ The first is the stromal tissue at the drainage angle. There is also evidence that the ciliary muscle acts a potential barrier, ³⁸ as does the presence of the an anatomic “compact zone” at the ora serrata. ³⁹ The behavior of eyes with cyclodialysis clefts also suggests that the potential scleral outflow capacity is not fully realized in vivo. Cyclodialysis clefts open a conduit between the anterior chamber and the suprachoroidal space, overcoming the normal anatomic barriers to uveoscleral outflow. As predicted by the current data, cyclodialysis clefts are commonly associated with a significant, often severe reduction in IOP. 

Conclusion

Scleral HC is of relevance to the treatment of increased IOP and influences the ocular pharmacokinetics of water-soluble drugs. Despite this, current knowledge is limited to the study of one human eye, with an indirect measure of transscleral fluid flow. The present study used a direct measure of HC and applied it to a range of ages and topographical locations. The results should improve our understanding of ocular fluid dynamics—particularly uveoscleral outflow—and diseases characterized by alterations in transscleral water movement. The data also suggest that the potential flow of water across the sclera is considerably more than occurs in vivo, and hence medical or surgical interventions that fully exploit this potential have considerable capacity to lower IOP. 

 Supported by the Special Trustees of St. Thomas’ Hospital, the Allerton Fund, and German Research Council Grant DGHi 758/1-1.

 Submitted for publication April 3, 2006; revised June 6, 2006; accepted September 22, 2006.

 Disclosure: T.L. Jackson, None; A. Hussain, None; A. Hodgetts, None; A.M.S. Morley, None; J. Hillenkamp, None; P.M. Sullivan, None; J. Marshall, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

The experimental setup. A, metal chamber; B, rubber O-ring; C, scleral tissue held in a tissue clamp (see Fig. 2 ); D, metal compression ring; E, locking screw; F, three-way tap connected to a syringe of degassed saline; G, glass water column; and H, digital micrometer.

Figure 2.

 

Schematic showing the Perspex tissue clamp (C) in Figure 1 . A, locking screw; B, tissue clamp; C, sclera.

Figure 3.

 

Photomicrograph of a section of human sclera after clamping. The tissue was fixed after a 24-hour experiment and shows the transition zone between tissue compressed in the clamp and the uncompressed tissue in the central aperture of the Ussing chamber. Despite tissue compression, the collagen fibrils remained intact and the area of compression extended for a minimal distance into the 4-mm central aperture that was used to assess HC.

Figure 4.

 

Age plotted against the specific HC (K) of human sclera in 18 donor eyes.

Figure 5.

 

HC compared with scleral thickness in 18 donor eyes (R ² = 0.270; P = 0.027).

Figure 6.

 

Mean scleral HC (±1SD) in 10 donor eyes with anterior, equatorial, and posterior trephines taken from each eye. The difference between equatorial and both anterior and posterior locations was not significant (P = 0.28 and P = 0.27, respectively).

Figure 7.

 

Plot of HC versus age, using anterior scleral specimens and all data sets combined (n = 28). The data show a reduction in HC with age (R ² = 0.206, P = 0.015).

The authors thank Richard J. Antcliff for providing the schematic drawings (Figs. 1 and 2) .