Hydrostatic Pressure of the Suprachoroidal Space

Home , Ora serrata, Outer plexiform layer, Pars plana

Kozuyuki Emi, Jonathan E. Pederson, and Carol B. Toris

The hydrostatic pressure of the suprachoroidal space was measured in 18 cynomolgus monkey eyes by one of two methods: (1) direct cannulation, or (2) silicone sponge implantation. The intraocular pressure (IOP) and suprachoroidal pressure were monitored simultaneously with the IOP being held at various levels between 5 and 60 mm Hg. In eyes with direct cannulation, at an IOP of 15 mm Hg, the pressure in the anterior suprachoroidal (supraciliary) space was 0.8 ± 0.2 mm Hg (n = 6, mean ± SE) below the IOP, but the posterior suprachoroidal pressure was 3.7 ± 0.4 mm Hg (n = 8) below the IOP. The suprachoroidal pressure in eyes with silicone sponge implant was 4.7 ± 0.6 (n = 7) mm Hg below the IOP. A change in IOP produced a corresponding change in the supraciliary space pressure. However, the pressure difference between the anterior chamber and the posterior suprachoroidal space increased at higher IOP. This pressure differential is the driving force for uveoscleral outflow. Invest Ophthalmol Vis Sci 30:233-238,1989

The suprachoroid lies between the choroid and the conformed to the ARVO Resolution on the Use of sclera and is composed of closely packed layers of Animals in Research. Each monkey was anesthetized long pigmented processes derived from each tissue.1 with intramuscular ketamine hydrochloride (20 The suprachoroidal space is a potential space provid- mg/kg) and intravenous sodium pentobarbital (25 ing a pathway for uveoscleral outflow and becomes mg/kg). Pupils were dilated with phenylephrine hy- an actual space in choroidal detachment. The hydro- drochloride and cyclopentolate hydrochloride. static pressure in the suprachoroidal space is an im- portant parameter for understanding intraocular Direct Cannulation with Polyethylene Tubing fluid dynamics and the mechanism of choroidal de- A deep lateral canthotomy and a temporal peri- tachment. The suprachoroidal hydrostatic pressure tomy were performed, followed by two full-thickness was first measured in 1961 by van Alphen, who in- scleral incisions 1.0 mm wide, one made radially at serted a needle into the suprachoroidal of cats. He the inferotemporal sclera 7 mm posterior to the reported that the suprachoroidal space pressure limbus and another made circumferentially at the su- (SCSP) was 1 or 2 mm Hg less than the intraocular perotemporal sclera 2 mm posterior to the limbus. pressure and that the pressure difference was unaf- After removal of a small amount of aqueous fluid fected by intraocular pressure.2 with a 3 mm needle knife, a tunnel was made into the In this study, the suprachoroidal space hydrostatic suprachoroidal space with a 0.5 mm fine spatula. A pressure (SCSP) was measured in monkeys with two tapered polyethylene tube (O.D. about 0.25 mm, I.D. different methods in order to reevaluate the intersti- about 0.05 mm), was inserted through the radial tial fluid hydrostatic pressure and to investigate the scleral incision into the posterior suprachoroidal relationship between suprachoroidal hydrostatic space (beneath the posterior pole) and fixed with 8-0 pressure and intraocular pressure. silk suture. The localization of the cannula was ob- served as a elevated retinal reflex line on the posterior Materials and Methods fundus. Another tapered cannula was introduced Nine cynomolgus monkeys of either sex weighing through the anterior scleral incision into the supracil- 3.2 to 4.5 kg were used in this study. Animal usage iary space and fixed as well. A 27-gauge needle connected to polyethylene tubing was inserted into the anterior chamber to monitor the intraocular pres- From the Department of Ophthalmology, University of Minne- sure. Another 27-gauge needle, which was connected sota, Minneapolis, Minnesota. to a fluid reservoir filled with sterile physiological Presented in part at the Association for Research in Vision and Ophthalmology meeting, May 2, 1988, Sarasota, Florida. saline, was introduced into the vitreous cavity Supported by NIH grant EY-03277. through the pars plana in order to control the intra- Submitted for publication: July 6, 1988; accepted September 2, ocular pressure. Each tubing was filled with sterile 1988. saline and connected to a 4-channel pressure trans- Reprint requests: Kazuyuki Emi, MD, Department of Ophthal- ducer (7754B, Hewlett Packard, Waltham, MA). The mology, University of Minnesota, Box 493 UMHC, Minneapolis, MN 55455. hydrostatic pressures at four different locations (ante-

230

Downloaded from iovs.arvojournals.org on 09/26/2021 234 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / February 1989 Vol. 30

anterior chamber

posterioi suprachoroidal space Fig. 1. Cannula and needle placement for direct cannulation Fig. 3. Needle placement for sponge implant method. method.

rior chamber, vitreous cavity, supraciliary space, su- cone sponge, 5X7X7 mm, (Scleral Sponge II, Dow prachoroidal space) were monitored simultaneously Corning, Midland, MI) was hollowed out (the inner (Fig. 1). The cannula introduced into the posterior cavity was about 3X5X5 mm.). Under intravenous suprachoroidal space was visible in the fundus (Fig. sodium pentobarbital anesthesia, a circumferential 2). For the first hour after the cannula placement, the scleral incision was made 7 mm wide on the temporal pressure measurement was performed without infu- surface, 7 mm from the limbus, avoiding the vortex sion. After the pressures became stable, the IOP was veins. After 0.5 ml of vitreous was aspirated through artificially maintained at several levels (10 mm Hg to the nasal pars plana, the choroidal tissue was gently 60 mm Hg) for 15 to 30 min by changing the height separated from the sclera using a dull spatula (0.5 of the reservoir bottle. Then the fluid infusion was mm). The silicone sponge was implanted in the su- stopped to allow recovery of spontaneous IOP. Fi- prachoroidal space and the scleral incision was su- nally, the IOP was lowered to 5 mm Hg by aspirating tured tightly with 8-0 silk sutures. At least 6 weeks aqueous humour from the cannula inserted into the after the sponge implantation, a 27-gauge needle was anterior chamber. introduced inside the implanted sponge capsule. Then, 27-gauge needles were introduced into the anterior chamber and the vitreous cavity (Fig. 3). The Silicone Sponge Implantation hydrostatic pressure measurements were made in the After the cannulation study, the right eyes of eight same manner as in the cannula study. At the termina- monkeys were used for sponge implantation. A sili- tion of the pressure measurements, about 50 jtl of sponge capsule fluid was aspirated and a plasma sam- ple was collected. The protein concentrations of the plasma and capsule fluid were measured by the Lowry method.3 Four eyes with sponge implants were enucleated and examined histologically.

Results Spontaneous Hydrostatic Pressure Difference About 40 min was required for the pressures to stabilize after cannulation. The intraocular pressure (IOP) stabilized at pressures between 4.5 and 13.5 mm Hg (average 9.3 mm Hg) under the intravenous pentobarbital anesthesia. As shown in Table 1, with direct cannulation, the hydrostatic pressure in the su- Fig. 2. Photograph of monkey fundus showing cannula in the praciliary space was 0.9 ± 0.2 mm Hg (mean ± SE, n posterior suprachoroidal space (arrows). = 9) less than the IOP (paired student t-test: P

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 2 SUPRACHOROIDAL SPACE PRESSURE / Emi er ol 235

Table 1. Spontaneous pressure measurements (mm Hg) Anterior Posterior cannula cannula Sponge IOP 9.4 ± 0.9 (9)* 9.2 ±0.9(10)t 9.3±1.2(7)f SCSP 8.4 ± 0.9 (9)* 5.8±0.5(10)t 5.1 ±1.2 (7)* IOP-SCSP 0.9 ± 0.2 (9)§" 3.5 ± 0.5 (10)§ 4.2 ±0.5(7)"

Each value indicates mean ± SE. ( ) = n. IOP: intraocular pressure. SCSP: suprachoroidal space pressure. *P < 0.05, ff < 0.001 (paired student t- test). §•"/> < 0.001 (unpaired student t-test).

< 0.05). The hydrostatic pressure in the posterior suprachoroidal space was 3.5 ± 0.5 mm Hg (n = 10) less than the IOP (paired student t-test: P < 0.001) and significantly lower than the hydrostatic pressure in the supraciliary space (unpaired student t-test: P < 0.001). The hydrostatic pressure in the sponge was 4.2 ± 0.5 mm Hg (n = 7) less than the IOP (P < 0.001) and also significantly lower than the supra- INTRAOCULAR PRESSURE (mm Hg) ciliary pressure (P < 0.001). There was no significant Fig. 4. Correlation between intraocular pressure and supraciliary difference between the hydrostatic pressure in the or suprachoroidal space pressure. Closed squares (sponge): mean pressure inside sponge. Open circles (posterior cannula): mean posterior cannula and the pressure in the sponge. pressure in cannula in posterior suprachoroidal space. Crosses (anterior cannula): mean pressure in cannula in supraciliary space. Effect of Raising the IOP on the SCSP Vertical bars indicate standard error. Solid straight line has unity slope. After the pressures stabilized spontaneously, the reservoir bottle was elevated to increase the IOP to a higher level. As the IOP increased, the supraciliary space pressure (anterior cannulation), the suprachoroidal space pressure (posterior cannulation) and the suprachoroidal pressure (sponge implantation) also increased but the magnitude of increase varied. An increase in IOP produced almost the same increase in the supraciliary space pressure but a smaller increase was measured in the SCSP with both posterior cannulation and sponge implantation (Fig. 4). By raising the IOP from 15 mm Hg to 60 mm Hg, the pressure difference between the IOP and the suprachoroidal hydrostatic pressure increased from 3.7 ± 0.4 mm Hg (n = 8) to 10.2 ± 1.3 mm Hg (n = 7) in the posterior cannulation (P < 0.001) and from 4.7 ± 0.6 mm Hg (n = 7) to 10.2 ± 0.8 mm Hg (n = 7) in the sponge implantation {P < 0.001, Fig. 5).

Effect of Lowering the IOP on the SCSP One hour after the infusion was stopped, the pressures returned to nearly the same levels as in the INTRAOCULAR PRESSURE (mm Hg) spontaneous measurement along a similar curve to Fig. 5. Correlation between intraocular pressure (IOP) and pres- that obtained by raising the IOP. However, when the sure difference between IOP and supraciliary or suprachoroidal IOP was lowered suddenly from high to low levels (eg, space pressure. Closed squares (sponge): mean pressure difference from 60 to 5 mm Hg) by aspirating aqueous humour, between IOP and inside sponge. Open circles (posterior cannula): the large pressure difference generated by the high mean pressure difference between IOP and cannula in posterior suprachoroidal space. Crosses (anterior cannula): mean pressure IOP decreased only by 1 or 2 mm Hg in the first 10 difference between IOP and cannula in supraciliary space. Vertical min after aspiration. The SCSP usually became nega- bars indicate standard error.

Downloaded from iovs.arvojournals.org on 09/26/2021 236 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / February 1989 Vol. 30

Fig. 6. Photo micrograph of monkey eye with chronic sponge implant. The choriocapillaris overlying the sponge is partly compressed or obliterated in the periph- eral portion, with overlying cystic degeneration in the outer plexiform layer of the retina. The retinal pigment epithelium is continuous (X48).

tive to atmospheric pressure at low IOP levels for but the retinal pigment epithelium overlying the more than 20 min. This large pressure difference re- sponge was continuous with a virtually normal ap- turned very slowly to the spontaneous level. When pearance. The choriocapillaris overlying the sponge the IOP was maintained at 5 mm Hg, the pressure was partly compressed and obliterated in the periph- difference between the IOP and the supraciliary space eral portion, where the sensory retina had a cystic pressure was 0.8 ± 0.2 mm Hg (n = 5), similar to that degeneration in the outer plexiform layer, as shown found in the normal (15 mm Hg) IOP studies (Fig. 5). in Figure 6. However, the pressure difference between the IOP and the SCSP changed significantly from the 15 mm Discussion Hg IOP studies (Fig. 5). In the posterior cannulation, the pressure difference was 1.7 ± 0.2 mm Hg (n = 7) Technical Problems vs 3.7 ± 0.4 mm Hg (n = 8) at the normal IOP (P The suprachoroidal space is normally narrow < 0.001). In the sponge implantation, the pressure (about 30 nm thick)1 and forms a transitional zone difference was 3.4 ± 0.6 mm Hg (n = 7) vs 4.7 ± 0.6 between the choroid and sclera. It contains layers of mm Hg (n = 7) at the normal IOP (P < 0.001). long pigmented collagenous processes forming a closely packed collagen mesh.1'4 Van Alphen mea- Protein Concentration of the Fluid in the Sponge sured the hydrostatic pressure of the suprachoroidal space in cat eyes.2 He introduced a 27-gauge needle The protein concentration of the sponge capsule into the suprachoroidal space, making an "absorb- fluid was 23.7 ± 3.0 mg/ml (mean ± SE, n = 6) and able space" by injection of a minute bolus of fluid. He the protein concentration of the plasma was 82.1 measured the pressure which fell to a plateau level ± 2.5 mg/ml (n = 6). (equilibrated pressure). However, the plateau was not always maintained consistently. Once the injected Histological Examination fluid was absorbed, the interstitial pressure measure- Four sponge capsules examined were situated be- ment became very unstable because of: (1) plugging tween the ora serrata and the equator in the supra- of the needle tip by tissue; (2) movement of the nee- choroidal space and were well encapsulated with fi- dle in the tissue; or (3) a ball-valve effect occurring at brous tissue. The interior of three capsules contained the tip of the needle. These problems has been dis- little tissue ingrowth. In one case, in which pressure cussed by Guyton et al5 with reference to pressure measurements were unsuccessful, the implanted measurements in the subcutaneous tissue. He em- sponge capsule was collapsed and filled with the con- phasized that the "equilibrated pressure" measured nective tissue. The choroid was slightly compressed by the needle is actually the total tissue pressure,

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 2 SUPRACHOROIDAL SPACE PRESSURE / Emi er ol 237

which includes solid tissue pressure in addition to usually higher than the interstitial fluid pressure. An- interstitial fluid pressure. The resulting measure- other possible explanation is the difference of IOP ments of Van Alphen would then be an overestimate under anesthesia. The IOP usually drops under gen- of the suprachoroidal space hydrostatic pressure. eral anesthesia but the anesthesia used by Van Al- In the present study, two different methods were phen and the IOP drop was not reported. In the cur- used to measure the interstitial fluid pressure of the rent study, the stabilized IOP under intravenous so- suprachoroidal space. In the cannulation method, the dium pentobarbital anesthesia ranged from 4.5 to tapered cannula was introduced deeply into the pos- 13.5 mm Hg. It is possible that deeper anesthesia terior suprachoroidal space. The plugging of the can- causes a larger drop in IOP. Van Alphen reported that nula tip often occurred and the "absorbable space" the SCSP simply followed the IOP. However, it is made by a bolus fluid injection was not useful for clear from the present study that the pressure differ- continuous pressure monitoring. When the cannula ence is significantly correlated to the level of IOP. was withdrawn slightly forming a "definite space," Species differences may also play a role in explaining plugging of the cannula tip was avoided. This slight these results. cannula manipulation did not cause detectable pressure leakage. This technique was very successful for Hydrostatic Pressure Differential in the Uveal Tract continuous measurement of SCSP above 5 mm Hg. This method has other potential problems such as During spontaneous measurement, the hydrostatic mechanical choroidal irritation or choroidal bleed- pressure in supraciliary space (anterior cannula) was ing. lower than in the anterior chamber (P < 0.05) and higher than in suprachoroidal space (posterior can- In the sponge implantation method, the implanted nula or sponge implant, P < 0.001). Therefore, there sponge was encapsulated with a thin layer of fibrous is a pressure differential from the anterior chamber to tissue and fixed securely in the suprachoroidal space. the supraciliary space and from the supraciliary space Partial choroidal compression and obliteration of the to the suprachoroidal space, which would be the choriocapillaris was noted, but the retinal pigment driving force for uveoscleral outflow. This also sug- epithelium was continuous. The inside of the hol- gests differential flow resistance in the uveal tract. lowed sponge, which was filled with interstitial fluid, There is a minimal barrier between the anterior was large enough to easily introduce a 27-gauge nee- chamber and the supraciliary space, but the ciliary dle. Continuous pressure measurements were suc- muscle does restrict the uveoscleral outflow since pi- cessful in most cases even at very low IOP. This 7 locarpine decreases uveoscleral outflow. It has been method is comparable to the implantation of a perfo- recently reported that the posterior uvea is separated rated plastic capsule in subcutaneous tissue.6 The in- from the anterior uvea by a "compact zone" at the side of the sponge communicates with the ambient ora serrata, which acts as an anterior structural suprachoroidal interstitial space and is free from the boundary, and by the sheet-like fibroblastic lamellae solid tissue pressure because of the sponge shell. of the lamina fusca, which are closely apposed along Therefore, the SCSP measured by this sponge 4 the choroid-scleral transition. Since the pressure de- method should be the interstitial fluid pressure. This cline from the supraciliary space to the suprachoroi- sponge implantation technique also has some possi- dal space is much larger than that from the anterior ble artifacts. One is a choroidal inflammation which chamber to the supraciliary space, the drainage of might be caused by mechanical choroidal separation aqueous humor into the suprachoroidal space may be or tissue reaction to the implanted material. Another more restricted by a compact zone and fibroblastic possible artifact is that the formed capsule membrane lamellae than by the ciliary muscle in the normal eye. might act as a semipermeable membrane. However, This is consistent with the report that the uveoscleral the connective tissue lining is permeable to protein outflow increases in eyes with experimental cilio- and the effectiveness of this membrane as an osmom- 5 choroidal detachment made by suprachoroidal injec- eter is very small. 5 tion of autologous serum.

Normal Pressure Difference Relationship between IOP and SCSP The spontaneous pressure difference between the Aqueous humor that flows into the suprachoroidal IOP and the SCSP obtained in this study was 3.5 mm space, leaves this space via the scleral channels9 or via Hg with the cannula method and 4.2 mm Hg with the the choriocapillaris.10 Such fluid inflow or outflow sponge method. These values are about twice those seems to be closely correlated with the hydrostatic reported by van Alphen.2 As mentioned above, this pressure in the suprachoroidal space. In this study, a might be because the pressure measured by needle is higher IOP generated a larger pressure difference be-

Downloaded from iovs.arvojournals.org on 09/26/2021 238 INVESTIGATIVE OPHTHALMOLOGY & VI5UAL 5CIENCE / February 1989 Vol. 30

tween the suprachoroidal space and the anterior choroidal space is the driving force for uveoscleral chamber, and a lower IOP generated a smaller pres- outflow. sure difference. Therefore, the change in IOP affects the inflow of aqueous humor and/or the outflow of Key words: suprachoroidal space, interstitial fluid pressure, suprachoroidal fluid. When the IOP decreased sud- oncotic pressure, intraocular pressure, cynomolgus mon- denly from high to low level, the large pressure differ- key ence generated by the high IOP did not decrease quickly and the SCSP sometimes became negative to Acknowledgments atmospheric pressure. Therefore, the main cause of Invaluable assistance was provided by Karlan Hunt. the pressure drop in the suprachoroidal space is not fluid leakage through scleral channels but fluid ab- References sorption by colloid osmotic gradient across the chor- 1. Hogan MJ, Alvarado JA, and Weddell JE (Eds.): Choroid, In iocapillaris. In fact, the protein concentration was Histology of the Human Eye. Philadelphia, W.B. Saunders, 23.7 mg/ml in the sponge and 82.1 mg/ml in the 1971, pp. 320-392. plasma. This large concentration difference is consis- 2. Van Alphen GWHM: On emmetropia and ametropia. Oph- tent with that found with immunoglobulins in the thalmologica 142(Suppl):47, 1961. human choroid" and that found with extra vascular 3. Lowry OH, Rosenbrough NJ, Farr AL, and Randall RJ: Pro- albumin in the monkey choroid.12 This protein con- tein measurement with the folin phenol reagent. J Biol Chem 193:265, 1951. centration difference is large enough to generate the 13 4. Kelly DE, Hageman GS, and McGregar JA: Uveal compart- osmotic absorptive force across the choriocapillaris. mentalization in the hamster eye revealed by fine structural In normal circumstances this colloid osmotic differ- and tracer studies: Implications for uveoscleral outflow. Invest ence is nearly constant. However, the choriocapillaris Ophthalmol Vis Sci 24:1288, 1983. is permeable to serum albumin and proteins in a mo- 5. Guyton AC, Taylor AE, and Granger HJ (Eds.): Measurement lecular sieving manner,14 indicating that the high hy- of the different types of tissue pressure. In Circulatory Physiol- ogy II: Dynamics and Control of the Body Fluids. Philadel- drostatic pressure in the suprachoroidal space inhibits phia, W.B. Saunders, 1975, pp. 22-52. the permeation of serum protein, facilitates the out- 6. Guyton AC: A concept of negative interstitial pressure based flow of the suprachoroidal fluid and reduces the pro- on pressures in implanted perforated capsules. Circ Res tein concentration in the suprachoroidal space. The 12:399, 1963. result is a larger pressure difference between the su- 7. Bill A and Wahlinder PE: The effects of pilocarpine on the prachoroidal space and the anterior chamber. On the dynamics of aqueous humor in a primate (Macaca irus). Invest Ophthalmol 5:170, 1966. other hand, a low hydrostatic pressure in the supra- 8. Pederson JE, Gaasterland DE, and MacLellan HM: Experi- choroidal space induces the permeation of serum mental ciliochoroidal detachment: Effect on intraocular pres- protein, decreases the outflow of the suprachoroidal sure and aqueous humor flow. Arch Ophthalmol 97:536, 1979. fluid and increases the protein concentration in the 9. Bill A: The aqueous humor drainage mechanism in the cyno- suprachoroidal space. The result is a smaller pressure molgus monkey (Macaca irus) with evidence for unconven- tional routes. Invest Ophthalmol 4:911, 1965. difference between the suprachoroidal space and the 10. Pederson JE, Gaasterland DE, and MacLellan HM: Uveo- anterior chamber. In the present study, the increase scleral aqueous outflow in the rhesus monkey: Importance of in IOP generated the larger pressure difference be- uveal resorption. Invest Ophthalmol 16:1008, 1977. tween the anterior chamber and the suprachoroidal 11. Allansmith MR, Whitney CR, McClellan BH, and Newman space. Nevertheless, the rate of uveoscleral outflow is LP: Immunoglobulins in the human eye. Arch Ophthalmol 89:36, 1973. reported to be relatively constant even if the IOP 1516 12. Pederson JE, Tons CB, Rice TJ, and Gregerson DS: Determi- changed, except at the extremely low IOP. This nation of the extravascular albumin concentration of the uvea suggests that the IOP may also affect the pressure using the ELISA method. ARVO abstracts. Invest Ophthalmol difference by changing the flow resistance of uveo- Vis Sci 29(Suppl):324, 1988. scleral outflow. 13. Prather JW, Gaar KA, and Guyton AC: Direct continuous recording of plasma colloidal osmotic pressure of whole blood. J Appl Physiol 24:602, 1968. In summary, the hydrostatic pressure difference 14. Chylack LT and Bellow AR: Molecular sieving in supracho- between the anterior chamber and the suprachoroidal roidal fluid formation in man. Invest Ophthalmol Vis Sci space is significantly correlated to the intraocular 17:420, 1978. pressure. This hydrostatic pressure difference is gen- 15. Bill A: Conventional and uveo-scleral drainage of aqueous hu- erated mainly by the colloid osmotic absorption of mour in the cynomolgus monkey (Macaca irus) at normal and high intraocular pressures. Exp Eye Res 5:45, 1966. the choroidal vessels and partly by the outflow of 16. Tons CB and Pederson JE: Effect of intraocular pressure on fluid across the sclera and emissaria. The pressure uveoscleral outflow following cyclodialysis in the monkey eye. differential from the anterior chamber to the supra- Invest Ophthalmol Vis Sci 26:1745, 1985.

Downloaded from iovs.arvojournals.org on 09/26/2021