Abstract
Purpose:
Develop a custom-made injector system for delivery of aqueous solutions to the suprachoroidal space (SCS). And to test the effect of intraocular pressure (IOP), volume, and location on distribution.
Methods:
SCS injectors were produced in the laboratory using readily available components. 50–100 μl of either fluorescein sodium, and/or indocyanine green (ICG) dye solutions were injected 6 mm posterior to the corneal limbus in either the superotemporal or inferonasal quadrants of the eyes of albino rabbits. Coverage of the injected dyes was documented in vivo. Distribution of 50- and 100-μl single-injection volumes were compared, as were 50-μL injections in two opposing quadrants. The vasoconstrictor endothelin-1 (ET-1) was injected into some eyes as a test of efficacy. To determine the possible effects of IOP on injection coverage, both eyes of 6 rabbits were injected with ICG in a crossover design with the IOPs of the eyes held manometrically at either 10 or 35 mmHg.
Results:
Successful SCS injections, as determined by fundus imaging, were done in more than 40 rabbit eyes [as well as 40 non-human primate (NHP) eyes in separate studies]. 100-μL single injections provided better coverage than 50 μl, but the most complete coverage was with two 50-μL injections in opposite quadrants, although gaps remained. No statistically significant difference in coverage between IOPs of 10 and 35 mmHg was observed.
Conclusions:
Construction of a reliable custom-made injector system is feasible. IOP at the pressures measured did not appear to affect coverage of the suprachoroidally injected ICG dye solution.
Keywords: suprachoroidal, injection, distribution, intraocular pressure
Introduction
There has been considerable interest in suprachoroidal delivery of drugs to the eye. This represents a new route of ocular treatment. Traditional approaches have included topical, subconjunctival, sub-Tenon’s, intravitreal or retrobulbar. Recently, a subretinal pharmaceutical for gene delivery has also been approved for human use. 1 Suprachoroidal delivery is directed towards the potential space between the highly vascularized ocular choroid and the adjacent sclera. There are only a few fixed points of attachment between the choroid and sclera including the short posterior ciliary arteries at the macula and the vortex veins in the periphery. Thus, at least in theory, aqueous solutions injected suprachoroidally should be distributed widely.
Early attempts at suprachoroidal injection employed a small bore (e.g., 30-gauge) hypodermic needle directed tangentially through the conjunctiva and sclera into the choroid. The injected fluid then inflates the suprachoroidal space (SCS). However, such an approach is fraught with complications including hemorrhage (due to the tangential path through the vascular choroid), as well as inadvertent subretinal or intravitreal injection. A related method is to make a tangential puncture partially through the sclera with a standard 30-gauge needle and then using a 34-gauge blunt needle with a 45-degree bevel connected to a Hamilton syringe to penetrate the remaining scleral fibers and access the SCS. Such a procedure has been successfully used in rats, 2,3 rabbits, 4 pigs, 2 and non-human primates (NHPs), 2 but, again, the method is technically difficult and prone to complications. Another method has been to insert catheters into the SCS, which requires a conjunctival peritomy followed by a careful incision in the sclera without severing the large choroidal vessels. The catheter is then inserted tangentially and directed posteriorly in hopes that it will remain in the SCS. Although a time-consuming and delicate procedure, it has the theoretical advantage of being able to direct the test material to a desired specific location, such as the macula. In practice, this works well for albino rabbits, 5 but the catheter is difficult to visualize in pigmented rabbits, pigs or NHPs. However, visualization has been achieved in NHPs using a catheter with a small flashing light on its tip (such as the iTrack™, Nova Eye Medical, Fremont, CA). 6,7
A major advance in suprachoroidal delivery occurred with the introduction of a small gauge hypodermic-type needle with a stop at various lengths. 8,9 The injection device is directed perpendicular to the globe and insertion made approximately 4 to 6 mm posterior to the corneal limbus in large animal eyes (rabbit, pig, dog, and NHP). The procedure is relatively simple and quick—suitable for both preclinical (i.e., animal) and clinical applications. A commercially available injector has been approved for human use (SCS Microinjector®, Clearside Biomedical, Alpharetta, GA) that has a 30-gauge sharp tip angled at approximately 30°, lengths of 900 and 1100 μm, and is used for injection of triamcinolone acetonide suspension in to the SCS (Xipere®, Clearside Biomedical, Alpharetta, GA, and Bausch & Lomb, Rochester, NY) to treat macular edema associated with uveitis. It is designed to be used as a rapid, routine clinical procedure, similar to intravitreal injections. 10,11 Luo et al. report using a suprachoroidal injector system by Chengdu Origen Biotechnology Co., Ltd. (Chengdu, China) but state that it is not commercially available. 12 Another perpendicular injector system that employs ultrasharp needles made of silicone crystal is being tested (Bella-vue by Uneedle, Enschede, Netherlands). 13 Because of the extreme sharpness of the silicone crystal, relatively high angle tips (around 45°) can be employed and still penetrate the sclera, which permits a larger diameter bore for suspensions such as triamcinolone acetonide. An additional company has developed a device and methodology to insert a small gauge needle tangentially into the SCS without need for conjunctival cutdown. This instrument utilizes a rounded tissue separator that is inserted and then retracted from the SCS prior to drug administration, which may improve suprachoroidal distribution (Everads Therapy Ltd, Tel Aviv, Israel). 14 de Smet, et al. 15 have described an SCS device that uses a 27-gauge insertion needle containing a flexible microcatheter. The needle is passed through the sclera at about 45-degree angle. Activation of a trigger then releases the microcatheter into the SCS. The microcatheter is illuminated, which helps to verify when it is correctly positioned in the SCS [Oxulumis Ophthalmic Administration Device (formerly Oxular Ltd, Oxford, UK but recently acquired by Regeneron Pharmaceuticals, Tarrytown, NY)]. Gyroscope Therapeutics (wholly owned by Novartis Pharma, Basil, Switzerland) has introduced an SCS injection system named TS-Micro. As is the case with the Everads device, TS-Micro injects tangentially, but rather than employing a spacer, a microcannula is passed directly into the SCS prior to injection. 16 A review of recent advances in suprachoroidal injection and potential new therapeutic applications can be found in Wu, et al. 17
Predecessors of the Clearside injector system have been thoroughly tested regarding drug distribution patterns within the SCS by Patel et al.8 in ex vivo human eyes and by Chiang et al. 18–21 in both in vivo and ex vivo New Zealand white rabbit eyes (as well as ex vivo human eyes). A question that still needs to be addressed is the effect of IOP on the distribution of aqueous solutions injected into the SCS in the living eye. Unfortunately, neither the Clearside, Chengdu, Uneedle, Everads, Oxulumis, nor the TS-Micro injectors are available for independent purchase for basic preclinical research. In this manuscript, we describe the construction of and engineering principles behind a simple, custom-made SCS injector system for delivery of aqueous solutions to the suprachoroidal space (SCS). And test the effect of intraocular pressure (IOP), volume, and location on the post-injection distribution of aqueous solutions.
Methods
All studies conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Visual Research and were approved by our institutional animal care and use committee.
Custom-made SCS injector.
To make the SCS injector, the 0.2 × 9 mm (34 gauge) TSK STERiJECT™, INViSIBLE NEEDLE™ (TSK, Japan), with a sharp tip beveled at 12.5°, was used. The needle was attached to a 1 mL syringe to allow for easier handling and manipulation. A 25-gauge microvitrectomy cannula (Alcon™ valved entry system, 5 mm long, Fort Worth, TX) was slid over the 34-gauge needle such that the plastic (colored) valved end was distal to the tip of the needle. Epoxy resin was applied connecting the plastic valved end of the microvitrectomy cannula to the hub of the 34-gauge needle (Figure 1). Before the resin hardened, the metal end of the cannula was adjusted under a dissecting microscope to leave a predetermined length of the 34-gauge needle tip exposed, typically 700 μm or 900 μm (Figure 2). The syringe was then attached to a Pelco® tissue rotator (Ted Pella, Inc.) to allow the epoxy resin to harden in a relatively uniform shape (Figure 3). The valve of the 25-gauge cannula held the measured distance steady while the epoxy resin hardened. After the epoxy resin hardening, the length of the needle was rechecked. Any needle whose length was not within ±30 μm of the target length was discarded. A total of more than 40 rabbits and 40 NHPs were injected in vivo. An additional 15 ex vivo injections have been performed using pig, rabbit, and human donor eyes.
Figure 1.
Custom suprachoroidal needle system. L, luer lock end of a 1 mL syringe. H, hub of a 34-gauge low dead-space needle. E, epoxy resin. V, valved head of a 25-gauge microvitrectomy cannula system. C, cannula portion of a 25-gauge microvitrectomy cannula system. T, exposed tip of 34 gauge needle (brace = 700 μm).
Figure 2.
Left frame, standard 30-gauge needle with a 12.5° beveled tip. Right frame, custom suprachoroidal 34-gauge needle (see Figure 1) with a 12.5° beveled tip. The exposed lumen of the 30-gauge needle is considerably longer than the 34-gauge needle (red ovals) such that it would necessarily span both the sclera (yellow band) and choroid with remaining exposed lumen external to the eye. If the 30-gauge needle were inserted further, the lumen would be partly in the vitreous space. The exposed lumen of the 34-gauge needle, by comparison, is short enough to be mostly blocked by the sclera (yellow band) with only the very tip of the lumen in the suprachoroidal space.
Figure 3.
Syringes and needles fastened horizontally with binder clips to a tissue rotator revolving at 10 cycles per minute to allow the epoxy resin to harden in a symmetrical fashion.
Injection procedure and imaging.
The custom SCS needles were attached to a 1-mL syringe and used to inject 50–100 μL of either fluorescein sodium (0.1%) and/or indocyanine green (ICG 0.0074%) diluted with Balanced Salt Solution (BSS®), 6 mm posterior to the corneal limbus (measured with a caliper) in either the superotemporal or inferonasal quadrants of the eyes of 6 New Zealand white (i.e., albino) rabbits (3 males and 3 females of from 6 months to one year old). The animals were anesthetized with a combination of ketamine hydrochloride (HCl) 30 mg/kg intramuscular (IM), xylazine 8 mg/kg IM, and butorphanol tartrate 0.3 mg/kg IM or buprenorphine 0.02 mg/kg-0.05 mg/kg IM or subcutaneous (SC). As necessitated, glycopyrrolate 0.1 mg/kg IM was also administered. To extend anesthesia, ketamine supplements of 10–20 mg/kg IM were administered. Coverage of the injected dyes was documented in vivo with a Heidelberg Spectralis® HRA + OCT (Heidelberg Engineering, Heidelberg, Germany) scanning laser ophthalmoscope using a 55 degree non-contact lens. 50- and 100-μL single-injection volumes were compared. 50-μL injections in two opposing quadrants were also done.
Effect of IOP on SCS injection distribution.
To test the effects of IOP on injection-volume coverage, both eyes of 6 rabbits (the same animals as were investigated in the previous sub-section) were injected superotemporally with 50 μL of ICG in a crossover design with the IOPs of the eyes held manometrically during the entire injection (via 30-gauge needle inserted into the anterior chamber connected to a reservoir) at either 10 or 35 mmHg. IOPs were verified throughout the experiment with a TONOVET® rebound tonometer (TVP; iCare Finland Oy). Coverage of the dye in the opposite (inferonasal) quadrant was measured quantitatively using a digital image intensity selection tool (Adobe Photoshop; (Adobe, San Jose, CA). Some eyes were then fixed in 4% paraformaldehyde within 10 minutes of SCS injection and euthanasia. The eyes were opened at the level of the pars plana, the retinas removed, and radial cuts made in the posterior segment to flatten the eyes. An epifluorescence microscope (Olympus BH2; Olympus America, Inc., Melville, NY) with a motorized stage and automatic image stitching (cellSens® software, Evident Scientific, Waltham, MA) ) was used to make composite separate images of the fluorescein and ICG distribution. The images were then combined digitally.
Effect of SCS injection on choroidal blood flow.
As part of another experiment (using the Clearside Biomedical SCS Microinjector®, Alpharetta, GA), 11 pigmented New Zealand red rabbits were pre-anesthetized with: midazolam 1–2 mg/kg IM, ketamine HCl 30 mg/kg IM, and buprenorphine 0.01–0.05 mg/kg IM or subcutaneous (SC). Animals were intubated, ventilated mechanically, and anesthetized with an oxygen/isoflurane mixture (3–5% isoflurane for induction, 1–3% isoflurane, or to effect, for maintenance). With the use of isoflurane anesthesia, supplementation with ketamine HCl up to 20 mg/kg IM and/or buprenorphine 0.02–0.05 mg/kg IM or SC was required to ensure stage III, plane 2 level of anesthetization. Heart rate and blood per cent oxygen saturation were monitored with a pulse oximetry device, ventilation efficacy with capnographic end-tidal CO2 and inspired CO2, and arterial blood pressure with Doppler sphygmomanometry. Rectal or oral temperature was maintained at 37º C to 40º C with a water circulating heating pad. Fluid supplementation was maintained throughout the procedure with Lactated Ringer’s solution 5–10 mL/kg/hour IV with 5-mL/kg IV boluses (as needed) to support and control the arterial blood pressure. Endothelin-1 (ET-1) (concentrations varying from 9.4×10−7 M to 2.5×10−6 M diluted with sterile water was injected into the SCS of the right eyes using microinjectors (Clearside Biomedical, Alpharetta, GA), either as a single injection of 100 μL or two injections of either 50- or 100-μL in separate locations. Similarly, sterile water vehicle alone was injected into the control left eyes. 10 to 20 minutes following the SCS injections, a thoracotomy was performed, and 6 million 15-μm fluorescent polystyrene microspheres (FM) were injected into the left ventricle of the heart. During the FM injection and for two minutes after the completion of the FM injection, an arterial reference blood sample (1 mL/minute) was collected. The animal was immediately euthanized with an overdose of Euthasol® and the eyes were enucleated. The eyes were fixed for 2 days in 4% paraformaldehyde. The globes were then bleached with potassium permanganate solution, flat mounted, and photomicrographed. Image analysis was used to locate each microsphere as previously described in detail. 22,23
Results
Custom-made SCS injector subjective results.
We have used our SCS injector on several studies, some of which employed good laboratory practice (GLP) protocols as part of pre-clinical drug development. Of the approximately 40 in vivo rabbit, and 40 NHP [rhesus or cynomolgus macaque (Katz AW, et al., ARVO Abstract #2623, 2023)] SCS ocular injections, there were only 2 or 3 occasions during which the SCS needle needed to be slightly repositioned. However, there was no detectable leakage of the injection solution prior to repositioning. There was no instance of inadvertent intravitreal injection. Due to the challenges in imaging the SCS in NHPs as well as confidentiality issues related to industry sponsored trials, the remainder of the Results section will concentrate on our work with rabbits.
Rate of SCS injected solution spread.
To determine the rate of SCS spread, a video was made of the fundus of an albino rabbit right eye as it was being injected with 50 μL ICG in BSS®. The injection site was 6 mm posterior to the corneal limbus in the superotemporal quadrant. The initial movement of the dye was towards the periphery but then migrated posteriorly, presumably after it reached the scleral spur. Nearly all the spread of the injected ICG dye solution was completed in approximately one minute (Figure 4 and supplemental video).
Figure 4.
Series of black-and-white still images taken from a video showing the fundus of a rabbit right eye injected with 50 μL ICG in BSS®. The injection site was 6 mm posterior to the corneal limbus in the superotemporal quadrant. The initial movement of the dye is towards the periphery but then migrates posteriorly, presumably after it has reached the scleral spur. The optic nerve and medullary ray appear dark. The short posterior ciliary artery insertion points along the visual streak fluoresce strongly with ICG (bright white spots in the last frame). Nearly all the spread of ICG solution was completed by the last frame (45.7 seconds).
Effect of SCS injection volume.
Comparisons were made of single injections of 50 μL and 100 μL of ICG in BSS® in 6 albino rabbits One eye was injected 6 mm posterior to the corneal limbus in the superotemporal quadrant with 50 μL. The fellow eye of the same rabbit was also injected 6 mm posterior to the corneal limbus in the superotemporal quadrant with 100 μL. There was greater coverage with 100 μL but ICG did not reach the inferonasal SCS. IOPs were 15 mmHg in both eyes prior to injection (Figure 5). Better coverage was obtained with two separate 50 μl injections—one superotemporal and one inferonasal. In the representative example shown in Figure 6, ICG was used for the superotemporal injection and fluorescein sodium for the inferonasal one. When digital composite images were made, it was evident that, although the coverage was extensive, some overlapped areas and some non-stained areas were present.
Figure 5.
Comparison of single injections of 50 μL vs 100 μL of ICG in BSS®. Left frame, right eye injected 6 mm posterior to the corneal limbus in the superotemporal quadrant with 50 μL. Right frame, left eye of same rabbit also injected 6 mm posterior to the corneal limbus in the superotemporal quadrant with 100 μL. There is greater coverage with 100 μL but the ICG solution did not reach the inferonasal suprachoroidal space. IOPs were 15 mmHg in each eye prior to injection.
Figure 6.
Suprachoroidal coverage following two 50-μL injections in the same right eye. Frame A, ICG injected superotemporally (digitally colored red). Frame B, Fluorescein sodium injected inferonasally (digitally colored green). Frame C, Digital merge of Frame A and Frame B images. SCS areas of ICG and fluorescein sodium overlap appear yellow (labeled “O”) and non-stained areas not covered by either dye appear dark (labeled “S”).
Peripheral extent of SCS injected solution spread.
Because fundus imaging in vivo with Spectralis scanning laser ophthalmoscopy does not capture ICG or fluorescein sodium in the far periphery in rabbits, some animals were sacrificed immediately after injection. The eyes were removed and fixed in 4% paraformaldehyde. When opened and imaged with an epifluorescence microscope, choroidal and scleral staining were still present although the borders of the stained areas were somewhat indistinct. However, it was apparent that the staining extended to the far periphery (Figure 7).
Figure 7.
Composite image of a postmortem left eye with the retina removed and flattened by radial cuts with an epifluorescence microscope and motorized stage. The ICG (red) and fluorescein sodium (green) staining of the sclera are evident as it was with in vivo SCS imaging. Although the borders of the staining patterns are indistinct, staining extends to the far periphery.
Effect of IOP on SCS injected solution spread.
The effect of IOP controlled manometrically on the distribution of aqueous solutions injected into the SCS was determined quantitatively in 6 albino rabbits in a crossover design. Because the entire coverage of the suprachoroidally injected dye cannot be imaged in vivo, a limited area that can be fully imaged was demarcated in the quadrant opposite (inferonasal) to the injection quadrant (superotemporal). The area of measurement (red border) consisted of a straight line starting at the edge of the optic nerve and perpendicular to the medullary ray. A second bounding line was drawn perpendicular to the first extending nasally. The extent of dye coverage was identified objectively using a digital image intensity selection tool (Figure 8). Box plots were made of the distribution in the inferonasal quadrant (opposite the superotemporal injection site) of 50 μL fluorescein sodium with the intraocular pressures manometrically held to either 10 mmHg or 35 mmHg. This represented only a small portion of the total coverage of the aqueous solution (see Figure 8). Although there is a trend towards decreased coverage at the higher pressure, it was not statistically significant (Figure 9 and Table 1).
Figure 8.
Example of method for quantitative determination of the spread of 50 μL ICG injected in the left eye. Because the entire coverage of the suprachoroidally injected dye cannot be imaged in vivo, a limited area that can be fully imaged is demarcated in the quadrant opposite (inferonasal) to the injection quadrant (superotemporal). The area of measurement (red border) consists of a straight line starting at the edge of the optic nerve and perpendicular to the medullary ray. A second bounding line is drawn perpendicular to the first extending nasally. The extent of dye coverage is identified objectively using a digital image intensity selection tool.
Figure 9.
Box plots of the distribution in the inferonasal quadrant (opposite the superotemporal injection site) of 50 μL ICG with the intraocular pressures manometrically held to either 10 mmHg or 35 mmHg. This represents only a small portion of the total coverage of the solution (see Figure 8), which accounts for some of the variability. Although there is a trend towards decreased coverage at the higher pressure, it is not statistically significant.
Table 1.
Individual crossover animal/injection data used to produce the box plots in Figure 9.
| IOP 10 ± 2 mmHg | IOP 35 ± 2 mmHg | ||||||
|---|---|---|---|---|---|---|---|
| Animal | Gender | Day | Eye | ICG Area* | Day | Eye | ICG Area* |
| 1 | M | 0 | OS | 10.2 | 0 | OD | 2.6 |
| 96 | OD | 6.8 | 96 | OS | 8.1 | ||
| 2 | M | 0 | OS | 11.5 | 0 | OD | 9.8 |
| 96 | OD | 11.1 | 96 | OS | 8.1 | ||
| 3 | M | 0 | OS | 12.7 | 0 | OD | 12.7 |
| 96 | OD | 4.1 | 96 | OS | 3.4 | ||
| 4 | F | 0 | OS | 17.1 | 0 | OD | 12.5 |
| 81 | OD | 14.7 | 81 | OS | 7.8 | ||
| 5 | F | 0 | OS | 4.9 | 0 | OD | 4.7 |
| 81 | OD | 1.1 | 81 | OS | 0.0 | ||
| 6 | F | 0 | OS | 1.2 | 0 | OD | 3.2 |
| 81 | OD | 1.3 | 81 | OS | 3.7 | ||
| Mean | 8.06 | Mean | 6.39 | ||||
| SEM | 1.11 | SEM | 0.81 | ||||
Optic disc areas.
Although there is a trend toward decreased coverage at the higher pressure, it is not statistically significant.
ICG, indocyanine green; IOP, intraocular pressure; OS, left eye; OD, right eye; SEM, standard error of the mean.
Potential volume effect on blood flow due to SCS injections.
To determine if there was a physiologic effect of an aqueous solution injected (using a Clearside SCS Microinjector®) into the SCS, a powerful vasoconstrictor, ET-1, was used. Choroidal blood flow was determined using non-recirculating fluorescent microspheres in 11 pigmented rabbits. Aqueous solution administered alone had no effect on the distribution of the microspheres, but when ET-1 was added, there was marked reduction in choroidal blood flow in a distribution that was like that of SCS-injected fluorescein sodium and ICG (Figure 10).
Figure 10.
Distribution of non-recirculating 15-μm fluorescent polystyrene microspheres in the rabbit choriocapillaris following a single 50-μL superotemporal SCS injection of vehicle (left frame) or ET-1 (right frame) in a pigmented New Zealand red rabbit. The microspheres were injected into the left ventricle of the heart 10 minutes after the last SCS injection. The density of the microspheres corresponds to blood flow, with the highest flow being along the visual streak. SCS injection of vehicle has no effect on the choroidal blood flow, but the ET-1 results in marked reduction of flow. The distribution of the reduction is like the distribution of ICG dye injected in the New Zealand white rabbits (Figure 4 and 5).
Discussion
Custom-made SCS injector.
Our custom-made SCS injector successfully delivers aqueous solutions into the SCS both rabbit and NHP (data not presented) eyes. The pattern of distribution of those solutions is not different qualitatively between injections using our SCS needles and those using the Clearside Biomedical SCS Microinjector®. The length of SCS needles made in the laboratory can be adapted to different animal models and/or injection locations. Although not appropriate for clinical studies, laboratory-made SCS needles are reliable, cost-effective, and convenient for use in preclinical drug development studies.
One limitation of this custom-made injector is that commercially available, small gauge sharp hypodermic needles have an angle of about 12.5°, which, for a 30-gauge needle, results in a lumen length that is greater than the thickness of the sclera (Figure 2). Using a 34-gauge needle gets around this problem but restricts the injector’s use of suspension solutions such as triamcinolone acetonide. The Clearside Biomedical SCS Microinjector®, by comparison, uses a larger diameter needle (approximately 30 gauge) with a taper of about 30° so that the length of the exposed lumen is like that of our needle. Although our system could, in theory, be used with a 30- or 27-gauge needle tapered to 30°, the only such needles available commercially (e.g., from the Hamilton Company, Reno, NV) are not sufficiently sharp to penetrate the sclera.
Other types of custom-made SCS injectors have been reported. Oli and Waikar24 describe the use of a modified 26-gauge needle to inject triamcinolone acetonide suprachoroidally in three human patients with pseudophakic cystoid macular edema. To create a stop, a 22-gauge angiocatheter was trimmed and passed over the 26-gauge needle such that 1 mm of the 26-gauge needle was exposed. The injection was made 4 mm posterior to the corneal limbus and the SCS needle directed perpendicular to the scleral surface. We have tried to use a similar system in our NHPs but found that the flexibility of the angiocatheter introduces a variable limiting usage (unpublished data). Too much manual pressure on the syringe with such a device can cause retinal perforation. Also, using a 22-gauge angiocatheter over a 26-gauge needle creates the potential for leakage of injectable material outside the sclera. Another, somewhat more elaborate method of making a custom-made needle has been reported in a series of articles by Marashi et al. 25–27 They removed the rubber stopper from a 1-mL syringe and mounted it over a trimmed luer slip syringe such that when a short 30-gauge needle was inserted through the stopper, 1000 μm of needle was exposed. As with our SCS injector, the Marashi custom-made needle is meant to be injected perpendicular to the sclera. Although we have not tried such a system for our rabbits or NHPs, it seems that getting the exact needle-exposure length would be more of a challenge than with our use of valved cannulae and epoxy resin. Furthermore, the relatively large size of the stopper obscures the injection site, which may make it difficult to know if there is leakage of injectable material external to the sclera.
Two companies have SCS injectors currently in development that they plan to make available without co-development agreements and that will be approved for human use. One is Oculogenex, Inc., which is using a design somewhat like our custom-made SCS injector (La Habra, CA, Hema Ramkumar, MD, CEO, personal communication). The other company is MedOne, Inc. (Sarasota, FL, Paco Rodriquez, Engineering Manager, personal communication and ARVO 2025 exhibitor booth) whose SCS injector has a design like the angiocatheter stop described by Oli and Waikar24 except that the flexible angiocatheter will be replaced by a stiffer sleeve.
Distribution of SCS injections.
New Zealand white rabbits were chosen for this study because their lack of melanin pigment in the retinal pigment epithelium (RPE) and choroid permits visualization of fluorescent solutions, such as ICG and fluorescein sodium in vivo. The distribution of these dyes when injected with our custom-made injector was like that which has been reported for dyes or small particles using the Clearside system or its predecessors. 9,18,19 When injected superotemporally, 50 μL of the aqueous dyes fill approximately half of the eye’s suprachoroidal space in less than one minute with little progression thereafter (Figure 4, see supplemental materials for video). Increased coverage was obtained with a single injection of 100 μL but the dyes still did not reach about a third of the eye (Figure 5). Two injections of 50 μL each—one superotemporal and one inferonasal produced the best coverage although there were some areas of overlap and of non-overlap near the vortex veins (Figure 6). Ex vivo examination of the injected eyes showed that the dyes reached the areas of the far periphery that could not be imaged easily with the scanning laser ophthalmoscope in vivo (Figure 7).
Effect of IOP on the distribution of SCS injections in vivo.
To our knowledge, this is the first attempt to study the effect of IOP on the distribution of solutions injected into the SCS in the living eye. Patel et al.8 tested the effect of IOP during injection ex vivo in human eyes as part of their efforts to optimize the delivery parameters of what eventually became the Clearside SCS injector. 8 Holding the needles with a micromanipulator, they looked at the success rate of SCS injection using various needle lengths and concluded that the chance of entering the SCS was greater when the IOP of the eye was held at 36 mmHg than at 18 mmHg, but they did not consider the distribution of the injected material. The reason for this is not clear. It may be that the early needles were not sufficiently sharp to penetrate the last few layers of sclera or that their micromanipulator system did not simulate the standard clinical practice of handheld injections. However, they did not look at the effect of IOP on distribution of SCS injected solutions. Although one might conclude a priori that if the injected material does not leak out of the eye or goes into the vitreous or subretinal space, it will all end up in the SCS regardless of IOP. Nevertheless, we wanted to test this because of concerns raised by sponsors (personal communications) and because it could be argued that high IOP might cause the injected solution to become more concentrated in a smaller area.
Since we could not image the entire SCS in vivo, we chose to quantitatively measure the area of coverage in the inferonasal SCS from superotemporal injection of 50 μL of the ICG dye to ensure the entire coverage in that quadrant could be imaged. The distal border of the fluorescent area was determined automatically by an image intensity selection tool to avoid observer bias (Figure 8). Although there was considerable variation between eyes and injections, there was no statistically significant difference when the manometric IOP was held at either 10 or 35 mmHg (Figure 9). This makes sense considering that the volume of injected material is the same regardless of initial IOP. Therefore, the primary consideration when injecting human eyes is whether adding to an already high IOP (such as for a glaucoma patient) could damage the optic nerve but not necessarily the distribution of the SCS injected material.
Effect of SCS injections on choroidal blood flow.
We wanted to determine 1) if SCS injections per se interfered with choroidal blood flow and 2) if it were possible to affect choroidal blood flow pharmacologically thus showing at least one functional consequence of SCS drug delivery was possible. For this we chose to inject ET-1, a powerful vasoconstrictor diluted with sterile water. Sterile water alone had no effect on the choroidal blood flow as measured by non-recirculating fluorescent microspheres, but the ET-1 solution markedly decreased the blood flow in a pattern like that of the injected dyes (Figure 8). Therefore, the volume of SCS injected material itself did not have a deleterious effect on the flow. While this experiment shows that drugs injected into the SCS can affect the choroidal vessels, it remains to be seen under what circumstances pharmaceutical agents will reach the RPE and/or pass through the RPE’s zonula occludens (the blood-retinal barrier). The demonstrated efficacy of triamcinolone acetonide injected in the SCS to treat macular edema associated with uveitis (possibly due to inflammatory breakdown of the blood-retinal barrier) is encouraging. 10,11
Conclusions.
The administration of therapeutic agents to the SCS represents an exciting new route of delivery to the eye. Until now, pre-clinical researchers have not had ready access to an SCS injector. Construction of a reliable, custom-made injector system is possible with limited resources. It is comparable qualitatively in efficacy to that reported for a commercially made system in both rabbits and NHPs. IOP at the pressures measured did not appear to affect coverage of the suprachoroidally injected ICG dye solution in rabbit eyes.
Supplementary Material
Funding
This project was supported by the National Institutes of Health (NIH) P30 EY016665 and S10 OD026957, the Wisconsin National Primate Research Center P51RR000167/P51OD011106, and Research to Prevent Blindness
Footnotes
Author conflict of interest statement
No Interests to disclose for any of the authors.
References
- 1.Maguire AM, Russell S, Wellman JA, et al. Efficacy, Safety, and Durability of Voretigene Neparvovec-rzyl in RPE65 Mutation-Associated Inherited Retinal Dystrophy: Results of Phase 1 and 3 Trials. Ophthalmology 2019;126(9):1273–1285, doi: 10.1016/j.ophtha.2019.06.017 [DOI] [PubMed] [Google Scholar]
- 2.Ding K, Shen J, Hafiz Z, et al. AAV8-vectored suprachoroidal gene transfer produces widespread ocular transgene expression. J Clin Invest 2019;129(11):4901–4911, doi: 10.1172/JCI129085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shen J, Kim J, Tzeng SY, et al. Suprachoroidal gene transfer with nonviral nanoparticles. Sci Adv 2020;6(27), doi: 10.1126/sciadv.aba1606 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hackett SF, Fu J, Kim YC, et al. Sustained delivery of acriflavine from the suprachoroidal space provides long term suppression of choroidal neovascularization. Biomaterials 2020;243(119935, doi: 10.1016/j.biomaterials.2020.119935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen M, Li X, Liu J, et al. Safety and pharmacodynamics of suprachoroidal injection of triamcinolone acetonide as a controlled ocular drug release model. J Control Release 2015;203(109–17, doi: 10.1016/j.jconrel.2015.02.021 [DOI] [PubMed] [Google Scholar]
- 6.Olsen TW, Feng X, Wabner K, et al. Cannulation of the suprachoroidal space: a novel drug delivery methodology to the posterior segment. Am J Ophthalmol 2006;142(5):777–87, doi: 10.1016/j.ajo.2006.05.045 [DOI] [PubMed] [Google Scholar]
- 7.Olsen TW, Feng X, Wabner K, et al. Pharmacokinetics of pars plana intravitreal injections versus microcannula suprachoroidal injections of bevacizumab in a porcine model. Invest Ophthalmol Vis Sci 2011;52(7):4749–56, doi: 10.1167/iovs.10-6291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Patel SR, Lin AS, Edelhauser HF, et al. Suprachoroidal drug delivery to the back of the eye using hollow microneedles. Pharm Res 2011;28(1):166–76, doi: 10.1007/s11095-010-0271-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Patel SR, Berezovsky DE, McCarey BE, et al. Targeted administration into the suprachoroidal space using a microneedle for drug delivery to the posterior segment of the eye. Invest Ophthalmol Vis Sci 2012;53(8):4433–41, doi: 10.1167/iovs.12-9872 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Merrill PT, Henry CR, Nguyen QD, et al. Suprachoroidal CLS-TA with and without Systemic Corticosteroid and/or Steroid-Sparing Therapy: A Post-Hoc Analysis of the Phase 3 PEACHTREE Clinical Trial. Ocul Immunol Inflamm 2023;31(8):1579–1586, doi: 10.1080/09273948.2021.1954199 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Khurana RN, Merrill P, Yeh S, et al. Extension study of the safety and efficacy of CLS-TA for treatment of macular oedema associated with non-infectious uveitis (MAGNOLIA). Br J Ophthalmol 2022;106(8):1139–1144, doi: 10.1136/bjophthalmol-2020-317560 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Luo S, Jiang H, Li Q, et al. An adeno-associated virus variant enabling efficient ocular-directed gene delivery across species. Nat Commun 2024;15(1):3780, doi: 10.1038/s41467-024-48221-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ciulla TA, Cunningham ET Jr. Suprachoroidal drug delivery: a versatile therapeutic platform. Expert Opin Drug Deliv 2024;21(12):1705–1713, doi: 10.1080/17425247.2024.2435461 [DOI] [PubMed] [Google Scholar]
- 14.Rotenstreich Y, Sher I, Lawrence M, et al. A Novel Device for Suprachoroidal Drug Delivery to Retina: Evaluation in Nonhuman Primates. Transl Vis Sci Technol 2023;12(6):3, doi: 10.1167/tvst.12.6.3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.de Smet MD, Goncerut M, Asmus F, et al. Refractory post-surgical cystoid macular edema managed following suprachoroidal microcatheterization and delivery of triamcinolone. BMC Ophthalmol 2023;23(1):367, doi: 10.1186/s12886-023-03110-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hedgeland M, Camacho Gonzalez S, Stampoulis D, et al. Evaluation of Suprachoroidal Injection in Cadaveric and Preclinical Models. Transl Vis Sci Technol 2025;14(3):20, doi: 10.1167/tvst.14.3.20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wu KY, Gao A, Giunta M, et al. What’s New in Ocular Drug Delivery: Advances in Suprachoroidal Injection since 2023. Pharmaceuticals (Basel) 2024;17(8), doi: 10.3390/ph17081007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chiang B, Kim YC, Edelhauser HF, et al. Circumferential flow of particles in the suprachoroidal space is impeded by the posterior ciliary arteries. Exp Eye Res 2016;145(424–431, doi: 10.1016/j.exer.2016.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chiang B, Venugopal N, Edelhauser HF, et al. Distribution of particles, small molecules and polymeric formulation excipients in the suprachoroidal space after microneedle injection. Exp Eye Res 2016;153(101–109, doi: 10.1016/j.exer.2016.10.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chiang B, Venugopal N, Grossniklaus HE, et al. Thickness and Closure Kinetics of the Suprachoroidal Space Following Microneedle Injection of Liquid Formulations. Invest Ophthalmol Vis Sci 2017;58(1):555–564, doi: 10.1167/iovs.16-20377 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chiang B, Wang K, Ethier CR, et al. Clearance Kinetics and Clearance Routes of Molecules From the Suprachoroidal Space After Microneedle Injection. Invest Ophthalmol Vis Sci 2017;58(1):545–554, doi: 10.1167/iovs.16-20679 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nork TM, Kim CBY, Shanmuganayagam D, et al. Measurement of regional choroidal blood flow in rabbits and monkeys using fluorescent microspheres. Arch Ophthalmol 2006;124(6):860–8, doi: 10.1001/archopht.124.6.860 [DOI] [PubMed] [Google Scholar]
- 23.Nork TM, Kim CBY, Munsey KM, et al. Regional choroidal blood flow and multifocal electroretinography in experimental glaucoma in rhesus macaques. Invest Ophthalmol Vis Sci 2014;55(12):7786–98, doi: 10.1167/iovs.14-14527 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Oli A, Waikar S. Modified inexpensive needle for suprachoroidal triamcinolone acetonide injections in pseudophakic cystoid macular edema. Indian J Ophthalmol 2021;69(3):765–767, doi: 10.4103/ijo.IJO_1464_20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Marashi A, Baba M, Zazo A. Managing solar retinopathy with suprachoroidal triamcinolone acetonide injection in a young girl: a case report. J Med Case Rep 2021;15(1):577, doi: 10.1186/s13256-021-03162-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Marashi A, Zazo A. Suprachoroidal injection of triamcinolone acetonide using a custom-made needle to treat diabetic macular edema post pars plana vitrectomy: a case series. J Int Med Res 2022;50(4):3000605221089807, doi: 10.1177/03000605221089807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Marashi A, Baba M, Abu Ghedda S, et al. A combination of suprachoroidal injection of triamcinolone using a custom-made needle and intravitreal Ziv-aflibercept every eight weeks to manage naive/denovo central DME: a single-center retrospective case series. Int J Retina Vitreous 2024;10(1):30, doi: 10.1186/s40942-024-00550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.