Retrospective Assessment of Carbonic Anhydrase Inhibitors in Topical or Episcleral Implant Form for the Treatment of Equine Glaucoma

ABSTRACT

Objective

To assess whether topical carbonic anhydrase inhibitors (CAIs) with or without episcleral brinzolamide implants control intraocular pressure (IOP) in horses with glaucoma.

Animals Studied

12 client‐owned horses diagnosed with glaucoma.

Procedures

Medical records of horses treated with topical CAIs for ≥ 14 days, with a subset receiving episcleral brinzolamide implants, were reviewed. Diagnosis, signalment, concurrent ocular disease, medications, IOP, visual status, and additional procedures were recorded. Treatment failure was defined as horses with IOP > 30 mmHg resulting in phthisis bulbi or requiring a procedure or enucleation.

Results

Secondary glaucoma was diagnosed in 8/12 (67%) eyes and primary glaucoma in 4/12 (33%) eyes. IOP was < 30 mmHg in 7/12 (58%) eyes after treatment with topical CAIs for a median of 28 days. Episcleral brinzolamide implants were placed in 7/12 eyes after a median of 58 days of medical therapy. Treatment failure occurred in 3/7 horses (43%) receiving brinzolamide implants at a median of 189 days and in 4/5 medical therapy‐only horses (80%) at a median of 51 days. While 6/12 eyes (50%) were visual at last follow‐up (median of 161 days), enucleation was the final outcome in 10/12 (83%) eyes.

Conclusions

Topical CAIs appear to result in temporary IOP control in a subset of horses, though some horses may be refractory. Episcleral brinzolamide implants may extend IOP control in horses with glaucoma, but prognosis for vision remains guarded and further optimization is needed.

1. Introduction

Glaucoma is a progressive neurodegenerative disease characterized by elevated intraocular pressures (IOP), which damage the neural retina and optic nerve, and result in blindness if left untreated. In horses, glaucoma is commonly secondary to other ocular diseases such as uveitis (including equine recurrent uveitis, ERU), trauma, or intraocular masses [1, 2, 3, 4, 5]. Primary glaucomas, arising due to inherent abnormalities of aqueous humor drainage, have been proposed to occur in horses but are poorly understood [1, 2]. Overall, the prevalence of equine glaucoma has been estimated at 0.07% [5]. Treatment is focused on addressing the underlying cause and managing IOP to slow the progression of the disease. Currently available medical and surgical treatments for glaucoma are usually directed at either reducing the production of aqueous humor (carbonic anhydrase inhibitors, beta blockers, transscleral cyclophotocoagulation, and intravitreal gentamicin), or improving aqueous humor outflow dynamics (prostaglandin analogs and gonioshunt procedures) [3].

Carbonic anhydrase inhibitors (CAIs) are the most commonly used medication for glaucoma, including in horses [3], and function by reducing the volume of aqueous humor produced by the ciliary body [6]. Topical dorzolamide and brinzolamide have both been shown to lower IOP in normal horses [7, 8, 9], but data on their efficacy in glaucomatous horses are lacking. Given the impact of medication noncompliance on treatment outcomes, implantable devices have been gaining increasing attention for the treatment of glaucoma in both people and animals [10]. This is of particular importance in horses, which can be difficult to treat with topical ocular medications in the absence of a subpalpebral lavage system. Episcleral silicone matrix brinzolamide implants, which have demonstrated both safety and IOP‐lowering ability in normal New Zealand white rabbits [11], are available for clinical use in horses. However, their safety and efficacy in horses have not been tested.

This retrospective case series aimed to assess the ability of topical CAIs with or without episcleral brinzolamide implants on IOP control in glaucomatous horses. Based on clinical experience, we hypothesized that both topical CAIs and implants would result in lower IOPs in horses, and that the implants would be well tolerated.

2. Materials and Methods

Animal owners or owners' representatives provided written consent for the treatment provided and the use of medical records and images for research purposes. An electronic search of the North Carolina State University College of Veterinary Medicine (NCSU CVM) medical records was performed between the dates of January 2013 and December 2023 with the keywords “equine” and “glaucoma,” or episcleral brinzolamide implant product code. Additional records were provided by Hawaii Veterinary Vision Inc. Cases were included if information regarding the diagnosis of glaucoma, based on elevated IOP (> 25 mmHg) and concurrent supporting clinical signs (corneal edema, episcleral injection, Haab's stria, etc.), was present, if topical CAIs were used for at least 14 days following diagnosis, and if the horse returned for at least two follow‐up appointments after beginning CAIs. Where the cause of glaucoma was noted in the record (primary glaucoma or secondary to ERU, other uveitis, or trauma), this was recorded. Cases were excluded if they did not receive any medical management before brinzolamide implant placement, enucleation, or glaucoma surgery, if a glaucoma surgery (gonioshunt placement, transscleral cyclophotocoagulation) or intravitreal gentamicin injection was performed before placement of brinzolamide implants, or if the horse was lost to follow‐up after implant placement.

All patients underwent complete ophthalmic examinations performed by a board‐certified ophthalmologist, including neuro‐ophthalmic examination, slit lamp biomicroscopy (SL‐17, Kowa), indirect ophthalmoscopy, applanation tonometry (TonoPen, AviaVet, Reichert Technologies), and fluorescein staining. Horses were sedated for exams with detomidine and/or xylazine given intravenously at varying doses depending on the individual case. Information extracted from the records included signalment, diagnosis (primary or secondary glaucoma), concurrent ocular disease, topical medications used, number of brinzolamide implants per eye, complications associated with brinzolamide implants, IOP, and visual status at diagnosis and subsequent visits. Secondary glaucoma was defined as glaucoma occurring due to another ocular disease (uveitis, heterochromic iridocyclitis and keratitis, cataracts, or trauma), while primary glaucoma was defined as glaucoma in the absence of a predisposing factor. Treatment failure was defined as an IOP > 30 mmHg resulting in phthisis bulbi or requiring glaucoma surgery, intravitreal gentamicin injection, or enucleation while on topical CAIs and/or with a brinzolamide implant in place. Visual status was approximated with menace responses.

The brinzolamide implants were composed of silicone (medical grade, Nusil Technology, Carpinteria, CA) and 12 mg brinzolamide (pharmaceutical grade, AcaChem Scientific, San Antonio, TX) to form a 20 mm × 2 mm implant, consisting of 30% brinzolamide by weight as a percentage of the weight of the implant (wt/wt). The implants were formulated at NCSU CVM as described by Smith et al. [11].

Episcleral implants were placed as has been described previously [12]. Briefly, horses were sedated and topical anesthetic (proparacaine HCL ophthalmic solution, USP 0.5%) was applied after the ocular surface was aseptically prepared. A 3 mm incision into the superior temporal bulbar conjunctiva and Tenon's capsule was made 5 mm posterior from the limbus, and a pocket in the episcleral space was created by blunt dissection. One to two implants were placed within the pocket, oriented parallel to the limbus, and the conjunctiva was closed with a single interrupted or cruciate suture.

All statistical analysis was performed in RStudio [13]. Descriptive statistics are presented as median and range.

3. Results

A total of 12 horses (12 eyes) fit the inclusion criteria, with seven receiving brinzolamide implants (implant group) and five being treated by medical management only (medical‐only group). Ages ranged from 0.67 to 22 years with a median of 12.5 years and no difference between groups (p = 0.68). Demographic and case summary information is presented in Table 1. Secondary glaucoma (N = 8) was more common than primary glaucoma (N = 4) and was most frequently associated with ERU (N = 5) in both the implant and medical‐only groups. Initial IOP ranged from 27 to 54 mmHg (median 38 mmHg) with no difference between groups (p = 0.14). One eye in each group was blind at diagnosis (cases 4 and 9), and the remainder were visual.

TABLE 1.

Summary of cases of horses with glaucoma receiving topical carbonic anhydrase inhibitors (CAIs) and episcleral brinzolamide implants (cases 1–7) or topical CAIs only (cases 8–12).

Group	Case	Signalment	Diagnosis	Eye	Initial IOP (mmHg)	Days on topical CAIs prior to implant	IOP after topical CAIs (mmHg)	Days to last IOP	Last IOP (mmHg)	Outcome
Brinzolamide implant	1	20 years Arabian mare	Primary glaucoma	OD	30	25	20	54	27	Blind, comfortable; implant + topical CAIs
	2	17 years Appaloosa gelding	Primary glaucoma	OS	27	9	NA	486	15 a	Transscleral cyclophotocoagulation b
	3	8 years Quarter Horse gelding	Primary glaucoma	OD	38	51	46	147	66	Intravitreal gentamicin injection
	4	21 years Paint gelding	Glaucoma secondary to ERU	OS	NA c	287	32	2209	28	Blind, comfortable; implant only d
	5	9 years Quarter Horse mare	Glaucoma secondary to HIK	OD	40	58	15	100	9	Visual, comfortable; implant + topical CAIs e
	6	13 years Oldenburg Warmblood mare	Glaucoma secondary to HIK and ERU	OD	33	122	NA f	189	51	Enucleation due to uncontrolled glaucoma and HIK
	7	12 years Appaloosa gelding	Glaucoma secondary to ERU	OS	51	356	57	668	16	Blind, comfortable; implant only
Topical CAIs only	8	10 years Thoroughbred gelding	Primary glaucoma	OS	54	30	17	311	48	Uncontrolled IOP on CAIs, visual g
	9	22 years Arabian gelding	Glaucoma secondary to ERU	OS	54	20	26	49	20	Blind, comfortable, on topical CAIs; enucleated due to uveitis and difficulty medicating
	10	8 month Quarter Horse filly	Glaucoma secondary to phacoemuls‐ ification	OS	48	7	15	18	43	Enucleated due to uncontrolled glaucoma on CAIs
	11	16 years Warmblood gelding	Glaucoma secondary to ERU	OS	37	11	19	51	68 h	Intravitreal gentamicin injection i
	12	10 years Quarter Horse mare	Glaucoma secondary to ERU	OD	35	12	12	175	68	Enucleated due to uncontrolled glaucoma on CAIs

^a Horse was having reported pressure spikes with values not recorded.

^b Gonioshunt surgery 6 days following transscleral cyclophotocoagulation, enucleated 1 month following gonioshunt surgery due to uncontrolled glaucoma.

^c IOP described as “high” but not recorded at diagnosis due to severe corneal edema precluding accurate readings.

^d Blind at diagnosis, implants replaced 354 days after initial implants and 429 days after second round of implants.

^e Enucleated 474 days after diagnosis due to severe corneal ulceration.

^f IOP reported as “high” in the record but values are not recorded.

^g Phthisis bulbi 754 days following diagnosis of glaucoma.

^h Not on topical CAIs.

ⁱ Enucleated 72 days after diagnosis of glaucoma due to uncontrolled IOPs.

2% dorzolamide–0.5% timolol (Cosopt) was the only topical CAI used for all eyes except case 12, in which 2% dorzolamide was used. Other topical medications included diclofenac (cases 6–8), neomycin/polymyxin/dexamethasone (cases 6, 8–10, and 12), prednisolone acetate (case 3), ofloxacin (case 10), moxifloxacin (case 12), voriconazole (case 12), atropine (case 12), and EDTA (cases 10 and 12).

Horses in the implant group (N = 7) were treated with topical CAIs for 9–356 days (median 58 days) prior to receiving implants, and only two horses had IOP < 30 mmHg at implantation (cases 1 and 5) (Table 1). Five of 7 total eyes had the number of implants placed recorded: four horses (cases 1 and 5–7) received two implants per eye, and one horse (case 3) received three. One horse (case 4) had implants replaced twice with 354–429 days between procedures, and another (case 2) had implants replaced at 486 days following the first placement prior to receiving transscleral cyclophotocoagulation 6 days later due to uncontrolled IOPs. Two horses (cases 6–7) received suprachoroidal cyclosporine implants concurrently with brinzolamide implants, and these were the only two horses to have transient conjunctival hyperemia and chemosis reported. Only two horses had an IOP reported in the record between implant placement and last follow‐up: case 4 had a 6.25% reduction in IOP (32 mmHg at implant placement vs. 30 mmHg at 19 days post‐implant), and case 7 had a 63.16% reduction in IOP (57 mmHg at implant placement vs. 21 mmHg at 42 days post‐implant). At a median last follow‐up of 189 days (range 54–2209 days), three of seven implant horses (cases 2, 3, and 6) required additional procedures or enucleation due to uncontrolled IOPs, with 2/3 (cases 3 and 6) reported as receiving topical CAIs at the time of enucleation. The remaining four horses (cases 1, 4, 5, and 7) had controlled IOPs—two were still receiving topical CAIs in addition to the brinzolamide implant. Three of the four eyes with controlled IOPs (cases 1, 5, and 7) were visual at implantation, and two became blind by last follow‐up (cases 1 and 7) (Table 1).

In the medical‐only group, all horses (N = 5) had an IOP < 30 mmHg at a median of 12 days on topical CAIs (range 7–30 days). At a median last follow‐up of 51 days (range 18–311 days), three horses (cases 10–12) required additional procedures or enucleation due to uncontrolled IOPs, and one horse (case 8) with uncontrolled IOP eventually progressed to phthisis bulbi (Table 1). One of the enucleated eyes (case 12) had developed fungal keratitis, which was medically managed, though blindness and an IOP of 68 mmHg were cited in the record as the reason for enucleation. One of the eyes that underwent an intravitreal injection (case 11) developed high IOP after discontinuing topical CAIs and was restarted on them, but ultimately IOP did not respond. The remaining horse (case 9) had controlled IOP at 49 days post‐diagnosis, but the eye was enucleated due to underlying uveitis and the horse's behavioral resistance to topical medications.

4. Discussion

The results of this study support the use of CAIs for initial medical management of equine glaucoma, though not all horses appear to be responsive, and many eventually become refractory. Seven horses (58%) had IOPs reduced to < 30 mmHg after treatment with dorzolamide ± timolol for a median of 28 days, while four horses (25%) continued to have IOPs > 30 mmHg. In the seven horses receiving brinzolamide implants, four horses were reported to be unresponsive to topical CAIs before implant placement, with two of those four (cases 3 and 6) requiring surgery because of uncontrolled IOPs following implant placement. Cases 4 and 7 had uncontrolled IOPs before implant placement but responded favorably to the brinzolamide implants, having controlled IOPs at last follow‐up while no longer being on topical CAIs. The topical CAI‐responsive horses that received implants (cases 1 and 5) maintained controlled IOPs at last follow‐up, and both were still on topical CAIs at that time. In the medical management only group, all five horses initially responded to topical CAIs, but four of the five developed uncontrolled IOPs by a median of 51 days. Altogether, horses responsive to initial treatment with topical CAIs are likely good candidates for medical management with topical CAIs, with the understanding that many of them will eventually become refractory. Horses not responsive to topical CAIs may require a glaucoma procedure or enucleation, or they may benefit from brinzolamide implants, as was seen in two horses in this study. The exact reason why some horses that were not initially responsive to topical CAIs were able to have controlled IOPs with the implants and others did not experience this benefit is unknown, but it could be related to differences in the chronicity of glaucoma, the underlying cause of glaucoma, or how effectively topical medications were being administered by owners.

There is minimal information in the literature on the use of brinzolamide for equine glaucoma. A single study demonstrated that topical 1% brinzolamide given twice daily lowers IOP in normotensive equine eyes by 5 mmHg [8], which is a greater effect than that reported for dorzolamide‐timolol (3.5 mmHg [9]) or dorzolamide alone (2 mmHg [7]). The effectiveness of these drugs on glaucomatous equine eyes is not reported in the literature, and most of what is known about the action of CAIs on aqueous humor production comes from species in which glaucoma is more thoroughly studied. Both dorzolamide and brinzolamide are thought to reduce aqueous humor production by competitively inhibiting carbonic acid from interacting with carbonic anhydrase isoenzyme II (CAII) in the ciliary body epithelium, resulting in decreased production of bicarbonate [6]. Action of CAII must be inhibited by at least 98% to result in an IOP‐lowering effect [14]. It is unknown what the most effective doses of either dorzolamide or brinzolamide are for the equine eye. In theory, a CAI implant will produce a steady state of release which could combat peak‐and‐trough fluctuations in IOP that likely occur with topical treatment alone and contribute to glaucoma progression, as seen in dogs and humans [15, 16, 17]. In New Zealand white rabbits, a single 12 mg brinzolamide implant resulted in an IOP reduction of 29.9% over 28 days and a vitreal brinzolamide concentration of 10 ng/mL [11], greatly exceeding the reported amount of 1.2 ng/mL brinzolamide needed to inhibit CAII in this species [6]. In vitro analysis suggested implant releases brinzolamide at a rate of 2.5 μg/day [11]. Most horses in this study received two implants per eye, consistent with a dose of 24 mg. Considering the larger vitreous volume of horses (26.14 mL [18]) relative to rabbits (1.5 mL [19]), a more concentrated implant or a higher number of implants may be needed to elicit the same IOP‐lowering effect. As two of the horses in the implant group which were visual on implant placement did become blind despite having controlled IOPs at last follow‐up, it is plausible that these horses continued to experience IOP spikes that contributed to progression of disease and that a higher dose of brinzolamide may be needed to prevent large diurnal fluctuations. Importantly, there were no significant adverse effects reported with placement of the implants. Two horses developed transient hyperemia and chemosis, but these were also the only two horses to receive suprachoroidal cyclosporine implants, making it difficult to determine which implants caused the effect. Optimization of brinzolamide implants to deliver a higher dose or use of additional implants may improve efficacy of these implants and require further study.

Prognosis for both vision and globe retention is poor in equine glaucoma. While six (50%) of the eyes in this study had at least some vision at last follow‐up, all but two of these eyes had enucleation as the final outcome. The majority of equine glaucoma cases are secondary to ERU [1, 2, 3, 20], as was seen in 41% of cases in this study. Inability to control the underlying cause of glaucoma will contribute to a poor outcome. A study on outcomes of ERU in horses in the southern United States found that 11.2% of horses presenting for ERU had concurrent glaucoma, and these horses were more likely to go blind or have enucleations performed compared to horses without elevated IOP [21]. Four horses in this study were diagnosed with primary glaucoma, due to absence of signs of uveitis or other causes. Primary glaucomas of the horse are rarely reported in the literature [1], and it is difficult to verify that the four horses described here truly had primary glaucoma without histopathological confirmation. In a previous histologic study of globes enucleated due to glaucoma, while only 48% of cases had clinical findings consistent with uveitis, 91% had histopathologic evidence of uveitis [2]. Only one of the four primary glaucoma horses had controlled IOPs with both the brinzolamide implant and topical CAIs at last follow‐up (case 1).

There are limitations in this study worth noting, including its retrospective nature (incomplete records, inconsistent data reporting, and lack of long‐term follow‐up) and small sample size, which lessen the statistical power of the study. Horses were afflicted by a variety of underlying causes of glaucoma, presenting confounding variables in the outcome. All but one of these cases (case 7) was seen by an academic referral institution and may represent the most extreme glaucoma cases. Additionally, brinzolamide implants may have been more likely to be placed in eyes refractory to topical CAIs, introducing a selection bias. It was not possible in this retrospective analysis to compare IOPs in horses receiving CAIs to IOPs in glaucomatous horses not receiving CAIs. Finally, IOP measurements between implant placement and last follow‐up in the brinzolamide group were sparse, making it difficult to determine the IOP‐lowering effect of the implants. Prospective data with long‐term follow‐up on topical and/or implantable CAIs in glaucomatous horses are needed to confirm our results.

Overall, there is still much to be learned about the pathophysiology of equine glaucoma and the pharmacological properties of CAIs in the equine eye, but this study provides support for the idea that topical CAIs may provide a period of controlled IOP in a subset of horses (58% of horses in this study). Episcleral brinzolamide implants may be most beneficial early in the disease process in horses, even in some that do not respond initially to dorzolamide, but optimization of the implant is needed.

Ethics Statement

The study was done in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. As this was a retrospective study, animal use approval was not required.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data are available upon reasonable request.