This article was originally published in a sponsored newsletter.

As we begin our sixth year of Retina Minute, I want to thank our loyal readers, as well as Sarah Fackler and the BroadcastMed team who work behind the scenes to create this monthly newsletter. For this edition, I am going to tell a story about the lecture I gave at Retina Subspecialty Day in San Francisco this past November.

Last summer, I listened to a podcast called Retina Synthesis hosted by Dr. Carmen Puliafito. In one episode, Dr. Puliafito interviewed my friend Dr. Roger Goldberg on central serous chorioretinopathy (CSCR), during which Dr. Goldberg mentioned there is currently no treatment for acute CSCR. However, I gave a lecture on the use of NSAIDs for the treatment of acute CSCR 10 years ago at Retina Subspecialty Day in New Orleans. After I was done yelling at my car radio, I realized that the message from that talk was not adopted by most of the retina community in the United States. My epiphany motivated me to redo my previous study to see if anything had changed in the last decade.

The initial theory was that NSAIDs, which are used to treat central macular edema (CME) and diabetic macular edema (DME), might work in some CSCR cases. In the original study, we compared q.i.d. dosing of bromfenac 0.09% and nepafenac for the treatment of acute CSCR in our practice to three other practices that observed patients and acted as controls. The three participating control practices were Cleveland Clinic's Cole Eye Institute (Dr. Rishi Singh), Charlotte Eye Ear Nose & Throat Associates (Dr. Andrew Antozsyk), and Illinois Retina Center (Dr. David Dodwell).

The q.i.d. dosing interval was aligned with McNamara et al’s findings that nonsteroidal drops penetrate the cornea, anterior chamber, choroid, and retina, but not the vitreous.¹ The medication must penetrate the sclera, as opposed to the cornea, to get into the retina, which requires a higher frequency of administration.

We defined acute CSCR as the resolution of edema on OCT and improvement of vision to 20/20 within a year. The time frame for the initial study was 2007 to 2013. Overall, 111 patients met the criteria. The control practices had 73 patients who met these criteria, while our practice—Medical Center Ophthalmology Associates (MCOA)—had 38 patients who were treated with COX-2-specific NSAIDs, bromfenac 0.09% or nepefenac, as explained above.

The mean number of days to resolution with observation in the control group was 131 days (range: 12 to 362 days). In the treated group, the mean number of days to resolution was 42 days with a range of 14 to 95 days. This difference is highly statistically significant (p<0.0002).

In the updated study, we compared our use of NSAIDs dosed q.i.d. to Georgia Retina (Dr. Ella H. Leung and Dr. David S. Chin Yee) who manage their acute patients by observation. We looked at patients from 2018 to 2023 and reviewed more than 1,600 charts. In the observation arm, 119 patients met the criteria for follow-up, but only 63 patients resolved with only observation. The mean number of days to resolution was 133 days, with a range of 40 to 329 days. In the MCOA group, 26 patients met the inclusion criteria and were treated with NSAID medication q.i.d.; however, bromfenac and nepafenac was replaced with ketorolac due to costs. These patients resolved in a mean of 89 days with a range of 27 to 264 days.

When comparing the data of the two groups in the more recent study, we saw that treatment with ketorolac q.i.d. decreased time to resolution by 44 days, which was significant (p<0.0295), but less so than the previous study. To understand why there was such a difference between the two studies, I traveled to Paris to meet with Dr. Francine Behar-Cohen. She suggested the following potential mechanisms:

1. NSAIDs, most notably COX-2 inhibitors, block prostaglandin receptor E2.
2. Aquaporin (AQP) water channels are involved with the development of macular edema and these channels are upregulated by prostaglandin E2.
3. Prostaglandin E receptor 2 (PTGER2) was increased in induced pluripotent stem cells treated with aldosterone or cortisol.
4. NSAIDs directly block calcium-activated potassium ion channels which have been implicated in increased choroidal permeability and may lead to CSCR.

While bromfenac and nepafenac are COX-2 inhibitors, ketorolac is a nonselective inhibitor and blocks both COX-1 and COX-2 inhibitors. As a result, the effect is not as strong in acute CSCR, leading to a longer time to resolution.

Nepefenac and bromfenac can be up to $600 per bottle. It is hard for most people to afford that price tag, especially when it is used q.i.d. as opposed to q.d. (i.e., the current dosing schedule for postoperative cataract surgery). However, one of my patients told me that generic bromfenac in stronger concentrations (i.e., 0.9%) is less than $100, so it is more affordable and can encourage compliance.

I hope this information will help you the next time you see an acute CSCR patient.

Reference(s):

1. McNamara T, Baklayan GA, Deshmukh HM, Patterson HM, Gow JA. Concentrations of radioactivity in ocular tissues after a single 0.09% topical dose of 14C-bromfenac ophthalmic solution. Invest Ophthalmol Vis Sci. 2006 May;47(13):Poster 5086. From ARVO 2006.

Michael A. Singer, MD is a board-certified Vitreoretinal Specialist and Director of Clinical Research at Medical Center Ophthalmology Associates in San Antonio, TX. He is currently the Clinical Professor of Ophthalmology at the University of Texas Health Science Center in San Antonio.