We all have burning questions that come up during regular clinical practice that we’d pose to an expert if we had the chance—whether it be a challenging diagnosis, a tough management decision, or knowing when to refer. Attendees of Eyes on Retina 2026 got the chance to pose any and all questions they had regarding retinal practice to Carolyn Majcher, OD, FAAO, FORS, who took up residency In the Hot Seat for a live Q&A. Here are some of her thoughts.
How effective is low-level light therapy for macular degeneration?
Photobiomodulation (PBM) is an FDA-approved treatment option for patients with intermediate-stage macular degeneration. In the LIGHTSITE III study, which looked at the effect of PBM on geographic atrophy (GA) onset, the researchers found PBM treatment resulted in a significantly lower rate of geographic atrophy (GA).1
At month 24, ~24% of patients in the sham group (who still technically received treatment, just a lower dose of light therapy) developed geographic atrophy; in the PBM group, this was reduced to ~7%. The researchers also observed an improvement of approximately one line in visual acuity among individuals treated with PBM.1 Although this was a post-hoc analysis and thus should be taken with caution, this suggests that we should perhaps utilize imaging technologies, like optical coherence tomography (OCT), to stratify intermediate age-related macular degeneration patients into low- and high-risk categories. For those within the high-risk category, who have reticular pseudodrusen, hyperreflective foci, large drusen load, etc., we should consider PBM to reduce the risk of future GA.
How do we differentiate between IRMA and neovascularization?
In patients with diabetic retinopathy (DR), we know IRMA, or intraretinal microvascular abnormalities, are dilated telangiectatic capillaries; although they’re abnormal blood vessels, they’re still intraretinal. This contrasts with neovascularization, which is preretinal (on top of the retina) and is the hallmark of proliferative diabetic retinopathy (PDR). The problem is that if we’re assessing potential IRMA, it's tough to determine where it is located based solely on the clinical examination, making multimodal imaging technologies, such as optical coherence tomography angiography (OCTA), key in assisting the identification of early neovascularization and distinguishing it from areas of IRMA.
These technologies allow depth resolution of abnormal vasculature to identify defining features. For example, observing neovascular fronds, typically on the optic disc in the parapapillary region or elsewhere in the retina, with attached fibrotic tissue, is a key sign that what you’re seeing isn’t IRMA and must be neovascularization. Using a 3D view, you’ll also be able to identify neovascularization as being in front of the retina as opposed to within—which can be done even just using OCT.
But if we focus in on OCTA, there's a preset called the vitreoretinal interface, through which we can observe the blood flow just above the retina in the posterior vitreous, which should be inherently avascular. If something does show up here on the vitreoretinal interface flow, it’s likely suggestive of neovascularization and PDR. Conversely, a patient with IRMA would be devoid of signal on the vitreoretinal interface.
Sometimes fluorescein angiography (FA) is suggested as a potential distinguishing modality. It can help with the determination of whether there's leakage from the abnormal blood vessels, which is more indicative of neovascularization compared to IRMA. Like OCT, fluorescein angiography can provide depth resolution. They’re both great technologies, but as optometrists, it's more within our scope, easier, and simpler to do non-invasive OCTA versus fluorescein angiography. But use the tools that you have available to you.
What are the risks of GLP-1 receptor agonists on the progression of diabetic retinopathy?
We know there’s an associated transient early worsening of DR, especially within the first 12 to 18 months, after beginning to take a GLP-1 receptor agonist. This risk factor is really directly related to how quickly and how much blood glucose drops upon starting the medication. As an example, if a patient drops from 12% to 6%, they’re at great risk of progressing or developing diabetic retinopathy during that early period.
The American Academy of Ophthalmology’s care recommendations are that patients who begin GLP-1 receptor agonists be given a baseline dilated comprehensive eye exam and color fundus photography, with OCT being considered in those at greatest risk. They also recommend more frequent follow-up within the first 12–18 months for those with preexisting DR or other risk factors—those with mild non-proliferative DR (NPDR) be monitored every ~6 months; those with moderate NPDR, every ~3–4 months; and severe NPDR, every ~4–6 weeks.
We know from the literature that the cardiovascular and diabetes control benefits outweigh DR risk temporarily worsening; it's important that patients continue these medications for the cardiovascular and diabetes benefits. We just need to watch their eyes closely during this time.
Could you explain the update in the Plaquenil toxicity screening guidelines?
Regarding hydroxychloroquine retinal toxicity, the screening guidelines were recently revised—ten years in the making—from the previous guidelines.2 In this updated version, OCT (spectral domain) and wide-pattern fundus autofluorescence (FAF) are now recommended as primary techniques. You’re no longer obligated to do a 10-2 or 24-2 visual field as primary screening—though visual field testing (in particular the 24-2C) should still be considered for secondary or conformity screening alongside multifocal ERG. The frequency of screening remains the same: a baseline examination, followed by annual screening at five years, though annual screening can be started before this point in patients with significant risk factors.
For OCT, obtain macular cubes with quantitative data alongside raster scans. The standard-of-care update highlights that the ellipsoid and interdigitation zones are the layers most likely to experience loss from plaquenil toxicity, typically in the parafoveal region. The update also emphasizes the importance of the quantitative data, highlighting that progressive thinning can be seen here before it’s visible on individual high-resolution raster scans, exemplifying why you should monitor both.
Finally, when it comes to the wide-field FAF done as part of the primary screening, it should ideally extend out 50–55o beyond the arcades. This is because it’s now thought that some patients, particularly those who are Asian, may develop extramacular toxicity as opposed to the more conventional peripheral pattern of hypertoxicity in the parafoveal region.
