Clinicians and researchers don’t shy away from hot topics in the field

This year's Congress on Controversies in Ophthalmology (COPHy) touched on a broad range of topics, including AI and neuro-ophthalmology. Image credit: ©arhendrix – stock.adobe.com

Ophthalmologists from around the globe convened for the 15th annual Congress on Controversies in Ophthalmology (COPHy) on 15 and 16 March, 2024. At this year’s meeting, the participants went face-to-face for a series of debates in Athens, Greece.

In their welcome letter, COPHy co-chairs Anat Loewenstein, MD, and Baruch Kuppermann, MD, PhD, thanked the experts who were not afraid to address controversial topics in the name of sharing knowledge.

“The Congress aims at reaching up-to-date recommendations to ongoing debates even when data remain limited, through evidence-based medicine, including expert opinion,” they wrote in the letter. Each topic was defended and dissected by a pair of ophthalmologists in back-to-back arguments. Following the spirited debate, a discussion period allowed participants to focus on the finer points and areas for nuance in the arguments presented.

Emphasis on AI

Following welcome remarks, the meeting opened with a keynote lecture on artificial intelligence (AI), arguably the most controversial topic in medical technology today. Ursula Schmidt-Erfurth, MD, professor and chair of the Department of Ophthalmology and Optometry, Medical University of Vienna, Austria, delivered the keynote, titled “How Good Are We Really Without AI Tools?”

Speaking to Ophthalmology Times Europe prior to the meeting, Prof Schmidt-Erfurth explained why ophthalmology is fertile ground for enhancing AI’s capabilities. “Ophthalmology is one of the most promising [AI] targets for many reasons, because doctors are already using digital imaging in large amounts and in huge volumes per image. So millions and millions of pixels, in millions of millions of patients’ retinal images, are the perfect playground for precision AI,” she explained. Prof Schmidt-Erfurth used the frequency of optical coherence tomography (OCT) as an example of the amount of data generated by diagnostic imaging. “We have to identify the right biomarkers that should trigger treatment and we have to identify them in the most precise, quantified way. AI can extract [these data] for us, so that we can use [them] to the benefit of our patients.”

The first debate of the session, immediately following the keynote, took the role of AI in medicine to its farthest reaches. Retina specialists Giuseppe Querques, MD, PhD, and Paolo Lanzetta, MD, both from Italy, shared opposing answers to the question, “Is AI ready to replace physicians?”

Prof Querques, associate professor, University Vita-Salute, IRCCS Ospedale San Raffaele, Milan, argued in the affirmative. “Artificial intelligence is not the future, but it’s the present,” Prof Querques told Ophthalmology Times Europe. He also said that AI is not necessarily replacing physicians, but filling the absence of one. “AI is particularly useful in areas where there’s a lack of the right number of physicians for managing patients,” he said. Whether the challenges are staffing issues, remote or rural clinics, or a lack of training resources, AI can bridge the gap where a human physician cannot.

In his rebuttal, Prof Lanzetta, chairman of the Department of Ophthalmology, University of Udine, and director, European Institute of Ocular Microsurgery, Udine, argued that AI is not able to replace many vital aspects of the physician-patient relationship.

“Machines will not replace empathy,” Prof Lanzetta said, “And we know how important empathy is in our relationships with our patients.” As an example, he cited patients with macular conditions, who require regular injections with anti-VEGF agents. For those patients, Prof Lanzetta said, having a warm, personal relationship with a human practitioner is an important part of the treatment protocol. “Artificial intelligence and related technology will definitely assist physicians in being more accurate, more efficient, by reducing errors and facilitating their daily activities,” Prof Lanzetta added. Envisioning a future where AI plays a larger role in patient care, Prof Lanzetta said, “[AI usage] will be more like collaboration between professionals for better managing and treating our patients.”

Protecting patients with best practices

Beyond the technological topics under discussion, participants also debated the best practices in a series of common situations practitioners and patients might navigate. John J. Chen, MD, PhD, professor of Ophthalmology and Neurology at the Mayo Clinic in Rochester,
Minnesota, US, advised caution as the best route when confronted with the scenario of an elderly patient presenting with acute visual loss. To support the argument for stroke evaluation, Prof Chen presented a case that was then discussed by other speakers on the panel. The patient, a 67-year-old man, presented with acute visual loss in the left eye. Examination showed retinal whitening and a cherry red spot, which is typically associated with central retinal artery occlusion (CRAO). Chen positioned the two choices available to a treating physician: perform ocular massage and anterior paracentesis, and provide a primary care physician (PCP) referral for an outpatient stroke workup, or immediately refer the patient to an emergency department (ED) or stroke center.

While the first choice was more traditional, Chen said, it could result in a lack of visual improvement and put the patient at risk for a subsequent, significant stroke before the patient reached their PCP. However, immediate care in the ED would present opportunities for intervention to preserve visual acuity (such as intravenous thrombolysis). A potential underlying cause could then be addressed and, if necessary, physicians in the ED could take action to prevent stroke (such as a carotid endarterectomy for significant carotid stenosis). As Chen emphasised, symptomatic stroke occurs in 3% to 5% of CRAO cases within 2 to 4 weeks of the embolic event. Conservative treatments, like ocular massage and anterior paracentesis, are no longer considered the best course of action in professional guidelines.

Neuro-ophthalmology topics take the stage

On the second day of the meeting, participants debated additional topics such as neuro-ophthalmology, diabetic retinopathy and surgical techniques. One debate focused on visual snow syndrome (VSS); Eleni Papageorgiou, MD, a consultant in paediatric ophthalmology and neuro-ophthalmology at the University Hospital of Larissa, Greece, argued in favor of a full workup for these patients. Per the International Headache Society, VSS requires the presence of continuous, dynamic, small dots across the entire visual field for longer than 3 months. Symptoms cannot be explained by another disorder and patients do not report migraine visual aura. In addition, at least two of the following criteria must be present: palinopsia, photophobia, nyctalopia, and/or positive visual phenomena, such as floaters and photopsias.

Dr Papageorgiou recommended that clinicians take a thorough patient history, perform a routine ophthalmic exam with color vision and visual field tests, and complete a range of diagnostic measures including OCT, fundus autofluorescence and electroretinography. There is no standard treatment for VSS, in large part because there are several potential neurologic causes. Visual cortex abnormalities, traumatic brain injury, occipital lobe epilepsy, use of hallucinogenic drugs, and neurodegenerative diseases could all contribute to VSS. Dr Papageorgiou noted that most cases of VSS are benign and relatively stable, and patients should be counseled as such. However, physicians should watch for factors such as new-onset VSS (especially in older patients), intermittent occurrences and development of unilateral or quadrant VS, and reassess the patient as necessary.

Modern Retina also spoke with Andrew G. Lee, MD, Herb and Jean Lyman Centennial Chair in Ophthalmology at Blanton Eye Institute, Houston Methodist Hospital in Houston, Texas, US, about the neuro-ophthalmology segment he chaired. Dr Lee argued the negative position in a debate about whether all patients with idiopathic intracranial hypertension (IIH) should have a lumbar puncture (LP), also known as a spinal tap.

“It is controversial whether a spinal tap needs to be performed in idiopathic intracranial hypertension, but it’s a little bit of a misnomer because idiopathic means you already did the spinal tap,” he explained. “But of course, in the real world, some patients refuse to have a spinal tap, or they can’t have one, or they’re on aspirin, or some other reason that they can’t really have a spinal tap. And [with] some patients, despite your best efforts, you can’t get the spinal tap. You’re sticking the needle in and nothing is coming out.”

Thus, the argument against performing an LP on all patients is more a question of realistic treatment protocols in an imperfect industry. Because the high pressure of IIH can cause vision changes and episodes of total sight loss, Dr Lee advocated for treating the underlying condition, even if a diagnosis via LP was impossible. “If it looks like idiopathic intracranial hypertension, we would treat it as such, as long as they don’t have any atypical features, or have progression or unusual findings,” he recommended. “The teaching point shouldn’t be we shouldn’t do it, we should try. But when you can’t get it, you just have to make do with what you have.”