Comment on the Review: Clarifying Steroid-Associated Glaucoma—from Association to an Actionable Closed-Loop [Response to Letter]

Dear editor

We sincerely appreciate the interest in our review, “Safe Use of Corticosteroids in Non-Infectious Uveitis”,¹ and we would like to thank the authors for their thoughtful comments, particularly regarding the critical area of surveillance for steroid-associated glaucoma.²

We greatly value the closed-loop surveillance algorithm you propose. We agree that incorporating objective parameters and actionable thresholds substantially enhances the translational value of the information for clinicians.

We would like to take this opportunity to offer our point of view and constructively discuss your algorithm.

We appreciate your suggestion to use the term “association” rather than “causality” when discussing glaucoma progression with intraocular pressures (IOP) within the normal range. Indeed, in the context of non-infectious uveitis (NIU) when describing the long-term findings of the MUST trial, we have used terminology that reflects a clinical grounded probability that steroid exposure, beyond and not merely IOP-related, contributes to the observed damage, based on longitudinal data.^3–6 In other words, although formal and directed association of causality requires additional prospective and evidence-based research into the underlying mechanisms, we consider the MUST clinical evidence sufficiently compelling to treat steroid as an active contributing factor when progression occurs under controlled IOP. Nonetheless, we fully agree that prospectively elucidating these mechanisms remains a key priority for future research.

About the suggested algorithm, we consider the easy-to-consult scheme remarkable. Regarding retinal nerve fiber layer (RNFL) thinning progression thresholds, in NIU it reflects a complex interplay. First, the normal age-related loss of nerve fiber, which differs among decades. Usually, patients with NIU range between 20 and 50 years old, when median RNFL loss span between 0.24 μm/year and 0.54 μm/year.^7–9 Moreover, individual factors (ie axial length, RNFL baseline thickness and RNFL region) may influence this parameter. Second, the inflammatory activity which may mask the damage and paradoxically increase RNFL thickness.¹⁰ Third, the RNLF thinning cause by steroid-related elevated IOP.

Weighing up these aspects, in the context of a high-risk inflamed eye exposed to long-acting steroids, we agree that it is appropriate to define progression as any rate of loss exceeding the “desired level” of stability (even lower than ≤1 μm/year). Notably, accepting such a low threshold implies treating even patients with normal progression of thinning, while avoiding undertreatment of patients who require treatment.

This threshold is intended as a sensitive trigger for early intervention, prioritizing neuroprotection in a risk-laden setting and prompting re-evaluation of whether the target IOP or the drug exposure are truly safe.

We would like to highlight the value of the structural sensitivity and the intensive schedule you propose, both of which are fundamental for neuroprotection in high-risk eyes.

We wish to emphasize our concern that, in practice, both visual field (VF) testing and optical coherence tomography (OCT) may yield false-negative or difficult-to-interpret results in the presence of active inflammation or sequelae such as uveitic macular oedema. For this reason, critical structural and functional assessments of progression (OCT and VF) require direct clinical evaluation that allow for a complete clinical evaluation and ensure that active inflammation is not the underlying cause of reduced visual acuity or structural deterioration.

In this setting, we would highlight the challenging role of suprachoroidal triamcinolone acetonide as an alternative local route positioning implant. While it stands as a good lower-IOP-risk option for prior steroid responders or glaucoma suspects, further studies with longer follow-up are warranted to determine whether or not this route of administration is safer than others. Also it has limited availability worldwide nowadays and requires a steeper learning curve compared to other implants and specialized equipment.

We once again appreciate your valuable contribution, which undoubtedly enriches the discussion on clinical practice guidelines and helps shift the focus from mere association towards an actionable risk-management strategy.