Challenges in the Clinical Management of Proliferative Diabetic Retinopathy: Treatment Choice and Follow-up

Tsui et al address the issue of whether there is an association between loss to follow-up or treatment approach with future likelihood of poor outcomes by performing a nested case-control study of a medical claims database. The authors examined the association of different types of prior treatment for PDR and a history of 6-month period of loss to follow-up with development of tractional retinal detachment (TRD). In this study, patients with PDR who progressed to TRD were matched with patients with PDR without TRD based on similar diagnosis year of PDR and inclusion criteria as of the index date. Tsui et al performed conditional logistic regression models for TRD with adjustment for several covariates and reported there was no difference in the odds of developing TRD between patients with vs without a history of a 6-month or greater period of loss to follow-up and between patients who received laser only vs those who received anti-VEGF injections only.

An unacknowledged limitation of this study lies in the retrospective nature of the data collection, which makes it impossible to determine whether there is a causal association between treatment type and TRD outcome or whether the results are confounded by unmeasured variables, such as physicians' clinical decision-making to avoid or prefer specific treatments for patients with eyes at higher risk of progression to TRD. It is possible that the authors have not identified a significant association of TRD with prior anti-VEGF treatment because patients in this cohort at higher risk of TRD were less likely to be treated with anti-VEGF therapy alone.

The article by Tsui et al highlights important challenges faced in the management of patients with PDR. A total of 50% or more of both the TRD and non-TRD subgroups were lost to follow-up for 6 or more months during the follow-up period. Ophthalmologists need better ways to ensure that patients at risk of vision loss from diabetic retinopathy have consistent and regular ophthalmic follow-up. Although loss to follow-up was happily not associated with TRD in this cohort, it is clear that poor visit adherence is a barrier to prompt diagnosis and timely treatment of vision-threatening disease. Large proportions of the TRD group (41.6%) and non-TRD group (64.5%) had no treatment. Although it is possible that all the untreated eyes in the non TRD group had mild PDR that did not warrant treatment, the fact that more than 40% of eyes that developed TRD remained untreated speaks to the challenges in treatment delivery that are faced by many patients with PDR globally.

Fortunately, there are a variety of approaches available for the treatment of PDR, including anti-VEGF alone, panretinal photocoagulation alone, a combination of these 2 treatment modalities and vitrectomy surgery. Yet despite these effective therapies, patients across the globe continue to lose vision from PDR. Although development of novel therapies for PDR is an important priority in the field, solutions to the problems of loss to follow-up and under treatment in patients with PDR are sorely needed and would have enormous positive public health impact.

Source: Jennifer K. Sun, MD, MPH; Danni Liu, MSPH; JAMA Ophthalmology

doi:10.1001/jamaophthalmol.2022.5057