This study characterized the rates of panretinal photocoagulation (PRP) and anti-VEGF medications before and after publication of the Diabetic Retinopathy Clinical Research Network Protocol S.
This retrospective, cross-sectional study analyzed the impact of DRCR.net Protocol S on the use of anti-VEGF injections versus PRP for the treatment of proliferative diabetic retinopathy (PDR) with and without diabetic macular edema (DME). Protocol S prepublication period was defined as January 1, 2012, through December 31, 2015, and the postpublication period was January 1, 2016, through September 30, 2019. A nationally representative claims-based database was used for analysis. Eyes with continuous enrollment and newly diagnosed PDR with no prior history of anti-VEGF injections or PRP were included.
Over the 8-year study period, 63.6% of PDR treatments were anti-VEGF injections versus 36.4% that were PRP. Among PDR eyes with DME, there was no significant difference in PRP use before and after Protocol S. In contrast, for treated eyes without DME, PRP rates declined slightly during the period before Protocol S publication, but declined significantly after publication (P<0.001). The use of anti-VEGF did not change significantly in the period prior to publication, but increased significantly following publication.
The retrospective nature as well as use of claims-based data are inherent limitations of this study. In addition, there are likely multiple other factors that led to the decline in use of PRP and the increase in the use of anti-VEGF injections outside of the publication of Protocol S. These factors confound the results of this study. For example, in January of 2016, there was a 66% decrease in reimbursement for CPT code 67228. PRP takes more time and skill than performing intravitreal injections. With the decrease in reimbursement for PRP and the lack of efficiency of performing the procedure in a busy office setting, intravitreal injections become more convenient to perform than PRP.
The results of this study indicate that randomized controlled trials have an impact on real-world practice patterns. Reimbursement as well as practice efficiency likely played a significant role in the decrease in PRP use over time. It is possible that because Protocol S demonstrated that ranibizumab is noninferior to PRP for PDR, this gave license to practitioners to avoid PRP. Understanding the inefficiency of PRP, retina specialists must also consider the risk of severe progression of PDR in the setting of noncompliance. We know that if a patient stops anti-VEGF injections, there is a significant risk of progression to a traction retinal detachment and possible permanent loss of vision. PRP can protect against this severe progression. The decline in PRP use demonstrated by this study is clear, but retina specialists need to consider noncompliance when choosing anti-VEGF injections over PRP.