This cohort study assesses the rate of retinal intervention in patients with minimal or no diabetic retinopathy within 2 years after their telemedicine screening.
Key Points
Question
What is the 2-year rate of retinal intervention in patients with minimal or no diabetic retinopathy on telemedicine screening?
Findings
In this cohort study, 11 of 69 364 patients with no baseline retinopathy and 11 of 9811 patients with minimal baseline retinopathy required a retinal intervention because of diabetic eye disease within 2 years of the baseline evaluation.
Meaning
The findings suggest that extending follow-up examinations for patients with minimal or no baseline diabetic retinopathy may be warranted because most may not need retinal intervention within 2 years of initial evaluation, provided this extension does not lead to worse follow-up in later years.
Abstract
Importance
Most patients with diabetes have little or no retinopathy on initial examination. Tracking the long-term outcomes of these patients may increase our understanding of how to best provide follow-up treatment.
Objective
To assess how many patients with minimal or no retinopathy require retinal intervention within 2 years of retinal evaluation.
Design, Setting, and Participants
This retrospective cohort study assessed patients who underwent screening for diabetic retinopathy within a telemedicine program at Kaiser Permanente Southern California and had minimal or no retinopathy on fundus photographs.
Exposure
Retinal interventions performed within 2 years of photographs.
Main Outcomes and Measures
Patients with minimal or no retinopathy on initial screening photographs taken in 2012 had their medical records searched for Current Procedural Terminology codes for intravitreal injections, retinal lasers, or pars plana vitrectomy. The medical records of patients identified as having received these interventions within 2 years of retinal evaluation were then manually reviewed for further characterization.
Results
Diabetic retinopathy screening photographs were taken for 116 134 patients (mean [SD] age, 58 [12.8] years; 54 582 [47.0%] female; 46 453 [40.0%] Latino). Of these patients, 79 445, including 69 634 patients without retinopathy and 9811 patients with minimal retinopathy, had 2 years of follow-up. Eleven patients without baseline retinopathy required treatment of diabetic retinopathy in the following 2 years (1 of 12 660 or 0.000079 patients per year), and 11 patients with minimal retinopathy required intervention during the same period (1 of 1784 or 0.000561 patients per year). In addition, retinal interventions were performed for conditions not directly related to diabetic eye disease in 44 patients without baseline retinopathy (1 of 3165 or 0.000316 patients per year) and 5 patients with minimal retinopathy at baseline (1 of 3924 or 0.000255 patients per year).
Conclusions and Relevance
These findings suggest that it is rare for patients with minimal or no baseline retinopathy to require retinal interventions in the 2 years after retinal evaluation. It appears that extending the recommended follow-up interval for low-risk patients may be reasonable as long as this does not lead to worse follow-up in later years, because most are unlikely to have vision-threatening disease that necessitates treatment.
Introduction
Retinopathy screening is a cornerstone in the health care management of patients with diabetes and central to preserving vision1,2; however, significant difficulties remain in achieving appropriate ophthalmic care for many patients. Telemedicine and teleophthalmology programs improve diabetic retinopathy detection and represent an increasingly popular means to help offset the increasing demand for diabetic retinal screening.1,2,3,4 Several methods of fundus photography have been used for teleophthalmology programs. The increased peripheral retinal visualization afforded by ultrawide field photography allows for increased detection of nondiabetic peripheral retinal changes and can result in higher grading of diabetic retinopathy severity compared with traditional nonmydriatic fundus photographs1,2,5,6; however, it remains unclear whether this alters practical clinical outcomes.
Although the importance of screening eye examinations is well established, the follow-up interval for patients remains sub judice. The American Academy of Ophthalmology recommends a minimum of yearly screening examinations, with shorter intervals recommended for those with increasing severity of diabetic retinopathy. The possibility of screening low-risk patients every 2 years has been advocated.7 An ideal screening interval would detect patients who require intervention to help prevent vision loss while limiting the number of unnecessary examinations.
Methods
The incidence of initiating retinal interventions in patients with minimal or no baseline diabetic retinopathy during 2 years was examined in this retrospective cohort study. The diabetic retinopathy telemedicine screening program at the Kaiser Permanente Southern California’s Eye Monitoring Center has been previously described and validated.3 Screening photographs consisted of two 45° fundus photographs: one centered on the optic nerve and one on the macula. Results were reviewed by a vitreoretinal specialist. Patients with moderate retinopathy or worse, including those with macular exudates, were referred for clinical evaluation. All the patients belong to the same integrated health care network, which includes primary care physicians and specialists, and any interventions rendered can be monitored through electronic medical records. Kaiser Permanente Southern California’s Institutional Review Board granted approval for this study. Informed consent was waived by the institutional review board because the research involves no more than minimal risk to participants.
Patients who received initial screening photographs in 2012 and who were graded as having minimal or no retinopathy and had at least 2 years of follow-up were included in this study. The medical records of patients with minimal or no retinopathy were cross-referenced with Current Procedural Terminology codes used for intravitreal injections, retinal lasers, and pars plana vitrectomy. Identified patients then had their medical records manually reviewed; the clinical indication and type of interventions were subsequently categorized. Vitrectomies that were performed at the time of cataract surgery, to remove retained lens fragments, or related to trauma were excluded.
Statistical Analysis
Descriptive analyses were performed to characterize the target population and to provide counts of interventions among patients with minimal or no retinopathy at baseline. Incidence rate of interventions per person-year were calculated.
Results
A total of 116 134 patients (mean [SD] age, 58 [12.8] years; 54 582 [47.0%] female; 46 453 [40.0%] Latino) had diabetic retinopathy screening photographs taken, with 81 544 patients having no retinopathy and 11 689 having minimal retinopathy. Of these patients, 69 634 (85.4%) without retinopathy and 9811 (83.9%) with minimal retinopathy had 2 years of follow-up.
Eleven patients without baseline retinopathy required the initiation of treatment for complications of diabetic retinopathy, and 44 patients required an intervention because of conditions not directly related to diabetic eye disease in the 2 years after their baseline photographs (Table 1). This rate represents 0.000079 patients (or 1 of 12 660) per year requiring intervention for diabetic retinopathy and 0.000316 patients (or 1 of 3165) per year requiring interventions for nondiabetic conditions among those without baseline retinopathy. Eleven patients with minimal retinopathy on initial screening photographs required the initiation of treatment related to diabetic retinopathy (1 of 1784 or 0.000561 patients per year), and 5 patients required intervention for nondiabetic conditions (1 of 3924 or 0.000255 patients per year) during 2 years (Table 2).
Table 1. Initiation of Retinal Interventions in Patients Without Retinopathy During 2 Years.
| Intervention | No. of Interventions for Diabetic Eye Disease (Type of Intervention)a | No. of Interventions Not for Diabetic Eye Disease (Type of Intervention)b | Total |
|---|---|---|---|
| Year 1 | |||
| PPV | 2 (2 NCVH) | 15 (8 RRD, 4 MH, 1 MH/ERM, 1 NCVH, 1 lymphoma) | 17 |
| Laser | 3 (2 PRP for PDR, 1 FML for ME) | 4 (3 Retinopexy, 1 PRP after BRVO) | 7 |
| Intravitreal injection | 2 (Anti-VEGF for NCVH) | 0 | 2 |
| Year 2 | |||
| PPV | 4 (NCVH) | 27 (7 MH, 4 RRD, 4 ERM, 3 MH/ERM, 4 NCVH, 1 melanoma, 1 RRD/ERM, 1 CMV retinitis/RRD, 1 posterior scleritis, 1 vitritis) | 31 |
| Laser | 4 (3 PRP for PDR, 1 FML for ME) | 2 (Retinopexy) | 6 |
| Intravitreal injection | 1 (Anti-VEGF for ME) | 2 (1 Anti-VEGF for BRVO with ME, 1 ocriplasmin for MH) | 3 |
| First 2 y combined | |||
| PPV | 6 | 42 | 48 |
| Laser | 7 | 6 | 13 |
| Intravitreal injection | 3 | 2 | 5 |
Abbreviations: BRVO, branch retinal vein occlusion; CMV, cytomegalovirus; ERM, epiretinal membrane; FML, focal macular laser; ME, macular edema; MH, macular hole; NCVH, nonclearing vitreous hemorrhage; PDR, proliferative diabetic retinopathy; PPV, pars plana vitrectomy; PRP, panretinal photocoagulation; RRD, rhegmatogenous retinal detachment; VEGF, vascular endothelial growth factor.
Eleven patients required intervention during the first 2 years; 2 patients required both PPV and laser, 1 patient required both PPV and injection, and 1 patient required PPV, laser, and injection.
Forty-four patients required intervention during first 2 years; 5 patients required both PPV and laser, and 1 patient required both PPV and injection. Three patients required interventions in both eyes (bilateral PPV) and were only counted once in the table.
Table 2. Initiation of Retinal Interventions in Patients With Minimal Retinopathy During 2 Years.
| Intervention | No. of Interventions for Diabetic Eye Disease (Type of Intervention)a | No. of Interventions Not for Diabetic Eye Disease (Type of Intervention)b | Total |
|---|---|---|---|
| Year 1 | |||
| PPV | 2 (1 TRD, 1 NCVH) | 2 (1 RRD, 1 ERM) | 4 |
| Laser | 5 (PRP for PDR) | 1 (Retinopexy) | 6 |
| Intravitreal injection | 3 (Anti-VEGF for ME) | 2 (1 Anti-VEGF for postoperative ME, 1 anti-VEGF for exudative age-related macular degeneration) | 5 |
| Year 2 | |||
| PPV | 0 | 1 (ERM) | 1 |
| Laser | 2 (1 FML for ME, 1 PRP for PDR) | 0 | 2 |
| Intravitreal injection | 1 (Anti-VEGF for ME) | 0 | 1 |
| First 2 y combined | |||
| PPV | 2 | 3 | 5 |
| Laser | 7 | 1 | 8 |
| Intravitreal injection | 4 | 2 | 6 |
Abbreviations: ERM, epiretinal membrane; FML, focal macular laser; ME, macular edema; MH, macular hole; NCVH, nonclearing vitreous hemorrhage; PDR, proliferative diabetic retinopathy; PPV, pars plana vitrectomy; PRP, panretinal photocoagulation; RRD, rhegmatogenous retinal detachment; VEGF, vascular endothelial growth factor.
Eleven patients required intervention during first 2 years; 1 patient required both PPV and laser, and 1 patient required both laser and injection.
Five patients required intervention during first 2 years; 1 patient required both PPV and laser.
The most common intervention for treatment of diabetic retinopathy was panretinal photocoagulation for proliferative diabetic retinopathy in patients with minimal (6 patients [0.06%]) and those without (5 patients [0.007%]) baseline retinopathy. Pars plana vitrectomy was the most common intervention not related to diabetic eye disease, with 42 patients (0.06%) without baseline retinopathy and 3 patients (0.03%) with minimal baseline retinopathy undergoing this procedure. The most common indications for pars plana vitrectomy were macular holes (with or without concurrent epiretinal membrane, n = 15 [0.02%]) and rhegmatogenous retinal detachments (n = 15 [0.02%]).
Discussion
The importance of diabetic retinopathy screening is well established; however, the ideal interval for evaluation remains unclear. In addition, the utility of expanded visualization of the peripheral retina is an area of interest given the increasing use of ultrawide field photography. Extending follow-up intervals in low-risk patients may help alleviate some of the strain on health care systems.
Silva et al5 detected more choroidal nevi (6.9% vs 5.3%, P < .001), chorioretinal atrophy or scarring (1.3% vs 0.6%, P < .001), retinal tears (0.3% vs 0%, P < .001), lattice and peripheral degenerations (0.7% vs 0%, P < .001), and vitreous detachments and floaters (1.8% vs 0%, P < .001) with ultrawide field photographs compared with conventional nonmydriatic fundus photographs; however, the clinical importance of detecting these changes remains questionable because most do not necessitate treatment or a change in future diagnostic approach. More patients undergoing imaging with ultrawide field imaging were identified as having diabetic retinopathy (38.4% vs 33.8%, P < .001) and vision-threatening diabetic retinopathy (14.5% vs 11.9%, P < .001) than with conventional nonmydriatic fundus photographs in another series by Silva et al.6 Although differences between imaging types in the studies by Silva et al5,6 were statistically significant, the percentage of affected patients and the absolute difference between photography methods were modest for most measures, raising the possibility that the practical utility of ultrawide field photographs would be limited for most patients. Some of the patients likely had peripheral findings that were not detected; however, not detecting these lesions appeared to be of limited consequence because most patients did not go on to require the initiation of treatment. This finding suggests that the status of the far peripheral retinal is unlikely to change the clinical course for most patients during 2 years. The incidence of retinal intervention was used as an indirect measure of visually significant disease.
Future investigations should help better refine screening intervals, potentially by factoring in additional variables to create more personalized risk stratification. Ultrawide field views of the retina confer several advantages and may represent the future standard care; however, transitioning to these cameras represents a significant cost for established practices and reading centers. Traditional fundus photography is likely sufficient in most patients, especially those who are found to have little or no retinopathy. At present, the role of ultrawide field photography in patients with diabetes is under active investigation.8
Limitations
A limitation of this study is the use of Current Procedural Terminology coding data to identify which patients underwent retinal procedures, which can lend itself to inaccuracies because of coding errors, as well as the descriptive nature of the analyses.
Conclusions
The results presented herein suggest that patients with minimal or no retinopathy are unlikely to require the initiation of retinal treatment for diabetic retinopathy within 2 years of screening, and as such, most patients may not benefit from an additional examination before the second year. Efforts to extend follow-up intervals would need to ensure that adherence at later years is not compromised. Most interventions were rendered for conditions not directly related to diabetic retinopathy. Additional study is required to determine the ideal screening intervals for patients with diabetes as well as the economic, clinical, and practical value to ultrawide field imaging.
References
- 1.Gupta A, Cavallerano J, Sun JK, Silva PS. Evidence for telemedicine for diabetic retinal disease. Semin Ophthalmol. 2017;32(1):22-28. doi: 10.1080/08820538.2016.1228403 [DOI] [PubMed] [Google Scholar]
- 2.Silva PS, Aiello LP. Telemedicine and eye examinations for diabetic retinopathy: a time to maximize real-world outcomes. JAMA Ophthalmol. 2015;133(5):525-526. doi: 10.1001/jamaophthalmol.2015.0333 [DOI] [PubMed] [Google Scholar]
- 3.Hudson SM, Contreras R, Kanter MH, Munz SJ, Fong DS. Centralized reading center improves quality in a real-world setting. Ophthalmic Surg Lasers Imaging Retina. 2015;46(6):624-629. doi: 10.3928/23258160-20150610-05 [DOI] [PubMed] [Google Scholar]
- 4.Mansberger SL, Sheppler C, Barker G, et al. . Long-term comparative effectiveness of telemedicine in providing diabetic retinopathy screening examinations: a randomized clinical trial. JAMA Ophthalmol. 2015;133(5):518-525. doi: 10.1001/jamaophthalmol.2015.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Silva PS, Cavallerano JD, Haddad NM, et al. . Comparison of nondiabetic retinal findings identified with nonmydriatic fundus photography vs ultrawide field imaging in an ocular telehealth program. JAMA Ophthalmol. 2016;134(3):330-334. doi: 10.1001/jamaophthalmol.2015.5605 [DOI] [PubMed] [Google Scholar]
- 6.Silva PS, Cavallerano JD, Tolls D, et al. . Potential efficiency benefits of nonmydriatic ultrawide field retinal imaging in an ocular telehealth diabetic retinopathy program. Diabetes Care. 2014;37(1):50-55. doi: 10.2337/dc13-1292 [DOI] [PubMed] [Google Scholar]
- 7.Taylor-Phillips S, Mistry H, Leslie R, et al. . Extending the diabetic retinopathy screening interval beyond 1 year: systematic review. Br J Ophthalmol. 2016;100(1):105-114. doi: 10.1136/bjophthalmol-2014-305938 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Aiello LP, Odia I, Glassman AR, et al. ; Diabetic Retinopathy Clinical Research Network . Comparison of early treatment diabetic retinopathy study standard 7-field imaging with ultrawide-field imaging for determining severity of diabetic retinopathy [published online October 18, 2018]. JAMA Ophthalmol.doi: 10.1001/jamaophthalmol.2018.4982 [DOI] [PMC free article] [PubMed] [Google Scholar]