Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Non-Mydriatic Fundus Camera Screening for Referral-Warranted Diabetic Retinopathy in a Northern California Safety-Net Setting

Brian C. Toy, MD; Tyler Aguinaldo, MD; Joseph Eliason, MD; James Egbert, MD

Abstract

BACKGROUND AND OBJECTIVE:

To compare nonmydriatic fundus photography (nFP) screening for diabetic retinopathy (DR) with clinical dilated fundus examination (DFE) in a safety-net setting.

PATIENTS AND METHODS:

This was a retrospective, institutional cohort study of 6,911 patients undergoing nFP screening for DR between 2008 and 2012. A subset of 1,521 patients underwent DFE, and clinical DR grade was compared with nFP DR grade.

RESULTS:

nFP screening demonstrated 17% prevalence of any DR, with moderate nonproliferative diabetic retinopathy (NPDR) or worse present in 5% of patients. Thirteen percent of photos were unreadable. When compared with DFE, sensitivity and specificity of nFP to detect moderate NPDR or worse were 91% and 97%, respectively. Area under the receiver operating characteristic curve was 0.97.

CONCLUSIONS:

The authors' study found 17% of patients screened with nFP had referral-warranted DR. These results demonstrate a use-case of telemedicine to screen large numbers of patients in a safety-net setting, but they also highlight the need for adequate specialty resources to care for referred patients.

[Ophthalmic Surg Lasers Imaging Retina. 2016;47:636–642.]

Abstract

BACKGROUND AND OBJECTIVE:

To compare nonmydriatic fundus photography (nFP) screening for diabetic retinopathy (DR) with clinical dilated fundus examination (DFE) in a safety-net setting.

PATIENTS AND METHODS:

This was a retrospective, institutional cohort study of 6,911 patients undergoing nFP screening for DR between 2008 and 2012. A subset of 1,521 patients underwent DFE, and clinical DR grade was compared with nFP DR grade.

RESULTS:

nFP screening demonstrated 17% prevalence of any DR, with moderate nonproliferative diabetic retinopathy (NPDR) or worse present in 5% of patients. Thirteen percent of photos were unreadable. When compared with DFE, sensitivity and specificity of nFP to detect moderate NPDR or worse were 91% and 97%, respectively. Area under the receiver operating characteristic curve was 0.97.

CONCLUSIONS:

The authors' study found 17% of patients screened with nFP had referral-warranted DR. These results demonstrate a use-case of telemedicine to screen large numbers of patients in a safety-net setting, but they also highlight the need for adequate specialty resources to care for referred patients.

[Ophthalmic Surg Lasers Imaging Retina. 2016;47:636–642.]

Introduction

Approximately 387 million individuals worldwide and 29.1 million Americans are affected by diabetes mellitus.1,2 Diabetic retinopathy (DR) and diabetic macular edema (DME) are common ophthalmic sequelae of diabetes, resulting from variable degrees of retinal capillary hyperpermeability and nonperfusion. Subsequent retinal ischemia and neovascularization may result in proliferative diabetic retinopathy (PDR).

The prevalence of retinopathy among patients with diabetes has been estimated to be 28.5%, and diabetes is the leading cause of blindness in adults in the United States.3 If detected and treated in a timely fashion, blindness from DR progression can be mitigated.4,5 Nevertheless, rates of DR screening remain low, particularly in the health care safety-net setting, perhaps related to insufficient provider availability, poor adherence to regular exams, and cost to the patient.6–8

The gold standard for diagnosing DR, as defined by the Early Treatment of Diabetic Retinopathy Study (ETDRS), is seven-field stereoscopic dilated fundus photographs graded by the modified Airlie House protocol.9 Unfortunately, implementing this in the primary care and safety-net settings can be difficult due to requisite cost, staff, and infrastructure, as well as patient acceptance of pharmacologic dilation and the small (1:20,000) risk of precipitating acute angle closure. Thus, the use of non-mydriatic fundus photography (nFP) in telemedicine has shown promise in screening for DR in these settings.10–15

At the time of this study, 18,800 patients received care for diabetes at Santa Clara Valley Medical Center (SCVMC). This safety-net health care system, which predominantly serves uninsured and underinsured patients, has a reported prevalence of DR of 27%, with a disproportionate burden of disease affecting patients of Latino race.16 Patients in this setting are at high risk for experiencing health and health care disparities. Therefore, improving access to screening, surveillance, and treatment for DR represents an important goal in this population; however, the large patient load presents a challenge for traditional models of patient care delivery. The present study supports the use of telemedicine, and specifically, nFP screening, as one part of a comprehensive system to screen for and treat diabetes and DR.

Patients and Methods

Study Design and Patient Selection

This retrospective, single-center, longitudinal cohort study included adult patients who underwent nFP DR screening as part of clinical care for diabetes at SCVMC, the safety-net hospital in Santa Clara County, CA, from April 2008 through September 2012. The study protocol was approved by the SCVMC institutional review board, complied with the Health Insurance Portability and Accountability Act, and adhered to the tenets of the Declaration of Helsinki.

Data Collection

Encounter data from separate electronic medical record systems (SCVMC Eye Clinic: 4D EyeCare [4D SAS, San Jose, CA] and SCVMC Diabetes Clinic: FileMaker Pro [FileMaker, Santa Clara, CA]) were abstracted and linked by patient. Of 16,251 patients receiving screening exams for DR during the study time period, 6,911 underwent nFP screening for DR, with some patients seen more than once, yielding a total of 9,125 patient encounters.

A Nidek NM-1000 (Nidek, Fremont, CA) nonmydriatic fundus camera located in the SCVMC Diabetes Clinic was used to acquire a single non-mydriatic 45°-field-of-view fundus image centered on the macula. The images were transmitted for later grading by a single board-certified ophthalmologist (JE) into one of four categories: no DR, minimal DR (microaneurysms only), referral-warranted DR (moderate nonproliferative diabetic retinopathy [NPDR] or worse, presence of lipid exudates), or photos unreadable (based on photo clarity and field of view). When the grade between fellow eyes differed, the more severe grade was assigned to the overall encounter. Patients with no or minimal DR were scheduled for follow-up examination within 1 year, either by fundus photography or with an inclinic dilated fundus exam (DFE). Patients with referral-warranted retinopathy or unreadable photos were referred for prompt ophthalmic evaluation and potential treatment at the SCVMC Eye Clinic.

Patients evaluated at the SCVMC Eye Clinic underwent DFE, including slit lamp biomicroscopy and indirect ophthalmoscopy. The DR clinical grade for this subset of patients was determined based on the five-level International Clinical Diabetic Retinopathy (ICDR) Disease Severity Scale:17,18 no retinopathy, mild NPDR, moderate NPDR, severe NPDR, or PDR. The presence or absence of diabetic macular edema (DME) was also noted. When noted, the etiology of unreadable nFP was also abstracted from the clinic note.

Statistical Analysis

Comparisons of clinic and nFP grades were performed using Stata (StataCorp; College Station, TX) to calculate descriptive, receiver operating characteristic (ROC), and multivariable logistic regression with generalized estimating equation analyses. Where possible, analyses were clustered by patient to account for the correlation between fellow eyes of a single patient, as well as longitudinal data from patients with more than one visit. Numeric values are reported as mean ± standard deviation, unless otherwise noted.

Results

Patient Demographics

Patients in the current study were identified from a larger cohort of patients with diabetes as those undergoing nFP screening for DR. This longitudinal study included 6,911 patients, yielding a total of 9,125 nFP screening visits over the study period.

Demographic and clinical data for patients at the time of their first nFP are summarized in Table 1. Nearly half (47%) of the patients were male. The mean ± standard deviation (SD) age was 55.4 years ± 12.2 years. Nearly half (46%) of patients were Latino, 24% of patients were Asian/Pacific Islander, and 17% were white. The mean ± SD age of diagnosis with diabetes was 46.8 years ± 11.7 years, and the mean duration of diabetes was 9.4 years ± 2.8 years. The mean ± SD serum hemoglobin A1c (HgbA1c) level was 8% ± 1.9%. Forty-one percent of patients were without health insurance, whereas nearly half (49%) had public government-sponsored health insurance.


Demographic and Clinical Characteristics of Patients Undergoing Non-Mydriatic Fundus Camera Screening for Diabetic Retionpathy

Table 1:

Demographic and Clinical Characteristics of Patients Undergoing Non-Mydriatic Fundus Camera Screening for Diabetic Retionpathy

The patients in this cohort had a variety of cardiovascular comorbidities. Forty-nine percent of patients had systemic hypertension. Seventy-four percent of patients were overweight, based on a body mass index (BMI) above 25 kg/m2, and the mean ± SD BMI was 31.7 kg/m2 ± 7.8 kg/m2. Seventy percent of patients had dyslipidemia. Thirty-two percent of patients had albuminuria, and 14% of patients had an estimated glomerular filtration rate less than 60 mL/min, indicating impaired renal function corresponding to stage 3 or worse chronic kidney disease.

Non-Mydriatic Fundus Camera Screening

A total of 9,125 pairs of nFP images were graded during the study period. Most (72%) encounters demonstrated no DR (Table 2), whereas 12% and 5% demonstrated minimal or referral-warranted retinopathy, respectively. From 856 patients, 1,187 (13%) encounters were graded as unreadable. A subset of 305 patients with unreadable photos presented to eye clinic, and the most common identified cause was cataract (53%) (Table 3). Excluding encounters with unreadable photos, the overall prevalence of any DR was 17%, with 5% of patients having referral-warranted retinopathy.


Non-Mydriatic Photo Grading

Table 2:

Non-Mydriatic Photo Grading


Etiology of Unreadable Photos

Table 3:

Etiology of Unreadable Photos

A total of 1,013 patients were referred for a complete ophthalmic examination after nFP camera screening. Of these patients, 711 (70%) presented to the SCVMC Eye Clinic. Of the remaining patients referred for further DR evaluation, 266 (26%) reported examination with an outside eye doctor, and 36 (4%) were lost to follow-up. An additional 810 patients without retinopathy on nFP were evaluated in clinic for DFE unrelated to DR screening, yielding a total of 1,521 patients evaluated by nFP who were also seen in our clinic. The nFP and clinical grades of the more severely affected eye for these patients are detailed in Table 4. Patients with moderate NPDR or worse may benefit from ophthalmic care and treatment.19,20 Thus, in order to evaluate the effectiveness of nFP to make this distinction, the clinical ICDR grade was dichotomized to no retinopathy/ mild NPDR versus moderate or severe NPDR/any PDR. Compared to the clinical ICDR grade, nFP was noted to have the following characteristics: sensitivity, 91%; specificity, 97%; positive predictive value (PPV), 85%; negative predictive value (NPV), 98%; kappa = 0.86. Including unreadable photos (dichotomizing the photo grade to no/minimal retinopathy vs. referral-warranted retinopathy/unreadable) and dichotomizing the clinical grade as before, the sensitivity of nFP screening to detect moderate NPDR or worse disease increased to 93%, specificity decreased to 75%, PPV decreased to 42%, NPV remained 98%, and kappa decreased to 0.45. The area under the receiver operating characteristic was 0.97.


Photo Grade and Corresponding Clinical Grade of DR and DME

Table 4:

Photo Grade and Corresponding Clinical Grade of DR and DME

Multivariable logistic regression identified patient factors that may be associated with referral-warranted disease (Table 5). Factors found to be associated with more severe retinopathy were as follows: male sex, uninsured status, Latino race, longer duration of diabetes, systemic hypertension, HgbA1c greater than 9%, and nephropathy with albuminuria. HgbA1c of 7% or less was associated with decreased risk of referral-warranted DR.


Patient Factors Associated With Referral-Warranted Diabetic Retinopathy

Table 5:

Patient Factors Associated With Referral-Warranted Diabetic Retinopathy

Discussion

The present study provides data on DR prevalence in a large and ethnically diverse safety-net population in Northern California. With nearly 7,000 patients examined longitudinally during more than 9,000 encounters spanning 4 years, to our knowledge, this represents one of the largest reports to date of telemedicine screening for DR. Others have recommended basing DR screening equipment within primary care clinics,21 which may account in part for the large screening volume in this study. With more than 35,000 identified patients with diabetes now in our population, there is great need for timely triage and appropriate ophthalmic referral.

Our protocol acquired single 45° nFP for subsequent grading. Most photos were judged of sufficient quality to grade, with 13% of photos deemed unreadable, similar to other studies utilizing nFP for DR.15,20,22–28 To address unreadable photos due to poor dilation and media opacity, we have now incorporated a protocol for selected patients to undergo mydriatic photography.29

Seventeen percent of screening encounters in this study demonstrated some degree of DR, with 5% having moderate NPDR or worse. Reports from other community-based groups have indicated similar DR prevalence ranging from 11% to 23%.15,22,30–33 A prior study of 686 patients at our county hospital undergoing in-clinic DFE for DR screening in 2011 demonstrated a prevalence of 27%.16 The difference in prevalence compared to the present study may be multifactorial. In both studies, Latino race was found to be associated with a higher DR prevalence. The proportion of Latino patients in that cohort was higher (54%), compared to the present cohort (46%). The clinic population from which the prior study was drawn included patients who had been previously screened (either in-clinic or with telemedicine) and sent for follow-up to that clinic, which may represent selection bias in that sample. Furthermore, in the present telemedicine study, all included patients were screened through the diabetes clinic, and as such, may have had more resources to assist them in glycemic control and other microvascular risk factor modification, resulting in a lesser degree of DR. Finally, it is possible that telemedicine screening may have missed cases of DR, but we examined a subset of patients who underwent both telemedicine and in-clinic screening.

A subset of 1,521 (22%) patients underwent full ophthalmic evaluation after fundus camera screening. For these patients, the degree of DR on clinical exam and nFP was compared, with the clinical exam used as the standard, as in previous similar studies.24,34 Prior studies have directly compared ophthalmoscopy and seven-field ETDRS fundus photography, considered the gold standard in determining the severity of DR, finding agreement within one step in 34% to 86% and kappa ranging from 0.4 to 0.75.20,35,36 Cases of disagreement were found to be rarely of clinical significance and occurred most often in early retinopathy. As in prior telemedicine studies, patients with at least moderate NPDR were referred for full ophthalmic examination to evaluate more closely for disease requiring intervention.37,38

In the present study, the sensitivity and specificity of nFP for detecting referable DR were 91% and 97%, respectively, in line with findings of a systematic review by the American Academy of Ophthalmology, which concluded that “single-field fundus photography can be used as a screening tool to identify patients with DR who require referral for ophthalmic evaluation and management.”39

Sixteen patients were graded to have no or minimal retinopathy on nFP but subsequently noted to have referral-warranted disease on clinical exam. They fell into two general categories: (1) borderline findings elucidated with selective mydriatic photography: five patients had borderline findings and underwent mydriatic photography that detected referral-warranted disease as part of an update to our protocol in 2011, and these patients were then seen in clinic; (2) confirmed false-negative reports: 11 patients were found in clinic to have moderate NPDR or worse disease. Five of these patients were noted to progress rapidly to PDR within several months, and one patient was noted to have neovascularization outside the field of view of nFP. These rapidly progressing patients all had HgbA1c greater than 10% and durations of diabetes of longer than 12 years. Of note, the mean ± standard error of mean HgbA1c level in this subgroup was 9.2% ± 0.5%, higher than the overall cohort of patients. These findings suggest there may be a role for selective dilation in patients with borderline findings and systemic risk factors for more severe DR. Indeed, our multivariable regression analyses indicated associations between referral-warranted DR and the following patient factors: male sex, uninsured status, Latino race, longer duration of diabetes, systemic hypertension, HgbA1c greater than 9%, and nephropathy with albuminuria. The presence of these factors may help identify patients who could benefit from more frequent photographic or in-clinic examination.

A limitation of the present study is the comparison of nFP and DFE in only a subset of patients, but this realistically represents the resource-limited environment of many safety-net systems. As such, we do not have DFE data for patients not subsequently seen in clinic.

It is possible that some patients who were referred for ophthalmic follow-up encountered barriers to care. We found that DR screening rates increased from around 40% to more than 60% over the course of implementing this telemedicine system (unpublished data; Tyler Aguinaldo, MD). Although loss to follow-up of referral from telemedicine to in-clinic examination in the present study was only 4%, we are currently assessing factors that may present additional barriers to care that were not addressed by our tele-medicine screening program.

Similar to other studies without stereoscopic photography, there was difficulty in accurately identifying DME without the presence of lipid exudates.40 Potential modifications to our screening protocol, with the goal of distinguishing patients who may have sight-threatening diabetic retinopathy, may include stereoscopic image capture,24,41 monochromatic red-free image capture,20,42 and the capture of additional fields.43 A recent study by the INSIGHT network found that fundus photography screening for DR may detect other ocular conditions,44 and we have now incorporated referral for these findings into our clinical protocol.

Employing a primary care-based nFP, the present study demonstrated the utility of telemedicine to screen for referral-warranted DR. The present study demonstrated good correlation with findings on standard clinical ophthalmic examination. These results demonstrate a use-case of telemedicine to screen a large number of patients in the safety-net setting, but they also highlight the need for adequate specialty resources to care for referred patients.

References

  1. Control CfD. National Diabetes Statistics Report: Estimates of Diabetes and its Burden in the United States, 2014. Atlanta, GA: US Department of Health and Human Services; 2014.
  2. Whiting DR, Guariguata L, Weil C, Shaw J. IDF diabetes atlas: global estimates of the prevalence of diabetes for 2011 and 2030. Diabetes Res Clin Pract. 2011;94(3):311–321. doi:10.1016/j.diabres.2011.10.029 [CrossRef]
  3. Zhang X, Saaddine JB, Chou CF, et al. Prevalence of diabetic retinopathy in the United States, 2005–2008. JAMA. 2010;304(6):649–656. doi:10.1001/jama.2010.1111 [CrossRef]
  4. No authors listed. Early photocoagulation for diabetic retinopathy. ETDRS report number 9. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(5 Suppl):766–785.
  5. Chew EY, Ferris FL 3rd, Csaky KG, et al. The long-term effects of laser photocoagulation treatment in patients with diabetic retinopathy: the early treatment diabetic retinopathy follow-up study. Ophthalmology. 2003;110(9):1683–1689. doi:10.1016/S0161-6420(03)00579-7 [CrossRef]
  6. Brechner RJ, Cowie CC, Howie LJ, Herman WH, Will JC, Harris MI. Ophthalmic examination among adults with diagnosed diabetes mellitus. JAMA. 1993;270(14):1714–1718. doi:10.1001/jama.1993.03510140074032 [CrossRef]
  7. Lee PP, Feldman ZW, Ostermann J, Brown DS, Sloan FA. Longitudinal rates of annual eye examinations of persons with diabetes and chronic eye diseases. Ophthalmology. 2003;110(10):1952–1959. doi:10.1016/S0161-6420(03)00817-0 [CrossRef]
  8. Wylie-Rosett J, Basch C, Walker EA, et al. Ophthalmic referral rates for patients with diabetes in primary-care clinics located in disadvantaged urban communities. J Diabetes Complications. 1995;9(1):49–54. doi:10.1016/1056-8727(94)00005-9 [CrossRef]
  9. No authors listed. Grading diabetic retinopathy from stereoscopic color fundus photographs--an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(5 Suppl):786–806. doi:10.1016/S0161-6420(13)38012-9 [CrossRef]
  10. Chabouis A, Berdugo M, Meas T, et al. Benefits of Ophdiat, a telemedical network to screen for diabetic retinopathy: a retrospective study in five reference hospital centres. Diabetes Metab. 2009;35(3):228–232. doi:10.1016/j.diabet.2008.12.001 [CrossRef]
  11. Conlin PR, Fisch BM, Orcutt JC, Hetrick BJ, Darkins AW. Framework for a national teleretinal imaging program to screen for diabetic retinopathy in Veterans Health Administration patients. J Rehabil Res Dev. 2006;43(6):741–748. doi:10.1682/JRRD.2005.08.0146 [CrossRef]
  12. Cuadros J, Bresnick G. EyePACS: an adaptable telemedicine system for diabetic retinopathy screening. J Diabetes Sci Technol. 2009;3(3):509–516. doi:10.1177/193229680900300315 [CrossRef]
  13. Ogunyemi O, George S, Patty L, Teklehaimanot S, Baker R. Teleretinal screening for diabetic retinopathy in six Los Angeles urban safety-net clinics: final study results. AMIA Annu Symp Proc. 2013;2013:1082–1088.
  14. Ogunyemi O, Terrien E, Eccles A, et al. Teleretinal screening for diabetic retinopathy in six Los Angeles urban safety-net clinics: initial findings. AMIA Annu Symp Proc. 2011;2011:1027–1035.
  15. Zimmer-Galler I, Zeimer R. Results of implementation of the DigiScope for diabetic retinopathy assessment in the primary care environment. Telemed J E Health. 2006;12(2):89–98. doi:10.1089/tmj.2006.12.89 [CrossRef]
  16. Day S, Ko M. Diabetic retinopathy in a multi-ethnic population in Santa Clara County, California. ARVO Annual Meeting Abstracts 2012. Invest Ophthalmol Vis Sci. 2012;53:987.
  17. Wilkinson CP, Ferris FL 3rd, Klein RE, et al. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology. 2003;110(9):1677–1682. doi:10.1016/S0161-6420(03)00475-5 [CrossRef]
  18. Wu L, Fernandez-Loaiza P, Sauma J, Hernandez-Bogantes E, Masis M. Classification of diabetic retinopathy and diabetic macular edema. World J Diabetes. 2013;4(6):290–294. doi:10.4239/wjd.v4.i6.290 [CrossRef]
  19. DDSG. Digital Diabetic Study protocol: diabetic retinopathy screening with digital imaging — a pilot program. Kaiser Permanente Northern California, 1996.
  20. Lin DY, Blumenkranz MS, Brothers RJ, Grosvenor DM. The sensitivity and specificity of single-field nonmydriatic monochromatic digital fundus photography with remote image interpretation for diabetic retinopathy screening: a comparison with ophthalmoscopy and standardized mydriatic color photography. Am J Ophthalmol. 2002;134(2):204–213. doi:10.1016/S0002-9394(02)01522-2 [CrossRef]
  21. Daskivich LP, Mangione CM. The promise of primary care-based screening for diabetic retinopathy: the devil will be in the details. Arch Intern Med. 2012;172(21):1678–1680. doi:10.1001/2013.jamainternmed.406 [CrossRef]
  22. Ogunyemi O, George S, Patty L, Teklehaimanot S, Baker R. Teleretinal screening for diabetic retinopathy in six Los Angeles urban safety-net clinics: final study results. AMIA Annu Symp Proc. 2013:1082–1088. eCollection 2013.
  23. Mizrachi Y, Knyazer B, Guigui S, et al. Evaluation of diabetic retinopathy screening using a non-mydriatic retinal digital camera in primary care settings in south Israel. Int Ophthalmol. 2014;34(4):831–837. doi:10.1007/s10792-013-9887-3 [CrossRef]
  24. Ahmed J, Ward TP, Bursell SE, Aiello LM, Cavallerano JD, Vigersky RA. The sensitivity and specificity of nonmydriatic digital stereoscopic retinal imaging in detecting diabetic retinopathy. Diabetes Care. 2006;29(10):2205–2209. doi:10.2337/dc06-0295 [CrossRef]
  25. Peng J, Zou H, Wang W, et al. Implementation and first-year screening results of an ocular telehealth system for diabetic retinopathy in China. BMC Health Serv Res. 2011;11:250. doi:10.1186/1472-6963-11-250 [CrossRef]
  26. Liesenfeld B, Kohner E, Piehlmeier W, et al. A telemedical approach to the screening of diabetic retinopathy: digital fundus photography. Diabetes Care. 2000;23(3):345–348. doi:10.2337/diacare.23.3.345 [CrossRef]
  27. Lopez-Bastida J, Cabrera-Lopez F, Serrano-Aguilar P. Sensitivity and specificity of digital retinal imaging for screening diabetic retinopathy. Diabet Med. 2007;24(4):403–407. doi:10.1111/j.1464-5491.2007.02074.x [CrossRef]
  28. Scanlon PH, Foy C, Malhotra R, Aldington SJ. The influence of age, duration of diabetes, cataract, and pupil size on image quality in digital photographic retinal screening. Diabetes Care. 2005;28(10):2448–2453. doi:10.2337/diacare.28.10.2448 [CrossRef]
  29. Murgatroyd H, Ellingford A, Cox A, et al. Effect of mydriasis and different field strategies on digital image screening of diabetic eye disease. Br J Ophthalmol. 2004;88(7):920–924. doi:10.1136/bjo.2003.026385 [CrossRef]
  30. Garg S, Jani PD, Kshirsagar AV, King B, Chaum E. Telemedicine and retinal imaging for improving diabetic retinopathy evaluation. Arch Intern Med. 2012;172(21):1677–1678. doi:10.1001/archinternmed.2012.4372 [CrossRef]
  31. Chin EK, Ventura BV, See KY, Seibles J, Park SS. Nonmydriatic fundus photography for teleophthalmology diabetic retinopathy screening in rural and urban clinics. Telemed J E Health. 2014;20(2):102–108. doi:10.1089/tmj.2013.0042 [CrossRef]
  32. Mansberger SL, Gleitsmann K, Gardiner S, et al. Comparing the effectiveness of telemedicine and traditional surveillance in providing diabetic retinopathy screening examinations: a randomized controlled trial. Telemed J E Health. 2013;19(12):942–948. doi:10.1089/tmj.2012.0313 [CrossRef]
  33. McGwin G, Friedman D, Haller J, et al. A multicenter study of diabetes eye screening in community settings in the United States. Association for Research in Vision and Ophthalmology Annual Meeting. ; 2013; Seattle, WA. .
  34. Scanlon PH, Malhotra R, Thomas G, et al. The effectiveness of screening for diabetic retinopathy by digital imaging photography and technician ophthalmoscopy. Diabet Med. 2003;20(6):467–474. doi:10.1046/j.1464-5491.2003.00954.x [CrossRef]
  35. Emanuele N, Klein R, Moritz T, et al. Comparison of dilated fundus examinations with seven-field stereo fundus photographs in the Veterans Affairs Diabetes Trial. J Diabetes Complications. 2009;23(5):323–329. doi:10.1016/j.jdiacomp.2008.02.010 [CrossRef]
  36. Moss SE, Klein R, Kessler SD, Richie KA. Comparison between ophthalmoscopy and fundus photography in determining severity of diabetic retinopathy. Ophthalmology. 1985;92(1):62–67. doi:10.1016/S0161-6420(85)34082-4 [CrossRef]
  37. Choremis J, Chow DR. Use of telemedicine in screening for diabetic retinopathy. Can J Ophthalmol. 2003;38(7):575–579. doi:10.1016/S0008-4182(03)80111-4 [CrossRef]
  38. Massin P, Aubert JP, Erginay A, et al. Screening for diabetic retinopathy: the first telemedical approach in a primary care setting in France. Diabetes Metab. 2004;30(5):451–457. doi:10.1016/S1262-3636(07)70142-8 [CrossRef]
  39. Williams GA, Scott IU, Haller JA, Maguire AM, Marcus D, McDonald HR. Single-field fundus photography for diabetic retinopathy screening: a report by the American Academy of Ophthalmology. Ophthalmology. 2004;111(5):1055–1062. doi:10.1016/j.ophtha.2004.02.004 [CrossRef]
  40. Wang YT, Tadarati M, Wolfson Y, Bressler SB, Bressler NM. Comparison of prevalence of diabetic macular edema based on monocular fundus photography vs optical coherence tomography. JAMA Ophthalmol. 2016;134(2):222–228. doi:10.1001/jamaophthalmol.2015.5332 [CrossRef]
  41. Raman R, Rani PK, Sharma T. The sensitivity and specificity of nonmydriatic digital stereoscopic retinal imaging in detecting diabetic retinopathy: response to Ahmed et al. Diabetes Care. 2007;30(5):e47. doi:10.2337/dc07-0052 [CrossRef]
  42. Delori FC, Gragoudas ES. Examination of the ocular fundus with monochromatic light. Ann Ophthalmol. 1976;8(6):703–709.
  43. Vujosevic S, Benetti E, Massignan F, et al. Screening for diabetic retinopathy: 1 and 3 nonmydriatic 45-degree digital fundus photographs vs 7 standard early treatment diabetic retinopathy study fields. Am J Ophthalmol. 2009;148(1):111–118. doi:10.1016/j.ajo.2009.02.031 [CrossRef]
  44. Owsley C, McGwin G Jr., Lee DJ, et al. Diabetes eye screening in urban settings serving minority populations: detection of diabetic retinopathy and other ocular findings using telemedicine. JAMA Ophthalmol. 2015;133(2):174–181. doi:10.1001/jamaophthalmol.2014.4652 [CrossRef]

Demographic and Clinical Characteristics of Patients Undergoing Non-Mydriatic Fundus Camera Screening for Diabetic Retionpathy

Patient Characteristics (n = 6,911)
Male, n (%)3,217 (47)

Age, mean ± SD55.4 ± 12.2 years

Race, n (%)White1,512 (17)
Latino4,139 (46)
API2,101 (24)
Black407 (5)
Arab/Middle East197 (2)
Indian96 (1)
Other623 (7)

Uninsured, n (%)2787 (40)

Duration of diabetes, mean ± SD9.4 ± 2.8 years

Hemoglobin A1c, mean ± SD8% ± 1.9%
< 6537 (6)
6 – 6.92,422 (28)
7 – 7.92,420 (26)
8 – 8.91,373 (16)
≥ 92,133 (25)

Systemic Hypertension, n (%)3,360 (49)

Albuminuria2,246 (33)

Non-Mydriatic Photo Grading

Non-Mydriatic Photo GradePhotosPatients
No retinopathy6,549 (72%)5,032 (73%)
Minimal retinopathy1,057 (12%)746 (11%)
Referral-warranted retinopathy332 (4%)277 (4%)
Photo unreadable1,187 (13%)856 (12%)
Total9,1256,911

Etiology of Unreadable Photos

Unreadable Photos, n (%)
Cataract162 (53%)
Refractive54 (18%)
PDR19 (6%)
Retinopathy18 (6%)
Glaucoma35 (11%)
Other17 (6%)
Total305

Photo Grade and Corresponding Clinical Grade of DR and DME

Clinical Grade of DR

Non-Mydriatic Photo GradeNo RetinopathyMild NPDRModerate NPDRSevere NPDRPDRTotal
No retinopathy79117110810
Minimal retinopathy72751220161
Referral-warranted retinopathy1315822854192
Photo unreadable21578321122358
Total1,09118512742761,521

Clinical Grade of DME

Non-Mydriatic Photo GradeDME AbsentDME PresentTotal
No retinopathy8055810
Minimal retinopathy14714161
Referral-warranted retinopathy13753190
Photo unreadable33624360
Total1,425961,521

Patient Factors Associated With Referral-Warranted Diabetic Retinopathy

Patient FactorsUnivariate ORMultivariate ORMultivariate P value
SexFemaleRefRef
Male1.051.43.03

Age, years1.000.98.03

RaceCaucasianRefRef
Latino1.211.29.02
API0.881.03.28
Black0.631.02.68
Arab/Middle East1.481.04.53
Indian1.911.24.08
Other0.961.02.57

Uninsured1.401.59.004

Duration of diabetes, years1.081.07< .001

Hemoglobin A1c
<7%0.550.95.003
7–9%RefRef
>9%1.541.04.04

Systemic Hypertension1.761.57.005

Albuminuria4.113.30< .001
Authors

From Byers Eye Institute, Stanford University, Palo Alto, CA (BCT, JE); the Department of Ophthalmology, Santa Clara Valley Medical Center, San Jose, CA (BCT, JE); and the Department of Medicine, Santa Clara Valley Medical Center, San Jose, CA (BCT, TA)

Presented at the Association for Research in Vision and Ophthalmology in Seattle, WA, in May 2013.

The authors report no relevant financial disclosures.

Address correspondence to Brian C. Toy, MD, Byers Eye Institute, Stanford University, 2452 Watson Court, Palo Alto, CA 94303; email: briantoy.stanford@gmail.com.

Received: January 23, 2016
Accepted: May 06, 2016

10.3928/23258160-20160707-05

Sign up to receive

Journal E-contents