Retinopathy Screening in Primary Care

Rationale for screening

Ideally, all patients with diabetes would receive their annual eye examination from an ophthalmologist with expertise in diabetic eye disease. For various reasons, patients may not have access to such an experienced ophthalmologist. In remote or medically underserved areas, the eye exam may be performed in the primary care office by a non-ophthalmologist, possibly with supervision from a centralized retina reading center. While such screenings lack important components of a comprehensive eye exam, they may permit more timely treatment of sight-threatening disease than would occur in the absence of any examination.[1,2]

This article reviews 3 methods of retinopathy screening proposed for US primary care and other non-ophthalmology settings: direct ophthalmoscopy, stereo fundus photography, and telescreening. Most of these techniques require instillation of mydriatic (dilating) eye drops to maximize the area of the retina available for examination and to maximize the sensitivity of the exam. Some screening methods do not require the use of dilating eye drops; these methods are described as nonmydriatic. The sensitivity and specificity of screening devices and methods vary, so these techniques are not necessarily interchangeable.

Direct ophthalmoscopy

The direct ophthalmoscope (Figure 1) is inexpensive, widely available, and portable, but it has a number of significant limitations as a screening tool for diabetic eye disease. Because it lacks stereo viewing capability, it may not detect diabetic macular edema (DME). By contrast, the biomicroscope typically used in an ophthalmology office has stereo viewing capability (Figure 2). For this and other reasons, the biomicroscope, when used in conjunction with a high-magnification contact lens, is considered the standard diagnostic instrumentation for diabetic retinopathy.[3,4] A further significant limitation of the direct ophthalmoscope is the wide variation in the sensitivity and specificity of this device to detect sight-threatening diabetic eye disease.[5] For example, remarkably high degrees of sensitivity and specificity relative to ophthalmologists using standard instrumentation were attained in one study of well-trained diabetologists (eg, endocrinology subspecialists) and community screeners (diabetologists: 82.5% sensitivity, 98% specificity; screeners: 83.3% sensitivity, 96.8% specificity).[6] By contrast, other studies have shown that non-ophthalmologists using direct ophthalmoscopes may miss around half of cases of diabetic retinopathy.[7,8] A study which compared non-ophthalmologists using direct ophthalmoscopy and ophthalmologists using indirect ophthalmoscopy against retina specialists using stereo fundus photographs found that the non-ophthalmologists could detect proliferative diabetic retinopathy (PDR) with 49% sensitivity and 84% specificity, while the ophthalmologists using indirect ophthalmoscopy (Figure 3) attained 96% sensitivity and 93% specificity.[7] A more recent study found that the sensitivity and specificity of direct ophthalmoscopy (relative to 7-field fundus photographs) to detect any degree of retinopathy was 46% and 93%, respectively, even though the exams had been performed by skilled ophthalmologists.[8] Nevertheless, in healthcare settings with limited resources, direct ophthalmoscopy through dilated pupils is accepted as part of the minimum standard of care proposed by the IDF for type 2 diabetes.[9]

Figure 1. Direct ophthalmoscopy is convenient but less sensitive than biomicroscopy.

Figure 2. Slit-lamp biomicroscopy is a standard ophthalmic diagnostic technique.

Figure 3. Indirect ophthalmoscopy is used by some ophthalmologists to view the peripheral retina.

Stereo fundus photography

In the US, stereo fundus photography is used to document disease severity and response to treatment, often in the context of a controlled clinical trial, less frequently in routine clinical management. In the UK, more extensive use is made of fundus photography for retinopathy screening.[10] Evidence indicates that the higher sensitivity and opportunities for quality assurance make photographs a better option for screening than direct ophthalmoscopy, as noted by the IDF.[9] Fundus photographs may also be used to detect ocular pathology other than diabetic retinopathy.[11]

Both trained photographers and trained readers are required to make effective use of photographic techniques. The gold standard in fundus photography procedures was established by the University of Wisconsin-Madison Department of Ophthalmology and Visual Sciences Fundus Photograph Reading Center using analog (eg, film) cameras for the Early Treatment of Diabetic Retinopathy Study (ETDRS). The Fundus Photograph Reading Center continues to provide training materials and photographer certification services for 7-field standard stereo color fundus photography.[12] The portions of the retina included in the 7 standard fields are shown in Figure 4. Other photographic protocols (and screening devices) may use fewer fields. Single-field screening by a trained reader compares favorably to 7-field stereo photographs, with a sensitivity of 61-90% and a specificity of 85-97%.[13]

Figure 4. Locations of the 7 standard ETDRS fields on the right and left eyes, respectively.

Image credit: Modified 7 Standard Fields and Fundus Reflex for Color Fundus Photography Tutorial, Fundus Photograph Reading Center. Available at: http://eyephoto.ophth.wisc.edu/photography/tutorial.

Since the ETDRS was conducted, digital cameras have become increasingly available. Digital images offer numerous technical advantages over analog images with respect to image generation and storage, but do not necessarily have the resolution and other qualities needed for accurate retinopathy screening or documentation. For example, one study comparing digital to analog photographs found that only 24% of the digital images were considered of good quality, while 93% of the analog images met this criterion.[14] Thus, digital photography cannot be substituted for analog photography without proper validation. The Fundus Photograph Reading Center offers certification for 6 different digital camera systems.[15]

Dilation of the pupil improves the technical and clinical utility of digital screening images. In particular, photographic image quality may improve markedly with dilation. A study of 300 eyes of 150 patients with diabetes conducted in France showed that dilation increased the number of eyes with good images for all fields studied from 7 to 160, and decreased the number of ungradeable eyes from 127 to 15.[16] This study found that dilation also increased the certitude of endocrinologists in diagnosing and grading diabetic retinopathy.[16]

Telescreening

The term telescreening may be applied to screening techniques in which clinical images are obtained in one physical location but interpreted in a different physical location. These methods range from mailing packages of photographs to an expert reader in a distant city to real-time electronic transmission of digital images to a remote reader.[17] Clinical and technical standards for telescreening procedures were established in 2004.[2] Some vendors of screening instruments have arranged access to retinopathy experts for use in conjunction with their products.[18,19] These systems use a proprietary automated camera to take digital images of various retinal fields. The field images are sent to the reading center for review by trained technicians. A supervising ophthalmologist reviews all abnormal images. The results are returned to the referring physician within one or two days. Improvements in patient care, as measured by increased rates of surveillance and treatment of diabetic retinopathy, have been reported with such systems in the primary care setting,[1] but US reimbursement policies for teleophthalmology procedures vary between insurers. For example, Aetna qualifies only two specific screening cameras for reimbursement, and restricts reimbursement to the initial screening, excluding payment for follow up of previously diagnosed retinopathy.[20] By comparison, Blue Cross Blue Shield lists systems from 5 manufacturers in its reimbursement policy, and will pay for initial and follow up screening according to the American Diabetes Association's recommendations.[21]

As is true of fundus photography, image quality may significantly limit the utility of teleophthalmology. In one study of real-time teleophthalmology using a direct ophthalmoscope equipped with a digital camera, poor image quality prevented classification of 36% of eyes of patients with diabetes.[22] Thus, formal quality assurance procedures are a prominent component of telescreening programs.[17,23,24]

Future Developments in Diabetic Retinopathy Screening

The rapidly increasing numbers of patients diagnosed with diabetes has significant implications for the ophthalmology workload.[25] Therefore, automated screening methods are being investigated to identify the patients most likely to require treatment. Both DME and NPDR have distinctive diagnostic features that lend themselves to automated detection: microaneurysms in NPDR, and altered retinal profile and thickness in DME.[26]

Automated screening methods depend to one extent or another upon imaging techniques. Consequently, image quality control is critical to the practicality of automated screening. A Danish study of 165 eyes of 83 patients found that pupil dilation improved the sensitivity of automated analysis of digital imaging from 89.9% to 97%.[27] Disqualifying poor quality images prior to grading can be used in conjunction with automated microaneurysm detection algorithms to improve test sensitivity. One very large study, using a training set of 1,067 images on 14,406 clinical images from 6,722 patients attained a sensitivity of 90.5% for detecting any retinopathy and 97.9% for detecting retinopathy requiring treatment.[28]

Optical coherence tomography (OCT), an instrument developed since the ETDRS, offers a sensitive, rapid, objective, and reproducible method to detect CSME.[29] One study comparing automated OCT to clinical examination as a screening technique for CSME yielded a sensitivity of 89% and a specificity of 86%.[30]

References

1. Wilson C, Horton M, Cavallerano J, Aiello LM. Addition of primary care-based retinal imaging technology to an existing eye care professional referral program increased the rate of surveillance and treatment of diabetic retinopathy. Diabetes Care. 2005;28:318-322.

2. Cavallerano J, Lawrence MG, Zimmer-Galler I, et al; American Telemedicine Association, Ocular Telehealth Special Interest Group; National Institute of Standards and Technology Working Group. Telehealth practice recommendations for diabetic retinopathy. Telemed J E Health. 2004;10:469-482.

3. American Academy of Ophthalmology. Preferred practice pattern: diabetic retinopathy. San Francisco, Calif: American Academy of Ophthalmology; 2003.

4. Early Treatment Diabetic Retinopathy Study research group. Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1. Arch Ophthalmol. 1985;103:1796-1806..

5. Hutchinson A, McIntosh A, Peters J, et al. Effectiveness of screening and monitoring tests for diabetic retinopathy: a systematic review. Diabet Med. 2000;17:495-506.

6. Pandit RJ, Taylor R. Quality assurance in screening for sight-threatening diabetic retinopathy. Diabet Med. 2002;19:285-291.

7. Sussman EJ, Tsiaras WG, Soper KA. Diagnosis of diabetic eye disease. JAMA. 1982;247:3231-3234.

8. Emanuele N, Klein R, Moritz T, et al. Comparison of dilated fundus examination by ophthalmologists with 7-field stero fundus photographs in the Veterans Affairs Diabetes Trial (VADT). American Diabetes Association 66th Annual Scientific Sessions. Washington, DC, June 9-13, 2006. Abstract 2215-PO.

9. IDF Clinical Guidelines Task Force. Global guideline for type 2 diabetes. Brussels: International Diabetes Federation, 2005. Available at: http://www.idf.org/webdata/docs/IDF%20GGT2D.pdf. Accessed September 13, 2005.

10. Harding S, Greenwood R, Aldington S, et al; Diabetic Retinopathy Grading and Disease Management Working Party. Grading and disease management in national screening for diabetic retinopathy in England and Wales. Diabet Med. 2003;20:965-971.

11. Barahimi B, Rao US, Recchia C, et al. Analysis of ophthalmic pathology other than diabetic retinopathy diagnosed by photographic screening of diabetic patients. Association for Research in Vision and Ophthalmology 2006 Annual Meeting. Ft Lauderdale, Fla, April 30-May 4, 2006. Abstract 3868/B908.

12. Modified 7-Standard Field Color Fundus Photography (7M-F) & Film Fluorescein Angiography (FA-F). December 30, 2004. Madison, Wisc: Fundus Photograph Reading Center, Department of Ophthalmology and Visual Sciences, University of Wisconsin-Madison. Available at: http://eyephoto.ophth.wisc.edu/Photography/Protocols/Mod7&FA-ver1.pdf. Accessed August 7, 2006.

13. 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;111:1055-1062.

14. Yogesan K, Constable IJ, Barry CJ, et al. Telemedicine screening of diabetic retinopathy using a hand-held fundus camera. Telemed J. 2000;6:219-223.

15. Fundus Photograph Reading Center. Digital System Evaluation. Available at: http://eyephoto.ophth.wisc.edu/DSES.html. Accessed August 7, 2006.

16. Deb-Joardar N, Germain N, Thuret G, et al. Screening for diabetic retinopathy by ophthalmologists and endocrinologists with pupillary dilation and a nonmydriatic digital camera. Am J Ophthalmol. 2005;140:814-821.

17. Cavallerano AA, Cavallerano JD, Katalinic P, et al; Joslin Vision Network Research Team. A telemedicine program for diabetic retinopathy in a Veterans Affairs Medical Center--the Joslin Vision Network Eye Health Care Model. Am J Ophthalmol. 2005;139:597-604.

18. Zeimer R, Zou S, Meeder T, Quinn K, Vitale S. A fundus camera dedicated to the screening of diabetic retinopathy in the primary-care physician's office. Invest Ophthalmol Vis Sci. 2002;43:1581-1587.

19. Fransen SR, Leonard-Martin, TC, Feuer WJ, Hildebrand PL; Inoveon Health Research Group. Ophthalmology. 2002;109:595-601.

20. Aetna. Diabetic Retinopathy Telescreening Systems. Clinical policy bulletin 0563. November 4, 2005. Available at: http://www.aetna.com/cpb/data/CPBA0563.html. Accessed August 7, 2006.

21. Blue Cross Blue Shield of North Carolina. Diabetic Retinopathy Telescreening. Corporate medical policy OTH8045. September 2005. Available at: https://www.bcbsnc.com/services/medical-policy/pdf/diabetic_retinopathy_telescreening.pdf. Accessed August 7, 2006.

22. Marcus DM, Brooks SE, Ulrich LD, et al. Telemedicine diagnosis of eye disorders by direct ophthalmoscopy. A pilot study. Ophthalmology. 1998;105:1907-1914.

23. Arun CS, Young D, Batey D, et al. Establishing ongoing quality assurance in a retinal screening programme. Diabet Med. 2006;23:629-634.

24. Schneider S, Aldington SJ, Kohner EM, et al. Quality assurance for diabetic retinopathy telescreening. Diabet Med. 2005;22:794-802.

25. Scanlon PH, Carter S, Foy C, Ratiram D, Harney B. An evaluation of the change in activity and workload arising from diabetic ophthalmology referrals following the introduction of a community based digital retinal photographic screening programme. Br J Ophthalmol. 2005;89:971-975.

26. Hejlesen O, Ege B, Englmeier KH, et al. TOSCA-Imaging--developing Internet based image processing software for screening and diagnosis of diabetic retinopathy. Medinfo. 2004;11(pt 1):222-226.

27. Hansen AB, Hartvig NV, Jensen MS, et al. Diabetic retinopathy screening using digital non-mydriatic fundus photography and automated image analysis. Acta Ophthalmol Scand. 2004;82:666-672.

28. Philip S, Fleming A, Goatman K, Sharp P, Olson J. Automated image quality assessment is an essential step for automating diabetic retinopathy detection. American Diabetes Association 66th Annual Scientific Sessions. Washington, DC, June 9-13, 2006. Abstract 2220-PO.

29. Duker J, Fujimoto J, Witkin A, Wojtkowski M. OCT3 and beyond: new developments. Managing the open globe calls for creativity and flexibility of surgical approach tailored to the specific case. Rev Ophthalmol. 2005;12(5). Available at: http://www.revophth.com/index.asp?page=1_719.htm. Accessed April 17, 2006.

30. Sadda SR, Tan O, Walsh AC, et al. Automated detection of clinically significant macular edema by grid scanning optical coherence tomography. Ophthalmology. 2006;113:1196.e1-2.