Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Combined Nonmydriatic Spectral-Domain Optical Coherence Tomography and Nonmydriatic Fundus Photography for the Detection of Age-Related Macular Degeneration Changes

Haytham I. Salti, MD; Rafic S. Antonios, MD; Sandra S. Haddad, MD; Rola N. Hamam, MD; Ziad F. Bashshur, MD; Nicola G. Ghazi, MD

Abstract

BACKGROUND AND OBJECTIVE:

Nonmydriatic fundus photography (FP) has been a suboptimal tool for detecting age-related macular degeneration (AMD) changes. This study sought to enhance the detection of AMD changes by combining nonmydriatic FP with nonmydriatic spectral-domain optical coherence tomography (SD-OCT).

PATIENTS AND METHODS:

The study population included 249 patients aged 65 years and older who were assessed for AMD changes using standard mydriatic biomicroscopic fundus examination. Each eye then underwent nonmydriatic FP in one session followed 1 week later with nonmydriatic FP coupled with nonmydriatic SD-OCT. Images were interpreted for detection of AMD changes, and findings were compared to the original mydriatic biomicroscopic examination.

RESULTS:

Nonmydriatic FP had 64% sensitivity, 97% specificity, and a kappa value of 0.67 in detecting AMD changes compared with the traditional mydriatic biomicroscopic examination. Combined nonmydriatic FP and nonmydriatic SD-OCT increased sensitivity to 91.5%, specificity to 98.6%, and kappa to 0.91.

CONCLUSION:

The addition of nonmydriatic SD-OCT to nonmydriatic FP enhances the detection of AMD changes.

[Ophthalmic Surg Lasers Imaging Retina. 2015;46:531–537.]

From the Department of Ophthalmology, American University of Beirut Medical Center, Beirut, Lebanon (HIS, RSA, RNH, ZFB); the Beirut Central Military Hospital, Beirut, Lebanon (SSH); and the King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia (NGG).

The authors report no relevant financial disclosures.

Address correspondence to Haytham I. Salti, MD, Department of Ophthalmology, American University of Beirut, Medical Center, PO Box 110236, Riad El Solh, Beirut 1107-2020 Lebanon; email: hs06@aub.edu.lb.

Received: December 03, 2014
Accepted: March 04, 2015

Abstract

BACKGROUND AND OBJECTIVE:

Nonmydriatic fundus photography (FP) has been a suboptimal tool for detecting age-related macular degeneration (AMD) changes. This study sought to enhance the detection of AMD changes by combining nonmydriatic FP with nonmydriatic spectral-domain optical coherence tomography (SD-OCT).

PATIENTS AND METHODS:

The study population included 249 patients aged 65 years and older who were assessed for AMD changes using standard mydriatic biomicroscopic fundus examination. Each eye then underwent nonmydriatic FP in one session followed 1 week later with nonmydriatic FP coupled with nonmydriatic SD-OCT. Images were interpreted for detection of AMD changes, and findings were compared to the original mydriatic biomicroscopic examination.

RESULTS:

Nonmydriatic FP had 64% sensitivity, 97% specificity, and a kappa value of 0.67 in detecting AMD changes compared with the traditional mydriatic biomicroscopic examination. Combined nonmydriatic FP and nonmydriatic SD-OCT increased sensitivity to 91.5%, specificity to 98.6%, and kappa to 0.91.

CONCLUSION:

The addition of nonmydriatic SD-OCT to nonmydriatic FP enhances the detection of AMD changes.

[Ophthalmic Surg Lasers Imaging Retina. 2015;46:531–537.]

From the Department of Ophthalmology, American University of Beirut Medical Center, Beirut, Lebanon (HIS, RSA, RNH, ZFB); the Beirut Central Military Hospital, Beirut, Lebanon (SSH); and the King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia (NGG).

The authors report no relevant financial disclosures.

Address correspondence to Haytham I. Salti, MD, Department of Ophthalmology, American University of Beirut, Medical Center, PO Box 110236, Riad El Solh, Beirut 1107-2020 Lebanon; email: hs06@aub.edu.lb.

Received: December 03, 2014
Accepted: March 04, 2015

Introduction

Nonmydriatic fundus photography (FP) has been used in numerous studies to screen for retinal pathologies such as diabetic retinopathy,1 glaucomatous optic nerve head cupping,2 and age-related macular degeneration (AMD).3 The utility of nonmydriatic FP in AMD has been questioned due to the high proportion of images in AMD screening studies judged to be unreadable.4 In addition, false negatives and false positives are seen more frequently in cases of AMD compared with other retinal pathologies.3 This is due primarily to the multiple artifacts and challenges associated with nonmydriatic imaging when dealing with eyes with lenticular opacities and older age-related changes, which can decrease media clarity and make it difficult to find minute retinal changes. All of these factors have limited the role of nonmydriatic FP in screening programs for AMD.

The use of optical coherence tomography (OCT) has been gaining momentum and popularity as an essential tool for retinal disease evaluation. Since its introduction in the 1990s, OCT has revolutionized the diagnosis and the understanding of certain blinding conditions.5,6 In addition, OCT has set a shift in the treatment paradigms of retinal pathologies, particularly AMD,7 Diabetic macular edema,8,9 and retinal vein occlusions.10,11 Although early generation OCT had a modest capability in detecting relatively small lesions, the more recent and higher density spectral-domain (SD) OCT acquisition protocols have led to a more detailed and quick capture of the macular area, discerning changes at a microscopic level.12 Although most OCT studies have been performed on patients with dilated pupils, a few recent studies that have explored the use of OCT in a nonmydriatic setting have yielded promising results.13,14

We propose complementing nonmydriatic FP with nonmydriatic SD-OCT to overcome the hurdles described previously and to improve on the readability, sensitivity, and specificity in detecting AMD changes in the retina in the nonmydriatic setting. This study compares our two-step technique to the gold standard of traditional mydriatic biomicroscopy and to nonmydriatic FP imaging alone. We also performed the same comparison for each of the AMD classic changes to evaluate our newly proposed combined technique with greater precision.

Patients and Methods

From June 2011 to January 2013, 491 consecutive fundi of 249 patients presenting to the Beirut Central Military Hospital ophthalmology service were examined using three different methods of evaluation to detect changes of AMD. Prior to commencement, the study was granted approval from the Institutional Review Board.

Fundi were labeled as having AMD if any of the retinal changes defined by the Age-Related Eye Disease Study Research Group were verified.15 Briefly, a diagnosis of AMD was made if any one or more of the following changes were found on fundus examination: definite soft distinct or indistinct drusen with sizes ranging from at least 63 µm to more than 250 µm; the presence of retinal pigment abnormalities including hyperpigmentation or depigmentation; central or noncentral geographic atrophy; signs of choroidal neovascularization causing retinal pigment epithelium (RPE) or associated neurosensory detachment; subretinal hemorrhage; and features of photocoagulation or scarring due to AMD. Inclusion criteria were patients age 65 and older, and signed informed consent. Exclusion criteria were patients with overt corneal opacities, patients using pupillary-constricting agents that would hinder proper nonmydriatic evaluation, and patients with extremely poor visual acuity with an inability to focus at a target for nonmydriatic fundus imaging.

All three methods of evaluation were interpreted by the same subspecialist (HIS). The first method consisted of a traditional pharmacologic dilated fundus examination using slit-lamp biomicroscopy and a Goldmann contact lens. Dilation was achieved using tropicamide 1% eye drops and phenylephrine 10% eyed drops instilled once in each eye. This first evaluation was performed by a retina subspecialist (HIS) 30 minutes after instillation of the eye drops. Patients’ demographic data and name were masked to the examiner.

Patients underwent a second method of evaluation that consisted of a 45° nonmydriatic fundus color image centered at the macula and taken after dark adaptation for 15 minutes. Acquisition was performed by a comprehensive ophthalmologist (SH) 1 week after the initial interpretation using a Kowa fundus camera (VX-10i, Kowa, Nagoya, Japan) set to the nonmydriatic mode.

The third method of evaluation consisted of non-mydriatic FP of the fovea (using the same camera) followed by nonmydriatic SD-OCT using Cirrus HD-OCT (Carl Zeiss, Dublin, California). The SD-OCT imaging was performed using a macular cube acquisition protocol with a scan pattern of 6 × 6–mm data cube (512 × 128 scan) consisting of a scanning density of 27,000 A-scans per second. All line scans of each macular study were reviewed for errors in automated retinal boundary detection, and those with errors were corrected manually. Next, the center of the fovea was localized, and changes in the area suggestive of AMD were noted.

Both imaging techniques had a dark adaptation period of 15 minutes preceding each to ensure some physiological pupillary dilatation. Acquisition also was performed approximately 1 week later by the same comprehensive ophthalmologist. For each fundus, both an nonmydriatic FP and a nonmydriatic SD-OCT were saved in an electronic file. These files were sent to a reading center with the demographic data and names of the patients masked, and the files were read by the same subspecialist. Fundus images were read using the IMAGEnet Lite viewer (Topcon Medical Systems, Paramus, New Jersey). These were displayed on a LeNovo ThinkVision L1711p monitor (Lenovo Group, Beijing, China) with a screen resolution of 1,280 × 1,024 pixels. At least 1 month separated the interpretation of each examination method to minimize the reader’s bias. Readability of fundus photographs and OCT images were assessed prior to performing statistical analysis. In addition to the detection of AMD, each specific change also was viewed and compared to traditional biomicroscopy.

SPSS version 20.0 (SPSS, Chicago, Illinois) was used for statistical analyses. Descriptive statistics were reported as number and percent for categorical variables. Sensitivity, specificity, and positive and negative predictive values of each technique were calculated for detection of AMD and for detection of each of the cardinal signs of AMD, namely, drusen, RPE changes, geographic atrophy, RPE and neurosensory detachment, periretinal hemorrhage, and fibrous scarring. Interobserver agreement (kappa) also was calculated to evaluate the agreement between the mydriatic examination and the nonmydriatic FP evaluation method, and between the mydriatic examination and the nonmydriatic FP coupled with the nonmydriatic SD-OCT evaluation method.

Results

Demographic characteristics and general medical status for the 249 patients are summarized in Table 1. Using the traditional mydriatic biomicroscopic evaluation, AMD changes were detected in 35% of the fundi evaluated. The different changes seen in our patients are summarized in Table 2. Briefly, the vast majority of patients had drusen and RPE changes (81.4% and 80.8%, respectively). End-stage changes, geographic atrophy and macular scarring, were found in 13.9% and 10.5% of patients, respectively, and a slightly smaller percentage had evidence of neovascular disease activity; 8.72% of patients had RPE or neurosensory detachment, and 9.3% of patients had some evidence of bleeding.

Demographic Characteristics and Medical Status of Study Population

Table 1:

Demographic Characteristics and Medical Status of Study Population

AMD-related Changes Seen in Study Population on Mydriatic Biomicroscopic Examination

Table 2:

AMD-related Changes Seen in Study Population on Mydriatic Biomicroscopic Examination

Using nonmydriatic imaging alone, 389 of the fundi were judged as readable (79%); 102 fundus images could not be interpreted. A comprehensive eye examination was performed to look into the possible reasons for the inability to interpret these images; Table 3 lists these potential hindering factors. For the 389 fundi with readable nonmydriatic images, AMD changes were detected in only 78 of the images or 20% of the fundi evaluated. Comparison was made to the initial technique using only the data of the same 389 fundi for both groups. Sensitivity and specificity were 64% and 97%, respectively. Positive predictive and negative predictive values were 89% and 87.4%, respectively. The kappa value was 0.67. Evaluating with greater detail each change using nonmydriatic FP compared to a mydriatic examination, the sensitivity for drusen, RPE changes, geographic atrophy, RPE or neurosensory detachment, periretinal hemorrhage, and fibrous scarring were 81.69%, 60.32%, 66.67%, 50%, 66.67%, and 62.5%, respectively, whereas the specificity for these changes was 98.12, 91.41, 99.47, 99.22, 99.48, and 99.48, respectively. The kappa value for each of these changes was 0.83, 0.508, 0.719, 0.438, 0.661, and 0.66, respectively.

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Alone

Table 3:

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Alone

For the third technique, nonmydriatic SD-OCT, 431 images were judged as readable (87.8%). Fundi that could not be interpreted again underwent a comprehensive eye examination to look into the possible causes of difficulty; Table 4 lists potentially limiting factors. In the 431 fundi that could be interpreted, age-related changes were detected in 31%. Comparison was made to the traditional examination with the same 431 fundi. The sensitivity and specificity were 91.5% and 98.6%, respectively. Positive and negative predictive values were 97% and 96%, respectively, and the kappa value was 0.91. The sensitivity for detecting specific changes such as drusen, RPE changes, geographic atrophy, RPE and neurosensory detachment, periretinal hemorrhage, and fibrous scarring was 89.55%, 89.74%, 92.31%, 90.91%, 69.23%, and 92.86%, respectively. The specificity for these changes was 97.64, 97.17, 99.04, 99.76, 99.04, and 99.52%, respectively. The kappa value for each of these changes was 0.88, 0.86, 0.82, 0.91, 0.68, and 0.89, respectively. Table 5 summarizes all the findings for the two techniques.

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Combined With Optical Coherence Tomography

Table 4:

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Combined With Optical Coherence Tomography

Sensitivity, Specificity, Positive and Negative Predictive Value, and Kappa for Detecting Different Levels of Age-related Macular Degeneration

Table 5:

Sensitivity, Specificity, Positive and Negative Predictive Value, and Kappa for Detecting Different Levels of Age-related Macular Degeneration

Discussion

Screening for AMD is important because identifying the disease in its early stages of treatable lesions yields a better visual outcome, a lesser treatment burden,16 and a better quality of life.17 In addition, as the number of older adults continues to increase, the prevalence of AMD also is expected to increase, and the number of retina specialist is not expected to match this rise. Therefore, the need for a sound cost-effective screening program is paramount. The recent technologic revolution in telemedicine has enticed investigators to take advantage of the breakthroughs in digital imaging to answer this need. Initial endeavors in telemedicine screening for AMD were reported by Klein et al,3 who compared both mydriatic and nonmydriatic digital imaging in various degrees of AMD to 35-mm film and demonstrated the superiority of the former compared to the latter. The same conclusion was reached by Lim et al18 in a similar study.

Clearly, digital dilated fundus photography would be an ideal tool for AMD screening. However, concerns about precipitating an acute glaucoma in an older patient19 refuted this technique of screening. Despite its inferiority to dilated imaging, nonmydriatic FP seemed like a safer telemedicine alternative, and many authors have advocated its use for screening.3,20–22

However, limitations of this technique have prompted researchers to search for further cost-effective steps with better results. Pupillary shadow, especially among older patients, poses a challenge that has yet to be overcome. In our series, with numbers similar to other recent published work in fundus photography alone, as shown in Table 6, we believe a good number of the unreadable images were too dim for interpretation because of dark pupils, pinpoint pupils, and poorly responsive pupils to mesopic conditions. Silva et al23 proposed a recent technology-related enhancement tool, the low flash image capture, to limit pupil constriction, especially in second eye imaging. We also have used different image manipulations to enhance image quality24 and detection of disease. In this series, adding the nonmydriatic SD-OCT improved readability, primarily in patients with small pupils and dark pupils, as shown in Table 4.

Published Readability of NMFP in AMD Screening Studies

Table 6:

Published Readability of NMFP in AMD Screening Studies

Pirbhai et al25 have reported findings missed by single-field fundus photography in AMD screening studies. Although futile at times, some of these changes could have some significance in terms of missing potentially treatable lesions. To avoid such “misses,” we added nonmydriatic SD-OCT. This improved readability from 79% to 87.8%. Sensitivity increased whether looking at the disease in general or looking at each specific change. Most parameters were improved as shown in Figure 1.

(A) Sensitivity, (B) specificity, and (C) kappa values using nonmydriatic fundus photography (FP) versus nonmydriatic FP combined with nonmydriatic OCT in the detection of drusen, retinal pigment epithelium (RPE) changes, geographic atrophy, RPE and neurosensory detachment, fibrous scarring, and periretinal hemorrhage.

Figure 1.

(A) Sensitivity, (B) specificity, and (C) kappa values using nonmydriatic fundus photography (FP) versus nonmydriatic FP combined with nonmydriatic OCT in the detection of drusen, retinal pigment epithelium (RPE) changes, geographic atrophy, RPE and neurosensory detachment, fibrous scarring, and periretinal hemorrhage.

Any change not detected by nonmydriatic FP was likely to be seen on nonmydriatic SD-OCT. In fact, the biggest enhancement seen was one of changes involving fluid or volume. Focal subretinal fluid from a pigment epithelial detachment and deep choroidal abnormalities might have been overlooked with nonmydriatic FP alone, especially with iris shadow; adding nonmydriatic SD-OCT overcomes this. Figure 2 illustrates a fundus image in which AMD changes could not be detected, whereas the nonmydriatic SD-OCT counterpart clearly shows deep retinal changes.

Nonmydriatic fundus photography does not detect drusen changes, primarily because of the iris shadow, whereas nonmydriatic OCT overcomes this shortcoming.

Figure 2.

Nonmydriatic fundus photography does not detect drusen changes, primarily because of the iris shadow, whereas nonmydriatic OCT overcomes this shortcoming.

By the same token, nonmydriatic SD-OCT has had a modest but promising role as a screening tool.13,14 When used alone, nonmydriatic SD-OCT missed small findings such as hard drusen or small intraretinal hemorrhages.13 However, combining nonmydriatic SD-OCT with fundus photography overcomes this minor shortcoming. Figure 3 demonstrates this advantage.

(A) Nonmydriatic fundus photography (FP) shows drusen, small in size, whereas the nonmydriatic OCT appears normal. (B) Nonmydriatic FP shows extrafoveal changes compatible with early age-related macular degeneration, whereas the nonmydriatic OCT does not reach this area.

Figure 3.

(A) Nonmydriatic fundus photography (FP) shows drusen, small in size, whereas the nonmydriatic OCT appears normal. (B) Nonmydriatic FP shows extrafoveal changes compatible with early age-related macular degeneration, whereas the nonmydriatic OCT does not reach this area.

In summary, nonmydriatic FP complemented with nonmydriatic SD-OCT was successful in mimicking a traditional dilated fundus examination when looking for AMD changes in an older population. In addition, manipulating the OCT capture with volumetric measures and adding the five line raster capture could possibly further enhance nonmydriatic SD-OCT in missing fewer changes. Advantages of nonmydriatic imaging whether performed by fundus photography or SD-OCT include the lower cost of image acquisition, patient comfort and safety, and speed of image acquisition, as well as the possibility of reducing the future rising burden on ophthalmologists by having trained paramedical staff perform the imaging.

Our study had a modest number of AMD fundi evaluated. Larger nationwide studies could further fine-tune our combined technique looking at the detection rate of each individual AMD change with either technique.

References

  1. Bursell SE, Cavallerano JD, Cavallerano AA, et al. Joslin Vision Network Research Team. Stereo nonmydriatic digital-video color retinal imaging compared with Early Treatment Diabetic Retinopathy Study seven standard field 35-mm stereo color photos for determining level of diabetic retinopathy. Ophthalmology. 2001;108(3):572–585. doi:10.1016/S0161-6420(00)00604-7 [CrossRef]
  2. Li HK, Tang RA, Oschner K, Koplos C, Grady J, Crump WJ. Tele-medicine screening of glaucoma. Telemed J. 1999;5(3):283–290. doi:10.1089/107830299312032 [CrossRef]
  3. Klein R, Meuer SM, Moss SE, Klein BE, Neider MW, Reinke J. Detection of age-related macular degeneration using a nonmydriatic digital camera and a standard film fundus camera. Arch Ophthalmol. 2004;122(11):1642–1646. doi:10.1001/archopht.122.11.1642 [CrossRef]
  4. Le Tien V, Streho M, d’Athis P, et al. Interobserver and intraobserver reliability of detecting age-related macular degeneration using a nonmydriatic digital camera. Am J Ophthalmol. 2008;146(4):520–526. doi:10.1016/j.ajo.2008.05.031 [CrossRef]
  5. Puliafito CA, Hee MR, Lin CP, et al. Imaging of macular diseases with optical coherence tomography. Ophthalmology. 1995;102(2):217–229. doi:10.1016/S0161-6420(95)31032-9 [CrossRef]
  6. Coker JG, Duker JS. Macular disease and optical coherence tomography. Curr Opin Ophthalmol. 1996;7(3):33–38. doi:10.1097/00055735-199606000-00006 [CrossRef]
  7. Montero JA, Ruiz-Moreno JM, Tavolato M. Follow-up of age-related macular degeneration patients treated by photodynamic therapy with optical coherence tomography 3. Graefes Arch Clin Exp Ophthalmol. 2003;241(10):797–802. doi:10.1007/s00417-003-0752-3 [CrossRef]
  8. Hee MR, Puliafito CA, Duker JS, et al. Topography of diabetic macular edema with optical coherence tomography. Ophthalmology. 1998;105(2):360–370. doi:10.1016/S0161-6420(98)93601-6 [CrossRef]
  9. Hee MR, Puliafito CA, Wong C, et al. Quantitative assessment of macular edema with optical coherence tomography. Arch Ophthalmol. 1995;113(8):1019–1029. doi:10.1001/archopht.1995.01100080071031 [CrossRef]
  10. Imasawa M, Iijima H, Morimoto T. Perimetric sensitivity and retinal thickness in eyes with macular edema resulting from branch retinal vein occlusion. Am J Ophthalmol. 2001;131(1):55–60. doi:10.1016/S0002-9394(00)00722-4 [CrossRef]
  11. Ip M, Kahana A, Altaweel M. Treatment of central retinal vein occlusion with triamcinolone acetonide: an optical coherence tomography study. Semin Ophthalmol. 2003;18(2):67–73. doi:10.1076/soph.18.2.67.15860 [CrossRef]
  12. Forooghian F, Cukras C, Meyerle CB, Chew EY, Wong WT. Evaluation of time domain and spectral domain optical coherence tomography in the measurement of diabetic macular edema. Invest Ophthalmol Vis Sci. 2008;49(10):4290–4296. doi:10.1167/iovs.08-2113 [CrossRef]
  13. Ouyang Y, Heussen FM, Keane PA, Pappuru RK, Sadda SR, Walsh AC. Juxtapapillary pigment epithelium detachment observed in asymptomatic participants using optical coherence tomography. Invest Ophthalmol Vis Sci. 2013;54(2):1144–1149. doi:10.1167/iovs.12-9903 [CrossRef]
  14. Medina FJ, Callen CI, Rebolleda G, Munoz-Negrete FJ, Callen MJ, del Valle FG. Use of nonmydriatic spectral-domain optical coherence tomography for diagnosing diabetic macular edema. Am J Ophthalmol. 2012;153(3):536–543e1. doi:10.1016/j.ajo.2011.08.008 [CrossRef]
  15. Age-Related Eye Disease Study Research Group. The Age-Related Eye Disease Study system for classifying age-related macular degeneration from stereoscopic color fundus photographs: the Age-Related Eye Disease Study Report Number 6. Am J Ophthalmol. 2001;132(5):668–681. doi:10.1016/S0002-9394(01)01218-1 [CrossRef]
  16. El-Mollayes GM, Mahfoud Z, Schakal AR, Salti HI, Jaafar D, Bashshur ZF. Intravitreal bevacizumab in the management of neovascular age-related macular degeneration: effect of baseline visual acuity. Retina. 2013;33(9):1828–1835. doi:10.1097/IAE.0b013e3182877a0d [CrossRef]
  17. Coleman AL, Yu F, Ensrud KE, et al. Impact of age-related macular degeneration on vision-specific quality of life: Follow-up from the 10-year and 15-year visits of the Study of Osteoporotic Fractures. Am J Ophthalmol. 2010;150(5):683–691. doi:10.1016/j.ajo.2010.05.030 [CrossRef]
  18. Lim JI, Labree L, Nichols T, Cardenas I. Comparison of nonmydriatic digitized video fundus images with standard 35-mm slides to screen for and identify specific lesions of age-related macular degeneration. Retina. 2002;22(1):59–64. doi:10.1097/00006982-200202000-00011 [CrossRef]
  19. Hancox J, Murdoch I, Parmar D. Changes in intraocular pressure following diagnostic mydriasis with cyclopentolate 1%. Eye (Lond). 2002;16(5):562–566. doi:10.1038/sj.eye.6700146 [CrossRef]
  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. Zimmer-Galler IE, Zeimer R. Feasibility of screening for high-risk age-related macular degeneration with an Internet-based automated fundus camera. Ophthalmic Surg Lasers Imaging. 2005;36(3):228–236.
  22. De Bats F, Vannier Nitenberg C, Fantino B, Denis P, Kodjikian L. Age-related macular degeneration screening using a nonmydriatic digital color fundus camera and telemedicine. Ophthalmologica. 2013; Dec12 [Epub ahead of print].
  23. Silva PS, Walia S, Cavallerano JD, et al. Comparison of low-light nonmydriatic digital imaging with 35-mm ETDRS seven-standard field stereo color fundus photographs and clinical examination. Telemed J E Health. 2012;18(7):492–499. doi:10.1089/tmj.2011.0232 [CrossRef]
  24. Salti HI, Nasrallah M, Haddad S, Khairallah W, Salti IS. Enhancing nonmydriatic color photographs of the retina with monochromatic views and a stereo pair to detect diabetic retinopathy. Ophthalmic Surg Lasers Imaging. 2009;40(4):373–378. doi:10.3928/15428877-20096030-04 [CrossRef]
  25. Pirbhai A, Sheidow T, Hooper P. Prospective evaluation of digital non-stereo color fundus photography as a screening tool in age-related macular degeneration. Am J Ophthalmol. 2005;139(3):455–461. doi:10.1016/j.ajo.2004.09.077 [CrossRef]

Demographic Characteristics and Medical Status of Study Population

Mean Age, yrs (range)70 (65–93)
Gender (M:F)1.19:1
% smoker26
% known hypertensive44
% known NIDDM29
% CAD27

AMD-related Changes Seen in Study Population on Mydriatic Biomicroscopic Examination

ConditionNumberFrequency (%)
Drusen14081.4
RPE changes13980.8
Geographic atrophy2413.9
RPE and neurosensory detachment158.72
Periretinal hemorrhages169.3
Periretinal fibrous scarring1810.5

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Alone

CauseNumber%
Lenticular opacities7270.6
Dark iris6765.7
Posterior capsule opacification2120.6
Poorly dilating pupil1918.6
Corneal opacity1312.7
Vitreous opacity22

Potential Causes of Poor Visualization of the Fundus by Nonmydriatic Imaging Combined With Optical Coherence Tomography

CauseNumber%
Lenticular opacities4566
Dark iris23.3
Posterior capsule opacification1931
Poorly dilating pupil1016
Corneal opacity1321
Vitreous opacity23.3

Sensitivity, Specificity, Positive and Negative Predictive Value, and Kappa for Detecting Different Levels of Age-related Macular Degeneration

DrusenRPE ChangesGeographic AtrophyRPE & Neurosensory DetachmentPeriretinal HemorrhagesFibrous Scarring
Fundus Photography Alone
Sensitivity83.33%83.33%80.00%80.00%66.67%75.00%
Specificity98.74%98.83%99.74%99.48%100.00%99.74%
Positive Predictive Value93.75%90.91%88.89%66.67%100.00%85.71%
Negative Predictive Value96.31%97.68%99.47%99.74%99.48%99.48%
Kappa0.8580.8520.8380.7230.7980.796

Fundus Photography with OCT
Sensitivity91.60%92.11%90.91%90.91%90.00%100.00%
Specificity96.67%98.59%99.76%99.76%99.52%99.76%
Positive Predictive Value92.31%93.33%90.91%90.91%81.82%92.86%
Negative Predictive Value96.35%98.31%99.76%99.76%99.76%100.00%
Kappa0.8850.9120.9070.9070.8540.962

Published Readability of NMFP in AMD Screening Studies

StudyReadable Images by NMFPAge RangeNo. of Fundi Evaluated
De Bats et al80%65–931,363
Ouyang et al89.40%18–77284
Le Tien et al85.50%73–98238
Current study79%65–93491

10.3928/23258160-20150521-04

Sign up to receive

Journal E-contents