Jump to comment:

• Re:Non-mydriatic fundus photography in epidemiological prevalence studies
  Kwadwo Owusu Akuffo
  Published on: 22 March 2016
• Non-mydriatic fundus photography in epidemiological prevalence studies
  Craig Wilde
  Published on: 22 March 2016

• Published on: 22 March 2016

  Re:Non-mydriatic fundus photography in epidemiological prevalence studies

  • Kwadwo Owusu Akuffo
  • Other Contributors:
    • 1 John M. Nolan,1 Tunde Peto, Jim Stack,Joanne Feeney, Rose Anne Kenny, Aisling M. O'Halloran, Rachel Moran, Cara Dooley, Hilary Cronin,Stephen Beatty

  We would like to thank Craig et al for their interest in our article.[1] TILDA was designed to provide accurate and reliable information regarding ageing in the Republic of Ireland in order to inform policy. Ophthalmic assessment, which included retinal photography, was a small component of the broader TILDA study. 1. We agree with Craig et al that mydriatic photography is preferable, when attempting to identify and grade age-rel...

  Show More

  We would like to thank Craig et al for their interest in our article.[1] TILDA was designed to provide accurate and reliable information regarding ageing in the Republic of Ireland in order to inform policy. Ophthalmic assessment, which included retinal photography, was a small component of the broader TILDA study. 1. We agree with Craig et al that mydriatic photography is preferable, when attempting to identify and grade age-related macular degeneration (AMD) in epidemiological studies. Unfortunately, this is not always feasible. 2. We believe, however, that the observed 2.2% prevalence of late AMD (in our sample) in the 75 years and over age-group, alluded to by Craig et al, is not attributable to our photographic technique. To illustrate our contention in this regard, we have done some example calculations (below), making diverse assumptions about prevalence among the participants whom we were unable to grade, thereby simulating several possible scenarios. Indeed, our own belief is that non-response bias, a feature of most social surveys, may be an important contributing factor, if indeed prevalence is under-estimated in our study. TILDA participants had the option of either a home-based or health centre assessment, and only the latter group were graded for AMD. Participants who opted for home-based assessment were more likely to be older, rate their health as poor, walk more slowly, more likely to self-report a disability than those who opted for health centre assessment.[2] Thus, the TILDA sample for our study was a relatively younger and healthier cohort, and the low prevalence of late AMD, which we reported, may reflect this. The high percentage of gradable photographs (over 96% in our sample, despite the lack of dilation) is also a likely consequence of this non-response bias. 3. Excluding ungradable participants, 9 out of 402 participants in our sample, aged 75 years and over, had late AMD - a prevalence rate of 2.2%. If dilation had been feasible, and all 108 ungradable participants had been successfully graded in consequence, how would this have affected the estimated prevalence? Thirty of these 108 participants were aged 75 years and over, and including these, we would have had 432 rather than 402 in this age group in our sample: a. Assuming that 30.8% of the 30 extra participants had late AMD (this figure of 30.8% comes from the over-80 subgroup in the Reykjavik Eye Study,[3]and represents by far the highest prevalence of late AMD of any subgroup listed in Table 7 of our article[1]), then 9 extra participants (30.8% of 30) in our study would have been classified as late AMD, and our estimated prevalence rate would then have been 18/432=4.2% rather than the 2.2% which we reported. b. Assuming 18.5% of these 30 participants had late AMD (18.5%, from the 85+ age group in the Blue Mountains Eye Study,[4] being the next highest figure in our Table 7), we would have reported 5 or 6 extra cases of late AMD, resulting in a prevalence of 14/432=3.2% or 15/432=3.5%. c. Assuming 7.1% prevalence (Beaver Dam Eye Study[5] which, like our study, and unlike both of the above studies, had 75 years and over as the oldest age grouping) the estimated prevalence would have been 11/432=2.5%. In conclusion, we concur with Craig et al that non-mydriatic photography represents a limitation of our study, which we acknowledged in our article.[1] However, we believe non-response bias is more likely to account for the lower-than-expected prevalence of late AMD in the 75 years and over subgroup of the TILDA sample. Funding: TILDA is funded by the Irish Department of Health, The Atlantic Philanthropies, and Irish Life plc. KOA and JMN are funded by the European Research Council (ERC). JMN is also funded by the Howard Foundation, Cambridge, UK. TP is funded by the NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, UK. Competing Interest: JMN and SB do consultancy work for nutraceutrical companies in a personal capacity and as directors of Nutrasight Consultancy Limited. All other authors report no potential conflict of interest. Contributors: KOA wrote the initial draft. All authors contributed to writing and critical revision of the article. Reference List [1] Akuffo KO, Nolan J, Stack J, Moran R, Feeney J, Kenny RA, et al. Prevalence of age-related macular degeneration in the Republic of Ireland. Br J Ophthalmol 2015 Aug;99(8):1037-44. [2] Cronin H, O'Regan C, Finucane C, Kearney P, Kenny RA. Health and aging: development of the Irish Longitudinal Study on Ageing health assessment. J Am Geriatr Soc 2013 May;61 Suppl 2:S269-S278. [3] Jonasson F, Arnarsson A, Sasaki H, Peto T, Sasaki K, Bird AC. The prevalence of age-related maculopathy in iceland: Reykjavik eye study. Arch Ophthalmol 2003 Mar;121(3):379-85. [4] Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia. The Blue Mountains Eye Study. Ophthalmology 1995 Oct;102(10):1450-60. [5] Klein R, Klein BEK, Linton KLP. Prevalence of Age-Related Maculopathy - the Beaver Dam Eye Study. Ophthalmology 1992;99(6):933-43.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 22 March 2016

  Non-mydriatic fundus photography in epidemiological prevalence studies

  • Craig Wilde, Ophthalmologist
  • Other Contributors:
    • Ali Poostchi, Winfried M Amoaku

  We read with interest the article by Akuffo et al reporting the prevalence of age related macular degeneration (AMD) in the Republic of Ireland and congratulate them on their findings. The authors report a recent meta-analysis demonstrating considerable heterogeneity in the prevalence of AMD across studies of European ancestry (1), with 20% variability being explained by the age-ranges used and 50% by study characterist...

  Show More

  We read with interest the article by Akuffo et al reporting the prevalence of age related macular degeneration (AMD) in the Republic of Ireland and congratulate them on their findings. The authors report a recent meta-analysis demonstrating considerable heterogeneity in the prevalence of AMD across studies of European ancestry (1), with 20% variability being explained by the age-ranges used and 50% by study characteristics. The TILDA age-specific prevalence for AMD in the over 75- year age group of 2.2% seems lower than we would expect. The authors report the measured prevalence was similar to that in the National Health and Nutrition Examination Survey (2). We would like to highlight this study shared one similar methodology that we feel may contribute to the low prevalence: that of non-mydriatic 45-degree digital colour fundus photography. The use of high resolution mydriatic stereo/non-stereoscopic digital imaging has been shown to be comparable at detecting AMD lesions when compared to film-based stereoscopic photographs (3), but the use of 45-degree non-mydriatic photography has several shortcomings. It reduces colour contrast and increases the frequency of poor quality/ungradable images, especially in individuals with small pupils and media opacities (3). It has been shown to have a low sensitivity (70%) for detecting AMD when compared to 30-degeree colour photography (4). Despite this, in the TILDA study over 96% of photographs were deemed gradable which is higher than some published rates (5). We understand dilation was not possible as AMD grading was not the only aim, but we would like to emphasise that when feasible, the use of mydriatic colour fundus photography is the preferred option for detecting AMD in epidemiological studies. It makes comparison to other published prevalence rates easier to interpret which was one of the fundamental ideas behind standardising prevalence studies with the use of The International Classification System.

  1. Rudnicka AR, Jarrar Z, Wormald R, Cook DG, Fletcher A, Owen CG. Age and gender variations in age-related macular degeneration prevalence in populations of European ancestry: a meta-analysis. Ophthalmology. 2012 Mar;119(3):571-80. PubMed PMID: 22176800. 2. Klein R, Chou CF, Klein BE, Zhang X, Meuer SM, Saaddine JB. Prevalence of age-related macular degeneration in the US population. Arch Ophthalmol. 2011 Jan;129(1):75-80. PubMed PMID: 21220632. 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 Nov;122(11):1642-6. PubMed PMID: 15534124. 4. 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 Feb;22(1):59-64. PubMed PMID: 11884880. 5. 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 Journal international d'ophtalmologie International journal of ophthalmology Zeitschrift fur Augenheilkunde. 2014;231(3):172-6. PubMed PMID: 24356326.

  Conflict of Interest:

  None declared

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top