Purchase this article with an account.
R.H. Muni, M. Altaweel, M.T. Tennant, B. Weaver, P.J. Kertes; Agreement Among Canadian Retina Specialists In The Determination Of Treatment Eligibility For Photodynamic Therapy In Age–Related Macular Degeneration . Invest. Ophthalmol. Vis. Sci. 2006;47(13):347.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
It is important to consider how well retina specialists agree in their angiographic classification of CNV. Our goal was to determine the inter–observer agreement among Canadian retina specialists in the determination of treatment eligibility for photodynamic therapy and the angiographic classification of CNV in exudative age–related macular degeneration (AMD). Agreement was also determined between Canadian retina specialists and the University of Wisconsin Fundus Photograph Reading Center (Reading Center).
All retina specialists in Canada were asked to participate in a web–based survey, which consisted of 24 cases of exudative AMD provided by the Reading Center. Participants were provided with a digital colour and/or red–free photo, in addition to the digital FA for each case. Participants were asked to indicate if they would treat the patient with PDT and to categorize the angiographic appearance of the CNV. The agreement among participants was determined for decision to treat and for angiographic classification, by calculating the intraclass correlation (ICC) and pairwise kappa's. The angiographic interpretation by participants was also compared to that of the Reading Center.
40 Canadian retina specialists participated in the survey. The mean response rate for decision to treat and lesion categorization were 99.45% and 99.06% respectively. The mean raw agreement with the Reading Center and the ICC among participants for decision to treat were 68.7% and 0.293(95% CI=0.184 – 0.419) respectively. The mean raw agreement with the Reading Center and ICC for lesion categorization were 65.42% and 0.430 (95% CI= 0.364 – 0.518) respectively. The pooled estimate of kappa between observers and the gold standard for treatment decision and lesion categorization were 0.31 (95% CI=0.216–0.403) and 0.561 (95% CI=0.503–0.620) respectively.
There is poor inter–observer agreement among Canadian retina specialists for decision to treat with PDT and moderate agreement for angiographic CNV categorization. There is moderate agreement between observers and the Reading Center for angiographic CNV categorization. Agreement with the Reading Center was worst for minimally classic lesions and best for predominantly classic lesions.
This PDF is available to Subscribers Only