Purchase this article with an account.
Florence Cabot, Sonia Yoo, Alain Saad, George Kymionis, Ana Paula Canto, Damien Gatinel; Refractive surgery evaluation and risk of corneal ectasia, inter-rater agreement and comparison to a new software: the SCORE analyzer. Invest. Ophthalmol. Vis. Sci. 2013;54(15):3138.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To assess the level of agreement between 5 refractive surgeons and a new software in detecting corneal ectasia risk during refractive surgery evaluation.
Retrospective multicenter study including 168 eyes of 84 patients. Three attending ophthalmologists subspecializing in refractive surgery and 2 cornea fellows reviewed 168 Orbscan (Bausch and Lomb, Rochester, NY) topographic quadmaps extracted from the Bascom Palmer Eye Institute refractive evaluation database. They rated the maps and determined whether a refractive surgery (Laser in situ keratomileusis [LASIK], or Photorefractive keratectomy [PRK]) was advisable or not. Topographic maps were divided in 3 different groups according to the following classification: LASIK or PRK (group 1), PRK only (group 2), No surgery (group 3). Only topographic indices were used to classify the maps. Age, daily activities, family history, symmetry between both eyes and refraction were not taken into account to rate the maps. The same maps were also screened by a new corneal ectasia risk detection software combined with the Orbscan: the SCORE analyzer (Technnolas Perfect Vision). Multirater kappa statistics were used to assess inter-rater agreement.
43.1% and 58.9% of topographic maps were classified in group 1 ; 37.1% and 2.9% in group 3 by cornea fellows. 75,6% , 46.2% and 55% of topographic maps were classified in group 1 ; 13.2% , 18.7% and 1.8% in group 3 by subspecialized attending ophthalmologists. 57.7% of topographic maps were classified in group 1 and 9.6% in group 3 by the SCORE analyzer (Technolas Perfect Vision, Munich, Germany). Inter-rater agreement was slight to fair compared with that expected by chance: the overall rate of agreement was 56% and the fixed marginal kappa coefficient was 0.24.
The inter-rater agreement between experienced refractive surgeons with different backgrounds is relatively low. Using the SCORE analyzer as an objective assessment in determining the patient’s candidacy for refractive surgery is valuable. Age, refraction, symmetry between eyes and family history are other factors that may be incorporated into the SCORE analysis in the future to make the assessment more robust.
This PDF is available to Subscribers Only