Purchase this article with an account.
S.M. Shah, S. Tatlipinar, H. Tabandeh, Q.D. Nguyen, E. Quinlan, I. Zimmer–Galler, P.A. Campochiaro; Quantitative Fluorescein Angiography (FA) for Assessment of Treatment Effects in Patients With Choroidal Neovascularization (CNV) . Invest. Ophthalmol. Vis. Sci. 2005;46(13):1396.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: FA is a valuable technique for identifying, classifying, and measuring the size of CNV lesions. The amount and rate of dye collection within and leakage from CNV is assessed subjectively. In this study we sought to develop and characterize a technique to quantify CNV size, dye collection and leakage. Methods: High resolution digital FA was performed on six patients with subfoveal CNV using a digital FA system. FAs were done before and 3 months after photodynamic therapy (PDT). Frames with adequate clarity and centration were analyzed by Matlab® to exclude underexposed or overexposed images. The border of each CNV lesion was outlined in selected frames and pixel area was measured using ImagePro Plus® by two independent observers and normalized with optic disc area. An overlay mask was drawn over the largest area of the lesion and its luminosity was computed. Six avascular areas of the background were chosen and their mean luminosity was subtracted from that of the lesion to obtain net CNV luminosity. Pixel area and luminosity of each frame were plotted against time from injection and best–fit curves were obtained. Results: The mean area under curve for pixel area/time (AUCP) was 504.61 at baseline and increased to 770.12 after treatment. AUC for luminosity/time (AUCL) was 10483 at baseline and 9883 after treatment. On clinical examination and review of FA, three patients showed an increase in lesion size with decreased leakage at month 3 compared to baseline, corresponding to an AUCP increase of 50.1% (597 at baseline to 900 after treatment), and an AUCL increase of 7.1% (7702 at baseline to 8250 after treatment; slight increase attributable to increased lesion size). One patient had decreased leakage with corresponding decrease in AUCP from 228 to165 and in AUCL from 12716 to 9773. Two patients showed no change in clinical findings or FA. Conclusions: We have developed a reproducible computer–based method to quantify FAs that can be used as an outcome measure in clinical trials or standard care. Such techniques may provide an objective, continuous scale index of disease progression or regression. This index and its graphical representation coupled with visual acuity measurements may be useful in detecting treatment responses in patients with CNV.
This PDF is available to Subscribers Only