Published in:
01-04-2014 | Retinal Disorders
Diagnosis of subretinal neovascularization associated with idiopathic juxtafoveal retinal telangiectasia — fluorescein angiography versus spectral-domain optical coherence tomography
Published in: Graefe's Archive for Clinical and Experimental Ophthalmology | Issue 4/2014
Login to get accessAbstract
Background
To compare ability of fluorescein angiography (FFA) and spectral-domain optical coherence tomography (SD-OCT) to diagnose subretinal neovascularization (SRNV) secondary to idiopathic juxtafoveal retinal telangiectasia (IJRT) type 2A.
Methods
Two masked observers evaluated FFA and SD-OCT images separately to diagnose the presence of SRNV in 65 eyes of 36 patients. A third masked observer diagnosed SRNV on color fundus photo. Presence of SRNV on color fundus photo was defined as presence of subretinal hemorrhage, thickening of retina, and/or visible membrane at the macula. Presence of SRNV on FFA was defined when there was hyperfluorescence in early phase with increase in intensity and size in the late phase; presence of membrane, large irregular lesion, hypofluorescence due to subretinal haemorrhage. SRNV on SD-OCT was defined as the presence of thickening of the retinal pigment epithelium–choriocapillaris complex with or without intraretinal fluid or subretinal fluid and associated with intraretinal hyperreflective area with retinal thickening.
Results
Interobserver agreement (Kappa) for diagnosis of SRNV on FFA and SD-OCT was 0.373 (95 % CI, 0.106–0.617) and 0.775 (95 % CI, 0.612–0.899) respectively. The sensitivity and specificity of FFA were 52.3 % and 70.0 % respectively. With regard to SD-OCT, the sensitivity and specificity were 72.7 % and 64.1 % respectively in reference to color photograph. The negative predictive value (NPV) of SD-OCT (80.6 %) was higher than FFA (73.7 %).
When we considered only the presence of subretinal hemorrhage on color photograph as a confirmed diagnosis of SRNV and compared the diagnostic ability of FFA and SD-OCT, we found that the FFA had poor sensitivity (58.3 %) but a better specificity (71.8 %) than SD-OCT, which had sensitivity of 75 % and specificity of 64 % (Tables 1 and 2). However, the negative predictive value (NPV) of SD-OCT (89.29 %) was slightly better than FFA (84.85 %).
Conclusion
Interobserver agreement between the observers was better for SD-OCT than for FFA in making the diagnosis of SRNV. SD-OCT is a better diagnostic modality than FFA for ruling out the presence of SRNV.