[Measurement of macular edema in retinal vein occlusion using optical coherence tomography and its visual prognosis].

OBJECTIVE: To examine the type of macular edema in patients with retinal vein occlusion by using optical coherence tomography (OCT). To compare the sensitivity and specificity between OCT and fundus fluorescein angiography (FFA). To investigate the visual prognosis and risk factors in patients with cystoid macular edema (CME). METHODS: Ninety-one eyes of 90 patients with various types of retinal vein occlusion were examined by OCT and FFA. There were 54 cases male and 36 cases female. The age of these patients ranged from 20 to 74 years old and averaged (57.8 +/- 13.8) years old. Right eye was affected in 45 cases, and left eye in 46 cases (both eye in 1 case). Central retinal vein, hemicentral retinal vein and branch retinal vein were affected in 54, 9 and 28 eyes, respectively. The average follow-up period was 6.10 months. The sensitivity and specificity of OCT and FFA were measured. The visual prognosis and risk factors were analyzed. Macular central retinal thickness of 54 opposite normal eye was measured for comparison. RESULTS: (1) The classification of macular edema by OCT was cystoid macular edema in 71 eyes (78.0%); subretinal fluid in 14 eyes and laminar macular hole in 1 eye. The minimal and maximal height of intraretinal cystoid space was 94 microm and 1317 microm, respectively and averaged (668.18 +/- 245.58) microm. The minimal and maximal height of macular central retinal thickness of 54 opposite normal eye was 110 microm and 236 microm, respectively and averaged (154.09 +/- 21.85) microm. The maximal height of subretinal fluid space was 1377 microm or even beyond the detective limit of OCT. (2) The sensitivity of OCT for detection of CME was 98.6% and the specificity was 100%. The sensitivity of FFA was 86.1% and the specificity was 100.0%. (3) The visual prognosis of 61 eyes with CME follow-up over 3 months: The difference between the initial and final VA in branch retinal vein occlusion was statistically significant (P < 0.01), while no difference between initial and final VA in central and hemicentral retinal vein occlusion (P > 0.05). The factors affected visual prognosis of CME included the duration, the presence of hemorrhage in cystoid space and the severity of occlusion, etc. In CME under 3 months, the visual prognosis after therapy is better than that of before the therapy (P < 0.01); while in eyes with duration more than 3 months, no difference of vision could be detected before and after the therapy (P > 0.05). CONCLUSIONS: OCT is a safe, high-resolution, non-invasive, reliable and reproducible new examine method for detecting CME. The visual prognosis of CME is poor. It is possible that the visual prognosis can be improved by early detecting CME using OCT.