PURPOSE: To evaluate transient, large visual acuity (VA) decreases, termed sporadic vision loss, during anti-vascular endothelial growth factor treatment for neovascular age-related macular degeneration (AMD). DESIGN: Cohort within a randomized clinical trial. METHODS: setting: Comparison of Age-Related Macular Degeneration Treatments Trials (CATT). study population: Total of 1185 CATT patients. main outcome measures: Incidence of sporadic vision loss and odds ratio (OR) for association with patient and ocular factors. Sporadic vision loss was a decline of ≥15 letters from the previous visit, followed by a return at the next visit to no more than 5 letters worse than the visit before the VA loss. RESULTS: There were 143 sporadic vision loss events in 122 of 1185 patients (10.3%). Mean VA at 2 years for those with and without sporadic vision loss was 58.5 (∼20/63) and 68.4 (∼20/40) letters, respectively (P < .001). Among patients treated pro re nata, no injection was given for 27.6% (27/98) of sporadic vision loss events. Multivariate analysis demonstrated that baseline predictors for sporadic vision loss included worse baseline VA (OR 2.92, 95% confidence interval [CI]:1.65-5.17 for ≤20/200 compared with ≥20/40), scar (OR 2.21, 95% CI:1.22-4.01), intraretinal foveal fluid on optical coherence tomography (OR 1.80, 95% CI:1.11-2.91), and medical history of anxiety (OR 1.90, 95% CI:1.12-3.24) and syncope (OR 2.75, 95% CI:1.45-5.22). Refraction decreased the likelihood of sporadic vision loss (OR 0.62, 95%CI: 0.42-0.91). CONCLUSIONS: Approximately 10% of CATT patients had sporadic vision loss. Baseline predictors included AMD-related factors and factors independent of AMD. These data are relevant for clinicians in practice and those involved in clinical trials.