gms | German Medical Science

Background: To understand visual outcomes in study eyes after release from protocol in participants enrolled in the IVAN clinical trial.

Methods: 532 participants eligible for inclusion after IVAN exit. 124 died before data collection. Approval was available for passive data collection on all survivors and those deceased. 199 of the survivors attended a research visit. Data from every visit was extracted along with VA and treatments administered to each eye. BCVA, LLA and macular imaging captured in attenders and in non attenders images from the most recent visit were obtained. The mean, median of DVA after IVAN exit by type of follow up in survivors and deceased patients and change in DVA, visit and injection rates for each year of follow up by BCVA category at IVAN exit was computed. Duration of follow up was length of time between IVAN exit and cessation of active management and used for rate of change in DVA/year, injection and visit rates A multivariable linear random effects model was used to test effects of covariates.

Results: The median DVA was 58.0 letters (IQR 34.0, 73.0), 14 letters worse than at IVAN exit with a median change -10 letters, [IQR -22.0 to -2.0]. One third had a VA better than 68 letters. One fifth (had a VA of worse than 33 letters. The multivariable model estimated the reduction in DVA during study eye monitoring to be -4.3 letters per year (95% CI 3.7 to 4.9). The only significant interaction was between age and time (p<0.001) with DVA deteriorating faster in older participants and in the group who died after IVAN exit but prior to data collection. Injection rate did not influence DVA decline

Conclusion: Despite intensive active management of nAMDwith anti VEGF the VA in study eyes deteriorated with an average loss of around 4 letters per year