Long-term impact of delayed follow-up due to COVID-19 lockdown on patients with neovascular age-related macular degeneration

BACKGROUND: During the first wave of the coronavirus disease 2019 (COVID-19) pandemic in 2020 outpatient care of neovascular age-related macular degeneration (nAMD) patients was severely reduced due to lockdown. Missed visits are known to be detrimental to patients in need of continued anti-vascular endothelial growth factor (VEGF) intravitreal injections (IVIs). The purpose of the study was to assess the effect of a month-long pause of regular visits and anti-VEGF IVIs in nAMD patients. METHODS: A retrospective study was performed. Patients were treated in a pro re nata (“as needed”) scheme. Distance (logMAR) and near (logRAD) visual acuity (VA), optical coherence tomography, delay between planned and actual visit date and the indication for IVI were assessed for 3 continous visits in the 6 months before lockdown (V-3, -2, -1) and the 2 visits after lockdown (V0, V + 1). For analysis of long-term impact, records for visits 1 years before and after lockdown (V-3, V + 2) were gathered. RESULTS: We included 166 patients (120 female, 46 male) with a median (range) age of 80.88 (59.8–99.36) years. Compared to V-1, distance VA was significantly worse at both V0 (0.27 ± 0.21 vs 0.31 ± 0.23 logMAR, p < 0.001) and V + 1 (0.27 ± 0.21 vs 0.30 ± 0.23 logMAR, p = 0.021). Near VA was significantly worse at both V0 (0.31 ± 0.21 vs 0.34 ± 0.22 logRAD, p = 0.037) and V + 1 (0.31 ± 0.21 vs 0.34 ± 0.22 logRAD, p = 0.02). Visit delay (VD) at V0 was significantly longer than at V + 1 (30.81 ± 20.44 vs 2.02 ± 6.79 days, p < 0.0001). Linear regression analysis showed a significant association between visit delay and a reduction of near VA between V-1 and V + 1 (p = 0.0223). There was a significant loss of distance VA (p = 0.02) in the year after the lockdown period (n = 125) compared to the year before. Loss of reading acuity was not significantly increased (p = 0.3). One year post lockdown, there was no correlation between VA change and visit delay after lockdown (p > 0.05). CONCLUSIONS: In nAMD patients whose visits and treatment were paused for a month during the first wave of the COVID-19 pandemic, we found a loss of VA immediately after lockdown, which persisted during follow-up despite re-established anti-VEGF treatment. In the short term, length of delay was predictive for loss of reading VA. The comparison of development of VA during the year before and after the lockdown showed a progression of nAMD related VA loss which may have been accelerated by the disruption of regular visits and treatment. TRIAL REGISTRATION: This article does not report the outcome of a health care intervention. This retrospective study was therefore not registered in a clinical trials database.