Comparison of current treatment strategies for AMD-related submacular haemorrhage: Real-world outcomes

PURPOSE: Comparison of available treatment modalities for neovascular AMD-related submacular haemorrhage including robot-assisted subretinal drug delivery. METHODS: Single centre, long-term, retrospective analysis of patients treated for nAMD-related submacular haemorrhage. For each treatment group, outcomes assessed included haemorrhage resolution and best corrected visual acuity (BCVA) change with sub-analyses by haemorrhage size. RESULTS: We identified 55 patients with classic nAMD-related submacular haemorrhage. We observed a preference for the surgical approach injecting subretinal TPA (manual or robotic) with intravitreal anti-VEGF and pneumatic displacement (30 cases: 8:10:12 ratio of small:medium:large haemorrhage), over intravitreal TPA/anti-VEGF/pneumatic displacement (13 cases: 4:5:4) or intravitreal anti-VEGF alone (12 cases 8:4:0). Subretinal TPA led to significant decrease in haemorrhage size by month 1 compared to intravitreal TPA or anti-VEGF monotherapy (p < 0.0001, Tukey test) with most cases resolving by 12 months irrespective of the treatment group. Mean BCVA change over 12 months was not significantly different between treatment groups, although by 12 months + 15 letter gain was observed post subretinal TPA, + 11 after intravitreal TPA and -7 letters loss after anti-VEGF monotherapy. A higher proportion of patients in the surgical group (53% subretinal and 69% intravitreal TPA) were associated with BCVA improvement, or stabilisation (33% and 15% respectively), while anti-VEGF alone led to more frequent BCVA deteriorations (50% of cases). Robot-assisted TPA injection showed comparable result to the manual technique. CONCLUSIONS: Subretinal TPA approach, manual and robotic, resulted most effective clearance of submacular haemorrhage in the early post-operative phase, with some potential clinical benefit for longer term visual function.