10 years of submacular haemorrhage (SMH): 101 cases at a tertiary centre – Outcomes and aetiology

Author: Haseeb Akram (United Kingdom)

Co-authors: Marco Isac, Niral Karia, Aman Chandra

Purpose

To describe aetiologies and visual outcome of the largest UK dataset of submacular haemorrhage treated with pneumatic (PD) and pars plana vitrectomy assisted displacement (PPV). We analysed descriptive and outcome differences in various aetiologies for SMH and after PD or PPV.

Setting/Venue

Ophthalmology Department, Southend University Hospital, Southend-On-Sea, Essex, UK

Methods

Retrospective case analysis of patients treated for SMH between April 2011 and April 2021. Primary outcome was logMAR Best corrected visual acuity (BCVA) change 3 months post treatment. BCVA compared pre-operatively with 3 months post operatively and the difference in BCVA (ΔBCVA) was analysed. We analysed demographic, and ocular features of patients, previous anti Vascular Endothelial Growth Factor (VEGF) treatment and surgical intervention. Surgical intervention included either (a) pneumatic displacement (PD) using tissue plasminogen activator (t-PA) and gas tamponade with or without anti-VEGF intravitreal injections; or (b) Pars Plana Vitrectomy (PPV) using tissue plasminogen activator (t-PA) and gas tamponade with or without anti VEGF subretinal injections.

Results

101 cases representing 94 eyes were identified. 5 were re-operated due to insufficient displacement and 2 were re-operated due to a subsequent re-SMH. Median age was 81.5 years with 59 (62.8%) female patients. Aetiologies included neovascular age-related macular degeneration (nAMD) (79 (84.0%)), macroaneurysm (11 (11.7%)), trauma (2 (2.1%)), unspecified (2 (2.1%)). 39 (41.5%) were either using antiplatelet or anticoagulant medication. 64 (68.1%) underwent PD; 5 (7.8%) of these subsequently underwent PPV due to insufficient displacement of haemorrhage. The remaining 31.9% underwent primary PPV. 20 (24.7%) of SMH caused by nAMD had had preceding anti-VEGF therapy. Median number of injections prior to SMH was 6 (range 2-28) and median number of weeks since last injection was 13.75 (range 2.5-150). 9 (11.4%) of nAMD eyes had a re-bleed after treatment. 8 (88.9%) of these occurred beyond 8 weeks of their last anti-VEGF injection. Other aetiologies did not re-bleed. Visual acuity analysis was available for 60 eyes. Overall, median ΔBCVA was -0.37 (+1.40 to -2.03) (p< .00001). Median ΔBCVA was -1.10 in macroaneurysm SMH and -0.30 in nAMD SMH (p=0.02). No significant difference in ΔBCVA was found between naïve nAMD and treated (p=0.32) or PD and PPV (p=0.09).

Conlusions

87% of SMH were caused by nAMD. 75.3% of these were previously untreated. 2.5% of nAMD SMH subsequently had a further SMH. Displacement of SMH by either PD or PPV significantly improved visual acuity. Visual acuity improvement was similar in eyes previously treated with anti-VEGF with naïve eyes, and between PD and PPV assisted displacement. Macroaneurysm SMH had significantly better visual improvement than nAMD SMH.

Financial Disclosure

None to declare