Men, Patients With Disciform Scars Have Greater Risk of Developing SMH, Study Shows

For patients treated with vascular endothelial growth factor (VEGF) inhibitors using intravitreal injections, submacular hemorrhages (SMH) could affect men and patients with disciform scars at baseline, according to a study published in Acta Ophthalmologica.

The study aimed to assess the estimated incidence of SMH with loss of vision secondary to age-related macular degeneration (nAMD) treated with VEGF inhibitors in clinical practice. The study also aimed to report the risk factors for developing SMH in nAMD-treated eyes, evaluate the 1-year treatment outcomes of SMH eyes compared with matched control eyes, and look for predictors of recovery of vision after 12 months.

The study was a retrospective analysis conducted from January 1, 2010, to December 31, 2020. The researchers used the observational database Fight Retinal Blindness! (FRB) registry for this study. Participants for this study were from Australia, France, Ireland, Italy, Netherlands, New Zealand, Spain, Singapore, Switzerland, and the United Kingdom.

Researchers used data from the AMD module of the FRB outcomes registry. Data included the visual acuity (VA), the activity of underlying choroidal neovascularization (CNV) lesion, treatment given, procedures, and ocular adverse events. Baseline exams captured demographic characteristics, the type and greatest linear diameter of the CNV, and whether the eye received prior treatment for the condition being audited.

Participants were excluded if they had any other cause of hemorrhage.

The researchers used a matched cohort with 5 controls; these were participants whose eyes did not develop SMH with loss of vision during treatment.

The study's main outcome was the estimated incidence of SMH with loss of vision over the study period. The secondary outcome was the cumulative rate of SMH and factors that predicted the development of SMH. Other outcomes included visual and treatment outcomes at 12 months after SMH. Researchers used locally weighted scatterplot smoothing regression to visualize longitudinal visual outcomes between SMH case eyes and control eyes.

The study included 6425 patients and 7642 eyes who had received 135,095 IVTs overall. Of the eligible eyes, 19% completed at least 5 years of follow-up. There were 105 eyes who developed SMH with loss of vision during the study period, 95 of which developed before December 31, 2019. There were 73 eyes that completed 12 months of follow-up after SMH diagnosis.

The median (IQR) incidence of SMH with loss of vision per year per 1000 patients and 10,000 injections was 4.6 (3.8, 5.6) and 7.8 (6.4, 9.4) respectively. Patients who lost their vision with SMH were more frequently those whose eyes had polypoidal choroidal vasculopathy.

The cumulative rate of SMH was 1.1% at 2 years, 1.7% at 4 years, 2.5% at 6 years, 3.2% at 8 years, and 4.4% at 10 years of follow up. The cumulative rate of SMH was 0.95% for 10 injections, 1.62% at 20 injections, 2.12% at 30 injections, 2.59% at 40 injections, and 3.01% at 50 injections The study did not find any significant increase in the probability of SMH with each injection.

The likelihood of developing SMH increased by 71% for male participants (HR 1.71; 95% CI, 1.26, 2.16). Age, presenting VA, and baseline angiographic size of the CNV lesion were not associated with SMH development. However, eyes with disciform scars at baseline had a significantly increased risk of SMH (HR 12.35; 95% CI, 10.76, 13.94).

Of the participants, 60% had their last injection fewer than 60 days before the development of SMH with loss of vision. CNV lesions tended to be graded as active at the last visit before it developed compared with their matched controls (48% vs 36%). The mean change in VA from prior SMH was ­–20 (–26, –15) 12 months after SMH. Of the participants 60% and 53% had 2-line and 3-line loss of vision respectively.

A quarter of patients with SMH recovered their VA prior to SMH. Patients with better VA at the last visit prior to SMH were less likely to recover their previous VA at 1 year (odds ratio [OR] 0.98; 95% CI, 0.95, 1.00). However, this was not statistically significant (OR 1.02; 95% CI, 1.00, 1.05). Eyes that had initial treatment did not have better outcomes at 1 year than eyes that continued with VEGF inhibitor monotherapy.

There were several limitations; use of self-reported data can lead to an undercount of events. Patients were not randomization during initial treatment; of note, researchers were only able to obtain additional data on initial SMH management on 58% of cases. Symptom duration was not recorded in the FRB database.

The researchers concluded that men and patients with disciform scars were at a greater risk of developing SMH. They also concluded that visual outcomes were poor for half of the SMH cases, with half of the participants losing at least 3 lines of vision a year later.

Reference

Gabrielle PH, Maitrias S, Nguyen V, et al. Incidence, risk factors and outcomes of submacular haemorrhage with loss of vision in neovascular age-related macular degeneration in daily clinical practice: data from the FRB! registry. Acta Ophthalmol. Published online March 23, 2022. doi:10.1111/aos.15137.