Purpose
Although outcomes for exudative age-related macular degeneration have been transformed by anti-VEGF treatment, the prognosis for large sub-foveal haemorrhages remains poor. As eyes with large haemorrhages were excluded from most clinical trials, there is uncertainty about the optimum management. We carried out a retrospective observational clinical study to determine the outcomes of vitrectomy combined with sub-retinal tPA and anti-VEGF in the management of these eyes.
Methods
We examined the records of 36 patients presenting with sub-foveal haemorrhage associated with age-related macular degeneration. All patients had a complete pars plana vitrectomy, followed by injection of sub-retinal tPA and anti-VEGF. A complete fluid air exchange was performed and the eye was filled with 20% SF6. The patient was positioned supine for four hours and then chin down overnight. Depth and area of sub-retinal haemorrhage were measured with Topcon OCT. Data was entered into a MS Access database.
Results
<br /> Median follow up was 28 weeks with a minimum of 13 weeks. Pre-operative vision ranged from 6/60 to HM. 30 eyes had a vision less than 6/96. The mean haemorrhage area was 38.9mm2(S.D 17.9). 64% of eyes achieved a best visual acuity of ≥6/60. This dropped to 53% at their latest follow-up appointment. The mean change in visual acuity was a gain of 0.64 logMAR, and 78% improved by at least 0.2 logMAR. 10 patients suffered a re-bleed, usually whilst they were still on anti-VEGF. There were no macular holes or retinal detachments. The mean Latest LogMAR was better if the duration of haemorrhage was less than one week (t-test, p=0.03); if the haemorrhage was <40mm2 (t-test, p= 0.04); and if the presenting vision was better than HM (t-test, p= 0.003)<br /> <br />
Conclusions
Our results indicate that visual outcomes with vitrectomy, sub-retinal tPA and anti-VEGF appear to be significantly better than the natural history of sub-macular haemorrhage in age-related macular degeneration. Apart from recurrent sub-retinal bleeding, complications were rare. As visual outcomes are dependent on pre-operative factors such as haemorrhage area, presenting visual acuity, and duration of the haemorrhage, it is essential that these parameters are fully reported in future case series to enable meaningful comparison of outcomes.