Eighty-eight studies were included (1,915 cases); three were level 1 randomised controlled trials, three were level 2 cohort studies, and the remainder were lower level cohort studies or case series.
Subfoveal choroidal neovascularisation (765 cases): overall, 28% showed improved, and 25% worsened, visual acuity. The complication rate was 50% (CI: 12, 100) and the recurrence rate was 22%.
Macular translocation (792 cases): overall, 31% showed improved, and 27% worsened, visual acuity. The complication rate was 71% (CI: 8, 100) and the recurrence rate was 16%.
Pigment epithelium transplantation (94 cases): overall, 22% showed improved, and 21% worsened, visual acuity. The complication rate was 61% (CI: 29, 100) and the recurrence rate was 4%.
Removal of subretinal haemorrhage (264 cases): overall, 62% showed improved, and 13% worsened, visual acuity. The complication rate was 39% (CI: 0, 100) and the recurrence rate was 16%.
Logistic regression showed that the surgical intervention was the only significant predictor of improvement (p<0.0004), but not of deterioration (p=0.0568), in visual acuity. Study design was a significant predictor of improvement (p<0.0001) and deterioration (p<0.0001).