Aims To report the incidence, risk factors and long-term outcomes of laser goniopuncture (LGP) in patients with previous deep sclerectomy (DS).

Methods Retrospective cohort study of 1765 eyes (1385 patients) undergoing DS with or without cataract surgery between 2001 and 2020 in two UK institutions. Kaplan-Meier was used to estimate LGP incidence. DS success after LGP was calculated for criteria A, B, and C defined as intraocular pressure (IOP) of ≤18, ≤15 and ≤12 mm Hg with 20%, 25% and 30% reduction, respectively. Cox regression was used to investigate factors associated with the risk of failure.

Results LGP had an estimated incidence of 33.3% (30.9%–35.6%), 56.3% (53.5%–58.9%) and 62.8% (59.7%–65.6%) at 1, 3, 5 years, respectively. Mean (±SD) IOP significantly (p<0.001) decreased from 21.2 (±6.0) mm Hg pre-LGP to 13.8 (±5.2) mm Hg and 12.9 (±4.7) mm Hg at 3 and 5 years post-LGP, respectively. Success rates at 3 and 5 years were, respectively, 40.9% (37.5%–44.6%) and 33.7% (30.3%–37.6%) for criterion A; 27.1% (24.0%–30.5%) and 22.3% (19.3%–25.7%) for criterion B and 13.9% (11.6%–16.7%) and 11.6% (9.5%–14.3%) for criterion C. In all models, higher pre-LGP IOP (p<0.001) and higher pre-LGP medication number (p<0.001) were associated with increased failure, while male gender (p≤0.004), intraoperative mitomycin C (p≤0.031), longer interval between DS and LGP (p≤0.01) with reduced failure.

Conclusion Most patients undergoing DS will eventually require LGP. LGP is effective at rescuing eyes with a failing DS. This study identifies several factors associated with LGP outcomes, knowledge of which may help clinicians predict LGP success.