Retrospective cohort study.
Consecutive glaucoma patients from our Glaucoma Progression Study with ≥5 Swedish interactive threshold algorithm standard 24-2 VFs from 1999 through 2008.
Disc photographs of all patients were evaluated for the presence of DH. Exclusion criteria were conditions other than glaucoma likely to affect the VF and insufficient number of VFs to create a slope after DH detection. Automated pointwise linear regression was used to determine the rate of VF loss after DH detection. Fast progression was defined as a global VF loss of ≥1.5 dB/year. Factors associated with a fast rate of VF loss after the detection of the DH were evaluated.
Assessed variables included baseline (age, gender, intraocular pressure [IOP], central corneal thickness, VF mean deviation [MD], presence of migraine, Raynaud's phenomenon, low blood pressure, and exfoliation syndrome) and intercurrent data (DH recurrence, fellow eye involvement, glaucoma surgery, and IOP reduction). Odds ratios (ORs) and 95 % confidence intervals (CIs) were determined for each variable.
Seventy-six eyes (76 patients; mean age, 68.3±10.9 years) were enrolled. Mean IOP and VF MD at the time of the DH detection were 16.6±3.8 mmHg and −5.6±5.7 dB, respectively. The mean global progression rate after DH was −1.1±1.3 dB/year (mean follow-up, 3.8±2.8 years). A rate of progression of ≥1.5 dB/year was found in 20 (26 % ) eyes. Multivariate logistic regression analysis revealed larger baseline MD (OR, 1.11; 95 % CI, 1.01–1.20; P = 0.03) and older age (OR, 1.06; 95 % CI, 1.01–1.13; P = 0.04) to be significant risk factors for fast progression after DH. Eyes with a baseline MD worse than −4.0 dB had a 270 % increased risk of fast progression compared with those with an MD better than −4.0 dB.
The presence of a DH in older subjects with a worse VF predicted further VF global MD deterioration by more than 5 dB within 4 years. These eyes should undergo careful and frequent disease surveillance and consideration should be given to more aggressive treatment.
The author(s) have no proprietary or commercial interest in any materials discussed in this article.