Risk factors for progression in glaucoma. The Groningen Longitudinal Glaucoma Study

Abstract

Abstract Purpose To investigate risk factors associated with visual field progression in glaucoma. Methods 221 patients with a reproducible glaucomatous visual field defect at baseline were followed prospectively with perimetry (HFA 30‐2). Three criteria for progression were used: the EMGT algorithm (GPA), a non‐parametric algorithm applied to mean deviation (MD; NPA) and MD slope. For progression according to GPA, the last field had to be labelled as possible or likely progression. For progression according to NPA, the last two or more consecutive fields had to have an MD value worse than the worse baseline MD value. Risk factor analyses were performed using Cox proportional hazard models (dependent variables: classification by GPA and NPA) and multiple linear regression (dependent variable: MD slope). Results Mean follow‐up was 5.4 years; on average 7.2 reliable fields were available. Mean MD at baseline and MD slope during follow‐up were ‐9.9 dB and ‐0.26 dB/yr for OD and ‐9.0 dB and ‐0.21 dB/yr for OS. Of 167 eligible right eyes, 45 showed progression with GPA and 69 with NPA. For OS, these numbers were 36 and 67 of 167. Mean IOP during follow‐up (per mmHg increase; GPA: HR 1.13, P=0.014; NPA: HR 1.10, P=0.016; MD slope: ‐0.042 dB/yr/mmHg, P=0.002) and disease stage (early versus moderate/severe glaucoma with MD = ‐6 dB as cut‐off point; GPA: HR 1.88, P=0.034; NPA: HR 2.05, P=0,001; MD slope: ‐0.20 dB/yr, P=0.010) were found to be significant risk factors. Less convincing associations were found for family history of glaucoma and age. Highest IOP ever measured, myopia, sex and cardiovascular disease were not associated with progression. Conclusion Higher mean IOP during follow‐up and worse disease stage at baseline were associated with progression.