Purpose
Trend-based analyses examining rates of visual field (VF) loss in glaucoma are useful for predicting risk of vision-related morbidity. Although patients with faster losses are more likely to require treatment escalation, little is known about rates that might trigger the clinician’s decision to intervene surgically. We performed a retrospective, observational clinical study to investigate prior rates of VF progression in patients attending for trabeculectomy.
Methods
We extracted VF data from patient charts for 112 eyes of 80 consecutive patients with glaucoma attending for trabeculectomy, including 30 patients referred from general ophthalmology clinics and 50 patients from specialist glaucoma clinics. Rates of change in standard automated perimetry (SAP) mean deviation (MD) were examined using linear regression and random coefficient models. Patients with neovascular, congenital or other forms of secondary glaucoma were excluded, as were eyes that had previous trabeculectomy. Data regarding intraocular pressure, number of glaucoma medications and number of medication changes prior to surgery were also recorded.
Results
Mean age at surgery was 63.9 ± 10.5 years. On average patients were followed for 5.9 ± 3.4 years prior to surgery with 6.6 ± 2.6 useable VFs per eye. Average MD at baseline was -7.3 ± 7.2 dB with a median deterioration of 0.72 dB per year. Accounting for possible confounding factors, patients referred from general ophthalmologists lost 1.06 dB per year compared to 0.78 dB per year in those referred from glaucoma specialists (P =0.070). Patients referred from general ophthalmologists had more medication changes prior to surgery (3.3 ± 1.4 versus 2.6 ± 1.3, P=0.007).
Conclusions
Patients attending for trabeculectomy had faster average rates of VF loss prior to surgery than published values for the general glaucoma population. Those managed by glaucoma specialists had fewer changes in medication and tended to slower rates of VF loss, although the latter did not reach statistical significance.