The National Survey of Trabeculectomy. I. Sample and methods

Purpose. The National Survey of Trabeculectomy was designed to evaluate current practices of glaucoma surgery in the United Kingdom and to determine the success and complication rates of trabeculectomy on a national basis. This paper reports the survey methods, levels of consultant activity, waiting times, indications for surgery and the demographic and clinical characteristics of the patient sample. Methods. Consultant ophthalmologists performing trabeculectomy in the United Kingdom were studied. Four consecutive patients undergoing trabeculectomy under each consultant prior to 18 June 1996 were retrospectively sampled. Patients were followed prospectively and evaluated 6 and 12 months after surgery. Data were collected by self-administered postal questionnaires. To determine the effects of selection and reporting bias a validation study of 14 randomly selected units was also conducted. Results. Three hundred and eighty-two consultants recruited 1454 eligible patients for analysis. The mean age of patients was 69.2 years (standard deviation 10.9) and 51.7% were male. The underlying diagnosis was primary open angle glaucoma in 89.2%, pseudoexfoliation glaucoma in 5.4%, normal tension glaucoma in 3.8% and pigmentary glaucoma in 1.6%. There was advanced visual field damage in 50.5% of the cohort by the time of listing. The main indications for surgery were failure of medication to control intraocular pressure in 57.1%, progressive visual field loss in 26.5% and progressive optic disc damage in 4.8%. Primary surgery was undertaken in 4.8% of patients. In 80% trabeculectomy was performed within 3 months of listing. However, almost a third of consultants considered individual patient's waiting time too long. Validation studies confirmed that systematic bias did not operate in the selection of patients for the survey or in the reporting of outcomes. Conclusion. The findings of this survey are representative of current practices of trabeculectomy by consultants throughout the United Kingdom and show considerable variation in practice. Failure to control intraocular pressure with topical medications was the main indication for surgery. Advanced glaucomatous visual field damage was present at the time of surgery in half the sample. Though most patients were operated on within 3 months of listing, almost a third of consultants considered the wait unacceptably long.