Abstract
How should I decide whether to start treatment and, if so, which treatment? Randomised trials show that pressure reduction by drugs, or trabeculoplasty by laser or surgery, will slow the loss of the visual field.5 6 Deciding whether to start treatment, and if so which treatment, depends on the extent to which the benefits outweigh the harm of adverse effects of treatment. If my visual field loss was stable rather than progressive, treatment would involve harm without any potential for benefit. Ophthalmological opinion was divided about whether I should start treatment immediately or be monitored to assess progression before deciding whether to start treatment. Assessing progression is difficult against the background of measurement variability in Humphrey's visual field testing.7 Glaucoma is unlikely to start before the 30s or 40s. I therefore estimated that if I had lost 30% of my left visual field in the past decade or two, that meant about 1.5-3% loss each year. It seemed likely that this estimate provided an adequate predictor of future progression.8 I decided to start treatment: with the threat to fixation in my left eye, it seemed unwise to wait until an imperfect test showed progression. With a long term condition like glaucoma, it seemed sensible to start with drugs administered by eye drops and consider the more invasive trabeculoplasty procedures only if control was not adequate. How health professionals could optimise care Discuss with your patients the benefits and harms of alternative management strategies, including not starting treatment immediately. Patients may choose different strategies, depending on how they weigh up the benefits against the harms.3 Use information from the patient's history to help predict whether disease is stable or progressive. Relying solely on future monitoring, in which changes in test results could be due to random biological or technical variability, may cause unnecessary delays in deciding management. Am I administering the treatment correctly? When administering eye drops, occluding the nasolacrimal duct by pressing on the nose near the corner of the eye, and keeping the eyes still and closed for a few minutes, will maximise local absorption and minimise systemic absorption.9 Keeping the drops refrigerated makes it easier to assess whether the drop actually got into the eye. I wonder how many drug failures—and subsequent decisions about surgery—are due to poor administration of eye drops. The problem must be even greater as one gets older or develops arthritis or tremor. How health professionals could optimise care Show patients how to administer drugs and observe and correct their technique. How do I know if the treatment is working sufficiently well? To decide whether to continue or change treatment, it seems common practice to measure intraocular pressure about a month after starting or changing treatment. For some drugs, treatment response can be assessed within a few days.10 A good way of assessing the effect of a drug on intraocular pressure is to instil the drug in one eye and compare pressures between the eyes; it is best if the person measuring the pressure is not aware which eye is being treated.10 I arranged with my ophthalmologist that I would treat only one eye and declare which it was only after pressures were measured and recorded. We based treatment decisions on two masked (blinded is the term used in other areas of medicine) pressure measurements a few days apart. When we were assessing the effect of an additional drug, single-eye trials with double measurement of pressure became even more important because of variabilty in my intraocular pressure while I was taking the first drug. The measured reduction in intraocular pressure needs to be compared to a treatment goal, based on the evidence of what pressure reduction is sufficient to slow progression. One trial shows that a pressure reduction of 30% slows the progression of visual field loss.5 The greater the pressure reduction, the more visual fields are spared.11 Even after optimising drug treatment, the pressure reduction has on average been only 15% in my left eye and 22% in my right eye. In my left eye, where further loss of vision might threaten fixation, it now seems reasonable to have laser trabeculoplasty. Evaluating how I respond to that can help inform later decisions about my right eye. How health professionals could optimise care Take account of variability in measurements during monitoring by using multiple masked measurements to compare eyes during different treatments. Decide treatment goals with your patients on the basis of the evidence from randomised trials and their perceptions of the benefits and harms of the treatment required. How should progression of the disease be monitored? Although there are promising new techniques for monitoring progression of glaucoma, Humphrey's visual field testing is the most direct measure of how much of the visual field a patient sees. It remains the commonly used technique to monitor functional loss. With what is known about the measurement error in testing,7 12 I was surprised that monitoring visual fields from the time of diagnosis does not routinely have as its starting point three or four measurements within the space of a few weeks. Unfortunately, many health services and medical insurance schemes do not fund multiple measurements at baseline. I had four initial tests so that it would be easier to decide if any later tests showed changes or fell within the limits of test variability in my fields at the start. Both for intraocular pressure and visual fields, simply plotting measurements over time aids decision making; practitioners use such graphs all too seldom for monitoring risk factors and diseases. How health professionals could optimise care To monitor progression, establish a baseline by doing multiple initial tests. For both intraocular pressure and visual field loss, simple graphical...