LeBlanc and Bellan both take issue with our conclusion,1 based on VF-14 questionnaire results, that cataract surgery is currently being performed for doubtful indications in a substantial proportion of patients. There is no perfect instrument to measure visual function, but the VF-14 was developed by ophthalmologists for their use in assessing cataract patients and is apparently acknowledged as the best tool there is. Anderson also acknowledges that on the VF-14 “a score of 88 is in fact an excellent outcome” and that “100 represents absolutely no visual disability.” It is therefore difficult to understand the decision to operate in the 15% of patients who scored above 95, and especially in the 4% of patients with the astonishing score of 100, at the time of preoperative assessment in our study. In choosing the VF-14 for our study we relied on the ophthalmology literature, the epidemiologists working in the University of British Columbia Department of Ophthalmology and the advice of the ophthalmologists associated with the project. The consensus remains (as quoted by Bellan himself) that the subjective VF-14 score correlates more strongly with visual function than any objective measurement of visual acuity made by the surgeon. Anderson's last paragraph seems to deny this accepted conclusion from cataract outcomes research, and he returns to suggesting that measured visual acuity, rather than the VF-14, is the most appropriate measure of outcome.
In claiming that the ophthalmologists involved in the project have reported “visual improvement” in 92.4% of patients, LeBlanc perpetuates the misapprehension that visual acuity as reported by the surgeon is a better measure of visual function than the VF-14 as reported by the patient and as used in our study. We agree that the question he suggests for determining patient satisfaction would be a good one in any evaluation of elective surgical outcomes. For example, it could be added as a final question in the postoperative application of the VF-14 questionnaire.
The reported results were restricted to patients undergoing first-eye surgery because the steering committee was uncertain how to deal with the 1-eye or 2-eye issue raised by Bellan, and current practice varies widely in relation to indications for and timing of surgery on the second eye. Bellan seems to be arguing for routinely operating on both eyes, but we must leave this question (for patients with or without mild cataract in the second eye or postoperative anisometropia) to be answered by ophthalmologists on the basis of research evidence.
Finally, we did not suggest that the VF-14 should be used with some kind of absolute threshold as the sole criterion of the need for surgery. As with any operation, the recommendation to proceed must be based on careful assessment of symptoms, functional impairment and findings in relation to the potential benefits and risks of the procedure. Surgery may indeed be indicated in some patients with minor visual impairment, but our conclusion that “the threshold indications for cataract surgery are now very low” remains valid (and is probably an understatement) on the basis of the evidence we obtained from the best instrument currently available.
Charles J. Wright Centre for Clinical Epidemiology and Evaluation Vancouver Hospital and Health Sciences Centre Vancouver, BC
Reference
- 1.