Evaluating elective surgery

Charles Wright and colleagues¹ concluded that indications for and outcomes of elective surgery can be evaluated systematically at reasonable cost. Their analysis focused on 6 procedures, including cataract surgery. For patients who had undergone this procedure, the authors highlighted the finding that “32%, 15% and 4% [of patients] had a preoperative visual function score of greater than 90, greater than 95 and 100 respectively (on a scale of 100).” They concluded that “the threshold indications for cataract surgery are now very low.” On the surface, this seems a reasonable claim if the measurement tool accurately quantifies visual impairment. However, the measurement tool used by Wright and colleagues,¹ the VF-14 (which was originally described in a report of the Patient Outcomes Research Trial [PORT]), only partly reflects visual impairment.² This tool is widely used because it correlates “more strongly with [patients'] overall self-rating of the amount of trouble they have with vision and of their satisfaction with vision than do any of several measures of their visual acuity.”² However, the correlation value (Spearman correlation coefficient) was only –0.45, which indicates that more than half of patients' trouble with vision was not accounted for by this measure. This led the PORT authors to conclude that “vision-related functional status measures, in conjunction with the global ratings by patients of their vision and visual acuity, will likely prove to be better indicators of the need for and outcome of cataract surgery than will visual acuity or a general measurement of functional status alone.”² By basing their assessment of visual difficulty on the Visual Function Assessment alone, Wright and colleagues¹ underestimated the magnitude of patients' vision problems.

The article also highlighted that 27% of patients had worse Visual Function Assessment scores after surgery than before. This result might lead one to wonder how much inappropriate surgery is being undertaken. Not surprisingly, the press has picked up on these concerns.^3,4 The explanation for this worrisome finding can be found in the more detailed report of this study (the Regional Evaluation of Surgical Indications and Outcomes [RESIO] project),⁵ which stated that “these data are for operations on the first eye even if both were eventually done.” The results of the original PORT study, which defined the VF-14 index, also indicate that the VF-14 score may progressively decline after cataract surgery if it is performed on only one eye.⁶ However, dramatic gains in functional status were seen for patients who were rechecked after the second eye was treated. The PORT authors stated that “patients who underwent surgery in both eyes demonstrated a 1.6-fold greater improvement in VF-14, were 2.1 times as likely to report no trouble with their vision, and 2.7 times more likely to be satisfied with their vision than patients who underwent surgery in only one eye. The magnitude of the difference in improvement in VF-14 between one- and two-eye surgery appears to be attributable to both the direct effect of second-eye surgery and to a decline in VF-14 between 4 and 12 months in the one-eye surgery group.”⁶

In summary, Wright and colleagues have used a tool not designed or appropriate for determining surgical thresholds to argue that thresholds for surgery are inappropriate. They have also suggested that inappropriate surgery is being done, while omitting the obvious explanation for the results, which was hinted at in the initial study report.⁵ The authors may be correct that these kinds of evaluations can be done at a reasonable cost, but given the interpretation offered, I am not at all surprised that they found a lack of enthusiasm among surgeons.

Lorne Bellan Misericordia General Hospital Winnipeg, Man.