Study pans surgery for eye disease.

A controversial surgical procedure for a serious eye disorder not only doesn't work, it may make vision worse in some cases, warns a study published this week. Each year in the United States, surgeons perform more than 1,000 of these operations, mainly on older people.

The disorder, known as nonarteritic anterior ischemic optic neuropathy, is characterized by sudden loss of vision due to a painless swelling of the optic disc, a region at the back of the eye where the retina and the optic nerve join. The retina converts visual images to electric impulses that travel along the optic nerve to the brain for processing.

Beginning in 1989, several small studies suggested that this surgical procedure, which eases swelling, could improve vision for people afflicted with the disorder. The operation proved very popular. Nonetheless, many ophthalmologists continued to note that the surgery's worth had yet to be proved by a rigorous test.

In 1992, the federal government decided to change that by launching the "first large, randomized clinical trial looking at the safety and efficacy of the procedure," says Scott M. Whitcup, clinical director of the National Eye Institute in Bethesda, Md. A report in the Feb. 22 Journal of the American Medical Association (JAMA) describes the trial's results.

Investigators at 25 U.S. clinics recruited 244 men and women with nonarteritic anterior ischemic optic neuropathy. They randomly assigned 119 to receive surgery. The remaining 125 got no surgical treatment, but researchers monitored their vision closely.

After 6 months, 43 percent of the no-surgery group could see three or more additional lines of letters on a standard eye chart. In contrast, just 33 percent of the surgery group demonstrated the same boost in performance.

The findings suggest that many such patients improve on their own, without intervention. Previous studies had indicated that very few people would experience such spontaneous clearing of their vision.

The trial further indicated that 24 percent of the surgery group actually lost ground at their 6-month checkup, dropping three or more lines of vision. Only 12 percent of those in the no-surgery group suffered this same

deterioration in their ability to read the chart.

"Three lines of vision is thought to be a significant amount of vision," Whitcup says. For someone with perfect eyesight, it would mean the risk of flunking the vision portion of the driving test, he adds.

Last October, the Data and Safety Monitoring Committee, an independent panel of scientists and ethicists appointed by the National Eye Institute to monitor the trial, recommended halting the study. After a careful review of the data, the institute issued a clinical alert to warn 25,000 doctors of the study's findings. "The National Eye Institute would recommend that this particular operation not be performed for this condition," Whitcup says.

Ophthalmologist Leonard A. Levin of the University of Wisconsin Medical School in Madison concurs. "Why do the surgery?" he told Science News. "It definitely doesn't help."

In an editorial accompanying the JAMA paper, Levin says the procedure had gained widespread acceptance in recent years because of the desperate nature of this eye disorder. "Faced with a patient acutely blinded in one eye by a disease for which there is no known treatment, a compassionate physician might well consider any reasonable option."

The outcome of the new study underscores the value of such research. "The results from this study, although negative, bear witness to the immense power of the prospective, randomized, controlled clinical trial," Levin says in his editorial.