American Academy of Ophthalmology Web Site:
Original URL:

May 2004


Stop the IOP Zigzag

By Miriam Karmel, Contributing Writer

Intraocular pressure has long been considered a major risk factor for glaucoma progression. Now, new evidence suggests that doctors pay attention to something else: fluctuation in IOP. In fact, some experts go so far as to say that a fluctuating pressure is a greater risk factor than mean pressure.

This finding may help explain one of the more confounding situations in glaucoma practice: Why does a patient’s condition worsen, despite IOP readings in the normal range? The answer appears to be that though the patient presents with a normal IOP, he may in fact have pressures bouncing all over the place, and something about the fluctuation causes progression.

New Perspective
Sanjay G. Asrani, MD, who studied the risk associated with diurnal IOP variations in patients with open-angle glaucoma, said: “We wanted to see why people whom we feel are well controlled . . . are still losing visual field. So we took these people who we thought were well controlled for their glaucoma and looked at their fluctuations.” Dr. Asrani, assistant professor of ophthalmology at Duke University Eye Center, found that large fluctuations in diurnal IOP are a significant risk factor for progression of visual field loss.

Beware the fluctuation effect.
Caption: The danger lies not just in an elevated IOP but also
in a fluctuating one, researchers now realize. The next step:
Determining whether there is a “magic number” in IOP
variability that can be tolerated.

In the meantime, a group of researchers headed by Joseph Caprioli, MD, unmasked the fluctuation effect
in a retrospective look at data collected for the Advanced Glaucoma Intervention Study (AGIS). “Fluctuation wasn’t our primary focus, but it turned out to be our most important finding,” said Dr. Caprioli, professor of ophthalmology at the University of California, Los Angeles. How important? “I think it will change the way we look at pressure control.”

Additional Ammunition
Other retrospective analyses, using data from the Collaborative Initial Glaucoma Treatment Study (CIGTS) and from a study in Olmsted County, Minn., support the fluctuation theory. The CIGTS data, based on 607 patients with newly diagnosed glaucoma who were followed over five years, revealed that higher IOP and IOP variability over time are strongly associated with visual field loss.

The fluctuation factor emerged again, when a Mayo Clinic group looked at the Olmsted County data to find out why people went blind from glaucoma. They found that the variability in IOP readings was higher in the group of patients who were going blind, compared with persons in an age-matched group who were not going blind. Though the mean IOP on treatment was similar between the groups, the range of variability of IOP was higher in the group going blind: 9.9 mmHg vs. 6.8 mmHg.

However, these findings were incidental to the overall study of why people went blind from glaucoma. And there is a danger in retrospective data analysis, noted Chris A. Johnson, PhD, director of diagnostic research at Devers Eye Institute in Portland, Ore. The exception to that rule may be the AGIS review, headed by Dr. Caprioli. Dr. Johnson reviewed the AGIS paper for a presentation at the Academy’s Annual Meeting last fall and discovered none of the confounding factors that he normally expects to find in a retrospective analysis. The study adjusted for all possible learning effects, he said.

“Normally, I have negative things to say about studies. This was an exception,” said Dr. Johnson, who added that he would like to see the methods used to study the AGIS data set, applied to other multicenter study data sets to see how they compare.

The AGIS findings were strong enough for Dr. Caprioli and his colleagues to conclude that IOP fluctuation is a significant and stronger risk factor than mean IOP for visual field worsening. But Dr. Asrani, who is convinced that fluctuation is important in glaucoma management, cannot, on the basis of his study, say that it is a stronger risk factor than mean IOP.

In his study, 64 patients (105 eyes) with office pressure readings in the normal range used a home tonometer to monitor pressures, five times a day, for five days.

Although mean home IOP and baseline office IOP were similar, the average IOP range over the five days of home tonometry was 10 ± 2.9 mmHg. Even after adjusting for office IOP, age, race, gender, and visual field damage at baseline, diurnal IOP range and IOP range over multiple days proved to be significant risk factors for progression. “The risk is more strongly associated with fluctuations than with high values,” Dr. Asrani concluded.

“I’m not advocating that the mean pressure is not important,” he said. “The mean is important. But one has to pay attention to fluctuation.”

What to Look For
Issues of stability.
In the past, Dr. Asrani might have wanted to lower a pressure of 14 mmHg to 12 mmHg. Now, he would be satisfied with an IOP of 14 mmHg, if it were stable. “I’ll accept that,” he said, explaining how that is preferable to a zigzag course of 9 mmHg one day, then 13 mmHg, then down to 8 mmHg, and back up to 14 mmHg. “Those individuals [if in the advanced stage of glaucoma], I am absolutely worried about, because they’ll have a high chance of progression.”

Dr. Asrani added that it is more important to have a stable IOP, if it is at the target for the level of cupping. “Even if it’s a little above the target, I would go with a stable pressure rather than a lower target.” Dr. Caprioli agreed. The findings, he said, “suggest that we not only need to lower pressure, but keep it at a constant level.”

Amount of variation. How much fluctuation is allowed? There’s no magic number, Dr. Caprioli said. “That’s the next step.” But Dr. Asrani’s study found that patients with a 3.1 mmHg fluctuation or less were less susceptible to progression than an eye with a 5.5 mmHg fluctuation, which had a sixfold chance of progressing. Typically, a normal range is about 3 mmHg, while swings of 4 to 5 mmHg and higher are cause for concern. “The concern is higher for those with advanced glaucoma, because they have little reserve,” he said.

Bottom Line
“I’m convinced that fluctuation is something that is very important in glaucoma management, [but] which we have not had the chance to pay attention to until now,” Dr. Asrani continued. “The mean level is so deeply ingrained in all of us that we have to shift our focus to also see fluctuation.”

Three Tips for Tx

With fluctuation now perceived to be a risk factor for glaucoma progression, clinicians have a new set of considerations.

Here are some tips on ways to incorporate the IOP fluctuation findings into glaucoma management:

Prescribe drugs that maintain IOP at a constant level. Prostaglandins and prostaglandin analogues, which have longer duration of action, are an appropriate first choice. Prostaglandins maintain a flatter diurnal curve and thus keep pressures from bouncing around, said Dr. Asrani. For nonresponders, he’ll try Cosopt (dorzolamide-timolol maleate) or Timoptic-XE (timolol maleate gel), which is long-acting.

Whatever you do, don’t deviate from the FDA-mandated regimen. For example, brimonidine (Alphagan P) should be taken three times a day, not two. “Use them the way they’re supposed to be [used],” Dr. Asrani said. In the meantime, improved drugs could be in the pipeline, because the drug companies are interested in this finding of fluctuation as a risk factor, Dr. Caprioli said.

Consider surgery. It’s more likely to keep pressure at a constant level, Dr. Caprioli said. Anecdotal evidence supports this, he said, noting that patients who maintained low pressures on multiple medications, yet got worse, improved after surgery. “We then operate on them, get a pressure no better than before surgery, but progression stops,” Dr. Caprioli said. “I attribute it to a reduction in fluctuation.”

Dr. Asrani concurred, noting that, in one small study, fluctuation after trabeculectomy flattened out completely.1

Schedule appointments at different times of day. If you see a patient in the morning, schedule the next appointment for the afternoon, Dr. Asrani advised. While Dr. Caprioli agreed, he cautioned, “One every six months is not going to get it.” A large number of measurements (20 to 30) are needed to get some estimation of fluctuation, which is why the AGIS data—collected twice a year for 10 years—were so valuable, he said.

1 Wilensky, J. T. et al. Trans Am Ophthalmol Soc 1994;92:377–381.

Dr. Johnson receives research support from Carl Zeiss Meditec, the manufacturer of the Humphrey Field Analyzer visual field device. Drs. Asrani and Caprioli have no related interests.