Reduction of intraocular pressure is currently the only known treatment for glaucoma. How aggressively we work to lower IOP is determined, at least in part, by the risk of further progression of disease. However, IOP is variable, and more than one IOP parameter could potentially be assessed and modified—at least in theory. Here, I’d like to discuss the pros and cons of different IOP-related parameters, in terms of their potential usefulness in the clinic.

Mean IOP vs. Peak IOP

The IOP parameter that has been most extensively characterized is mean IOP, which has been the primary parameter assessed in most of the large clinical trials looking at risk factors for progression, including the Collaborative Normal-Tension Glaucoma Study, the Advanced Glaucoma Intervention Study, the Early Manifest Glaucoma Trial and the Ocular Hypertension Treatment Study. The data from those studies confirmed that mean IOP is a very important predictor of both the risk of glaucoma progression and the risk of developing glaucoma in patients who have ocular hypertension.

Ironically, although most of our clinical trial data tells us about the significance of mean IOP, most of us are probably basing our clinical decisions more on peak IOP than mean IOP. In a typical busy practice, we’ll create a pressure target—a peak IOP that we want the patient to remain below—and then we’ll check the patient’s pressure periodically as he or she comes in for routine visits. If the measured IOP is above the target IOP, that’s when we tend to react and alter our therapy. We may recheck the IOP to ensure that the measurement was not anomalous, but we are still reacting to a peak IOP and not the mean.

Unfortunately, we have much less objective information about the clinical significance of peak IOP than we do about mean IOP. Peak IOP is not what the clinical trials were originally designed to look at (although peak IOP has been looked at retrospectively). This doesn’t mean that mean IOP is more important than peak IOP; we simply have more data regarding mean IOP.

In any case, the practical reality is that when you can only take one measurement every few months, thinking in terms of peak IOP rather than mean IOP makes more sense; it allows us to proceed despite the limitations of the data we have available to us. That’s why a typical clinical practice operates that way, and that tendency is unlikely to change until we have tools that give us better data.

The Fluctuation Factor

Fluctuation of IOP as an independent risk factor for glaucoma is an area that has been of significant interest recently. Many patients with seemingly good IOP control (as measured in the clinic) continue to get worse, and IOP fluctuation is a plausible contributor to this. Although the large clinical trials have provided extensive useful data regarding mean IOP, they were not designed to reveal whether any other IOP parameters might be useful as well. Nevertheless, retrospective data analysis to determine the significance of IOP fluctuations has been performed on the data from most of the major clinical trials in glaucoma.

Unfortunately, research into the significance of IOP fluctuation hasn’t produced a lot of conclusive information so far, for two reasons. First, there are many levels of fluctuation and multiple ways to analyze them. Second, the technology to monitor fluctuation is still in its infancy, so the data we have to base predictions on, and our ability to collect data, are very limited.

One issue is the scale of time over which we look at fluctuations. We know from animal studies that fluctuations may be occurring almost constantly on a second-to-second and minute-to-minute basis. Larger-scale fluctuations also occur over the course of a 24-hour period, as has been documented in animals as well as 24-hour human sleep-lab studies. In addition, there are even longer fluctuations that occur over weeks to months, and others that occur over years. The relative importance of these different levels of fluctuation is not yet clear.

At the minute-to-minute level, data collected using implantable transducers in rabbits and monkeys demonstrate that pressures fluctuate very significantly over very brief periods of time, and do so almost constantly.1,2 The clinical significance of these short-term fluctuations, however, is unknown. Continuous 24-hour IOP monitoring is one of the holy grails of glaucoma research, and there have been notable technological advances over the past few years. Nevertheless, devices for routine clinical use that can deliver the type of IOP data we can currently gather from animals are not yet available. Until we can measure these fluctuations in humans, their significance in the development and progression of glaucoma will likely remain unclear.

We do have some limited data regarding the impact of diurnal (daytime) fluctuations. One study done by Sanjay Asrani, MD, for example, looked at diurnal fluctuations in IOP measured using home tonometry and found that glaucoma patients who had larger fluctuations over the course of a day were more likely to progress.3 But home tonometry is not available to most patients, and those study results should be repeated using other measurement methods.

In terms of circadian variability, it’s well-established based on the sleep laboratory research of Robert Weinreb, MD, and John Liu, PhD, and colleagues, at the Shiley Eye Center, University of California San Diego, that there’s a rise in intraocular pressure when patients lie down to sleep. (For example, see figure above.) However, the magnitude of that rise is greater in normal patients than in glaucoma patients, so the difference between diurnal and nocturnal IOP is unlikely to be an independent risk factor for glaucoma. However, systemic blood pressure is lower at night in most people, and this could combine with a higher IOP at night to compromise blood flow to the optic nerve. For now, whether or not the nocturnal IOP rise is an independent risk factor for glaucoma remains to be demonstrated.

The Large-scale Trials

Most of the published research concerning IOP fluctuation as a risk factor for glaucoma has examined long-term fluctuations over months and years, using the data from the large clinical trials. However, what those studies have found is somewhat contradictory. Several papers have been published based on data from the AGIS study, for example. The first paper found that patients with larger standard deviations in IOP between visits were more likely to progress, using visual field scores as a measure of severity. However, a subsequent analysis took into account the fact that patients who have glaucoma progression probably have a change in their therapy in order to lower their pressure, a change that would automatically create long-term IOP variability. Once they took that into account, it was only patients who had low mean IOPs who seemed to have IOP fluctuation as a risk factor for further progression.

The reason for that is not really clear. One explanation might be that we’re thinking about IOP variability in the wrong way. Most studies look at IOP variability in terms of things like range and standard deviation. But if you think about IOP variability in terms of a percentage change instead of an absolute change, then at a high pressure, say a pressure of 30 mmHg, a 2-mmHg fluctuation is a pretty small percentage change up or down. On the other hand, if you’re at 10 mmHg and you have that same 2 mmHg variability, that’s a large percentage change. That 2-mmHg fluctuation may actually have a greater effect in terms of movement of the lamina cribrosa and deformation of the optic nerve head at 10 mmHg than at 30 mmHg.

Other studies have also looked at the impact of long-term inter-visit IOP fluctuation. The CIGTS study compared medical treatment for glaucoma vs. surgical treatment; IOP fluctuation data from that study indicated that IOP fluctuations were a risk factor for progression in medically treated patients but not in surgically treated patients. That appeared to be true even though the magnitude of the fluctuation the patients experienced was pretty much the same in the two groups. Again, an explanation for these findings is not clear.

Another related, as yet unpublished finding was presented at the American Glaucoma Society Annual Meeting last spring by Mae O. Gordon, PhD, based on data from the OHTS trial. That data indicates that IOP variability in the observation group—those not being treated—did not correlate with an increased risk of converting to glaucoma. But in patients who were being treated medically, IOP variability did increase the risk of converting from ocular hypertension to glaucoma.

The idea that medical treatment may somehow potentiate the effect of IOP fluctuations on progression is interesting, but again, we don’t have a good explanation for these findings at this time.

Maximizing Measurements

With better technology in the future, we should be able to gather much more IOP-related information. In the meantime, there are a few things we can do to take advantage of what we currently know about mean and peak IOP and the possible impact of IOP fluctuation.

First of all, we can try to maximize the value of the in-office measurements we take today with a few simple strategies:

• Take IOP measurements of each patient at different times of day. Research has clearly shown that a single IOP measurement is not a good indicator of the patient’s IOP pattern over a 24-hour period. Work done by Tony Realini, MD, for example, looked at the repeatability of pressure measurements taken at a specific time of day on different days, and how well they reflect diurnal patterns.4.5 That data showed that one measurement is really not predictive of subsequent IOP measurements taken at the same time of day. So, doing multiple measurements at different times of day would be helpful.

• Take IOP measurements more often. This would allow us to get a better sense of the patient’s mean IOP. Of course, taking more measurements would also involve getting the patient into the office more frequently, which would be difficult when we’re all seeing more and more patients in a shorter period of time. (It would also be more of a burden for the patient.) And, it might raise questions regarding reimbursement. So practical considerations make this a challenge to implement.

One change that might help would be the development of a means for patients to accurately measure their IOP at home. Some instruments that don’t require an anesthetic have been tried for home use (such as the air-puff tonometer and rebound tonometer); however, using them at home involves significant cost for the patient and may not be suitable for patients with limited dexterity. And even though this type of arrangement would be better than having one measurement every three or six months, it still might not give us the information we really want. What we really need is a good profile of what the IOP is doing over the course of a 24-hour day, and over multiple days.

To date, we don’t have conclusive proof that fluctuating IOP impacts a patient’s risk of progressing, but I believe there’s enough suggestive evidence that fluctuation may be a risk factor for glaucoma—at least in some populations—that it’s worth tailoring therapy to minimize fluctuation, as long as there’s minimal risk to the patient.

Here are a few things we can do to accomplish this:

• Consider tailoring medications to minimize fluctuation. For example, certain medications provide smoother IOP control; any medications that suppress aqueous production will cause larger IOP fluctuations throughout the day, just because they reduce fluid flow but don’t change the flow resistance. In that situation anything that would normally perturb your IOP anyway, such as drinking a bottle of water, will cause your pressure to go up, and the rise will be larger than if you were on a different medication that improved outflow facility. (See diagram, facing page.) For that reason, using a medication such as a prostaglandin analog, which enhances aqueous humor outflow, might provide a better quality of IOP control—and less fluctuation—throughout the day.

You can also choose to avoid medications that don’t work at night. Drs. Weinreb and Liu and colleagues at the Shiley Eye Center have produced a large body of work from their sleep laboratory demonstrating that not all medications are effective at night. They have demonstrated that beta blockers and alpha agonists have minimal efficacy at night, while prostaglandin analogs and carbonic anhydrase inhibitors continue to have good nocturnal efficacy, although less than during the daytime. If the choice of medication doesn’t result in increased risk or prohibitive cost for the patient, selecting prostaglandin analogs or carbonic anhydrase inhibitors may benefit the patient by helping to limit 24-hour IOP fluctuation.

• Consider performing laser trabeculoplasty. One of the less-often discussed results of using a procedure such as selective laser trabeculoplasty is that IOP fluctuation is reduced.6,7 The likely explanation is that it enhances outflow facility, leading to a more consistent IOP over the course of the day. In addition, it’s a very low-risk procedure. Given that it smoothes out IOP fluctuation, it makes sense to try it in patients whose IOP seems well-controlled but who continue to get worse.

In the United States, clinicians still don’t perform SLT very frequently, often because of the perception that it doesn’t reduce IOP as much as other alternatives.I believe that perception is partly the result of the frequent use of SLT as a last resort. Once a patient is already on maximum medical therapy, even adding another drop will have minimal effect, so it shouldn’t be a surprise that SLT doesn’t cause a major change under these conditions.

In my experience, you’ll see a larger effect if you perform SLT earlier in the course of treatment, before the patient is on multiple medications. I think it’s a reasonable option to offer earlier in the treatment spectrum, not only to reduce IOP, but also to achieve a better quality of IOP control—i.e., reduced fluctuation—than you might get with something like a beta-blocker.

• Consider the impact of body position. In clinical situations we virtually always measure pressure with the patient sitting in a chair at a slit lamp, but it’s long been known that pressure varies with body position. Certainly IOP increases when we lie down. In fact, some studies have found that measuring pressure when the patient is supine is somewhat predictive of IOP at night. (See diagram, p. 105.)

Our group did a study in which we measured IOP in different body positions: sitting upright or sitting with neck flexed or extended, lying on your back, or lying on your side.8 It turned out that just about any body position results in a pressure that is higher than when you’re sitting upright at a slit lamp.

One useful ramification of this is that we can advise our patients to be aware of this factor. Certainly we all need to lie down and sleep, but individuals with glaucoma can try to avoid sleeping face down; sleeping face down can produce much higher intraocular pressures, not just from the body position, but also because of compression against the eye. I also advise patients who do yoga not to do head-stand positions, and patients who have highly asymmetric disease and sleep on their side may benefit from sleeping on their back or on the side that puts the worse eye above the better eye. Sleeping in a head-up position can also reduce the nocturnal IOP peak. However, the clinical benefit of adopting or avoiding specific head and body positions is currently unknown.

No Guarantees

Unfortunately, we don’t have any conclusive evidence that reducing fluctuation will prevent or minimize progression, and doing a controlled, randomized study to demonstrate this would be difficult. We know from existing studies that we have to manage mean IOP; it’s unethical not to. So anyone with glaucoma in the study would have to be treated to lower the mean IOP. If both groups are lowered to the same level, any remaining difference caused by fluctuation might be tiny in comparison and difficult to demonstrate, especially given today’s imprecise means of measuring fluctuation. That means the study would have to be very large to reveal a significant difference—assuming a difference appeared.

In the meantime, we can’t even be certain what the variability we encounter in the clinic really means. Most of the IOP fluctuation that we’re measuring may just be because of the time of day, or because someone had a cup of water right before they came in for their exam. So sorting out what is really reflective of IOP variability over a 24-hour period is going to be difficult. This will hopefully change as new technology for continuous monitoring of IOP emerges, but for now we have to operate on the assumption that fluctuation may be important to a patient’s prognosis, and do what we can to minimize it when the risk to the patient is acceptable. REVIEW 

Dr. Sit is an associate professor of ophthalmology at the Mayo Clinic in Rochester, Minnesota.

1. McLaren JW, Brubaker RF, FitzSimon JS. Continuous mea-surement of intraocular pressure in rabbits by telemetry. Invest Ophthalmol Vis Sci 1996;37:6:966-75.
2. Downs JC, Burgoyne CF, et al. 24-Hour IOP Telemetry in the Nonhuman Primate: Implant System Performance and Initial Characterization of IOP at Multiple Timescales. Invest Ophthalmol Vis Sci 2011;52:10:7365-7375.
3. Asrani S, Zeimer R, Wilensky J, Gieser D, Vitale S, Lindenmuth K. Large diurnal fluctuations in intraocular pressure are an independent risk factor in patients with glaucoma. J Glaucoma 2000;9:2:134-42.
4. Realini T, Weinreb RN, Wisniewski SR. Diurnal intraocular pressure patterns are not repeatable in the short term in healthy individuals. Ophthalmology 2010;117:9:1700-4. doi: 10.1016/j.ophtha.2010.01.044. Epub 2010 Jun 16.
5. Realini T, Weinreb RN, Wisniewski S. Short-term repeatability of diurnal intraocular pressure patterns in glaucomatous individuals. Ophthalmology 2011;118:1:47-51. doi: 10.1016/j.ophtha.2010.04.027. Epub 2010 Aug 14.
6. El Mallah MK, Walsh MM, Stinnett SS, Asrani SG. Selective laser trabeculoplasty reduces mean IOP and IOP variation in normal tension glaucoma patients. Clin Ophthalmol 2010;4:889-93.
7. Lee AC, Mosaed S, Weinreb RN, Kripke DF, Liu JH. Effect of laser trabeculoplasty on nocturnal intraocular pressure in medically treated glaucoma patients. Ophthalmology 2007;114:666–670.
8. Malihi M, Sit AJ. Effect of head and body position on intraocular pressure. Ophthalmology 2012;119:5:987-91. doi: 10.1016/j.ophtha.2011.11.024. Epub 2012 Feb 17.
9. Modified from Liu JHK, Zhang X, Kripke DF, Weinreb RN. Twenty-four-Hour Intraocular Pressure Pattern Associated with Early Glaucomatour Changes. Invest Ophthalmol Vis Sci 2003;44:1586-1590.
10.  Brubaker RF. Targeting Outflow Facility in Glaucoma Management. Surv Ophthalmol 2003;48:4:S17-20.