crystal ball is not really needed to diagnose glaucoma but I think that sometimes it might help.[*] In fact, given the usual criteria for such a diagnosis — a crystal ball might be thought essential. Why do I say this? Let's look at the situation facing the ophthalmologist when dealing with the 3 criteria necessary to make a diagnosis:

Elevated IOP

The Academy of Ophthalmology lists an IOP of 23 mm Hg as ‘suspicious’ for the presence of glaucoma. But this value has been ratcheted upward over the years which demonstrates that even this august body is unsure of what value constitutes glaucoma. And why not? By the late 1960's and early 1970's, it had become evident that only about 10% of people with an IOP of 22 mm Hg or more would develop glaucomatous damage. We've already pointed out (<What in Blazes is Glaucoma?) that elevated IOP is not glaucoma. Elevated IOP is only a risk factor and is not prognostic. Whilst it is true that statistically, patients with elevated IOP's at the time of diagnosis seem to lose vision much more rapidly than those who didn't — we're right back to: what's an elevated IOP?

What about 30 mm Hg? Surely that's a patient with glaucoma? Nope. In the absence of the other 2 criteria, this patient is classified as having ocular hypertension (OHT). Well ... how about 40 mm Hg? Is it glaucoma? Maybe not; depends on how long that IOP has been that high. And even in the presence of the other 2 indicia — what else is going on? Are cells and flare present? Has this eye been injured? Is the patient on steroids? And in the absence of the indicia has the patient just guzzled a six-pack? Danced with wolves?

Should we lower the IOP in such an individual? Yes, because the shut-down pressure for the central vascular system of the eye is 45 mm Hg. At 40 mm Hg the blood flow is being compromised even if the optic nerve fibers could withstand this level of IOP for some hours or days. Interestingly, even with induced IOP's of 90 mm Hg, some choroidal pulse has been recorded.

More than 40% of ‘glaucoma’ patients have a ‘normal’ IOP at the time of diagnosis which now begs the question: what's a normal IOP? From a functional standpoint, a ‘normal’ IOP is one that does not result in glaucomatous optic nerve head damage.[**] Because not all eyes respond similarly to a particular IOP level, a normal pressure cannot be represented as a specific measurement. Therefore, the most that we can expect is to determine their relative chance of developing glaucoma at different pressure levels given the knowledge of the distribution of IOP in general populations and in populations of individuals with glaucomatous damage.

Optic nerve cupping

What if the patient has a large amount of cupping but the IOP is ‘normal'. Weeel ... considering what I said about the IOP in general I'll add that an enlarged cup doth not glaucoma make either. That's because a certain number of our citizens are traipsing about with no idea whether they have glaucoma or not by this criteria. To put it another way: it's normal for some people to have optic nerve cups that are larger than ‘normal’; perhaps we should say ‘than average’ (see also Optic Disk Changes).

Figure 1. optic nerve cupping

There is no standard pattern for the development of glaucomatous cupping of the optic nerve head (ONH) or disc. This cupping may begin as a symmetrical enlargement of the physiological cup, but usually some portion of the rim thins more rapidly than another. It is frequently the inferior temporal or — less commonly — the superior temporal rim that becomes thinner first. Occasionally this thinning is localized and seen as a notch or less frequently a pit at the disc rim. If notches are present both inferiorly and superiorly, the cup becomes vertically oval like a football.

What about if the IOP is elevated as well? Elevated? What does that mean (see above)? If by elevated you mean over 23 mm Hg then I can only say that I would drop this patient on the ‘suspicious’ pile, schedule frequent visual fields (VF's) and move on. That's because the jury is out on the significance of enlarged cups all by their ownsome. However, for now, progressive optic disc cupping without visual field loss is considered an early indicator of glaucoma, assuming no other disease process is occurring.

Visual field(s)

Here we're on more solid ground but not if we're just talking about an enlarged blind spot or just a generalized suppression of retinal sensitivity. Some would arm wrestle with me about the latter believing as they do that suppression of overall sensitivity is an early sign of glaucoma. But it isn't because it very much depends upon the clarity of the lens. Thus one might see considerable fluctuation in sensitivity as well as blind spot size in a diabetic not in good control. Not to mention an individual with early cortical cataract changes. And then there's the fact that certain medicatiuons reduce retinal sensitivity ...

How about nasal notching (been doing your homework I see)? Even here I'd want to see that change persist over several fields taken over a few months. Because even the visual fields can change back and forth and can sometimes be influenced by medication.

So what's a person to do? Well, in our infinite wisdom, we've decided that it generally takes all three to be present — elevated IOP, enlarged optic nerve cup, and a characteristic change in the VF (though I really lean on the latter lacking else) — before we reach for our prescription pads and even then we may wish we could employ the services of our old crystal ball. You'll forgive me for being flippant, but sometimes it's enough to make a grown man weep!

Is it any wonder that general ophthalmologists tend to go by the IOP alone? So what's the harm in that? We're playing it safe. Right?

Wrong! There's plenty of potential harm in that practice and we're not playing it safe. We'll talk about that in the next section.

* I used to have two crystal balls on my desk. The second was in case I needed a second opinion.

** This answer is akin to the question: ‘How long is a piece of string?’ The answer: ‘Twice the distance from the center to the end,’ is not all that definitive.