P.D. vs Visual Axis Distance

[Dr. Bill Stacy]

There are 2 instruments I use regularly where I can see a corneal reflex that I believe corresponds to the visual axis or line of sight. The simple one is the pupillometer where the reflex is a small dot of light generated by a small fixation light. The more complex one is an autorefractor/keratometer/topographer where the reflex is a little box that I believe is computer generated and seems to also correspond to the visual axis, assuming the patient is actually looking at the target and there is no eccentric fixation (which is pretty certain if acuity is good and the macula is intact). Oh, and yes, the autorefractor also gives a p.d. reading.

I was wondering if both instruments actually are measuring the visual axes separations as opposed to the pupillary distance, which I would usually think of as the distance between the (centers of ) the pupils. At least that's how I understood it many years ago before monocular p.d.s existed, PALs were a figment of someone's imagination, and really the only thing we had was a mm rule.

So maybe someone can bring me up to date on this. Should we call it something like "visual axis separation at the corneal plane" or something like that rather than "P.D." or what? What brought this subject to my attention today was a case where both reflexes on both instruments were grossly offset from the centers of the pupils. Like 3mm out and 1mm down o.u. I didn't think to do it at the time, but I'm guessing my regular ruler P.D. would be about 6mm off from his visual axes separation. When he comes in for dispensing, I'll check that. Meanwhile any thoughts on this arcane subject?

[Barry Santini]

Both instruments are founded on the assumption that the corneal reflex is coincident with the intersection of the visual axis eith the cornea. This is because it is also assumed that the CC of the corneal and the first nodal point of the VA are coincident, or very close. This has proven to be good set of working assumptions, except when it ain't. Then, only a subjective measurement, rather than an objective one, will do.

B

[Dr. Bill Stacy]

OK my geometric optics are a bit fuzzy, but I should be able to understand what you're saying. What is the "CC of the corneal"? That's new terminology to me.

[Barry Santini]

CC = Center of Curvature

[Dr. Bill Stacy]

I think my use of CC (corneal curvature) is more commonly used than "center of corneal curvature" which is kind of archaic and not clearly defined in a brief lookup that I did. I do understand nodal points but since both loci really are points that I cannot see or measure it's kind of moot. Now I can see and locate the corneal reflex and have to assume it is very close to coincident with the point at which the visual axis intersects the cornea. It is the standard for placement of the fitting cross on PALs and for locating OC heights. I've seen your posts on subjective p.d. measurements and was wondering if the instrumentation is described anywhere or is it under development. I have no idea how it would be done, at least in general terms.

[Barry Santini]

Tom Clark, OD, sells a modified Essilor pupilometer that allows the subject to move the sliders and take their own PD.

Very good!

And very reasonable.

B

[Dr. Bill Stacy]

Tom Clark, OD, sells a modified Essilor pupilometer that allows the subject to move the sliders and take their own PD.

Very good!

And very reasonable.

B

I'm assuming the thing uses mirrors or prisms to allow the patient to see what we see through the pupillometer, namely his eyes, one at a time, and he moves the little line to overlap the corneal reflex. Is that correct? If so, then it isn't "subjective" at all; it's a "self measuring " objective device measuring what we measure without the mirrors/prisms. Or is there something else I'm missing?

[Barry Santini]

Nope. The testee actually is lining up the sliders with the light source, instead of the ECP.

Completely subjective. Like the Blink/Eye Netra device.

B

[Dr. Bill Stacy]

If that is actually what the testee/testor (in this case it's one in the same) is doing, then it is still an objective measurement rather than a subjective impression. If I measure my own 2x4 instead of having the carpenter do it, it's still an objective measurement, unless I'm just taking a look at the board without the tape, in which case it would be not only a subjective determination, it would be a complete guesstimate and pretty worthless. There of course remains the accuracy question. But I'm still having trouble with the patient being able to fixate the light source, which is placed at optical infinity and the measuring hatchmark, which normally is near the corneal plane where it would be so far out of focus as to be unusable. If he puts the hatchmark into focus, the fixation light is then WAY out of focus and subject to a huge variation in placement. Unless I see some reason to indicate that this thing might actually work, I'm sticking with the industry standard. (am i right, there is no on line diagram as to his "invention"?)

[Uilleann]

I'm curious what the measured percentage of the population exhibits a visual axis / corneal reflex deviation greater than ANSI tolerance for a spectacle lens. And what is that deviation in mm?

[Dr. Bill Stacy]

I'm curious what the measured percentage of the population exhibits a visual axis / corneal reflex deviation greater than ANSI tolerance for a spectacle lens. And what is that deviation in mm?

I think ANSI says about 1 mm per eye. The reason I started this thread is I noticed a huge decentration of the autorefractor target (lock on device, similar to a fighter plane's lock on device), and am accustomed to seeing a mm or so of devation of this target from the center of the pupil, which are both plainly visible. This time the little box was nearer to the pupil margin than to the pupil center, I estimated about 3 mm. I got to thinking about this issue and intend to re-measure his monocular p.d.s to see if I get, as expected, that same deviation. If so, I'll double check with a mm rule which will be to the center of the pupils per my customary technique of outer limbus to inner limbus measurement. I'll report back what I find, but I'm guessing I'll be discarding any and all mm rule p.d. measurements in favor of what I believe to be much closer to his actual visual axes separation.

[Barry Santini]

I'm curious what the measured percentage of the population exhibits a visual axis / corneal reflex deviation greater than ANSI tolerance for a spectacle lens. And what is that deviation in mm?

Studies have been done that show up to 25% up the population deviation more than 1mm monocularly, and as much as 2.5mm.

B

[Dr. Bill Stacy]

Could you site a study? And how did these studies determine the difference between visual axis and corneal reflex?

[Barry Santini]

Could you site a study? And how did these studies determine the difference between visual axis and corneal reflex?

Here's one:

[Dr. Bill Stacy]

Great, except that who has a synoptophore or orbscan laying around for P.D. measurments? Seriously, these were comparing angles lambda and kappa using sophisticated techniques, not measuring differences between the corneal reflex and the point of visual axis intersection on the cornea, which is what we were talking about.

Apples and oranges.

[Barry Santini]

Great, except that who has a synoptophore or orbscan laying around for P.D. measurments? Seriously, these were comparing angles lambda and kappa using sophisticated techniques, not measuring differences between the corneal reflex and the point of visual axis intersection on the cornea, which is what we were talking about.

Apples and oranges.

Actually doc, they were! And they were detailing the greater consequences of not relying on corneal reflex for ablation centration.

B

[Dr. Bill Stacy]

So are you advocating that we specify angles kappa or lambda when we order PAL's? Or do we have to calculate the decentration ourselves? Naw, forget it. I'm not buying an orbscan that's for sure. And I'm really really skeptical of that "subjective" pupillometer. For now I'm just going to crosscheck my good old mm rule with the pupillomter and the autorefractor.

[Lelarep]

What's the benefit of a patient subjective pupilometer vs. an objective pupilometer as read by someone trained to do so?

[Barry Santini]

What's the benefit of a patient subjective pupilometer vs. an objective pupilometer as read by someone trained to do so?

The theoretical assumptions underpinning the use of the corneal reflection for visual axis measurement are not valid for a significant percentage of the population..B

[Dr. Bill Stacy]

I found Dr. Clark's description of the problem in a 4 year old article on the subject located at:
http://optometrytimes.modernmedicine...-axi?page=full

I think he makes a huge error in defining angle kappa as the problem with the standard reflection pupillometer. This would only be true if the instrument were measuring the pupillary axis, which it isn't. I don't know how close the pupillary reflections are to the visual axis, but I think it is very, very close. The pupillary axis is defined as the line THROUGH THE CENTER OF THE PUPIL THAT IS NORMAL TO THE CORNEAL SURFACE. That is not what pupillometers measure, as anyone who has ever used one can plainly see. The corneal reflex is almost ALWAYS decentered from the pupil center, so it CANNOT be located on the pupillary axis.

[Barry Santini]

Doc, your are still missing the main point: The VA is not reliably located by the CR for a significant % of the population

[Dr. Bill Stacy]

Neither you nor Dr. Clark have proven that at all. Referencing the pupillary axis is significant in refractive surgery, but it has nothing at all to do with pupillometry.

Dr. Clark has driven you into a dead end street. It's no wonder there no such instrument being produced. It's another case of a little bit of knowledge getting someone in trouble.

[MakeOptics]

Using a corneal reflex pupilometer the distance measured is the coincidence of the purkinje reflections from the surfaces, the first surface is the cornea and the brightest reflex, the second surface is the back of the cornea and often indistinguishable from the first surface, the third and forth are reflections from the crystalline lenses front and back surface respectively. Knowing this we can get a close approximation of the visual axis, but any irregularities on any of the four surfaces can lead to erroneous measures. Even the new tech in offices has to make some assumptions so the PD is the best centration measure we have and has been descretized as the measure between the visual axii's, for ophthalmic purposes it's the best substitute.

[Dr. Bill Stacy]

I have enough trouble seeing the first purkinje in a pupillometer to be looking for the 3rd and 4th. As you say the 2nd will not be visible as its really part of the 1st in a pupillometer. And what would I do with it if I could see it?

[Uilleann]

The only time I tend to see a difference to speak of with the 1st and 2nd, and 3rd & 4th reflexes, is with post Sx IOL pts. Which of course have those lovely IOLs that wiggle and jiggle all over the place. But even then it's rarely more than 0.5 - 1.0 mm away from the primary reflex in the CRP. It would seem that even if this scenario was indeed as 'significant' as the data has lead some to conclude, that it would require something of a 'perfect storm' in optical terms for a pt to perceive it at all, let alone experience significant discomfort or all out non-adapt to a given lens placement.