It seems to me that Martin is talking about better clinical
measurements, as opposed to research methods.

DrG

On 10/20/06 5:43 PM, in article
Xns9862D2C019612scottseidmanmindspri@130.133.1.4, "Scott Seidman"
<namdiesttocs@mindspring.com> wrote:

Actually, my good friend Joel Miller is NOT an engineer. He is a vision
scientist. Demer is an engineer and a pediatric ophthalmologist.

I am also an engineer (electrical engineer by training and practice) also,
as well as a pediatric ophthalmogist, particularly regarding strabismus. I
was the one who worked with Miller on muscle sideslip, and I hand-fabricated
the first "buckle strain gauge" and did the implantation surgey in the late
80's. Millers big work is ORBIT, a computer program that simulates
strabismus and strabismus surgery. The data it uses is based on human and
monkey orbital and muscle data. The problem with using it to model a
particular patient is that you do not have individual's data regarding the
many parameters and measurements for the model, so one uses the averaged
data. He originally worked with David A. Robinson at Hopkins, where I met
him during my Wilmer strabismus fellowship. I got interested in his
simulation work and followed him out to San Francisco to the
Smith-Kettlewell Insitute, and collaborated with him for 6 years.

The modeling heps one understand certain situations, and to simulate certain
uncommon conditions in order to help understand them, as mentioned above.
However, surgery on a simple esotrope or exotrope is not aided by such
modeling, because it still is not modeling the particular patient. Also, the
model is basically biomechanical, based on the mechanical properties of the
muscles and the orbit, and cannot take into account the particular neurology
of the eye aiming, and cannot fathom the complexities of changes in the
nerve signals.

Clinical strabismus angle measurements are not done with a steel ruler - I
don't know what the poster was describing. One uses prisms and cover
testing. The prisms may be loose, or a prism bar (which he may have confused
with a steel ruler?) This is the standard method of testing. The other
methods mentioned are used in research - scleral search coils are the
preferred techniques with monkeys, for example. Video oculography and IR
oculography are good for measuring eye movements, esp. dynamic, but not so
good at alignment. And, no one uses these clinically in the office.

My main interest at the Institute was in functional electrical stimulation
of eye muscles, but it is impractical, and is not useful for the temporary
realignment as described. If the eyes are capable of almost aligning at
times, then the problem is the control, not the muscles. Someone in LA at a
"famous eye institute" is working on such electrical stimulation right now.
(I gave him my original experimental data, which is mostly unpublished.)

I have always said that someday history books will look back at eye muscle
surgery and laugh at it, because for the most part, the muscles are
innocent. [[This brings to mind an old Star-Trek movie, where the crew comes
back to the 1980's or so. Checkhov hits his head and has a subdural
hemorrhage. He is in the operating room awaiting neurosurgery to drain the
blood. Bones gets the OR team out, goes in, and waves his instutment over
Checkhov, who gets up off the table and is fine. Bones makes a comment,
something like "Barbarians! What were they thinking? They were going to
drill holes in his head!" (The current standard treatment of subdurals is
burr holes to drain the blood.)]]

It is the brain sending the wrong signals, but we still don't know why it
happens, and what we can do to change it. Scientists at Smith-Kettlewell and
other institutes have been studying oculomotor control for many years, and
still don't have the answer. In exotropia, the divergent muscle signals are
too strong relative to the convergent muscle signals.

Botox does move the yes to a new alignment position, and if the brain wanted
to then keep it centered, it would. Typically, constant exotropes who are
injected drift right back out once the Botox has worn off, unless they are
intermittent exotropes with good control. In that case, the Botox may make a
moderate term of improvement, but it is still not usually permanent. It
depends on the degree of fusion that remains. Surgery is also often not
permanent. Variable exotropes have the lowest long-term success rate with
surgical correction than almost any other type of strabismus.

As to cosmetic success vs functional success - if the eyes are aligned into
a better position, then IF there is fusion lurking in the brain, it will
hold the eyes straight later on. Any stereo that is there will also make
itself known, but there is not real technique to bring out stereo that is
not there. It develops in early infancy, and may not have developed due to
genetic/brain reasons, rather than being caused by having poor eye alignment
from the get-go. So if surgery is done, the aim is to get the eys in the
rnage where alignment and fusion could take place if it is there, which is
in the range where it is also cosmetically acceptable, usually within about
8 prisms of truly straight. In that range, if there is fusion, the remaining
angle will be taken care of by the brain. Also, if there was intermittent
exotropia, simply reducing the angle make greatly improve control - the
surgical eye alignment can err on the exotropic side significantly and still
markedly improve the control. Once usually aims for an initial
over-correction, aiming the eyes in (esotropic) about 8-15 prisms, right
after surgery, since the yes tend to go out some as it heals. However, if
the angle remains esotropic, you get double vision. And if the child is less
than 5 years old or so, they may develop supression from the new esotropia,
and lose whatever fusion they had - so being a well-controlled intermittent
exotrope is preferable to being a small-angle constant esotrope.


David Robins, MD
Board certified Ophthalmologist
Pediatric ophthalmology and adult strabismus subspecialty

I've seen this in some of the papers I gone through.

I designed a set of goggles with Solidworks; got them SLA printed and
fitted them with 2K x 2K IR-only cameras as well as a set of LCD
shutters and front-surface partial 45 degree mirrors.  The goggles also
have IR and visible-range illuminators aimed at each eye.  Each camera
has an optical filter that passes IR only.  The LCD shutters (one per
eye) allow you to occlude each eye as desired (or both, I guess).  By
interpolation you could get some pretty accurate measurements of x, y
and theta.

I learned a lot while watching him play, read, etc.  Not enough to go
much past observation, more than likely due to the fact that I am not
trained in this area at all.  What I do know is that each eye works
exceedingly well when called upon to do so (by occluding the other
eye).

Thanks for the other info.

-Martin

Yup, we've been there.  They recommended surgery.

-Martin