Dr. Bernard Grolman, D.O.S. Interview on his invention at AO - The Non-Contact Tonometer

AO NCT ~1975

My name is Dr. Bernard Grolman, my mother always called me Bernard, my friends call me Bernie.  I started working at American Optical's main plant in Southbridge Massachusetts in 1955… that was the year of the great flood.  As a matter of fact I thought that my career at American Optical was going to be short lived because I was only there a little while when the Quinebaug river overflowed and some of the AO buildings were under water up to the 3rd floor.  There were serious doubts that manufacturing operations could ever be resumed. However, by round-the-clock manual labor extending over a period of several months, this seemingly impossible task was achieved.  So I worked at AO and now Reichert from 1955 until I retired in 1985… 30 years.  And without the loss of a day I continue as a consultant to the company.

The beginnings of the NCT
Anyway, regarding the NCT… In not too many instances is the need for a specific kind of instrumentation so obvious as it was with the non-contact tonometer.  Optometrists, at the time (late 1960's) could not perform Goldmann tonometry because it required the use of a topical anesthetic.  Only physicians with medical degrees were permitted by law to use topical anesthetics and one could not measure intra-ocular pressure with out the use of a topical anesthetic.  So large segments of the population were denied the protection that results from detection of Glaucoma.  And it was obvious to me that a solution to this problem was needed.  Now, it had occurred to other people that the use of air and non-physical contact was a possible answer to the need I have described.  So I, like several others who were engaged in this kind of activity, was interested in measuring the change in reflectance as a consequence of distortion of the cornea under the influence of an air pulse.

But at that time we were attempting to measure the amplitude of the reflectance.  I, like the others, found that amplitude was the consequence of the curvature of the cornea, the position of the cornea, and the reflectance from the cornea, which (by the way) could change by a factor of 3 or 4 times between blinks.  And so measuring distortion of the cornea based on amplitude of reflectance was not a fruitful avenue.  Others gave up the chase, but I was determined to make the idea work.  I found (with a bit of luck) that by introducing a beam of light and monitoring the reflection of this light during the air pulse, one could determine when applanation (or flattening of an area the cornea) had been achieved.  I was then able to measure the time it took to reach that applanated or flattened state, and use this time as a correlate of intra-ocular pressure.  It would have been preferable to measure the air pressure required to applanate the eye by utilizing a pressure transducer inside the air chamber of the instrument.  However, at that time, in the early 60's, pressure transducers of adequate accuracy or response time were not available.  But time to applanation was very easy to measure.  So the NCT I, the original non-contact tonometer, did not measure intra-ocular pressure with a pressure transducer, but rather it deduced IOP based on time.  We then empirically determined, by comparison with Goldmann measurements, a calibration curve for the non-contact tonometer.

We introduced the NCT I in June of 1972 at the World Optical Fair in St. Louis.  We had a booth there… and you couldn't get near it.  They were 10, 12, 20 deep waiting to get their turn to be measured.  As a side note, I remember prior to the launch of the instrument, the director of Sales and Marketing and I had a series of friendly arguments about what we should charge for the NCT.  He wanted to make it competitive with other tonometry instruments, which at the time were selling for maybe $2200, but I was arguing that this was a totally different ball game.  And when we released the product at the show he cursed himself... he said he could have charged anything he wanted for this instrument.  We didn't even need to advertise for the first 2 or 3 years because we simply couldn't make them fast enough to fill the orders.

How it works
The air pulse, from all NCT's, has to produce enough force to push the cornea past applanation into a slight concavity.  Otherwise you can't identify when full applanation takes place.  The system sends a beam of light onto the cornea and monitors the reflected light.  Only when the cornea takes on a mirror-like surface, due to flattening, does the light beam produce the maximum amount of reflected light (monitored by a sensor).  The NCT I and NCT II used time as a correlate of intra-ocular pressure.  So the longer it took for the applanation to occur, the higher the pressure.  When applanation was detected, the current to the single-stroke solenoid (which creates the air puff) is cut off.

I spent perhaps 5 or 6 months in an attempt to document (by the way no other manufacturer has ever done this) that we actually produced applanation and that the signal that corresponds and tells us that applanation has occurred coincides with applanation.  By utilizing a high-speed motion picture study at 5000 frames per second, we were able to establish that we did indeed produce applanation and that the maximum signal that corresponds with applanation occurs the very same instant.  And, at 5000 frames per second, the very next frame after applanation does indeed show the beginning of a slight concavity.  We go from applanation to concavity in one frame.  I introduced these films to the International standards committee on tonometry in Paris and it was acknowledged as an applanation tonometer.

Why it is a superior test method
The measurement of intra-ocular pressure by NCT takes milliseconds.  But the pressure within the eye is not really constant.  It pulsates.  That is, there is a cardiac-related ocular pulse, which can vary from a millimeter to as much as 4 mmHg per heartbeat in a normal human eye.  And because the measurement takes only milliseconds it's done completely at random relevant to the cardiac cycle.  And so, we therefore recommend using the average of 3 measurements to arrive at a reliable average IOP measurement.  Now, with the Goldmann applanation tonometer, the operator can see the oscillation of the 2 half rings and most operators look for the mid-point.  But some look for the maximum.  It all depends on the operator's technique.  But one of the important advantages of the non-contact tonometer over Goldmann is that non-contact tonometry is objective.  There is no question of technique or skill.  No judgement is required.

In one classic study, a German researcher did a first set of measurements with Goldmann, a second set of measurements with an NCT, and a third set of measurements with the Goldmann again.  He found that the NCT measurements were closer to each of the Goldmann measurements than the Goldmann measurements were to each other.

Advances in NCT technology
In the first NCT's, the force of the air puff produced a concavity of considerable depth… of greater depth than was required in order to identify when applanation had actually occurred.  In most cases, by the time the air puff shut off, the force developed would have been enough to applanate an eye with a significantly higher IOP.  This is why many people who have been tested with older NCT's fear the puff so much.  Today's puffs are softer.  Not only are they softer, but they are quieter because we have superior ways of dampening the sound and much better ways of shutting down the pump sooner.  Much, much sooner than we were able to in the older models.  Besides that, we are now using pressure rather than time as the correlate of IOP, so we fashion the air pulse ramp differently.  Instead of it being, as in the first models, a linear pulse, it is now nonlinear, which results in a softer puff for the majority of eyes measured.

The significant change between what we did in the 60's and what we do today lies in the advance of electronic technology.  Today we have pressure transducers that respond fast enough to be used inside the air chamber of the NCT.  Their output is a direct correlate of intra-ocular pressure.  Instead of measuring time, which carries the requirement that each linear air pulse be exactly the same, we can now measure air pressure directly.  Therefore, it is no longer critical that each air pulse be linear and exactly the same.

Also, to align the first NCT I used a classical optical spherometer system, which is what is used to measure the radius of curvature of contact lenses.  It's at least a 200-year-old system.  It's the old "red dot in a ring" system.  It allowed me to align the system relative to the eye in 3 dimensions.  When you operated the instrument's joystick, so as to move air tube towards the cornea, so that the working plane of the microscope objective was coincident with the center of the curvature of the cornea, and it was aligned latterly and vertically, you would see an image of a red dot… but only when you had it aligned in the 3 dimensions would you see a clear image of the red dot.  It was a game… it was like a video game.  If you hit the trigger and the red dot had moved out, it wouldn't allow you to take a measurement.  In the current system, the AT550, it's completely different.  It uses two cameras, which detect the image of two light sources that are reflected from the cornea.  Each array has an X and Y address for every pixel.  If the images of those 2 LED's reside at any address other than 0, 0, stepper motors drive the system with the objective of restoring those images to 0, 0 addresses (optimal alignment on the corneal apex).  When each of the 2 light images reside at 0, 0, the system fires automatically.  So now, other than to instruct a patient on how to position their head, the operator needs to do nothing.  The system drives itself to achieve the optimum alignment.

Submitted by David Taylor; his email about the above interiew is below:

Attached you will find an interview I did with Bernie Grolman back in 2002.

The Lineage of the NCT goes like this:

1972 - NCT I
1980 - NCT II
1989 - Xpert NCT (first micro-processor based NCT)
1994 - Xpert Plus NCT (semi-auto alignment introduced)
1999 - AT550 (first fully automated alignment NCT)
2001 - PT100 (first totally cordless, hand held NCT)
2003 - AT555 (minor updates to AT550)
2005 - Ocular Response Analyzer (measures IOP and corneal biomechanics - able to provide "corneal compensated" IOP)
2007 - Reichert 7 introduced (http://www.reichertoi.com/reichert7.html) - 7th generation NCT

Hope this helps.


~1975 Model NCT donated by Reichert to the Optical Heritage Museum (July 2009); Dr. David Luce shown at right

Return to In Memory of Bernie Grolman

AO Instruments

AO Recollections