1146: the future of IOL calculations

This video features the keynote lecture that John Ladas, MD, PhD gave at the UCLA Jules Stein Eye Institute annual meeting in June 2021. You may not realize it, but you are approaching IOL calculations wrong and this talk will show you the light. If you think that by optimizing your A-constant and using a good formula is sufficient, but that will only give you about 80% accuracy. Invest the time to watch and learn from this talk that will show you the future of IOL calculations.

click to learn from this outstanding lecture on the future of IOL calculations:


  1. Another source of error to consider is the variance in IOL power from the manufacturing of IOL’s. It may be worth checking with the manufacturers. Years ago, a leading IOL company told me that an IOL was deemed acceptable if the power measured falls within 0.5 D of the target. It could be different now.

    Furthermore we customarily refract in 0.25 D increments, which limits how refined we can make our postop refractive measurements. Autorefraction could be helpful in refining the postop refraction. However, vision is a psycho-physiologic phenomenon. We all have encountered patients who would not accept the full objective refraction, and we have to modified the spherical and astigmatic corrections in order to satisfy the patient subjectively.

    Just some food for thought.

    1. Great food for thought. So happy you are thinking about this.

      To your first point, eventually “exact” powers will be provided with each iol. Deep learning algorithms such as our will incorporate it seamlessly. (An addition of .2 diopter of iol power will have a much different effect on a “short” eye as opposed to a “long” eye)….agreed? Also, ACD will be important…again…intimately related to AL but not mutually exclusive.

      To your second point…I agree with you but we have to have a goal of something…I could envision a future that determines what the best possible average refraction for a given eye is for “psycho-physiologic” success. I do know that a computer with deep leaning will take us there…not a human like myself guessing the “fudge factors” necessary to achieve that.,


Leave a Reply