Most people consider high myopia to be about -10 diopters or greater. We have covered this topic extensively in the past. But what about patients who are -20 diopters or more? This patient has a long-standing refraction of -21.5 in the right eye and when we measure the axial length it registers as 36 mm. That is an ultra high level of myopia and cataract surgery can be life-changing for this patient, but it is also quite challenging.
The first consideration is to ensure that the posterior segment is free from dangerous pathology which could lead to vision loss. I strongly recommend sending these patients to a retina specialist before attempting cataract surgery. The retinal periphery will be examined closely, any weak areas can be address with laser retinopexy, and the macular status can be evaluated. I value the input of my retinal colleagues so much that I also send the patient back for retinal consultation after the successful cataract surgery.
The IOL power estimation is not nearly as accurate in these eyes with extreme axial lengths. Absolutely do not use the print-out of third generation formulae from your biometer. These patients will often have a negative power IOL and the lens design will be a meniscus type. Our goal is to err on the side of residual myopia and in this regard, no matter the method you use for calculation, always aim for at least a little post-operative myopia. Taking a patient from -21.5 diopters to -2.0 diopters is a great success, but having this same patient end up at +2 diopters is not.
I recommend these methods for IOL power estimation for ultra high myopes:
- Barrett Universal II
- Wang-Koch Modification to Holladay 1
- Ladas Super Formula Artificial Intelligence
- Hill-RBF method (if parameters are “in bounds”)
Credit must be given to Professor Graham Barrett for his elucidation of the issues with calculations when it comes to negative power IOLs. An excellent explanation is here. The Wang-Koch group from Baylor University published groundbreaking work a decade ago which fitted a regression formula to highly myopic eyes to come up with an adjustment to the axial length to improve accuracy.
The Adjusted AL = 0.83 * measured AL + 4.27 and then this new value is plugged into Holladay 1 for the IOL power.
Using artificial intelligence algorithms, John Ladas, MD, PhD has come up with the Ladas Super Formula AI to better calculate these challenging eyes. Warren Hill MD also has an innovative approach using his Hill-RBF method and if the data you input is “in bounds” then it delivers a good result for a specific IOL. I encourage you to try all of these methods to see which serves you best.
In this patient, here is what we determine using the Barrett and Ladas AI formulae:
We have chosen the Alcon MA60MA three-piece acrylic IOL in a power of -3.0 diopters (minus power) for the right eye to give a post-op result of mild myopia of about -0.5 to -1.0 which should be excellent. Note how different this is from the original biometer print-out which called for a -7.0 diopter IOL, which certainly would have resulted in an undesirable, hyperopic post-op surprise.
In the future when you come across a patient with ultra high myopia, refer back to this article and when in doubt, remember to always choose the higher IOL power so that we err on the side of myopia. We certainly owe a debt of gratitude to the pioneers who have helped in the IOL power calculations for ultra high myopic eyes: Jack Holladay MD, Wolfgang Haigis PhD, Graham Barrett MD, Warren Hill MD, Doug Koch MD, Li Wang MD, PhD, and John Ladas MD, PhD.
Addendum: Should you use the special A-constants for negative-power IOLs? That was an idea from about 10 years ago by Wolfgang Haigis PhD. (Intraocular lens calculation in extreme myopia. Haigis W. JCRS 2009; 35(5): 906-911). This method has since fallen out of favor due to the advances such as the Wang-Koch axial length adjustment (Wang L, et al. Optimizing intraocular lens power calculations in eyes with axial lengths above 25.0 mm. JCRS 2011; 37:2018-2027.) Newer methods such as Ladas Super Formula 2.0, Barrett Universal, and Hill-RBF 2.0 will all do a good job. In 2018 and beyond, I disagree with using different A-constants for negative power IOLs. Instead, the best advice is to use the methods that I outline in the article.
All content is © 2018 by Uday Devgan MD. All rights reserved.