IOL power calculation not as straightforward as it seems

A young ophthalmologist gives pearls on how to be successful in a crucial step of modern cataract surgery.

Issue: July/August 2014

ByNino Hirnschall

Patient expectations after cataract surgery are continually increasing, and unaided postoperative visual acuity is an important correlate of patient satisfaction. Nino Hirnschall, MD, PhD, resident at the Hanusch Hospital, Vienna, and researcher at the Vienna Institute for Research in Ocular Surgery, both directed by Oliver Findl, MD, MBA, gives us an introductory guide to making the best-informed IOL lens power decisions and reminds us that this may not always be straightforward.

Anthony Khawaja, MB BS, MA(Cantab), MPhil, FRCOphth
Chair of the SOE Young Ophthalmologists committee

Many residents think that IOL power calculation is a relatively straightforward procedure and may completely trust whatever the optical biometry device tells them concerning IOL power. However, every young cataract surgeon should be aware of the fact that an accurate preoperative calculation of IOL power is crucial for the postoperative satisfaction of the patient. This article aims at giving a basic insight into the topic of IOL power calculation, important shortcomings of current techniques and some ideas of what the future most likely promises.

Nino Hirnschall

Even with modern measurement techniques, such as optical biometry and the latest-generation of IOL power calculation formulae, there is a degree of unpredictability of the postoperative manifest refraction outcome. Approximately 10% to 20% of patients need refractive correction of more than ±1 D spherical equivalent after surgery. These refractive surprises result in poor unaided visual acuity and reduced patient satisfaction.

What is the main source of error for the postoperative refraction?

Norrby investigated the error distribution of different factors on postoperative manifest refraction, and it turned out that by far the main source of error is the prediction of the postoperative IOL position. Other factors, such as the axial eye length measurement and the measurement of the cornea, are a much smaller source of error.

Which eyes are at risk of having a postoperative refractive surprise?

IOL power calculation is especially difficult for short and long eyes. In short eyes, this error is even more relevant because short eyes usually need a high IOL power. The prediction of the position is, therefore, even more important. In contrast, long eyes usually receive low-powered IOLs. Additionally, patients who have undergone corneal refractive treatment are at risk of an unexpected postoperative refraction.

A basic understanding of IOL power calculation

To understand modern IOL power calculation, one has to understand the nature of regression formulae: How could the axial eye length (in mm) and the corneal radii (in mm) be used to predict the postoperative manifest refraction (in diopters)? Let’s assume you would have these three parameters from 100 patients. In a next step, you can create a multiple linear regression model for every patient using this formula:

Rx = x * AL + y *K +A

What this formula means is that the postoperative refraction (Rx) is equal to the axial eye length (AL) times a weighting factor (ie, AL * 0.3) plus/minus the keratometry times a weighting factor. Furthermore, regression always requires a constant “A,” which compensates for the offset from the origin.

The aim of a regression model is to create a stable model that can be used in new patients other than the 100 patients the formula was derived from. It should be noted that the A constant has to be optimized for different types of IOLs separately. If we are a bit more critical, we have to admit that there is an element of “fudging” to this because we have a deviation between the aimed and measured outcome and correct for it by changing a virtual factor, namely the constant, without knowing where the error derives from. Most of the modern IOL power calculation formulae, including SRK/T, Holladay I and II, Hoffer Q, Haigis and others, use a combined model. That means that they use exact physical models (“theoretical formulae”) but add a regression component to it, including one or more constants that can be optimized. One difference between modern IOL power calculation formulae is the type and number of parameters that are used to predict the postoperative IOL position. To be more precise, these formulae do not predict the real postoperative IOL position but a virtual non-measureable factor called effective lens position.

Which parameters do different formulae use to predict the effective lens position?

One parameter that all formulae use to predict postoperative IOL position is the axial length. Some formulae, such as SRK/T, Holladay I and II, and Hoffer Q, also use the preoperatively measured corneal radii. Another factor that is often used with the Haigis, Holladay II, Olsen and Norrby formulae is the preoperatively measured anterior chamber depth. Some formulae, such as the Olsen, Holladay II and Norrby, also take the preoperative lens thickness into account.

However, despite taking all these parameters into account, cataract surgery still fails as a refractive procedure in very long and very short eyes. Imagine that you have a patient with an axial eye length of 20.5 mm who wants a multifocal IOL. The chance that this patient will have a refractive surprise of more than 1 D is greater than 20%.

What about calculating the IOL power using intraoperative measurements?

One novel method for selecting IOL power is to use intraoperative optical coherence tomography scans to measure the anterior chamber depth in the aphakic eye under standardized conditions. The idea is the following: During cataract surgery after removing the lens, in the aphakic state, the anterior lens capsule is taut and in a position close to where the anterior IOL haptic will be after implanting the IOL. Essentially, this method allows potentially more accurate assessment of true postoperative lens position. In recent studies it was shown that this novel concept results in a significantly better prediction of the postoperative IOL position compared with preoperative measurements.

References:
Aristodemou P, et al. J Cataract Refract Surg. 2011;doi:10.1016/j.jcrs.2010.07.032.
Findl O, et al. J Cataract Refract Surg. 2001;doi:10.1016/S0886-3350(00)00699-4.
Hirnschall N, et al. Invest Ophthalmol Vis Sci. 2013;doi:10.1167/iovs.13-11991.
Lundström M, et al. J Cataract Refract Surg. 2012;doi:10.1016/j.jcrs.2012.03.006.
Norrby S. J Cataract Refract Surg. 2008;doi:10.1016/j.jcrs.2007.10.031.
Olsen T. Acta Ophthalmol Scand. 2007;doi:10.1111/j.1600-0420.2007.00879.x.

For more information:
Nino Hirnschall, MD, PhD, can be reached at VIROS, Hanusch Krankenhaus, Heinrich-Collin Straße 30, 1140 Vienna, Austria; +43-1-91021-57564; email: nino.hirnschall@googlemail.com.
SOE Young Ophthalmologists website: http://soevision.org/yo

Disclosure: Hirnschall has a proprietary interest in the intraoperative OCT measurements for IOL power calculation as patent assignee.