gms | German Medical Science

30. Internationaler Kongress der Deutschen Ophthalmochirurgen (DOC)

11.05. - 13.05.2017, Nürnberg

DOC KEY NOTE LECTURE: The evolution of IOL power calculation

Meeting Abstract

Suche in Medline nach

  • Thomas Olsen - Aarhus Kommunehospital, University Eye Clinic, Aarhus, Dänemark

30. Internationaler Kongress der Deutschen Ophthalmochirurgen. Nürnberg, 11.-13.05.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocH 13.5

doi: 10.3205/17doc012, urn:nbn:de:0183-17doc0123

Veröffentlicht: 27. April 2017

© 2017 Olsen.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe



The quest for accurate IOL power calculation was born in 1949 when Harold Ridley implanted the first IOL. The surgery was successful – the patient still could not see – but the refractive error was found to be minus 20 D! However, it was not until the 1970s when ultrasound biometry gradually became integrated into the preoperative cataract evaluation that formulas evolved to estimate the IOL power from the K-reading and the axial length. The first formulas were the so-called ‘theoretical formulas’ (Fyodorov, Werner-Ostholt-Gernet, Binkhorst, Hoffer, Colenbrander and others) based on ‘thin-lens’ calculations with assumed models for the Estimated Lens Plane (ELP) after surgery.

In the 1980s a group of empirical formulas were developed. As opposed to the theoretical formulas these formulas were based on regression analysis of empirical data only (Sanders-Retzlaff-Kraff (SRK I, SRK II) and others). They were found to work well for most cases and gained wide-spread popularity. However, refractive surprises were still seen in out-of-range eyes.

In the 1990s, when the surgical technique was boosted by phacoemulsification, small-incision and capsulorhexis technique – allowing for standardized in-the-bag placement of the IOL – the demand increased for higher refractive control and formulas were developed that combined the theoretical approach with empirical modelling and back-calculation of the ELP (Holladay 1+2, SRK/T, Hoffer Q, Haigis). Many of these formulas are still in use today.

The introduction of laser biometry around year 2000 (Zeiss IOLMaster) marked a quantum leap in the history of IOL power calculation. What used to be the largest source of error now turned into one of the least significant. Consider the difference between a 0.20 mm error with ultrasound as compared to a 0.02 mm error with the laser. Because an axial length error of 1.0 mm corresponds to a 2.5 D error in the spectacle plane, only 0.05 D error can be attributed to the optical measurement of axial length. Therefore, the limiting factor for accurate IOL calculation is now in the determination of corneal power and/or in the prediction of the ELP – that is, in the formula itself. Thick-lens or ray-tracing formulas have evolved to address the limitations of the thin-lens formulas with improved models for the prediction of the IOL position (Preussner, Olsen, Barrett).

Recently, the old battle between theoretical and empirical methods has been re-ignited as newer statistical methods using pattern recognition have been claimed to be superior to other methods (Hill RBF). Future studies will show what is the best approach for accurate IOL power calculation.