IOL power calculation in eyes with keratoconus is challenging for several reasons, which resemble those potentially leading to refractive errors in eyes with previous corneal refractive surgery.

1. Calculating the corneal power with the usual keratometric index (n = 1.3375) can lead a wrong value. This fictitious index allows us to achieve the refractive power of the whole cornea from the anterior corneal curvature measurements only, only on condition that the ratio between the anterior and posterior corneal curvatures is within normal limits. The ratio is disrupted by corneal ectasia.

2. Keratometers and corneal topographers provide inaccurate measurements of corneal curvature if this is asymmetric. These instruments assume that such curvature is constant along a given meridian, which is not true in most keratoconic eyes. Moreover, when the keratoconus apex is located over the entrance pupil, the ectactic area can induce discrepancies between the corneal curvature readings along the paracentral ring and those in central corneal area.

3. Vergence formulas were developed for normal eyes and we do not know whether the ELP prediction is still accurate in eyes with keratoconus, due to the typically steep corneal curvature and increased anterior chamber depth.

4. Tear film irregularities are often observed in eyes with irregular corneas and can limit the repeatability of corneal curvature measurements.

**Results with standard formulas**

In 2019 Savini et al compared the outcomes of 5 traditional formulas in eyes with keratoconus: Barrett Universal II, Haigis, Hoffer Q, Holladay 1 and SRK/T.1 Eyes were categorized in Stage I, II and III, according to Krumeich.2 These were the main results:

All formula provided a mean hyperopic outcome.

Eyes with more advanced categories of keratoconus yielded worse results.

In eyes with Stage I keratoconus (K<48 D), the most accurate formula was the SRK/T, which provided a mean prediction error (PE) of +0.44 ±0.79 D, a median absolute error (MedAE) of 0.43 D and 61.9% with a PE within ±0.50 D. It can be postulated that the relatively good performance of the SRK/T is due to its tendency to overestimate IOL power in eyes with high K values:3 this tendency is likely to counterbalance the average hyperopic outcomes of the remaining formulas in eyes with keratoconus.

In eyes with Stage II keratoconus (K between 48 and 53 D), the SRK/T was still the most accurate formula, but all results were worse compared to Stage I.

· In eyes with Stage III keratoconus (K>53 D), the mean PE was hyperopic by 3 D with all formulas and very few to no eyes had a PE equal or lower than ±0.50 D.

Later on, two studies assessed the accuracy of two IOL power formulas specifically developed for eyes with keratoconus: the Barrett True-K formula for keratoconus and the Kane formula for keratoconus. In the first study (which did not include the Barrett formula as it was not created at the time of publication), the Kane formula provided the most accurate outcomes.4 In the second study, the Barrett True-K formula for keratoconus with measured posterior corneal astigmatism was the most accurate (87.5% of eyes with a PE within ±0.50 D), followed by the SRK/T (59.4%) Kane formula for keratoconus (53.1%).

**Future developments**

In order to avoid the keratometric error, we can calculate the corneal power by means of ray tracing. All Scheimpflug cameras and anterior segment OCT display these values, which can be entered into IOL power formulas. Theoretically, since corneal power by ray tracing is lower than the corneal power calculated with the keratometric index, specifically optimized constants should be used. Unfortunately, these are not usually available, so that it is hard to apply this solution. Further studies are needed to assess whether using corneal power by ray tracing with or without optimized constants can improve the results of IOL power calculation in eyes with keratoconus. Until these studies are available, it is wise to compare the keratometric and the ray traced corneal power: if a difference higher than 1.00 D is observed, caution should be warranted before choosing the IOL power.

**References**

1 - Savini G, Abbate R, Hoffer KJ, et al. IOL power calculation in eyes with keratoconus. J Cataract Refract Surg 2019;45:576-581.

2 - Krumeich JH, Daniel J, Knülle A. Live-epikeratophakia for keratoconus. J Cataract Refract Surg 1998;24:456-463.

3 - Melles RB, Holladay JT, Chang WJ. Accuracy of intraocular lens calculation formulas. Ophthalmology 2018;125(2):169-178.

4 - Kane JX, Connell B, Yip H, McAlister JC, Beckingsale P, Snibson GR, Chan E. Accuracy of intraocular lens power formulas modified for patients with keratoconus. Ophthalmology 2020;127:1037-1042.

5 - Ton Y, Barrett GD, Kleinmann G, Levy A, Assia EI. Toric intraocular lens power calculation in cataract patients with keratoconus. J Cataract Refract Surg 2021;47:1389-1397.

## Comentários