17 Keratoconus and Cataract Surgery Abstract Irregular astigmatism associated with keratoconus usually leads to a dissatisfied patient with suboptimal outcomes postcataract surgery. This chapter expresses all the concerns and the guidelines to be followed to minimize the postsurgery refractive error and also highlights the intraocular lens (IOL) power calculations and the choice of IOL options that are available. Keywords: keratoconus, hard contact lens, toric IOL, astigmatism, irregular astigmatism, corneal optics One of the greatest challenges in modern cataract surgery is that of getting the refractive target right. Patients want and expect good refractive outcomes from cataract surgery today. From the most basic surgical perspective, however, surgeons want to replace the cloudy crystalline lens with a pristine intraocular lens (IOL) that allows all light to pass through it. Keratoconus (and other conditions causing irregular corneal astigmatism) adds a level of complexity to this aspect. Not only is biometry a lot more challenging, but also, sometimes, despite the newly placed IOL being pristine and in the correct position and alignment, the best-corrected acuity does not improve as much as what was expected. In eyes with keratoconus and cataract, both the irregular corneal optics and the cloudy lens optics are playing a role in the reduced vision. To manage patient expectations best, it is prudent to know what the cornea’s contribution is to the reduced vision before embarking on the cataract surgery. Just like cataracts, keratoconic corneas also come is varying grades of distortion. For this chapter, perhaps the Amsler–Krumeich classification will work best to grade corneal irregularity. Preoperative vision, both UDVA (uncorrected distance visual acuity) and BSCVA (best spectacle corrected visual acuity) as well as the visual history some years prior to cataract surgery are also very helpful. For most cataract patients, the cause of the recent reduction in vision will be the cataracts as keratoconus tends to be more stable at ages at which cataracts typically occur. If the vision prior to the onset of cataracts could be well corrected with spectacles to a level of 6/12 or better, then cataract surgery with or without toric implants should provide similar or better best-corrected spectacle vision. Good vision with soft contact lenses (CLs), including toric soft CLs, may well be a similar indicator; however, soft lenses can mask a level of irregularity that may be significant and best to seek out records of best spectacle corrected vision from the individual responsible for fitting the lens. If the patient requires rigid gas permeable CLs (RGPs) or scleral or mini-scleral CLs to see well, cataract surgery may provide disappointing visual results until the corneal irregularity is addressed. Of course, patients can continue to wear RGPs or scleral CLs following cataract surgery, but when patient expectations are considered, it is usually desirable to provide the best UDVA wherever possible. Today addressing corneal irregularities is more successful than years before with the use of topography-guided laser ablation and intracorneal ring segments ( Fig. 17.1 Intraoperative image of a Bitoric lens implant in a patient with keratoconus following earlier implantation of a Ferrara ring. The patient was an elderly gas permeable contact lens wearer who historically had best spectacle corrected visual acuity (BSCVA) of 6/21 and became contact lens intolerant. A Ferrara ring was used to improve her corneal shape with improved BSCVA to 6/15. Microincisional cataract surgery was performed through a scleral incision and a 3.0-diopter Bitoric lens implant inserted. Postoperatively her uncorrected visual acuity improved to 6/12 and BSCVA to 6/9 with a refraction of + 1.25 – 1.00 × 160. • Belin’s ABCD classification1: • RETICS (Redes Temáticas de Investigación Cooperativa en Salud—Thematic Networks of Cooperative Research in Health). This classification is based mainly on CDVA but also includes higher-order aberrations (HOAs) like coma, asphericity, internal astigmatism, and H-RMS, thereby providing a functional classification of visual acuity. Using RETICS, toric IOLs can be implanted in eyes with grade 1 and selected eyes with grade 2. For higher grades, it makes sense to try and regularize the cornea first by means of intrastromal corneal ring segments (ISCRS) or topography-guided photorefractive keratectomy (TG-PRK). IOL power calculations use corneal data in two ways: • To apportion optical power as part of the entire optical system (e.g., 43 diopters). The cornea provides 66% of the refractive power of the eye, so even a 2% error can lead to a miscalculation of 1.00 diopters. • As a predictor of where the IOL will land up within the eye (ELP or effective lens position). The effect of ELP can be summarized by this rule of thumb: a 1-mm shift in the anteroposterior position of the IOL causes a 10% change in the effective IOL power. An eye with a 23.00-diopter IOL that lands up 0.5 mm more anteriorly than predicted will be 1.15 diopters more myopic than expected. In biometry, the cornea is characterized by the K-value, a single number that assigns refractive power to the cornea. Normally, this would suffice in regular corneas, but when the cornea is irregular and multifocal, a single number fails to provide sufficient data for biometry. Which part of the multifocal cornea is the patient using when looking in the distance? Is this the same part of the cornea when looking up close? Does this change during the day or during the visual task depending on fatigue, working distance and position, nature of the visual task, and so the list of variables goes on? What single K-value is going to be able to provide the best information for the IOL formulae? To add to the error for potential, consider how all keratometers work. They project an image onto the cornea and then measure the size of the reflected image. This size of the image is then converted to a measure of corneal steepness in millimeters of radius. A flatter cornea leads to a larger image and a steeper cornea to a smaller image. Given the mechanics, one can now clearly see how something as simple as misalignment may impact a corneal steepness measurement. The steepness or curvature is converted to a dioptric power using the formula D = (i – 1)/r, where i is the refractive index (RI) of the cornea and r is the radius of curvature in meters. This formula is an approximation as it ignores the posterior cornea curvature. A further error may creep in with the assumption that the RI of the cornea is 1.3375 for all humans. It is, however, highly likely that there is wide variation in the RI of the human cornea, both in its virgin state and following refractive surgery. It is likely to be different too in abnormal corneas such as we find in keratoconus. Because we cannot measure the corneal RI for any individual patient, we substitute with an assumption. The central corneal shape, the most important corneal power for most everyday tasks, is best described as an aspheric–toric surface. Aspheric because it is typically steeper in the center and flatter in the periphery and toric because most corneas have some degree of regular astigmatism, even if very subtle. Over and above these lower-order aberrations, there are also HOAs or irregularities. When these are amplified as with keratoconus, they can take on clinical and visual significance. These irregularities are usually described with Zernike polynomials or Fourier analysis. Despite the multitude of factors that contribute to corneal power, the cornea has been characterized by the single K-value from the earliest IOL formulae to those more recently like the Olsen and Holladay.2 This K-number represents an arbitrary paraxial corneal power using an assumptive corneal index of refraction due to the absence of data concerning the posterior corneal power with routine biometry. This works reasonably well for regular corneas where the assumptions that are made regarding the RI and the posterior corneal curvature are relatively accurate. In these corneas, HOAs do not play a major role and their contribution to the refractive power of the cornea is almost negligible. Ray-tracing modeling has confirmed these findings too: Okulix (Tedics Peric & Jöher) and PhacoOptics (IOL Innovations) use thick-lens models and can further calculate spherical aberration. Newer devices like the Sirius tomographer (CSO) use ray tracing with a thick-lens model to predict ELP and they do not assume total corneal power, but rather measure the posterior cornea directly and then calculate the total corneal power. These corneas have high degrees of HOAs and the typical anterior cornea: posterior cornea curvature relationship (Gullstrand’s ratio) can be distorted. Additionally, the devices measuring these corneas are also affected by the irregularity. Keep in mind that the posterior corneal surface is imaged through the irregular anterior surface that due to its irregularity will affect the quality of the posterior corneal data. Additionally, the RI of the cornea continues to be an assumption and is not directly measured. The most common irregular corneas encountered are those following previous corneal refractive surgery, corneal scars, dry eye, and keratoconus. It is critical to mention dry eye here as it will impact the quality of the biometry corneal data and the topographic data. Correct diagnosis of the corneal condition is paramount as different conditions will affect the calculation process differently. Examining the corneal topography is critical and biometry alone does not suffice. Corneal HOAs are also measured and quantified and are normally around 0.40 ± 0.15 µm2. Higher levels than this indicate some level of corneal irregularity. In keratoconus, the most important HOA is normally vertical coma and most keratoconus corneas are also hyperprolate inducing abnormally high levels of negative spherical aberration. As with all corneal irregularities or whole eye wavefront measurements, pupil size directly impacts the total HOA level—the larger the pupil, the greater the HOAs. Fortunately, following cataract surgery, pupil size is often smaller and typically between 4.0 and 5.0 mm and thus HOAs (by convention measured at 6.0 mm) are less relevant. As previously mentioned, RGP CLs can also be very valuable to assess the corneal contribution to the reduced visual acuity when corrected distance visual acuity (CDVA) appears to be less than expected given the density of the cataract and the retinal health. In keratoconus specifically, a problem of optical decentration exists, but it is different to the decentration that we deal with following decentered corneal ablative refractive procedures. In keratoconus, the posterior cornea steepens along with the anterior cornea, decreasing the anterior-to-posterior ratio (A-P ratio; unlike in ablative corneal surgery where the posterior corneal curvature typically remains unchanged.) This leads to an overestimation of the actual corneal net power. A second factor is the ELP, which is more posterior than in normal eyes due to the anterior displacement of the cornea. Both factors shift the refractive prediction error toward hyperopia. This is the very reason that in mild to moderate keratoconus, a regular vergence formula can be used, targeting mild myopia. When multifocality within the optically active visual axis is pronounced as with more advanced keratoconus, ray-tracing software should be used to select the best IOL power. If possible, increase the ELP, as there is no algorithm specific for keratoconic eyes. A simple alternative is to target a slightly myopic refraction. This chapter is specifically about cataract surgery in keratoconic eyes, but having a basic understanding of the methodology of calculating IOL power in other corneal irregular conditions is helpful. The main change after corneal ablative surgery (PRK and laser in situ keratomileusis [LASIK]) and small incision lenticule extraction (SMILE; stromal lenticule removal) is the change of the anterior corneal radius. The posterior corneal radius remains unchanged for the most part. Spherical aberration can be a problem as well as ELP if anterior corneal data only are used to predict it. Vergence formulas that use the double-K method for ELP calculations and use regression analysis to correct the Gullstrand ratio (A-P corneal curvature ratio) can be used here. These include the double-K Sanders–Retzlaff–Kraff (SRK)/T, the Holladay, Haigis-L, Shammas-PL, and the Barrett True-K formulas. Intraoperative biometry also has value3 to help refine the IOL power choice and toricity, and this applies to keratoconic corneas too. Decentered corneal surgery impairs the optical performance of the cornea irrespective of whether it was LASIK, PRK, SMILE, or RK, mainly due to induced coma that can be vertical, horizontal, or oblique. The greater the corneal irregularity, the more it makes sense to use exact optics rather than paraxial optics. The refraction itself is prone to the same issues with the multiple refractive corrections that can be successfully utilized depending on which part of the cornea the patient looks through. Ray tracing has the potential to add value here with the central corneal data being imported into the ray-tracing software and then retinal image quality being assessed with direct ray tracing assuming different power IOLs in the model. IOL power is the main aim of biometry and in patients with corneal astigmatism, toric IOLs have a significant role to play in improving UDVA. With keratoconus, the astigmatism is frequently irregular and selecting a toric IOL is more challenging. This challenge is further complicated when there has been previous corneal surgery like intrastromal corneal rings, corneal collagen crosslinking (CXL) or keratoplasty. Modern toric IOLs have been shown to provide excellent efficacy, predictability, and safety when used to correct corneal astigmatism during cataract and refractive lens exchange surgery in normal eyes, especially in eyes with more than 1.5 diopters of astigmatism. There is a paucity of peer-reviewed articles on toric IOLs in keratoconus, however.4,5,6,7,8,9,10,11,12,13 There are also some theoretical concerns about the use of toric lenses with high levels of cylinder. Some companies manufacture lenses to order and can deliver customized lenses with very high cylinders. The optical performance of high-magnitude toric lenses can be influenced based on whether the lens astigmatism is on a single surface, often the back or “bitoric,” whereby astigmatism is split and shared on the front and back surfaces (AT Torbi, Zeiss, Jena, Germany). Another concern is understanding the axis to use for toric lens alignment. Often the axis of astigmatism is nonorthogonal and asymmetric and the true alignment axis is based on the vector analysis considering the magnitude of astigmatism at each meridian. One of the authors (SD) has for over 10 years used the axis of the S2 aberration for astigmatism derived from corneal topography to align toric lenses and has found this to be a very useful and accurate method in keratoconus and form fruste keratoconus ( In keratoconus, the visual axis is typically along the slope of the cone and not at the apex of the cornea (except with central keratoconus) and hence K-readings may be falsely high for biometry purposes. The software employed in the IOL calculators was validated on normal eyes and will not necessarily transfer to keratoconus eyes either. Most normal eyes show the corneal apex being coincidental with the visual axis.8 It makes sense therefore to use toric IOLs in eyes with stable mild to moderate keratoconus with relatively regular central corneal astigmatism. An exciting lens option for use in keratoconus is the IC8 Lens (AcuFocus). This is a small-aperture lens designed to provide increased depth of focus and works on the principle of the pinhole. The benefit of the use of this lens is the pinhole effect also reduces the impact of corneal astigmatism as well as irregularity,14 both of relevance in keratoconus. Kermani implanted the lens unilaterally in four patients and bilaterally in two patients with keratoconus with unaided visual acuities of 6/12 or better in all cases along with improved unaided near vision (personal communication). Additionally, his colleague Georg Gerten, MD, implanted an IC8 lens in a patient with keratoconus that had previous intracorneal rings placed and obtained a very good visual outcome.15
17.1 Introduction
Fig. 17.1) and if the cornea has been regularized prior to cataract surgery, improved outcomes can be expected.
17.1.1 Other Grading Systems for Keratoconus
Anterior corneal curvature.
Posterior corneal curvature (Back).
Corneal pachymetry.
Corrected distance visual acuity (CDVA; Distance vision, corrected).
17.1.2 Intraocular Lens Power Calculations in Keratoconus
17.1.3 Biometry, Intraocular Lens Formulae, and Corneal Optics
17.2 Biometry in Irregular Corneas
17.3 Intraocular Lens Considerations
17.3.1 Toric Lenses
Fig. 17.2). The magnitude of astigmatism at the cornea can also contribute to the decision-making process of the amount of astigmatism that might wish to be considered when planning the procedure. Again in one of the author’s (SD) experience, correction beyond 6.0 diopters can lead to problems from abnormal optics as well as increased error in terms of calculation and residual astigmatism if misaligned by even small levels normally considered within tolerance (±5 degrees).
17.4 Other Intraocular Lens Options
Ento Key
Fastest Otolaryngology & Ophthalmology Insight Engine
