The Centration Dilemma in Refractive Corrections: Why Is It Still a Dilemma and How to Cope?

Abstract

(1) Background: Defining the optimum center for laser refractive corrections is difficult, with many of the available approaches having pros and cons. Decentered ablations result in undesirable side effects like halos, glare, monocular diplopia, and a reduction in visual acuity; (2) Methods: The ideal centration in refractive corrections should fulfil three requirements: covering a scotopic pupil; respecting the visual axis; and minimizing tissue removal. The implications of different centration strategies are discussed and shown graphically; (3) Results: Oversized asymmetric offset ablation faces fewer difficulties in registering static cyclotorsion, features less tissue wastage compared to a symmetric offset, and includes a certain amount of coma (and trefoil) in the profile, benefiting eyes with a pupil offset, which typically present with relevant amounts of coma and trefoil corneal aberrations due to decentered optics; (4) Conclusions: There is a need for a flexible choice of centration in refractive procedures to design customized and non-customized treatments optimally. An ideal optical zone covering the pupil with the widest entrance may be as important as a centration reference.

1. Introduction

Defining the optimum center for laser refractive corrections is difficult with many of the available approaches, each of them having their pros and cons. The human eye is an asymmetrical optical system where each element has its own optical axis, making the choice for the ideal reference very challenging. The real cornea is not a rotationally symmetrical volume. Typically, refractive corrections are designed with two different centration references that can be detected easily and measured with the currently available technologies (pupil centration (PC) and corneal vertex centration (CV)) [1].

Pupil centration may be the most extensively used centration method for several reasons. For a patient who fixates properly, PC defines their line of sight (LOS), which is the reference axis recommended by the OSA for representing wavefront aberrations. The pupil boundaries are the standard references observed by eye-tracking devices. Moreover, the entrance pupil can be well represented by a circular or oval aperture, like the most common ablation areas. But PC is not necessarily the reference based on which the patient is actually driving the visual axis during manifest refraction. More importantly, PC is unstable and changes with pupil size [2,3,4]. Centering on the pupil offers the opportunity to minimize the optical zone (OZ) size (and hence the ablation depth and volume), but the change in the pupil dimensions is not concentric, which could potentially mismatch the circular or oval ablation areas. The diagnostics for measuring aberrations are performed under the conditions of a scotopic pupil, which are not maintained when using laser correction devices.

The visual axis can be determined as the nodal ray from the fixation point that strikes the foveola with zero transverse chromatic aberrations. CV represents a stable morphologic reference and is the optical reference for corneal topographies, but it is difficult to map CV to a laser. The light reflex is non-constant and dependent on the gaze of the observer. The Purkinje images will be seen differently depending on the surgeon’s eye dominance and eye balance and the stereopsis angle of the microscope, making it difficult to manage under a laser. The subject-fixated coaxially sighted corneal light reflex (SF-CSCLR) avoids these dependencies and represents both corneal Purkinje images. The SF-CSCLR has been postulated to lie closer to the corneal intercept of the visual axis than PC [5], however, it is not possible to map it under the laser. For a higher angle kappa, the corneal reflex can result in perceived coma induction, as higher-order aberrations (HOAs) are measured with respect to PC with aberrometers. Since eye trackers follow the pupil boundaries, ablations centered using the pupillary offset (the distance between PC and CV) numerically consider CV as the reference. It must be noted that in less prevalent oblate corneas, the point of maximum curvature (corneal apex) might be off-center and not well represented by CV. In these cases, PC is probably more stable.

Decentered ablations during refractive surgeries can give rise to undesirable side effects, such as halos, glare, monocular diplopia, and a reduction in visual acuity [6,7]. The ideal centration in refractive corrections should fulfil three requirements: first, covering the scotopic pupil; secondly, respecting the visual axis; and thirdly, minimizing the tissue removal. These three requirements may apparently contradict each other. Covering the scotopic pupil means centering on the scotopic pupil and using a large optical zone, which violates the second requirement. Similarly, respecting the visual axis will mean leaving a crescent out of the scotopic pupil or using an extra-large optical zone, violating the first and third requirements.

The implications of different centration strategies are depicted in Figure 1.

Centering the ablation optical zone according to PC (a photopic pupil) under the laser and not considering a scotopic pupil will leave a crescent-like unablated area, leading to night vision problems (Figure 1A).

Enlarging the ablation zone to cover the scotopic pupil may present the challenge that static cyclotorsion control (SCC) is not registered correctly due to less of the iris area being captured (Figure 1B).

Centering according to (scotopic or photopic) PC means HOAs like coma and trefoil present in cases with large pupil offsets.

Centering according to CV using a symmetric offset without considering the scotopic pupil (Figure 1C) will not induce HOAs, but night vision problems may appear due to an even more pronounced crescent-like unablated area compared to PC. The severity of the night vision problems could be reduced due to the Stiles–Crawford effect and a surgeon may sometimes even need to consider the compromise that night vision problems only affect nighttime vision, whereas HOAs always affect vision.

Oversized symmetric offset ablation (Figure 1D) increases the ablation depth centrally in myopia and peripherally in hyperopia corrections, resulting in tissue wastage.

A compromise between ablations according to PC and CV is seen in an asymmetric offset, where the manifest refraction is referenced against CV, while higher-order aberrations are referenced to the pupil center (Figure 1E). Asymmetric offset centration may also present the challenge that SCC is not registered correctly due to less of the iris area being captured.

Oversized asymmetric offset (Figure 1F) ablation (oversized by 0.2–0.3 mm) faces fewer difficulties in registering SCC. Figure 1F uses the photopic pupil as the boundary since under the laser, the pupil size is typically photopic, which eye trackers can easily follow. It uses an asymmetric offset (a photopic pupil offset) to numerically determine CV as the reference and slightly oversizes the OZ to account for the shift in the center from a photopic pupil to a scotopic pupil. Yet the oversize value is much smaller compared to that for a pupil vertex offset. It features less tissue wastage compared to a symmetric offset and includes a certain amount of coma (and trefoil) in the profile, benefiting eyes with a pupil offset, which typically present with relevant amounts of coma (and trefoil) as corneal aberrations, essentially due to the decentered optics.

2. Comparison of Various Centration Methods

The pros and cons of each strategy are summarized in

Table 1. Summary of the pros and cons of the different centration strategies in refractive corrections.

	Photopic Pupil	Scotopic Pupil	Symm Offset (No Oversize)	Symm Offset (Oversized)	Asymm Offset
Reference	Photopic pupil center	Scotopic pupil center	Corneal vertex	Corneal vertex	Corneal vertex (for Rx) Pupil center (for HOAs)
Concentric to…	Photopic pupil	Scotopic pupil	Corneal vertex	Corneal vertex	Pupil
Easiness	Maximum	Dim laser lights	Corneal data	Corneal data	Corneal data
Min OZ	Scotopic pupil size	Scotopic pupil size	Scotopic pupil size	Scotopic pupil size + 2 × offset	Pupil size + 2 * pupil center shift
Tissue wastage	Minimum	Minimum	Minimum	Maximum	Moderate
HOAs	Coma Trefoil	Coma Trefoil	Coma Trefoil	Minimum	Minimum
Further issues	Crescent	None	Crescent	Waste of tissue	None
Topography	Normal	Normal	Best	Best	Better
Wavefront-guided	YES	YES	YES	YES	YES

.

Here, the color codes depict the severity of the effect, with green being least severe, yellow moderately severe, and red most severe.

Based on these comparisons, our personal ranking for centration in refractive procedures is as follows:

• An asymmetric offset (oversized by 0.2–0.3mm);
• Symmetric offset centration (oversized by 0.6–0.7mm) could be best for hyperopia (with large enough flaps);
• An asymmetric offset;
• Photopic pupil centration;
• Symmetric offset centration;
• Scotopic pupil centration.

An oversized asymmetric offset is recommended before oversized symmetric offset centration. Although both methods require more tissue removal, an asymmetric offset is oversized by 0.2–0.3 mm, representing +7% extra depth and a +14% larger volume, while symmetric offset centration is oversized by 0.6–0.7 mm, representing +21% extra depth and a +46% larger volume. This results in an oversized asymmetric offset still saving −12% depth and −22% volume compared to oversized symmetric offset centration. Eyes with a pupil offset usually show relevant amounts of coma (and trefoil) as corneal aberrations (in the corneal wavefront), essentially due to decentered optics. By using an asymmetric offset, one includes some amount of coma and trefoil in the profile due to the decentered optics of that particular eye, adding natural customization to the treatment. The optical quality of the human eye varies across the visual field. The directional sensitivity of cone photoreceptors (the Stiles–Crawford (SC) effect) to light passing through near the edge of the pupil enhances human vision by reducing the sensitivity to visual stimuli for light that exhibits significant optical aberrations and diffraction. The computed modulation transfer function and the contrast sensitivity function with the SC effect modeled optically are superior to those without the SC effect [8,9]. Wavefront-guided ablations show less increases in coma and higher-order aberrations despite a comparatively smaller optical zone compared to conventional LASIK techniques [10]. In theory, even under consideration of the SC effect and wide-field vision (as opposed to on-axis foveal vision), an ideal OZ covering the pupil with the widest entrance may be helpful in avoiding patient-reported glare and has been shown to result in improved clinical outcomes [11]; this may be as important as the centration reference.