Laser-assisted in situ keratomileusis (LASIK) and CRT for orthokeratology are two techniques that attempt a complete independence from conventional compensation as spectacles or traditional contact lenses in myopic patients.1–3 Both techniques use a similar principle to achieve myopia correction, flattening the anterior corneal surface (ACS), thus reducing the total power of the eye but they are substantially different in the way they produce such effect. LASIK removes stromal tissue while CRT does a redistribution of the corneal tissue. In both cases the peripheral cornea is supposed to remain unchanged.4

There is an increase in corneal asphericity, changing from its initially prolate shape (Q<0) to an oblate contour (Q>0), being flatter in the centre than in the paracentral zone surrounding the treatment area.1,5–8 However even if the ACS has been classically defined by a unique value of corneal asphericity, or two values corresponding to the orientations of the two principal meridians,9 corneal asphericity changes significantly depending on the peripheral reference points. An evidence of this is the different values of Q that can be obtained in the same individuals using different corneal topographers. In a previous study we have observed those differences in Q values for different reference diameters and the different shift of Q values from the more central to the more peripheral reference points according to the corneal astigmatism.10 It is also expected that these multi-aspheric concept of the cornea can also differ significantly depending on the corneal refractive treatment being performed.

With the development of techniques for measuring the optical quality of the eye, several studies have allowed a better knowledge of the optical quality of the corneal surface after LASIK3,11,12 or orthokeratology.13,14 Both refractive techniques significantly increase the ocular higher order aberrations,1 particularly third and fourth order aberrations.13,15,16

Alterations in corneal asphericity and the corresponding increase in optical aberrations have a significant impact not only in the quality of vision,17 but also on contrast sensitivity12,14 and other visual functions as night vision disturbances.3

Since the treatment zone varies according to treatment technique and the cornea's response to the different correction procedures,4,18 and because the cornea possesses different degrees of asphericity according to the corneal zone being analysed,10 it is important to study asphericity values for different corneal diameters in order to fully characterise this important property that defines the post-surgical corneal contour and evaluate its impact on the higher order aberrations induced as a consequence of such changes.19 Furthermore, Patel's analysis of the corneal shape within the apical zone of operated eyes and normal eyes suggests that the corneal contour would be better defined using different conic sections with different shape factors depending on the corneal region to be represented.20

Thus, the purpose of this study was to evaluate corneal asphericity for different corneal diameters after refractive treatments (CRT, LASIK SL and LASIK CL), as well as to correlate these values with corneal high order aberration (cHOA). To the best of our knowledge, this is the first report characterizing the multi-aspheric description of the cornea after surgical and non-surgical corneal refractive interventions for myopia correction.

The analysis of corneal Q for different diameters and its alterations as a consequence of surgical and non-surgical procedures for visual compensation allows us to obtain a more complete description of post-LASIK and post-CRT corneal contour, as well as to better differentiate how each of these strategies for myopia compensation works. The authors have recently shown that in normal corneas, Q becomes more negative as it is more peripherally analysed, especially for more astigmatic corneas.10 However, the differences for normal corneas are small varying from −0.10 to −0.20 for corneas with astigmatism below 3 diopters and from −0.15 to −0.35 for those with astigmatism over 3 diopters. These values refer to Q calculations computed from 3mm (Q3) and 7mm (Q7) reference diameters. This demonstrates that even for the normal, non-treated cornea, a single value of corneal asphericity cannot suffice to accurately describe the corneal topography.

After the interventions evaluated in this study the cornea presents positive Q values, and these values become lower as the reference point for calculation moves towards periphery. In the case of refractive surgery, for an 8mm diameter (Q8), Q even becomes negative, whereas for ortokeratology it remains positive. However, it comes to attention that the differences are much larger for corneas that have undergone ortokeratology with alterations in the value of Q from 0.9 for Q3 to 0.17 for Q8. These results reveal another interesting phenomenon, which is the fact that post-treatment outcomes of Q vary considerably according to the corneal zone analysed for the different treatments. In clinical terms a given instrument (videokeratoscope) can apparently indicate a post-surgical prolate shape if takes the reference point more towards periphery and other could show an oblate shape if considering a more central reference point. This can also limit our ability to identify difference in the outcomes of different refractive treatments; for example, for a corneal topographer considering the peripheral zones at 4mm (equivalent to the Q8 in this study, differences in Q will be masked showing no statistical significant differences among treatments). Conversely, the differences are maximized as we go closer to the centre of the treatment zone (i.e. Q3). Therefore, this multi-aspheric analysis of Q allows us to obtain more representative and “unbiased” information about post-treatment corneal Q-value irrespective of the instrument used to obtain the measurements. In a certain way it will act as a normalization procedure.

As far as current knowledge is concerned, no studies are known that simultaneously analyse the effects caused by standard LASIK surgery, by customized LASIK surgery and by orthokeratology on cHOA and corneal asphericity (Q). The results of the present work allow us to characterize the changes in cHOA and asphericity confirming that different treatments have a significantly different impact on these descriptors of corneal shape and optical properties. One of the purposes of this study was the analysis of the optical quality of the corneal front surface among these three techniques, an effort was made to ensure that baseline values of refractive error were comparable for the different groups, since several studies have reported increases of high order aberrations according to the amount of refractive error.13,26,27

As seen on the results (Fig. 1), in the post-treatment we found higher increase of high order aberrations in CRT compared to LASIK surgery. Corneal Q also changed for all three techniques from the initially prolate shape to positive values (oblate shape). The alterations in corneal Q after the treatments represent an important impact on HOA and especially on spherical-like aberrations and, therefore, on the optical function of the eye14,28,29 and its visual experience as spherical aberration is one of the most relevant optical errors influencing on the degradation of the image quality.30,31

Similar results were found by Anera et al., who analyzed mean values of corneal Q for 24 eyes submitted to LASIK surgery. In their study they found an initial Q of −0.12 changing to a mean value of +0.41 3 months after surgery.6 The study recently published by Anera et al., in which they analyse the differences between LASIK and CRT with pupil values of 5mm, for values of refractive error, age and number of eyes similar to ours, they have found increases in third, fourth and fifth order aberrations, for both emmetropization processes, with these increases being higher for CRT. In the same study the Q values found were higher after CRT (+0.45±0.42) than after LASIK (+0.13±0.12). This is in agreement with the results of the present study, although those authors have not performed the analysis for different Q zones.14

Our results show that spherical-like aberrations are those which suffer the highest increase after the treatments (fourth order and spherical-like aberrations). This is not surprising considering the limitations in pre-treatment astigmatism, thus reducing the potential increase in secondary astigmatism and other aberrations, and the requirement of well centred treatments thus reducing the amount of coma-like aberration. However, the values of coma-like aberrations were higher for CRT cases, which agrees with the fact that for this therapy there is a certain degree of decentring compared with LASIK being responsible for such aberration.22 An alternative explanation for this difference could be found on potentially different reference points for both treatments. However, in our study, both LASIK and CRT treatments were intended to be centred on pupillary area and VolCT analysed the HOA with reference to pupil centre as well, thus equalizing this potential source of error among treatments. Both spherical-like and coma-like aberrations are those which produce a larger deterioration in the optical quality of the eye, whereas aberrations of a higher order (5th, 6th and above) exert a lesser influence on vision.30,31 The increase in spherical-like aberrations and the changes in Q are related by the different topographical profiles created after each intervention (smaller optical zone and higher corneal steepening at the edge of the optical zone for CRT compared to LASIK procedures).4 From our analysis, the higher correlations between spherical-like aberration and Q values obtained for different diameters were found for Q5, Q6 and Q7. This might be expected considering that the spherical-like aberration has been obtained for a 6mm diameter zone.

One of the limitations of this study was to have obtained corneal aberrations only before and after the treatments, and not having measured total optical aberrations of the eye. As is commonly known, total aberrations are the result of the combination of corneal and internal aberrations.32,33 However, it is to be expected that the alterations observed in total aberrations of the eye follow the same pattern than those observed on the ACS, where the largest aberrations occur, and such was demonstrated by Anera et al.’s14 recent study both for refractive LASIK surgery and CRT. In this study Q values obtained in each corneal location were averaged among individuals. As stated by de Ortueta and Arba-Mosquera, using simple arithmetic's to average asphericity of different individuals might present some limitations. The authors propose a mathematical methodology to compute Q values in order to provide a more consistent result of the average asphericity from different individual corneas.34 Moreover, both Q values and cHOA are obtained from the same data, thus must reflect similar changes as the present study shows. However, we still believe that there is a rationale to explore both of them. Q values show the mathematical representation of shape changes and are used in clinical practice quite frequently. Thus, it is important to highlight how they can vary depending on the area being analysed, particularly after reshaping procedures. This fact is even more critical after orthokeratology given the larger changes between Q3 and Q8. At the same time, cHOA provide an insight on the optical quality of the surface and provide information about the visual quality after these procedures. Another limitation of the present study is the potential decentration between the treated areas which are targeted regarding the pupil centre and the aberration analysis carried out with reference to the corneal centre as analysed by the corneal topographer. As reported by other authors, this differences can have an impact on comatic aberrations.35

The new concept of multi-asphericity for the ACS has been applied in defining the normal cornea by the authors10 and has now been applied to the modified corneas by means of different refractive procedures, surgical and non-surgical. Its results have different implications; a unique value of asphericity might confound the results of different refractive treatments whereas a more complex determination of asphericity at different diameters can elucidate significant differences in the behaviour of those treatments, and other field of application could be the definition of the functional optical zone of the cornea previously defined by others as the zone where the corneal curvature will vary only within a narrow interval of power.36 Finally, the multi-aspheric modelisation of the human cornea (before and after refractive interventions) will help to design optical devices that mimic the natural aspheric nature of the cornea for example with the purpose of fitting contact lenses or to develop optical devices that compensate or reinforce a certain desirable refractive pattern for the whole eye from the asphericity pattern of the ACS.

The present results show that there is deterioration in terms of multi-aspheric description of the ACS compared to the pre-treatment situation. Although the present study does not report data on the total wavefront aberration, it is expected that those changes reported here will be closely related to a degradation of the optical quality of the eye.14 That same deterioration found by Anera et al.14 was larger for the CRT group than for LASIK surgery which agrees with our results; in addition we have shown absence of significant differences between standard LASIK surgery and customized LASIK surgery regarding multi-aspheric corneal shape description after the interventions and the inherent aberrations generated.

In summary, literature reports values of corneal asphericity in normal populations that range from −0.01 to −0.80.37–40 Indeed, as shown here, both in normal but above all, in corneas altered by refractive procedures, the same instrument might report values of asphericity completely different.

The authors have no conflicts of interest to declare.