Influence of Cap Thickness on Corneal Curvature and Corneal Biomechanics After SMILE: A Prospective, Contralateral Eye Study

Patients and Methods

Examinations and Measurements

All patients underwent a thorough ophthalmic examination that included measurement of intraocular pressure with a non-contact tonometer, ultrasound pachymetry, slit-lamp examination, and fundus examination. MRSE, uncorrected distance visual acuity (UDVA), CDVA, and corneal Scheimpflug tomography (Pentacam HR; Oculus Optikgeräte) were measured preoperatively and at 1 day and 1, 3, and 6 months postoperatively. The corneal curvature was measured as the mean keratometry (Km) at 2-, 4-, and 6-mm zones from the pupil center with Pentacam software (Figure A, available in the online version of this article). The Corvis ST was used to evaluate the changes in corneal biomechanics at 6 months postoperatively.

Figure A. Anterior and posterior surfaces of the corneal curvature measurement codes.

Results

One patient was not followed up at 1 week and 1 month postoperatively and another was not followed up at 1 week postoperatively. The baseline characteristics are compared between the two groups in Table 1. The only statistically significant between-group difference was in residual stromal bed thickness (P < .05).

Table 1: Comparison of Baseline Characteristics Between the Eyes in the Two Study Groups (N = 50)

There was no significant between-group difference in UDVA at any time point (P > .05; Table 2). At 1 day, 1 week, and 1, 3, and 6 months postoperatively, MRSE was slightly smaller in the 110-µm group (−0.193 ± 0.474, −0.128 ± 0.363, −0.122 ± 0.468, −0.135 ± 0.367, and −0.16 ± 0.383 D, respectively) than in the 140-µm group (−0.311 ± 0.444, −0.234 ± 0.368, −0.23 ± 0.436, −0.218 ± 0.384, and −0.255 ± 0.348 D, respectively); none of the differences were significant (P = .086, .068, .101, .111, and .05, respectively). The safety and efficacy indices were 1.07 ± 0.11 and 1.03 ± 0.12, respectively, in the 110-µm group and 1.08 ± 0.11 and 1.04 ± 0.12 in the 110-µm group. The outcomes of refractive surgery are shown in Figure 1.

Table 2: Comparison of Uncorrected Distance Visual Acuity After Small Incision Lenticule Extraction Between the Two Study Groups

Figure 1. Standard six graphs comparing the refractive results between the two study groups. (A) Uncorrected distance visual acuity (UDVA) outcomes, (B) change in corrected distance visual acuity (CDVA), (C) attempted spherical equivalent refractive change plotted against the achieved spherical equivalent refractive change at 6 months postoperatively, (D) spherical equivalent refractive accuracy, (E) refractive astigmatism at 6 months postoperatively, and (F) mean spherical equivalent plotted as a function of time. D = diopters

The postoperative changes in anterior surface keratometry (ΔKm-ant) in the 4-mm zone at 1 day and 1, 3, and 6 months were significantly higher in the 110-µm group (3.670 ± 0.635, 3.824 ± 0.658, 3.714 ± 0.675, and 3.693 ± 0.671 D, respectively) than in the 140-µm group (3.566 ± 0.665, 3.723 ± 0.631, 3.578 ± 0.649, and 3.593 ± 0.645 D; P = .043, .045, .003, and .049, respectively). At 3 months postoperatively, the ΔKm-ant in the 6-mm zone was significantly higher in the 110-µm group than in the 140-µm group (3.539 ± 0.569 vs 3.455 ± 0.574 D; P = .035), with no significant difference at any other follow-up point (Figure 2). There was no statistically significant change in posterior surface keratometry after SMILE in either group.

Figure 2. Comparison of changes in keratometry values on the anterior surface postoperatively (ΔKm-ant) (n = 48). *P < .05. D = diopters

There was no significant difference in any of the corneal biomechanical parameters between the study groups before surgery (P > .05). However, by 6 months after surgery, there were significant changes in all of these parameters after SMILE except for the time until highest concavity. Significant differences were found in second applanation length (AT2), peak distance at highest concavity, biomechanically corrected intraocular pressure, integrated radius, and deformation amplitude at highest concavity (DA) ratio postoperatively between the two groups (Table A, available in the online version of this article). Furthermore, the changes in AT2, DA, and integrated radius were smaller in the 110-µm group than in the 140-µm group (Table 3).

Table A: Comparison of Preoperative and Postoperative Corneal Biomechanical Parameters in Each Study Group and Between the Two Groups (N = 50)

Table 3: Comparison of Change in Corneal Biomechanical Parameters Between Before and 6 Months After Surgery in the Two Study Groups (N = 50)

Discussion

We found that patients with a 110-µm cap thickness showed greater changes in the curvature of the anterior surface in the 4-mm zone at each follow-up point and in the 6-mm zone at 3 months after surgery. The study findings suggest that cap thickness affects the change in curvature of the anterior surface. Moreover, the changes in AT2, DA, and integrated radius in the group with a cap thickness of 110 µm were smaller than those in the group with a cap thickness of 140 µm. To our knowledge, this is the first contralateral eye study to compare the changes in corneal curvature at different zones from the pupil center and in corneal biomechanics after SMILE using the new parameters provided by the Corvis ST.

The study found no significant postoperative difference in terms of MRSE and visual acuity. This is consistent with the findings of Liu et al.,⁹ who compared cap thicknesses of 120 and 140 µm and found that cap thickness has no effect on MRSE or visual acuity. Liu et al.¹⁰ reported that patients with a 110-µm cap had better visual acuity than those with a 150-µm cap in the first 24 hours after surgery but that the difference had disappeared at 1 week after surgery. They proposed that modification of the corneal shape would be easier using a 110-µm cap thickness. However, in the current study, there was no significant difference in early visual acuity between the two study groups, possibly because there was only a slight (30-µm) difference between the two corneal cap thicknesses used.

Lee et al.¹¹ suggested that the thicker the cap, the more difficult it is to flatten the anterior surface, and that it is necessary to add MRSE correction to obtain the same flattening of the corneal anterior surface curvature and the same refractive correction outcome. We similarly found that the change in anterior surface curvature was smaller for a thicker cap in the 4-mm zone, but also found that there was no significant effect of cap thickness in the 2-mm and 6-mm zones at 6 months postoperatively. A previous study showed that the anterior cornea is important for maintenance of the corneal curvature.² Furthermore, Müller et al.¹² found that the interwoven arrangement of the collagen lamellae was responsible for the stability of the anterior corneal curvature. Patients with a smaller cap thickness showed more variation in anterior corneal curvature after surgery, which may have been due to the removed lenticule including more of the anterior stroma. Furthermore, although not statistically significant, the MRSE was slightly smaller in the 110-µm group than in the 140-µm group (approximately 0.10 D), which can be explained by the difference in ΔKm-ant of approximately 0.10 D at the 4-mm zone. Güell et al.¹³ found that MRSE correction needed to be increased by 3% for every 10-µm increase in corneal cap thickness to compensate for the possible loss of laser energy. The laser energy was set up at 130 nJ in the study by Güell et al. and at 145 nJ in the current study. It is possible that when the laser energy setting is high enough, the effect of energy loss will be reduced because of the increase in thickness of the corneal cap. The different laser energy setting may explain why there was no statistically significant effect of cap thickness on MRSE in the current study. Our study also found small but statistically insignificant changes in the keratometry of the posterior surface at the 2-, 4-, and 6-mm zones after surgery, indicating that a cap thickness in the range of 110 to 140 µm does not interfere with the stability of the posterior surface. Therefore, we hypothesized that there would be less change in the curvature of the anterior surface and the same degree of refractive error correction, at least in the 4-mm zone, if a thicker cap is used, and that this slight difference would not affect MRSE or UDVA.

In this study, we found statistically significant changes in all biomechanical parameters except for time until highest concavity, establishing that most of these parameters can reflect changes in corneal biomechanics after SMILE, which is consistent with previous studies.^8,14 Wu et al.¹⁵ found that the changes in some corneal biomechanical parameters (A1T, A2T, peak distance at highest concavity, and DA) after SMILE were similar for cap thicknesses of 110 and 130 µm, as did Liu et al.¹⁰ for cap thicknesses of 110 and 150 µm. However, in our study, there were statistically significant differences in 5 of 14 postoperative corneal biomechanical parameters between the groups and the changes in AT2 and DA were smaller than those in the group with a cap thickness of 110 µm. Moreover, we used the new dynamic corneal response parameters to compare the changes in corneal biomechanics between our two study groups. The integrated radius is the integrated area under the curve of the inverse concave radius, and the higher the integrated radius, the softer the cornea.^5,8 We found that the change in integrated radius in the 140-µm cap group was larger than that in the 110-µm cap group. However, any conclusions must be drawn with caution. A mathematical model predicted that the postoperative corneal tensile strength of a thicker corneal cap was greater than that of a thinner one when the preoperative central corneal thickness and thickness of the lenticule were the same, but this model did not consider the effect of the side incision.¹⁶ The thicker the corneal cap, the deeper the side-cut incision. Kohlhaas et al.¹⁷ and Scarcelli et al.¹⁸ demonstrated that the tangential tensile strength was greater for the anterior stroma than for the posterior stroma using stress-strain mechanical testing and Brillouin optical microscopy, respectively. In our study, the mean preoperative spherical equivalent was −4.90 ± 0.96 D and the mean preoperative central corneal thickness was 551.0 ± 25.4 µm regardless of whether the corneal cap thickness was 110 or 140 µm, and the lenticule was removed mainly in the anterior one-third of the stromal lamellae. Designing a thicker corneal cap to preserve more of the anterior stromal lamellae has no clear advantage in terms of reducing the changes in biomechanics that can occur after SMILE.

This study has some limitations. In particular, the number of samples was small and the follow-up duration was short. Further clinical studies that include a larger number of samples and a longer follow-up duration are necessary to confirm our result. Moreover, given that there is little relevant research that includes the corneal biomechanical parameters investigated in this trial, a further study on Corvis ST parameters will be needed to help explain our findings.

We found that use of a thicker corneal cap is associated with less change in the anterior surface curvature without any effect on visual acuity or correction of refractive error.