A Comparative and Prospective Study of Corneal Consumption and Corneal Biomechanics After SMILE and FS-LASIK Performed on the Contralateral Eyes With High Myopic Astigmatism

Abstract

Purpose: The purpose of this study was to compare the actual corneal consumption and subsequent corneal biomechanical changes after small incision lenticule extraction (SMILE) and femtosecond laser-assisted in situ keratomileusis (FS-LASIK) in eyes with high myopic astigmatism.

Methods: In this prospective, randomized study on high myopic astigmatic eyes, 74 patients with similar refractive error received SMILE in one eye and FS-LASIK in the contralateral eye were enrolled. The central stromal reduction was calculated by preoperative and postoperative central stromal thickness (CST) measurements by the RTVue optical coherence tomography (OCT). Changes in corneal volume (CV) were evaluated by Pentacam HR. The corneal biomechanical parameters were measured with Corvis ST II.

Results: At the 6-month postoperative visit, the mean actual central stromal reduction achieved with SMILE was significantly less than that achieved with FS-LASIK (108.23 ± 16.52 µm vs. 122.66 ± 27.33 µm, P < 0.001). The residual central corneal thickness (CCT), CV 3 mm, CV 5 mm, and stiffness parameter at first applanation (SP-A1) after SMILE were all greater than those of FS-LASIK (all P < 0.05). Moreover, the residual CCT, CV 3 mm, and CV 5 mm were positively correlated with Ambrósio relational thickness through the horizontal meridian (ARTh) and SP-A1, while negatively correlated with the integrated radius (IR) and deformation amplitude (DA) ratio 2.0 mm in both the SMILE and FS-LASIK groups (all P < 0.05).

Conclusions: In patients with high myopia astigmatism, SMILE required fewer corneal stromal tissues and thus better corneal biomechanics than FS-LASIK when correcting a similar spherical equivalent refraction. Moreover, SP-A1 may be a more sensitive indicator for evaluating corneal biomechanical changes after refractive surgeries.

Translational Relevance: Fewer corneal stromal consumption with SMILE surgery may contribute to better corneal biomechanics in high myopic astigmatic eyes in comparison to FS-LASIK, which could provide reliable evidence for the appropriate choice of refractive surgery for patients with high myopic astigmatism.

Introduction

As the most commonly performed laser refractive surgical techniques, small-incision lenticule extraction (SMILE) and femtosecond laser-assisted laser in situ keratomileusis (FS-LASIK) have been proven to be safe, effective, predictable, and stable for correcting myopia and myopic astigmatism.^1–3 Both surgical methods have their advantages and disadvantages, and, in clinical practice, the appropriate surgical treatment could be selected according to the characteristics of the patients. However, there is no consensus on which surgical method is more suitable for patients with high astigmatism. Moreover, few studies have been conducted on patients with high astigmatism, thus further research is still needed. 

Corneal consumption and residual corneal stromal thickness (RST) are important indicators for postoperative safety and stability evaluation. However, inconsistencies in the planned central stromal reduction with achieved central stromal reduction were observed in laser refractive surgeries.^4–8 For example, our previous study⁴ indicated that the central stromal reduction was overestimated by 20.05 ± 5.92 µm with SMILE and underestimated by 8.21 ± 8.14 µm with FS-LASIK in patients with high myopia. Whereas Liu et al.⁹ revealed that the central corneal thickness (CCT) reductions were overestimated by 23.06 ± 6.97 µm and 28.29 ± 13.92 µm in patients with myopia with SMILE and FS-LASIK, respectively. We speculated that these studies mainly used interindividual comparisons, which could lead to interpatient variability biases. Herein, a rigorous pared-eye design was needed to yield reliable results regarding which surgery suits astigmatism better. Moreover, in cylinder correction, stromal ablation concentrated along the flat median other than the steep median, leaving an unsymmetric residual stromal profile.¹⁰ Given this phenomenon, corneal volume (CV), which is measured by the Pentacam HR, might be more sensitive and comprehensive in reflecting the actual changes of corneal tissues than corneal thickness alone. In previous studies, CV has been used to evaluate the corneal tissue changes upon refractive surgery¹¹ and predict and diagnose keratoconus^12,13 clinically. Thus, quantifying CV in astigmatic eyes could provide more detailed evidence to clarify whether SMILE needs to consume more cornea tissue than FS-LASIK in cylindrical correction. 

In addition, corneal ablation accompanied by biomechanical changes occurs during refractive surgeries, which might increase the risk of corneal ectasia. Several recent studies^14–19 have reported inconsistencies in the comparison of corneal biomechanical properties between SMILE and FS-LASIK. Wu et al.¹⁴ demonstrated a higher corneal hysteresis (CH) and a corneal resistance factor (CRF) in SMILE than those in FS-LASIK by the Ocular Response Analyzer (ORA). Abd El-Fattah et al.¹⁸ revealed a better corneal biomechanical performance (greater integrated radius [IR] and stiffness parameter at first applanation [SP-A1]) in SMILE among high myopia by Corvis-ST II, compared with FS-LASIK. Nevertheless, no significant difference in postoperative corneal biomechanical strength was observed between SMILE and FS-LASIK in the findings of Sefat et al.¹⁶ and Liu et al.¹⁷ using the Corvis-ST II. These discrepancies might be due to individual bias and different measuring devices.²⁰ The ORA and Corvis-ST have been designed to evaluate corneal biomechanical properties in vivo. However, the Corvis-ST could monitor the process of corneal deformation in real-time with a high-speed Scheimpflug camera and then could provide more corresponding corneal biomechanical parameters compared to ORA.²¹ These new parameters could better evaluate the changes in corneal biomechanics after different corneal refractive surgery and the diagnosis of keratoconus.^22–24 Additionally, these abovementioned studies mostly focused on patients with myopia, however, central optical ablation is mainly designed for myopic correction during the refractive surgery, whereas more mid-peripheral ablation is involved in astigmatic treatment, which might affect corneal biomechanics differently. Accordingly, the corneal biomechanical changes in the surgical corrected astigmatic eyes and the difference between SMILE and FS-LASIK surgery still need further investigation. 

Therefore, we conducted this randomized contralateral eye study to eliminate the influence of individual differences, and objectively evaluate the actual changes in corneal consumption with CV and CCT and biomechanical changes in patients with high myopic astigmatism who underwent the SMILE and FS-LASIK surgeries. 

Patients and Methods

Participants and Design

In this prospective, randomized, paired-eye study, we recruited 74 patients with myopia and myopic astigmatism between May 2022 and November 2022 at the Zhong Shan Ophthalmic Center, Sun Yat-Sen University. Each patient underwent SMILE in one eye and FS-LASIK in the contralateral eye. A random number generator assigned either SMILE or FS-LASIK to an eye. This study complied with the tenets of the Declaration of Helsinki and was approved by the Ethics Board of the Zhongshan Ophthalmic Center of Sun Yat-Sen University (Identifier No. 2020KYPJ159). The informed written consent was obtained from all participants. The inclusion criteria were as follows: age between 18 and 40 years, stable refraction for more than 2 years, the manifest refraction spherical equivalent (SE) less than −10.50 diopters (D), differences in SE and cylinder between the paired eyes ≤1 D, and a minimum RST more than 250 µm for SMILE and 300 µm for FS-LASIK. The exclusion criteria were as follows: history of ocular trauma or surgery, ocular surface diseases like severe dry eye, corneal degeneration, active ocular or systemic disease, and keratoconus or suspicious corneal topography. 

Patient Examination

Preoperative examinations included uncorrected distance visual acuity (UDVA) and corrected distance visual acuity (CDVA), manifest and cycloplegic refraction, slit-lamp and dilated fundus examination, corneal tomography and CV of a 10-mm diameter region using Scheimpflug camera (Pentacam HR; Oculus Optikgeräte GmbH, Wetzlar, Germany). Corneal biomechanics were measured using a Corvis ST II (software version 1.6r2015; Oculus, Wetzlar, Germany). CCT and central corneal epithelial thickness were measured by the RTVue XR optical coherence tomography (OCT) Avanti (RTVue OCT; Optovue). The actual amount of central corneal stromal thickness (CST) reduction was calculated as our previous study reported.⁴ All the patients were followed up at 1 week, and 1, 3, and 6 months after surgery. 

Surgical Technique

All surgical procedures were performed by the same fully qualified refractive surgeon (author K.M.Y.) on the same day. In the SMILE procedure, manual corneal marking was performed before surgery for accurate axis alignment, as described in our previous studies.^25,26 The VisuMax 500-kHz femtosecond laser (Carl Zeiss Meditec AG) was used to create a refractive lenticule and a 2-mm incision at 130 degrees with an intended cap thickness of 100 to 120 µm. The lenticule diameter was set as 6.0 to 6.8 mm and the transition zone at 0.1 mm during cylinder correction. Surgical procedures were centered on the coaxially sighted corneal light reflex (CSCLR). After the scanning procedure, the lenticule was dissected using a blunt dissector and extracted through the side cut. The interface was then flushed with a balanced salt solution. 

In the FS-LASIK procedures, the surgeon used the VisuMax femtosecond laser platform to create in situ keratomileusis flaps. The intended flap thickness was 95 µm and the flap diameter was varied from 8.1 to 8.5 mm with a superior hinge position. After the flap was lifted, ablation of the stromal bed was performed using the AMARIS 750S excimer laser (Schwind eye tech solutions; Kleinostheim, Germany) with pupil tracking enabled. Excimer ablation was subsequently performed with a similar programmed optical zone (POZ) as the paired eye in the SMILE procedure. 

Postoperative medications included topical 0.5% levofloxacin eyedrops (Tarivid; Santen Pharmaceutical, Osaka, Japan) and 0.25% tobramycin and dexamethasone eyedrops (Maxidex; Alcon Laboratories, Fort Worth, TX, USA) 4 times per day for the first week. Subsequently, 0.1% fluorometholone eye drops (Tarivid) were administered 4 times per day for 3 weeks. In addition, preservative-free lacrimal substitutes were used as needed. 

Corneal Biomechanics Measurement

The corneal biomechanical parameters were measured by a Corvis ST II analyzer (Oculus, Germany), which is a non-contact tonometer with high-speed Scheimpflug visualization of the dynamic process of corneal deformation under external air pulse. The Corvis ST II device noninvasively measured the following parameters: first applanation time (A1T), first applanation length (A1L), first applanation velocity (A1V), second applanation time (A2T), second applanation length (A2L), second applanation velocity (A2V), highest concavity time (HCT), highest concavity peak distance (HCPD), deformation amplitude ratio 2.0 mm (DA ratio 2.0 mm), integrated radius (IR), Ambrósio relational thickness to the horizontal profile (ARTh), SP-A1, stress-strain index (SSI), and corneal biomechanical index (CBI). All measurements were performed by the same experienced investigator (author X.T.H.) to eliminate possible interobserver variability. Each measurement was taken three times for good repeatability, and the best value was used for analysis. 

Statistical Analysis

All statistical analyses were performed using the IBM SPSS software (version 26.0; SPSS, Inc.). Mean ± standard deviation was used for quantitative variables. Data were tested for normality using the Kolmogorov–Smirnov test. Wilcoxon signed-rank tests were used to compare the nonparametric data (CBI) between SMILE and FS-LASIK preoperatively and postoperatively. Group comparisons for normally distributed data (corneal tissue and other corneal biomechanical parameters) were performed using the independent sample t-test, and preoperative and postoperative comparisons were performed using the paired t-test. The Bonferroni correction was used to adjust for multiple comparisons. Spearman correlation analysis was performed to evaluate the relationship between CCT, CV, and the biomechanical parameters postoperatively. The P values less than 0.05 were considered statistically significant. 

Results

Demographics

This study included 148 eyes of 74 patients with myopic astigmatism of greater than 2.00 D. Seventy-four eyes were treated with SMILE and the contralateral eyes were treated with FS-LASIK. The mean age of the enrolled patients was 23.70 ± 5.37 years (range = 17 to 39 years). Forty-five patients (61%) were men, and 29 patients (39%) were women. Twenty-six patients (35%) had a history of wearing contact lenses before surgery, including 18 patients with soft contact lenses and 8 patients with orthokeratology lenses. Eight patients had a history of eye rubbing. All surgeries were performed successfully with no observed complications. As is shown in Table 1, there were no significant differences in baseline characteristics, including CDVA, cylinder, SE, CCT, CV, and the optical zone diameter between SMILE and FS-LASIK (all P > 0.05). A larger attempted lenticule thickness (LT) and a lower predicted RST were observed in the SMILE group than those in the FS-LASIK group (P < 0.001). 

Table 1.

 

Demographic Data and Characteristics of Patients Treated With SMILE and FS-LASIK

 

Comparison of Planned and Achieved Central Stromal Thickness Reduction Between SMILE and FS-LASIK

As shown in Figure 1, preoperatively, the mean predicted CST reduction was 128.75 ± 19.63 µm in the SMILE group and 114.42 ± 22.03 µm in the FS-LASIK group (P < 0.001). However, at 6 months postoperatively, the mean actual CST reduction achieved with SMILE was significantly less than that achieved with FS-LASIK (108.23 ± 16.52 µm vs. 122.66 ± 27.33 µm, P < 0.001). The amount of CST reduction was overestimated by 20.52 ± 8.20 µm (range = –1 to 40 µm) in the SMILE group and underestimated by 8.58 ± 9.89 µm (range = –16 to 31 µm) in the FS-LASIK group (both P < 0.001). Thus, the residual CCT in SMILE was greater than that in FS-LASIK (424.45 ± 23.77 vs. 412.01 ± 30.12 µm, P < 0.001; see Fig. 1). Likewise, as shown in Figure 2, the CV 3 mm and CV 5 mm in SMILE were larger than those in FS-LASIK (3.28 ± 0.17 vs. 3.20 ± 0.21 mm³ and 10.03 ± 0.48 vs. 9.85 ± 0.56 mm³, respectively, both P < 0.05), whereas no significant difference was found in the CV 7 mm and CV 10 mm between the 2 surgeries at 6 months postoperatively (all P > 0.05). 

Figure 1.

 

Comparison of corneal thickness, corneal stromal thickness, actual corneal stromal reduction, and residual stromal thickness preoperatively and postoperatively between SMILE and FS-LASIK surgeries. ***P < 0.001, **P < 0.01, *P < 0.05, statistically significant for data between SMILE and FS-LASIK.

Figure 2.

 

Comparison of corneal volumes of different areas preoperatively and postoperatively between SMILE and FS-LASIK surgeries. ***P < 0.001, **P < 0.01, *P < 0.05, statistically significant for data between SMILE and FS-LASIK.

Comparison of Corneal Biomechanical Parameters Between SMILE and FS-LASIK

The corneal biomechanical parameters of SMILE and FS-LASIK pre- and postoperatively are shown in Table 2 and Figure 3. No significant differences in preoperative biomechanical parameters were observed between the SMILE and FS-LASIK groups (all P > 0.05). At 6 months postoperatively, HCR, ARTh, and SP-A1 decreased significantly, whereas HCPD, DA ratio 1.0 mm, DA ratio 2.0 mm, and IR increased significantly after SMILE and FS-LASIK (all P < 0.05). Moreover, the A1V and A1L were significantly smaller while SP-A1 was significantly larger in the SMILE group than those in the FS-LASIK group (all P < 0.05), whereas no significant difference in other biomechanical parameters was observed between the two surgeries (all P > 0.05). 

Table 2.

 

Corneal Biomechanical Parameters Before and at 6 Months After SMILE and FS-LASIK

 

Figure 3.

 

Comparison of corneal biomechanics preoperatively and postoperatively between SMILE and FS-LASIK surgeries. ***P < 0.001, **P < 0.01, *P < 0.05, statistically significant for data between SMILE and FS-LASIK.

Correlation Analysis Between Corneal Biomechanical Parameters and the CCT and CV

The results of the simple linear regression among postoperative CCT, CV, and the biomechanical parameters (DA ratio 2.0 mm, IR, SP-A1, ARTh, and SSI) are presented in Table 3. At 6 months postoperatively, the residual CCT, CV 3 mm, and CV 5 mm were all positively correlated with ARTh and SP-A1, while negatively correlated with the IR, DA ratio 2.0 mm, and CBI in the SMILE and the FS-LASIK groups (all P < 0.05). Moreover, the CV 7 mm was positively correlated with SP-A1 and negatively correlated with DA ratio 2.0 mm and CBI in both SMILE and FS-LASIK (all P < 0.05). CV 10 mm was weak but significantly positively correlated with SP-A1 in the SMILE group (r = 0.268, P = 0.030). Moreover, SSI was positively correlated with CCT, CV 3 mm, and CV 5 mm in the FS-LASIK group (all P < 0.05), whereas no significant correlation was found in the SMILE group (all P > 0.05). 

Table 3.

 

Correlation Analysis Between Corneal Biomechanical Parameters and CCT and CV Postoperatively

 

Discussion

Despite several studies have investigated the variations of corneal consumption^4,9,27–31 and corneal biomechanics^14–19 between the SMILE and FS-LASIK surgeries, the lack of standard comparison between techniques has led to varying reports with contradicting conclusions. Herein, our randomized paired-eye design with not only similar SE correction but also a similar POZ set by the same experienced refractive surgeon, eliminates individual bias as much as possible, which increases the credibility of the conclusions. This self-control-designed study shows that fewer corneal stromal consumption with SMILE surgery may contribute to better corneal biomechanics in high myopic astigmatic eyes in comparison to FS-LASIK. 

Our results reveal that SMILE consumes fewer corneal stromal tissues than FS-LASIK in correcting the same refractive error, establishing a corneal-saving role of SMILE surgery in high astigmatic eyes. Our study has clarified previous controversies.^4,27,29–31 For example, Luo et al.⁴ and Tian et al.²⁷ reported that the SMILE procedure could preserve more corneal tissue than FS-LASIK when correcting similar refractive errors. However, other scholars^29–31 reported contradictory conclusions that the reduction of CCT after SMILE surgery was higher than LASIK surgery. The discrepancy may be attributed to individual corneal responses to the femtosecond laser or measurement error. Moreover, interference from corneal epithelial remodeling was removed by analyzing corneal consumption based on corneal stromal thickness changes in our study. Additionally, notably, due to the asymmetrical ablation between flat meridians and steep meridians in high astigmatic eyes,³² this study further innovatively evaluated the CV to quantify the corneal tissue changes, which could characterize the corneal consumption pattern more comprehensively than CCT after laser refractive surgery.^11,33 Our results demonstrate that the postoperative changes in the CV of the central 5 mm cornea were consistent with the CCT, whereas for peripheral cornea, no significant difference was detected in CV 7 mm and CV 10 mm between SMILE and FS-LASIK, which may partly be supported by the finding of Wei et al.³³ that the CV mainly decreased in the peripheral region after SMILE while decreasing centrally following FS-LASIK. However, Schuh et al.²⁹ observed a significant overall greater loss of corneal volume of 7 mm diameter region in SMILE than FS-LASIK. The reason for the discrepancy may be explained by various excimer laser platforms, different surgical parameter designs, and different individual responses. Thus, the special and rigorous paired-eye design with a similar optical zone of the present study could further increase the credibility of our result that FS-LASIK surgery consumes more corneal tissue than that of SMILE surgery no matter in correcting myopia or astigmatism. 

In the present study, we also found that SMILE yielded a larger SP-A1 than FS-LASIK postoperatively, which indicated a better biomechanical strength after SMILE than FS-LASIK, which further clarified the inconsistencies in the previous literature.^14–19 Several recent studies^14,15,18 demonstrated that SMILE was superior in preserving corneal biomechanical integrity, whereas other studies^16,17,19 indicated otherwise, showing no significant differences in postoperative corneal biomechanics between SMILE and FS-LASIK. 

These discrepancies might be induced by the individual variation between the enrolled subjects, the difference between varied biomechanical examination devices, and different sensitivity of corneal biomechanics parameters. In our study, we measured biomechanical parameters such as the DA ratio 2 mm, SP-A1, IR, SSI, and so on, whereas we only found there were significant differences in SP-A1 between SMILE and FS-LASIK after surgery. Despite the SSI being a newly introduced corneal biomechanical parameter and could evaluate the corneal material properties,^34,35 SSI could not sensitively detect the biomechanical difference between the SMILE and FS-LASIK in the present study. Two factors may account for this phenomenon. First, SSI is almost independent of intraocular pressure and CCT. Second, Lu et al.³⁶ demonstrated that SSI showed relatively poor repeatability among the post-laser vision correction (LVC) cornea. Thus, the SSI may not be a sensitive indicator reflecting the biomechanical changes after refractive surgery. However, SP-A1 is defined as resultant pressure divided by deflection amplitude at the time of first applanation, which could strongly reflect the deformation resistance of the cornea.^37,38 Previous studies^24,39,40 have illustrated that SP-A1 could distinguish subtle biomechanical differences between mild keratoconus and normal eyes and found SP-A1 was significantly associated with DA ratio 2 mm, HCPD, and other corneal biomechanical parameters. This may be the reason why only SP-A1 detected the difference after SMILE and FS-LASIK in this study, suggesting that SP-A1 would be a sensitive biomarker for evaluating corneal biomechanics after laser surgeries. 

Furthermore, in the present study, we first investigate the relationship between postoperative CV in different regions and the corneal biomechanical parameters by Corvis-ST. The correlation analysis shows that CCT and CV 5 mm are significantly correlated with more corneal biomechanical parameters of high astigmatic eyes postoperatively, emphasizing the important role of central corneal volume in maintaining postoperative biomechanics. In light of this observation, SMILE surgery, which consumes less central corneal volume, may be safer than FS-LASIK in high astigmatic correction. The relatively better postoperative corneal biomechanics achieved in SMILE were partly associated with the greater postoperative CCT and CV in SMILE, compared with that in FS-LASIK. 

This study has some limitations. First, as Khamar et al.⁴¹ and Kamiya et al.⁴² described in their study, neither flap nor cap would sustainably affect biomechanical parameters after refractive surgery, thus the differences in thickness and diameter between the cap and flap in SMILE and FS-LASIK, respectively, were not taken into consideration. Second, Corvis ST measures corneal biomechanics based on a Scheimpflug image only on the horizontal meridian, which could be a bias when evaluating results of high astigmatic error corrections, with differential corneal consumption in different meridians. Third, the study had a relatively short follow-up period of 6 months, a longer follow-up with a larger sample size would allow a more thorough biomechanical comparison among more laser surgical procedures. 

In conclusion, in this prospective contralateral eye study, fewer corneal consumption and better corneal biomechanics were observed in the high astigmatic eyes following SMILE surgery, compared with that of FS-LASIK surgery in correcting a similar spherical equivalent. This study could provide reliable evidence for the appropriate choice of refractive surgery for patients with high myopic astigmatism. Moreover, SP-A1 could more sensitively detect the corneal biomechanical changes after corneal refractive surgery. 

Acknowledgments

Supported by grants from the Zhi Xing Shi Jie-Ophthalmic Clinical Research Support Project (Grant No. BFC-KECS-LC-20220930-2) and the Guangzhou Science Technology and Innovation Commission (Grant No. 202201011410). The research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest. 

Author Contributions: Study concept and design: X.T.H., J.Z., and K.M.Y.; data collection: X.T.H., P.C., N.Y., Y.Q.L., and H.W.; analysis and interpretation of data: X.T.H., N.Y., and K.M.Y.; writing the manuscript: X.T.H., P.C., and K.M.Y.; critical revision of the manuscript: J.Z. and K.M.Y.; supervision: J.Z. and K.M.Y. 

Disclosure: X. Hou, None; P. Chen, None; N. Yu, None; Y. Luo, None; H. Wei, None; J. Zhuang, None; K. Yu, None