Comparison of visual outcomes after femtosecond laser-assisted LASIK versus flap-off epipolis LASIK for myopia

The refractive error of myopia is commonly corrected by eyeglasses, contact lens, implantable contact lens [1], and corneal refractive surgery [2]. In the early 1990s, photorefractive keratectomy (PRK) was first introduced for the surgical correction of myopia [3]; laser ablation refractive surgery was widely applied for anterior segment operation. With advances in the techniques used for epithelium removal, femtosecond laser-assisted in situ keratomileusis (femto-LASIK) and epipolis LASIK (epi-LASIK) have emerged as innovative approaches in the field of refractive surgery.

Depending on whether it was performed with or without flap creation using a microkeratome, the epi-LASIK technique is divided into two types: flap-on and flap-off technique. Ang RE et al. [4] and Zhang Y et al. [5] reported that flap-off epi-LASIK with mitomycin C (MMC) results in lesser pain and corneal haze, and faster visual recovery, while visual results, refractive outcomes, contrast sensitivity (CS), and higher-order aberrations (HOAs) were comparable with those of flap-on epi-LASIK.

Numerous studies have compared the visual outcomes of femto-LASIK and flap-on epi-LASIK (flap creation using a microkeratome). Greater corneal backscattering [6], faster recovery of corneal sensation, lesser degree of spherical aberration (SA), and some CS values [7], and superior outcomes of visual acuity were observed in an early stage [8] after femto-LASIK compared to flap-on epi-LASIK. However, Kezirian GM et al. [9] reported that femto-LASIK and flap-on epi-LASIK were associated with equivalent visual outcomes during the first 3 months postoperative period. Wen D et al. [2] performed a network meta-analysis to compare visual outcomes and quality between these two techniques and found that there were no statistically significant differences in either visual outcomes (efficacy and safety) or visual quality (HOAs and CS); however, they reported that the outcome of femto-LASIK was more predictable than any other type of surgery. Moreover, in the current study, the outcomes were evaluated by Pentacam, which uses a Scheimpflug camera to determine the corneal tomography and topography, thereby providing more detailed corneal biomechanical information [10‐12].

The aim of the present study was to compare the visual outcomes and corneal biomechanical properties changes between femto-LASIK and flap-off epi-LASIK.

Forty eyes of 27 patients were divided into two groups based on whether a flap was created by femtosecond laser during surgery (20 eyes, femto-LASIK) or not (20 eyes, flap-off epi-LASIK). The characteristics of the two groups are summarized in Table 1. There were no significant differences in the baseline ophthalmic characteristics between both groups.

Table 2 shows the comparative evaluation of the pre- and postoperative changes between the two groups. There were no significant differences between the two groups with regard to the flattest keratometry reading (K₁), steepest keratometry reading (K₂), CCT, or Q-value (Ant. and Post.). Differences between pre- and postoperative K₁, K₂, CCT, and Q-value (Ant.) were significant for both the groups (all P < 0.05 in femto-LASIK; all P < 0.001 in flap-off epi-LASIK).

Changes in the corneal thickness spatial profile (CTSP) are shown in Table 3. There were no statistically significant differences in preoperative and postoperative CTSP values between the two groups at corneal ring diameters of 0-mm, 2-mm, 4-mm, and 8-mm (all P > 0.05); however, it was significantly thinner in flap-off epi-LASIK than femto-LASIK at a ring diameter of 6-mm (P = 0.039) after surgery. Further details are shown in Table 3.

The changes in UDVA and CDVA are shown in Fig. 1. The mean changes in logarithm of the minimum angle of resolution (logMAR) UDVA (improvement) were significant in both groups, 2 years postoperatively: from 1.00 ± 0.31 logMAR to − 0.01 ± 0.02 logMAR in femto-LASIK and from 1.12 ± 0.45 logMAR to 0.00 ± 0.00 logMAR in flap-off epi-LASIK (all P < 0.001). The improvement was more significant for femto-LASIK at 1 day (0.03 ± 0.06 logMAR in femto-LASIK and 0.54 ± 0.31 logMAR in flap-off epi-LASIK) and 1 week postoperatively (0.02 ± 0.05 logMAR in femto-LASIK and 0.14 ± 0.13 logMAR in flap-off epi-LASIK) (P < 0.001 and P = 0.019). There were statistically significant differences in CDVA between femto-LASIK and flap-off epi-LASIK at 1 day (0.00 ± 0.00 logMAR in femto-LASIK and 0.07 ± 0.14 logMAR in flap-off epi-LASIK, P = 0.026) and 1 week postoperatively (0.00 ± 0.00 logMAR in femto-LASIK and 0.06 ± 0.08 logMAR in flap-off epi-LASIK, P = 0.009).

The mean preoperative spherical equivalent refraction values were − 5.94 ± 2.23 D (femto-LASIK) and − 5.94 ± 1.62 D (flap-off epi-LASIK), respectively (P = 0.904). The postoperative refraction showed significantly higher myopic refraction errors in flap-off epi-LASIK group than femto-LASIK at 1 day (0.03 ± 0.52 D in femto-LASIK, and − 0.84 ± 0.77 D in flap-off epi-LASIK, P < 0.001) and 1 week postoperatively (− 0.04 ± 0.56 D in femto-LASIK, and − 0.81 ± 0.98 D in flap-off epi-LASIK, P = 0.009), and there were statistically significant improvements in refraction errors in both groups from 1 day after refractive surgery (all P < 0.001) (Fig. 2).

Table 4 and Table 5 show the changes in HOAs of the front, back, and total cornea in femto-LASIK and flap-off epi-LASIK. There was a significant reduction in vertical coma (Z_3,-1) aberration (from − 0.086 ± 0.251 μm to − 0.393 ± 0.335 μm), horizontal secondary astigmatism (Z_4,2) aberration (from 0.013 ± 0.051 μm to − 0.113 ± 0.113 μm), and induction of SA (Z_4,0) (from 0.271 ± 0.132 μm to 0.479 ± 0.139 μm) between pre- and post-femto-LASIK in the front corneal HOAs (P = 0.021, P = 0.001, and P = 0.001, respectively). In terms of total corneal HOAs changes, there was a significant reduction in vertical coma (Z_3,-1) aberration (from − 0.128 ± 0.215 μm to − 0.368 ± 0.328 μm), horizontal secondary astigmatism (Z_4,2) aberration (from − 0.007 ± 0.055 μm to − 0.122 ± 0.117 μm), and induction of SA (Z_4,0) (from 0.168 ± 0.061 μm to 0.430 ± 0.137 μm) between pre- and post-femto-LASIK (P = 0.007, P = 0.004, and P < 0.001, respectively). However, in terms of back corneal HOAs changes, there was a significant induction of vertical coma (Z_3,-1) aberration, (from 0.013 ± 0.025 μm to 0.027 ± 0.027 μm), reduction of oblique trefoil (Z_3,-3) aberration (from − 0.026 ± 0.042 μm to − 0.055 ± 0.037 μm), and oblique tetrafoil (Z_4,-4) aberration (from 0.006 ± 0.030 μm to − 0.008 ± 0.029 μm) between pre- and post-femto-LASIK (P = 0.015, P = 0.046, and P = 0.049, respectively). In flap-off epi-LASIK, there was only significant induction of SA (from 0.250 ± 0.128 μm to 0.626 ± 0.232 μm, and from − 0.156 ± 0.033 μm to 0.556 ± 0.227 μm) between pre- and postoperative in the front and total corneal HOAs (all P < 0.001). In the back corneal HOAs, there was a significant induction of horizontal secondary astigmatism (Z_4,2) aberration (from − 0.001 ± 0.016 μm to 0.007 ± 0.018 μm) and reduction of SA (Z_4,0) (from − 0.156 ± 0.033 μm to − 0.163 ± 0.037 μm) between pre- and postoperative periods (P = 0.027 and P = 0.011).

When we compared the postoperative corneal HOA changes between the two groups, the increment in SA (Z_4,0) was higher in flap-off epi-LASIK than femto-LASIK: 0.626 ± 0.232 μm and 0.479 ± 0.139 μm in the front cornea, 0.556 ± 0.227 μm and 0.430 ± 0.137 μm in the total cornea, respectively (P = 0.016 and P = 0.017). With regard to the back corneal HOAs, there were significant differences in vertical coma (Z_3,-1) aberration: 0.027 ± 0.027 μm (femto-LASIK) and 0.001 ± 0.034 μm (flap-off epi-LASIK); horizontal secondary astigmatism (Z_4,2) aberration: − 0.008 ± 0.012 μm (femto-LASIK) and 0.007 ± 0.018 μm (flap-off epi-LASIK); oblique tetrafoil (Z_4,-4) aberration: − 0.008 ± 0.029 μm (femto-LASIK) and 0.015 ± 0.026 μm (flap-off epi-LASIK), respectively (P = 0.018, P = 0.007, and P = 0.022, respectively) (Fig. 3).

Many studies have investigated whether flap creation using a femtosecond laser (femto-LASIK) is more effective than that using a microkeratome (flap-on epi-LASIK) [6‐9]. However, in the present study, we compared the outcomes between femto-LASIK and flap-off epi-LASIK. Previously, Kalyvianaki MI et al. [14] reported that flap-on epi-LASIK and flap-off epi-LASIK produced equivalent visual and refractive results for the treatment of low and moderate myopia. Furthermore, Na KS et al. [15] found that flap-off epi-LASIK yielded superior visual recovery and corneal re-epithelialization than flap-on epi-LASIK surgery in the early postoperative period.

Corneal haze with decreased corneal transparency is typically determined by corneal backward light scattering. It has been reported that ablation volume may increase the degree of backscattering [16], and cases of severe myopia that require more ablation may require a higher dose of MMC during the refractive procedure [17, 18]. Sia RK et al. [19] and Chen J et al. [20] reported that MMC was beneficial for the reduction of corneal haze, without delaying epithelialization. The present study demonstrated little difference between the two techniques. Significantly better visual and refractive outcomes were associated with femto-LASIK than flap-off epi-LASIK at 1 day and 1 week postoperatively, with no additional significant differences during the remaining follow-up.

Myopic or hyperopic refractive surgery aims to correct the corneal shape by changing the keratometric power [4, 21]. Huang J et al. [22] and Jain R et al. [23] obtained high degree of repeatability for corneal curvatures after LASIK using a Scheimpflug camera, with no significant difference between the automatic and manual keratometric readings [24]. In this study, we used the Scheimpflug camera to evaluate the outcomes after refractive surgery. We found that both procedures showed a statistically significant decrease in CCT, keratometry readings, and ACD values after surgery. Dai ML and associates [25] reported that the ACD was shallower in LASIK than in non-operated myopic eyes.

The surface ablation technique can help avoid numerous surgical complications arising from the creation of a lamellar corneal flap required in LASIK, and can theoretically provide more stable corneal biomechanics. Shih PJ et al. [26] demonstrated corneal biomechanical simulation of stress concentration after refractive surgery, and they proposed that both surface and stromal ablation techniques caused stress in an obliquely downwards direction after surgery. We postulated that these changes of corneal biomechanical properties may influence the changes in corneal SA after corneal refractive surgery.

The concept of CTSP was first introduced by Ambrosio R Jr. et al. [27]. Furthermore, Buhren J et al. [28] found that the posterior aberrations and thickness spatial profile data did not markedly improve discriminative ability over that of anterior wavefront data alone. In our study, we used CTSP to evaluate changes in corneal thickness at different corneal diameters. We found that CTSP changes were significantly smaller in flap-off epi-LASIK than femto-LASIK at a corneal ring diameter of 6-mm; the CTSP changes in the central region were greater than that at the mid-periphery. In addition, the corneal HOAs at the 6.5-mm diameter were significantly different in the front and total HOAs of SA, while few significant differences were found in posterior HOAs of vertical coma aberration, oblique trefoil aberration, and oblique tetrafoil aberration. We postulated that these changes in CTSP may influence the changes in corneal HOAs, and may also affect the Q-value (8 mm) changes after LASIK, in a manner dependent on the size of the optical zone being treated.

The effect of SA on the depth of focus has been investigated using adaptive optics systems [29]. The depth of focus, by definition, is relatively insensitive to focal length and subject distance for a fixed f-number. Typically, myopia is a condition in which light is focused in front of the retina rather than on it. However, corneal refractive surgery is a type of refractive surgery that ablates the corneal tissue to change the accommodation power. Wallace HB et al. [30] found that ACD was significantly reduced by 0.10 mm with accommodation, and statistically significant changes in corneal curvatures were seen in all participants with accommodation.

The principle of refractive surgery is to induce positive SA shifts for the correction of myopia, and negative shifts for hyperopic correction [31, 32]. Moreover, the concept of the SCHWIND Amaris 750S excimer laser involves using the optimized aspheric profile [13] to prevent surgically induced HOAs, especially SA and coma aberration. Although the amount of corneal SA and asphericity are intrinsically related, they provide a 2:1 correspondence between corneal and ocular SA [33]. However, in the present study, there was significant increment in SA: 0.479 ± 0.139 μm in femto-LASIK and 0.626 ± 0.232 μm in flap-off epi-LASIK, and the logMAR UDVA achieved − 0.01 ± 0.02 logMAR in femto-LASIK and 0.00 ± 0.00 logMAR in flap-off epi-LASIK at 2 years postoperatively.

Total corneal refractive power involves compensation for negative posterior refractive power by positive anterior refractive power. Steepening of the anterior corneal surface increases the positive refractive power; when both surfaces bulge similarly, the anterior surface induces far greater absolute refractive changes than the posterior surface. According to our results, the patterns of corneal HOA changes were similar, while changes in front and total corneal HOAs were significantly different after both corneal refractive surgeries.

The induced changes in corneal asphericity (Q-value) and SA after laser ablation are key factors associated with the selection of candidates for refractive surgery. Scheimpflug imaging provided reliable measurements, consistent with those reported in the literature; there was a positive change in the Q-value of the anterior surface after myopic ablation and a negative change after hyperopic ablation [34].

Corneal aberrations are usually positive, aberrations of the lens are usually negative, and the total SA changes more than other HOAs with accommodation. Moreover, ocular wavefront aberrations are primarily created in the cornea and lens, and are strongly affected by several factors, including the accommodative state [35], pupil diameter [36], tear film [37], age [38], and pupil entrance decentration [39]. We found a statistically significant difference in postoperative SA between the two different surgical techniques, but found no clinically significant difference up to 2 years postoperatively; femto-LASIK produced superior visual outcomes to flap-off epi-LASIK in the early postoperative stage.

A meta-analysis shows that there were no statistically significant differences in either visual outcomes or visual quality between different corneal refractive surgery techniques, and that femto-LASIK shows a better predictability than any other type of surgery. However, this study was limited by the small sample size; therefore, studies involving a larger population of patients are necessary to ensure more dependable results [40].

Refractive surgery has been regarded as an excellent surgical option, negating the need for contact lenses or glasses. Our study results indicated that both femto-LASIK and flap-off epi-LASIK were safe, effective, and predictable refractive surgeries. Femto-LASIK would be a better surgical option that provides lesser postoperative SA after surgery and superior visual outcomes in the early postoperative stage. Preoperative corneal thickness should be considered when choosing corneal refractive surgery in clinical practice.