Albeit rare, iatrogenic postoperative ectasia represents a major complication after laser visual correction.1 Although the evaluation of topography, tomography, epithelial thickness mapping, and corneal biomechanical properties have increased our ability to preoperatively detect corneal susceptibility,2,3 ectasia is occasionally being reported after corneal refractive laser surgeries.
With the recent emergence and successful implementation of the femtosecond laser, several advances have emerged in refractive surgery. In 2011, the small incision lenticule extraction (SMILE) technique was introduced, enabling the use of a femtosecond laser to create a stromal lenticule, which could be removed through a small peripheral incision.4 This minimally invasive technique is characterized by postoperative comfort5 and good predictability and safety when treating high myopia and myopic astigmatism.6 Eight years after the first reported outcomes, more than 2 million procedures have already been performed.4
From the biomechanical perspective, different surgical techniques affect the cornea differently. In 2013, a mathematical model estimated potential differences in postoperative tensile strength after photorefractive keratectomy (PRK), laser in situ keratomileusis (LASIK), and SMILE surgeries. In such a model, SMILE was the technique with the highest residual stromal tensile strength, followed by PRK and LASIK.7 In 2017, Lee et al.8 reported differences in biomechanically corrected intraocular pressure and dynamic corneal response parameters between PRK and femtosecond laser–assisted LASIK. In the same year, using two-dimensional stress-strain testing to characterize elastic and viscoelastic properties of a soft tissue, Spiru et al.9 demonstrated that SMILE is superior in preserving corneal biomechanics when compared to flap-based procedures. In 2019, using the same methodology, our group obtained similar biomechanical stability when identical refractive corrections were applied using either PRK or SMILE.3 Therefore, unlike mathematical model predictions,7 PRK and SMILE are biomechanically similar when measured in ex vivo human fellow eye corneas.3 Regardless, cases of ectasia after PRK or SMILE have been reported.10–14
In previous published reports, the ectasia after SMILE and PRK could be largely attributed to the presence of misidentified risk factors or even to preoperatively overlooked form fruste keratoconus.10,11,14,15 More recently, new parameters have been introduced and may improve patient selection in the future, even when currently known risk factors are negative12; the preclinical reduction of collagen or lysyl-oxidase (LOX) levels illustrate this new horizon for optimizing preoperative evaluations.12 Whether the risk factor is preoperatively identifiable or not, once ectasia is established postoperatively, cross-linking (CXL) is the procedure of choice for halting the progression.16
The biomechanical stability achieved after CXL has several potential pathways.17 Ultimately, CXL has the ability to increase corneal stiffness through an oxidative process within the stromal extracellular matrix and increase resistance of corneal stroma against enzymatic digestion.18 By doing so, CXL may be able to not only stabilize progressive forms of ectasia, but also often improve corneal shape and corrected visual acuity.19 Due to the different organization of collagen fibrils in the anterior and posterior stroma,20 as well as the selective removal of distinct stromal regions during PRK and SMILE, differences in CXL efficacy between corneas treated with the aforementioned techniques might be expected. In the current study, pairs of ex vivo human corneas were subjected to either SMILE or PRK and subsequently to mechanical stress cycles for preconditioning. Finally, CXL was performed and biomechanical properties were measured. The purpose of the current study was to evaluate the biomechanical stiffening effect of CXL in paired ectatic models of human corneas following SMILE or PRK.
Materials and Methods
A total of 26 human paired corneas from 13 individuals were obtained from different eye banks. The corneas were unsuitable for corneal transplantation and released for research purposes. Ultrasonic corneal pachymetry was performed on all corneas before both procedures. For standardization, PRK was performed on corneas from the right eye, whereas corneas from the left eye underwent SMILE. Afterward, the corneas were preconditioned to align collagen fibrils in the direction of the test load and increase the repeatability of measurements.21,22 CXL was performed and elastic modulus was determined in both groups.
To clarify nomenclature, the PRK-CXL group had refractive treatment with PRK before preconditioning and CXL, whereas the SMILE-CXL group had a SMILE refractive procedure before preconditioning and CXL. The steps are explained in detail below.
To standardize treatment, and because the epithelium does not contribute biomechanically, it was mechanically removed in all corneas using an artificial anterior chamber (Katena Products, Inc., Parsippany, NJ) and a surgical blade. This step also ensures that differences in epithelial transparency do not interfere with the laser procedure. In addition, intraocular pressure was controlled to approximately 20 mm Hg. An infusion system with an adjustable height of approximately 27 cm above the working level was used to achieve reproducible intraocular pressure in all eyes, and palpation with two fingers by the same surgeon was also performed. The MEL 90 excimer laser (Carl Zeiss Meditec AG, Jena, Germany) and the VisuMax 500-kHz femtosecond laser (Carl Zeiss Meditec AG) were used to perform PRK and SMILE, respectively. The refractive correction was −10.00 diopters (D) sphere and −0.75 D cylinder at 0° with a 7-mm zone using either surface ablation (PRK) or a 130-µm cap (SMILE). For SMILE, a medium-sized contact glass was used for corneal applanation and the energy profile used was 130 nJ with a 4.5-µm track/spot distance. The SMILE incision arc length was 3 mm. All surgical procedures were performed on the same day and the specimens were preserved in Optisol-GS storage medium (Bausch & Lomb, Rochester, NY) for transport. To reduce bias due to hydration status, paired corneas were subjected to different laser treatments 1 to 2 hours apart.
Corneoscleral buttons were mounted on a customized two-dimensional holder and placed within a stress-strain device (Z0.5; Zwick GmbH & Co., Ulm, Germany). Corneas from both groups underwent stretching (preconditioning) with two stress cycles up to 9 N (570 kPa stress). Then, CXL was performed on all corneoscleral buttons.
In both groups, accelerated epithelium-off CXL with total fluence of 5.4 J/cm2 was performed as described previously.23 In brief, 0.1% riboflavin solution (Vitamin B2; Streuli Pharma AG, Uznach, Switzerland) diluted in phosphate buffered saline was applied on the corneas every 3 minutes for 30 minutes, followed by ultraviolet-A irradiation of 9 mW/cm2 for 10 minutes at 365 nm.
Biomechanical characterization was performed on entire corneoscleral buttons using a two-dimensional extensometer, similarly to our previous studies.3,9,24 To control intraocular pressure, the load was applied three-dimensionally on the posterior corneal surface using a 10-mm diameter support and indenter. Tensile strength measurements were performed with the same commercial stress-strain extensometer (Z0.5; Zwick GmbH & Co.) used previously. In all corneas, two conditioning cycles with forces between 0.03 and 9 N were performed. Then, the stress-relaxation test was performed using 9 N for 120 seconds. The strain was recorded as a function of stress and converted into tensile strain in line with the geometric context (detailed equations are presented in previous studies).9,24 The elastic modulus was determined between 0.5% and 2.0% of strain from the second conditioning cycle. Homogeneous material properties and no difference between central and peripheral corneal thickness were assumed.
Statistical analysis was performed with SPSS Statistics software (version 23.0.0; IBM Corporation, Armonk, NY). Normality was tested with both Kolmogorov–Smirnov and Shapiro–Wilk tests. Subsequently, a two-tailed paired t test was applied for comparisons between the two groups. Confidence intervals of 95% (95% CI) were applied.
The average thickness across both groups was 730 µm. Corneal thickness was not significantly different between the SMILE-CXL and PRK-CXL groups (P > .05).
There was a normal distribution in both groups according to both Kolmogorov–Smirnov (P = .134 for both groups) and Shapiro–Wilk (P = .150 for PRK-CXL and P = .084 for SMILE-CXL) tests. The elastic modulus as a function of strain (strain ranging from 0.5% to 2.0%) shows a partial overlap of curves (Figure 1). The PRK-CXL group had a mean of 17.2 ± 5.3 MPa (95% CI: 14.1 to 20.5) and the SMILE-CXL group had a mean of 14.1 ± 5.0 MPa (95% CI: 11.1 to 17.2). There was no significant difference in the effective elastic modulus between the two groups (P = .093) (Figure 2).
Median of stress as a function of strain in both groups. Upper curve corresponds to the photorefractive keratectomy with corneal cross-linking (PRK-CXL) group and lower curve represents the small incision lenticule extraction with CXL (SMILE-CXL) group.
Columnar graph of mean and standard deviations of the elastic modulus between the photorefractive keratectomy with corneal cross-linking (PRK-CXL) (17.2 ± 5.3 MPa) and and small incision lenticule extraction with CXL (SMILE-CXL) (14.1 ± 5.0 MPa) groups, which was statistically equivalent (P = .093).
Corneal biomechanics plays a central role in the results and stability after refractive surgery.25 The purpose of this study was to evaluate the increase in corneal stiffness following CXL in the rare event of corneal ectasia occurring after PRK or SMILE. There was no statistically significant difference in the elastic modulus between groups, indicating a similar CXL stiffening effect regardless of whether a surface ablation (PRK) or a cap-based procedure (SMILE) was performed.
Spiru et al.9,26 demonstrated under ex vivo conditions that flap-based procedures weaken the cornea more than cap-based procedures and that corneas undergoing SMILE are approximately 47% stronger than those undergoing the same refractive correction with LASIK. On the other hand, until recently there was little evidence on whether SMILE and PRK would produce a similar biomechanical impairment.27 Recent data from our group indicated that both methods may result in similar ex vivo biomechanical stability in human corneas.3 The elastic modulus was 11.9 ± 4.9 and 9.6 ± 4.2 MPa in fellow-eye human corneas treated with PRK and SMILE, respectively.3
In the current study, we induced mechanical stress to mimic postoperative ectasia, and subsequently treated these corneas with CXL. Corneas from both groups received the same refractive correction. Because the initial conditions applied in the current investigation were identical to those of the aforementioned study,3 the CXL stiffening effect could be inferred more accurately. Under the same conditions, elastic modulus increased to 17.2 ± 5.3 MPa after PRK-CXL and 14.1 ± 5.0 MPa after SMILE-CXL, which represents a CXL stiffening effect of 44.5% and 46.8% in corneas with previous surface ablation or lenticule extraction, respectively. Thus, under similar circumstances, a 130-µm cap creation did not reduce the effect of CXL.
Corneal stiffening after CXL has been studied extensively. Epithelium-off standard protocol with irradiance of 3 mW/cm2 for 30 minutes (total fluence of 5.4 J/cm2) has been shown to be a successful treatment for patients with ectasia. Recently, several other protocols have been developed to accelerate the procedure. However, due to diffusion capacity and oxygen consumption, the biomechanical effect of CXL decreases significantly when using higher irradiances with shorter irradiation times, and this is particularly evident in laboratory settings.28,29 Clinically, CXL outcomes are more challenging to evaluate. Although the amount of CXL required to individually stabilize an ectasia is not known, protocols using 9 mW/cm2 for 10 minutes (total fluence 5.4 J/cm2) seem to represent a good compromise between biomechanical effectiveness and reduction of treatment time.23,30 Our group showed in porcine virgin corneas that accelerated CXL using 9 mW/cm2 for 10 minutes provided a stiffening effect of 33%.29 Although slightly larger, the estimated increase of 44.5% or 46.8% quoted above is in line with previous data.29 The difference between the porcine and human corneal model may be explained by the condensed anterior stroma and the presence of Bowman's layer in human corneas, which is virtually absent in porcine corneas. Moreover, such differences might also be due to postmortem stromal hydration status and morphology between porcine and human corneas.
Although surface ablation and cap-based procedures may be biomechanically similar3 and respond similarly to CXL, it is worth stressing that proper preoperative analysis remains critical in every refractive surgery candidate. Postoperative ectasia has a direct relation to the preoperative2 and postoperative25 biomechanical properties of the cornea. In addition, each procedure currently covers a different range of refractive errors. Accordingly, identification of preoperative risk factors and the choice of the surgical technique should continue to be evaluated on an individual basis, despite the biomechanical similarity revealed herein.
The study has some limitations. The ectasia model used after refractive surgery may not fully mimic the in vivo situation. Biological responses such as abnormal regulation of matrix metalloproteinases-1 may have an additional effect on the organization of collagen.31 In vivo evaluations are desirable and potentially provide additional information, but systems recently developed for this purpose (eg, air-puff tonometers or Brillouin microscopy) have shown inconclusive results or are not yet well established in medical practice. Also, the prestretched corneas are biomechanically different from those with keratoconus or form fruste keratoconus. Nevertheless, postoperative ectasia is considered a weakening of corneal integrity coupled with excessive tissue removal.25 In our model, following extensive tissue removal, stress was induced to simulate the onset of a cyclic cascade of biomechanical de-compensation. Although stress-strain extensometry is the gold standard for the evaluation of biomechanical properties, the method is destructive and can only be performed in ex vivo tissue. Therefore, we were not able to estimate the real weakening effect of the preconditioning cycles that the corneas received prior to CXL. Nonetheless, published results indicate that the biomechanical weakness induced after refractive surgery appears to be lower than the stiffening effect after CXL.24 This might be clearly observed with the stiffening effect after CXL in both groups and also clinically supports the therapeutic effect of CXL in postoperative ectasia.16 Given that donor age and time after donor death are variable, one might argue that a further limitation of the study would be on corneal hydration status, which certainly plays a central role in biomechanical analysis. Notwithstanding, this bias factor could be eliminated once paired corneas from the same donors were used.
Both experimental groups achieved similar biomechanical stability when measured experimentally on ex vivo human fellow corneas. Our data suggest that, in the event of postoperative ectasia, the biomechanical improvement achieved by CXL may be similar regardless of whether the primary surgery was PRK or SMILE.
- Kohlhaas M. [Iatrogenic keratectasia: a review]. Klin Monbl Augenheilkd. 2015;232(6):765–772.25853948
- Ambrósio R Jr, Lopes BT, Faria-Correia F, et al. Integration of Scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection. J Refract Surg. 2017;33(7):434–443. doi:10.3928/1081597X-20170426-02 [CrossRef]28681902
- Spiru B, Torres-Netto EA, Kling S, Lazaridis A, Hafezi F, Sekundo W. Biomechanical properties of human cornea tested by two-dimensional extensiometry ex vivo in fellow eyes: PRK versus SMILE. J Refract Surg. 2019;35(8):501–505. doi:10.3928/1081597X-20190730-02 [CrossRef]31393988
- Sekundo W, Kunert KS, Blum M. Small incision corneal refractive surgery using the small incision lenticule extraction (SMILE) procedure for the correction of myopia and myopic astigmatism: results of a 6 month prospective study. Br J Ophthalmol. 2011;95(3):335–339. doi:10.1136/bjo.2009.174284 [CrossRef]
- Ganesh S, Brar S, Pawar A. Matched population comparison of visual outcomes and patient satisfaction between 3 modalities for the correction of low to moderate myopic astigmatism. Clin Ophthalmol. 2017;11:1253–1263. doi:10.2147/OPTH.S127101 [CrossRef]28740361
- Sánchez-González JM, Alonso-Aliste F. Visual and refractive outcomes of 100 small incision lenticule extractions (SMILE) in moderate and high myopia: a 24-month follow-up study. Graefes Arch Clin Exp Ophthalmol. 2019;257(7):1561–1567. doi:10.1007/s00417-019-04349-4 [CrossRef]31119423
- Reinstein DZ, Archer TJ, Randleman JB. Mathematical model to compare the relative tensile strength of the cornea after PRK, LASIK, and small incision lenticule extraction. J Refract Surg. 2013;29(7):454–460. doi:10.3928/1081597X-20130617-03 [CrossRef]23820227
- Lee H, Roberts CJ, Kim TI, Ambrósio R Jr, Elsheikh A, Yong Kang DS. Changes in biomechanically corrected intraocular pressure and dynamic corneal response parameters before and after transepithelial photorefractive keratectomy and femtosecond laser-assisted laser in situ keratomileusis. J Cataract Refract Surg. 2017;43(12):1495–1503. doi:10.1016/j.jcrs.2017.08.019 [CrossRef]
- Spiru B, Kling S, Hafezi F, Sekundo W. Biomechanical differences between femtosecond lenticule extraction (FLEx) and small incision lenticule extraction (SmILE) tested by 2D-extensometry in ex vivo porcine eyes. Invest Ophthalmol Vis Sci. 2017;58(5):2591–2595. doi:10.1167/iovs.16-20211 [CrossRef]28494492
- Reznik J, Salz JJ, Klimava A. Development of unilateral corneal ectasia after PRK with ipsilateral preoperative forme fruste keratoconus. J Refract Surg. 2008;24(8):843–847. doi:10.3928/1081597X-20081001-14 [CrossRef]18856241
- Randleman JB, Caster AI, Banning CS, Stulting RD. Corneal ectasia after photorefractive keratectomy. J Cataract Refract Surg. 2006;32(8):1395–1398. doi:10.1016/j.jcrs.2006.02.078 [CrossRef]16863983
- Shetty R, Kumar NR, Khamar P, et al. Bilaterally asymmetric corneal ectasia following SMILE with asymmetrically reduced stromal molecular markers. J Refract Surg. 2019;35(1):6–14. doi:10.3928/1081597X-20181128-01 [CrossRef]30633782
- Sinha Roy A, Shetty R. Ectasia after SMILE: correct interpretation of biomechanical hypothesis. J Refract Surg. 2017;33(1):66. doi:10.3928/1081597X-20161018-03 [CrossRef]28068452
- Mattila JS, Holopainen JM. Bilateral ectasia after femtosecond laser-assisted small incision lenticule extraction (SMILE). J Refract Surg. 2016;32(7):497–500. doi:10.3928/1081597X-20160502-03 [CrossRef]27400083
- El-Naggar MT. Bilateral ectasia after femtosecond laser-assisted small-incision lenticule extraction. J Cataract Refract Surg. 2015;41(4):884–888. doi:10.1016/j.jcrs.2015.02.008 [CrossRef]25747164
- Hafezi F, Kanellopoulos J, Wiltfang R, Seiler T. Corneal collagen crosslinking with riboflavin and ultraviolet A to treat induced keratectasia after laser in situ keratomileusis. J Cataract Refract Surg. 2007;33(12):2035–2040. doi:10.1016/j.jcrs.2007.07.028 [CrossRef]18053900
- Kling S, Hammer A, Netto EAT, Hafezi F. Differential gene transcription of extracellular matrix components in response to in vivo corneal crosslinking (CXL) in rabbit corneas. Transl Vis Sci Technol. 2017;6(6):8. doi:10.1167/tvst.6.6.8 [CrossRef]29242757
- Hayes S, Kamma-Lorger CS, Boote C, et al. The effect of riboflavin/UVA collagen cross-linking therapy on the structure and hydrodynamic behaviour of the ungulate and rabbit corneal stroma. PLoS One. 2013;8(1):e52860. doi:10.1371/journal.pone.0052860 [CrossRef]23349690
- Richoz O, Mavrakanas N, Pajic B, Hafezi F. Corneal collagen cross-linking for ectasia after LASIK and photorefractive keratectomy: long-term results. Ophthalmology. 2013;120(7):1354–1359. doi:10.1016/j.ophtha.2012.12.027 [CrossRef]23582990
- Morishige N, Wahlert AJ, Kenney MC, et al. Second-harmonic imaging microscopy of normal human and keratoconus cornea. Invest Ophthalmol Vis Sci. 2007;48(3):1087–1094. doi:10.1167/iovs.06-1177 [CrossRef]17325150
- Cheng S, Clarke EC, Bilston LE. The effects of preconditioning strain on measured tissue properties. J Biomech. 2009;42(9):1360–1362. doi:10.1016/j.jbiomech.2009.03.023 [CrossRef]19394022
- Hatami-Marbini H, Rahimi A. Effects of bathing solution on tensile properties of the cornea. Exp Eye Res. 2014;120:103–108. doi:10.1016/j.exer.2013.11.017 [CrossRef]
- Lang PZ, Hafezi NL, Khandelwal SS, Torres-Netto EA, Hafezi F, Randleman JB. Comparative functional outcomes after corneal crosslinking using standard, accelerated, and accelerated with higher total fluence protocols. Cornea. 2019;38(4):433–441. doi:10.1097/ICO.0000000000001878 [CrossRef]30681515
- Kling S, Spiru B, Hafezi F, Sekundo W. Biomechanical weakening of different re-treatment options after small incision lenticule extraction (SMILE). J Refract Surg. 2017;33(3):193–198. doi:10.3928/1081597X-20161221-01 [CrossRef]28264134
- Roberts CJ, Dupps WJ Jr, . Biomechanics of corneal ectasia and biomechanical treatments. J Cataract Refract Surg. 2014;40(6):991–998. doi:10.1016/j.jcrs.2014.04.013 [CrossRef]24774009
- Spiru B, Kling S, Hafezi F, Sekundo W. Biomechanical properties of human cornea tested by two-dimensional extensiometry ex vivo in fellow eyes: femtosecond laser-assisted LASIK versus SMILE. J Refract Surg. 2018;34(6):419–423. doi:10.3928/1081597X-20180402-05 [CrossRef]29889296
- Yildirim Y, Ölçücü O, Basci A, et al. Comparison of changes in corneal biomechanical properties after photorefractive keratectomy and small incision lenticule extraction. Turk J Ophthalmol. 2016;46(2):47–51. doi:10.4274/tjo.49260 [CrossRef]27800259
- Hammer A, Richoz O, Arba Mosquera S, Tabibian D, Hoogewoud F, Hafezi F. Corneal biomechanical properties at different corneal cross-linking (CXL) irradiances. Invest Ophthalmol Vis Sci. 2014;55(5):2881–2884. doi:10.1167/iovs.13-13748 [CrossRef]24677109
- Torres-Netto EA, Kling S, Hafezi N, Vinciguerra P, Randleman JB, Hafezi F. Oxygen diffusion may limit the biomechanical effectiveness of iontophoresis-assisted transepithelial corneal cross-linking. J Refract Surg. 2018;34(11):768–774. doi:10.3928/1081597X-20180830-01 [CrossRef]30428097
- Hashemian H, Jabbarvand M, Khodaparast M, Ameli K. Evaluation of corneal changes after conventional versus accelerated corneal cross-linking: a randomized controlled trial. J Refract Surg. 2014;30(12):837–842. doi:10.3928/1081597X-20141117-02 [CrossRef]25437483
- Matthews FJ, Cook SD, Majid MA, Dick AD, Smith VA. Changes in the balance of the tissue inhibitor of matrix metalloproteinases (TIMPs)-1 and -3 may promote keratocyte apoptosis in keratoconus. Exp Eye Res. 2007;84(6):1125–1134. doi:10.1016/j.exer.2007.02.013 [CrossRef]17449031