Small incision lenticule extraction (SMILE) is now a well-established surgical technique for correcting myopia,1 so there is great interest in extending this for the treatment of hyperopia. To date, there have been two studies2,3 investigating the feasibility of using the refractive lenticule extraction technique for hyperopia correction. In these studies, the femtosecond lenticule extraction (FLEx) procedure was used rather than SMILE to ensure the lenticule extraction could be easily achieved and assessed.
In the first study,2 using a 200-kHz VisuMax femtosecond laser (Carl Zeiss Meditec AG, Jena, Germany), the lenticule profile was designed with a 6-mm optical zone and a pseudotransition zone created by an oblique lenticule side-cut angle providing a total lenticule diameter between 6.2 and 7.6 mm. Although it was shown that hyperopic treatment was feasible, there was a loss of corrected distance visual acuity (CDVA) in some eyes and the rate of refractive regression was unacceptable in comparison to modern excimer laser correction.4–7 This was understood to be most likely due to the lenticule shape, small optical zone, and centration of the lenticule on the pupil.
Following this initial study, we redesigned the geometry of the lenticule (Figure A, available in the online version of this article), drawing on the refractively stable outcomes with the MEL 80 (Carl Zeiss Meditec) ablation profiles using a large 7-mm optical zone and 2-mm transition zone, with centration on the coaxially sighted corneal light reflex.7,8 First results of 9 eyes with spherical hyperopia treated by FLEx using this updated lenticule profile were recently published, showing improved safety and refractive stability.3
Geometry of lenticule parameters for the hyperopic small incision lenticule extraction treatment.
In parallel, we initiated another study at the Tilganga Institute of Ophthalmology in Nepal to investigate SMILE for hyperopia. This was a multi-phase study, with the initial phases designed as a feasibility study in poorly sighted eyes to optimize energy settings and demonstrate topographic safety before proceeding to investigate visual and refractive outcomes in sighted eyes. The first safety factor to be evaluated was optical zone centration, given the well-known detrimental effect of decentered treatments.9 The centration technique for hyperopic SMILE is the same as for myopic SMILE, for which centration has been shown to be similar to eye-tracker–based myopic LASIK.10–12
The aim of the current study was to compare optical zone centration between hyperopic eyes treated by SMILE and appropriately matched eyes treated by LASIK, using both axial and tangential topography difference maps.
Patients and Methods
This was a prospective case series of consecutive hyperopic SMILE procedures by three experienced surgeons (DZR, KRP, GIC) using the 500-kHz VisuMax femtosecond laser. Inclusion criteria were attempted correction in the maximum hyperopic meridian between +1.00 and +7.00 D, astigmatism up to 6.00 diopters (D), medically suitable for SMILE, normal corneal topography, and no corneal disease. The study consisted of four phases: phase I consisted of at least 4 blind eyes (CDVA of 20/200 or worse), phase II consisted of at least 6 densely amblyopic eyes (CDVA between 20/100 and 20/200), phase III consisted of at least 10 mildly amblyopic eyes (CDVA of between 20/40 and 20/80), and phase IV consisted of up to 200 sighted eyes (CDVA of 20/40 or better). The transition to each subsequent phase was made once safety had been demonstrated for the previous group. All patients treated between May 2014 and September 2015 were included for the topographic centration analysis, although recruitment for phase IV is continuing for visual and refractive outcomes analysis. The data from the four phases were combined into a single population given the small number of eyes in the early phases. Ethics approval was obtained from the Nepal Health Research Council.
A complete ophthalmologic examination was performed prior to surgery by an in-house optometrist, as has been described previously.1 The preoperative examination also included corneal Placido topography by ATLAS (Carl Zeiss Meditec), corneal tomography by Pentacam (Oculus Optikgeräte GmbH, Wetzlar, Germany), corneal tomography and epithelial thickness mapping by RTVue optical coherence tomography (Optovue, Fremont, CA), pupillometry (VIPTM-200; Neuroptics, Irvine, CA), and handheld ultrasound pachymetry (SP-3000; Tomey Corporation, Nagoya, Japan).
The following criteria were applied for planning the procedure. First, the predicted postoperative residual stromal thickness under the lenticule must be greater than 250 µm. Second, the attempted correction was limited such that the predicted postoperative keratometry was less than 51.00 D, calculated using the manifest refraction at the corneal plane added to the preoperative keratometry at the axis of treatment by vector analysis.
All SMILE treatments were performed using the 500-kHz VisuMax femtosecond laser. Figure A (available in the online version of this article) shows the geometry of the incisions performed for hyperopic SMILE. In all eyes, two 2-mm incisions were created at 30° and 150° (superonasally and superotemporally). Intended cap thickness was 120 µm in all eyes. Cap diameter used was between 8.8 and 9 mm using a medium “contact glass” (treatment interface) in all eyes. Optical treatment zone diameter was between 6.3 and 6.7 mm, with a 2-mm transition zone. The minimum lenticule thickness was 30 µm in all eyes. The spot and track distance were 4.5 µm for the cap and lenticule interfaces and 2 µm for the lenticule side-cut and small incision. The energy level was set to 170 nJ in the VisuMax software.
The surgical technique was essentially identical to that used for myopic SMILE, as has been described previously.1,3,10,13 During the SMILE procedure, the patient is raised to the contact glass of the femtosecond laser. At the moment of contact between the individually calibrated curved contact glass and the cornea, a meniscus tear film appears, at which point the patient is able to see the fixation target clearly because the vergence of the fixation beam is adjusted according to the individual eye's refraction. At this point, the surgeon instructs the patient to look directly at the green light and, once in position, the corneal suction ports are activated to fixate the eye in this position. In this way, the patient aligns the visual axis and hence the corneal vertex14 to the vertex of the contact glass, which is centered to the laser system and the center of the lenticule to be created. The centration is confirmed by the surgeon by comparing the relative positions of the corneal reflex and pupil center to the Placido eye image obtained by the ATLAS topography scan. However, if the surgeon is not satisfied with the centration of the docking, the suction is released and the docking procedure is repeated.
The femtosecond laser tissue separation sequence consisted of creating the lower lenticule interface using a “spiral in” pattern, followed by the transition zone using a “spiral out” pattern, then the upper lenticule interface in a “spiral out” pattern, and finally the small incision. Total suction time was approximately 35 seconds, independent of refractive error treated. The lenticule dissection was performed as described previously1,3,10,13 using the MMSU1297 Reinstein lenticule separator instrument (Malosa Medical, Halifax, United Kingdom). The lenticule was then hydrated with balanced salt solution on the corneal surface to distend it for visual inspection for completeness and edge smoothness.
On completion, the patient was brought to the operating room slit lamp, fluorescein was instilled, and gentle distention of the cap was achieved by centrifugal stroking using a dry micro-spear to ensure that any redundant cap was redistributed to the periphery.
Postoperative Course and Evaluation
Patients were instructed to instill ofloxacin 0.1% together with prednisolone acetate (1%) four times a day for 1 week, as a standard protocol for broad-spectrum prophylaxis, and wear plastic shields for sleeping during the first week. The surgeon reviewed the patient 1 day postoperatively and an optometrist examined the patient at 1, 3, and 12 months as described previously.1 The 3-month examination included ATLAS topography to generate a difference map to evaluate optical zone centration.
LASIK Control Group
To be able to draw a comparison between the centration of hyperopic SMILE and LASIK, two control groups were retrospectively generated using a randomized, computerized program from the patient database of hyperopic LASIK at London Vision Clinic, London, United Kingdom. The control group was generated from eyes that had been treated with the VisuMax femtosecond laser and MEL 80 or MEL 90 excimer lasers. The first control group consisted of eyes treated using a 7-mm optical zone, with the second control group including 6.5-mm optical zone treatments. Each eye was individually matched where possible for attempted sphere correction (±0.75 D), attempted cylinder correction (±0.75 D), and with 3-month ATLAS topography data available. Informed consent and permission to use any clinical data for analysis and publication were obtained preoperatively from each patient included as part of our standard and routine clinical protocol.
LASIK Surgical Protocol
The VisuMax laser was programmed for flap thicknesses between 90 and 120 µm. Flap diameter was programmed to be 8 mm using the small contact glass for the 6.5-mm optical zone group and 8.8 mm using the medium contact glass for the 7-mm optical zone group. Side-cut for all VisuMax flaps was set as an in-cut of 55° from the normal. A 4.5-mm superior hinge was used in all cases. Treatments were centered on the coaxially sighted corneal light reflex.8,15 During surgery, the coaxially sighted corneal light reflex was determined before the flap was lifted as the first Purkinje reflex using the Seiler method,16 seen as the patient fixated coaxially with the aiming beam and the view of the surgeon's contralateral eye through the operating microscope. The coaxially sighted corneal light reflex was used as the best approximation of the intersection of the visual axis with the cornea.14
Difference maps of both the tangential and axial curvature were generated for each eye using the preoperative and 3-month postoperative ATLAS topography scans with the scale set to Standard (ANSI Z80.23). The difference map images were imported into Microsoft PowerPoint 2010 (Microsoft Corporation, Redmond, WA), and a previously prepared grid and set of concentric circles was overlaid. The lines of the grid were distributed equally with 0.1-mm steps and the concentric circles had radii increasing in 0.2-mm steps, between 4- and 7-mm radii (Figure B, available in the online version of this article). The difference map was enlarged on the screen so that the center of the optical zone could be confidently visualized within half a step. The concentric circles were used to visually align and superimpose the best-fitting circle to the optical zone defined as the central zone up to the mid-peripheral power inflection point. The central grid was then used to determine the centration offset of the optical zone with reference to the corneal vertex, with the optical zone center represented by the (0,0) coordinate of the central grid and the corneal vertex represented by the center of the topography map. The x- and y-coordinates of the location of the optical zone center with reference to the corneal vertex was measured to the nearest 0.05 mm using the central grid. This process was performed for both axial and tangential difference maps.
A central grid and set of concentric circles was generated and overlaid on the tangential difference map to identify the location of the optical zone. The central grid consisted of 10 × 10 0.1-mm steps. The concentric circles were created with radii increasing in 0.2-mm steps from 4 to 7 mm. The concentric circles were used to visually align and superimpose the best-fitting circle to the optical zone defined as the central zone up to the mid-peripheral power inflection point. The centration offset of the optical zone to the corneal vertex was determined according to the central grid.
The x-coordinates obtained for left eyes were mirrored in the vertical axis so that nasal/temporal characteristics of right and left eyes could be combined. The location of the center of the optical zones, relative to the corneal vertex (plotted as the origin), were displayed by plotting the x- and y-coordinates on a 360° polar plot. The vectorial mean and standard deviation ellipse were calculated for each population using principal component analysis to find the orientation with the greatest standard deviation. The distribution of the centration offset was evaluated using a cumulative frequency histogram. The Student's t test was used to evaluate the difference in centration offset. Microsoft Excel 2010 (Microsoft Corporation) was used for data entry and statistical analysis. A P value less than .05 was defined as statistically significant.
A total of 69 eyes of 43 patients underwent hyperopic SMILE between April 2014 and September 2015, of which 3-month data were available in 60 eyes of 37 patients (87% follow-up). Table 1 presents demographic data including number of eyes, age, preoperative CDVA, white-to-white diameter, spherical equivalent treated, and cylinder treated for the SMILE group and the two LASIK control groups.
Figure 1 shows axial topography difference maps for 5 randomly selected cases for each group. Figures 2–4 show the 360° polar plot for the center of the optical zone for each group. Table 1 includes the data for the centration offset of the optical zone relative to the corneal vertex. Figure 5 shows a cumulative histogram for the centration offset magnitude for the three groups.
Examples of ATLAS topography (Carl Zeiss Medite AG, Jena, Germany) tangential difference maps for small incision lenticule extraction (SMILE), 6.5-mm LASIK, and 7-mm LASIK. The perimeter of the optical zone is identified by the red circle. The center of the optical zone is indicated by the white cross and the corneal vertex is indicated by the black cross.
Plot showing the optical zone center relative to the corneal vertex for the small incision lenticule extraction group. The red data point indicates the vector mean and the red oval indicates the standard deviation (SD) ellipse.
Plot showing the optical zone center relative to the corneal vertex for the 6.5-mm LASIK group. The red data point indicates the vector mean and the red oval indicates the standard deviation (SD) ellipse.
Plot showing the optical zone center relative to the corneal vertex for the 7-mm LASIK group. The red data point indicates the vector mean and the red oval indicates the standard deviation (SD) ellipse.
Cumulative histogram showing the distribution of centration offset magnitude of the optical zone center relative to the corneal vertex for the small incision lenticule extraction (SMILE), 6.5-mm LASIK, and 7-mm LASIK groups.
The centration offset magnitude was lower for the SMILE group compared to the 6.5-mm LASIK group (P < .001) and the 7-mm LASIK group (P = .007). In the SMILE group, the center of the optical zone was distributed evenly between the four quadrants, with a slight tendency toward inferior and temporal directions. In both LASIK groups, the center of the optical zone was located in the inferonasal quadrant in the majority of eyes.
In the current study, the achieved topographical centration offset of the optical zone was found to be similar between fixation-based hyperopic SMILE and eye-tracker–based LASIK matched for hyperopia treated. All eyes achieved centration within 1 mm of the corneal vertex, with a mean offset magnitude of 0.23 mm for the SMILE group, 0.31 mm for the 7-mm LASIK group, and 0.33 mm for the 6.5-mm LASIK group. The scatter of centration about the mean (± standard deviation) was similar between groups. There was a difference in the distribution with a clear skew in the inferonasal direction for the LASIK groups, compared with a relatively even spread in all directions for the SMILE group.
Due to the nature of the study design, the majority of the eyes in the SMILE group were amblyopic, with a CDVA of 20/50 or worse in 42% of eyes and 20/100 or worse in 13% of eyes. In comparison, the majority of eyes (> 92%) in the LASIK groups had a CDVA of 20/40 or better. Therefore, it could be considered that these results represent a worst-case scenario for the SMILE group given that patients with dense amblyopia may have had greater difficulty in fixating during docking with the contact glass and will also likely present with a larger angle kappa.
The noticeable difference between the two procedures was that the optical zone center was inferonasal for most LASIK eyes, whereas the optical zone was evenly distributed for SMILE eyes. This is explained by the difference in how the treatment is centered. In the MEL 80 excimer laser, centration was set by the surgeon onto the coaxially sighted corneal light reflex, whereas in SMILE, patients effectively auto-aligned to the contact glass onto their visual axis by coaxial fixation on the light. The even distribution of the optical zone center in SMILE suggests that this process is an effective method of centration. Given that it is not possible to identify the visual axis, the topographic analysis was performed using the corneal vertex as the reference point. Therefore, it is possible that some of the scatter in the SMILE group might be due to the difference between the visual axis and corneal vertex.
In contrast, the tendency for the optical zone to be shifted inferonasally in the LASIK groups is most likely due to the presence of an angle kappa, which is common in high hyperopia and usually inferonasal.17 Although the coaxially sighted corneal light reflex and corneal vertex are usually well aligned, this shift may be partially explained by a difference between these two points. However, another explanation is that there was a parallax error introduced by viewing through the operating microscope with the contralateral eye. We studied this internally and found that a 200-µm temporal offset was required following the initial alignment (personal communication, Dan Z. Reinstein). This is similar to the offset found in the current study. If this adjustment had been used for the population in the current study, it would have shifted mean distribution closer to the corneal vertex; however, it is unlikely that this would have resulted in a statistical improvement compared to SMILE given that the standard deviations were similar.
The VisuMax laser uses a spherical coupling contact-glass, akin to a gonioscopy lens, where size is selected according to treatment diameter and white-to-white diameter. Angle kappa results in a relative rotation of the geometric center of the aspheric cornea away from the coaxial fixation target that passes through the center of the spherical contact glass surface. Thus, angle kappa results in an asymmetry of the curvature on either side of the coaxially fixating corneal vertex. This means that the best-fit sphere of the rotated aspheric surface of the cornea can lead to small offsets in the ‘accurvation’ of the cornea.
In this study, the medium contact glass was used in all eyes and the white-to-white diameter was less than 11.8 mm in 17% (20/60) of eyes with a minimum of 10.8 mm in one case. For some eyes with smaller white-to-white diameter, contact glass eye fixation was achieved with a portion of the peripheral suction ports applied on the limbal conjunctiva; however, in this series there were no instances of suction loss or significant treatment decentration observed.
The LASIK and SMILE groups were matched for attempted refractive correction. They would ideally have also been matched for age, gender, and ethnicity, but this was not possible due to the previously treated hyperopic LASIK population available in London.
The centration for hyperopic SMILE with a mean offset of 0.24 mm was slightly greater than that reported for myopic SMILE, where the mean offset has been reported as 0.17,12 0.20,10 and 0.3211 mm. However, this could be expected due to the larger angle kappa present in high hyperopia and the amblyopic population.
We have demonstrated that the safety of hyperopic SMILE in terms of optical zone centration is similar to hyperopic LASIK with the MEL 80 laser. Future phases of this study will evaluate additional aspects of safety, including achieved optical zone diameter and induced aberrations, before investigating the visual and refractive outcomes of hyperopic SMILE for a non-amblyopic population.
- Pradhan KR, Reinstein DZ, Carp GI, Archer TJ, Gobbe M, Dhungana P. Quality control outcomes analysis of small-incision lenticule extraction for myopia by a novice surgeon at the first refractive surgery unit in Nepal during the first 2 years of operation. J Cataract Refract Surg. 2016;42:267–274. doi:10.1016/j.jcrs.2015.09.026 [CrossRef]
- Blum M, Kunert KS, Vossmerbaumer U, Sekundo W. Femtosecond lenticule extraction (ReLEx) for correction of hyperopia - first results. Graefes Arch Clin Exp Ophthalmol. 2013;251:349–355. doi:10.1007/s00417-012-2064-y [CrossRef]
- Sekundo W, Reinstein DZ, Blum M. Improved lenticule shape for hyperopic femtosecond lenticule extraction (ReLEx((R)) FLEx): a pilot study. Lasers Med Sci. 2016;31:659–664. doi:10.1007/s10103-016-1902-2 [CrossRef]
- de Ortueta D, Arba Mosquera S, Baatz H. Aberration-neutral ablation pattern in hyperopic LASIK with the ESIRIS laser platform. J Refract Surg. 2009;25:175–184. doi:10.3928/1081597X-20090401-03 [CrossRef]
- Kanellopoulos AJ. Topography-guided hyperopic and hyperopic astigmatism femtosecond laser-assisted LASIK: long-term experience with the 400 Hz eye-Q excimer platform. Clin Ophthalmol. 2012;6:895–901. doi:10.2147/OPTH.S23573 [CrossRef]
- Reinstein DZ, Archer TJ, Gobbe M, Silverman RH, Coleman DJ. Epithelial thickness after hyperopic LASIK: three-dimensional display with Artemis very high-frequency digital ultrasound. J Refract Surg. 2010;26:555–564. doi:10.3928/1081597X-20091105-02 [CrossRef]
- Reinstein DZ, Couch DG, Archer TJ. LASIK for hyperopic astigmatism and presbyopia using micro-monovision with the Carl Zeiss Meditec MEL80. J Refract Surg. 2009;25:37–58.
- Reinstein DZ, Gobbe M, Archer TJ. Coaxially sighted corneal light reflex versus entrance pupil center centration of moderate to high hyperopic corneal ablations in eyes with small and large angle kappa. J Refract Surg. 2013;29:518–525. doi:10.3928/1081597X-20130719-08 [CrossRef]
- Mrochen M, Kaemmerer M, Mierdel P, Seiler T. Increased higher-order optical aberrations after laser refractive surgery: a problem of subclinical decentration. J Cataract Refract Surg. 2001;27:362–369. doi:10.1016/S0886-3350(00)00806-3 [CrossRef]
- Reinstein DZ, Gobbe M, Gobbe L, Archer TJ, Carp GI. Optical zone centration accuracy using corneal fixation-based SMILE compared to eye tracker-based femtosecond laser-assisted LASIK for myopia. J Refract Surg. 2015;31:586–592. doi:10.3928/1081597X-20150820-03 [CrossRef]
- Lazaridis A, Droutsas K, Sekundo W. Topographic analysis of the centration of the treatment zone after SMILE for myopia and comparison to FS-LASIK: subjective versus objective alignment. J Refract Surg. 2014;30:680–686. doi:10.3928/1081597X-20140903-04 [CrossRef]
- Li M, Zhao J, Miao H, et al. Mild decentration measured by a Scheimpflug camera and its impact on visual quality following SMILE in the early learning curve. Invest Ophthalmol Vis Sci. 2014;55:3886–3892. doi:10.1167/iovs.13-13714 [CrossRef]
- Reinstein DZ, Archer TJ, Gobbe M. Small incision lenticule extraction (SMILE) history, fundamentals of a new refractive surgery technique and clinical outcomes. Eye Vis. 2014;1:3. doi:10.1186/s40662-014-0003-1 [CrossRef]
- Pande M, Hillman JS. Optical zone centration in keratorefractive surgery: entrance pupil center, visual axis, coaxially sighted corneal reflex, or geometric corneal center?Ophthalmology. 1993;100:1230–1237. doi:10.1016/S0161-6420(93)31500-9 [CrossRef]
- Chang DH, Waring GO 4th, . The subject-fixated coaxially sighted corneal light reflex: a clinical marker for centration of refractive treatments and devices. Am J Ophthalmol. 2014;158:863–874. doi:10.1016/j.ajo.2014.06.028 [CrossRef]
- Seiler T. Basics, centration. In: Seiler T. Refractice Surgery of the Cornea. Stuttgart: Enke; 2000.
- Basmak H, Sahin A, Yildirim N, Papakostas TD, Kanellopoulos AJ. Measurement of angle kappa with synoptophore and Orbscan II in a normal population. J Refract Surg. 2007;23:456–460.
|Parameter||SMILE||LASIK 6.5 mm||LASIK 7 mm|
|Planned optical zone (mm)||6.3 to 6.7||6.5||7|
|Eyes (patients)||60 (37)||60 (51)||60 (55)|
|SEQ treated (D)||+5.61 ± 0.96; +3.20 to +6.50||+4.83 ± 0.80; +3.13 to +6.30||+5.54 ± 0.95; +3.13 to +7.00|
|Cyl treated (D)||−0.96 ± 0.62 (0.00 to −2.75)||−0.93 ± 0.71 (0.00 to −2.75)||−1.00 ± 0.73 (0.00 to −2.75)|
|Age (y)||29 ± 7 (19 to 52)||52 ± 13 (21 to 72)||45 ± 14 (20 to 69)|
|White to white diameter (mm)||12.2 ± 0.55 (10.8 to 13.3)||11.9 ± 0.38 (11.0 to 12.6)||12.4 ± 0.44 (11.1 to 13.3)|
|CDVA 20/20 or better||9%||62%||65%|
|CDVA 20/25 or better||15%||82%||82%|
|CDVA 20/40 or better||58%||92%||93%|
|CDVA 20/100 or better||87%||95%||98%|
|Centration offset (mm)||0.23 ± 0.15 (0.00 to 0.61, 95% CI: 0.21 to 0.25)||0.33 ± 0.14 (0.14 to 0.85, 95% CI: 0.30 to 0.36 mm)||0.31 ± 0.19 (0.05 to 0.85, 95% CI: 0.29 to 0.33 mm)|