Corresponding Author: In-Cheon You, MD, PhD. Department of Ophthalmology, Jeonbuk National University Medical School, 20 Geonji-ro, Deokjin-gu, Jeonju 54907, Korea. Small incision lenticule extraction (SMILE), first introduced in 2011 with the VisuMax 500 femtosecond laser (Carl Zeiss Meditec) is a flapless corneal refractive procedure that involves the creation and removal of an intrastromal lenticule through a small corneal incision [1-3]. SMILE has been extensively validated for the treatment of myopia and myopic astigmatism, demonstrating excellent refractive stability and reduced incidence of dry eye and flap-related complications compared with laser in situ keratomileusis (LASIK) [4-6]. Its biomechanical advantages have contributed to its widespread adoption.
The VisuMax 800 (Carl Zeiss Meditec), which received South Korean regulatory approval for hyperopia correction in 2023, incorporates faster laser pulse frequencies, reduced suction time, and sophisticated platforms—CentraLign and OcuLign—to assist with vertex alignment and cyclotorsion compensation [7]. These features theoretically improve the precision of hyperopic refractive correction.
Hyperopic SMILE, however, presents unique challenges. Unlike myopic treatments, hyperopic correction requires peripheral stromal steepening and a larger transition zone, increasing susceptibility to epithelial remodeling, regression, and centration-dependent optical changes. However, as ablation profile design evolved and larger optical and transition zones were adopted, refractive stability improved through a reduction in epithelial remodeling associated with a more consistent stromal curvature gradient [8-12]. Furthermore, decentration and cyclotorsion are more critical in hyperopic treatments because peripherally induced optical errors contribute more significantly to postoperative astigmatism and aberrations.
To the best of the authors’ knowledge, several studies have reported the clinical outcomes of hyperopic SMILE performed using the VisuMax 500 platform [13-17]. However, published studies have not evaluated the clinical outcomes of hyperopic SMILE using the VisuMax 800 system. Therefore, the aim of this pilot study was to evaluate early refractive, visual, vector astigmatic, and centration outcomes following hyperopic SMILE performed with the VisuMax 800 in a South Korean clinical setting. We hypothesized that early outcomes would be comparable to those previously reported with the VisuMax 500 and other refractive surgeries such as LASIK and that centration accuracy would be improved due to the enhanced guidance systems of the VisuMax 800.
Materials and Methods
Ethics statement
This retrospective study adhered to the tenets of the Declaration of Helsinki and received approval from the Public Institutional Review Board of the South Korean Ministry of Health and Welfare (No. P01-202509-01-033). Written informed consent for surgery was obtained from all patients.
Study design and patient selection
Medical records were reviewed for consecutive patients who underwent hyperopic SMILE using the VisuMax 800 at Onnuri Eye Hospital between December 1, 2024, and October 31, 2025. Only eyes that completed a minimum of three months of postoperative follow-up were included. The prevalence of hyperopia in South Korea is lower than that of myopia, leading to limited domestic experience and case volume [18].
Patients were eligible if they were at least 18 years of age and had manifest hyperopia between +1.00 and +7.00 diopters (D) with astigmatism of 4.00 D or less. Additional inclusion criteria required a corrected distance visual acuity (CDVA) of 20/40 or better, a central corneal thickness of at least 500 μm, and a predicted residual stromal bed thickness of at least 250 μm. The preoperative CDVA threshold of 20/40 was selected to reflect real-world hyperopic patients, who often present with reduced best-corrected visual acuity due to amblyopia, latent hyperopia, or long-standing refractive blur. Excluding such eyes would limit the external validity of this pilot study and reduce its relevance to routine clinical practice. Given the exploratory nature and limited case volume of hyperopic SMILE using the VisuMax 800, a pragmatic inclusion criterion was adopted to evaluate feasibility and safety across a representative clinical spectrum. Exclusion criteria consisted of prior corneal surgery, clinically significant corneal disease, keloid diathesis, glaucoma, and retinal pathology.
Preoperative examinations
All patients underwent a standardized preoperative evaluation on both eyes that included slit-lamp biomicroscopy, uncorrected distance visual acuity (UDVA) and CDVA measurement, manifest refraction and cycloplegic refractions, noncontact tonometry and pachymetry (NT-530P, Nidek Co.), axial length and white-to-white distance (IOLMaster 700, Carl Zeiss Meditec), endothelial cell density and central corneal thickness (UD-800, Tomey Corp.), corneal tomography (Galilei G4 dual rotating Scheimpflug analyzer, Ziemer Ophthalmic Systems), and scotopic pupil size measurement (VIP-300, NeurOptics Inc.). In addition, the treatment center is examined by IOLMaster 700 based on the corneal vertex. In-house optometrists examined both manifest refraction and cycloplegic refractions on separate days following a standardized protocol to ensure maximum plus and cylinder determination [19].
Perioperative care
During preparation, the operative area was cleaned and draped. Approximately 3 minutes before surgery, topical anesthesia was administered using 0.5% proparacaine hydrochloride eye drops (Alcaine, Alcon). After sterilization, the height of the patient bed was adjusted so that the corneal apex was gently applanated to the contact glass.
Surgical technique
All procedures were performed by a single experienced refractive surgeon (YTC) using the VisuMax 800. Treatment planning was based on manifest refraction with confirmation through cycloplegic refraction.
Laser parameters included a cap thickness of 110 or 120 μm, a cap diameter between 8.7 and 8.9 mm, and an optical zone diameter of either 6.0 or 6.2 mm. A 2.0-mm transition zone was applied in all cases. The minimum lenticule thickness was 15 μm. Spot and track spacing were set at 4 μm for the cap and lenticule interfaces, 15 μm for the side cut, and 2 μm for the small incision. The laser energy setting was +2 mJ, 2 MHz, 4.5 μm in VisuMax software.
The CentraLign displays the corneal vertex and the treatment center, with the value derived from preoperative IOLMaster 700 measurements. After the centration, while docking the cornea, the operator uses a joystick to align the OcuLign, displaying the horizontal axis of the eye, to control the cyclotorsion.
After confirming at least 70% contact, suction was applied to stabilize the eye. The posterior lenticule surface was created from the periphery toward the center, followed by the anterior surface from the center toward the periphery. Then, side cut and incision were created at 11 o’clock. Lenticule dissection was performed with a micro-spatula, and the lenticule was extracted using McPherson forceps (M. Blum design, Geuder). The stromal pocket was irrigated with balanced salt solution (Alcon) at the conclusion of the procedure.
Postoperative care
Postoperatively, patients were instructed to instill 1.5% levofloxacin eye drops (Cravit, Santen Pharmaceutical Co.) four times daily for 5 days, 0.1% fluorometholone eye drops (Opti-V, Reyon Pham.) four times daily for 5 weeks, and 0.1% sodium hyaluronate (Tearinfree, DHP Korea) as needed. Follow-up examinations were performed at 1 week, 1 month, and 3 months postoperatively.
Outcome measures and definitions
Visual and refractive outcomes were evaluated according to the nine standard graphs for refractive surgery [20]. Efficacy was assessed by comparing postoperative UDVA with preoperative CDVA. Safety was evaluated by changes in CDVA. Predictability was defined as the percentage of eyes within ±0.50 D and ±1.00 D of the intended postoperative spherical equivalent (SEQ). Stability was evaluated by changes in SEQ across postoperative visits. Line changes were defined as “unchanged” when the difference in CDVA was less than one line (−0.9 to +0.9 lines) and a line gain or loss defined as ±1.0 to 1.9 lines.
Vector astigmatism analysis
Astigmatic changes were evaluated using the Alpins vector method in accordance with the recommendations of the Journal of Refractive Surgery [20,21]. Target-induced astigmatism (TIA) represented the intended refractive correction. Surgically induced astigmatism (SIA) represented the achieved vector change from preoperative to postoperative refractive astigmatism. The difference vector quantified the magnitude and axis of residual astigmatism. The correction index (SIA/TIA) was used to determine undercorrection or overcorrection, while the index of success (difference vector/TIA) represented the accuracy of astigmatic correction. Angle of error (AE) defined as the angular deviation between the SIA and TIA vectors. Vector means were calculated by averaging J0 and J45 values. One eye with TIA = 0 was excluded from analysis.
Decentration and corneal higher-order aberrations analysis
Postoperative optical zone decentration and higher-order aberrations (HOAs) were assessed using Galilei G4 dual rotating Scheimpflug analyzer. Postoperative vertex-based decentration was measured using a difference map of the tangential curvature from the preoperative and 3 months follow-up examination postoperatively. The corneal HOAs were analyzed over a 6-mm central diameter, and the root mean square (RMS) values of HOAs including the spherical aberration and coma were calculated. Coma was analyzed as a signed Zernike term rather than as coma RMS.
Statistical analysis
Data entry and organization were performed in Microsoft Excel 2019 (Microsoft). Statistical analyses were conducted using R software ver. 4.5.2 (R Foundation for Statistical Computing). Owing to the small sample size, all variables were treated as nonparametric. Results are reported as medians with interquartile ranges (IQRs). The Wilcoxon signed-rank test was used for paired comparisons, and the Mann-Whitney U-test was used for comparisons between high and low hyperopia groups. Correlations were examined using Spearman rank correlation coefficient. Statistical significance was defined as p < 0.05. Vector and astigmatism analysis followed the recommendations of the Journal of Refractive Surgery [20].
Results
Eleven eyes from seven patients (three males and four females) who underwent hyperopic SMILE and completed at least 3 months follow-up were included. Baseline ocular characteristics are presented in Table 1. The median preoperative SEQ was +2.50 D (IQR, 1.94-4.12 D), and the median refractive cylinder was −0.75 D (IQR, −1.75 to −0.38 D). Cap thickness was 110 μm in three eyes and 120 μm in eight eyes. The optical zone diameter was 6.0 mm in nine eyes and 6.2 mm in two eyes. Mean lenticule thickness was 114 μm (IQR, 80-118 μm).
Visual outcomes and efficacy
At 3 months postoperatively, two eyes (18.2%) achieved a UDVA of 20/20 or better, and eleven eyes (100%) achieved UDVA of 20/32 or better. Postoperative UDVA was within one line of the preoperative CDVA in ten eyes (90.9%), and one eye gained two lines. The median efficacy index was 0.90 (IQR, 0.80-0.95). Correlation of preoperative CDVA with postoperative UDVA is illustrated in Fig. 1A and 1B.
Safety
At 3 months postoperatively, CDVA remained unchanged in seven eyes (63.6%) and decreased by one line in four eyes (36.4%). No eye lost two or more lines. The median safety index was 0.86 (IQR, 0.80-1.00). Line changes are shown in Fig. 1C.
Predictability and stability
At 3 months, all eyes were within ±1.00 D of intended SEQ, and eight eyes (72.7%) were within ±0.50 D. The median postoperative SEQ was +0.25 D (IQR, 0.00-0.50 D). Sequential SEQ measurements demonstrated mild early overcorrection with gradual regression tendency toward slight hyperopia: −0.13 D at 1 week, 0.00 D at 1 month, and +0.25 D at 3 months. Refractive accuracy and stability are shown in Table 2 and Fig. 1D-1F.
Astigmatism and vector analysis
Postoperative refractive cylinder was within 1.00 D in all eyes and within 0.50 D in nine eyes (81.8%) as shown in Fig. 1G. The arithmetic mean correction index was 1.02, indicating slight overcorrection. The median AE was 0.00° (IQR, −2.85° to 0.00°), and eight eyes (80.0%) had an AE within ±15°. Vector parameters are shown in Table 3, Fig. 1H, Fig. 1I, and Fig. 2A-2D.
Higher-order aberrations
Galilei-derived HOAs within a 6-mm zone, including RMS, spherical aberration, and coma, are summarized in Table 4. At 3 months postoperatively, RMS HOA increased by +0.38 μm (IQR, 0.09 to 0.52 μm). Spherical aberration decreased by −0.45 μm (IQR, −0.64 to −0.34 μm), and coma decreased by −0.12 μm (IQR, −0.35 to 0.05 μm).
Optical zone decentration
The median vertex-based decentration was 0.064 mm (IQR, 0.036 to 0.091 mm). A significant positive correlation was found between postoperative decentration and RMS HOA (rho = 0.709, p = 0.014). No significant correlation was observed between decentration and UDVA, CDVA, residual SEQ, spherical aberration, or coma. Full correlation data are available in Table 5.
Comparison of high and low hyperopia
A subgroup analysis compared outcomes between eyes with preoperative hyperopia of +4.00 D or more and those with lower degrees of hyperopia. Residual SEQ at 3 months did not significantly differ between the two groups (p = 0.849). Spherical aberration was significantly greater in the high-hyperopia group (p < 0.05), although RMS HOA, coma, and decentration values were not significantly different. Findings are summarized in Table 6.
Complications
No cases of suction loss, lenticule retention, aborted procedures, epithelial defects, epithelial ingrowth, or corneal erosions occurred. Interface haze was not analyzed because haze severity was not graded. Dry eye symptoms were reported in seven eyes and managed with topical eye drop therapy based on treatment guidelines proposed by Kim et al. [22]. No cases required surgical reintervention.
Discussion
To our knowledge, this is the first report of hyperopic SMILE performed using the VisuMax 800 platform. Early visual and refractive outcomes demonstrated favorable efficacy, safety, and predictability, while centration accuracy and AE were comparable to or better than those previously reported for the VisuMax 500 platform. However, because the number of eyes included in the present study was small, direct comparisons with other reports should be interpreted with caution.
The efficacy index of 0.90 in this cohort indicates that postoperative UDVA approached preoperative CDVA, although only 18.2% of eyes achieved 20/20 UDVA. This proportion is lower than the 68% reported in multicenter studies of hyperopic SMILE using the VisuMax 500 at 12 months [17]. However, the percentage of eyes achieving a UDVA of 20/32 or better was higher in this study than in other studies (58%-62%) [16]. This discrepancy may be attributed to a lower degree of hyperopia and a shorter follow-up duration in the present study. A previous study summarized modern hyperopic LASIK outcomes and reported a wide range of efficacy results, with UDVA ≥20/20 achieved in 6% to 90% of eyes [17]. This variability is likely attributable to the inherent challenges of the hyperopic correction, such as increased HOAs and postoperative regression. In addition, domestic studies of other hyperopic refractive procedures reported efficacy indices of 0.93 for photorefractive keratectomy (PRK), 0.88 for LASIK, and 0.44 for LASEK (laser epithelial keratomileusis) at 3 months postoperatively; however, at 12 months, the efficacy index increased to 1.02 for PRK and 0.94 for LASIK [23,24]. Compared with previous reports, including the domestic studies, the present study demonstrated comparable efficacy outcomes and is expected to show similar results at 12 months follow-up.
At 3 months postoperatively, CDVA remained within one line of the preoperative CDVA in all eyes, and no eye lost two or more lines. These results were comparable to those of previous studies as well as other refractive procedures, which reported a 0.0% to 1.2% incidence of eyes losing two or more lines of CDVA [16,17,23,25].
Predictability outcomes were encouraging. All eyes were within ±1.00 D of intended SEQ, and nearly three-quarters were within ±0.50 D at 3 months, which is comparable to or exceeds the early results reported with the VisuMax 500 platform, with 53% to 59% within ±0.5 D [15,16]. In studies of hyperopic LASIK, 43% to 95% of eyes achieved a postoperative SEQ within ±0.50 D at 12 months [26,27]. Consistent with the influence of preoperative SEQ, higher degrees of correction were associated with a lower proportion of eyes achieving a postoperative SEQ within ±0.50 D. In domestic studies with preoperative SEQ values similar to those in the present study, hyperopic SMILE performed using the VisuMax 800 platform resulted in a higher proportion of eyes within ±0.50 D.
Although longer follow-up is required to assess refractive regression, a mild early overcorrection observed at 1 week postoperatively, followed by gradual regression toward slight hyperopia, suggests that early regression or epithelial remodeling may have begun during this period. Previous studies have described similar patterns in hyperopic corneal procedures, in which peripheral epithelial thickening and central epithelial thinning contribute to early refractive changes [16,17,28-30]. However, based on previous reports [16,17], SMILE appears to be less prone to postoperative SEQ changes after 3 months compared with other refractive procedures, which have demonstrated refractive changes exceeding 0.1 D change per month [23,24]. In the present study, a hyperopic shift was observed between postoperative 1 week and 3 months. Further follow-up is warranted to clarify long-term refractive stability. Nevertheless, the stability of CDVA and the absence of significant safety concerns support the favorable short-term visual outcomes of this procedure.
Astigmatic correction was generally accurate, as indicated by a correction index near unity and a small AE, consistent with previous reports on hyperopic SMILE and hyperopic LASIK [17]. Hyperopic astigmatic corrections are particularly sensitive not only to axis misalignment but also to magnitude errors. Because lenticule geometry requires greater peripheral steepening, even small deviations in magnitude or axis can disproportionately increase residual astigmatism and HOAs [31].
Posture-related ocular cyclotorsion is a known major cause of intraoperative cyclotorsion, which can induce misalignment and leave residual astigmatism if not carefully managed [32-36]. However, with the aid of the OcuLign system in VisuMax 800 platform, cyclotorsion can be easily detected and adjusted. The low AE observed in this cohort indicates effective axis alignment, likely supported by the OcuLign cyclotorsion compensation system [31].
In this study, the results of the AE in two eyes were outlier compared with the other eyes, possibly because a total of three eyes, including these two, had a TIA of 0.50 D or less. AE tends to be unstable when the magnitude of TIA is small [21].
HOA profiles reflected the expected characteristics of hyperopic SMILE. Although RMS HOA increased at 3 months postoperatively, spherical aberration and coma decreased, suggesting early postoperative epithelial remodeling and controlled peripheral steepening. Previous studies have demonstrated that HOA values may continue to evolve beyond 12 months [16,17], and the present short-term results may therefore represent transitional rather than stabilized optical changes. Unlike previous studies of hyperopic SMILE reported an increase in coma after surgery [17], signed coma values in the present study decreased toward zero, suggesting improved centration rather than worsening asymmetric aberration. This finding may be attributable to improved centration achieved with the VisuMax 800 platform. Notably, the two eyes with a 6.2 mm optical zone diameter demonstrated decreased RMS HOA postoperatively, while eyes with 6.0 mm diameter showed an increase. Given the small number of cases, further studies with larger sample sizes are required to clarify the influence of optical zone diameter on postoperative HOAs changes.
Optical zone centration is particularly critical in hyperopic refractive surgery because peripheral stromal steepening renders postoperative optical quality more sensitive to centration errors. In the present study, the median vertex-based decentration was lower than that reported in previous studies of hyperopic and myopic SMILE procedures using the VisuMax 500 platform [13,37,38]. This finding may reflect cumulative improvements in surgical planning and execution, including greater attention to centration in hyperopic treatments and the routine use of vertex-referenced topographic measurements. Although the contribution of specific centration-assist features cannot be isolated in this study, the observed reduction in decentration provides a plausible explanation for the favorable HOA profile identified at 3 months, consistent with the known relationship between centration accuracy and optical aberrations.
The significant association between decentration and RMS HOA, but not with spherical aberration or coma, suggests that decentration primarily increases overall optical irregularity rather than inducing a consistent directional aberration. As a global metric, RMS HOA may therefore be more sensitive to subtle centration errors in hyperopic SMILE. This finding differs from earlier SMILE analysis, in which decentration was more strongly associated with coma and spherical aberration [13,31,39]. This discrepancy may be attributable to the small sample size or the limited variability in centration values in the present study.
Subgroup analysis revealed that eyes with higher hyperopia developed significantly greater spherical aberration than those with lower hyperopia, despite similar residual SEQ values. In contrast, previous studies have reported that greater degrees of hyperopia are more prone to undercorrection and regression [17]. Similar trends have been observed in hyperopic LASIK, in which larger corrections are associated with greater epithelial compensation, increased regression, and residual hyperopia [28-31]. The absence of significant differences in residual SEQ in the present study may be attributable to the short follow-up duration or differences in postoperative epithelial dynamics. The greater spherical aberration observed in higher-hyperopia eyes is consistent with the intrinsic optical characteristics of hyperopic corneal refractive correction in which larger peripheral transition zones induce steeper curvature gradients and increased aberrations. Such geometric changes inherently increase positive spherical aberration, particularly in eyes requiring higher degrees of correction. Importantly, this increase should not be interpreted as a sign of surgical decentration or procedural inaccuracy. Rather, the observed aberration profile reflects the optical trade-offs of hyperopic correction and has been similarly reported in hyperopic LASIK and SMILE literature.
No major perioperative complications occurred, and the absence of suction loss contrasts with isolated cases reported for the VisuMax 500 platform. The shorter suction duration of the VisuMax 800 (median, 21.0 seconds) likely enhances patient cooperation and reduces the risk of suction breaks [38]. Postoperative complications related to visual quality, including interface or corneal haze and dry eye symptoms, were not quantitatively analyzed because standardized grading systems and structured data collection were not employed. Although no clinically significant haze, corneal ectasia, infection, or other vision-threatening complications were observed during routine postoperative examinations, the absence of formal grading precluded retrospective quantitative assessment, particularly in eyes with reduced postoperative CDVA. This limitation should be considered when interpreting the safety outcomes of the present study.
This study has several limitations. Its retrospective, single-center design and small sample size limit statistical power and generalizability. In addition, the short follow-up duration precludes assessment of long-term refractive stability and optical outcomes, particularly in eyes with higher degrees of hyperopia. Although this study primarily focused on objective refractive and optical outcomes, subjective visual quality is an important aspect of hyperopic refractive surgery. Symptoms such as halos, glare, and night vision disturbances have been reported after hyperopic correction due to peripheral optical changes and induced HOAs. These subjective outcome measures, however, were not systematically collected in this pilot study. Nevertheless, no clinically significant visual complaints were spontaneously reported during routine postoperative follow-up.
Future studies should incorporate validated questionnaires for subjective visual quality, contrast sensitivity testing, standardized haze grading, and anterior-segment optical coherence tomography imaging to better characterize epithelial remodeling and stromal interface changes. Finally, hyperopic correction is known to carry a higher risk of regression and lower predictability compared with myopic correction [40]. Therefore, longer-term follow-up with larger cohorts will be necessary to fully establish the long-term safety, stability, and visual quality outcomes of hyperopic SMILE using the VisuMax 800 platform.
Despite these limitations, the present study provides important preliminary evidence that hyperopic SMILE performed with the VisuMax 800 platform is a feasible and effective treatment for hyperopia in South Korean patients. The OcuLign and CentraLign of the VisuMax 800, which may be particularly beneficial for surgeons during the learning curve for accurate optical zone positioning, may help reduce residual astigmatism and minimize decentration and associated optical aberrations. Larger, prospective, multicenter studies with longer follow-up are needed to determine the long-term stability of refractive outcomes and to optimize treatment strategies for higher degrees of hyperopia.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Print
