Journal of Refractive Surgery

Original Article Supplemental Data

Comparison of FDA-Reported Visual and Refractive Outcomes of the Toric ICL Lens, SMILE, and Topography-Guided LASIK for the Correction of Myopia and Myopic Astigmatism

Majid Moshirfar, MD, FACS; Anisha N. Somani, BS; Mahsaw N. Motlagh, BA; Uma Vaidyanathan, BS; Jorgen S. Sumsion, BS; James R. Barnes, BS; Yasmyne C. Ronquillo, MD, MSc

Abstract

PURPOSE:

To compare U.S. Food and Drug Administration (FDA)–reported visual and refractive outcomes following surgical correction of myopia and myopic astigmatism using the Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK).

METHODS:

FDA summary of safety and effectiveness data (SSED) were analyzed for each of the three platforms. Primary outcomes measured were efficacy, safety, stability, and accuracy of refractive correction. Stratified mean refractive spherical equivalent (MRSE) data were assessed.

RESULTS:

One hundred twenty-four Toric ICL patients (210 eyes), 357 SMILE patients (357 eyes), and 212 TG-LASIK patients (249 eyes) were included. SMILE eyes had a significant improvement in postoperative uncorrected distance visual acuity with respect to preoperative corrected distance visual acuity from 3 to 12 months (P < .001), whereas TG-LASIK had no further improvement from 3 to 12 months (P = .79). For preoperative MRSE greater than 10.00 diopters (D), there was a significant difference in the percentage of eyes achieving postoperative MSRE within ±0.50 D between Toric ICL (66%) and SMILE (100%) (P < .001). SMILE was consistently more accurate than Toric ICL for cylinder within ±0.25 D (P < .001), ±0.50 D (P < .001), and ±1.00 D (P = .0014).

CONCLUSIONS:

All three platforms analyzed in this study had excellent efficacy, safety, stability, and accuracy. Stratified analysis revealed that SMILE may be comparable to Toric ICL for patients with high myopia or myopic astigmatism, and SMILE may have a longer visual recovery compared to TG-LASIK than previously indicated.

[J Refract Surg. 2019;35(11):699–706.]

Abstract

PURPOSE:

To compare U.S. Food and Drug Administration (FDA)–reported visual and refractive outcomes following surgical correction of myopia and myopic astigmatism using the Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK).

METHODS:

FDA summary of safety and effectiveness data (SSED) were analyzed for each of the three platforms. Primary outcomes measured were efficacy, safety, stability, and accuracy of refractive correction. Stratified mean refractive spherical equivalent (MRSE) data were assessed.

RESULTS:

One hundred twenty-four Toric ICL patients (210 eyes), 357 SMILE patients (357 eyes), and 212 TG-LASIK patients (249 eyes) were included. SMILE eyes had a significant improvement in postoperative uncorrected distance visual acuity with respect to preoperative corrected distance visual acuity from 3 to 12 months (P < .001), whereas TG-LASIK had no further improvement from 3 to 12 months (P = .79). For preoperative MRSE greater than 10.00 diopters (D), there was a significant difference in the percentage of eyes achieving postoperative MSRE within ±0.50 D between Toric ICL (66%) and SMILE (100%) (P < .001). SMILE was consistently more accurate than Toric ICL for cylinder within ±0.25 D (P < .001), ±0.50 D (P < .001), and ±1.00 D (P = .0014).

CONCLUSIONS:

All three platforms analyzed in this study had excellent efficacy, safety, stability, and accuracy. Stratified analysis revealed that SMILE may be comparable to Toric ICL for patients with high myopia or myopic astigmatism, and SMILE may have a longer visual recovery compared to TG-LASIK than previously indicated.

[J Refract Surg. 2019;35(11):699–706.]

Refractive error is the most common cause of correctable visual impairment and is estimated to affect nearly 6 billion people by the year 2050.1 Approximately 1 in 6 people in the world are myopic,2 and the increasing prevalence of myopia has important implications for non-surgical and surgical management. Laser in situ keratomileusis (LASIK) remains the dominant surgical procedure to correct refractive error. With an estimated economic productivity loss per annum cost of $202 billion,1 the economic argument alone is reason to consider visual outcomes in patients having refractive surgery. Technological advances in surgical modalities, including the Visian Toric Implantable Collamer Lens (ICL) (STAAR Surgical, Monrovia, CA), small incision lenticule extraction (SMILE), and topography-guided LASIK, have high safety profiles as evaluated by the U.S. Food and Drug Administration (FDA). Each modality is unique in its indications, duration of use, and age criteria, as demonstrated in Table A (available in the online version of this article).

FDA Indications for Use and Range Used in the Study

Table A:

FDA Indications for Use and Range Used in the Study

When LASIK first emerged, patients with severe myopia and myopic astigmatism were considered high risk for postoperative iatrogenic corneal ectasia.3,4 The ICL then became known as a safe and effective treatment method for patients requiring large degrees of refractive correction.5 More recently, SMILE has developed as a procedure that avoids potential flap-related complications.6 It is essential to understand the differences in these modalities to optimize patient safety and visual outcomes. The objective of our study was to compare visual and refractive outcomes, obtained from FDA summary of safety and effectiveness data (SSED), following the correction of myopia and myopic astigmatism with these three platforms.

Patients and Methods

This study analyzed publicly available FDA SSED from three approved platforms for astigmatism correction: Visian Toric ICL (PMA P030016/S001),7 VisuMax femtosecond laser (Carl Zeiss Meditec, Inc., Dublin, CA) used in SMILE (PMA P150040/S003),8 and Wave-Light Allegro Topolyzer in conjunction with the Wave-light Allegretto Wave Eye-Q Laser System (Alcon Laboratories, Inc., Fort Worth, TX) used in TG-LASIK (PMA P020050/S12).9 These procedures will be referred to as Toric ICL, SMILE, and TG-LASIK, respectively, in this article.

This study analyzed primary, secondary, and stratified visual outcomes among treatment groups. There were six primary outcomes measured: (1) cumulative postoperative uncorrected distance visual acuity (UDVA) (efficacy); (2) postoperative UDVA compared to preoperative corrected distance visual acuity (CDVA) (efficacy); (3) change in lines of CDVA from preoperative to postoperative measurements (safety); (4) refractive error over time (stability); (5) postoperative mean refractive spherical equivalent (MRSE) (accuracy); and (6) correction ratio (accuracy). Each primary outcome represented the efficacy, safety, stability, and accuracy of the refractive correction platforms. Two secondary visual outcomes were included in this study: (1) contrast sensitivity and higher order aberrations (HOAs) and (2) subjective measures, such as glare and halos.

For continuous variables, summary statistics were analyzed using one-way analysis of variance F-tests as described by Cohen,10 Student's t tests or Welch's t tests as appropriate, and Tukey HSD post-hoc analysis. The tests were computed with R Statistical Software (Vienna, Austria) using packages Basic Statistics and Data Analysis and rpsychi.11 Two-tailed hypothesis tests were performed and evaluated with a P value of less than .05 indicating statistical significance. For categorical variables, summary statistics were analyzed using chi-square tests.12

Lack of raw data prevented the authors from running tests such as the Shapiro-Wilk's W-test or Kolmogorov–Smirnov test to verify the assumption of normality for the statistical analyses. Although these tests were not performed, the authors assumed normality of sampling distribution due to the large sample sizes.13–15 Summary statistics demonstrated no violations of homoscedasticity for the analysis of variance and Student's t tests.

The preoperative mean refractive spherical equivalent was stratified to compare the three platforms. Each stratified category was only analyzed if the sample size was 10 eyes or greater. Stratified cylinder data were not available for Toric ICL or SMILE and could not be analyzed.

Astigmatism correction was evaluated using the correction ratio (CR), which was calculated from the surgically induced refraction correction (SIRC) divided by the intended refraction correction (IRC): SIRC/IRC = CR.

Results

Patient Demographics

One hundred twenty-four Toric ICL patients (210 eyes), 357 SMILE patients (357 eyes), and 212 TG-LASIK patients (249 eyes) were included. The three platforms were significantly different in baseline refractive characteristics (Table 1).

Patient Demographics

Table 1:

Patient Demographics

Efficacy

UDVA was comparable across the three platforms (Figure 1). At 1 week, fewer eyes achieved 20/20 UDVA with SMILE (44.2%) compared to Toric ICL (76.6%) (P < .001); 1-week data were not reported for TG-LASIK. However, at 12 months, there was a similar percentage of eyes achieving 20/20 visual acuity. A similar pattern emerged at 6 months, All three platforms had nearly 100% of eyes with visual acuity of 20/40 or better at both 3 and 12 months.

Uncorrected distance visual acuity (UDVA) of Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK) patients at multiple time points. SMILE had a statistically significant lower percentage of patients achieving 20/16 visual acuity compared to both Toric ICL and TG-LASIK at 6 months (P < .001); however, there was no significant difference at 12 months.

Figure 1.

Uncorrected distance visual acuity (UDVA) of Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK) patients at multiple time points. SMILE had a statistically significant lower percentage of patients achieving 20/16 visual acuity compared to both Toric ICL and TG-LASIK at 6 months (P < .001); however, there was no significant difference at 12 months.

Postoperative UDVA versus preoperative CDVA data were available for SMILE and TG-LASIK only (Figure 2). At 3 months, a greater percentage of eyes reported postoperative UDVA worse than preoperative CDVA for SMILE versus TG-LASIK. This same pattern occurred at 12 months for SMILE versus TG-LASIK. For SMILE at 3 versus 12 months, there was a significant decrease in the percentage of eyes that reported postoperative UDVA worse than preoperative CDVA. For TG-LASIK, there was no significant change in the percentage of eyes with postoperative UDVA worse than preoperative CDVA at 3 versus 12 months.

Postoperative uncorrected distance visual acuity (UDVA) compared to preoperative corrected distance visual acuity (CDVA) at 3 and 12 months. Small incision lenticule extraction (SMILE) had a significantly higher percentage of eyes that gained lines at 12 months compared to 3 months (P < .001). Compared to topography-guided laser in situ keratomileusis (TG-LASIK), SMILE had a higher percentage of eyes that lost lines at both 3 months (P < .001) and 12 months (P = .009).

Figure 2.

Postoperative uncorrected distance visual acuity (UDVA) compared to preoperative corrected distance visual acuity (CDVA) at 3 and 12 months. Small incision lenticule extraction (SMILE) had a significantly higher percentage of eyes that gained lines at 12 months compared to 3 months (P < .001). Compared to topography-guided laser in situ keratomileusis (TG-LASIK), SMILE had a higher percentage of eyes that lost lines at both 3 months (P < .001) and 12 months (P = .009).

Safety

Toric ICL, SMILE, and TG-LASIK appear to be equally safe in preserving CDVA (Table 2). At 12 months, loss of CDVA of two or more lines occurred in 3 eyes (1.4%) with Toric ICL, 0 eyes with SMILE, and 1 eye (0.41%) with TG-LASIK. For Toric ICL, loss of CDVA of two or more lines resolved in one eye at the 17-month postoperative visit. CVDA loss of two or more lines was transient for TG-LASIK and resolved by the next postoperative visit. The rate of surgical reintervention was higher at 3.8% (8 eyes) for Toric ICL compared to 0.29% (1 eye) for SMILE (P < .001). TG-LASIK did not report surgical reintervention rates. At 12 months, fewer than 10% of eyes had loss of CDVA (Figure 3). There was no significant difference between the three platforms in the percentage of eyes that reported worse postoperative versus preoperative CDVA (P = .57).

Safety Parameters at 12 Months

Table 2:

Safety Parameters at 12 Months

Change in lines of corrected distance visual acuity (CDVA) (Snellen) at 12 months. Toric ICL = Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA); SMILE = small incision lenticule extraction; TG-LASIK = topography-guided laser in situ keratomileusis

Figure 3.

Change in lines of corrected distance visual acuity (CDVA) (Snellen) at 12 months. Toric ICL = Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA); SMILE = small incision lenticule extraction; TG-LASIK = topography-guided laser in situ keratomileusis

Stability

Neither mean sphere nor mean cylinder changed by more than 0.50 D in any of the three platforms from 1 to 12 months (Figure 4). From 1 to 3 months, 3 to 6 months, and 9 to 12 months, MRSE and mean refractive cylinder (MRCYL) changed less than 1.00 D in 99% and 98% of eyes, respectively. Of note, Toric ICL provided data from 6 to 12 months; for ease of comparison, these data were grouped into the 9- to 12-month groups.

Sphere and mean refractive cylinder (MRCYL) over time for Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK). The original U.S. Food and Drug Administration cylinder values for Toric ICL are reported as negative values for easier graphical comparison.

Figure 4.

Sphere and mean refractive cylinder (MRCYL) over time for Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL), small incision lenticule extraction (SMILE), and topography-guided laser in situ keratomileusis (TG-LASIK). The original U.S. Food and Drug Administration cylinder values for Toric ICL are reported as negative values for easier graphical comparison.

Accuracy

Fewer Toric ICL eyes (77%) achieved within ±0.50 D of intended MRSE compared to SMILE eyes (95%) and TG-LASIK eyes (95%) at the 12-month time point (P < .001) (Figure 5). More than 97% of eyes achieved within ±1.00 D of intended MRSE for all three platforms (P = .097).

Percentage of eyes that achieved within ±0.50 and ±1.00 diopters (D) of attempted mean refractive spherical equivalent (MRSE) at 12 months. Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL) had a significantly lower percentage of eyes within ±0.50 D compared to small incision lenticule extraction (SMILE) and topography-guided laser in situ keratomileusis (TG-LASIK) (P < .001).

Figure 5.

Percentage of eyes that achieved within ±0.50 and ±1.00 diopters (D) of attempted mean refractive spherical equivalent (MRSE) at 12 months. Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL) had a significantly lower percentage of eyes within ±0.50 D compared to small incision lenticule extraction (SMILE) and topography-guided laser in situ keratomileusis (TG-LASIK) (P < .001).

SMILE was consistently more accurate than Toric ICL within smaller degrees of cylinder (Figure 6). At 12 months, 76.4% of SMILE eyes were within ±0.25 D of attempted cylinder compared to 40.2% of Toric ICL eyes (P < .001); 92% of SMILE eyes were within ±0.50 D compared to 66% of Toric ICL eyes (P < .001); 97.6% of SMILE eyes were within ±1.00 D of attempted cylinder compared to 91.2% of Toric ICL eyes (P = .0014); and 100% of SMILE eyes were within ±2.00 D of attempted cylinder compared to 99.5% of Toric ICL eyes (P = .22).

Percent of patients who achieved mean refractive cylinder within ±0.25, ±0.50, ±1.00, and ±2.00 diopters (D) of attempted correction at 3 and 12 months. Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL) had a significantly lower percentage of eyes achieving within ±0.25 D (P < .001), ±0.50 D (P < .001), and ±1.00 D (P = .0014) at 12 months compared to small incision lenticule extraction (SMILE).

Figure 6.

Percent of patients who achieved mean refractive cylinder within ±0.25, ±0.50, ±1.00, and ±2.00 diopters (D) of attempted correction at 3 and 12 months. Visian Toric Implantable Collamer Lens (STAAR Surgical, Monrovia, CA) (Toric ICL) had a significantly lower percentage of eyes achieving within ±0.25 D (P < .001), ±0.50 D (P < .001), and ±1.00 D (P = .0014) at 12 months compared to small incision lenticule extraction (SMILE).

Vector analysis data were available for SMILE and TG-LASIK, stratified by 1.00 to 2.00 D and 2.00 to 3.00 D of preoperative cylinder (Figure A, available in the online version of this article). Both platforms had a CR above 0.89 with no significant differences (P > .05). Other stratified groups were not compared due to limited data.

Correction ratio for small incision lenticule extraction (SMILE) and topography-guided laser in situ keratomileusis (TG-LASIK). The surgically induced refraction correction for SMILE and TG-LASIK was measured at 6 and 3 months, respectively.

Figure A.

Correction ratio for small incision lenticule extraction (SMILE) and topography-guided laser in situ keratomileusis (TG-LASIK). The surgically induced refraction correction for SMILE and TG-LASIK was measured at 6 and 3 months, respectively.

Stratified Analyses

Toric ICL, SMILE, and TG-LASIK were stratified by preoperative MRSE of less than 7.00 D, 7.00 to 10.00 D, and greater than 10.00 D for comparison. SMILE values are from the 6-month time point and TG-LASIK values are from the 3-month time point. The data showed the MRSE remained relatively stable over time for all three modalities; thus, these stratifications were likely consistent over time. Stratified cylinder data were not available for Toric ICL or SMILE.

The percentage of eyes achieving MRSE within ±0.50 D and 20/20 UDVA was analyzed for these stratified groups. Table 3 summarizes the results of stratified analyses. The TG-LASIK group had no eyes with greater than 10.00 D.

Percentage of Eyes Achieving 20/20 UDVA, Within ±0.50 D, 20/40 UDVA, and Within ±1.00 D for Stratified MRSE Groupsa

Table 3:

Percentage of Eyes Achieving 20/20 UDVA, Within ±0.50 D, 20/40 UDVA, and Within ±1.00 D for Stratified MRSE Groups

Contrast Sensitivity and HOAs

Mesopic contrast sensitivity (without glare) was evaluated using the mean log contrast sensitivity change at 3, 6, and 12 cycles per degree (cpd) from baseline to 3 and 6 months (Figure B, available in the online version of this article). Data were provided by SMILE and TG-LASIK; the platforms had no statistically significant differences (P > .05). It was not possible to compare HOAs between the platforms because TG-LASIK evaluated only corneal aberrations, Toric ICL did not provide data for HOA, and SMILE data did not include the aberrometry acquisition method (corneal versus total wavefront).

Mesopic contrast sensitivity values for 3, 6, and 12 cycles per degree (cpd) at the 6-month time point. SMILE = small incision lenticule extraction; TG-LASIK = topography-guided laser in situ keratomileusis

Figure B.

Mesopic contrast sensitivity values for 3, 6, and 12 cycles per degree (cpd) at the 6-month time point. SMILE = small incision lenticule extraction; TG-LASIK = topography-guided laser in situ keratomileusis

Subjective Measures

Subjective symptoms (moderate and severe) of glare, halos, starbursts, and double vision were experienced by less than 4% of eyes and were not significantly different for Toric ICL and TG-LASIK at 12 months. Toric ICL (3.0%) had a significantly higher percentage of eyes with severe dryness compared to TG-LASIK (0%) (P = .0096), whereas TG-LASIK (3%) had a significantly higher percentage with moderate fluctuation in vision compared to Toric ICL (0%) (P = .001). Moderate dryness was reported in 6% of Toric ICL eyes compared to TG-LASIK (P = .395). Patient satisfaction was 100% after Toric ICL and 98.4% after TG-LASIK. These data were unavailable for SMILE.

Discussion

All three platforms analyzed in this study had excellent efficacy, safety, stability, and accuracy of refractive correction. Each platform performed markedly above the FDA efficacy threshold required for market approval, with nearly 100% of eyes achieving UDVA of 20/40 or better. With regard to safety, all platforms were concluded to be equally safe in preserving CDVA, showing either no change or gain of CDVA in at least 90% of treated eyes at 12 months. More than 97% of eyes were within ±1.00 D of intended MRSE across all three platforms. Mesopic contrast sensitivity was similar between SMILE and TG-LASIK, and subjective measures such as glare, halos, and starbursts were similar between Toric ICL and TG-LASIK.

Our analysis also revealed several unexpected trends. In the UDVA analysis of SMILE versus TGLASIK at 3 and 12 months, we reported significant differences that may correlate with improvement in efficacy over time (Figure 2). At 3 months, a higher percentage of eyes reported lower efficacy (postoperative UDVA compared to preoperative CDVA) of SMILE compared to TG-LASIK. In a similar analysis, Kanellopoulos16 reported that fewer eyes achieved 20/20 postoperative UDVA for SMILE compared to TG-LASIK, attributing this difference to topography customization and intraoperative compensation by the excimer laser. However, in our study, eyes undergoing SMILE had a significant improvement in efficacy of refractive correction between 3 and 12 months, whereas the majority of TG-LASIK eyes showed excellent efficacy at 3 months. Our analysis indicates that UDVA at 3 months is a possible prognostic indicator of UDVA over time for TG-LASIK, because fewer than 1% of eyes gained UDVA lines over the next 9 months. Although previous studies only measured short-term outcomes at 3 months, the results of our study may indicate subtle corneal remodeling after SMILE that occurs even up to 12 months after surgery, correlating with improvement in UDVA.17

The above findings apply to our results for stratified analyses because for the less than 7.00 D MRSE group, TG-LASIK performed significantly better than SMILE in the percentage of eyes that had postoperative visual acuity of 20/20 or better. However, the percentage of eyes that achieved MRSE within ±0.50 D was not significantly different between the two platforms. We postulate that because the data were taken from different time points (SMILE at 6 months and TG-LASIK at 3 months), UDVA for SMILE eyes would have continued to improve over time due to corneal remodeling, and additional refractive correction would not have improved visual acuity. Therefore, we do not conclude that TG-LASIK is superior to SMILE solely based on the limited data provided by the FDA.

In eyes with preoperative MRSE of greater than 10.00 D, there was a significant difference for postoperative MSRE within ±0.50 D between Toric ICL (66%) and SMILE (100%). There was also a lower percentage of eyes achieving 20/20 UDVA for Toric ICL (73%) versus SMILE (90%), although this difference was not statistically significant. Toric ICL is considered the standard of treatment for severely high myopia, and therefore we expected this device to perform superiorly in this cohort of individuals with myopia measuring greater than 10.00 D. However, in this subset of eyes, SMILE performed significantly better in achieving postoperative MRSE within ±0.50 D. It is important to mention that only 10 eyes with greater than 10.00 D underwent SMILE compared to 68 eyes with Toric ICL. Despite this relatively small population, this is a novel finding that has not been accounted for to date. Although the consensus understanding among refractive surgeons is that Toric ICL is the treatment of choice in severely high myopia, the current study indicates that SMILE provides comparable outcomes for these patients. However, inadequate preoperative corneal thickness may limit the candidacy of patients with high myopia for the SMILE procedure.

In addition to the distinct outcomes of SMILE and Toric ICL for postoperative MRSE, SMILE was consistently and significantly more accurate than Toric ICL within ±0.25 D, ±0.50 D, and ±1.00 D of cylinder. This strengthens the possibility that SMILE may be a comparable option to Toric ICL. Previous studies have speculated that the seemingly larger change in MRSE and MRCYL may be attributed to the difference in refractive cylinder correction because Toric ICL comes in 0.50 D increments, whereas the other modalities are in 0.25 D increments.18 This may also account for the lower accuracy of MRSE in Toric ICL compared to SMILE for eyes with preoperative MRSE of greater than 10.00 D, as well as the statistically significant 3.5% higher rate of surgical reintervention for Toric ICL compared to SMILE. Interestingly, Toric ICL had a similar percentage of eyes achieving 20/20 UDVA compared to SMILE at 12 months. We speculate that Toric ICL maintains visual acuity, despite a seemingly less accurate refractive correction, due to its anatomic placement behind the iris, which is closer to the optical nodal point and allows for relatively greater light magnification. Of note, we were not able to stratify these results by cylinder, and it is unknown whether eyes with high myopic astigmatism disproportionately affected our cylinder accuracy results.

Limitations

Our study had several limitations. The SSED for both SMILE and TG-LASIK included eyes with levels of myopia or astigmatism outside of the approved FDA indications for each technique. The Toric ICL SSED did not provide enough information to determine whether eyes outside of the approved indications were included in the study. Most of the data provided grouped all eyes together. This may limit the external validity of the current study because eyes with degrees of refractive error that fell outside of the FDA indications may have worse postoperative outcomes.

The three platforms have disparate inclusion criteria that may also affect postoperative outcomes. Toric ICL is the only modality offered to patients with astigmatism as high as 4.00 D of cylinder, whereas SMILE and TG-LASIK are only approved for up to 3.00 D of cylinder. In the patient sample, the average preoperative MRSE of Toric ICL was more than twice that of TG-LASIK and more than 1.5 times that of SMILE. Only 10 eyes with greater than 10.00 D MRSE underwent SMILE compared to 68 eyes with Toric ICL. Although we attempted to stratify by MRSE to minimize confounding variables, our ability to stratify was limited without raw data. This may affect the postoperative accuracy results because it is known that patients with higher myopic astigmatism have worse outcomes compared to patients with stable, moderate refractive error.19,20 Additionally, SMILE included 50 eyes with sphere only and TG-LASIK included 38 eyes with sphere only. All eyes that received Toric ICL had myopic astigmatism, which may have caused more difficulty in achieving favorable visual outcomes. These differences in sample size and baseline level of refractive error may be confounding factors in our comparison. Therefore, despite proper statistical significance, further investigation of raw data with larger sample sizes and homogenous groups is indicated.

Lack of raw data prevented the use of standardized graphs as proposed by Reinstein and Waring21 and Alpins.22 Additionally, analyses were limited by the available data; Toric ICL did not include HOA data and SMILE data did not include the aberrometry acquisition method (corneal versus total wavefront), preventing HOA analysis. Ideally, FDA-reported data should be published in a standardized format to improve the understanding of efficacy and safety in all patient groups. From a social aspect of health care, the authors also noted the overwhelming majority of patients to be white, limiting the generalizability of the current study to various ethnic groups. We speculate that socioeconomic status plays a role in accessibility to refractive procedures and may affect outcomes.

References

  1. Holden BA, Fricke TR, Wilson DA, et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. 2016;123(5):1036–1042. doi:10.1016/j.ophtha.2016.01.006 [CrossRef]26875007
  2. Kapetanakis VV, Chan MPY, Foster PJ, Cook DG, Owen CG, Rudnicka AR. Global variations and time trends in the prevalence of primary open angle glaucoma (POAG): a systematic review and meta-analysis. Br J Ophthalmol. 2016;100(1):86–93. doi:10.1136/bjophthalmol-2015-307223 [CrossRef]
  3. Ambrósio R Jr, Wilson S. LASIK vs LASEK vs PRK: advantages and indications. Semin Ophthalmol. 2003;18(1):2–10. doi:10.1076/soph.18.1.2.14074 [CrossRef]12759854
  4. Schallhorn SC, Amesbury EC, Tanzer DJ. Avoidance, recognition, and management of LASIK complications. Am J Ophthalmol. 2006;141(4):733–739. doi:10.1016/j.ajo.2005.11.036 [CrossRef]16564812
  5. Bhikoo R, Rayner S, Gray T. Toric implantable collamer lens for patients with moderate to severe myopic astigmatism: 12-month follow-up. Clin Exp Ophthalmol. 2010;38(5):467–474. doi:10.1111/j.1442-9071.2010.02273.x [CrossRef]20584028
  6. Piñero DP, Teus MA. Clinical outcomes of small-incision lenticule extraction and femtosecond laser-assisted wavefront-guided laser in situ keratomileusis. J Cataract Refract Surg. 2016;42(7):1078–1093. doi:10.1016/j.jcrs.2016.05.004 [CrossRef]27492109
  7. United States Food and Drug Administration. Visian Toric ICL Summary of Safety and Effectiveness Data (P030016/S001). https://www.accessdata.fda.gov/cdrh_docs/pdf3/P030016S001b.pdf. Accessed June 19, 2019.
  8. United States Food and Drug Administration. VisuMax Femtosecond Laser Summary of Safety and Effectiveness Data (P150040/S003). https://www.accessdata.fda.gov/cdrh_docs/pdf15/p150040s003b.pdf. Accessed June 19, 2019.
  9. United States Food and Drug Administration. ALLEGRETTO WAVE® Eye-Q Excimer Laser Summary of Safety and Effectiveness Data (P020050/S12). https://www.accessdata.fda.gov/cdrh_docs/pdf2/P020050S012b.pdf. Accessed June 19, 2019.
  10. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Toronto: Technometrics; 1988.
  11. R Development Core Team. R: A language and environment for statistical computing. Vienna: R Found Stat Comput; 2016. doi:10.1017/CBO9781107415324.004 [CrossRef]
  12. Pearson KX. On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Lond Edinb Dublin Philos Mag J Sci. 2009;Series 5:157–179. doi:10.1080/14786440009463897 [CrossRef]
  13. Glass GV, Peckham PD, Sanders JR. Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Review of Educational Research. 1972;42(3):237–288. doi:10.3102/00346543042003237 [CrossRef]
  14. Harwell MR, Rubinstein EN, Hayes WS, Olds CC. Summarizing Monte Carlo results in methodological research: the one- and two-factor fixed effects ANOVA cases. J Educ Stat. 1992;17(4):315–339. doi:10.3102/10769986017004315 [CrossRef]
  15. Lix LM, Keselman JC, Keselman HJ. Consequences of assumption violations revisited: a quantitative review of alternatives to the one-way analysis of variance F test. Rev Educ Res. 1996;66(4):579. doi:10.2307/1170654 [CrossRef]
  16. Kanellopoulos AJ. Topography-guided LASIK versus small incision lenticule extraction (SMILE) for myopia and myopic astigmatism: a randomized, prospective, contralateral eye study. J Refract Surg. 2017;33(5):306–312. doi:10.3928/1081597X-20170221-01 [CrossRef]28486721
  17. Hansen RS, Lyhne N, Grauslund J, Vestergaard AH. Small-incision lenticule extraction (SMILE): outcomes of 722 eyes treated for myopia and myopic astigmatism. Graefes Arch Clin Exp Ophthalmol. 2016;254(2):399–405. doi:10.1007/s00417-015-3226-5 [CrossRef]
  18. 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
  19. Kymionis GD, Tsiklis NS, Astyrakakis N, Pallikaris AI, Panagopoulou SI, Pallikaris IG. Eleven-year follow-up of laser in situ keratomileusis. J Cataract Refract Surg. 2007;33(2):191–196. doi:10.1016/j.jcrs.2006.11.002 [CrossRef]17276257
  20. Sugar A, Rapuano CJ, Culbertson WW, et al. Laser in situ keratomileusis for myopia and astigmatism: safety and efficacy: a report by the American Academy of Ophthalmology. Ophthalmology. 2002;109(1):175–187. doi:10.1016/S0161-6420(01)00966-6 [CrossRef]11772601
  21. Reinstein DZ, Waring GO III, . Graphic reporting of outcomes of refractive surgery. J Refract Surg. 2009;25(11):975–978. doi:10.3928/1081597X-20091016-01 [CrossRef]19921764
  22. Alpins NA. A new method of analyzing vectors for changes in astigmatism. J Cataract Refract Surg. 1993;19(4):524–533. doi:10.1016/S0886-3350(13)80617-7 [CrossRef]8355160

Patient Demographics

ParameterToric ICLSMILETG-LASIKPa
Eyes (n)210357249
Gender, male/female (n)55/69148/20993/119.786
Age, mean ± SD (years)35 ± 6.8b33.1 ± 7.2b34.0 ± 9.3.020
Preoperative sphere (D)−10.36c−4.81 ± 2.39b−4.01 ± 2.57b< .001
Preoperative MRSE (D)−9.38 ± 2.67b−5.48c−4.61 ± 2.43b< .001
Preoperative cylinder (D)1.95 ± 0.84b−1.33 ± 0.80b,d−1.19 ± 1.23b,d< .001

Safety Parameters at 12 Months

CDVAToric ICLSMILETG-LASIK
CDVA loss ⩾ 2 lines3/210 (1.4%)0/348 (0%)1/244 (0.41%)
CDVA worse than 20/40 (if preoperative CDVA was 20/20 or better)a0/210 (0%)0/348 (0%)0/225 (0%)

Percentage of Eyes Achieving 20/20 UDVA, Within ±0.50 D, 20/40 UDVA, and Within ±1.00 D for Stratified MRSE Groupsa

ParameterToric ICLSMILETG-LASIKP
Interval12 months6 months3 months
UDVA 20/20
  < 7.00 D9484b93b.0047
  7.00 to 10.00 D848593.28
  > 10.00 D7390N/A.25
MRSE within ±0.5 D
  < 7.00 D859494.102
  7.00 to 10.00 D829185.23
  > 10.00 D66b100bN/A< .001
UDVA 20/40
  < 7.00 D979999.62
  7.00 to 10.00 D9898100.52
  > 10.00 D91100N/A.32
MRSE within ±1.00 D
  < 7.00 D10010099.22
  7.00 to 10.00 D999698.53
  > 10.00 D94100N/A.43

FDA Indications for Use and Range Used in the Study

ParameterToric ICLSMILETG-LASIK
Age (y) criteria> 21 and < 45≥ 22≥ 18
Time on market10 months9 months6 years
SSED-reported range of refractive correction
  Spherical equivalent (D)Not providedUp to −11.50Up to −9.00
  Cylinder (D)a−1.00 to −4.00Up to −3.000.00 to −6.00
  Sphere (D)−3.00 to −20.00−1.00 to −10.00Up to −9.00
FDA-approved indication
  Spherical equivalent (D)−3.00 to −15.00Up to −10.00Up to −9.00
  Cylinder (D)−1.00 to −4.00−0.75 to −3.00Up to −3.00
  Sphere (D)Not provided−1.00 to −10.00Up to −8.00
Authors

From John A. Moran Eye Center, Department of Ophthalmology and Visual Sciences, University of Utah School of Medicine, Salt Lake City, Utah (MM); Utah Lions Eye Bank, Murray, Utah (MM); Hoopes Durrie Rivera Research Center, Hoopes Vision, Draper, Utah (MM, YCR); McGovern Medical School at The University of Texas Health Science Center at Houston, Houston, Texas (ANS, UV); the Department of Ophthalmology, University of Arizona College of Medicine, Tucson, Arizona (MNM); University of Utah School of Medicine, Salt Lake City, Utah (JSS); and Virginia Commonwealth School of Medicine, Richmond, Virginia (JRB).

Supported by Research to Prevent Blindness, Inc., New York, New York.

The authors have no financial or proprietary interest in the materials presented herein.

The authors thank Phillip C. Hoopes, MD, for his contributions to the article.

AUTHOR CONTRIBUTIONS

Study concept and design (MM, ANS, MNM, YCR); data collection (MM, ANS, MNM, JRB, YCR); analysis and interpretation of data (MM, ANS, MNM, UV, JSS, YCR); writing the manuscript (MM, ANS, MNM, UV, JSS, JRB, YCR); critical revision of the manuscript (MM, ANS, MNM, UV, YCR); statistical expertise (ANS, MNM, JSS); administrative, technical, or material support (MM, YCR); supervision (MM, YCR)

Correspondence: Majid Moshirfar, MD, FACS, Hoopes Durrie Rivera Research Center, Hoopes Vision, 11820 S. State Street Suite #200, Draper, UT 84020. E-mail: cornea2020@me.com

Received: July 16, 2019
Accepted: September 30, 2019

10.3928/1081597X-20190930-01

Sign up to receive

Journal E-contents