Ametropia, a major cause of reversible vision impairment, is addressed by corrective refractive surgeries.1 The safety, efficacy, and predictability of corneal refractive surgeries have improved with developments in surgical techniques and the introduction of the femtosecond laser.2,3 However, deviations after refractive correction exist with any type of corneal refractive surgery.4,5
The degree of myopia, laser energy, patient age, optical zone diameter, and procedural room environment are known to influence refractive surgery outcomes.6–10 The higher the degree of myopia, the greater is the undercorrection after surgery.6 Low laser energy may be associated with more undercorrection and a faster and better visual recovery after small incision lenticule extraction (SMILE).6,7 Older patients achieve more refractive change with the same intended dioptric correction compared to younger patients.8 Environmental factors such as higher humidity and temperature correlate with more undercorrection in laser in situ keratomileusis (LASIK).9 A larger optical zone induces more overcorrection than a smaller optical zone.10 Preoperative keratometry also affects correction outcomes after refractive surgery.11 Steeper corneas have more overcorrection with laser epithelial keratomileusis (LASEK) and LASIK.11,12 However, the effect of preoperative keratometry on refractive outcomes after SMILE, a flapless and all-femtosecond laser refractive procedure without excimer laser ablation and flap creation,13 is yet to be investigated.
The purpose of this study was to evaluate the relationship between preoperative keratometry and postoperative residual spherical equivalent in patients undergoing SMILE.
Patients and Methods
Patients and Study Design
The data of right eyes belonging to 515 consecutive patients treated with SMILE for myopia or myopic compound astigmatism at the Tianjin Eye Hospital, Tianjin Medical University, from July 2018 to April 2019 were retrospectively analyzed. The study design was approved by the ethics committee of the Tianjin Eye Hospital and it adhered to the tenets of the Declaration of Helsinki. Before study initiation, informed consent was obtained from each patient. Inclusion criteria were age 18 years or older, corrected distance visual acuity (CDVA) of 20/25 or better, myopic astigmatism correction of less than 2.50 diopters (D), and stable refraction for the past 2 years. Exclusion criteria were the presence of active ocular disease, previous ocular surgery or ocular trauma, suspected keratoconus, and systemic diseases such as diabetes or connective tissue disorders. Patients were requested to discontinue the use of soft and rigid contact lenses for 2 and 4 weeks before surgery, respectively.
Preoperative keratometry and postoperative residual refraction were recorded. Full ophthalmologic examination preoperatively and at postoperative day 1, day 7, month 1, and month 3 included measurements of uncorrected distance visual acuity (UDVA), CDVA, manifest and cycloplegic refraction, slit-lamp biomicroscopy, and Scheimpflug topography (Pentacam HR; Oculus Optikgeräte GmbH). Efficacy index is the ratio of mean postoperative UDVA to mean preoperative CDVA. Safety index is the ratio of mean postoperative CDVA to mean preoperative CDVA.
All SMILE procedures were conducted in the Tianjin Eye Hospital by a single surgeon (YW). The postoperative goal for all treatments in this study was emmetropia. The VisuMax 500-kHz femtosecond laser (Carl Zeiss Meditec AG) was used. The parameters used included an optical zone of 6.5 to 7 mm, a cap diameter of 7.5 to 8 mm, a predetermined cap thickness of 120 µm, and an energy of 125 to 160 nJ. The side cut was placed at the 12-o'clock position of the cornea with an angle of 90 degrees and a circumferential width of 2 to 3 mm. The posterior plane (spiral in), border, anterior plane (spiral out), and side cutting were scanned successively. Following removal of the lenticule, the incision was flushed with balanced salt solution (Alcon Laboratories, Inc). All patients received ofloxacin 0.3% (Tarivid; Santen, Inc.) four times a day for 3 days and fluorometholone 0.1% (Flumetholon; Santen, Inc.) four times a day for 2 weeks, and then tapered off over the following 2 weeks.
SPSS software version 20.0 (IBM Corporation) was used for data analysis. The Kolmogorov-Smirnov test was used to confirm data normality. To compare the refractive outcomes between eyes with flat corneas and eyes with steep corneas, we segregated the cohort into thirds. The lower third with the flattest keratometry readings was identified as the “flat group” and the upper third with the steepest keratometry readings was identified as the “steep group.”11 Postoperative spherical equivalent between the two groups was compared using the unpaired two-tailed t test after eliminating bias for age, sex, and preoperative spherical equivalent by 1:1 propensity score match (PSM) analysis. Pearson correlation and univariate regression analyses were used to determine the relationship between preoperative keratometry and postoperative spherical equivalent. The proportion of correction is the ratio of achieved spherical equivalent to attempted spherical equivalent, reflecting the relative refractive outcomes. The first quartile with the lowest preoperative spherical equivalent and the fourth quartile with the highest preoperative spherical equivalent were also identified, and the analysis was repeated in each quartile. A P value of .05 or less was considered statistically significant. Data are expressed as the mean ± standard deviation.
A total of 515 right eyes of 515 patients (261 men, 254 women) were analyzed. The mean age of the patients was 24 ± 6 years (range: 18 to 45 years). Preoperatively, the mean spherical refraction was −5.29 ± 1.85 D (range: −1.50 to −9.25 D), the mean cylinder was −0.76 ± 0.56 D (range: 0.00 to −2.50 D), the mean spherical equivalent was −5.67 ± 1.87 D (range: −1.63 to −9.75 D), and the mean keratometry was 43.10 ± 1.30 D (range: 38.90 to 47.00 D). Three months after SMILE, the mean spherical refraction was −0.04 ± 0.16 D, and the mean spherical equivalent was −0.07 ± 0.18 D. Preoperative mean keratometry of 515 eyes displayed a normal distribution (P = .79) (Figure 1).
Distribution of preoperative mean keratometry in 515 eyes. D = diopters
Standard Refractive Analyses
Standardized graphs of the SMILE results are displayed in Figure 2. In total, 474 (92%) eyes exhibited a postoperative distance UDVA of 20/20 or better. The mean efficacy index at 3 months postoperatively was 1.17 ± 0.21. The CDVA was the same in 118 (23%) eyes, whereas 237 (46%) eyes gained one line and 155 (30%) eyes gained two or more lines at postoperative month 3. Five (1%) eyes lost one line, and no eyes lost two lines of CDVA at postoperative month 3. The mean safety index was 1.23 ± 0.18. The relationship between attempted and achieved correction was high, with a correlation coefficient of 0.99. Of all eyes, 505 (98%) eyes were within ±0.50 D and 515 (100%) eyes were within ±1.00 D of the attempted refraction at postoperative month 3. The astigmatism of all eyes was within ±0.50 D at 3 months after SMILE.
Visual and refractive outcomes of small incision lenticule extraction (SMILE). (A) Cumulated Snellen corrected distance visual acuity (CDVA) before surgery and uncorrected distance visual acuity (UDVA) after surgery in 515 eyes from 515 patients treated with SMILE targeted for emmetropia. (B) Distribution of the change in Snellen lines of CDVA before surgery to UDVA at 3 months after SMILE. (C) Distribution of the change in Snellen lines of CDVA 3 months after SMILE. (D) Achieved (3 months after SMILE) versus attempted change in spherical equivalent refraction (SEQ). (E) Distribution of the error in spherical equivalent refraction (attempted subtracted by achieved change). (F) Distribution of refractive astigmatism before surgery and 3 months after surgery. D = diopters
Refractive Outcomes in Flat and Steep Corneas
PSM analysis was employed to overcome the selection biases between the flat and steep groups. Finally, 93 pairs of eyes were matched by 1:1 PSM analysis. As shown in Table 1, after matching two groups for age, sex, and preoperative spherical equivalent, both postoperative spherical equivalent and proportion of correction were significantly different between the flat and steep groups. The difference in refractive outcomes was small; however, greater undercorrection was found in the steep group (P = .001). Eyes in the flat and steep groups achieved 100.02% ± 2.85% and 98.15% ± 3.37% myopia correction, respectively. The difference between the two groups was significant (P < .001).
t Test in Flat and Steep Groups After Propensity Score Match Analysis
Effect of Preoperative Keratometry on Refractive Outcomes
The correlation between the mean preoperative keratometry and the postoperative spherical equivalent was significant (r = −0.215, P < .001); the steeper the cornea, the more undercorrection the refractive correction resulted in. Figure A (available in the online version of this article) shows the scatterplot of regression analysis of data from all eyes. Each diopter of steeper keratometry resulted in 0.52% (0.026 D) more undercorrection (R2 = 0.037, P < .001). To avoid a possible influence of preoperative refractive status, the regression analyses were repeated in the first (lowest preoperative spherical equivalent) and fourth (highest preoperative spherical equivalent) quartiles, respectively. Table 2 shows the results. The difference in postoperative spherical equivalent between the two quartiles was significant (P < .001). Correlation between the mean preoperative keratometry and postoperative spherical equivalent was significant in the lower preoperative myopia group (R2 = 0.059, P = .006), but not significant in the higher myopia group (R2 = 0.001, P = .809), indicating that corneas with a lower preoperative myopia were more influenced by the preoperative corneal curvature.
Scatterplot of the regression analysis between mean preoperative keratometric values and (A) postoperative spherical equivalent (SE) and (B) proportion of correction. D = diopters
Regression Analysis Between Preoperative Keratometry and Postoperative SE in the First and Fourth Preoperative SE Quartiles
This study found a negative correlation between preoperative keratometry and refractive outcomes of SMILE for myopia correction. That is, the refractive outcome of SMILE was more undercorrected for steeper corneas, especially in the eyes with lower preoperative myopia.
Standard refractive analyses showed that SMILE was safe, effective, and predictable in this study. Age, sex, preoperative spherical equivalent, and the sequence of the procedure are reported to affect the refractive outcomes of refractive surgery.6,8,14,15 All eyes in this study were right eyes because the right eye is the first to be operated on in our center. After matching for age, sex, and preoperative spherical equivalent in the flat and steep groups, we found that more undercorrection (both absolute and relative outcomes) occurred in eyes with steeper corneas. Thus, it is supposed that eyes with flatter corneas tend to achieve more refractive change than eyes with steeper corneas for the same attempted dioptric correction in SMILE. The correlation analysis also showed that a steeper cornea experienced more undercorrection. Because a high corneal power is associated with increasing myopic refraction,16 it is possible that more refractive change will be required for steeper corneas compared to flatter corneas for the same attempted dioptric correction. In the current study, each diopter of steeper keratometry resulted in 0.03 D more undercorrection (P < .001). Although the change per diopter may appear deceptively small, it could translate into large undercorrection or overcorrection in very steep or very flat corneas, and hence is worth considering in such cases. It should be noted that in this univariate regression model, the R2 value was small (R2 = 0.037, P < .001). To further improve the effectiveness of the model to predict postoperative spherical equivalent, multivariate regression models including more relevant predictive factors as covariates should be conducted in future studies.
It has been reported that the degree of preoperative ametropia correlates with the final outcome.6,14 Consistent with previous studies, more undercorrection was found in the quartile with the highest preoperative spherical equivalent in this study. Although the undercorrection was greater in the eyes with higher myopia, the correlation between preoperative keratometry and residual spherical equivalent was not significant in the quartile with the highest preoperative spherical equivalent, but was significant in the quartile with the lowest preoperative spherical equivalent. These results indicate that steep keratometry is a possible contributor to the undercorrection that occurred in the eyes with low preoperative myopia, but the influence of keratometry is not evident in the eyes with high myopia. In fact, we further analyzed the correlation between preoperative spherical equivalent and residual spherical equivalent in two quartiles and found the correlation was significant in the quartile with the highest preoperative spherical equivalent, but not significant in the quartile with the lowest preoperative spherical equivalent (data not shown). It is supposed that instead of keratometry, preoperative spherical equivalent was the dominant factor influencing refractive outcomes in the eyes with high myopia. In SMILE, the depth of the gap between the cap and residual stromal bed depends on the degree of myopia.17 The gap increases with increasing myopia and causes an anterior shift of the posterior corneal surface resulting in more undercorrection.17 Thus, the influence of the degree of preoperative myopia on postoperative refractive outcomes may be more evident when the preoperative myopia is higher.
The effect of keratometry on LASIK correction outcomes has been studied widely. Rao et al12 reported increased undercorrection in eyes with flat corneas and preoperative myopia from −10.00 to −12.00 D. Pérez-Santonja et al18 reported a tendency toward undercorrection in flatter corneas when correcting high myopia of −8.00 to −20.00 D. Contrary to LASIK studies, we found that steeper corneas show more refractive undercorrection after SMILE. The differences between the results of LASIK and SMILE can be attributed to the stromal flap created in LASIK.11,19 The lack of intact lamellar structure constraint and the relaxed peripheral lamellar segment expansion could induce a central cornea flattening in LASIK.20 In addition, the preoperative refractive status may mask the relationship between keratometry and refractive outcomes. Christiansen et al21 reported better visual outcomes in flatter corneas and increased undercorrection in steeper corneas of eyes with moderate myopia of −2.00 to −5.99 D after LASIK. Esquenazi and Mendoza22 found undercorrection of hyperopia in eyes with preoperative keratometry of greater than −45.00 D after LASIK. These results are contrary to other studies that included patients with preoperative myopia of greater than −8.00 D.12,18 In the current study, interactions between preoperative keratometry and preoperative spherical equivalent on postoperative spherical equivalent was tested by including the interaction term (ie, preoperative keratometry × preoperative spherical equivalent) in the multiple linear regression model for the entire dataset. However, the interaction term was not statistically significant (P = .730). The possible reason is the limited range of preoperative spherical equivalent in this study. The degree of ametropia should be taken into account when analyzing the effect of preoperative keratometry on refractive outcomes.
Our study revealed the occurrence of greater under-correction after SMILE in eyes with steeper corneas. Steep corneal curvature may be a predictor of under-correction occurring in eyes with low myopia. However, the effect of preoperative keratometry on the final refractive outcomes is not significant in eyes with high myopia. Validation of the study findings in larger data sets may improve the predictability of current SMILE nomograms.
- Kim TI, Alió Del Barrio JL, Wilkins M, Cochener B, Ang M. Refractive surgery. Lancet. 2019;393(10185):2085–2098. doi:10.1016/S0140-6736(18)33209-4 [CrossRef]
- Han T, Zheng K, Chen Y, Gao Y, He L, Zhou X. Four-year observation of predictability and stability of small incision lenticule extraction. BMC Ophthalmol. 2016;16(1):149. doi:10.1186/s12886-016-0331-0 [CrossRef]
- Sekundo W, Gertnere J, Bertelmann T, Solomatin I. One-year refractive results, contrast sensitivity, high-order aberrations and complications after myopic small-incision lenticule extraction (ReLEx SMILE). Graefes Arch Clin Exp Ophthalmol. 2014;252(5):837–843. doi:10.1007/s00417-014-2608-4 [CrossRef]
- Vestergaard A, Ivarsen AR, Asp S, Hjortdal JØ. Small-incision lenticule extraction for moderate to high myopia: predictability, safety, and patient satisfaction. J Cataract Refract Surg. 2012;38(11):2003–2010. doi:10.1016/j.jcrs.2012.07.021 [CrossRef]
- Chayet AS, Assil KK, Montes M, Espinosa-Lagana M, Castellanos A, Tsioulias G. Regression and its mechanisms after laser in situ keratomileusis in moderate and high myopia. Ophthalmology. 1998;105(7):1194–1199. doi:10.1016/S0161-6420(98)97020-8 [CrossRef]
- Cui T, Wang Y, Ji S, et al. Applying machine learning techniques in nomogram prediction and analysis for SMILE treatment. Am J Ophthalmol. 2020;210:71–77. doi:10.1016/j.ajo.2019.10.015 [CrossRef]
- Donate D, Thaëron R. Lower energy levels improve visual recovery in small incision lenticule extraction (SMILE). J Refract Surg. 2016;32(9):636–642. doi:10.3928/1081597X-20160602-01 [CrossRef]
- Rao SN, Chuck RS, Chang AH, LaBree L, McDonnell PJ. Effect of age on the refractive outcome of myopic photorefractive keratectomy. J Cataract Refract Surg. 2000;26(4):543–546. doi:10.1016/S0886-3350(99)00465-4 [CrossRef]
- Walter KA, Stevenson AW. Effect of environmental factors on myopic LASIK enhancement rates. J Cataract Refract Surg. 2004;30(4):798–803. doi:10.1016/j.jcrs.2004.01.001 [CrossRef]
- Liu M, Sun Y, Wang D, et al. Decentration of optical zone center and its impact on visual outcomes following SMILE. Cornea. 2015;34(4):392–397. doi:10.1097/ICO.0000000000000383 [CrossRef]
- de Benito-Llopis L, Teus MA, Sánchez-Pina JM, Gil-Cazorla R. Influence of preoperative keratometry on refractive results after laser-assisted subepithelial keratectomy to correct myopia. J Cataract Refract Surg. 2008;34(6):968–973. doi:10.1016/j.jcrs.2008.01.027 [CrossRef]
- Rao SK, Cheng AC, Fan DS, Leung AT, Lam DS. Effect of pre-operative keratometry on refractive outcomes after laser in situ keratomileusis. J Cataract Refract Surg. 2001;27(2):297–302. doi:10.1016/S0886-3350(00)00746-X [CrossRef]
- 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]
- Wang M, Zhang Y, Wu W, et al. Predicting refractive outcome of small incision lenticule extraction for myopia using corneal properties. Transl Vis Sci Technol. 2018;7(5):11. doi:10.1167/tvst.7.5.11 [CrossRef]
- Hjortdal JØ, Vestergaard AH, Ivarsen A, Ragunathan S, Asp S. Predictors for the outcome of small-incision lenticule extraction for myopia. J Refract Surg. 2012;28(12):865–871. doi:10.39 28/1081597X-20121115-01 [CrossRef]
- Muthu Krishnan V, Jayalatha K, Vijayakumar C. Correlation of central corneal thickness and keratometry with refraction and axial length: a prospective analytic study. Cureus. 2019;11(1):e3917. doi:10.7759/cureus.3917 [CrossRef]
- Ganesh S, Patel U, Brar S. Posterior corneal curvature changes following refractive small incision lenticule extraction. Clin Ophthalmol. 2015;9:1359–1364. doi:10.2147/OPTH.S84354 [CrossRef]
- Pérez-Santonja JJ, Bellot J, Claramonte P, Ismail MM, Alió JL. Laser in situ keratomileusis to correct high myopia. J Cataract Refract Surg. 1997;23(3):372–385. doi:10.1016/S0886-3350(97)80182-4 [CrossRef]
- Estopinal CB, Mian SI. LASIK flap: postoperative complications. Int Ophthalmol Clin. 2016;56(2):67–81. doi:10.1097/IIO.0000000000000107 [CrossRef]
- Roberts C. Biomechanics of the cornea and wavefront-guided laser refractive surgery. J Refract Surg. 2002;18(5):S589–S592.
- Christiansen SM, Neuffer MC, Sikder S, Semnani RT, Moshirfar M. The effect of preoperative keratometry on visual outcomes after moderate myopic LASIK. Clin Ophthalmol. 2012;6:459–464.
- Esquenazi S, Mendoza A. Two-year follow-up of laser in situ keratomileusis for hyperopia. J Refract Surg. 1999;15(6):648–652.
t Test in Flat and Steep Groups After Propensity Score Match Analysis
|Parameter||Flat Group||Steep Group||P|
|No. of eyes||93||93||–|
|Age (years)||23.51 ± 5.58||23.87 ± 5.87||.664|
|Sex (male, %)||54%||54%||1.000|
|Preoperative SE (D)||−4.98 ± 1.80||−5.01 ± 1.74||.927|
|Mean keratometry (D)||41.40 ± 0.74||44.49 ± 0.71||< .001|
| Range||38.90 to 42.60||43.70 to 47.00|
|Postoperative SE (D)||−0.02 ± 0.14||−0.09 ± 0.16||.001|
|Proportion of correction (%)||100.02 ± 2.85||98.15 ± 3.37||< .001|
Regression Analysis Between Preoperative Keratometry and Postoperative SE in the First and Fourth Preoperative SE Quartiles
|Parameter||Preoperative SE (D)||P|
|First Quartile||Fourth Quartile|
|< −4.375||> −7.125|
|No. of eyes||129||118||–|
|Preoperative SE (D)|
| Mean ± SD||−3.17 ± 0.76||−7.96 ± 0.62||< .001|
| Range||−1.63 to −4.38||−7.25 to −9.75|
|Postoperative SE (D)|
| Mean ± SD||−0.01 ± 0.12||−0.12 ± 0.19||< .001|
|Linear regression equation||Y = −0.020*X + 0.842||Y = −0.004*X + 0.044|