Journal of refractive surgery最新文献

Purpose: To develop a model for predicting postoperative keratometry in children undergoing cataract surgery with primary intraocular lens (IOL) implantation.

Methods: The longitudinal retrospective study included all children who underwent bilateral cataract surgery and had available intraoperative and postoperative keratometry data. Variables that could influence postoperative keratometry were analyzed, and a generalized estimating equation regression model was used to predict postoperative keratometry.

Results: One hundred twenty eyes of 60 children were included. The mean age at surgery was 2.95 ± 2.78 years (range: 0.12 to 10.95 years), the mean age at the last follow-up was 7.25 ± 4.61 years (range: 0.50 to 20.41 years), and the mean follow-up was 2.40 ± 2.33 years (range: 0.25 to 10.37 years). Keratometry measurements were obtained in both eyes during 510 postoperative visits. The mean keratometry values before surgery and at the last follow-up were 45.32 ± 2.57 diopters (D) (range: 40.75 to 53.88 D) and 44.62 ± 2.25 D (range: 40.63 to 51.25 D), respectively. Preoperative mean keratometry, age at cataract surgery, and age at follow-up were statistically significant predictors of postoperative keratometry. The model to predict postoperative keratometry = 3.1304 + 0.9388 × (baseline keratometry) + 1.8294 × Log10 (baseline age), -1.1336 × Log10 (age at follow-up), -0.7045 × Log10 (baseline age) × Log10 (age at follow-up) was determined. The correlation between measured and estimated keratometry was R = 0.92. Using this model, a hypothetical patient who had surgery at 1 year of age with a mean keratometry value of 44.00 D would be estimated to have a keratometry value of 42.94 D at 21 years of age.

Conclusions: The model is a good predictor for future keratometry after bilateral pediatric cataract surgery with an IOL implant. [J Refract Surg. 2025;41(3):e207-e212.].

Purpose: To report a deep learning neural network on anterior segment optical coherence tomography (AS-OCT) for automated detection of different keratorefractive laser surgeries-including laser in situ keratomileusis with femtosecond microkeratome (femto-LASIK), LASIK with mechanical microkeratome, photorefractive keratectomy (PRK), keratorefractive lenticule extraction (KLEx), and non-operated eyes-while also distinguishing between myopic and hyperopic treatments within these procedures.

Methods: A total of 14,948 eye scans from 2,278 eyes of 1,166 patients were used to develop a deep learning neural network algorithm with an 80/10/10 patient distribution for training, validation, and testing phases, respectively. The algorithm was evaluated for its accuracy, F1 scores, area under precision-recall curve (AUPRC), and area under receiver operating characteristic curve (AUROC).

Results: On the test dataset, the neural network was able to detect the different surgical classes with an accuracy of 96%, a weighted-average F1 score of 96%, and a macro-average F1 score of 96%. The neural network was further able to detect hyperopic and myopic subclasses within each surgical class, with an accuracy of 90%, weighted-average F1 score of 90%, and macro-average F1 score of 83%.

Conclusions: Neural networks can accurately classify a patient's keratorefractive laser history from AS-OCT scans, which may support treatment planning, intraocular lens calculations, and ectasia assessment, particularly in cases where electronic health records are incomplete. This represents a step toward transforming OCT from a diagnostic to a more comprehensive screening tool in refractive clinics. [J Refract Surg. 2025;41(3):e248-e256.].

Purpose: To evaluate the clinical and patient-reported outcomes of the toric version of a refractive extended depth of focus intraocular lens (IOL).

Methods: This was a prospective observational study including 26 eyes of 26 patients (age range: 58 to 88 years) who underwent cataract surgery with implantation of the LuxSmart toric IOL (Bausch & Lomb GmbH). Changes in distance, intermediate (80 cm), and near (40 cm) visual acuities, manifest refraction, defocus curve, distance photopic contrast sensitivity (CSV-1000 test), and patient-reported outcomes (Catquest-9SF questionnaire) were evaluated during a 3-month follow-up.

Results: Mean 3-month postoperative uncorrected distance, corrected distance, distance-corrected intermediate, and distance-corrected near visual acuities of 0.10 ± 0.11, 0.03 ± 0.08, 0.06 ± 0.13, and 0.38 ± 0.22 logarithm of the minimum angle of resolution (logMAR) were found, respectively. Uncorrected and corrected distance visual acuity improved significantly with surgery (P < .001). In the defocus curve, mean visual acuity was 0.30 logMAR or better for defocus between +0.50 and -2.00 D, without significant changes during the follow-up (P ⩾ .071). The postoperative spherical equivalent was within ±0.50 and ±1.00 diopters (D) in 92% and 96% of eyes, respectively. Surgically induced astigmatism prediction error ranged between -0.37 and 0.80 D, with a mean value of 0.11 ± 0.28 D. Mean 3-month postoperative contrast sensitivity was within or close to the upper limit of the normality range for most spatial frequencies. Mean 3-month postoperative overall Rasch calibrated Catquest score was -1.68 ± 1.04, with 96.2% of patients satisfied with the vision achieved. IOL rotations of 5º or less were observed in 92.3% of eyes.

Conclusions: The toric extended depth of focus IOL evaluated provides a good distance and intermediate functional vision in eyes with preexisting corneal astigmatism, with functional near visual performance for most patients. [J Refract Surg. 2025;41(3):e238-e247.].

Purpose: To report the visual and tomographic outcomes 1 year after porcine collagen lenticule implantation in eyes with advanced keratoconus.

Methods: Patients older than 18 years with advanced keratoconus having central corneal thickness (CCT) greater than 350 µm and visual acuity of 20/40 or better with contact lens, but intolerant to contact lenses, were included. A femtosecond laser-assisted stromal pocket was created at a depth of 140 µm. A Xenia implant (Gebauer Medizintechnik GmbH) was inserted into the stromal pocket using a glide. Uncorrected (UDVA) and corrected (CDVA) distance visual acuity testing, slit-lamp evaluation, corneal topography using Pentacam (Oculus Optikgeräte GmbH), and anterior segment optical coherence tomography using Optovue (XR-Avanti) was performed preoperatively and at 1 week, 1, 3, and 6 months, and 1 year after Xenia lenticule implantation.

Results: Nine eyes of 9 patients (6 men and 3 women) with a mean age of 28.4 years underwent Xenia lenticule implantation. Mean UDVA and CDVA improved from 1.43 ± 0.3 to 0.78 ± 0.17 logMAR (P = .0087) and 0.89 ± 0.13 to 0.45 ± 0.03 logMAR (P = .0005) at 1 year. The preoperative mean flat keratometry (K1) was 61.29 ± 3.76 diopters (D), steep keratometry (K2) was 68.09 ± 4.41 D, CCT was 354.71 ± 9.30 µm, average keratometry (Kmean) was 64.50 ± 3.98 D, and maximum keratometry (Kmax) was 77.07 ± 6.01 D. All keratometry values showed statistically significant flattening at 1 year with mean K1 of 40.70 ± 3.96 D (P = .0003), K2 of 43.80 ± 4.04 D (P = .0005), Kmean of 42.20 ± 4.00 D (P = .0003), and Kmax of 61.53 ± 4.70 D (P = .0152). There were no episodes of graft rejection. Two eyes had stromal melt within 3 months postoperatively, requiring lenticule explantation.

Conclusions: Significant improvement in UDVA, CDVA, and keratometry values was noted in patients undergoing Xenia lenticule implantation in advanced keratoconus. Preliminary results demonstrate that Xenia lenticule implantation could be an alternative to keratoplasty and human stromal lenticule implants in cases with advanced keratoconus. [J Refract Surg. 2025;41(3):e272-e279.].

Purpose: To evaluate the repeatability of angle kappa measurements obtained from the Pentacam (Oculus Optikgeräte GmbH), Sirius (Costruzione Strumenti Oftalmici), and Keratron Scout (BVI), and to assess the agreement with those obtained from the Amaris 750S (SCHWIND eye-tech-solutions).

Methods: This study included 110 patients scheduled to undergo transepithelial photorefractive keratectomy or femtosecond laser-assisted laser in situ keratomileusis. Before surgery, each patient underwent angle kappa measurement using three instruments: Pentacam, Sirius, and Keratron. Prior to corneal ablation with the excimer laser during surgery, the angle kappa was measured again using an Amaris 750S. Within-subject standard deviation, test-retest repeatability, within-subject coefficient of variation, and intraclass correlation coefficient (ICC) were calculated to assess the repeatability of the angle kappa measurements using the three instruments before surgery. A Bland-Altman plot was used to evaluate the agreement between these three instruments and the Amaris 750S to measure the angle kappa.

Results: For the kappa offsets measured by Sirius, Keratron, and Pentacam, the ICC values were 0.948, 0.950, and 0.931, respectively. For the measurement of the kappa axis using the three instruments, the ICC values were 0.996, 0.997, and 0.994, respectively. The 95% limits of agreement (LoA) of kappa offset between Keratron and Amaris 750S was from -0.15 to +0.09 mm, which was the narrowest among the three instruments compared with the Amaris 750S, and the narrowest 95% LoA of kappa axial was also between the Keratron and Amaris 750S.

Conclusions: The results measured before Keratron, whether kappa offset or kappa axial, were relatively consistent with those measured intraoperatively using the Amaris 750S. [J Refract Surg. 2025;41(2):e213-e219.].

Purpose: To evaluate and compare non-corneal intraocular higher order aberrations (HOAs) in keratoconic and normal myopic eyes.

Methods: Eighty-eight keratoconic and 106 normal myopic eyes were examined using high-resolution (1) pyramidal ocular wavefront sensor PERAMIS (designed by CSO for SCHWIND eye-tech-solutions GmbH) and (2) anterior segment optical coherence tomography (AS-OCT) MS-39 (CSO). Intraocular HOAs were calculated by subtracting the total corneal aberrations measured by the AS-OCT from the total ocular aberrations measured by the pyramidal aberrometer. Aberrations were reported at an optical zone of 5.5 mm and were referenced to the pupil center.

Results: Every component of absolute intraocular HOAs was larger in keratoconic eyes than in myopic eyes in regard to magnitude and vertical and horizontal components (P < .05). Intraocular HOAs were also more scattered in keratoconic eyes. For trefoil aberrations, the axes of total corneal and intraocular centroid were opposing, almost 60 degrees away in keratoconic eyes. This mirrored the difference in orientation between anterior and posterior corneal trefoil. For each of coma and trefoil, 15.9% of keratoconic and 0% of myopic eyes had intraocular aberrations of 0.50 diopters (D) or greater. A total of 3.4% of keratoconic eyes and 0% of myopic eyes had total intraocular HOAs of 1.00 D or greater (all P < .05).

Conclusions: Intraocular HOAs in keratoconus are larger than in normal eyes and can either offset or support their total corneal counterparts. Intraocular HOAs should be computed and taken into consideration when planning customized corrections in eyes with keratoconus. [J Refract Surg. 2025;41(3):e189-e198.].

Purpose: To evaluate and compare the effect of pupil size, spherical aberration (SA), decentration, and tilt on the optical performance of five different monofocal intraocular lenses (IOLs).

Methods: Four aspheric IOLs (Vivinex, Hoya; SN60WF, Alcon Laboratories, Inc; ZCB00, Johnson & Johnson Surgical Vision; Akreos Adapt AO, Bausch & Lomb) with different SA values and one spherical IOL (SN60AT, Alcon Laboratories, Inc) were tested using an OptiSpheric IOL PRO 2 optical bench. The IOL diopter was measured at different apertures. The optical quality of the IOLs was evaluated using the modulation transfer function (MTF), through-focus MTF, and images of the United States Air Force Target test with different apertures and corneal SA values. The IOLs were also measured while they were decentered and tilted.

Results: The smaller the aperture, the wider the through-focus MTF of all tested IOLs, and the best image quality was observed with a 3-mm aperture. The diopter of aspheric IOLs with negative SA were negatively correlated with aperture size, but the result was opposite for spherical IOLs. The IOL power difference was more than 0.50 diopters between different apertures. All aspheric IOLs had similar MTF curves with a 3-mm aperture. Aspheric IOLs had the best visual quality with the same or similar corneal SAs. Most of the tested IOLs had lower tolerances in tilt than in decentration.

Conclusions: Pupil size, SA value, the degree of decentration, and tilt had different effects on the optical performance of aspheric IOLs. The results can provide guidance for surgeons in clinical practice. [J Refract Surg. 2025;41(3):e220-e230.].

Purpose: To evaluate the efficacy, predictability, and safety of laser in situ keratomileusis (LASIK) for the correction of myopia and myopic astigmatism using the SCHWIND Amaris 1050RS excimer laser (SCHWIND eye-tech-solutions).

Methods: This was a prospective cohort study of 1,420 eyes with myopia or myopic astigmatism that underwent LASIK in a single tertiary center between January 2017 and April 2022. The IntraLase iFS femtosecond laser (Johnson & Johnson Vision) was used for flap creation and excimer laser ablation was performed using the SCHWIND Amaris 1050RS with aspheric aberration-free ablation profile, asymmetric pupillary offset, and seven-dimensional eye tracking. Preoperative mean spherical equivalent was -5.81 ± 2.41 diopters (D) (range: -0.63 to -13.63 D).

Results: At 3 months, 95.1% and 99.6% of eyes achieved uncorrected distance visual acuity of 20/20 and 20/40 or better, respectively. The mean efficacy index was 0.99 ± 0.06 and the mean safety index was 1.0 ± 0.02. For predictability, 97.7% and 99.7% of eyes were within ±0.50 and ±1.00 D of the intended correction, respectively. The mean postoperative spherical equivalent was +0.02 ± 0.21 D, mean cylinder was 0.13 ± 0.25 D, mean vector-analyzed magnitude of cylindrical error was 0.09 ± 0.22 D, mean cylindrical angle of error was 0.64 ± 6.79 degrees, and mean cylindrical vector difference was 0.15 ± 0.26 D. None had significant surgical complications.

Conclusions: This study demonstrates excellent efficacy, predictability, and safety of the SCHWIND Amaris 1050RS excimer laser for the correction of myopic astigmatism. The seven-dimensional eye tracker was effective in compensating for eye movements intraoperatively. [J Refract Surg. 2025;41(3):e231-e237.].