Moderator
Abhay R. Vasavada, MS, FRCS
Panelists
James T Murphy III, MD, ABO
Maria S. Romero, MD, ABO
Viewing Papers
Expand a paper title to the right to view the paper abstract and authors. Use the video link to jump to that poster in the session.
Presenting Author
Mostafa Khalil, MBBS
Co-Authors
Amber Blood BA, MBBS, Ahmed Hassane MBBS
Presenting Author
Paul M. Mann II, MD
Co-Authors
Phillip Brunson OD
Purpose
To determine patient satisfaction with vision after Phorcides Planned Contoura topography-guided LASIK.
Methods
This study was a single site, single-arm, prospective, observational study of subject satisfaction after successful bilateral LASIK surgery. A total of 104 eyes of 52 subjects at one site were targeted. Subjects were assessed 3 months post-operatively. Clinical evaluations included measurement of visual acuity, manifest refraction, topography, aberrometry, and administration of the ocular surface disease index (OSDI) and the patient reported outcomes with LASIK (PROWL) questionnaires.
Results
A total of 50 subjects (100 eyes) completed the study. The mean satisfaction score at 3 months postoperative was 53.2 higher than the mean preoperative score (p <.0001). Mean 3 scores were for double image, glare, halo, and starburst were 7.4, 13.7, 18.0, and 29.2 higher, respectively, postoperatively compared to preoperatively. All differences were significant (p <.0001) with the exception of double image. Mean score on the OSDI questionnaire was 27.0 � 22.4 preoperatively, compared to 7.5 � 6.7 at 3 months postoperatively (p <.0001).
Conclusion
The results of this study suggest improved patient reported satisfaction, reduced dry eye symptoms, and overall decrease in visual disturbances following Phorcides planned Contoura LASIK.
Presenting Author
Amin Masjedi, MD
Purpose
To evaluate the agreement of placido-scheimpflug tomographer and swept-source anterior segment optical coherence tomography (AS-OCT), namely, SIRIUS and CASIA-2 OCT, in measuring the crystalline lens rise, anterior chamber depth (ACD) and central corneal thickness (CCT) in myopic phakic eyes who are candidates for implantable collamer lens (ICL).
Methods
A total of 49 right eyes from 49 myopes were investigated in this cross-sectional study. The anterior chamber depth (ACD), central corneal thickness (CCT) and crystalline lens rise (CLR) in all subjects were measured during one session by the same operator. The agreement between the two systems was evaluated using the Bland-Altman plots and 95% limits of agreement (LoA).
Results
Data were collected for 49 eyes (49 right eyes) of 49 patients (mean age 29.94 years � 7.2 [SD]) and mean refractive error -8.83�4.17[SD]. The mean values measured by SIRIUS and CASIA-2, respectively, were as follows: CLR, 88 � 180 micron and 160 � 195 micron; ACD, 3.31 � 0.31 mm and 3.29 � 0.32 mm; CCT, 496 � 40 micron and 492 � 37 micron. Area under the curve for CLR was 80% (AUC=80%, 95% CI: 65-95%). According to Youden�s index, Sirius lens rise greater than 85 can be assumed as Tomey CLR greater than 300 with sensitivity=90% [95% CI:60-98%] and specifity=64% [95% CI:48-77%].
Conclusion
Interdevice agreement analysis suggests that interchangeable measurements between two devices can be achieved for ACD and CCT, but not for CLR.
Presenting Author
Zhanyu Su, MD
Co-Authors
Zhaoxiang Wang MD, Kangjun Li MD
Purpose
This study explores the large language model (LLM) ChatGPT, particularly focusing on their image analysis and long-term memory functions, to predict the ICL vault following ICL-V4c surgery in patients with low anterior chamber depth (ACD) based on preoperative examination data.
Methods
In this retrospective case-control study, 160 eyes from patients with low ACD undergoing ICL-V4c surgery were analyzed. Preoperative OCT images and key metrics (age, ACD, ACA, WTW, CLR, and SE) were standardized into case reports to precisely predict vault outcomes and probabilities. Subjects were divided into training, validation, and test groups at an 8:1:1 ratio. Vault predictions using ChatGPT4 and 4o models were compared on the same test set, both before and after inputting the full training data into ChatGPT�s �memory� function. The study evaluated predictive accuracy, recall rate, weighted F1 score, and AUC for both models.
Results
Both ChatGPT4 and ChatGPT4o showed significant improvements in prediction accuracy of postoperative vault height using the �memory� function. Post-memory, both models achieved 75% prediction accuracy within �200 �m of the actual values, with ChatGPT4 achieving 50% of predictions within �100 �m, compared to 31.25% for ChatGPT4o. ChatGPT4o recorded an accuracy of 56.3%, recall of 42.9%, weighted F1 score of 0.46, and an AUC of 0.571 (P=0.634). In contrast, ChatGPT4 demonstrated an accuracy of 81.3%, recall of 57.1%, weighted F1 score of 0.73, and an AUC of 0.865 (P=0.015).
Conclusion
ChatGPT4 shows superior performance in predicting vault in low ACD patients, significantly outperforming GPT4o. LLMs hold significant potential for postoperative predictions. The continued refinement of LLMs, integrating long-term memory and data input for clinical applications, shows great promise.
Presenting Author
David Ballesteros, MD
Co-Authors
Erik Navas Villar MD, Karen Saldana MD, Nora Robledo OD, Ivonne Curiel Arce MD, Arturo Chayet MD
Purpose
To determine the accuracy of percentage of tissue altered (PTA) and ablation depth (AD) measurements using optical low-coherence reflectometry (OLCR) by comparing it to preoperative predicted and postoperative achieved values in myopic Femto-LASIK.
Methods
A prospective longitudinal study was conducted on 95 eyes of 48 consecutive patients who underwent femto-LASIK surgery for myopia from december 2023 to april 2024 at Codet Vision Institute (Tijuana, M�xico). Central corneal thickness (CCT) was measured with Pentacam before flap creation. Flap thickness (FT) was measured using OLCR, and AD was measured intraoperatively using OLCR and postoperatively using subtraction pachymetry with Pentacam data. The predicted PTA was calculated using the Munnerlyn formula and the wavenet planning station (EX500), while the achieved PTA was calculated by adding the measured FT and the achieved AD (OLCR, 1-week, and 3-month), divided by the preoperative CCT.
Results
The average patient age was 26.8 years, with a mean preoperative MSE of -4.49 D and a mean flap thickness of 91.58 �m. The mean ablation depths (AD) for the Munnerlyn formula, Wavenet Planning Station (EX500), OLCR, and Pentacam (1-week and 3-months) were 59.55, 66.78, 80.25, 73.16, and 70.05 �m, respectively. A mean difference of 13.47 �m was observed between the planned ablation depth (EX500) and the achieved depth measured using OLCR (P < .001). Similarly, the mean percentage of tissue altered (PTA) for these methods was 26.23%, 27.55%, 31.04%, 29.71%, and 29.11%, with a mean difference of 3.49% between the estimated (EX500) and achieved (OLCR) PTA (P < .001).
Conclusion
The achieved ablation depth, measured with OLCR, was 20% greater than the predicted value from the planning station (EX500), leading to a 13% increase in the PTA. Accurate measurement of flap thickness and ablation depth is essential for the precise calculation of PTA in LASIK surgical planning.
Presenting Author
David Sung Yong Kang, MD
Co-Authors
Kangyoon Kim MD, Byunghoon Chung MD, Samuel Arba Mosquera PhD, Tae-im Kim PhD, MD
Purpose
To evaluate the effect of using a novel optical zone (OZ) calculation formula incorporating topographic offset on postoperative corneal higher-order aberrations (HOAs) and visual outcomes after keratorefractive lenticule extraction (KLEx).
Methods
This study was a retrospective case series of 780 eyes that underwent KLEx (Visumax 800 (Carl Zeiss Meditec), ATOS (Schwind eye-tech-solution)) for myopic astigmatism. A novel formula for calculating attempted OZ for KLEx was derived from multilinear linear regression analyses of the postoperative effective OZ, spherical equivalent, and topographical offset. Correlations between OZ ratios (planned OZ / aimed effective OZ) and induced corneal HOAs were analyzed.
Results
The mean UCVA and BCVA were comparable for all planned OZs at day 1, 2weeks and 2 months postoperatively. The induction of total corneal HOAs, coma and spherical aberration was significantly lower in subgroup with high OZ ratios than in subgroup with low OZ ratios (P< />
Conclusion
KLEx performed using an OZ derived from the novel formula demostrated less induction of ccornel HOAs. Efforts to optimize OZ are crucial for achieving better surgical outcomes in KLEx.
Presenting Author
Maria Lee, BSc
Co-Authors
Yeabsira Mesfin BSc, Albert Xu PhD, Julie Schallhorn MD, Madeline Yung MD
Purpose
To perform a systematic review of the currently existing methods for astigmatism management in topography-guided ablations, and to perform a meta-analysis to determine which strategies result in the best patient outcomes.
Methods
Online databases were searched for experimental and observational studies reporting patient outcomes after topography-guided ablations (LASIK or PRK). Studies stating the method for determining target astigmatism and 3-month LASIK post-operative data on primary outcomes were included in a meta-analysis. Primary outcomes were: 1) percentage of eyes with post-operative residual spherical equivalent (SE) within 0.5D of target, and 2) percentage of eyes with post-operative best corrected distance visual acuity (BCDVA) of 20/20 or better. After a risk of bias assessment, high-risk studies were excluded from analyses. Descriptive statistics were collated and a network meta-analysis was performed.
Results
Of 693 studies identified, 53 studies describing 9267 eyes met inclusion criteria, including 20 randomized controlled trials. Common laser platforms were Alcon WaveLight EX500 with Contoura software and Nidek CXII with FinalFit software. Nomograms for Contoura included manifest refraction (MR), topography-modified refraction (TMR), Phorcides Analytic Engine (PAE), and Alcon-recommended. Preliminary analysis revealed the average percentages of eyes with residual SE within 0.5D as 86.3%, 83.2%, 97.8%, 65.1% and 92.9% for MR, TMR, PAE, Alcon-recommended and FinalFit, respectively. Average percentages of eyes with BCDVA of 20/20 or better were 75.7%, 83.9%, 95.3%, 94.0% and 85.0%, respectively.
Conclusion
MR, TMR, PAE, Alcon-recommended, and FinalFit nomograms provide excellent refractive outcomes. Data analysis was limited by the low number of comparative studies reporting primary outcomes. Additional high-quality comparative studies are required to determine the optimal algorithm for determining target astigmatism in topography-guided ablations.
Presenting Author
Abha D Shah, MS
Co-Authors
Rohit Shetty FRCS, Pooja Khamar MD, PhD
Purpose
Photorefractive keratectomy (PRK) procedures may result in corneal haze, significantly impacting postoperative vision. Ocular surface inflammation is recognized as one of the contributing factors to PRK haze. This study aims to evaluate pre-operative inflammatory biomarkers and their influence on post-PRK haze with and without treatment.
Methods
In this study, 1050 patients undergoing PRK surgery were enrolled. Tear samples were collected preoperatively, and inflammatory biomarkers were analyzed using the Bio-M-Pathfinder kit. Among them, 75 patients had elevated biomarkers, primarily TNF-alpha and/or combinations of IL-6, MMP-9, and IL-17A. These patients were randomly divided into two groups: Group 1 (n=37 eyes) received placebo treatment, while Group 2 (n=38 eyes) received targeted treatment for their specific elevated biomarkers. PRK was performed 1 month after biomarker-specific treatment, and both groups followed the same postoperative regimen. Postoperative haze incidence and grading were recorded for each group.
Results
In Group 1, which received only placebo treatment for raised inflammatory biomarker, 7 patients developed post PRK haze of grade 1 [raised TNF-a(1fold)], 4 patients had haze of grade 2 [raised TNF-a +MMP-9(2fold)] and 3 patients developed haze of grade 3 [raised TNF-a+IL-17a(3fold)] or above. In Group 2, which had received specific targeted therapy for respectively raised biomarkers, 1 patient had developed a haze of grade 2, and 1 patient had a grade 1 haze. Rest patients did not show any clinical science of haze. The incidence of haze between groups 1 and 2 was statistically significantly different: with group 2 being lower (p< />
Conclusion
Ocular Inflammation driving PRK haze is a known fact. Preoperative evaluation of the ocular surface becomes a quintessential part of the refractive evaluation in patients undergoing PRK. Also, pre-treating the ocular surface with specific biomarker-targeted therapy may significantly reduce the incidence of PRK haze and optimize the visual�outcomes.
Presenting Author
Pooja Khamar, MD, PhD
Co-Authors
Rohit Shetty FRCS, Abhijit Roy PhD, Francis Mathew PhD
Purpose
This study examines the impact of corneal biomechanics on intraoperative opaque bubble layer (OBL) formation, lenticule dissection ease, and postoperative visual outcomes in kerato-lenticule extraction (KLEX). By linking biomechanics with surgical parameters, the goal is to optimize intraoperative techniques and enhance postoperative recovery.
Methods
Here�s a more concise version of the methods: This prospective, observational cohort study included 300 eyes undergoing KLEX, performed by a single surgeon using consistent technology. Preoperative assessments utilized Pentacam AXL Wave for corneal topography, Corvis ST for biomechanics and elasticity (Young's modulus), and PSOCT for collagen imaging. Intraoperative OBL formation and lenticule dissection difficulty were recorded. Postoperative visual quality was evaluated with the Quality of Vision (QoV) questionnaire on day 1, at 1 week, and 1 month. Data were correlated with visual recovery and patient satisfaction.
Results
All patients achieved a visual acuity of 20/20 on postoperative day 1 (POD-1) according to Snellen�s chart. In 110 eyes with increased corneal elasticity (>100 N/m�) and thicker Bowman�s membrane (>17 ?m), intraoperative grade 2-3 OBL was observed, alongside tougher dissection (grade 3-4). These eyes exhibited a slower recovery in QoV scores. The remaining 190 eyes, which demonstrated normal corneal biomechanics, elasticity (<90 n/m�),="" collagen="" orientation,="" and="" bowman�s="" membrane="" thickness=""><15 ),="" had="" grade="" 1="" obl="" and="" easier="" dissection="" (grade="" 1),="" leading="" to="" faster="" qov="" recovery.="" despite="" these="" differences,="" both="" groups="" exhibited="" comparable="" qov="" scores="" by="" the="">
Conclusion
Corneal biomechanics, OBL formation, and intraoperative dissection are closely interrelated and significantly influence the speed of visual recovery. These findings suggest that future optimization of laser parameters, such as energy, spot size, and track distance, could enhance postoperative outcomes in lenticule extraction procedures.
Presenting Author
David Smadja, MD
Co-Authors
Sara Sella MD, Itay Lavy MD, Tomer Batash MD
Purpose
To validate the use of corneal epithelial refractive power map in the surgical planning for laser enhancement to avoid unexpected postoperative refractive surprise
Methods
A proprietary optical formula was developed by using the corneal epithelial index of refraction and a simplified approximation of the Munnerlyn equation to convert the epithelial thickness map into a vergence map. The formula was applied to the preoperative epithelial thickness maps (AS-OCT Optovue RTVue-100) of 12 patients that were candidates for laser vision retreatment due to unexpected surprise. The generated refractive power of the preoperative corneal epithelial pattern was used to adjust the refraction to be corrected after epithelial removal. Mean absolute differences and correlations with the postoperative refractive errors at 6 months were analyzed.
Results
The pre-retreatment spherical equivalent was +0.6 � 1.43 D (range: -1.5 to + 1.75D). At the 6-month follow-up visit, 83.3% (n = 10/12) of eyes treated were within �0.50 D and 91.7% (n= 11/12) of eyes were within � 1.00 D of intended correction. None of the eyes lost two or more lines of CDVA at the 6-month follow-up visit. Of the 10 eyes with plano target, 90% had an uncorrected distance visual acuity of 20/20 or better at 6 months.
Conclusion
The adjustment of the refractive surgical planning for the correction of unexpected refractive surprise post-PRK using the calculated preoperative epithelial refraction within the central 3mm enables high accuracy in visual and refractive outcomes.
To log in, click the teal "Login" button in the upper right-hand corner of this page. If you are logged in but still do not have access, please check your 2025 Annual Meeting registration.
