Surgeon seniority was not a significant determinant of surgically induced astigmatism after phacoemulsification cataract surgery, and machine-learning models based on preoperative clinical data provided only limited predictive value, particularly for vector astigmatism outcomes.
Abstract
To evaluate whether surgeon seniority influences postoperative astigmatic change after phacoemulsification cataract surgery and to assess the performance of machine-learning models in predicting postoperative astigmatic outcomes using preoperative clinical data.
This study included 169 eyes of 169 patients who underwent phacoemulsification cataract surgery. Surgeons were categorized as residents (
n
= 101) or specialists (
n
= 68). Astigmatism was quantified using both classical magnitude-based analysis and vector decomposition. Potential determinants of postoperative astigmatic change were evaluated using multivariable regression analyses. Machine-learning models, including Gradient Boosting, K-Nearest Neighbors, Decision Tree, and Support Vector Machine models, were used to predict classical and vector astigmatism outcomes. Model performance was assessed using mean absolute error, root mean square error, and the coefficient of determination.
Of the 169 eyes, 101 were operated on by residents and 68 by specialists. Postoperative astigmatic change did not differ significantly between resident- and specialist-performed procedures in either classical or vector analyses. Within the resident group, increasing surgical seniority was not associated with postoperative astigmatic change. In multivariable analysis, preoperative cylindrical error was the only significant predictor in the vector model, whereas no significant predictor was identified in the classical model. Overall, machine-learning models demonstrated limited predictive performance, with only modest utility for classical astigmatism and poor explanatory performance for vector outcomes.
Surgeon seniority was not a significant determinant of surgically induced astigmatism after phacoemulsification cataract surgery. Machine-learning models based on preoperative clinical data provided only limited predictive value, particularly for vector astigmatism outcomes. Incorporating intraoperative variables may improve future predictive performance.
Abstract Purpose This study compared traditional statistical models with machine learning algorithms for predicting surgically induced astigmatism after cataract surgery, aiming to identify the most accurate and generalizable method among linear regression, regression trees, random forests, and neural networks. Methods Retrospective analysis was performed on 321 eyes (321 patients) undergoing phacoemulsification at a tertiary center. Data were standardized and surgically induced astigmatism was modeled using three vector-based outcomes: KEQ.post (keratometric equivalent power), KAST0.post (horizontal astigmatism component), and KAST45.post (oblique component). Preoperative variables served as predictors. Data were split into training (60%) and test (40%) sets. Four predictive models were developed and evaluated on unseen data. Performance was assessed using mean squared prediction error, and variable importance analyses identified key predictors. Results Linear regression achieved the best out-of-sample performance across all outcomes (e.g. KEQ.post mean squared prediction error = 0.043). Tree-based models performed slightly worse, while neural networks showed substantial overfitting with markedly higher test errors. Preoperative astigmatism and corneal radii were the strongest predictors. Conclusion KAST0 and KAST45 were not predictably modeled, whereas changes in KEQ.post were. Multivariable linear regression provided the most accurate and reliable predictions, while more complex machine-learning models (especially neural networks) overfit the limited dataset and offered no clinical advantage. The relationship between preoperative biometrics and postoperative equivalent power appears predominantly linear, though larger datasets may enhance machine-learning performance.
Amanda Pan, K. P. Kaiser, Stefan Raidl et al.· Current Eye Research· 0 citations
OBJECTIVE
To estimate surgically induced astigmatism (SIA) and the flattening effect produced by an on-axis incision in patients undergoing phacoemulsification cataract surgery.
STUDY DESIGN
A quasi-experimental study. Place and Duration of the Study: Department of Eye, Combined Military Hospital, Lahore, Pakistan, from May to September 2024.
METHODOLOGY
A total of 142 eyes underwent phacoemulsification with a 2.75-mm limbal incision placed on the steep axis. Keratometric (K) readings, including flat K (K1) and its axis, steep K (K2) and its axis, and astigmatism and its axis, were measured by the same technician before surgery and again 6 weeks postoperatively using an optical biometer (Tomey AL 100). Uncorrected visual acuity (UCVA) and corrected distance visual acuity (CDVA) were recorded for all patients preoperatively and postoperatively. SIA was calculated using the ASSORT online calculator, and changes in astigmatism, UCVA, and CDVA were analysed using the Wilcoxon signed-rank test (SPSS version 26), with statistical significance set at p < 0.05.
RESULTS
The mean age of participants was 62.9 ± 6.9 years, with 60.6% males. Preoperative astigmatism averaged 0.88 ± 0.52 dioptres (D) and decreased to 0.65 ± 0.48 D postoperatively, representing a mean reduction of 0.23 ± 0.30 D (p < 0.001). The mean SIA was 0.58 ± 0.50 D, and target-induced astigmatism was 0.45 ± 0.45 D. Both UCVA and CDVA improved significantly (p <0.001). Nearly 70% of eyes gained three or more Snellen lines of corrected vision.
CONCLUSION
A 2.75-mm on-axis limbal incision during phacoemulsification produces low and predictable SIA, a statistically significant reduction in corneal astigmatism and substantial improvement in visual acuity. This approach provides a practical, cost-neutral option for astigmatism management in cataract surgery.
KEY WORDS
Phacoemulsification, Astigmatism, Cataract extraction, Cornea/surgery, Visual acuity.
Sadia Humayun, Amna Tariq, Samar Fatima et al.· Journal of the College of Ph...· 0 citations
Background: Cataract remains a leading cause of visual impairment worldwide, with surgical intervention being
the definitive treatment. Phacoemulsification and small incision cataract surgery (SICS) are commonly performed
techniques, yet comparative postoperative outcomes require further evaluation.
Aim: To compare postoperative visual outcomes and complication rates following phacoemulsification and SICS
in patients with senile cataract.
Methodology: A hospital-based, observational study was conducted on 200 patients (100 per group) at
Department of Ophthalmology, Bhagwan Mahaveer Institute of Medical Sciences,Pawapuri, Bihar, India. Patients
underwent either phacoemulsification or SICS and were followed for 28 days. Postoperative uncorrected visual
acuity (UCVA), best-corrected visual acuity (BCVA), and intraoperative and postoperative complications were
recorded. Statistical analysis was performed using independent t-tests and chi-square tests.
Results: Mean UCVA at 28 days was 0.52 ± 0.48 logMAR (~6/18) in the phaco group and 0.61 ± 0.55 logMAR
(~6/24) in the SICS group (p = 0.184). Normal vision (≥6/18) was achieved in 58% of phaco patients and 50% of
SICS patients (p = 0.256). Intraoperative and postoperative complication rates were comparable between groups,
with no statistically significant differences.
Conclusion: Both phacoemulsification and SICS provide favorable postoperative visual outcomes with similar
safety profiles. Phacoemulsification shows a modest, non-significant advantage in early visual recovery, but both
techniques remain effective and safe.
Maninee Suman, Pradeep Kumar, Nandani Priyadarshini· International Journal of Pha...· 0 citations
Background: Phacoemulsification is the standard therapy for cataracts; however, postoperative changes in intraocular pressure (IOP) may lead to complications. This study observed changes in IOP on the first and sixth postoperative days and the associated risk factors. Objective: To analyze changes in IOP after phacoemulsification and determine the frequency of related risk factors. Methods: A prospective cohort study was conducted on 31 eyes from 27 patients who underwent uncomplicated phacoemulsification at H. Abdul Manap Regional Hospital Jambi from July to September 2025. IOP was measured on postoperative day 1 and day 6. Data were analyzed using the Wilcoxon Signed Rank test with a significance level of p < 0.05. Results: The majority of patients were male (68%) and aged 60–74 years (48%). Mean IOP decreased significantly from 17.16 mmHg on postoperative day 1 to 14.00 mmHg on day 6 (p = 0.001). Knowledge of wound care and medication adherence were considered good. Corneal edema was observed in 4 cases on day 1 and was not observed on day 6. Postoperative inflammatory symptoms also decreased. Conclusion: Phacoemulsification resulted in a significant reduction in IOP and demonstrated good postoperative outcomes.
Hanum Chaira Bahri Aeswarani, Vonna Riasari, Hasna Dewi et al.· International Journal of Hea...· 0 citations
OBJECTIVE
To compare early postoperative changes in central corneal sensitivity following conventional phacoemulsification and femtosecond laser-assisted cataract surgery (FLACS).
METHODS
This prospective two-arm cohort study included 74 patients (148 eyes) with age-related cataract who underwent either phacoemulsification (88 eyes) or FLACS (60 eyes) performed by a single surgeon between July 2024 and March 2025. Central corneal sensitivity was measured using a Cochet-Bonnet esthesiometer preoperatively and at 30 and 90 days. Outcome assessors were masked to the surgical technique. Propensity score matching and generalized estimating equations were applied to adjust for baseline differences and within-patient correlation. The primary endpoint was sensitivity at 90 days adjusted for baseline.
RESULTS
Baseline sensitivity was slightly higher in FLACS eyes (46.08 ± 8.87 mm) compared with phacoemulsification (42.63 ± 7.18 mm, P = 0.012). At 30 days, both groups demonstrated significant reductions from baseline (P < 0.001), with no significant intergroup difference. By 90 days, sensitivity in the phacoemulsification group recovered partially but remained below baseline (40.23 ± 7.80 mm, P = 0.001), whereas sensitivity in the FLACS group returned near baseline (45.25 ± 9.13 mm, P = 0.07). Adjusted analysis favored FLACS at 90 days (mean difference: 5.02 mm; 95% CI: 2.2-7.8; P < 0.001). In the FLACS group, younger age and higher baseline sensitivity were associated with better recovery.
CONCLUSION
Both techniques caused early postoperative hypoesthesia, but FLACS demonstrated faster and more complete recovery of corneal sensitivity by 90 days. This neurosensory advantage may be particularly relevant for patients predisposed to ocular surface dysfunction.
Isabella Passarelli Giabardo Marques, C. L. Budib, C. R. Hilgert et al.· Indian Journal of Ophthalmol...· 0 citations
Objective The study aimed to develop a predictive model for preoperative estimation of CI operative duration by identifying independent clinical and procedural factors that significantly influence surgical time. Design Single-centre retrospective observational study. Settings King Abdullah Ear Specialist Center (KAESC). Participants A total of 216 cochlear implant patients were enrolled between 2019 and 2025. Main outcome measures The primary outcome was surgical time (minutes), defined as the duration from skin incision to wound closure. Secondary measures included demographic, clinical, and intraoperative factors evaluated as predictors of surgical duration. Statistical analysis was performed using SPSS version 26 and R version 4.2.2. Univariate and multivariate generalized linear models were used to identify independent predictors of surgical time and develop a final predictive model. Results Surgical time was longer in females (180 vs. 150 min, p = 0.015), bilateral (190 min, p < 0.00001) and primary/revision cases (240 min, p = 0.003), and surgeries involving fellows or residents (p < 0.00001), whereas the round window approach (p = 0.036) and sealing the insertion route (p = 0.025) were associated with shorter times. In multivariate analysis, male gender was associated with a reduction in time by 16.6 min with CI 95% (−0.36, −32.77) (p = 0.0464), unilateral implantation by 62 min (p < 0.00001), and the round window technique by 28.5 min (p = 0.0445), while older age was associated with an increase in time by 0.13 min (p = 0.0034). A Consultant/Fellow/Resident team added 43 min (p = 0.0090). Conclusion The developed predictive model is a useful instrument for improving operating room management and customizing care plans in various CI practice environments.
M. Aljehani, F. Albassam, Fida Almuhawas et al.· Frontiers in Surgery· 0 citations