Deviation between programmed and actual capsulotomy diameter in femtosecond laser–assisted cataract surgery: a clinical study of influencing factors and predictive modeling Authors: Zhao Tan, Fei Yuan Published: 2025 ABSTRACT AbstractBackgroundDifferences between actual and programmed capsulotomy diameter were found in femtosecond laser–assisted cataract surgery (FLACS). This study aimed to evaluate the deviation between the programmed (PDC) and actual (ADC) diameters of the capsulotomy in FLACS, identify the influencing factors, and establish a predictive model to improve the accuracy of capsulotomy size.MethodsThis retrospective study included 47 eyes from 47 adult patients who underwent FLACS. The ADC was measured using Image Pro Plus version 6.0 based on intraoperative video images, and the diameter deviation of the capsulotomy (DDC) was calculated by subtracting the PDC from the ADC. Correlation analysis was used to assess the relationships of demographics and ocular biometrics with ADC and DDC. Multiple linear regression was conducted to develop a model for predicting ADC.ResultsThe mean ADC (5.99 ± 0.34 mm) was significantly greater than the PDC (5.67 ± 0.15 mm, P < 0.001) in the overall cohort. The average DDC was 0.33 ± 0.31 mm, with 63.8% of eyes exhibiting a DDC greater than 0.20 mm. Correlation analysis showed that age was significantly negatively correlated with both ADC and DDC, and this correlation remained significant after adjusting for other ocular biometrics. Stepwise multiple linear regression identified age and PDC as independent predictors for ADC, yielding the following formula: ADC = 2.531 − 0.011·Age + 0.741·PDC (F = 14.265, P < 0.001; adjusted R2 = 0.366, Durbin-Watson statistic = 2.294).ConclusionAge was identified as the most significant factor influencing DDC. The proposed regression model for ADC may aid surgeons in preoperative planning for a more accurate and individualized capsulotomy diameter.Keywords: Femtosecond laser, Capsulotomy, Deviation diameter, Predictive model Background For individuals with cataracts, surgery remains the only definitive treatment. Continuous curvilinear capsulorhexis (CCC), introduced by Gimbel et al. [1] in 1990, is a critical step in cataract surgery. Studies have established that an irregular, decentered, or improperly sized capsulorhexis can induce IOL tilt, decentration, or longitudinal movement [2–4] during capsular fibrosis and shrinkage. These positional changes of the IOL may lead to refractive errors and increased higher-order aberrations [3–5], thus degrading visual quality—especially in eyes with premium IOLs [6, 7]. Therefore, to avoid these complications, the widely recognized ideal CCC is circular, well-centered, and slightly smaller than the intraocular lens (IOL) optic to ensure complete anterior capsule coverage. This configuration facilitates subsequent surgical steps, ensures IOL centration, and reduces lens epithelial cell proliferation and migration [8–10]. Nevertheless, achieving a perfect CCC is challenging, even for experienced surgeons. In 2009, the femtosecond laser technique was first introduced into cataract surgery [11]. Compared with manual capsulorhexis, femtosecond laser-assisted capsulotomy provides superior circularity and centration, as well as reduced IOL tilt and decentration [12–17]. However, studies have reported a discrepancy between the actual diameter of the capsulotomy (ADC) and the programmed diameter of the capsulotomy (PDC) [18]. This deviation can compromise anterior capsule overlap of the IOL optic, potentially affecting the effective lens position (ELP) and postoperative visual quality. Therefore, to help surgeons achieve the ideal capsulotomy diameter, this study aimed to identify factors influencing capsulotomy diameter deviation and to develop a predictive regression model for ADC. This model will aid in preoperative planning for a more accurate capsulotomy size. Methods This retrospective study enrolled patients who received femtosecond laser-assisted cataract surgery (FLACS) at the Department of Ophthalmology, Zhongshan Hospital, between March 2023 and October 2023. This study was approved by the Medical Ethics Committee of Zhongshan Hospital Affiliated to Fudan University (B2023-313R) and was conducted in accordance with the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants for the use of their clinical data. The inclusion criteria were as follow: (1) diagnosis of cataract; (2) availability of high-quality surgical images in which the anterior capsulotomy edge was clearly visible, intact, and non-distorted. The exclusion criteria were: (1) a history of intraocular surgery or ocular trauma; (2) intraoperative complications such as anterior capsule tear, posterior capsule rupture, or significant intraoperative miosis that impaired visualization of the anterior capsule opening and IOL optic. Patient data were retrospectively extracted from the hospital’s electronic medical record system. A meticulous review of medical records and surgical videos against these criteria identified 47 eyes from 47 patients for inclusion in the analysis. Surgical technique All FLACS procedures at our center followed the established standardized protocol detailed below. Patients received 0.1% pranoprofen (three times daily) and 0.5% levofloxacin (four times daily) for three consecutive days prior to surgery. Before the procedure, pupils were dilated with a 0.5% tropicamide and phenylephrine compound, and topical anesthesia was achieved using 0.5% proparacaine hydrochloride. An experienced surgeon performed FLACS procedures with VICTUS platform (VICTUS Laser System, Bausch & Lomb). A centered circular capsulotomy was created on the anterior capsule. The target capsulotomy diameter was programmed smaller than the IOL optic to ensure 360-degree overlap. Although 5.0–5.5 mm is generally recommended for IOLs with a 6.0 mm optic, 5.5–5.8 mm were selected for two reasons: to accommodate IOLs with a larger 6.25 mm optic (Rayner Superflex Aspheric 920 H), and to facilitate intraoperative maneuverability in eyes with harder nuclear cataracts [19]. All programmed diameters remained smaller than the corresponding IOL optics, reflecting a tailored, real-world surgical protocol. Lens fragmentation was performed using an eight-piece cross pattern. A 3.2-mm biplanar clear corneal main incision was created at 135°, followed by a 1.5-mm single-plane side-port incision at 15°. Following laser treatment, patients were transferred to a sterile operating room for phacoemulsification (Centurion, Alcon Laboratories). After nuclear removal and cortex aspiration, foldable IOLs were accurately implanted into the capsular bag. Detailed IOL specifications are listed in Table 1. All surgical procedures were video-recorded for subsequent analysis. Table 1Specifications of IOLs used in this studyIOLManufacturerMaterialDesignOptic/overall diameter (mm)ZXR00J&JHydrophobic Acrylic1-Piece/ C-Loop6.00/13.00ZA9003J&JHydrophobic Acrylic3-Piece/C-Loop6.00/13.00MA60ACAlconHydrophobic Acrylic3-Piece/C-Loop6.00/13.00MA60MAAlconHydrophobic Acrylic3-Piece/C-Loop6.00/13.00Superflex 920 HRaynerHydrophilic Acrylic1-Piece/closed C-Loop6.25/12.50PC 525 ErgomaXOphtecHydrophilic Acrylic1-Piece/Plate6.00/10.50Abbreviations: J&J Johnson & Johnson Vision Specifications of IOLs used in this study Abbreviations: J&J Johnson & Johnson Vision Clinical data collection Ocular biometric parameters including axial length (AL), flat keratometry (K1), steep keratometry (K2), white-to-white (WTW), anterior chamber depth (ACD), and lens thickness (LT) were obtained from the optical biometer (IOL Master 700, Carl Zeiss Meditec AG). The AL is defined as the distance from the anterior surface of the cornea to the retinal pigment epithelium. The WTW distance is the horizontal corneal diameter. K1 is defined as the curvature power of the flattest meridian of the anterior corneal surface, and K2 as that of the steepest meridian. The ACD is defined as the distance between the anterior surface of the cornea and the anterior surface of the lens. LT is the axial thickness of the crystalline lens from the anterior to the posterior lens capsule. Age and nuclear hardness classification were obtained from the electronic medical record system. Nuclear hardness of cataracts was graded according to the Lens Opacities Classification System III (LOCS III) [20]. Capsulotomy diameter measurement Surgical videos were reviewed to capture images under standardized conditions for ADC measurement. To eliminate potential mechanical effects of the ocular viscoelastic device (OVD) on capsulotomy size, images were taken after complete OVD aspiration. Furthermore, to minimize measurement bias from capsulotomy distortion, only images where the anterior capsule edge was clearly visible, intact, and non-distorted were selected for analysis. Eligible images were imported to an image analysis software (Image Pro Plus software, version 6.0) for ADC measurement, with the IOL optic diameter as a reference scale (Fig. 1). The mean value of three measurements from an experienced observer was used for analysis. Intra-observer reproducibility was quantified using the intraclass correlation coefficient (ICC) based on a two-way mixed-effects model for absolute agreement. The diameter deviation of the capsulotomy (DDC) was calculated by subtracting the PDC from the ADC. Measurement of the actual capsulotomy diameter. (a) The blue arrows indicate the edge of the anterior capsule opening, while the yellow arrows indicate the edge of the IOL optic. (b) The vertical yellow solid line represents the diameter of the IOL optic, and the horizontal red solid line represents the actual diameter of the capsulotomy. The IOL shown is Ophtec PC 525 ErgomaX, with a 6.0 mm optic diameter Statistical analysis Comparisons between the two PDC subgroups were conducted using independent-samples t-tests for normally distributed variables or the Mann–Whitney U test for non-normally distributed variables. Analysis of variance with Bonferroni post hoc testing was used for comparisons among three or more groups. The Wilcoxon signed-rank test was applied to compare PDC and ADC. Pearson or Spearman correlation tests were used to assess the relationships between ocular biometrics and ADC or DDC. Following this, partial correlation analyses were conducted to control for confounding variables. A stepwise multiple linear regression was conducted to develop a predictive model for ADC. The Shapiro-Wilk test was used to examine residual normality, and multicollinearity was assessed by variance inflation factors (VIF). Homoscedasticity was verified by plotting standardized residuals against predicted values, and the Durbin-Watson test was used to check for autocorrelation. The predictive performance and generalizability of the regression model were evaluated via leave-one-out cross-validation (LOOCV) in Python (version 3.13), quantified by the mean squared error (MSE) and the cross-validated R². Post hoc power analysis using G*Power software (version 3.1.9.6) was performed to determine whether the power of sample size was significant. Statistical analyses were performed using IBM SPSS Statistics version 26.0, and graphical visualization was conducted using GraphPad Prism version 9. A two-tailed P-value of < 0.05 was considered statistically significant. Results A total of 47 participants (47 eyes) aged between 32 and 92 years were enrolled. The demographics and ocular biometrics are summarized in Table 2. When classified by the PDC into 5.5 mm and 5.8 mm subgroups, no significant intergroup differences were observed in age, AL, K1, K2, LT, ACD, or WTW (all P > 0.05; Table 3). Table 2Study population characteristicsClassificationN = 47Sex (F/M)27/20Eye (R/L)23/24Age (y)72 (60,83)AL (mm)24.76 (23.78,25.84)K143.68 (42.44,44.18)K244.49 (43.54,45.15)ACD (mm)3.10 ± 0.48LT (mm)4.55 ± 0.51WTW (mm)11.69 ± 0.41LOCS IIINO12 (4.3%)NO24 (8.5%)NO311 (23.4%)NO415 (31.9%)NO510 (21.3%)NO65 (10.6%)Abbreviations: AL axial length, ACD anterior chamber depth, LT lens thickness, WTW white-to-white, NO nuclear opalescence Study population characteristics Abbreviations: AL axial length, ACD anterior chamber depth, LT lens thickness, WTW white-to-white, NO nuclear opalescence Table 3Characteristics of different PDC groupsClassification5.5 mm (N = 21)5.8 mm (N = 27)statistics P Age (y)70.48 ± 15.9867.04 ± 15.35t = 0.750.457ACD (mm)(mm)3.11 ± 0.433.09 ± 0.53t = 0.190.849LT (mm)4.65 ± 0.474.47 ± 0.53t = 1.240.222WTW (mm)(mm)11.73 ± 0.3911.65 ± 0.44t = 0.670.509AL (mm)25.99 ± 2.4324.55 ± 1.80Z = -1.940.053K142.83 ± 2.2143.50 ± 1.74Z = -1.120.261K243.86 ± 2.5544.53 ± 1.96Z = -1.130.257Abbreviations: ACD anterior chamber depth, LT lens thickness, WTW white-to-white, AL axial length Characteristics of different PDC groups Abbreviations: ACD anterior chamber depth, LT lens thickness, WTW white-to-white, AL axial length The intra-observer reproducibility for ADC measurements was excellent (ICC = 0.984, 95% CI: 0.974–0.990). ADC was significantly greater than PDC in the overall cohort (5.99 ± 0.34 mm vs. 5.67 ± 0.15 mm; Z = − 5.622, P < 0.001). This significant difference was also observed in both subgroups: in the 5.5 mm group, the mean ADC was 5.85 ± 0.32 mm (Z = − 3.945, P < 0.001), and in the 5.8 mm group, it was 6.11 ± 0.31 mm (Z = − 4.078, P < 0.001). In the overall cohort, the DDC averaged 0.33 ± 0.31 mm (range: − 0.25 to 1.22 mm). No statistically significant difference in DDC was found between the 5.5 mm and 5.8 mm PDC subgroups (0.35 ± 0.32 mm vs. 0.31 ± 0.31 mm; Z = − 0.343, P = 0.732). Notably, the majority of cases (63.8%) exhibited a DDC greater than 0.20 mm, while 29.8% had a minor deviation (0–0.20 mm), and only 6.4% showed a DDC ≤ 0 mm. Correlation analysis Simple bivariate correlation analyses showed that ADC was negatively correlated with age (rs = -0.460, P = 0.001) and LT (r = -0.402, P = 0.005). Similarly, DDC was negatively associated with age (rs = -0.416, P = 0.004) and LT (rs = -0.378, P = 0.009), but positively associated with AL (rs = 0.475, P = 0.001) and ACD (rs = 0.377, P = 0.009) (Table 4). Notably, partial correlation analyses identified age as the sole independent factor significantly associated with both ADC (r = -0.420, P = 0.003) and DDC (r = -0.362, P = 0.012) after adjusting for AL, ACD, and LT. Table 5 summarizes the mean DDC across different age subgroups. Table 4Correlation analysis of variables with ADC and DDCVariablesADCDDC r s P * r s P * AL (mm)0.2470.0940.4750.001K10.1540.3010.1200.421K20.1810.2240.1480.322ACD (mm)0.2770.059#0.3770.009LT (mm)-0.4020.005#-0.3780.009WTW (mm)-0.0100.949#0.0710.634Age (y)-0.4600.001-0.4160.004Abbreviations: ACD anterior chamber depth, AL axial length, DDC deviation of the diameter of the capsulotomy, LT lens thickness, PDC predicted diameter of capsulotomy, WTW white-to-white# Pearson correlation analysis, * Spearman correlation tests Correlation analysis of variables with ADC and DDC Abbreviations: ACD anterior chamber depth, AL axial length, DDC deviation of the diameter of the capsulotomy, LT lens thickness, PDC predicted diameter of capsulotomy, WTW white-to-white # Pearson correlation analysis, * Spearman correlation tests Table 5DDC in different age groupsAge (y) N DDC (mm)30–59100.58 ± 0.3960–69130.35 ± 0.3170–79110.25 ± 0.18≥ 80130.18 ± 0.21statisticsF = 4.324 P 0.009Abbreviations: DDC diameter deviation of the capsulotomy DDC in different age groups Abbreviations: DDC diameter deviation of the capsulotomy Multiple linear regression model A stepwise multiple linear regression model identified age and PDC as significant independent predictors of ADC, yielding the following equation: ADC = 2.531 − 0.011·Age + 0.741·PDC (F = 14.265, P < 0.001; adjusted R2 = 0.366, Durbin-Watson statistic = 2.294). The model demonstrated moderate goodness-of-fit, with a scatter plot revealing a moderate correlation between predicted and observed ADC values (R² = 0.393, P < 0.0001; Fig. 2A). All regression assumptions were met: standardized residuals were normally distributed (Shapiro-Wilk test, P = 0.109) and homoscedastic (Fig. 2B), with no evidence of multicollinearity (VIF = 1.012 for both predictors). The model’s robustness was further supported by minimal overfitting, which was demonstrated by the close agreement between unadjusted R² (0.393) and adjusted R² (0.366) and a cross-validated R² of 0.298 from LOOCV. Assessment of the multiple linear regression model. (A) Scatter plot of observed versus predicted ADC values. The solid line represents the best-fit linear regression (Observed ADC = 1.000 × Predicted ADC – 3.996 × 10⁻⁶; R² = 0.393, P < 0.0001). (B) Scatter plot of standardized residuals versus predicted ADC values. The absence of a discernible pattern supports the assumption of homoscedasticity. ADC actual diameter of the capsulotomy Power analysis A post hoc power analysis was conducted to quantify the statistical power of our regression model. With an effect size (f²) of 0.647, a sample size of 47, and two predictors, at α = 0.05, the achieved power (1-β) was 0.998. Discussion Capsulorhexis is critical in cataract surgery, as it influences the effective lens position (ELP) and the development of posterior capsular opacification (PCO) [10, 21, 22]. Previous studies have demonstrated that a 360° anterior capsule overlap of the IOL optic contributes to a lower incidence of PCO and improved IOL centration [23–25]. Li et al. [10] found a significant correlation between the degree of anterior capsule coverage and ELP. Hollick et al. [8] reported that large capsulorhexis (6.0–7.0 mm) without optic overlap resulted in a 66.2% incidence of PCO one year postoperatively, whereas smaller capsulorhexis (4.5–5.0 mm) with complete coverage of the IOL optic led to only 32.7%. PCO primarily develops from the proliferation, migration, and epithelial-mesenchymal transition (EMT) of residual lens epithelial cells [26]. An oversized capsulorhexis compromises anterior capsule-IOL adhesion, which in turn fails to create a mechanical barrier against cell migration, thereby promoting PCO [8, 27–29]. Furthermore, a capsulorhexis exceeding 6.5 mm may jeopardize the zonular attachment, threatening IOL stability. However, an excessively small capsulorhexis can trigger excessive lens epithelial cell proliferation and anterior capsule fibrosis [30], thereby increasing the risk of anterior capsule contraction syndrome (ACCS) [31], especially in patients with comorbidities such as retinitis pigmentosa [32], pseudoexfoliation [31, 33], uveitis [34], and diabetes mellitus [35]. ACCS typically results in significant visual impairment due to opacification along the visual axis. Furthermore, in the process of capsular bag fibrosis and shrinkage, asymmetric capsular contraction could exert asymmetric forces on the IOL, leading to decentration and tilt [36–38]. In eyes with pre-existing zonular weakness, this process can progress to the more serious complication—capsular bag-IOL complex dislocation [39, 40]. To mitigate this risk, Joo et al. [30] suggested a capsulorhexis diameter of 5.5–6.0 mm. As a general principle, a capsulorhexis diameter slightly smaller than the IOL optic is recommended to ensure sufficient overlap while minimizing complications. Accordingly, for IOLs with a 6.0 mm optic, a capsulorhexis diameter of 5.0–5.5 mm is advised to achieve the optimal 0.25–0.5 mm anterior capsule overlap [41]. In practice, the final size remains a clinical decision, tailored to individual patient factors and surgical expertise. Compared with manual capsulorhexis, the femtosecond laser technique significantly improves capsulotomy size accuracy [15, 42, 43], as evidenced by a minimal deviation (0.06 ± 0.03 mm) between the capsule disk diameter and PDC [43]. Nevertheless, the actual capsulotomy diameter was found to be significantly larger than the PDC in pediatric patients undergoing FLACS [18], with a maximum DDC of 1.99 mm. Luo et al. [44] also observed significantly greater ADC than PDC in patients under 20 years of age, with a DDC of 0.67 ± 0.04 mm. Although the numerical differences between ADC and PDC are small, they may affect IOL position and visual quality. Consequently, achieving a precise and predictable ADC is essential for optimizing visual outcomes. Our study found a significant difference between ADC and PDC, with a mean DDC of 0.33 ± 0.31 mm. This finding differs from that of Luo et al. [44], who reported a high level of agreement between ADC and PDC in patients older than 20 years. Akaishi et al. [45] also reported minimal diameter deviations in adults, with an ADC of 5.30 ± 0.16 mm when the PDC was 5.30 mm. The interstudy discrepancy may be attributed to differing preset PDC values: Luo et al. used 5.2 mm, and Akaishi et al. employed 5.3 mm, corresponding to the anterior capsule’s thickest midperipheral region [46], which typically measures around 5.25 mm in diameter [9]. Capsulotomy performed within this region may benefit from greater mechanical resistance and structural stability, consequently reducing ADC variation. This anatomical basis may explain the minimal deviation reported in their studies compared to our findings. Correlation analysis revealed that age was negatively associated with DDC, aligning with previous reports [18, 44]. This finding may be explained by age-related biomechanical changes in the anterior capsule, which consists primarily of hyaluronic acid and type IV collagen. With aging, the capsule thickens and loses elasticity [46], thereby limiting its distension after laser cutting and leading to a reduced DDC. Regarding ocular biometrics, Liao et al. [47] reported that in pediatric patients (2–6 years), AL was positively associated with the anterior capsulotomy enlargement index, whereas ACD was negatively associated. They proposed that a shallow anterior chamber might promote capsulotomy enlargement by increasing intracapsular pressure and capsule tension, and suggested that the effect of AL on the anterior capsulotomy diameter may be mediated by the laxity of the zonular fibers [47]. Additionally, Luo et al. [44] reported a negative correlation between LT and DDC. In our cohort, AL, ACD, and LT also showed significant correlations with DDC in bivariate analysis. However, after adjusting for age, these associations were no longer significant, indicating that age is the dominant factor influencing DDC. These results indicate that the effects of ocular biometrics are likely confounded by age, with their independent influence becoming insignificant once age is accounted for. To predict ADC, we developed a multiple linear regression model, which serves as a practical tool to guide preoperative PDC selection. The model demonstrated moderate predictive power (adjusted R² = 0.366), confirming its potential clinical utility. At the same time, the substantial unexplained variance indicates that ADC is influenced by additional factors beyond the current model. Most cases in our study demonstrated an enlargement of ADC relative to PDC. The mechanical interaction between the IOL and the capsular bag may be a potential factor. Since IOLs typically have a greater overall diameter than the crystalline lens, the haptics may exert a continuous expansile force on the capsular bag. This IOL-capsule interaction is complex and likely influenced by haptic design, overall IOL length, and material properties, meriting future investigation. While OVDs could theoretically cause transient capsular expansion, we consider its effect on the final, measured diameter to be minimal, as intraoperative observations confirmed that enlargement often visible prior to OVD injection and all ADC measurements were definitively taken after complete OVD aspiration. Notably, a small subset of cases exhibited a reduction in capsulotomy diameter. A plausible mechanism for this phenomenon is the intraoperative repositioning of the anterior capsule. Following lens extraction, the loss of lenticular support can cause the anterior capsule to shift posteriorly toward the equatorial plane, pulling the capsulotomy edge centripetally and thereby reducing the final ADC. Additionally, femtosecond laser parameters such as laser energy and pulse frequency may alter the microstructure and biomechanical properties of the anterior capsule, thereby influencing its stretchability and final measured diameter. However, these remain speculative considerations. Further studies are warranted to explore the influence of laser settings on capsulotomy size. This study has several limitations. First, the modest sample size (n = 47) for a multivariable regression may affect the model’s stability and limit its generalizability. Although the model achieved moderate predictive power (adjusted R² = 0.366), the substantial unexplained variance necessitates external validation in larger cohorts before any clinical application. Second, our ADC measurement protocol relied on the IOL optic as an internal reference scale. Consequently, any minor discrepancy between the nominal and actual IOL diameter could introduce a systematic bias in the absolute ADC values. Third, the inclusion of a variety of IOL designs, while reflecting real-world clinical practice, precludes definitive conclusions about the impact of specific IOL designs on capsulotomy diameter. Finally, the absence of postoperative follow-up data prevents any assessment of long-term capsulotomy dynamics, which are critical for understanding final IOL centration and stability. Future studies with larger sample sizes, homogeneous IOL cohorts, and longitudinal follow-up are needed to validate our model and define the personalized capsulotomy size for specific patient profiles and IOL types. Conclusion In conclusion, significant deviations exist between actual and programmed capsulotomy diameters in FLACS. Age was identified as the primary factor influencing DDC, likely due to age-related changes in anterior capsule elasticity. Our stepwise regression model, incorporating age and PDC, provides a practical tool for predicting ADC. This model may assist clinicians in achieving more precise capsulotomy size, thereby contributing to superior IOL position and visual outcomes. Acknowledgements Not applicable. Abbreviations Programmed diameter of the capsulotomy Actual diameter of the capsulotomy Diameter deviation of the capsulotomy Femtosecond laser–assisted cataract surgery Continuous curvilinear capsulorhexis Intraocular lens Axial length Flat keratometry Steep keratometry Lens thickness Lens Opacities Classification System III Anterior chamber depth White-to-white Nuclear opalescence Intraclass correlation coefficient Ocular viscoelastic device Effective lens position Posterior capsular opacification Anterior capsule contraction syndrome Mean squared error Leave-one-out cross-validation Author contributions Z.T. and F.Y. designed the study. Z.T. collected the data, performed the statistical analysis, and drafted the manuscript. F.Y. revised it for important intellectual content. All authors read and approved the final version of the manuscript. Funding No financial support or grants were received for this study. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Medical Ethics Committee of Zhongshan Hospital Affiliated to Fudan University (B2023-313R) and was conducted in accordance with the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants for the use of their clinical data. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.