A pilot prospective study based on scanning OCT: the intraoperative vault model predicts postoperative vault

Abstract

Purpose

To evaluate the correlation between intraoperative vault measured by optical coherence tomography (OCT) and postoperative vault, and to assess dynamic changes in vault following EVO implantable collamer lens (ICL) implantation.

Setting

Eye and ENT Hospital of Fudan University.

Design

Prospective observational study.

Methods

This prospective study enrolled 14 patients (5 males, 9 females) who underwent EVO ICL implantation at the Eye and ENT Hospital of Fudan University between July and August 2024, involving a total of 27 eyes. The mean age of the patients was 25.5 ± 4.4 years (range: 19 to35 years), and the average manifest refraction spherical equivalent was − 7.84 ± 2.25 D (range: −11.63 to − 2.00 D). Intraoperative OCT was used to measure real-time vault height without the application of viscoelastic agents. Central vault was measured using CASIA-2 immediately after ICL implantation and reassessed at 1 day, 1 week, and 1 month postoperatively. At each follow-up, visual acuity, subjective and objective refraction, and anterior segment parameters were recorded.

Results

The mean intraoperative vault was 937.30 ± 402.27 μm. Postoperative vault measurements demonstrated a progressive decline: 912.00 ± 269.96 μm at 2 hours, 777.43 ± 252.19 μm at 1 day, 722.22 ± 226.37 μm at 1 week, and 675.98 ± 224.34 μm at 1 month. A strong correlation was observed between intraoperative and postoperative vault values, with an r² of 0.827 at 2 hours. Bland–Altman analysis indicated a systematic bias of 39.63 μm, with 92.6% of eyes falling within the 95% limits of agreement. The anterior chamber depth to both the crystalline lens and ICL remained stable throughout the postoperative period.

Conclusion

The vault measured by intraoperative OCT during EVO ICL implantation closely corresponded with postoperative measurements obtained via CASIA-2, and vault decreased in the early postoperative period.

Introduction

The implantable collamer lens (ICL) has been widely recognized as an effective and safe option for refractive surgery, providing stable visual and refractive outcomes with a low incidence of complications. However, one of the major challenges in ICL implantation is the accurate prediction of postoperative vault—a critical parameter for minimizing complications [1]. Excessively high vaults can lead to pupillary block, goniosynechia, and angle-closure glaucoma, whereas insufficient vaults may increase the risk of anterior subcapsular cataract and lens opacity [2, 3]. Improper vault remains the leading cause of ICL explantation, as both overly high and low vaults significantly elevate the likelihood of requiring ICL replacement [4]. A more comprehensive understanding of ciliary sulcus anatomy facilitates targeted ICL realignment in cases of abnormal vault, offering a potentially simpler and less invasive alternative to lens exchange.

The integration of intraoperative optical coherence tomography (OCT) systems into surgical microscopes offers surgeons a valuable tool to assess intraoperative parameters in real time and make informed clinical decisions during surgery. This technology has already been successfully applied in various intraocular procedures, including corneal transplantation, glaucoma, and retinal surgeries [5–7]. More recently, intraoperative OCT has been employed to assess real-time ICL vault, demonstrating its utility in predicting postoperative vault [8–10]. By identifying abnormal vaults intraoperatively, surgeons can perform immediate ICL repositioning or replacement, thereby reducing the need for secondary postoperative interventions [11]. In addition, intraoperative OCT monitoring of anterior segment parameters offers valuable insights into the dynamic changes of vault in the early postoperative period.

In this prospective study, we utilized intraoperative OCT to measure real-time vault and anterior segment parameters during ICL implantation. We further analyzed their correlation with postoperative measurements to investigate early postoperative vault changes and explore the utility of intraoperative OCT in guiding surgical decisions.

Methods

Patients

This study employed a randomized, single-masked design, involving 27 eyes from 14 consecutive patients who underwent bilateral ICL V4c implantation at the Eye and ENT Hospital of Fudan University between July and August 2024. Inclusion criteria required a minimum anterior chamber depth (ACD) of 2.8 mm and an endothelial cell density of at least 2500 cells/mm². Patients with any significant intraoperative or postoperative complications were excluded. Postoperative ocular evaluations included slit-lamp biomicroscopy, intraocular pressure measurements, subjective and objective refraction assessments, and both uncorrected and corrected distance visual acuity. Anterior segment and vault measurements were obtained using OCT at various time points: intraoperatively, immediately post-surgery, and at 1 day, 1 week, and 1 month postoperatively.

This study was approved by the Ethics Committee at the Eye and ENT Hospital of Fudan University, and all protocols complied with the Declaration of Helsinki. Written informed consent was obtained from all participants after a thorough explanation of the study’s potential benefits and risks.

Surgical techniques

All surgical procedures were performed by an experienced surgeon (XZ). The size of the ICL was determined based on the horizontal diameter of the cornea and the ACD. All eyes received Staar Surgical’s EVO ICL V4c model, with sizes customized to WTW measurements (range: 12.1–13.7 mm). A 3.0 mm clear corneal incision was created temporally, followed by the horizontal implantation of the ICL V4c into the posterior chamber. The surgical techniques employed were comprehensive and adhered to protocols previously described in earlier reports [12]. Postoperatively, each patient was prescribed a topical antibiotic (0.5% levofloxacin; Santen, Japan) and 1.0% prednisolone acetate (Pred Forte, USA), both administered four times daily for one week, along with pranoprofen (Senju, Japan) four times daily for two weeks.

Intraoperative vault measurement

Intraoperative vault measurements were obtained using a digital 4-dimensional (4D) and swept-source OCT (SS-OCT)-guided ophthalmic surgical microscope (TowardPi Medical Technology Co., Ltd, Beijing, China). Fourier-domain OCT systems measure the interference spectrum either spatially with a spectrometer (spectral-domain OCT) or temporally by rapidly sweeping the wavelength using a tunable laser source (SS-OCT). The integrated SS-OCT technology within the surgical microscope enables precise OCT imaging of both the anterior and posterior eye segments during surgery. The device’s technical specifications include a wavelength of 830 nm, an axial resolution of 5 μm (in air), and a scanning speed of 35,000 A-scans per second. Vault measurements were performed using an integrated caliper, with the built-in scale factory-calibrated for accuracy and routinely verified. These intraoperative measurements were recorded by the same surgeon (XZ) after the removal of the ophthalmic viscosurgical device, ensuring that the ACD was close to physiological conditions.

Postoperative vault measurement

Postoperative central vault measurements were obtained using the CASIA-2 OCT system, which utilizes a swept-source laser operating at a wavelength of 1310 nm. The CASIA-2 system captured 128 cross-sectional images in approximately 2.4 s. Vault measurements were conducted with the OCT scan beam positioned centrally, as this alignment minimizes both extrinsic and intrinsic distortions, ensuring that the aspect ratio does not influence the accuracy of the results. Figure 1 shows the OCT scan beams used for vault measurements in both the intraoperative OCT and CASIA-2 OCT systems. Anterior segment parameters recorded from each examination included anterior chamber depth to the crystalline lens (ACD-L), anterior chamber depth to the ICL (ACD-ICL), and the distance from the ICL to the crystalline lens (ICL-L). All measurements were performed by a skilled technician using a double-blind approach to ensure objectivity and consistency.

Fig. 1.

Illustration of OCT scanning systems for intraoperative and postoperative measurements

Statistical analyses

Statistical analyses were performed using IBM SPSS Statistics version 20.0 (IBM Corp., Armonk, USA). For quantitative variables, the data were expressed as mean ± standard deviation. The Kolmogorov–Smirnov test was used to assess the normality of the distribution. Paired t-tests were conducted to compare means of related samples, with a threshold of P < 0.05 indicating statistical significance. Correlations between variables were examined through simple linear regression analysis. To assess the agreement between intraoperative and postoperative vault measurements, Bland-Altman analysis was applied to evaluate the consistency of the measurements.

Results

Patients characteristics

A total of 27 eyes from 14 patients (mean age: 25.5 ± 4.4 years; range: 19–35 years) were included in the analysis. The cohort consisted of 5 males and 9 females, with a mean manifest refraction spherical equivalent (MRSE) of −7.84 ± 2.25 D (range: −11.63 to −2.00 D). Descriptive statistics for the patient sample are provided in Table 1. The ICL sizes used in this study ranged from 12.1 mm to 13.7 mm (mean: 12.9 ± 0.4 mm), with size distribution as follows: 12.1 mm (n = 3, 11.1%), 12.6 mm (n = 8, 29.6%), 13.2 mm (n = 12, 44.4%), and 13.7 mm (n = 4, 14.8%).

Table 1.

Baseline patient characteristics

Characteristic	Value
Patients (eyes, n)	27, 14
Age (years)	25.5 ± 4.4 (19 to 35)
Sex (Male/Female)	5 (35.7%)/9 (64.3%)
Refractive errors (D)
Spherical	−7.38 ± 2.16 (−11.00 to −1.75)
Cylindrical	−0.97 ± 0.50 (−1.50 to 0.00)
MRSE	−7.84 ± 2.25 (−11.63 to −2.00)
Corneal front Km (D)	43.11 ± 1.90 (40.2 to 46.6)
WTW (mm)	11.90 ± 0.47 (11.3 to 12.9)
ACD (mm)	3.15 ± 0.21 (2.83 to 3.54)
ICL size (mm)	12.90 ± 0.40 (12.10 to 13.70)

D diopters, MRSE manifest refractive spherical equivalents, Km maximum single point keratometry, WTW white-to-white distance, ACD anterior chamber depth, ICL implantable collamer lens

Data were presented as mean ± standard deviation (range or proportion)

Changes in anterior segment parameters and vault

The mean vault measured intraoperatively by OCT was 937.30 ± 402.27 μm. Postoperative vault measurements were as follows: 912.00 ± 269.96 μm at 2 hours, 777.43 ± 252.19 μm at 1 day, 722.22 ± 226.37 μm at 1 week, and 675.98 ± 224.34 μm at 1 month. The correlation coefficients between intraoperative and postoperative vaults were 0.827 (95% CI: 0.684–0.912), 0.735 (95% CI: 0.582–0.837), 0.811 (95% CI: 0.678–0.899), and 0.778 (95% CI: 0.635–0.876), respectively. When stratified by myopia severity (Table 2), 11 eyes (40.7%) were classified as low-to-moderate myopia and 16 eyes (59.3%) as high myopia. While high myopic eyes showed numerically higher vault values at all time points, no statistically significant differences were observed between the two groups (all P > 0.05). The vault reduction from intraoperative to 1-month follow-up was similar between groups (−230.91 ± 178.56 μm vs. −283.75 ± 201.34 μm, P = 0.451). Table 3 provides details of vault and anterior segment parameter changes over time. As illustrated in Fig. 2, ACD-L showed an increase at 2 hours postoperatively compared with intraoperative values and remained stable through 1 month. ACD-ICL increased at 2 hours post-surgery and continued to rise until 1 month. ICL-L remained similar to intraoperative values at 2 hours post-surgery but exhibited a sharp decline by 1 day, followed by a slight downward trend until 1 month.

Table 2.

Vault measurements stratified by myopia severity

Group	Eyes	MRSE Range (D)	Mean MRSE (D)	Vault (µm)					Vault Change* (µm)
				Intraoperative	2-Hour	1-Day	1-Week	1-Month
Low-to-moderate Myopia (< −6.00 D)	11 (40.7%)	−2.00 to −5.75	−4.18 ± 1.35	856.36 ± 348.25	834.55 ± 231.67	715.45 ± 209.34	678.18 ± 198.45	625.45 ± 186.23	−230.91 ± 178.56
High Myopia (≥−6.00 D)	16 (59.3%)	−6.25 to −11.63	−9.67 ± 1.89	992.19 ± 427.83	962.50 ± 287.94	817.81 ± 273.56	752.50 ± 243.78	708.44 ± 246.89	−283.75 ± 201.34
P value	-	-	< 0.001	0.387	0.223	0.298	0.412	0.362	0.451

*Vault change calculated as 1-month vault minus intraoperative vault. MRSE manifest refractive spherical equivalent, D diopters. Data presented as mean ± standard deviation or n (%). P values calculated using independent t-test for continuous variables and chi-square test for categorical variables

P<0.05 was marked as bold

Table 3.

Changes in anterior segment parameters and vault

Variables	PD (mm)	ACD-L (µm)	ACD-ICL (µm)	ICL-L (µm)
Intraoperative	6.46 ± 1.35	2923.47 ± 298.56	1687.23 ± 431.84	937.30 ± 402.27
2-Hour	6.23 ± 0.89	3118.92 ± 172.34	1943.18 ± 327.65	912.00 ± 269.96
1-Day	3.91 ± 0.76	3107.84 ± 241.73	2008.47 ± 293.12	777.43 ± 252.19
1-Week	5.91 ± 0.78	3102.56 ± 219.83	2087.91 ± 298.47	722.22 ± 226.37
-1Month	5.89 ± 0.69	3101.29 ± 227.94	2126.75 ± 282.18	675.98 ± 224.34

PD pupil diameter, ACD-L anterior chamber depth to the crystalline lens, ACD-ICL anterior chamber depth to the ICL, ICL-L distance from the ICL to the crystalline lens

Fig. 2.

Changes in anterior segment parameters and vault

Correlation between anterior segment parameters and vault

Correlation analysis revealed significant positive correlations between intraoperative and postoperative vault values at all measured time points: 2 hours, 1 day, 1 week, and 1 month (Fig. 3). The r² value for the correlation between intraoperative and 2-hour postoperative vault was 0.684, indicating that the vault measured at 2 hours postoperatively could be predicted from the intraoperative vault. Additionally, Bland-Altman analysis demonstrated a systematic deviation of 39.63 μm between intraoperative and 2-hour postoperative vaults (Fig. 4), with 25 out of 27 eyes (92.6%) falling within the 95% limits of agreement.

Fig. 3.

Correlation of intraopertive vault and postoperative vault. The intraoperative vault was positively correlated with the vault of 2 hours (A), 1 day (B), 1 week (C) and 1 month (D) after ICL implantation

Fig. 4.

Bland-Altman plot comparing the level of agreement between intraoperative vault and immediate postoperative vault. The vertical axis represents the difference between these measurements and the horizontal axis shows the corresponding mean value. The 95% limit of agreement values is indicated using dashed lines, and the middle bold line represents the mean difference between these measurements

Discussion

This study represents the first prospective investigation employing intraoperative OCT to predict postoperative vault, signifying a meaningful advancement in ICL implantation surgery. The novel application of real-time intraoperative OCT demonstrated a strong correlation between intraoperative and postoperative vault measurements across multiple time points, providing valuable insights into the dynamic changes of vault in the early postoperative period.

Specifically, a robust correlation was observed between intraoperative vault and postoperative vault at 2 hours after surgery. This suggests that intraoperative OCT serves as a reliable tool for predicting postoperative vault. Consistent with findings from the Zaldivar study [8], which reported a high correlation at 4 h post-surgery (r = 0.850) and a slightly lower correlation at 24 h (r = 0.745), our results further support the effectiveness of intraoperative OCT in predicting early postoperative vault changes, especially within the first few hours following ICL implantation.

Postoperative vault changes, particularly during the early phase, were more pronounced. This study identified a rapid decrease in vault within 1 day after surgery, closely associated with the posterior shift of the ICL. Several factors may contribute to this phenomenon, though some mechanisms require further investigation. Based on our observed data, the primary evidence-supported factor is the compression effect of the iris, particularly due to postoperative mydriasis, which directly correlates with the observed vault reduction pattern. This finding aligns with previous research demonstrating iris pressure effects on ICL positioning [13, 14]. Secondly, the gradual fixation and embedding of the ICL haptics in the ciliary sulcus lead to increased stability of the ICL, further influencing vault changes, especially from 1 week to 1 month post-surgery, as evidenced by the progressive ACD-ICL changes in our measurements. However, several proposed mechanisms remain hypothetical and warrant future investigation, including the precise role of aqueous humor dynamics in vault changes and the detailed process of sulcus adaptation over time. The immediate effects of anterior chamber filling post-surgery were minimal, resulting in only slight vault changes. Analysis of the ACD-L and ACD-ICL in this study further revealed that while the positions of the ICL and the lens were largely consistent immediately after surgery, the ICL exhibited a backward shift over time, while the lens position remained stable. This asynchronous movement suggests that the ICL retraction is the main cause of the vault decline, with the iris’ contribution to this process diminishing over time.

In this study, the “one-step non-viscoelastic technique” was employed to mitigate the influence of viscoelastic agents on intraoperative and postoperative vault measurements [15]. This approach proved effective in avoiding potential confounding factors, as supported by Torbey et al., who reported a significant decrease in vault after intraoperative removal of viscoelastic agents [16]. Furthermore, the postoperative pupil diameter observed in this study, averaging around 6.0 mm, closely matched the intraoperative pupil size. Previous research has demonstrated that both light- and drug-induced pupil constriction can cause the ICL to retreat, subsequently affecting the vault [13, 14]. By maintaining a consistent pupil size during and after surgery, this study minimized the potential impact of this factor. It is also important to highlight that different intraoperative and postoperative OCT devices were utilized across various studies, potentially leading to discrepancies such as image distortion. These differences may introduce subtle errors, warranting further exploration in future investigations to ensure the accuracy and comparability of vault measurements across different OCT systems.

On the first day after surgery, ACD-ICL increased while ACD-L remained unchanged, indicating a backward shift of the ICL, while the lens position remained relatively stable. This observation aligns with findings from our previous studies, where pilocarpine-induced pupil constriction resulted in iris pressure on the ICL, leading to its retraction [17]. In this study, the noticeable contraction of pupil on the first postoperative day, along with the similar anterior segment structural changes, suggests that the iris is likely the primary factor contributing to ICL retraction. From the first day to the first month, the vault continued to decline, ACD-ICL consistently increased, and ACD-L remained stable. These changes reinforce the conclusion that ICL retroposition is the primary reason for the decrease in vault. Although pupillary size showed minimal variation from one week to one month post-surgery, the trend of ICL retraction persisted. We postulate that this retrogression may be attributed to several factors. First, the gradual fixation and backward movement of the ICL ring could contribute to its repositioning. Second, the compression exerted by the ICL ring on the soft tissue of the ciliary sulcus may facilitate its adaptation and embedding over time. However, this hypothesis warrants confirmation through further imaging techniques, such as ultrasound biomicroscopy (UBM). Additionally, during the first week post-surgery, as the pupil gradually dilated, the iris compression effect from pupil contraction diminished, potentially making the phenomenon of ICL retraction more pronounced.

Intraoperative OCT not only enables early prediction of vault but also facilitates real-time intraoperative adjustments during ICL surgery. This capability empowers surgeons to modify ICL positioning intraoperatively based on real-time feedback, thereby reducing the likelihood of postoperative complications related to suboptimal vault height and enhancing overall surgical outcomes.

The innovation of this study lies in the introduction of a novel intraoperative OCT combined with the “one-step non-stick technique,” enabling accurate vault measurements. Additionally, we have delved deeper into the mechanisms underlying vault changes, particularly focusing on the impact of ICL retroposition during the early postoperative period. Furthermore, this study provides a comparative analysis of the positions of the ICL and lens, elucidating the changes that occur at various stages following surgery. This comprehensive approach not only enhances our understanding of vault dynamics but also offers practical insights for optimizing surgical techniques and improving patient outcomes.

In this prospective study, the vault and anterior segment parameters during ICL implantation were monitored in real time by intraoperative OCT, and the correlation analysis was conducted with the postoperative parameters, so as to explore the rule and influencing factors of the early vault change.

Several limitations should be acknowledged. First, the relatively small sample size (27 eyes) and short follow-up period (1 month) may limit the generalizability of our findings. Second, the long-term predictive accuracy of intraoperative OCT measurements requires validation in larger cohorts with extended follow-up durations. Future studies should aim to expand the sample size, incorporate longer-term follow-up, and include additional imaging techniques such as UBM to validate and further elucidate the mechanisms underlying vault changes.

Conclusion

This study employed intraoperative OCT to monitor vault and anterior segment parameters in real time during ICL implantation. Correlation analyses demonstrated that intraoperative vault measurements are reliable predictors of postoperative vault height. Furthermore, the dynamic changes in anterior segment structures observed postoperatively provide insights into the mechanisms underlying early vault fluctuations. These findings underscore the clinical value of intraoperative OCT in enhancing the precision of ICL implantation and optimizing surgical outcomes.

Authors’ contributions

Conceptualization: ZT and YY: investigation: all authors; formal analysis: ZT, JZ and YS; writing original draft: ZT and YY; review and editing: SL and XZ. funding acquisition: XZ: supervision: SL and XZ.

Funding

This work was supported by National Natural Science Foundation of China (Grant No. 82371096); Project of Shanghai Science and Technology (Grant No. 20410710100); Project of Shanghai Xuhui District Science and Technology (Grant No. XHLHGG202104); Shanghai Engineering Research Center of Laser and Autostereoscopic 3D for Vision Care (Grant No. 20DZ2255000). The funding agencies had no role in study design, data collection and analysis, interpretation of data, or writing the manuscript.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study adhered to the tenets of the Declaration of Helsinki. The protocols were approved by the Ethics Committee of the Eye and ENT Hospital of Fudan University. Written informed consent was obtained from all participants after a thorough explanation of the study’s potential benefits and risks.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.