Idiopathic Macular Hole Area to Foveal Avascular Zone Ratio and Its Effects on Visual Acuity Before and After Surgery

Abstract

Purpose: This study investigates factors influencing visual acuity logarithmic minimum angle of resolution (logMAR) in idiopathic macular holes (IMH), with a focus on the foveal avascular zone (FAZ).

Methods: En face images from optical coherence tomography and optical coherence tomography angiography of 152 patients with stages 2, 3, and 4 IMH were analyzed. The minimum area (MA), base area, and FAZ of the macular hole were quantified, and the ratio of minimum area to FAZ (MFR) was calculated. In addition, central subfield thickness was extracted. Relationships between preoperative and postoperative visual acuity and these parameters, along with age, sex, axial length, and stage, were assessed using univariate and multivariate analyses.

Results: The study included 113 patients with high-quality images (113 eyes; median age, 69 years; interquartile range, 65–73 years). Multivariate analysis of factors significantly associated with pre- and postoperative visual acuity identified MFR as the only consistent independent factor (preoperative: β = 0.280, P < 0.05; postoperative: β = 0.357, P < 0.01).

Conclusions: The ratio of macular hole area to the FAZ may be a potentially important morphofunctional parameter influencing visual acuity outcomes in patients with IMH. These findings suggest that MFR could be useful in assessing surgical prognosis, although further research with larger, diverse cohorts is needed.

Translational Relevance: This study bridges the gap between basic retinal morphology and clinical outcomes by identifying MFR as a predictor of visual acuity in patients with IMH. Incorporating MFR into preoperative evaluations could improve surgical prognostication.

Introduction

Idiopathic macular hole (IMH) presents as a neurosensory retinal defect at the macular center, causing significant impairment to central vision.¹ Optical coherence tomography (OCT) technology has spurred extensive research into IMH's structural aspects² and postoperative functional predictors.^3–5 Surprisingly, limited attention has been given to the preoperative morphology and function of IMH.^6–11 The advent of OCT angiography (OCTA) has facilitated postoperative morphological evaluations, including the foveal avascular zone (FAZ), and explored its relationship with IMH.^12–14 Additionally, the correlation between FAZ and visual acuity in various conditions has underscored its morphofunctional significance.^15,16 Studies indicate that, in IMH-affected eyes, the FAZ is morphologically aligned with the holes, diminishing in size with the postoperative closure of these lesions.^12–14 However, the preoperative FAZ area does not correlate with preoperative visual acuity.^13,14 OCTA imaging, particularly high-density raster scans, offer detailed blood flow information and high-resolution OCT en face images. Shahlaee et al.¹⁷ used this technology to quantitatively assess vascular and cystic changes before and after macular hole surgery. Recent developments introduced two macros: the Kanno–Saitama macro (KSM) for FAZ extraction from OCTA images^18,19 and the KSM2 for hole extraction from OCT en face images.²⁰ These tools offer efficient measurements characterizing IMH and FAZ with high reproducibility. 

This study aimed to identify factors associated with visual acuity in the morphofunctional evaluation of IMH, emphasizing the pivotal role of FAZ. Through the utilization of KSM and KSM2 macros, we conducted a comprehensive assessment of IMH-affected eyes, focusing on the morphofunctional significance of FAZ. 

Subjects and Methods

Patients and Study Design

This retrospective study obtained approval from the Ethics Committee of Saitama Medical University Hospital, in accordance with the Declaration of Helsinki, and acquired patient consent (Institutional Review Board 19079.01). Data from 152 patients with Gass classification stages 2, 3, and 4 IMH who underwent Swept-source OCTA imaging at our institution between February 2018 and June 2024 using the PLEX Elite 9000 (Carl Zeiss Meditec, Inc., Dublin, CA) were retrospectively analyzed. Collected data included preoperative best-corrected visual acuity (BCVA), slit-lamp microscopy, noncontact tonometry (TONOREF II, Nidek, Gamagori, Japan), fundus photography (CX-1, Canon, Tokyo, Japan), ocular axial length measurement (Optical Biometer OA-2000, Tomei Corporation, Nagoya, Japan), and retinal thickness analysis via spectral domain OCT (SD-OCT) (Spectralis HRA 2, Heidelberg Engineering, Heidelberg, Germany). Exclusion criteria comprised patients with an axial length of 26 mm or greater, diabetic retinopathy and maculopathy, glaucoma, traumatic MH, retinal detachment, prior endophthalmitis surgery, or poor-quality en face images from OCTA imaging. Surgery was performed by several experienced surgeons. All patients except pseudophakia first underwent phacoemulsification and intraocular lens (IOL) implantation. Standard three-port, sutureless, pars plana vitrectomy and internal limiting membrane peeling were performed. In all cases, the macular holes were closed. Air or sulphur hexafluoride (SF₆) was used as an intravitreal tamponade. Postoperatively, patients were advised to posture face down by day from 3 to 7 days. Postoperative BCVA was measured within 1 month after surgery, after the intraocular gas had disappeared. 

OCTA Imaging and En Face Images

The PLEX Elite 9000 uses a swept laser source with a central wavelength of 1050 nm and a bandwidth of 100 nm, delivering an estimated axial resolution of 5 µm and a transverse resolution of approximately 14 µm.²¹ In this study, 3 × 3-mm macula-centered images were used. The OCTA images used to measure the FAZ were en face images of the retinal surface layer, extending from the inner limiting membrane to the inner nuclear layer. These images were constructed using the built-in segmentation software. En face OCT images of the minimum area (MA) and base area (BA) of the IMH were extracted using the slab custom function within the segmentation software to enhance the accuracy of hole area delineation. The slab segmentation specifications were “Offsets/Top: ILM –70 Bottom: RPEFit –70” for MA and “Offsets/Top: RPEFit –65 Bottom: RPEFit –35” for BA (Figs. 1A, 1B). The settings for this custom segmentation are quite simple. The segmentation of the MA involves shifting the default Retina Slab segmentation toward the vitreous side by 70 µm. The BA is defined using the RPEFit method for the top and bottom segmentation lines, with a distance of 30 µm set between them, and then shifted toward the vitreous side by 65 µm. By moving the segmentation toward the vitreous side, we decrease the influence of high-intensity signal reflections within the hole area in each en face image generation (Figs. 1C, 1D), thereby preventing misextraction. Images with OCT signal intensity of 8 out of 10 or higher were eligible for analysis. 

Figure 1.

 

Hole extraction procedure. The images in the left column represent the relationship between the minimum diameter (MD) and minimum area (MA), and those in the right column show the relationship between the base diameter and base area (BA). (A and B) Slab segmentation of MA and BA (white dotted line). (C and D) OCT en face images of each area. (E and F) Mask images of each area extracted by KSM2. OCT en face images (C, D) created by each area's custom segmentation (A, B) are output. When the output images (C, D) are imported into ImageJ and KSM2 is activated, mask images (E, F) of each extracted area are generated. The mask images for each area (E and F) represent the macular hole regions at the same height level as the MD (double-headed arrow in A) and the base diameter (double-headed arrow in B). When the bottom segmentation line is in contact with the retinal pigment epithelium, a high-intensity signal reflection is generated within the en face image's hole area (C and D), which can lead to erroneous extractions. The slab segmentation settings used in this study were adjusted to minimize this effect while obtaining measurement values at the same level as those measured by conventional methods.

Two Area Extraction Macros

The KSM is designed specifically for FAZ extraction, using binarization and dilation-erosion processes.^18,19 For this study, we used a modified version of KSM, incorporating enhancements in noise processing and adjusting the “Enlarge” setting to 4 pixels.¹⁹ This modified version extends the capabilities of the previously reported macro.¹⁸ OCTA images from the device were imported into ImageJ software (http://rsb.info.nih.gov/ij, accessed February 8, 2021), enabling efficient FAZ extraction within seconds of macro activation. 

The KSM2, in contrast, is designed for hole extraction, using binarization processes.²⁰ The OCT en face images representing the MA and BA, as previously described, underwent basic noise processing and binarization within ImageJ (Figs. 1C–F). Owing to differences in size (300 × 300 pixels) for the OCT en face image compared with OCTA images, the former was pre-enlarged to 1024 × 1024 pixels (without interpolation) to match the size of the OCTA image. The Phansalkar method,^22,23 designed for selecting darker regions, was used for binarization (Fig. 2). After image enlargement and macro activation, the hole area extraction process took only seconds (see Text document, Supplemental Digital Content 1, which contains the code for both macros, and Video, Supplemental Digital Content 2, illustrating a macro video). 

Figure 2.

 

Differences in extraction of the hole area using binarization methods. (A) Original image. (B) Otsu method of global thresholding. (C) Niblack method of local adaptive thresholding. (D) Phansalkar method of local adaptive thresholding. The Phansalkar method (D) is the best at extracting the darkest part of the hole.

It is advisable to adapt these macros as needed for different image capture models.^19,24 

Exploration of Visual Acuity–Associated Factors Using Regression Analysis

Preoperative and postoperative BCVA was converted into logarithmic minimum angle of resolution (logMAR) units for statistical purposes. Factors associated with preoperative and postoperative BCVA were explored using simple regression analysis. These encompassed the FAZ of the superficial retina and the MA and BA of the hole, which were extracted using the aforementioned macros. The three area values were calculated after applying magnification correction using the axial length.²⁵ From these area values, the area ratio of MA to FAZ (MFR) was calculated as an indicator of the relationship between MA and FAZ. In studies using conventional methods, the minimum diameter (MD) and base diameter (BD) are used as factors of interest. In this study, as a factor corresponding to the MD in the conventional method, the MA was adopted, and as a factor corresponding to the base diameter, the BA was adopted. Additionally, as a factor corresponding with the height in the conventional method, the central subfield thickness (CST) within the 1-mm circle of the Early Treatment Diabetic Retinopathy Study grid obtained from SD-OCT retinal thickness analysis was included in the analysis. Age, sex, axial length, and stage were added to these factors, resulting in the final factors considered: age, sex, axial length, stage, FAZ, MFR, MA, BA, and CST. Finally, factors that showed significant differences in simple regression analysis were examined as independent factors associated with pre- and postoperative BCVA using multiple regression analysis (see the Supplemental Digital Content 1, for detailed information on the morphological factors). 

Additional Image Evaluation

As an additional evaluation, the number and percentage of cases with epiretinal membrane (ERM) and epiretinal proliferation (EP) surrounding the hole were assessed, following previously reported methods,^26–28 using OCT en face images obtained from SD-OCT B-scan images and OCTA imaging. Observations included 9 mm × 9 mm horizontal and vertical B-scan images centered on the macula, along with raster scan data obtained from retinal thickness analysis. OCT en face images were acquired with a 6 mm × 6 mm field of view. The en face images were segmented using previously reported settings.²⁷ The presence or absence of ERM and EP was determined by two masked observers (J.K. and H. Ibuki). In cases of disagreement, a senior physician (H. Ishii) made the final decision. 

Statistical Analysis

Data normality was assessed via the Shapiro–Wilk test. Given the non-normal distribution of the data, nonparametric tests were selected. Continuous variables are presented as medians (interquartile range). Patient characteristics such as sex, pseudophakia, and the presence or absence of ERM or EP were evaluated using a binomial test. Stages were evaluated using a chi-square test. Pre- and postoperative BCVA was evaluated with the Wilcoxon signed rank test. The examination of independent factors associated with preoperative and postoperative BCVA used univariate and multivariate analyses. Statistical significance was set at a P value of less than 0.05. All statistical analyses were performed using R software (version 3.6.3; R Foundation for Statistical Computing, Vienna, Austria). 

Results

In this investigation of IMH, 113 eyes from 113 patients were included out of the initial 152 cases. Exclusions were attributed to poor en face image quality (15 cases), BA segmentation errors (7 cases), and axial lengths exceeding 26 mm (17 cases). Table 1 displays the patient characteristics. The median patient age was 69 years (quartile range, 65–73 years). No significant differences were observed in terms of sex or stage (P = 0.259 and P = 0.746, respectively). Pseudophakia was significantly less prevalent, occurring in 19 of 113 eyes (17%) (P < 0.001). ERM or EP was present in a total of 37 of 113 eyes (33%), significantly fewer than the others (P < 0.001). Cases with ERM also included some with concurrent EP. The median preoperative BCVA was 0.523 logMAR (Snellen 20/67) and the median postoperative BCVA was 0.301 logMAR (Snellen 20/40). Postoperative BCVA was significantly better than preoperative BCVA (P < 0.001). 

Table 1.

 

Patient Characteristics

 

Regression Analysis of Independent Factors Affecting Visual Acuity

Table 2 presents the results of univariate and multivariate analyses investigating the relationship between preoperative BCVA and factors of interest. Figure 3A shows scatter plots of factors significantly associated with preoperative BCVA in univariate regression analysis. The significant factors associated with preoperative BCVA were FAZ, MFR, MA, BA, and CST, which were also identified as the final independent factors. A strong positive correlation was observed between MFR and MA (rho = 0.936, P < 0.001, Spearman's rank correlation), which led to their analysis in separate models. As a result of multiple regression analysis, the independent factors associated with preoperative BCVA in model 1, which excluded MFR, were MA and CST (β = 0.363, P < 0.05; β = 0.227, P < 0.05, respectively). In model 2, which excluded MA, the independent factors were FAZ, MFR, and CST (β = 0.218, P < 0.05; β = 0.280, P < 0.05; β = 0.211, P < 0.05, respectively). 

Table 2.

 

Factors Associated With Preoperative BCVA: Univariate and Multivariate Analyses

 

Figure 3.

 

Scatter plots of factors significantly associated with preoperative and postoperative BCVA. (A) Scatter plot of factors significantly associated with preoperative BCVA. In the univariate regression analysis for preoperative BCVA, significant associations were found with FAZ, MFR, MA, BA, and CST (β = 0.354, P < 0.001; β = 0.451, P < 0.001; β = 0.482, P < 0.001; β = 0.458, P < 0.001; and β = 0.297, P < 0.01, respectively). (B) Scatter plot of factors significantly associated with postoperative BCVA. In the univariate regression analysis for postoperative BCVA, significant associations were observed with FAZ, MFR, MA, BA, CST, and axial length (β = 0.203, P < 0.05; β = 0.512, P < 0.001; β = 0.457, P < 0.001; β = 0.438, P < 0.001; β = 0.263, P < 0.01; and β = −0.193, P < 0.05, respectively).

Table 3 presents the results of univariate and multivariate analyses investigating the relationship between postoperative BCVA and factors of interest. Figure 3B shows scatter plots of factors significantly associated with postoperative BCVA in univariate regression analysis. The factors showing a significant association with postoperative BCVA were axial length, FAZ, MFR, MA, BA, and CST, which were also identified as the final independent factors. After multiple regression analysis, the only independent factor associated with postoperative BCVA in model 1 was axial length (β = –0.271, P < 0.01). In model 2, the independent factors were axial length and MFR (β = –0.255, P < 0.01; β = 0.357, P < 0.01, respectively). 

Table 3.

 

Factors Associated With Postoperative BCVA: Univariate and Multivariate Analyses

 

Preoperative Images of Representative Cases

Figure 4 shows images of representative cases illustrating the relationship between preoperative BCVA and each morphological factor. All images were cropped to a size of 1.5 × 1.5 mm, centered on the FAZ. In addition to the measurements of each morphological factor, we included the value obtained by subtracting the MA from the FAZ area (remaining FAZ area) to illustrate the relationship between preoperative BCVA and each morphological factor using representative case images (see the Supplemental Digital Content 1 for detailed information on the morphological factors). Cases 1 through 4 depict MAs arranged by size, correlating with changes in BCVA values (cases 1–4, Fig. 4B). Larger MA sizes were associated with lower acuity values. However, while comparing BA sizes, case 3 appeared larger than case 4, deviating from the MA ranking (case 3, Fig. 4C). The sequence of MFR and Remaining FAZ area sizes mirrored that of MA. In case 5, despite a larger MA than in case 2, BCVA values were superior in case 5. A comparison of their MFRs indicated that case 5 had a lower MFR than case 2. Case 6 was characterized by a larger remaining FAZ area than those in cases 1 and 2. However, BCVA values were lower than those in cases 1 and 2, correlating with higher MFRs in both cases. Thus, even in cases where the association between preoperative BCVA and MA appeared contradictory, using MFR helped to explain the reasoning. 

Figure 4.

 

Preoperative Images of representative cases. (A) FAZ. (B) MA. (C). BA. (D) Horizontal B-scan image. The white outlines in (A, B, and C) represent the FAZ. Case 1: Visual acuity is 0.6. MA area is 28 µm². BA area is 64 µm². Remaining FAZ area is 282 µm². MFR is 0.09. Case 2: Visual acuity is 0.4. MA area is 39 µm². BA area is 238 µm². Remaining FAZ area is 222 µm₂. MFR is 0.149. Case 3: Visual acuity is 0.15. MA area is 284 µm². BA area is 877 µm². Remaining FAZ area is 169 µm². MFR is 0.627. Case 4: Visual acuity is 0.09. MA area is 350 µm². BA area is 689 µm². Remaining FAZ area is 153 µm². MFR is 0.696. Case 5: Visual acuity is 0.5. MA area is 51 µm². BA area is 148 µm². Remaining FAZ area is 322 µm². MFR is 0.137. Case 6: Visual acuity is 0.3. MA area is 86 µm². BA area is 343 µm². Remaining FAZ area is 306 µm². MFR is 0.219.

Discussion

In this study, our focus was to identify factors associated with BCVA in the morphofunctional evaluation of IMH, specifically emphasizing the significance of FAZ. Notably, MFR consistently emerged as an independent factor associated with BCVA both preoperatively and postoperatively, underscoring its potential importance in visual outcomes. 

Table 4 summarizes findings from a previous study investigating the correlation between preoperative BCVA and morphometric factors. Previous reports highlighted significant correlations of BCVA primarily with MD and base diameter.^6–11 Careful interpretation is essential, given potential variations in measurement methods for these diameters and areas among authors. This study used the area value of each lesion, extracted from en face images, as an index. Notably, significant associations were observed for MA and BA, aligning with outcomes from previous studies. In addition, one out of three reported cases for retinal height showed significant correlation.^6,9,10 In this study, a significant association with CST was identified. Assuming that CST reflects retinal height to some extent, the discrepancy between studies is interesting. Based on the results of MA and BA in this study, CST seems to be a factor that reflects the height of the hole. Although two previous reports investigated the correlation between FAZ area and BCVA, neither found a significant correlation.^13,14 However, this study observed a significant association. This difference in results may likely be attributed to the disparity in sample sizes, with the prior studies comprising 28 and 51 cases, respectively, compared with 113 cases in the current investigation. Additionally, this study introduced an index to represent the relationship between FAZ and MA: the ratio of MA to FAZ (i.e., MFR). Interestingly, MFR consistently demonstrated a significant association with BCVA, both pre- and postoperatively, highlighting its potential role as an independent predictor of visual outcomes in patients undergoing treatment for IMH. 

Table 4.

 

Summary of Findings on Correlations Between Preoperative BCVA and Morphometric Factors From Previous Studies

 

In the preoperative regression analysis, an association between BCVA and age was anticipated, particularly considering that pseudophakia accounted for 17% of all patient characteristics. However, no significant association for age was observed (P = 0.513). In conjunction with these results, the OCT signal intensity in this study was 8 out of 10 or higher in all cases, suggesting that the impact of intermediate media opacity in this investigation appeared to be minimal. Furthermore, similar results were obtained in the postoperative analysis, where IOLs were inserted in all cases. This outcome seems to substantiate these findings. In an additional evaluation of cases with ERM or EP, ERM was observed in 16% and EP in 17% of cases. These proportions suggested that ERM was somewhat lower, although EP seemed to be comparable, compared with those reported in previous studies.^26–28 The OCT en face images used for observation used a customized segmentation developed by Ishida et al.²⁷ Images generated by their segmentation were considered highly suitable for observing preretinal abnormal tissue. Their segmentation was a simple modification of the existing settings. Many OCTA devices offer segmentation customization features, and once these settings are saved, they can be applied in the same manner as the default settings. With such a straightforward customization as theirs, validation is feasible for anyone. The settings developed in this study are similarly simple. There have been reports using default settings to measure hole diameter from OCT en face images.¹¹ Thus, it is possible to extract the hole area using default settings. However, segmentation in the macular region is prone to errors, particularly in the inner retinal layers.²⁹ Consequently, Philippakis et al.³⁰ selected the full-thickness retinal slab for MD measurement, as in the present study. Furthermore, in this study, a setting was adopted in which the segmentation was slightly shifted toward the vitreous side to improve extraction accuracy. 

In the preoperative multiple regression analysis, model 1 (excluding FMR) revealed significant associations between MA and CST (β = 0.363, P < 0.05; β = 0.227, P < 0.05, respectively). Model 2 (excluding MA) indicated significant associations with MFR, FAZ, and CST (β = 0.280, P < 0.05; β = 0.218, P < 0.05; β = 0.211, P < 0.05, respectively). These findings suggest that MA, MFR, FAZ, and CST are independent associated factors in relation to the analyzed outcomes The specific anatomical nature of the central fovea, lacking an inner retina and containing only cones,³¹ logically establishes that MA is significantly associated with BCVA. Additionally, a positive association was found between FAZ and preoperative BCVA. This finding can be interpreted as indicating that, as the FAZ becomes wider, BCVA decreases. There is a significant positive correlation between MA and FAZ (rho = 0.610, P < 0.001, Spearman's rank correlation), which is likely the reason for this relationship. Indeed, Shin et al.³² suggested that eyes with a larger FAZ tend to form larger IMHs. This indicates a close association between MA and FAZ. Additionally, MFR demonstrated a stronger association with preoperative BCVA than FAZ (β = 0.280, P < 0.05; β = 0.218, P < 0.05, respectively). Therefore, MFR seems to be a factor that emphasizes the functional role of FAZ. 

In the postoperative multiple regression analysis, model 1 revealed axial length as an independent associated factor for postoperative BCVA (β = –0.271, P < 0.01), whereas model 2 identified both axial length and MFR as independent associated factors (β = 0.255, P < 0.01; β = –0.357, P < 0.01, respectively). The unexpected negative association with axial length suggests several possible interpretations. For instance, Mori et al.³³ cited the report by Guyer et al.,³⁴ highlighting fluid movements as a potential cause of tangential traction. Assuming that posterior vitreous detachment progresses more slowly in eyes with shorter axial length, tangential traction owing to fluid movements could gradually accumulate damage to the macular retina over time in these eyes, potentially explaining the observed association with BCVA. Furthermore, tangential traction owing to fluid movements has been observed during fixation shifts in measurement. In addition, in the postoperative multiple regression analysis, MFR was the only independent associated factor among the morphological factors constituting the IMH. Thus, MFR was the only consistent independent associated factor for BCVA both preoperatively and postoperatively. The size of the FAZ may be altered by tissue distortion owing to axial or even radial drift as the hole increases in size. The change in FAZ size may have a negative effect on its association with BCVA. Nevertheless, despite the possibility of such effects, this study revealed a significant association between MFR and BCVA. This finding highlights a direct and significant role of FAZ in the visual function of IMH. Furthermore, it is compelling and logical that the preoperative ratio of MA to FAZ (i.e., MFR) shows a strong association with postoperative visual function once the hole has closed, indicating the structural importance of MFR in recovery. 

Provis et al.³⁵ investigated the role of FAZ as an indicator of high visual acuity in the central retina, with a specific emphasis on foveal hypoplasia. In a recent study, Okumichi et al.³⁶ explored the association between the structure of the central fovea and visual acuity in nanophthalmos. This study reported a significant correlation between visual acuity and deep FAZ size in the case group in which FAZ formation was observed. However, owing to the small sample size, we could not establish a clear relationship between morphology and function definitively. In this study, a significant association was found between preoperative and postoperative visual acuity and FAZ, which seems to support this hypothesis. Furthermore, a stronger association with the ratio reflecting the degree of FAZ impairment (i.e., MFR) suggests that the area of FAZ holds particular significance in terms of both morphology and function. From these insights, we deduced that the FAZ is a crucial area, not only for high visual acuity, but also as a significant factor in the relationship between morphology and functional impairment in IMH. The findings of this study open avenues for further research, promising a deeper understanding of the role of FAZ in IMH and other retinal diseases. We anticipate that subsequent studies will further elucidate these relationships, enhancing our understanding of the intricate interplay between retinal morphology and visual function. 

This study has several limitations. First, the small sample size may restrict the generalizability of our findings. Future investigations with larger cohorts are required to establish a more robust understanding of the intricate relationship between morphology and function. Second, the calculation of MFR requires measurement of the FAZ area, making it impossible to use MFR for predictions without OCTA equipment. Although other methods were considered, the necessity for FAZ information renders their implementation impractical. Third, our study did not account for the potential influence of retinal cysts and fluid cuffs, which are pivotal findings in the evaluation of morphology and function. Further studies are warranted to elucidate the effects of these factors. Fourth, the exact duration of IMH could not be precisely determined owing to its dependency on subjective symptoms, leading to the adoption of a stage classification approach. However, the classification did not show a significant association with BCVA in either the preoperative or postoperative analyses. Fifth, this investigation did not explore whether the MFR is superior to the MD as a prognostic indicator for IMH. Instead, MA was incorporated as a candidate prognostic parameter in lieu of MD. As a result of the postoperative multivariate analysis, MFR was identified as the only independent associated factor among the morphological factors, suggesting that it may serve as a more useful predictive factor than MD. Lastly, the scope of this study was confined to the immediate postoperative period. It is anticipated that future longitudinal studies will illuminate the long-term postoperative FAZ morphology, thereby enhancing our understanding of the IMH-FAZ interplay. 

In conclusion, the morphological factors associated with visual acuity before and after surgery for IMH included FAZ, MFR, MA, BA, and CST. Among these, the consistent emergence of MFR as an independent factor related to visual acuity both preoperatively and postoperatively is particularly intriguing and highlights the potential significance of FAZ in the morphological and functional evaluation of IMH. Further research in this area is eagerly anticipated to address these limitations and deepen our understanding of this complex topic. 

Acknowledgments

The authors thank Editage for grammar and spelling checks. 

Supported in part by a grant to T.S. from a JSPS KAKENHI Grant Number 19K09976. 

Author Contributions: JK, T Shoji, and KS: Designed and conducted the study; JK, T Shoji, H Ishii, H Ibuki, RW, YY, and T Sasaki: Data collection; JK, T Shoji, and KS: Data analysis and interpretation; JK: Writing; T Shoji and KS: Critical revision; JK, T Shoji, H Ishii, H Ibuki, RW, YY, T Sasaki, and KS: Manuscript approval. 

Institutional Review Board Statement: This study was conducted in accordance with the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of Saitama Medical University. 

Informed Consent Statement: Informed consent was obtained from all participants involved in the study. 

Data Availability Statements: The datasets generated and/or analyzed in the current study are available from the corresponding author upon reasonable request. 

Disclosure: J. Kanno, None; T. Shoji, None; H. Ishii, None; H. Ibuki, None; Y. Yoshikawa, None; T. Sasaki, None; K. Shinoda, None