May 2023
Volume 12, Issue 5
Open Access
Retina  |   May 2023
Bidirectional Dimples After Internal Limiting Membrane Peeling for a Macular Hole
Author Affiliations & Notes
  • Young Ho Kim
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Myung-Sun Song
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Ariunaa Togloom
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Kyung-Sook Yang
    Department of Biostatistics, Korea University College of Medicine, Seoul, Korea
  • So Min Ahn
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Cheolmin Yun
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Jaeryung Oh
    Department of Ophthalmology, Korea University College of Medicine, Seoul, Korea
  • Correspondence: Jaeryung Oh, Department of Ophthalmology, Korea University Medicine, 73 Goryeodae-ro Sungbuk-ku, Seoul 02841, Korea. e-mail: [email protected]; [email protected] 
Translational Vision Science & Technology May 2023, Vol.12, 23. doi:https://doi.org/10.1167/tvst.12.5.23
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Young Ho Kim, Myung-Sun Song, Ariunaa Togloom, Kyung-Sook Yang, So Min Ahn, Cheolmin Yun, Jaeryung Oh; Bidirectional Dimples After Internal Limiting Membrane Peeling for a Macular Hole. Trans. Vis. Sci. Tech. 2023;12(5):23. https://doi.org/10.1167/tvst.12.5.23.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To investigate microstructural changes and prognosis associated with retinal surface dimples after internal limiting membrane (ILM) peeling for macular holes (MHs).

Methods: We analyzed swept-source optical coherence tomography (SS-OCT) images of patients who underwent surgery for idiopathic MHs. The inner retinal dimples on SS-OCT images were classified into three types: unidirectional, bidirectional, and complicated bidirectional dimples.

Results: Dimples were found in 97.1% of the 69 eyes (69 patients) during a mean follow-up period of 14.0 ± 11.9 months after MH surgery. Of the eyes with dimples, 83.6% had bidirectional dimples. The proportion of eyes with dimples increased from 55.3% at 1 month postsurgery to 95.5% at 3 months and 97.9% at 6 months postsurgery. However, the proportion of eyes with complicated bidirectional dimples gradually increased from 1 month (29.8%) to 3 months (46.3%) and 6 months (64.6%) postsurgery. In the multivariable generalized estimating equation model, complicated bidirectional dimples occurred more frequently in eyes with shorter axial length (P = 0.039) and longer follow-up duration (≥6 months; at 6 months: P = 0.001; at 12 months: P = 0.009).

Conclusions: Changes in retinal layers associated with retinal surface dimples after ILM peeling can occur at different retinal depths and over different time courses. These findings suggest the progression of dimple-associated remodeling of the underlying retinal layer.

Translational Relevance: Various types of dimples can be used as surrogates to evaluate structural changes and outcomes of MH surgery.

Introduction
Changes in the inner retinal layer, associated with an irregular shape of the inner retinal surface, have been reported after macular hole (MH) surgery. In earlier studies,13 the dimple-like changes in the inner retinal layers were suggested to represent a dissociation of the optic nerve fiber layers (DONFLs). Mitamura and Ohtsuka3 proposed that the appearance of DONFLs is caused by internal limiting membrane (ILM) peeling and is unrelated to the loss of optic nerve fibers. Spaide4 showed focal areas of thinning of the ganglion cell layer and suggested that this change, rather than DONFL, underlay the formation of inner retinal dimples. Müller cell damage due to ILM peeling has been suggested as a cause of dimple appearance,48 which suggests that dimple appearance is a consequence of a common healing reaction to the surgical trauma to the ILM in patients with MH. 
In previous studies, the change associated with the dimple appearance was thought to be localized on the surface,13,9 and the presence of dimples has not been suggested to be a factor influencing the patient's final visual prognosis in terms of central vision.2,7,1012 However, in other studies,13,14 it was shown that the effect of ILM peeling was not limited to the inner retinal surface but to the deeper retinal layers. This suggests that the effect of the dimples may be more extensive than previously thought. However, the microstructural alterations in the deeper retina underlying the inner retinal dimples have not been clearly elucidated. 
In this study, we hypothesized that the effects of the changes associated with dimple appearance could be transmitted differently beyond the surface of the retina to deeper layers and that they could have various influences on the microstructural features in both superficial and deep retinal layers. We classified the changes in the underlying retinal layers associated with dimple appearance into three classes and evaluated the associated prognoses. 
Methods
This retrospective study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of the Korea University Hospital. We retrospectively reviewed the medical records of consecutive patients diagnosed with MH who underwent primary vitrectomy with ILM peeling between September 2015 and June 2022 and were followed up using swept-source optical coherence tomography (SS-OCT) at the Korea University Hospital. 
We included patients in whom the MH was successfully closed using a single surgery and who had a minimum postoperative follow-up period of 3 months. The exclusion criteria were as follows: (1) patients who were diagnosed with traumatic MH; (2) those with high myopia with axial length exceeding 28.5 mm and/or MH combined with myopic tractional maculopathy such as myopic foveoschisis, retinoschisis, or foveal detachment; (3) those who underwent reoperation for unclosed MH after vitrectomy; (4) those with diabetic retinopathy or diabetic macular edema; (5) those with MH caused by other secondary causes, including type 2 macular telangiectasia, age-related macular degeneration, tractional retinal detachment, laser injury, and uveitis; (6) those with previous vitreoretinal disease or a history of having undergone vitreoretinal surgery; and (7) those with poor preoperative and postoperative image quality (signal strength index <50) that limited image analysis. 
One of the three right-handed surgeons (JO, YHK, or SMA) performed pars plana vitrectomy with their preferred choice of 25-, 23- or 20-gauge three-port vitrectomy systems. Combined cataract surgery was performed at the surgeon's discretion. After core vitrectomy, posterior vitreous detachment was induced when needed. ILM peeling was performed using end-gripping micro-forceps after indocyanine green staining. The epiretinal membrane was removed if present. ILM peeling was completed symmetrically around the fovea in a semicircular manner for approximately 3 to 4 disc-diameters. After fluid–air exchange, intravitreal gas tamponade was performed using 14% perfluoropropane (C3F8). Postoperatively, patients were instructed to maintain a face-down position for 7 days. 
Before surgery, all patients underwent a comprehensive ophthalmic examination, including slit-lamp biomicroscopy, fundus photography, and SS-OCT. Preoperative data from medical records, including sex, age, symptom duration, best-corrected visual acuity (BCVA), and axial length, measured using laser interferometry (IOL Master; Carl Zeiss Meditec, Jena, Germany), were collected. All patients were examined using the same SS-OCT device after surgery. The first postoperative OCT scan was acquired about 3 to 5 weeks after surgery when the gas had dissipated and macular scans were possible. OCT scans were performed at 3 and 6 months after surgery and then approximately every 6 months thereafter. Follow-up OCT images for image analysis were selected at the closest visit, followed by 1, 3, 6, and 12 months postoperatively and at the last visit. When the follow-up period was less than 12 months, the images and data from the most recent visit of each patient were used as the last visit. 
Optical Coherence Tomography
For preoperative measurements, we obtained macula volume scans with a nominal scan area of 12 × 9 mm, encompassing both the macula and optic disc using an SS-OCT device (DRI OCT Triton; Topcon Corp., Tokyo, Japan). Basal and minimal hole diameters and height were measured using the built-in software in the SS-OCT device, and the MH index was calculated as the ratio of the MH height to the diameter of the MH base, as previously described.15,16 For postoperative measurement of dimples, we used high-resolution B-scan images acquired using the five-line cross-scan protocol of the SS-OCT device (Supplementary Fig. S1). The five-line cross-scan protocol covered 9 mm, centered on the macula, and consisted of 10 B-scans with five lines in each of the horizontal and vertical directions, at 250-µm intervals. Each high-resolution B-scan image was obtained by averaging the eight B-scan images at each location. To determine foveal integrity, we used both macular volume and five-line cross-scans. 
Classification of the Dimples and Determination of Foveal Integrity
In the five-line cross-scan, the dimple was determined as the dent of the inner retinal surface on postoperative SS-OCT. The dimples were classified as unidirectional and bidirectional dimples (Fig. 1). Unidirectional dimples were defined as dimples without any underlying retinal changes. Bidirectional dimples were defined as when the dimple was associated with both a dent on the inner retinal surface and upward bowing of the interface between the inner plexiform layer (IPL) and inner nuclear layer (INL). Bidirectional dimples were graded into uncomplicated and complicated bidirectional dimples. A complicated bidirectional dimple was identified when the bulging of the hyporeflective area representing the outer nuclear layer (ONL) on OCT scans was apparent. 
Figure 1.
 
Unidirectional and bidirectional dimples. A unidirectional dimple (white arrowheads) was defined as a dimple without underlying retinal change. A bidirectional dimple (yellow and red arrows) was defined as when the dimple was associated with both a dent in the inner retinal surface and upward bowing of the interface between the inner plexiform layer and inner nuclear layer (yellow and red dotted lines). Bidirectional dimples were graded into uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples. A complicated bidirectional dimple was defined as when bulging of the outer nuclear layer was apparent (red dashed lines).
Figure 1.
 
Unidirectional and bidirectional dimples. A unidirectional dimple (white arrowheads) was defined as a dimple without underlying retinal change. A bidirectional dimple (yellow and red arrows) was defined as when the dimple was associated with both a dent in the inner retinal surface and upward bowing of the interface between the inner plexiform layer and inner nuclear layer (yellow and red dotted lines). Bidirectional dimples were graded into uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples. A complicated bidirectional dimple was defined as when bulging of the outer nuclear layer was apparent (red dashed lines).
In the line and volume scans, foveal integrity after MH surgery was determined and classified into glial and nonglial sealing groups, as previously reported.17,18 Glial sealing was determined when the moderately to highly reflective tissue seen on OCT involved the external limiting membrane and photoreceptor layers.17 Two experts (YHK and JO) blinded to the date of surgery examined each image and determined the type of dimple and the foveal integrity, with consensus reading. 
Statistical Analysis
The Kolmogorov–Smirnov test was used to determine the normality of distribution of continuous variable data. Continuous variables with a normal distribution were expressed as means ± standard deviations, while those without were expressed as median and interquartile range (IQR). Categorical variables were described using numbers (%). Comparisons between groups were performed using independent t-test or the Wilcoxon's rank-sum test for continuous variables and chi-square test or Fisher's exact test for categorical variables. Visual acuity measured using the Snellen visual acuity chart was converted to the logarithm of minimum angle of resolution (logMAR) visual acuity for the statistical analysis, and paired t-test was used to compare pre- and postoperative BCVA. The occurrence of dimples according to postoperative follow-up period was analyzed using univariate and multivariable generalized estimating equations (GEEs) to account for the repeated measurements in each eye with compound symmetry correlation structures. Each final multivariable GEE model was selected using the stepwise variable selection method. To investigate the correlation between final visual acuity and several parameters, univariate regression analyses were conducted using the Pearson's correlation coefficient (r). Then, multiple regression analysis with the backward elimination method was performed, using variables that had P < 0.10 in the univariate analysis. A P value <0.05 for two-sided tests was considered statistically significant. All statistical analyses, except GEE, were performed using the Statistical Package for the Social Sciences (version 20.0; IBM Corp., Armonk, NY, USA). Univariate and multivariate GEE analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA). 
Results
General Characteristics
This study included 69 eyes of 69 patients with idiopathic MH (Table 1). The mean age of the patients was 65 ± 7.4 years. The proportion of women (62.3%) was higher than that of men (37.7%). The mean follow-up duration was 14.0 ± 11.9 months. After MH surgery, the MHs in all eyes were closed, and their preoperative BCVA (logMAR) (median [IQR], 0.69 [0.52–1.00]) had improved at the last examination (0.22 [0.08–0.52]) (P < 0.001). 
Table 1.
 
General Characteristics (N = 69)
Table 1.
 
General Characteristics (N = 69)
Occurrence of Various Types of Dimples and Pre- and Postoperative Factors
At the last visit, 67 (97.1%) patients had at least one type of dimple in the study eye (Table 2). All types of dimples were observed on both the horizontal and vertical scans centered on the fovea (Supplementary Fig. S2). Among 67 eyes with dimples, 56 (83.6%) eyes showed bidirectional dimples. Of the 56 eyes with bidirectional dimples, uncomplicated and complicated bidirectional dimples were observed in 40 (71.4%) and 16 (28.6%) eyes, respectively (Fig. 2). 
Table 2.
 
Occurrence and Type of Dimples after Macular Hole Surgery
Table 2.
 
Occurrence and Type of Dimples after Macular Hole Surgery
Figure 2.
 
Representative cases of dimple progression from unidirectional dimples (white arrowheads) to uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples.
Figure 2.
 
Representative cases of dimple progression from unidirectional dimples (white arrowheads) to uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples.
Baseline and postoperative factors, as well as the vitrectomy systems used, were not different between patients with dimples, bidirectional dimples, or complicated bidirectional dimples at the last examination and those without (Supplementary Table S1). The number of dimples, bidirectional dimples, or complicated bidirectional dimples did not correlate with age, symptom duration, hole diameter, or MH index (Supplementary Table S2). 
Factors Related to the Development of Various Types of Dimples
Results of the univariate and multivariate GEE analysis to test the association between the occurrence of various types of dimples and several preoperative factors are shown in Supplementary Table S3 and Table 3. In the multivariate GEE analysis that included age and follow-up duration, the occurrence of any type of dimple at the last visit was found to be associated with age (odds ratio [OR], 1.06; 95% confidence interval [CI], 1.00–1.13; P = 0.041) and follow-up duration after MH surgery (all P < 0.001) (Supplementary Table S3). 
Table 3.
 
Univariate and Multivariable Generalized Estimating Equation Model for Eyes with Bidirectional and Complicated Bidirectional Dimples After Macular Hole Surgery
Table 3.
 
Univariate and Multivariable Generalized Estimating Equation Model for Eyes with Bidirectional and Complicated Bidirectional Dimples After Macular Hole Surgery
In the multivariable GEE model that included age, axial length, basal hole diameter, minimal hole diameter, hole height, MH index, and follow-up duration, the occurrence of bidirectional dimples was significantly associated with axial length, basal hole diameter, minimal hole diameter, hole height, and follow-up duration (all P < 0.05) (Table 3). The occurrence of bidirectional dimples increased with longer follow-up duration (all P < 0.001), and they were more frequently observed in eyes with shorter axial lengths (OR, 0.75; 95% CI, 0.58–0.96; P = 0.021). 
In the multivariable GEE model that included age, axial length, and follow-up duration, complicated bidirectional dimples occurred significantly more frequently in eyes with shorter axial lengths (OR, 0.78; 95% CI, 0.62–0.99; P = 0.039) and longer follow-up duration (≥6 months; at month 6, P = 0.001; at month 12, P = 0.009) (Table 3). 
Occurrence of Various Types of Dimples over Time
At 1 month after surgery, eyes with dimples were only observed in 55.3% of the patients (Table 2). Compared to month 1, the occurrence of dimples increased significantly by 95.5% at month 3 (OR, 18.41; 95% CI, 3.72–91.16; adjusted P < 0.001), 97.9% at month 6 (OR, 40.26; 95% CI, 3.39–478.24; adjusted P = 0.001), and 100% at month 12 after surgery in multiple pairwise comparisons with Tukey's adjustment in multivariable GEE models (Supplementary Table S4). Similarly, the occurrence of eyes with bidirectional dimples, including uncomplicated and complicated bidirectional dimples, was 38.3% at month 1, and it increased to 74.6 % by month 3 but plateaued at 81.3% by month 6 and 87.5% by month 12. Compared to month 1, bidirectional dimples were more frequently observed at month 3 (OR, 5.32; 95% CI, 1.69–16.73; adjusted P < 0.001), month 6 (OR, 7.27; 95% CI, 1.92–27.55; adjusted P < 0.001), and month 12 (OR, 12.79; 95% CI, 2.61–62.66; adjusted P < 0.001). However, the occurrence of complicated bidirectional dimples did not differ between postoperative months 1 (29.8%) and 3 (46.3%) (adjusted P = 0.317). They were observed in 64.6% of the eyes at month 6 and in 59.4% of the eyes at month 12. The increase was statistically significant for both months 6 (OR, 4.21; 95% CI, 1.35–13.13; adjusted P = 0.006) and 12 (OR, 3.58; 95% CI, 1.03–12.48; adjusted P = 0.043) as compared to month 1. 
Bidirectional Dimple and Visual Prognosis
BCVA did not differ significantly between eyes with and without bidirectional dimples or complicated bidirectional dimples at month 3 (P = 0.107, P = 0.076, respectively), month 6 (P = 0.083, P = 0.070, respectively), or month 12 (P = 0.583, P = 0.899, respectively). The final BCVA did not differ between eyes with and without bidirectional dimples (Supplementary Table S5). However, the median final BCVA of 40 eyes with complicated dimples (0.19 [0.06–0.30]) was better than that of those without complicated dimples (0.40 [0.10–0.70]) (P = 0.039). Eyes with glial sealing at the fovea had worse vision than those without glial sealing (P < 0.001). 
In univariate analysis, the final BCVA was correlated with symptom duration, basal hole diameter, minimum hole diameter, MH index, preoperative BCVA, and mean number of dimples (all P < 0.05) (Supplementary Table S6). However, it did not correlate with the number of bidirectional or complicated bidirectional dimples. In multiple regression analysis to identify factors influencing final visual acuity, symptom duration, preoperative BCVA, basal hole diameter, follow-up duration, mean number of dimples, presence of complicated bidirectional dimples, and presence of glial sealing at the last visit were included. In the final model, the presence of glial sealing (β = 0.451, P < 0.001) and basal hole diameter (β = 0.324, P = 0.001) were statistically significant factors influencing the final BCVA (Table 4). The presence of complicated bidirectional dimples was included in the final model and had borderline significance (β = −0.185, P = 0.051). 
Table 4.
 
Multiple Linear Regression Analysis with Backward Elimination Method for Risk Factors of Final Visual Acuity in 69 Patients Including Symptom Duration, Preoperative Best-Corrected Visual Acuity, Basal Hole Diameter, Follow-up Duration, Mean Number of Any Type of Dimple, Presence of Complicated Bidirectional Dimples, and Presence of Glial Sealing at the Last Visit
Table 4.
 
Multiple Linear Regression Analysis with Backward Elimination Method for Risk Factors of Final Visual Acuity in 69 Patients Including Symptom Duration, Preoperative Best-Corrected Visual Acuity, Basal Hole Diameter, Follow-up Duration, Mean Number of Any Type of Dimple, Presence of Complicated Bidirectional Dimples, and Presence of Glial Sealing at the Last Visit
Discussion
Changes in the inner retinal surface after ILM peeling for MH surgery, known as dimples or DONFL appearance, have previously been reported. Mitamura et al.1 presented that dimples were limited to the retinal nerve fiber layer (RNFL) thickness. Other studies reported that these changes were limited to only RNFL or ganglion cell layers and had no adverse effects on visual function. These studies described them as inner retinal dimples from the viewpoint of the innermost retinal surface or entire retina from the retinal pigment epithelium to the ILM. However, other studies have reported that these structural alterations may affect the deeper retina underlying the inner retinal dimples.13,14 In the present study, we classified dimples as unidirectional or bidirectional depending on whether any alterations were observed in the IPL or INL. On OCT B-scan images, we observed that a large number of inner retinal dimples had a microstructural change that caused the lower boundary of the IPL layer to bulge upward (Fig. 1). We have named the inner retinal dimples with these microstructural changes bidirectional dimples from the viewpoint of the middle layers of the retina, including the IPL and INL. Of the patients with dimples, 83.6% had bidirectional dimples involving the IPL and INL. These findings suggested that ILM peeling may lead to changes in layers deeper than the surface of the retina or inner retinal layer. However, it is unclear why unidirectional dimples occur in some cases and bidirectional dimples occur in others. One possibility is that the incidence of the different types of dimples is related to the differences in baseline characteristics. In the current study, the incidence of bidirectional dimples was associated with baseline characteristics such as axial length and hole configuration. Another possibility is that bidirectional dimple is a consequence of damage to the ganglion cells. Using intraoperative OCT imaging, Runkle et al.7 showed that an acute increase in inner retinal thickness following ILM peeling can induce dimples in eyes with MH. This result may suggest that more tractional force exerted to peel the ILM can induce more damage to ganglion cells and affect Müller cell integrity. 
In this study, 71.4% of the patients with bidirectional dimples had complicated bidirectional dimples involving the ONL. The pathogeneses of complicated bidirectional dimples is not clear. In a recent study, Tao et al.19 suggested that distorted retinal layers could be associated with ONL bumps under dimples, and changes in the macular configuration after MH surgery may have induced this distortion. Another possibility is that ONL thickening could have been induced by a change in the configuration of Henle's fiber layer (Fig. 3). Ikeda et al.20 suggested that apoptosis of ganglion cells can occur because of detachment from the surrounding extracellular matrix. Upward traction induced by severe shrinkage of the ganglion cell layer may have dragged Henle's fiber layer through the INL and outer plexiform layer.19 In the OCT B-scan, both Henle's fiber layer and ONL have similar reflectivity and are observed as hyporeflective areas.21,22 It is not easy to distinguish Henle's fiber layer from the ONL in some areas of OCT scans. However, in some other areas, Henle's fiber layer has higher reflectivity than the ONL, and it can be distinguished from the reflectivity of the ONL.2325 The dragged Henle's fibers may have appeared as bulging of the ONL in the OCT images. 
Figure 3.
 
Change in the reflectivity of Henle's layer. Dimple progression was associated with a change in the reflectivity of Henle's layer from month 6 (top) to month 18 (bottom) after internal limiting membrane peeling for macular hole surgery.
Figure 3.
 
Change in the reflectivity of Henle's layer. Dimple progression was associated with a change in the reflectivity of Henle's layer from month 6 (top) to month 18 (bottom) after internal limiting membrane peeling for macular hole surgery.
In our patients, the timeline of progression of complicated bidirectional dimples was different from those of other types of dimples. Dimples were found in approximately half of the patients at 1 month after surgery. This proportion had increased by 3 months after surgery and then plateaued thereafter. However, complicated bidirectional dimples were found in less than 30% of the patients at 1 month after surgery, and the proportion gradually increased from months 3 to 6. This finding supports the hypothesis that the development of complicated bidirectional dimples is related to the wound-healing process. Depending on the degree of trauma, remodeling of the retinal layer underlying the dimple may have occurred over a different time course. Tao et al.19 showed that areas with dimples can be associated with ONL bumps in the perifoveal temporal macula after ILM peeling.19 However, in our study, which used both horizontal and vertical line scans, complicated bidirectional dimples involving the ONL were not limited to the temporal macula but were also observed in the superior and inferior perifoveal areas. These observations suggest that a late change in retinal configuration can occur as a consequence of remodeling of retinal layers involving the whole area in which the ILM was peeled. 
It was initially thought that dimple appearance could cause vision abnormalities; however, in subsequent studies, a significant correlation between visual acuity and dimples was ruled out. In several previous studies, no correlation was found between the presence of dimples and BCVA.2,7,1012,26,27 In a previous study,14 postoperative BCVA did not differ significantly based on the depth of the dimples. Ito et al.2 suggested that eyes with DONFL have no functional abnormalities in microperimetric observation. However, in recent studies,8,19 the possibility of a relationship between dimples and visual function was presented. Distortions in the retinal layers have been proposed to underlie the pathogenesis of scotomas. In our study, the presence of complicated bidirectional dimples was associated with the final BCVA. However, in a model, including preoperative and postoperative factors, such as basal hole diameter and glial sealing, which are known to be related to visual prognosis after MH surgery, the relationship between complicated bidirectional dimples and final BCVA was not significant. Moreover, final vision was better in eyes with complicated dimples than those without complicated dimples. This suggests that the presence of complicated dimples is not a significant factor that prevents favorable visual outcomes. The results of our study suggest that the retinal changes associated with dimples, whether mild or severe, do not adversely affect visual prognosis. 
This study was not free from the limitations of a small number of cases and a retrospective design. We did not measure microperimetry, and we were not able to measure the effect of various dimples on scotoma. After ILM peeling, dimples occurred around the macular center. We only evaluated dimples in a limited area covered by the five-line cross-scans in the retrospective study. Five-line cross-scans have a higher resolution than the volume scan, and it was easy to determine the presence of various types of dimples. While the five-line cross-scans included both horizontal and vertical scans, they underestimated the extent of the dimples in the diagonal area of the macula. Additionally, while reviewing the color fundus photographs of the built-in image viewer, we assessed the SS-OCT images, but the 250-µm intervals between each OCT scan might have prohibited us from analyzing the images from the same position. Radial scans that evenly cover the entire area of the macula would have been better for determining the presence or the type of dimple. 
In conclusion, dimples progressed with thinning of the ganglion cell and IPL over time. They were also associated with changes in the underlying INL and ONL. These findings were suggestive of the progression of dimple-associated remodeling of the underlying retinal layer. 
Acknowledgments
Supported by the Korea University grant (K2208351) and by the Korea Medical Device Development Fund grant funded by the Korean government (Ministry of Science and ICT, Ministry of Trade, Industry and Energy, Ministry of Health & Welfare, Ministry of Food and Drug Safety) (Project Number: 1711174316, RS-2020-KD000026). 
Disclosure: Y.H. Kim, None; M.-S. Song, None; A. Togloom, None; K.-S. Yang, None; S.M. Ahn, None; C. Yun, None; J. Oh, None 
References
Mitamura Y, Suzuki T, Kinoshita T, Miyano N, Tashimo A, Ohtsuka K. Optical coherence tomographic findings of dissociated optic nerve fiber layer appearance. Am J Ophthalmol. 2004; 137(6): 1155–1156. [CrossRef] [PubMed]
Ito Y, Terasaki H, Takahashi A, Yamakoshi T, Kondo M, Nakamura M. Dissociated optic nerve fiber layer appearance after internal limiting membrane peeling for idiopathic macular holes. Ophthalmology. 2005; 112(8): 1415–1420. [CrossRef] [PubMed]
Mitamura Y, Ohtsuka K. Relationship of dissociated optic nerve fiber layer appearance to internal limiting membrane peeling. Ophthalmology. 2005; 112(10): 1766–1770. [CrossRef] [PubMed]
Spaide RF. “Dissociated optic nerve fiber layer appearance” after internal limiting membrane removal is inner retinal dimpling. Retina. 2012; 32(9): 1719–1726. [CrossRef] [PubMed]
Nakamura T, Murata T, Hisatomi T, et al. Ultrastructure of the vitreoretinal interface following the removal of the internal limiting membrane using indocyanine green. Curr Eye Res. 2003; 27(6): 395–399. [CrossRef] [PubMed]
Yooh HS, Brooks HL, Jr, Capone A, Jr, L'Hernault NL, Grossniklaus HE. Ultrastructural features of tissue removed during idiopathic macular hole surgery. Am J Ophthalmol. 1996; 122(1): 67–75. [PubMed]
Runkle AP, Srivastava SK, Yuan A, et al. Factors associated with development of dissociated optic nerve fiber layer appearance in the PIONEER intraoperative optical coherence tomography study. Retina. 2018; 38(suppl 1): S103–S109. [PubMed]
Steel DH, Dinah C, White K, Avery PJ. The relationship between a dissociated optic nerve fibre layer appearance after macular hole surgery and Muller cell debris on peeled internal limiting membrane. Acta Ophthalmol. 2017; 95(2): 153–157. [CrossRef] [PubMed]
Kishimoto H, Kusuhara S, Matsumiya W, Nagai T, Negi A. Retinal surface imaging provided by Cirrus high-definition optical coherence tomography prominently visualizes a dissociated optic nerve fiber layer appearance after macular hole surgery. Int Ophthalmol. 2011; 31(5): 385–392. [CrossRef] [PubMed]
Christensen UC, Krøyer K, Sander B, et al. Value of internal limiting membrane peeling in surgery for idiopathic macular hole stage 2 and 3: a randomised clinical trial. Br J Ophthalmol. 2009; 93(8): 1005–1015. [CrossRef] [PubMed]
Kusuhara S, Matsumiya W, Imai H, Honda S, Tsukahara Y, Negi A. Evaluating dissociated optic nerve fiber layer appearance using en face layer imaging produced by optical coherence tomography. Ophthalmologica. 2014; 232(3): 170–178. [CrossRef] [PubMed]
Kim YJ, Lee KS, Joe SG, Kim JG. Incidence and quantitative analysis of dissociated optic nerve fiber layer appearance: real loss of retinal nerve fiber layer? Eur J Ophthalmol. 2018; 28(3): 317–323. [CrossRef] [PubMed]
Nukada K, Hangai M, Ooto S, Yoshikawa M, Yoshimura N. Tomographic features of macula after successful macular hole surgery. Invest Ophthalmol Vis Sci. 2013; 54(4): 2417–2428. [CrossRef] [PubMed]
Kim KY, Yu SY, Kim M, Kim ES, Kwak HW. Morphological change of inner retinal layer on spectral-domain optical coherence tomography following macular hole surgery. Ophthalmologica. 2013; 230(1): 18–26. [CrossRef] [PubMed]
Kusuhara S, Teraoka Escaño MF, Fujii S, et al. Prediction of postoperative visual outcome based on hole configuration by optical coherence tomography in eyes with idiopathic macular holes. Am J Ophthalmol. 2004; 138(5): 709–716. [CrossRef] [PubMed]
Ruiz-Moreno JM, Staicu C, Piñero DP, Montero J, Lugo F, Amat P. Optical coherence tomography predictive factors for macular hole surgery outcome. Br J Ophthalmol. 2008; 92(5): 640–644. [CrossRef] [PubMed]
Oh J, Yang SM, Choi YM, Kim SW, Huh K. Glial proliferation after vitrectomy for a macular hole: a spectral domain optical coherence tomography study. Graefes Arch Clin Exp Ophthalmol. 2013; 251(2): 477–484. [CrossRef] [PubMed]
Oh J, Smiddy WE, Flynn HW, Jr, Gregori G, Lujan B. Photoreceptor inner/outer segment defect imaging by spectral domain OCT and visual prognosis after macular hole surgery. Invest Ophthalmol Vis Sci. 2010; 51(3): 1651–1658. [CrossRef] [PubMed]
Tao J, Yang J, Wu Y, et al. Internal limiting membrane peeling distorts the retinal layers and induces scotoma formation in the perifoveal temporal macula. Retina. 2022; 42(12): 2276–2283. [CrossRef] [PubMed]
Ikeda T, Nakamura K, Sato T, Kida T, Oku H. Involvement of anoikis in dissociated optic nerve fiber layer appearance. Int J Mol Sci. 2021; 22(4): 1724. [CrossRef] [PubMed]
Bringmann A, Unterlauft JD, Barth T, Wiedemann R, Rehak M, Wiedemann P. Müller cells and astrocytes in tractional macular disorders. Prog Retin Eye Res. 2022; 86: 100977. [CrossRef] [PubMed]
Staurenghi G, Sadda S, Chakravarthy U, Spaide RF. Proposed lexicon for anatomic landmarks in normal posterior segment spectral-domain optical coherence tomography: the IN•OCT consensus. Ophthalmology. 2014; 121(8): 1572–1578. [CrossRef] [PubMed]
Lujan BJ, Roorda A, Knighton RW, Carroll J. Revealing Henle's fiber layer using spectral domain optical coherence tomography. Invest Ophthalmol Vis Sci. 2011; 52(3): 1486–1492. [CrossRef] [PubMed]
Mrejen S, Gallego-Pinazo R, Freund KB, Paques M. Recognition of Henle's fiber layer on OCT images. Ophthalmology. 2013; 120(6): e32–33.e31. [CrossRef] [PubMed]
Lujan BJ, Roorda A, Croskrey JA, et al. Directional optical coherence tomography provides accurate outer nuclear layer and Henle fiber layer measurements. Retina. 2015; 35(8): 1511–1520. [CrossRef] [PubMed]
Kumar V, Dubey D, Kumawat D, et al. Role of internal limiting membrane peeling in the prevention of epiretinal membrane formation following vitrectomy for retinal detachment: a randomised trial. Br J Ophthalmol. 2020; 104(9): 1271–1276. [PubMed]
Demirel S, Abdullayev A, Yanık Ö, Batıoğlu F, Özmert E. Evaluation of ganglion cell-inner plexiform layer thickness after vitreoretinal surgery with internal limiting membrane peeling in cases with idiopathic macular hole. Turk J Ophthalmol. 2017; 47(3): 138–143. [CrossRef] [PubMed]
Figure 1.
 
Unidirectional and bidirectional dimples. A unidirectional dimple (white arrowheads) was defined as a dimple without underlying retinal change. A bidirectional dimple (yellow and red arrows) was defined as when the dimple was associated with both a dent in the inner retinal surface and upward bowing of the interface between the inner plexiform layer and inner nuclear layer (yellow and red dotted lines). Bidirectional dimples were graded into uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples. A complicated bidirectional dimple was defined as when bulging of the outer nuclear layer was apparent (red dashed lines).
Figure 1.
 
Unidirectional and bidirectional dimples. A unidirectional dimple (white arrowheads) was defined as a dimple without underlying retinal change. A bidirectional dimple (yellow and red arrows) was defined as when the dimple was associated with both a dent in the inner retinal surface and upward bowing of the interface between the inner plexiform layer and inner nuclear layer (yellow and red dotted lines). Bidirectional dimples were graded into uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples. A complicated bidirectional dimple was defined as when bulging of the outer nuclear layer was apparent (red dashed lines).
Figure 2.
 
Representative cases of dimple progression from unidirectional dimples (white arrowheads) to uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples.
Figure 2.
 
Representative cases of dimple progression from unidirectional dimples (white arrowheads) to uncomplicated (yellow arrows) and complicated (red arrows) bidirectional dimples.
Figure 3.
 
Change in the reflectivity of Henle's layer. Dimple progression was associated with a change in the reflectivity of Henle's layer from month 6 (top) to month 18 (bottom) after internal limiting membrane peeling for macular hole surgery.
Figure 3.
 
Change in the reflectivity of Henle's layer. Dimple progression was associated with a change in the reflectivity of Henle's layer from month 6 (top) to month 18 (bottom) after internal limiting membrane peeling for macular hole surgery.
Table 1.
 
General Characteristics (N = 69)
Table 1.
 
General Characteristics (N = 69)
Table 2.
 
Occurrence and Type of Dimples after Macular Hole Surgery
Table 2.
 
Occurrence and Type of Dimples after Macular Hole Surgery
Table 3.
 
Univariate and Multivariable Generalized Estimating Equation Model for Eyes with Bidirectional and Complicated Bidirectional Dimples After Macular Hole Surgery
Table 3.
 
Univariate and Multivariable Generalized Estimating Equation Model for Eyes with Bidirectional and Complicated Bidirectional Dimples After Macular Hole Surgery
Table 4.
 
Multiple Linear Regression Analysis with Backward Elimination Method for Risk Factors of Final Visual Acuity in 69 Patients Including Symptom Duration, Preoperative Best-Corrected Visual Acuity, Basal Hole Diameter, Follow-up Duration, Mean Number of Any Type of Dimple, Presence of Complicated Bidirectional Dimples, and Presence of Glial Sealing at the Last Visit
Table 4.
 
Multiple Linear Regression Analysis with Backward Elimination Method for Risk Factors of Final Visual Acuity in 69 Patients Including Symptom Duration, Preoperative Best-Corrected Visual Acuity, Basal Hole Diameter, Follow-up Duration, Mean Number of Any Type of Dimple, Presence of Complicated Bidirectional Dimples, and Presence of Glial Sealing at the Last Visit
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×