Abstract
Background
Epiretinal membrane (ERM) formation, frequently encountered in retinal outpatient practice, is commonly caused by idiopathic and diabetic factors. In this surgically treated condition, choroidal changes occur alongside macular changes following ERM peeling.
Methods
A total of 36 patients who underwent ERM peeling were included in the study: 18 in the idiopathic group and 18 in the diabetic group. Preoperative and postoperative one-month macular and choroidal measurements were obtained using the Heidelberg Engineering Spectralis spectral-domain optical coherence tomography device. The CVI was calculated manually after segmenting the choroid using the ImageJ software.
Results
No statistically significant differences were found between the groups in terms of age or sex. Preoperatively, statistically significant differences were found between the groups in terms of central foveal thickness (CFT), superior inner retinal thickness (SIRT), temporal inner retinal thickness (TIRT), and inferior inner retinal thickness (IIRT). Postoperatively, a significant difference was observed only in superior outer retinal thickness (SORT) and the CVI of the Haller layer. When the preoperative and postoperative periods were compared within the idiopathic group, significant differences were found in best-corrected visual acuity (BCVA), CFT, and total CVI. In the diabetic group, preoperative and postoperative comparison revealed significant differences in BCVA, CFT, SIRT, TIRT, IIRT, nasal inner retinal thickness, SORT, Sattler-CC CVI, and total CVI.
Conclusion
Although significant differences in retinal thickness were found between the two groups preoperatively, particularly in CFT and in certain sectors, these differences disappeared postoperatively, except for the SORT sector. This finding indicated that, although retinal thickness was greater in diabetic ERM before surgery, the ERM surgery itself may have produced similar effects on retinal thickness in both the idiopathic and diabetic ERMs in the early postoperative period. The choroidal structures were similar between the groups preoperatively; postoperatively, a difference was observed only in the CVI of the Haller layer. This finding suggests that surgery-related inflammatory changes and the associated increase in CVI were more evident in the diabetic group, despite similar preoperative CVI values between idiopathic and diabetic ERM cases.
Keywords: Idiopathic, Diabetic, Epiretinal membrane, Vitreoretinal surgery
Background
The formation of an epiretinal membrane (ERM) is commonly due to primary and secondary causes and is frequently encountered in retinal outpatient practice. Primary causes are idiopathic, and there are many different theories regarding the mechanism of formation. The cause cited most often is glial fibroproliferation on the internal limiting membrane (ILM) following posterior vitreous detachment (PVD) [1, 2]. Secondary causes include various systemic and ocular conditions, primarily diabetes, as well as hypertension, central retinal vein occlusion, uveitis, and panretinal photocoagulation. Among these, diabetic ERM is the most common [3, 4].
An ERM typically presents with vision loss and metamorphopsia. Treatment for ERM is surgical. The decision to proceed with surgery is based on changes in the retinal layers associated with distortion of the foveal contour as well as the patient’s symptoms. During surgery, the ERM is peeled away in conjunction with pars plana vitrectomy (PPV) [5, 6].
The choroidal vascular index (CVI) is an index that began to be measured with the development of enhanced deep imaging optical coherence tomography (EDI-OCT), and it is used as a biomarker for certain diseases [7, 8]. Initially, choroidal thickness was measured using EDI-OCT; however, since understanding later developed that choroidal thickness is influenced by various factors, such as age, gender, axial length, and intraocular pressure (IOP), the CVI became preferred. Furthermore, while choroidal thickness measures only a single cross-section, the CVI measures an area. CVI has emerged as a preferred quantitative imaging biomarker for assessing choroidal vascular status and structural changes in chorioretinal diseases. In the calculation of the CVI, the choroid is divided into two parts: the stromal and luminal areas. The CVI is obtained by dividing the luminal area by the total choroidal area [9–11].
Although ERM is a macular disease that causes changes in the macula, it is also likely to affect the adjacent choroid. For this reason, many studies have evaluated the CVI following ERM surgery. These studies generally calculated the total CVI [8, 12, 13]. In our study, the choroid was segmented into the Haller and Sattler-choriocapillaris (CC) layers, and the CVI of each was calculated separately. To the best of our knowledge, this study is the first to evaluate macular and choroidal changes in idiopathic and diabetic ERMs from this perspective. This study compared the postoperative nine-quadrant macular thicknesses and the CVIs of the Haller and Sattler-CC layers in both idiopathic and diabetic ERMs and evaluated whether these could be used as biomarkers of postoperative anatomical and functional recovery.
Methods
This retrospective, cross-sectional study was conducted in accordance with the principles of the Declaration of Helsinki. Informed consent was obtained from all participants. Local ethical approval was obtained from the Ethics Committee of Izmir City Hospital under decision number 2026/43. The study included 18 patients with idiopathic retinopathy and 18 patients with diabetic retinopathy who underwent uncomplicated PPV plus ERM peeling due to ERM.
The medical records of the patients included in the study were reviewed. The inclusion criteria were patients with idiopathic or diabetic ERM who had been clinically diagnosed with unilateral ERM, had undergone surgical removal of the ERM, and had a follow-up period of at least one month. The exclusion criteria were as follows: the absence of intravitreal anti-VEGF or steroid injections due to macular oedema, the presence of uveitis, age-related macular degeneration, significant lens opacities, axial myopia (axial length > 26 mm), severe hyperopia (axial length < 21 mm), vascular retinal diseases, chorioretinal diseases, the absence of panretinal photocoagulation, glaucoma, a history of previous retinal or cataract surgery, and a history of ocular trauma.
Best-corrected visual acuity (BCVA) was measured for all of the patients. A detailed ophthalmological examination was performed, including anterior segment and fundus examinations. All patients underwent a comprehensive ophthalmological examination, which included measurement of the BCVA using the Snellen chart, slit-lamp biomicroscopy, and dilated fundus examination.
To minimize the potential effects of environmental factors and diurnal variation on OCT-derived retinal and choroidal parameters, all OCT examinations were performed at the same time of day (9–12 a.m.). All OCT images were acquired using a Heidelberg Engineering Spectralis spectral-domain optical coherence tomography (SD-OCT) device (Heidelberg Engineering, Heidelberg, Germany). Standard macular OCT scans were performed to analyze the retinal thickness. Images with motion artifacts, segmentation errors, or signal intensity below 20 dB were excluded from the analysis.
The retinal thickness measurements were automatically calculated according to the Early Treatment Diabetic Retinopathy Study (ETDRS) grid protocol. The macula was divided into nine sectors consisting of the central fovea and the upper, temporal, lower, and nasal quadrants of the inner and outer rings. Average retinal thickness values for each sector were recorded separately for statistical analysis. The sectors evaluated were as follows.
Central foveal thickness (CFT)
Superior inner retinal thickness (SIRT)
Temporal inner retinal thickness (TIRT)
Inferior inner retinal thickness (IIRT)
Nasal inner retinal thickness (NIRT)
Superior outer retinal thickness (SORT)
Temporal outer retinal thickness (TORT)
Inferior outer retinal thickness (IORT)
Nasal outer retinal thickness (NORT)
Choroidal vascular index analysis
All OCT images were acquired using the Heidelberg Engineering SA-OCT system. The EDI-OCT mode was used to acquire choroidal images for the CVI analysis.
The CVI was calculated by converting the selected choroidal area into binary images using the ImageJ software (version 1.53j, National Institutes of Health, Bethesda, MD, USA) and was defined as the ratio of the lumen area to the total choroidal area. CVI measurements were performed using the ImageJ software, as previously described by Agrawal et al., and segmentation was performed according to the protocol described by Sim et al. [14–16]. The subfoveal choroidal area between the retinal pigment epithelium and the choroid-scleral junction was manually selected. The selected image was converted to an 8-bit image and binarized using the Niblack automatic local thresholding method. Dark pixels corresponding to the luminal area (LA) were automatically identified. The stromal area (SA) was calculated by subtracting the lumen area from the total choroidal area (TCA) [17]. (Fig. 1) The submacular CVI was calculated as the ratio of LA to TCA. All the measurements were performed independently by two researchers (RB and MC) using the same predefined protocol. To minimise interobserver variability and manual measurement error, both researchers applied identical segmentation criteria and ImageJ settings. The researchers performed the measurements independently and used the mean of the two sets of measurements for statistical analysis.
Fig. 1.
Representative images of the choroidal vascularity index (CVI) measurement process in a diabetic patient (left column) and an idiopathic case (right column). (A) Original EDI-OCT scan of the diabetic eye, with the subfoveal choroidal boundary outlined manually. (B) The same scan converted to an 8-bit image and binarized in ImageJ using the Niblack local thresholding method; the outlined region defines the total choroidal area (TCA). (C) Binarized image after color thresholding, in which the dark pixels represent the luminal area (LA) and the light pixels the stromal area (TCA minus LA). (D-F) The identical sequence in the idiopathic eye: (D) original EDI-OCT scan with the manually outlined choroidal boundary, (E) binarized image defining the TCA, and (F) binarized image showing the LA and the stromal area. CVI was calculated as the ratio of the luminal area to the total choroidal area (CVI = LA / TCA). All measurements were performed on the same horizontal scan centered on the fovea. Scale bars are shown on each panel
Surgical method
All of the surgeries were performed by a single surgeon (RB). Following retrobulbar anesthesia, a 3-trocar 25-gauge vitrectomy was performed using the Constellation Vision surgical system (Alcon Surgical, Fort Worth, TX, USA). Intraoperative triamcinolone was used to induce posterior vitreous detachment following a core vitrectomy. The ERM was made visible using dual-dye. At this stage, only the ERM was peeled using ILM forceps; the ILM itself was not peeled. The peripheral retina was examined and no tears or holes were observed. No patient underwent photocoagulation. A balanced saline solution was used for intraocular tamponade, and the sclerotomies were closed with 7.0 Vicryl to prevent post-operative hypotony.
Statistical analysis
In this study, statistical analyses were performed using IBM SPSS Statistics for Windows, version 20.0 (IBM Corp., Armonk, NY, USA). A post-hoc sensitivity analysis was performed using G Power based on the actual sample size. With 18 patients per group, a two-sided α level of 0.05 and 80% statistical power, the minimum detectable effect size was Cohen’s d = 0.961 for independent-group comparisons and Cohen’s dz = 0.701 for paired preoperative–postoperative comparisons. The normality of continuous variables was assessed using the Shapiro–Wilk test. Normally distributed continuous variables were expressed as mean ± standard deviation (SD), whereas non-normally distributed variables were expressed as median (interquartile range [IQR]). For within-group comparisons of preoperative and postoperative measurements, normally distributed paired data were analyzed using the paired-samples t-test, whereas non-normally distributed paired data were analyzed using the Wilcoxon signed-rank test. For comparisons between the idiopathic and diabetic ERM groups, normally distributed continuous variables were analyzed using the independent-samples t-test, whereas non-normally distributed variables were analyzed using the Mann–Whitney U test. Categorical variables were compared using Fisher’s exact test. A two-sided p value of < 0.05 was considered statistically significant.
Results
A total of 36 patients were included in the study. In the idiopathic group, there were 9 (50%) women and 9 (50%) men; in the diabetic group, there were 8 (44.4%) women and 10 (55.6%) men. The mean age in the idiopathic group was calculated as 71.4 ± 7.3 years, whilst in the diabetic group it was 68.0 ± 6.2 years. There were no significant differences between the groups in terms of gender or age (p = 0.738, p = 0.136). BCVA improved significantly in both groups after surgery, with no significant differences between the groups either preoperatively or postoperatively (Table 1).
Table 1.
Comparison of preoperative and postoperative retinal parameters in patients with idiopathic and diabetic epiretinal membrane
| Parameter | Idiopathic ERM (N = 18) | Diabetic ERM (N = 18) | P values | |||||
|---|---|---|---|---|---|---|---|---|
| Preoperative (Mean ± SD) |
Postoperative (Mean ± SD) |
Preoperative (Mean ± SD) |
Postoperative (Mean ± SD) |
p1 Idiopathic Preop–Postop |
p2 Diabetic Preop–Postop |
p3 Idiopathic Preop vs DiabeticPreop |
p4 Idiopathic Postop vs DiabeticPostop |
|
| BCVA | 0.2 (0.23) | 0.5 ± 0.2 | 0.1 (0.25) | 0.5 ± 0.3 | *0.006ᵃ | *<0.001ᵇ | 0.141ᵈ | 0.843ᶜ |
| CFT (µm) | 469.0 (49) | 369.2 ± 81.0 | 533.4 ± 147.7 | 388.5 (91) | *0.002ᵇ | *0.006ᵇ | *0.027ᵈ | 0.635ᵈ |
| SIRT (µm) | 437.0 (68) | 382.0 (38) | 503.5 (150) | 374.0 ± 36.2 | 0.262ᵇ | *0.001ᵃ | *0.018ᵈ | 0.375ᵈ |
| TIRT (µm) | 416.2 ± 63.1 | 376.5 (21) | 481.5 (378) | 368.5 (48) | 0.176ᵇ | *0.001ᵇ | *0.004ᵈ | 0.150ᵈ |
| IIRT (µm) | 421.5 (82) | 382.5 (52) | 492.0 (274) | 383.0 (36) | 0.321ᵇ | *0.005ᵃ | *<0.001ᵈ | 0.692ᵈ |
| NIRT (µm) | 437.0 (105) | 391.3 ± 57.6 | 461.5 (194) | 402.0 (42) | 0.085ᵇ | *0.004ᵃ | 0.327ᵈ | 0.516ᵈ |
| SORT (µm) | 366.0 (100) | 350.8 ± 45.5 | 369.0 (80) | 322.7 ± 28.0 | 0.394ᵇ | *0.010ᵃ | 0.646ᵈ | *0.032ᶜ |
| TORT (µm) | 323.0 (64) | 327.5 (79) | 369.5 (185) | 358.0 (118) | 0.862ᵇ | 0.118ᵇ | 0.150ᵈ | 0.635ᵈ |
| IORT (µm) | 356.9 ± 55.4 | 355.2 ± 51.9 | 382.5 (157) | 352.0 (166) | 0.933ᵇ | 0.647ᵃ | 0.255ᵈ | 0.334ᵈ |
| NORT (µm) | 368.5 ± 58.3 | 359.5 (34) | 368.0 (113) | 350.5 (35) | 0.979ᵇ | 0.170ᵃ | 0.912ᵈ | 0.110ᵈ |
BCVA, best-corrected visual acuity; CFT, central foveal thickness; SIRT, superior inner retinal thickness; TIRT, temporal inner retinal thickness; IIRT, inferior inner retinal thickness; NIRT, nasal inner retinal thickness; SORT, superior outer retinal thickness; TORT, temporal outer retinal thickness; IORT, inferior outer retinal thickness; NORT, nasal outer retinal thickness; SD, standard deviation; µm, micrometer
ᵃWilcoxonSigned-RankTest; ᵇPairedSample T-Test; ᶜIndependentSamples T-Test; ᵈMann-Whitney U Test
*p < 0.05 (Statistically significant); p1: Idiopathic preop–postop; p2: Diabetic preop–postop; p3: Idiopathic preop vs. Diabetic preop; p4: Idiopathic postop vs. Diabetic postop
CFT decreased significantly after surgery in both groups. Preoperatively, CFT was significantly greater in the diabetic ERM group than in the idiopathic ERM group; however, this difference was no longer significant postoperatively (Table 1).
Among the inner ETDRS sectors, retinal thickness decreased significantly after surgery in all four sectors in the diabetic ERM group, whereas no significant postoperative changes were observed in the idiopathic ERM group. Preoperatively, the superior, temporal, and inferior inner retinal thicknesses were significantly greater in the diabetic ERM group, whereas no significant difference was observed in the nasal inner sector. None of these inner-sector parameters differed significantly between the groups postoperatively (Table 1).
Among the outer ETDRS sectors, a significant postoperative reduction was observed only in the superior outer retinal thickness (SORT) in the diabetic ERM group. SORT also differed significantly between the groups postoperatively. No significant within- or between-group differences were observed in the temporal, inferior, or nasal outer sectors (Table 1).
Subfoveal choroidal thickness (SFCT) did not change significantly after surgery in either group, and no significant intergroup differences were observed. Sattler–choriocapillaris (CC) CVI increased significantly after surgery only in the diabetic ERM group. CVI of the Haller layer did not change significantly within either group but was significantly higher in the diabetic ERM group postoperatively. Total CVI increased significantly after surgery in both groups, with no significant intergroup differences either preoperatively or postoperatively (Table 2).
Table 2.
Comparison of preoperative and postoperative choroidal parameters in patients with idiopathic and diabetic epiretinal membrane (ERM)
| Parameter | Idiopathic ERM (N = 18) | Diabetic ERM (N = 18) | p | |||||
|---|---|---|---|---|---|---|---|---|
| Preop (Mean ± SD) |
Postop (Mean ± SD) |
Preop (Mean ± SD) |
Postop (Mean ± SD) |
p1 Idiopathic Preop–Postop |
p2 Diabetic Preop–Postop |
p3 Idiopathic Preop vs. Diabetic Preop |
p4 Idiopathic Postop vs. Diabetic Postop |
|
| SFCT (µm) | 257.8 ± 35.1 | 249.5 (20) | 255.0 (82) | 268.0 (46) | 0.832b | 0.516b | 0.410d | 0.274d |
| Sattler-CC CVI (%) | 61.0 (13.8) | 72.4 ± 5.9 | 62.0 (3.3) | 66.0 (16) | 0.682b | *0.011 b | 0.256c | 0.150d |
| Haller CVI (%) | 57.1 ± 5.0 | 56.0 (6) | 56.6 ± 5.8 | 59.0 (1) | 0.640ᵃ | 0.051b | 0.782c | *0.024 d |
| Total CVI (%) | 59.9 ± 4.5 | 60.0 (4) | 58.2 ± 4.5 | 61.0 (4) | *0.004 b | *0.002 b | 0.725d | 0.182d |
CVI: Choroidal vascularity index; Sattler-CC: Sattler choriocapillaris layers; SFCT: Subfoveal choroidal thickness; SD: Standard deviation;µm: Micrometer
aWilcoxon signed-rank test; bPaired sample t-test;cIndependent samples t-test;dMann-Whitney U test
*p < 0.05 (statistically significant); p1: Idiopathic preop–postop; p2: Diabetic preop–postop; p3: Idiopathic preop vs. Diabetic preop; p4: Idiopathic postop vs. Diabetic postop
Discussion
In this study, visual, retinal, and choroidal changes in patients who underwent surgery for idiopathic and diabetic ERM were compared in the early postoperative period. In the preoperative period, significant differences were detected between the two groups in terms of CFT, SIRT, TIRT, and IIRT, whereas in the postoperative period, significant differences were observed only in the SORT and the CVI of the Haller layer. In the intragroup evaluation, significant changes were observed in terms of the BCVA and CFT in both groups, while additional significant changes in retinal thickness across several quadrants and in the CVI parameters were detected in the diabetic ERM group. These findings indicated that the early anatomical response after surgery was more pronounced in the diabetic than in the idiopathic ERM cases.
ERM is a condition frequently encountered in the retinal outpatient clinic, and its surgery results in significant changes in macular morphology in the short term. The adjacent choroid is also affected by these changes. In recent years, CVI has emerged as a complementary OCT-based biomarker to choroidal thickness for assessing choroidal vascular status. Therefore, CVI may provide additional information regarding the extent of choroidal involvement in idiopathic and diabetic ERM [8, 12, 13].
In addition to the CVI, several parameters were evaluated in our study, the first of which was the BCVA. Yüksel et al., in their study comparing idiopathic and diabetic ERMs before and after surgery, found that the BCVA significantly improved in both groups at postoperative months 1, 3, and 6 compared with the preoperative period [18]. Similarly, several studies in the literature reported an improvement in BCVA after surgery [19, 20]. In this respect, our study was consistent with the literature: The BCVA significantly improved in the postoperative period in both the idiopathic and the diabetic groups. This finding indicated that surgery effectively improved visual acuity in both groups in the early postoperative period.
However, when the groups were compared with each other, no significant difference was found between the preoperative or the postoperative BCVA values. This finding was inconsistent with most studies in the literature [21–23]. In these studies, the BCVA in the idiopathic ERM group was reported to be significantly better than in the diabetic ERM group. This difference was associated with poorer integrity of the ellipsoid zone (EZ) in the diabetic group compared with the idiopathic group.
As mentioned in the limitations section, the integrity of the EZ was not evaluated for either group in our study; therefore, the similarity in the BCVA between the groups may have been related to the absence of EZ disruption in these patients.
One of the key parameters evaluated in our study was the change in retinal thickness across the nine ETDRS macular sectors. Yüksel et al. found that, in the idiopathic group, retinal thickness decreased significantly at postoperative month 1 in all quadrants except IORT and NORT. In the diabetic group, it decreased in all quadrants [18]. Other studies, however, focused primarily on changes in the CFT and did not pay attention to changes in retinal thickness by quadrant [20, 24]. In our study, the CFT decreased significantly in both the idiopathic and diabetic groups during the early postoperative period. However, this reduction was more pronounced in the diabetic group, where the retinal thickness decreased significantly in all quadrants (CFT, SIRT, TIRT, IIRT, NIRT, and SORT). In the idiopathic group, however, there were no significant quadrant-based changes in retinal thickness in the postoperative period compared with the preoperative period. These findings may have been related to the etiology of diabetic ERM and suggest that retinal thickness in cases of diabetic ERM may be more susceptible to traction caused by ERM, thereby potentially resulting in a more favorable anatomical response after surgery [23]. In addition, postoperative changes in vascular endothelial growth factor (VEGF) levels after ERM peeling may have contributed to the observed reduction in retinal thickness, particularly in the diabetic ERM group [25].
Similarly, in the present study, preoperative CFT, SORT, TIRT, and IIRT values were significantly higher in the diabetic ERM group than in the idiopathic ERM group. This difference may be related to diabetes-associated retinal changes. However, in the early postoperative period, the difference between the groups persisted only in the SIRT sector. These findings suggest that, despite greater preoperative retinal thickness in diabetic ERM eyes, ERM surgery resulted in a comparable reduction in retinal thickness in both idiopathic and diabetic ERM groups during the early postoperative period.
Another important parameter examined in our study was the CVI, which was calculated after the choroid was segmented. D’Aloiso et al. reported that the CVI decreased in the idiopathic group one month after ERM surgery, whereas it increased in the diabetic group [22]. In their study, Hondur et al. also noted that the postoperative CVI decreased in cases of idiopathic ERM [26]. Chen et al., however, reported that postoperative intraocular inflammation increased significantly following phacoemulsification surgery, which may have contributed to the observed changes in choroidal vascularity [27]. Ercan et al. reported that the CVI decreased following macular hole surgery [13].
Since previous cataract or retinal surgery could have affected the CVI in all of the surgical cases evaluated in our study, we excluded patients with a history of such surgery. This exclusion criterion may have enhanced the methodological quality of our study compared with previous studies. In our study, the total CVI increased significantly in the idiopathic group only in the postoperative period compared with the preoperative period, whereas both the Sattler-CC CVI and total CVI increased significantly in the diabetic group. The increase in total CVI in idiopathic ERM and the increase in both Sattler–CC CVI and total CVI in diabetic ERM may be explained by the significant impact of diabetes on both the retina and the choroid, which may contribute to diabetic chorioretinal alterations [28]. Furthermore, vitrectomy can decrease blood flow to the eye and increase oxygen levels in the posterior segment in patients with diabetes [29]. These changes may also affect the CVI.
Although no significant difference was found in the CVI values between the idiopathic and diabetic ERM groups during the preoperative period, during the early postoperative period, the CVI of the Haller layer was significantly higher in the diabetic group than in the idiopathic group. This suggested that, despite having similar preoperative CVI values, idiopathic and diabetic ERMs may respond differently to surgery, with the diabetic group experiencing more pronounced inflammation and associated CVI increases. Finally, in this study, no significant difference was observed in the SFCT in either the preoperative–postoperative comparisons within the groups or the intergroup comparisons. This suggests that CVI may be a more specific imaging biomarker than SFCT for detecting surgery-related choroidal vascular alterations in ERM patients.
This study had several limitations. First, the study’s retrospective nature, the small patient group, and the short follow-up period of only one month are among the main limitations. Additionally, the failure to grade the preoperative ERM in patients, the lack of assessment of the EZ integrity, and the lack of information on the stage of diabetes in the patients with ERM were also significant limitations. Furthermore, the relatively small sample size represents an important limitation of this study and may increase the risk of type II error. The sensitivity analysis indicated that the present sample size was primarily sufficient to detect moderate-to-large or large effects; therefore, smaller true differences may have remained undetected. Consequently, nonsignificant findings should be interpreted cautiously and should not necessarily be considered evidence of equivalence between the groups.
Conclusions
This study observed significant differences in retinal thickness between idiopathic and diabetic ERM patients in the preoperative period, particularly in the CFT and certain sectors. However, these differences disappeared after ERM peeling in the postoperative period, except in the SORT sector. This suggested that, despite the retinal thickness being greater in the diabetic ERM patients preoperatively, ERM surgery may have had a similar effect on retinal thickness in both idiopathic and diabetic ERM patients in the early postoperative period. Moreover, the choroidal structures were similar between the groups in the preoperative period. However, a difference was observed only in the CVI measurement of the Haller layer postoperatively. This suggested that, while the idiopathic and diabetic ERMs had similar CVI values preoperatively, the increase in CVI induced by surgery may have been more pronounced in the diabetic group.
In conclusion, ERM surgery is associated with early visual improvement and a reduction in CFT in idiopathic and diabetic ERM cases. The more widespread changes in retinal thickness and CVI parameters observed in the diabetic ERM group suggested that structural alterations to the retina–choroid complex caused by diabetes may affect the postoperative response. However, supporting these findings require prospective studies with larger sample size sand longer follow-up period than those used for this study.
Abbreviations
- ERM
Epiretinal membrane
- OCT
Optical coherence tomography
- SD-OCT
Spectral-domain optical coherence tomography
- EDI-OCT
Enhanced depth imaging optical coherence tomography
- CVI
Choroidal vascularity index
- BCVA
Best-corrected visual acuity
- ETDRS
Early Treatment Diabetic Retinopathy Study
- CFT
Central fovea lthickness
- SIRT
Superior inner retinal thickness
- TIRT
Temporal inner retinal thickness
- IIRT
Inferior inner retinal thickness
- NIRT
Nasal inner retinal thickness
- SORT
Superior outer retinal thickness
- TORT
Temporal outer retinal thickness
- IORT
Inferior outer retinal thickness
- NORT
Nasal outer retinal thickness
- SFCT
Subfoveal choroidal thickness
- LA
Luminal area
- SA
Stromal area
- TCA
Total choroidalarea
- CC
Choriocapillaris
- VEGF
Vascular endothelial growth factor
- IOP
Intraocular pressure
- ILM
Internal limiting membrane
- SPSS
Statistical Package for the Social Sciences
- µm
Micrometer
Author contributions
All authors have made significant contributions to the reported study, whether in the concept, study design, implementation, data collection, analysis, or interpretation; have participated in the preparation, revision, or critical review of the manuscript; have given final approval of the version to be published; have agreed on the journal to which the article will be submitted; and have accepted responsibility for all aspects of the work.
Funding
The authors have stated that they did not receive any financial support.
Data availability
Data supporting the findings of this study are available upon request from the corresponding author. The data are not publicly available, as they contain information that could compromise the confidentiality of the study participants.
Declarations
Ethics approval and informed consent
This retrospective, cross-sectional study was conducted in accordance with the principles of the Declaration of Helsinki. Local ethical approval was obtained from the Ethics Committee of Izmir City Hospital (2026/43). Informed consent was obtained from all participants.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Gediz BS, Doguizi S, Ozen O, Sekeroglu MA. Is choroidal vascularity index a useful marker in different stages of idiopathic epiretinal membranes? Photodiagnosis Photodyn Ther. 2021;33:102110. [DOI] [PubMed]
- 2.Johnson MW. Posterior vitreous detachment: evolution and complications of its early stages. Am J Ophthalmol. 2010;149:371–e3821. [DOI] [PubMed] [Google Scholar]
- 3.Xiao W, Chen X, Yan W, Zhu Z, He M. Prevalence and risk factors of epiretinal membranes: a Systematic review and Meta-Analysis of Population-Based Studies. BMJ Open. 2017;7(9):e014644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nicholson BP, Zhou M, Rostamizadeh M, Mehta P, Agrón E, Wong W, Wiley H, Nussenblatt R, Sen HN. Epidemiology of epiretinal membrane in a large cohort of patients with uveitis. Ophthalmology. 2014;121(12):2393–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Schechet SA, Devience E, Thompson JT. The effect of internal limiting membrane peeling on idiopathic epiretinal membrane surgery, with a review of the literature. Retina. 2017;37(5):873–80. [DOI] [PubMed] [Google Scholar]
- 6.Miguel AI, Legris A. Prognostic factors of epiretinal membranes: a systematic review. J Fr Ophtalmol. 2017;40(1):61–79. [DOI] [PubMed] [Google Scholar]
- 7.Ikuno Y, Maruko I, Yasuno Y, Miura M, Sekiryu T, Nishida K, Iida T. Reproducibility of retinal and choroidal thickness measurements in enhanced depth imaging and high-penetration optical coherence tomography. Invest Ophthalmol Vis Sci. 2011;52:5536–40. [DOI] [PubMed] [Google Scholar]
- 8.Ruan K, et al. Short-term postoperative changes in the choroidal vascularity index in patients with a unilateral epiretinal membrane. BMC Ophthalmol. 2023;23(1):64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ersan I, Tufan HA, Arikan S, Kara S, Gencer B, Hondur AM. Effect of reduced meal frequency during ramadan fasting on retinal and choroidal thickness. Semin Ophthalmol. 2017;32(4):418–21. [DOI] [PubMed] [Google Scholar]
- 10.Agrawal R, Chhablani J, Tan KA, Shah S, Sarvaiya C, Banker A. Choroidal vascularity index in central serous chorioretinopathy. Retina. 2016;36(9):1646–51. [DOI] [PubMed] [Google Scholar]
- 11.Agrawal R, Ding J, Sen P, et al. Exploring choroidal angioarchitecture in health and disease using choroidal vascularity index. Prog Retin Eye Res. 2020;77:100829. [DOI] [PubMed] [Google Scholar]
- 12.Kim DI, Bae KW, Hwang DDJ. Longitudinal Change of Choroidal Thickness after Pars Plana Vitrectomy for Idiopathic Epiretinal Membrane. J Clin Med. 2022;11(23):6950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ercan ZE, Gokgoz G, Yilmaz G. Choroidal vascularity index changes with phacovitrectomy for vitreoretinal interface disorders. Indian J Ophthalmol. 2022;70(8):2998–3001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Agrawal R, Gupta P, Tan KA, et al. Choroidal vascularity index as a measure of vascular status of the choroid: Measurements in healthy eyes from a population-based study. Sci Rep. 2016;6:21090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Dawn A, Sim PA, Keane H, Mehta S, Fung J, Zarranz-Ventura M, Fruttiger PJ, Patel CA, Egan. Adnan Tufail; Repeatability and Reproducibility of Choroidal Vessel Layer Measurements in Diabetic Retinopathy Using Enhanced Depth Optical Coherence Tomography. Invest Ophthalmol Vis Sci. 2013;54(4):2893–901. [DOI] [PubMed] [Google Scholar]
- 17.Shozo Sonoda T, Sakamoto T, Yamashita M, Shirasawa E, Uchino H, Terasaki. Masatoshi Tomita; Choroidal Structure in Normal Eyes and After Photodynamic Therapy Determined by Binarization of Optical Coherence Tomographic Images. Invest Ophthalmol Vis Sci. 2014;55(6):3893–9. [DOI] [PubMed] [Google Scholar]
- 18.Yüksel K, Karaküçük Y, Özkaya A, Pekel G, Baz Ö, Alagöz C, Yazıcı AT. Comparison of photoreceptor outer segment length in diabetic and idiopathic epiretinal membranes. Eye (Lond). 2015;29(11):1446–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fleissig E, Zur D, Moisseiev E, Keren S, Ohana O, Barak A. Five year follow-up after epiretinal membrane surgery: a single-center experience. Retina. 2018;38:1415–9. [DOI] [PubMed] [Google Scholar]
- 20.Romano MR, Cennamo G, Schiemer S, Rossi C, Sparnelli F, Cennamo G. Deep and superficial OCT angiography changes after macular peeling: idiopathic vs diabetic epiretinal membranes. Graefes Arch Clin Exp Ophthalmol. 2017;255(4):681–9. [DOI] [PubMed] [Google Scholar]
- 21.Romano MR, Ilardi G, Ferrara M, Cennamo G, Allegrini D, Pafundi PC, et al. Intraretinal changes in idiopathic versus diabetic epiretinal membranes after macular peeling. PLoS ONE. 2018;13(5):e0197065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.D’Aloiso R, et al. Choroidal Vascularity Index Fluctuations in Epiretinal Membranes in Vitreoretinal Surgery: Comparison Between Idiopathic and Diabetic Ones. Transl Vis Sci Technol. 2023;12(12):9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ozturk M, Guven D, Kacar H, Karapapak M, Demir M. Functional and Morphological Results of Epiretinal Membrane Surgery in Idiopathic versus Diabetic Epiretinal Membranes. Semin Ophthalmol. 2021;36(5–6):366–72. [DOI] [PubMed] [Google Scholar]
- 24.Rabina G, et al. Epiretinal Membrane in Patients with Diabetic Macular Edema. Ophthalmologica. 2022;245(5):446–54. [DOI] [PubMed] [Google Scholar]
- 25.Rizzo S, Savastano A, Finocchio L, Savastano MC, Khandelwal N, Agrawal R. Choroidal vascularity index changes after vitreomacular surgery. Acta Ophthalmol. 2018;96(8):e950–5. [DOI] [PubMed] [Google Scholar]
- 26.Hondur AM, Aribas YK. Choroidal macrovascular and capillary alterations in eyes with idiopathic epiretinal membranes. Arq Bras Oftalmol. 2024;87(3):e20220369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Chen H, et al. Short-term changes of choroidal vascular structures after phacoemulsifcation surgery. BMC Ophthalmol. 2018;18(1):81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cao J, McLeod DS, Merges CA, Lutty GA. Choriocapillaris degeneration and related pathologic changes in human diabetic eyes. Arch Ophthalmol. 1998;116:589–97. [DOI] [PubMed] [Google Scholar]
- 29.Siegfried CJ, Shui YB. Intraocular oxygen and antioxidant status: new insights on the effect of vitrectomy and glaucoma pathogenesis. Am J Ophthalmol. 2019;Jul:203:12–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data supporting the findings of this study are available upon request from the corresponding author. The data are not publicly available, as they contain information that could compromise the confidentiality of the study participants.