• Users Online: 155
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2017  |  Volume : 18  |  Issue : 3  |  Page : 160-165

Retinal ellipsoid zone/external limiting membrane restoration after diabetic macular edema treatment

Ophthalmology Departement, Kasr Al Ainy Medical School, Cairo University, Cairo, Egypt

Date of Submission31-Jan-2017
Date of Acceptance27-Apr-2017
Date of Web Publication17-Oct-2017

Correspondence Address:
Nehal M. Samy El Gendy
43 Gameat el Dewal AlArabia street, Al Mohandeseen District, Giza
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/DJO.DJO_13_17

Rights and Permissions

Purpose The aim of this study was to determine which clinically significant diabetic macular edema treatment modality results in early restoration of the ellipsoid zone and external limiting membrane (EZ/ELM).
Patients and methods This was a retrospective study where the medical records of patients with type 2 diabetes mellitus and clinically significant diabetic macular edema were reviewed. Patients who had valid scans before treatment and 6 months after treatment were included. Disruption of EZ/ELM was graded and compared. Patients were divided into six groups according to the treatment received: group A received argon laser treatment; group B received micropulse diode laser treatment; group C received antivascular endothelial growth factor (anti-VEGF) treatment; group D received anti-VEGF+laser therapy; group E received triamcinolone acetonide injection (TAI); and group F received TAI+laser therapy.
Results A total of 293 eyes of 200 patients were included (group A: 47, group B: 40, group C: 52, group D: 54, group E: 51, and group F: 49 eyes). Improvement in EZ/ELM integrity was 38.3% in group A, 45% in group B, 59.6% in group C, 37% in group D, 54.9% in group E, and 44.9% in group F. The difference between groups A and B was not statistically significant (P=0.3). The differences between group A and groups C, D, and E were statistically significant (P=0.002, 0.006, and 0.02, respectively). The difference between groups A and F was not significant (P=0.2). The difference between groups C and D was significant (P=0.001), whereas the difference between groups E and F was not significant (P=0.15).
Conclusion Anti-VEGF monotherapy showed the highest percentage of early EZ/ELM restoration. Argon laser alone or argon laser combined with anti-VEGF delayed EZ/ELM healing. Micropulse laser therapy was equivalent to argon laser therapy. TAI results were comparable with anti-VEGF results. Laser addition did not compromise TAI results.

Keywords: diabetic macular edema, ellipsoid zone and external limiting membrane, SD-OCT

How to cite this article:
El Gendy NS. Retinal ellipsoid zone/external limiting membrane restoration after diabetic macular edema treatment. Delta J Ophthalmol 2017;18:160-5

How to cite this URL:
El Gendy NS. Retinal ellipsoid zone/external limiting membrane restoration after diabetic macular edema treatment. Delta J Ophthalmol [serial online] 2017 [cited 2022 Jan 26];18:160-5. Available from: http://www.djo.eg.net/text.asp?2017/18/3/160/216917

  Introduction Top

Macular edema is a major cause of visual deterioration in diabetic retinopathy. Several therapeutic modalities have been investigated, including grid laser photocoagulation [1], vitrectomy [2], or intravitreal injection of antivascular endothelial growth factor (anti-VEGF) [3] or triamcinolone acetonide [4]. The efficacies of these therapies have been evaluated by best-corrected visual acuity and macular thickness measurement using optical coherence tomography. With improved visualization of the retinal architecture offered by Fourier domain optical coherence tomography (FD-OCT), the retinal outer layers including photoreceptors can be evaluated accurately. Several recent studies have highlighted the value of the ellipsoid zone (EZ) integrity in retinal diseases such as retinitis pigmentosa, central serous chorioretinopathy, branch retinal vein occlusion, and macular hole [5],[6],[7],[8].

The association between integrity of EZ and visual acuity in diabetic macular edema has been evaluated in many studies [8],[9],[10]; however, limited information is available about the early effects of different treatment modalities.

In the present study, we tried to figure out which treatment modality results in better early restoration of EZ/external limiting membrane (ELM) integrity. By better understanding the effects of different treatment modalities on retinal microstructures, we can better understand the effects and mechanisms of action of each treatment modality.

  Patients and methods Top

All protocols of the present study conformed to local laws and were compliant with the principles of the Declaration of Helsinki. The study was approved by the Local Ethics Committee of Cairo University. This was a retrospective study in which the medical records and FD-OCT scans of diabetic patients with clinically significant macular edema who received different treatments were reviewed. The reviewed cases were treated by the same physician during 2013–2016. Only cases with valid scans in which the EZ/ELM could be clearly evaluated before treatment and 6 months after treatment were included in the study. Scans were obtained from FD-OCT RTvue (model RT100; Optovue Inc., Fremont, California, USA). Scans were routinely performed using vertical and horizontal 6-mm line scans (1024 axial scans 16 scans for 16 384 total data points averaged to a single scan image). Scans passing through the central fovea were reviewed. Areas where reflectivity signals in the retinal pigment epithelium (RPE) were attenuated by media opacity or hyper-reflective lesions in the inner retinal layers were excluded. Eyes with diffuse macular edema, cystoid macular edema (CME) (with largest cyst <60% of macular thickness: CME I, CME II [11]) were included, whereas eyes with CME III or CME IV [11], serous retinal detachment, posterior hyaloidal traction, and traction retinal detachment were excluded. This was carried out for better homogeneity and comparability between treatment groups. Patients with ischemic macular edema detected as areas of macular capillary nonperfusion by fundus fluorescein angiography were also excluded. Eyes with other conditions that can cause macular thickening, such as venous occlusion or vitreomacular traction, were also excluded. Glaucomatous patients were also excluded. Central subfield thickness (CST) before and after treatment in each treatment modality group was recorded. Jain et al. [12] classified EZ/ELM disruption as follows:
  1. Grade 0: no disruption of ELM or EZ.
  2. Grade 1: ELM disruption with intact EZ.
  3. Grade 2: both EZ and ELM are disrupted.

However, as we were reviewing post-treatment scans, we expected to find cases with regenerated ELM without EZ.

Therefore, we modified grade 1 of Jain et al.’s [12] classification as follows:

Grade 1: intact external limiting membrane or ellipsoid zone

To overcome the variability in reflectivity of EZ by optical coherence tomography, Murakami et al.’s [13] classification was used as a guide, where EZ was classified as follows:
  1. Intact line: highly reflective and continuous (physiological).
  2. Faint line: continuous but with lower reflectivity (still considered intact).
  3. Disrupted line: absent or discontinuous.

Relatively homogenous and continuous ELM was considered intact.

Patients receiving the following treatment modalities were found to be suitable for the present study:
  1. Group A: argon laser grid treatment.
  2. Group B: micropulse diode laser treatment.
  3. Group C: anti-VEGF injections (all patients received at least three injections on a monthly basis).
  4. Group D: anti-VEGF followed by argon laser grid therapy.
  5. Group E: triamcinolone acetonide injection (TAI).
  6. Group F: TAI followed by argon laser grid therapy.

Postvitrectomy cases were excluded as all cases had some degree of vitreomacular traction in the pretreatment scans.

Argon green laser light (OPTO Advant Green Laser; Optos Plc., Dunfermline, Scotland, UK) therapy was routinely performed with 50–100-μm spot size and with an exposure time of 100 ms in our practice. The power is usually adjusted by slowly increasing the laser power until a light gray–white (just visible) burn is obtained. Eyes in the subthreshold micropulse laser group were treated with a 810-nm diode laser with the following parameters: 75–125 μm spot size, 15% duty cycle of 300 ms and power starting with 1000 mW, and if a laser reaction was noted a 200-mW decrease in power was set until no visible mark was seen.

Bevacizumab was the treatment of choice in our practice as the anti-VEGF as it was economical. Patients received bevacizumab (Avastin, Genentech; Roche, Basil, Switzerland) 1.25 mg/0.05 ml monthly for 3 months and then received pro re nata (PRN). TAIs are administered more frequently in developing countries such as Egypt (compared with intravitreal implants) owing to economic causes too. Cases receiving triamcinolone acetonide (Kenalog) 0.1-ml 4 mg/ml (Triesence/Trivaris is alcohol-free preparation that is FDA approved for intraocular use) injections were included. Cases undergoing argon laser photocoagulation were also included.

Statistical analysis

Data management and analysis were performed using Excel 2010 software (Microsoft Corp., Redmond, Washington, USA) statistical analysis system. Numerical data are expressed using means and SDs or medians and ranges. Categorical data are expressed as percentages. Comparisons between the two groups with respect to numerical variables were carried out by the Student’s t test. The χ2 test was used to compare between groups with respect to categorical data. All P-values were two-sided. P-values less than 0.05 were considered statistically significant.

  Results Top

The medical data of 318 eyes of 225 patients with type 2 diabetes mellitus and diabetic maculopathy indicated for treatment according to the early treatment diabetic retinopathy study (ETDRS) study were reviewed, from 2013 to 2016. Twenty-five eyes were excluded: 12 eyes because of bad image quality, six eyes because of lack of complete medical data, and six eyes because of a short follow-up period. Medical records of 293 eyes of 200 patients were thus included in the present study, with 47 eyes in group A, 40 eyes in group B, 52 eyes in group C, 54 eyes in group D, 51 eyes in group E, and 49 eyes in group F.

The patients’ age ranged from 49 to 73 years with a mean±SD of 58.4±5.8 years. Duration of diabetes ranged from 9 to 20 years with a mean±SD of 12±5.2 years. Patient demographic data for each group are illustrated in [Table 1]. Average CST changes are shown in [Table 2]. Percentage of patients in each grade of EZ/ELM integrity grading at baseline are shown in [Table 3]. EZ/ELM grading changes after treatment are shown in [Table 4]. Improvement in EZ/ELM integrity was found in 38.3% of eyes in group A, 45% in group B, 59.6% in group C, 37% in group D, 54.9% in group E, and 44.9% in group F.
Table 1 Demographic data of patients in groups A, B, C, D, E, and F

Click here to view
Table 2 Central macular thickness, before treatment and after treatment, and % change

Click here to view
Table 3 Percentage of patients in each grade of ellipsoid zone and external limiting membrane integrity at baseline

Click here to view
Table 4 Changes in ellipsoid zone and external limiting membrane grading after treatment

Click here to view

The difference in the percentage of cases with EZ/ELM improvement between groups A and B was not statistically significant (P=0.3). The differences between group A on the one hand and groups C, D, and E on the other were statistically significant (P=0.002, 0.006, and 0.02, respectively). However, the difference between groups A and F was not statistically significant (P=0.2). The difference between groups C and D was significant (P=0.001), whereas the difference between groups E and F was not statistically significant (P=0.15). Comparing groups C and E, the difference was not statistically significant (P=0.5). The difference between groups D and F was also not significant (P=0.5).

Most of the cases in group A received only one laser treatment session, with only two cases requiring another session; this was performed 3 months after the initial session (average±SD=1.04±0.2 laser sessions). Only four cases in group B required another treatment session (average±SD=1.1±0.3). This was also performed 3 months after the initial treatment.

Group C received an average±SD of 3.14±0.5 bevacizumab injections, whereas group D received an average±SD of 3.2±0.5 bevacizumab injections.

Group E received an average±SD of 1.25±0.6 injections. Group F received a comparable number of injections with group E (1.24±0.5).

EZ/ELM evaluation is shown in [Figure 1].
Figure 1 (a) Pretreatment scan of a case from group E, Fourier domain optical coherence tomography (FD-OCT), showing intact ellipsoid zone (EZ) but not external limiting membrane (ELM) layers as well as grade II cystoid macular edema. Note areas of very faint lines of EZ correspond to inner retinal exudates. (b) FD-OCT of group C, pretreatment scan, showing diffuse macular edema and grade 2 EZ/ELM. (c) A case from group C after treatment. Note the almost complete restoration of ELM, whereas EZ is still faint at the center.

Click here to view

  Discussion Top

The present study showed that cases that received argon laser therapy either alone or preceded by intravitreal bevacizumab injections were associated with the lowest percentage of EZ/ELM restoration within the first 6 months of treatment. There was no statistically significant difference between the argon laser group and the micropulse therapy group. Intravitreal injection of bevacizumab alone showed the highest percentage of EZ/ELM restoration, followed by TAI. However, the difference was not statistically significant. Whether or not intravitreal injections of anti-VEGF or steroids have a toxic effect on retinal tissue, resulting in delayed healing, is not clear. Among all, which treatment modality is less injurious to the outer retinal layers is also not yet known. Long-term results of different treatment modalities depend on many factors. On the other hand, short-term results may be more related to the mode of treatment and its mechanical and chemical insult on the retinal layers.

The correlation between visual acuity and integrity of photoreceptor markers (EZ/ELM complex) was clarified by previous studies [8],[9],[10]. In the present study, EZ/ELM integrity was not correlated with visual acuity. On the other hand, we were interested in knowing which treatment modality is linked to early restoration of the EZ/ELM complex. Most of the previous studies reported results 12 months after treatment.

Jain et al. [12] demonstrated for the first time that in diabetic maculopathy ELM disruption occurs first followed by EZ. This was based on the study conducted by Omri et al. [14], who considered ELM to be a third retinal barrier, and once interrupted fluid accumulation occurs in the macula.

Whether ELM acts as a scaffold for EZ to regenerate or it acts as a barrier that protects EZ from toxins or fluid accumulation is not clear. In either theory, it is logical to think that ELM has to be restored first for EZ to regenerate. In such cases, ELM was expected to be seen in recovering cases to be intact before the EZ. That is why we modified Jain et al.’s [12] classification to fit to the postoperative evaluation in the present study. In degenerative retinal diseases, on the other hand, it is believed that disorganization seems to occur in a stepwise manner: where the EZ is affected before the ELM [15],[16],[17].

Large intraretinal cysts seem to mechanically compress ELM/EZ layers [13]. To have a more fair comparison between the six groups, we excluded cases with large cysts as classified by Helmy and Atta Allah [11]. Diffuse retinal edema may be also associated with some sort of outer retinal layer compression and decreased thickness. This may explain the good results achieved by TAI compared with other treatments. The rapid and more dramatic decompression of the outer retinal layers may contribute to this. Cases indicated for TAI are usually associated with high pretreatment thickness. Knowing the antiedematous effects of steroids, and the higher reduction in retinal thickness after, may explain this aspect.

The basis of the micropulse laser is that it allows some degree of heat dispersion during the treatment. This allows more energy to be applied to the RPE with less collateral damage to the outer retina (compared with argon laser). In the present study, there was no difference between the argon laser-treated group (38.3%) and the micropulse laser group (45%). Figueira et al. [18] reported equal effectiveness of both treatment modalities with regard to central macular thickness (CMT) and visual acuity (VA) 12 months after treatment, but they did not address early photoreceptor marker changes. To the best of our knowledge, the present study is the first study comparing early EZ/ELM restoration between these two treatment modalities. The on and off energy delivery during micropulse laser treatment allows more energy to be delivered to the RPE without damaging nearby layers such as EZ. However, the present study did not support this aspect. This may indicate that ELM restoration (which is less expected to be affected by laser therapy, compared with EZ layer) is more important than EZ insult during treatment. It may also be explained by the fact that we reviewed scans 6 months after treatment: may be micropulse therapy has an upper hand earlier directly after laser shots insults. In addition, it may also indicate that the initial effect of EZ is equal in both treatment modalities.

It is not clear whether intravitreal injections have a neuroregenerative effect as claimed by Mori et al. [19] or whether it is the early regression of the edema that allows early restoration of the EZ/ELM complex. Our results did not contradict Mori et al.’s [19] findings, as the highest percentage of EZ/ELM restoration was recorded in eyes treated with bevacizumab. However, this claimed neuroprotective result seems to be compromised greatly by laser burns. On the other hand, the triamcinolone only group showed comparable results with the anti-VEGF group, showing that the antiedema and anti-inflammatory effects of steroids compete with the neuroprotective effect of anti-VEGF. It is to be noted that adding laser therapy to TAI did not compromise the results as it did with anti-VEGF. This could be explained by the anti-inflammatory effect of steroids that could have regulated the photocoagulation insult. In addition, it could be explained by the fact that eyes with combined treatments (groups D and F) received their laser session 1 month after the last injection. All cases receiving anti-VEGF treatment had at least three injections (average±SD=3.2±0.5), whereas those that received TAI had an average±SD of 1.24±0.5 injections. Thus, post-treatment scans were closer in time to the laser session performed in group D compared with group F.CST reduction may have enhanced the early effect on outer retinal layer restoration, as cases with highest reduction of CST were associated with good restoration of EZ/ELM (groups E and D).

The present study is limited by its retrospective nature. Controlled, randomized, prospective studies are recommended for better evaluation of different treatment modalities with regard to EZ/ELM restoration. A monthly comparison of EZ/ELM changes may better explain the effects of different therapies. Hyper-reflective foci in the inner retina may compromise interpretation of outer retinal layer changes. Hyper-reflective foci are usually more common with higher retinal edema. TAI cases were associated with the highest pretreatment CST (the difference was statistically not significant). Artifacts in the form of disrupted outer layers may occur. Thus, in groups E and F, the restoration of EZ/ELM layers may be due to underestimated integrity of these layers before treatment. However, while interpreting pretreatment scans, we considered this fact. Jain et al.’s [12] classification was chosen as it does not depend on the exact length of disruption, a factor that we found more suitable for diabetic macular edema evaluation (compared with degenerative retinal diseases).

EZ/ELM restoration may reflect healing. On the other hand, deteriorated cases may reflect a direct injury or toxicity more. Groups A, D, and F showed the highest numbers of deteriorated cases (two, three, and three, respectively) compared with zero cases in groups B and C and only one case in group E.

In conclusion, the present study showed that bevacizumab as monotherapy is associated with the highest rate of EZ/ELM restoration. It also showed that edema control by TAIs enhanced outer retinal layer restoration. In addition, it highlighted a potential, less-injurious effect of anti-VEGF compared with laser therapy (at least early). TAI may have a protective effect in reducing argon laser burns. It showed that micropulse laser treatment was not superior to argon laser treatment. Bevacizumab may have a neuroprotective effect; however, this effect is markedly compromised by adjuvant laser therapy.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Bresnick GH. Diabetic maculopathy: a critical review highlighting diffuse macular edema. Ophthalmology 1983; 90:1301–1317.  Back to cited text no. 1
Lewis H, Abrams GW, Blumenkranz MS, Campo RV. Vitrectomy for diabetic macular traction and edema associated with posterior hyaloidal traction. Ophthalmology 1992; 99:753–775.  Back to cited text no. 2
Massin P, Bandello F, Garweg JG, Hansen LL, Harding SP, Larsen M et al. Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE study): a 12-month, randomized, controlled, double-masked, multicenter phase II study. Diabetes Care 2010; 33:2399–2405.  Back to cited text no. 3
Martidis A, Duker JS, Greenberg PB, Rogers AH, Puliafito CA, Reichel E et al. Intravitreal triamcinolone for refractory diabetic macular edema. Ophthalmology 2002; 109:920–927.  Back to cited text no. 4
Oishi A, Otani A, Sasahara M, Kojima H, Nakamura H, Kurimoto M et al. Photoreceptor integrity and visual acuity in cystoid macular oedema associated with retinitis pigmentosa. Eye (Lond) 2009; 23:1411–1416.  Back to cited text no. 5
Monnet D, Levinson RD, Holland GN, Haddad L, Yu F, Brézin AP. Longitudinal cohort study of patients with birdshot chorioretinopathy. III. Macular imaging at baseline. Am J Ophthalmol 2007; 144:818–828.  Back to cited text no. 6
Sayanagi K, Sharma S, Kaiser PK. Photoreceptor status after antivascular endothelial growth factor therapy in exudative age-related macular degeneration. Br J Ophthalmol 2009; 93:622–626.  Back to cited text no. 7
Sakamoto A, Nishijima K, Kita M, Oh H, Tsujikawa A, Yoshimura N et al. Association between foveal photoreceptor status and visual acuity after resolution of diabetic macular edema by pars plana vitrectomy. Graefes Arch Clin Exp Ophthalmol 2009; 247:1325–1330.  Back to cited text no. 8
Maheshwary AS, Oster SF, Yuson RM, Cheng L, Mojana F, Freeman WR. The association between percent disruption of the photoreceptor inner segment-outer segment junction and visual acuity in diabetic macular edema. Am J Ophthalmol 2010; 150:63–67.  Back to cited text no. 9
Abdollahi A, Esshghabadi A, Faghihi H, Mirshahi A. The relationship between central macular photoreceptor status and final visual acuity in resolved diabetic macular edema by nonsurgical treatment. J Clin Exp Ophthalmol 2011; 2:157.  Back to cited text no. 10
Helmy YM, Atta Allah HR. Optical coherence tomography classification of diabetic cystoid macular edema. Clin Ophthalmol 2013; 7:1731–1737.  Back to cited text no. 11
Jain A, Saxena S, Khanna VK, Shukla R, Meyer C. Status of serum VEGF and ICAM-1 and its association with external limiting membrane and inner segment outer segment junction disruption in type 2 diabetes mellitus. Mol Vis 2013; 19:1760–1768.  Back to cited text no. 12
Murakami T, Nishijima K, Akagi T, Uji A, Horii T, Ueda-Arakawa N et al. Optical coherence tomographic reflectivity of photoreceptors beneath cystoid spaces in diabetic macular edema. Invest Ophthalmol Vis Sci 2012; 53:1506–1511.  Back to cited text no. 13
Omri S, Omri B, Savoldelli M, Jonet L, Thillaye-Goldenberg B, Thuret G et al. The outer limiting membrane (OLM) revisited: clinical implications. Clin Ophthalmol 2010; 4:183–195.  Back to cited text no. 14
Hagiwara A, Mitamura Y, Kumagai K, Baba T, Yamamoto SY. Photoreceptor impairment on optical coherence tomographic images in patients with retinitis pigmentosa. Br J Ophthalmol 2013; 97:237–238.  Back to cited text no. 15
Aizawa S, Mitamura Y, Hagiwara A, Sugawara T, Yamamoto S. Changes of fundus autofluorescence, photoreceptor inner and outer segment junction line, and visual function in patients with retinitis pigmentosa. Clin Experiment Ophthalmol 2010; 38:597–604.  Back to cited text no. 16
Wakabayashi T, Oshima Y, Fujimoto H, Murakami Y, Sakaguchi H, Kusaka S et al. Foveal microstructure and visual acuity after retinal detachment repair: imaging analysis by Fourier-domain optical coherence tomography. Ophthalmology 2009; 116:519–528.  Back to cited text no. 17
Figueira J, Khan J, Nunes S, Sivaprasad S, Rosa A, de Abreu JF et al. Prospective randomised controlled trial comparing sub-threshold micropulse diode laser photocoagulation and conventional green laser for clinically significant diabetic macular oedema. Br J Ophthalmol 2009; 93:1341–1344.  Back to cited text no. 18
Mori Y, Suzuma K, Uji A, Ishihara K, Yoshitake S, Fujimoto M et al. Restoration of foveal photoreceptors after intravitreal ranibizumab injections for diabetic macular edema2016. Sci Rep 2016; 6:39161.  Back to cited text no. 19


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4]

This article has been cited by
1 Short-term outcomes of intravitreal dexamethasone in relation to biomarkers in diabetic macular edema
Shilpi Harshal Narnaware,Prashant Keshao Bawankule,Dhananjay Raje
European Journal of Ophthalmology. 2021; 31(3): 1185
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Patients and methods
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded153    
    Comments [Add]    
    Cited by others 1    

Recommend this journal