Statistics from Altmetric.com
Staining of the ILM in macular surgery
We read with interest the article by Li and colleagues about trypan blue staining of the vitreomacular interface during vitrectomy.1 We congratulate the authors on their work. In particular, we appreciate their critical approach of testing trypan blue for staining of the internal limiting membrane (ILM) and epiretinal membrane (ERM) as well as their comments on potential untoward effects of indocyanine green (ICG) in macular surgery. We would like to comment on two remarks concerning the ultrastructural findings on the retinal side of the ILM following ILM removal with and without the use of ICG.
We agree with Li and colleagues that fragments of glial cells are commonly found in ILM specimens. Ultrastructurally, they appear as tiny fragments of Müller cell membranes adherent to and enclosed within the undulations of the retinal side of the ILM. These glial structures had been described in detail by Eckhardt and colleagues, and are in accordance with previous work of our group, in an investigation of the ultrastructure of the vitreomacular interface of 93 specimens in 91 consecutive patients with macular holes, epiretinal membranes, diffuse diabetic macular oedema, and vitreomacular traction syndrome without the use of indocyanine green or other dyes (unpublished data).2
We also agree with the authors, that the surgical technique and the underlying disease may influence the amount of glial structures adherent to the retinal side of the ILM, as these structures are predominantly found within undulations and folds of the ILM.3,4 However, we would like to emphasise the effect of ICG in this context.
Firstly, there are obvious differences between ILM specimens removed with and without the use of ICG not only in terms of quantity of glial structures but in terms of quality. A continuous layer of cell membranes, undetermined cellular debris, and entire footplates of Müller cells were commonly observed following ICG assisted peeling of the ILM, whereas such structures had never been found in the series of 93 unstained specimens described above.5
Secondly, all stained and unstained specimens having been investigated by electron microscopy were removed by one experienced surgeon (AK). Beside the use of ICG, there was no change of the surgical technique. Moreover, retinal elements as described above were not found before the introduction of the dye at our institution in September 2000, nor after having stopped ICG staining in April 2001.
Thirdly, in an experimental setting in human donor eyes published recently, retinal structures adherent to the undulating side of the ILM as described above could be found following the application of ICG to the macula only.6 No attempt of peeling or any other mechanical approach to the vitreomacular interface was made in these eyes. However, the ILM was detached from the macula. Retinal elements were adherent to the retinal side of the ILM showing an identical morphology like those obtained during vitrectomy with ICG assisted ILM removal.6
Therefore, in our experience there is increasing evidence that at least some commonly used preparations of ICG may affect the ultrastructure of the inner retina, and are primarily responsible for obvious differences in the ultrastructure of the surgically removed ILM. ILM removal by itself results in removal of tiny fragments of Müller cell membranes. Their morphological and functional implications for the macula remain unknown.
Finally, we would like to encourage the authors to follow their promising approach of staining the ILM and ERM with trypan blue. In our institution, single specimens which had been stained and peeled using trypan blue revealed no evidence of retinal damage (submitted data).
We wish to thank Gandorfer and colleagues for their interest in our paper1 and for the kind comments and encouragement with regard to our work. Certainly, these correspondents are compiling evidence concerning the effect of indocyanine green (ICG) on the retina in both their published and unpublished studies.2,3
In our report, we restricted our comments regarding retinal damage and dye usage to the specimens wherein an epiretinal membrane (ERM) was present.1 Evidence of retinal damage was observed in four of these five specimens, mostly in the form of neural and glial elements adherent to the retinal side of the inner limiting membrane (ILM). The apparent lack of such elements in some ERM specimens may reflect partial separation of the ILM due to traction from the membrane before surgery.4 Nevertheless, in one of our specimens a substantial fragment of neuroretina was also present.1 We have, indeed, long considered such fragments as potential confounding factors in immunohistochemical studies of surgically excised ERMs.5,6 Since they are present in ERMs removed without the use of dye, we cannot blame their presence on these surgical aids. Perhaps these fragments are avulsed as a result of enhanced adhesion between the ERM and retina via glial anchorage sites running through dehiscence in the ILM.5 It is clear that our investigation does not exclude an effect of ICG on the retina2,3 and we wholeheartedly agree with Gandorfer and coworkers that agents such as trypan blue warrant further evaluation as aids to ERM and ILM peeling.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.