Article Text

Deep lamellar keratoplasty with complete removal of pathological stroma for vision improvement
  1. Department of Ophthalmology, Ichilov Hospital, 6 Weitzman Street, Tel-Aviv 64239, Israel
    1. Eye Clinic, Southend Hospital, Prittlewell Chase, Westcliff on Sea, Essex SS0 0RY
      1. J SUGITA
      1. Sugita Eye Hospital, 5-1-30 Sakae
      2. Naka-ku, Nagoya 460, Japan

        Statistics from

        Editor,—We read with great interest the article by Sugita and Kondo and the editorial by Aggarwal.1 2 We congratulate the authors on this paper which can contribute to the revival of lamellar keratoplasty. The authors report on deep lamellar keratoplasty in which the stroma is excised to the extent that only Descemet’s membrane remained. We wish to comment that we have previously reported on a similar procedure.3-5 We used a full thickness graft, since in the eyes operated by us the endothelium was compensated and graft clarity was maintained by the donor’s corneal endothelium. It seems to us that deep lamellar keratoplasty with a full thickness graft can also be a useful procedure in cases with decompensated endothelium.



        Editor,—I am pleased by the interest and debate generated by the editorial and also the paper by Sugita and Kondo.

        As mentioned in the editorial the concept of deep lamellar keratoplasty is not new. Its most attractive feature is almost complete eradication of the risk of graft rejection (as with lamellar keratoplasty) together with the potential of good visual acuity (as with penetrating keratoplasty).

        Patients most likely to benefit from such a procedure are thus those with normal functioning endothelium but a diseased stroma. The problem to date has been (and still remains) that in such cases dissection to Descemet’s membrane is time consuming and technically relatively difficult. Many have tried to improve the technique, including using air to help identify and separate the deep stromal fibres from Descemet’s membrane. Archila in 1985, Price in 1989, and Chau et al in 1992.

        The paper by Sugita and Kondo was interesting because they describe another technique to dissect the deep stroma from Descemet’s membrane. They showed excellent results in a large series of cases using this technique.

        Loewenstein and Lazar in their letter refer to their “Intentional retention of Decemet’s membrane in keratoplasty for the surgical treatment of bullous keratopathy”. This is indeed an interesting concept.

        In bullous keratopathy as the stroma is already hydrated it is much easier to identify and separate the deep stroma from Descemet’s membrane.

        As they themselves state in their paper, for the graft to function microperforations in Descemet’s membrane are necessary and the production of a secondary aqueous filled chamber. I suspect that the donor endothelium is hence exposed to the recipient’s immunological system in the same way as in penetrating keratoplasty, and hence the main advantage of deep lamellar keratoplasty (reducing the risk of rejection) is also probably lost. In all their patients Decemet’s membrane opacified to some degree and this must reduce the final visual outcome. Indeed the best visual outcome they report is 6/60.

        The only advantage of this procedure over penetrating keratoplasty, in bullous keratopathy, is that the anterior chamber is not completely opened (only by microperforations). As the authors state this may be an advantage in some eyes where one does not wish to open the anterior chamber completely.



        Editor,—Loewenstein and Lazar have proposed that deep lamellar keratoplasty (DLK) using a full thickness graft with intentional opening of the central Descemet’s membrane in the host recipient lamellar bed creating a double anterior chamber can be useful in the management of bullous keratopathy.

        I have had three cases of persistent double anterior chamber. In each case a fairly large accidental perforation of Descemet’s membrane occurred during the DLK procedure. In each case, a full thickness donor cornea with presumed functioning endothelium was used, although donor endothelial quality had been judged marginal before surgery. All three grafts have remained clear for 1–4 years and a second anterior chamber in each case has persisted. These cases were not included in our series, the recipients did not have known endothelial disease, and the management of endothelial disease was not considered an indication for DLK. It is likely that the construction of an appropriately sized defect in the host Descemet’s membrane for their proposed procedure may be difficult, the objective of a second anterior chamber may not always be realised, and if not it is likely that much of the donor endothelium will die. However, it is intriguing to consider whether the donor endothelium in such a second anterior chamber may be protected from possible deleterious effects of aqueous streaming compared with donor endothelial cells facing the anterior chamber in conventional penetrating keratoplasty–an analogy would be the shore within a protected bay compared with the shore facing an ocean.

        We believe that what Loewenstein and Lazar have proposed is conceptually different from DLK as discussed in our paper. Whether their procedure is a reasonable alternative to penetrating keratoplasty for the routine management of corneal endothelial disease or for selected cases of endothelial disease where minimal intraocular surgery is desired awaits the test of time.

        View Abstract

        Request permissions

        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.