New techniques for the surgical treatment of keratoconus

Until recently, penetrating keratoplasty (PK) was the gold standard surgical treatment for keratoconus; visual and refractive outcomes are often good, but the risk of endothelial rejection associated with this technique remains. Because keratoconus patients are often young, surgeons are concerned about exposing them to the lifelong immunologic risk, which may require both therapeutic and prophylactic steroids for relatively long periods, possibly inducing glaucoma and cataract formation. This has caused some to attempt to find other ways of treating their patients.

During the last decade several techniques of lamellar keratoplasty (LK) have been revisited with the purpose of retaining the advantages of PK surgery while avoiding removal of healthy endothelium from the recipient cornea. However, hand-dissection of the deep stromal layers is technically difficult and the quality of the surfaces obtained is seldom compatible with 20/20 vision. To improve the visual results, some authors have used fluid or air ("big bubble" technique) to separate the whole corneal stroma from Descemet's membrane. This technique is particularly difficult to learn and, even in the hands of experienced surgeons, is complicated by micro- or macro-perforation in a relatively high number of cases, making conversion to PK necessary.

One particular technique, microkeratome-assisted LK for keratoconus, has, in my opinion, the advantage of being a standardized, technically easy procedure, which yields smooth dissected surfaces compatible with 20/20 vision.

It should be noted that PK and LK surgery do not always yield satisfactory results. For example, a standard 8 mm penetrating or lamellar graft will completely encompass a small, nipple-like cone located centrally, whilst a broad cone located more peripherally will be partly spared by conventional surgery. In this case, the irregular shape of the peripheral recipient rim left in place may give rise to an irregularly shaped cornea postsurgery.

In order to avoid this undesirable outcome, grafts that are undersized by 0.25 to 0.5 mm should be used; once the graft is sutured into the recipient bed under tension, the cone is flattened and a normal corneal shape is restored.

The risk of recipient bed resistance to the flattening effect of a lamellar graft and the threat of corneal bulging once sutures are removed are kept to a minimum by removing as much recipient cornea tissue as possible. Caution must, however, be exercised with microkeratome-assisted keratectomies as they are not finely titratable, hence to minimize the risk of intraoperative perforation, a minimal recipient bed thickness of approximately 100 μm must be retained.

As published in a recent report (Ophthalmology, June 2005), results of LK for keratoconus with undersized grafts compare favourably with those of PK surgery. As early as one year after surgery, the interface between donor graft and recipient bed is barely visible even at high magnification of the slit-lamp. Best-spectacle corrected visual acuity (BSCVA) is 20/40 or better in nearly all patients and refractive astigmatism is within 4 diopters (D) in most cases.