Dear Editor,

We read with great interest the paper by Senoo et al1 and we would like to draw your attention to a complication of the approach described. Some years ago, relying on our experience in deep sclerectomy, we tried a similar approach in two cases: we first performed a double scleral flap exposing Descemet membrane at limbus and then started the dissection. Although in both cases we had a good exposure of Descemet's membrane at surgery without perforation, in one patient we observed an unusual complication on the first postoperative day: blood was present between the donor cornea and the recipient Descemet's membrane marking a channel from the sclerotomy site. We had to reopen the scleral flaps and wash out the blood in the interface by introducing air in the anterior chamber. The patient did not suffer any further complications after this treatment.

We discontinued this approach not only for the possibility of this complication but also because in our experience the time required to prepare the limbal access is equal to that of a classical mechanical procedure. The high percentage of double anterior chamber reported for the limbal approach (45%) is another point worth consideration. Since we observe a rate of less than 3% with mechanical dissection, the suspecion arises that the higher rate reported for the new technique may be due to either viscoelastic persistence in the interface or filtration through trabeculodescemetic membrane at the sclerotomy site: in the presence of an intact descemetic membrane and functioning endothelium, percolation of aqueous humor at the sclerotomy site may maintain a double chamber. It would be of interest to verify this hypothesis with ultrasound biomicroscopy.

Finally, we wonder why the author did not try to resolve the long persistence of a double chamber (often a sign of unrecognized perforation) by introducing air in the anterior chamber. In this procedure, the risk of a pseudo Urrets-Zavalia syndrome2 may be controlled by both avoiding overinflation of the chamber and strictly monitoring the patient over hours.

Marco Nardi, Gianluca Guidi, Marino De Luca
Neuroscience Department, University of Pisa, Pisa, Italy

Correspondence to:
M Nardi MD
Pza Varanini 2
55100 Lucca Italy
marco.nardi@med.unipi.it

References:

1. Senoo T, Chiba K, Terada O et Al. Deep lamellar keratoplasty by deep parenchima detachment from the corneal limbs. Br J Ophthalmol 2005; 89:1597-1600.

2. Maurino V, Allan BD, Stevens JD, Tuft SJ. Fixed dilated pupil (urrets-Zavalia syndrome) after air/gas injection after deep lamellar keratoplasty for keratoconus. Am J Ophthalmol 2002; 133(2): 266-8.