Importance of Axial Length and Functional Corneal Endothelial Cells in Descemet Membrane Endothelial Keratoplasty
We read with great interest the article by Hayashi et al. The authors have noted that uneventful graft insertion into an anterior chamber is the most important determinant of successful surgery during Descemet membrane endothelial keratoplasty (DMEK), especially in small eyes with a shorter axial length and a shallow anterior chamber.1 We have similar findings comparing endothelial cell density (ECD) with visual acuity, highlighting the importance of axial length and functional corneal endothelial cells in DMEK. There are no studies, to our knowledge, reporting a direct correlation of axial length or functional endothelial cells with endothelial cell loss (ECL) and long-term graft survival after DMEK. Moreover, most of the publications show the correlation and success rate of DMEK based on endothelial cell survival and less complications in the postoperative follow-up period. Hence, we set out to correlate ECL with axial length of the eye and ECD with visual acuity.
This retrospective study was approved by the Institutional Ethical Review Board of Riga Stradins University (acceptance n. E-9(2) decision n.29/29.09.2015). Twenty-five patients with pseudophakic bullous keratopathy between January 2016 and March 2017 in Riga, Latvia, were treated with DMEK. The tissues were prepared using the stripping method as previously described by Dapena et al2 and were implanted by injecting a DMEK roll in the recipient eye. All patients were treated for 1 week with topical fluoroquinolone (Vigamox; Alcon, Fort Worth, TX) drops in addition to a 1-month course of 1% prednisolone acetate (Allergan, Irvine, CA) 4 times a day after surgery. ECD was measured using a specular microscope (SP-3000P; Topcon Corp.) at 1, 3, and 6 months after surgery. The axial length was measured using the IOLMaster 500 (Zeiss, Germany). Data are presented as mean (±SD). The calculated probability values were 2 tailed, and data showing P <0.05 were deemed to be statistically significantly different. Statistical analyses were performed using IBM SPSS, version 23.0.
The average age of the donors was 59.7 ± 6.8 (mean ± SD) years, and the mean ECD before stripping was 2710 ± 102 cells/mm2. No occurrence of graft rejections was observed. The relationship between ECL and visual improvement over the first 6 months showed no strict correlations. Visual acuity did not improve or reduced with a decrease in ECD over 6 months (P > 0.05), resulting in a speculation that the function of each cell is more important than the overall number of cells present in DMEK. Dapena I et al3 found that the learning curve in DMEK did not correlate with clinical outcomes (best-corrected visual acuity and ECD), but rather with the presence of a functional graft. From our clinical experience and evidence over the years, we believe that it is not merely the number of cells that contributes to clarity of the cornea, but the function of residual cells present in the transplanted graft. Improvements and changes in visual acuity are not directly linked to reduction of cells, and there is no direct or logarithmic correlation that was found. The evidence also suggests that the final improvements in vision are not linked to the vision before surgery, suggesting that replacement of the endothelium with new donor tissue in a cloudy cornea will give anatomical and optical replacement of previous function even if advanced decompensation is present. Furthermore, clinical consultations and statistical data show that the clarity and function of the cornea, referred by the patient, are not strictly linked to the number of endothelial cells. It supports the hypothesis that to maintain the corneal stroma dry, the function of every single endothelial cell present is more important than the total number of cells.