We read with interest the article “Changes in corneal endothelial cells after Ahmed glaucoma valve implantation: 2-year follow-up” by Lee and associates. The authors conclude that the corneal endothelial cell loss increased with time: 15.3% at 12 months and 18.6% at 24 months after surgery, on average, in all measured areas. These data and the conclusion would have serious implications on the management decisions in refractory glaucoma and need to be put in perspective in the context of the results from the TVT study as well as the increasing enthusiasm for glaucoma drainage device as a primary surgery for primary glaucoma.

The authors do concede that there might be inborn errors in endothelial cell count as well as the inability to count cells in the same area that could potentially affect their conclusions. We would like to highlight some of the issues in the endothelial cell count as reported in this study. These considerations might limit the ability to extrapolate the conclusions of this study.

On reviewing previous publication by the authors, it appears that the authors used a fixed-frame method (dot method) for endothelial cell counts and assessment of other parameters. The fixed-frame method is inherently associated with more errors. In addition, there are sampling errors associated with the possibility of irregular distribution of cells throughout the endothelium. If there is extensive pleomorphism the number of cells present within a frame may not be representative of the entire population of endothelial cells.

The cell count in the superior temporal area in the control group was 2152 ± 658 at baseline and increased to 2554 ± 585 at 1 month. This increase is tabulated to be 1.4% but actually will work out to around 20%. Even the subsequent rows in the same column document endothelial counts much above baseline visit. We would like to know if this single value was a typographical error, because every other value shows progressive reduction in the cell count.

It is surprising that with progressive endothelial cell loss the authors did not find any increase in pleomorphism or polymegathism. Endothelial wounds in humans heal almost exclusively by migration and enlargement of endothelial cells.

The percentage of cell loss increased to 15.3% at 12 months and 18.6% at 24 months after surgery. The 95% confidence intervals for the percentage loss of cells have not been reported in the publication. When we calculated the same, we found that the percentage loss of cells at 6 (95% CI: −8.7 to −14.2), 12 (95% CI: −11.0 to −19.5), 18 (95% CI: −12.5 to −20.7), and 24 months (95% CI: −13.5 to −23.7) showed overlapping confidence intervals.

Considering all the above, the conclusion that “there is a progressive loss of endothelial cells after AGV surgery” is probably undecided and needs further studies.