June 2013
Volume 54, Issue 15
ARVO Annual Meeting Abstract  |   June 2013
Corneal Endothelial Cell Loss after Endothelial and Penetrating Keratoplasty for Endothelial Disease
Author Affiliations & Notes
  • Sanjay Patel
    Ophthalmology, Mayo Clinic, Rochester, MN
  • Keith Baratz
    Ophthalmology, Mayo Clinic, Rochester, MN
  • Jay McLaren
    Ophthalmology, Mayo Clinic, Rochester, MN
  • Lori Bachman
    Ophthalmology, Mayo Clinic, Rochester, MN
  • William Bourne
    Ophthalmology, Mayo Clinic, Rochester, MN
  • Footnotes
    Commercial Relationships Sanjay Patel, None; Keith Baratz, Assessing the likelihood of developing Fuchs Corneal Dystrophy (P); Jay McLaren, None; Lori Bachman, None; William Bourne, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2013, Vol.54, 1752. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sanjay Patel, Keith Baratz, Jay McLaren, Lori Bachman, William Bourne; Corneal Endothelial Cell Loss after Endothelial and Penetrating Keratoplasty for Endothelial Disease. Invest. Ophthalmol. Vis. Sci. 2013;54(15):1752.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

To compare changes in the corneal endothelium after three different keratoplasty techniques for the treatment of endothelial disease.


Patients with corneal endothelial disease (predominantly Fuchs dystrophy) were enrolled in two consecutive prospective studies at Mayo Clinic, Rochester, MN. In a randomized controlled trial, 28 eyes (26 patients) received penetrating keratoplasty (PK) or deep lamellar endothelial keratoplasty (DLEK) with a 9 mm scleral incision. In a consecutive prospective observational study, 52 eyes (45 patients) received Descemet-stripping endothelial keratoplasty (DSEK) with a 5 mm scleral incision. Endothelial images were acquired through 3 years (or 5 years for some eyes after DSEK) by using confocal microscopy (ConfoScan 3 or 4, Nidek Technologies). Endothelial cell density (ECD) and morphology were determined by the same masked observer by digitizing the apices of cells (HAI Cell Analysis System) in images acquired at various intervals after keratoplasty. Endothelial cell loss (ECL) was the percentage of cells lost from preoperative ECD of the donor tissue, as measured by the eye bank that provided the donor tissue. Endothelial variables were compared between techniques by using generalized estimating equation models to account for any correlation between fellow eyes of the same subject.


Donor diameter was 7.6 ± 0.1 mm (mean ± sd) for PK, 7.9 ± 0.1 mm for DLEK, and 8.2 ± 0.3 mm for DSEK. Preoperative donor ECD did not differ between treatments (PK, 2,845 ± 306 cells/mm2; DLEK, 2,749 ± 364 cells/mm2; DSEK, 2,925 ± 374 cells/mm2; p≥0.10). Mean ECL is summarized in the Table. At one month, there was a trend toward higher ECL after DSEK than after DLEK and PK (p≥0.31, minimum detectable difference, 17% [α=0.05, β=0.20]). At 24 and 36 months, ECL after DSEK was lower than it was after DLEK (p≤0.004) and PK (p≤0.03). At 60 months, ECL after DSEK was similar to that at 36 months after DLEK and PK. By 3 years, there were 0, 1, and 5 graft failures after PK, DLEK, and DSEK, respectively.


Despite a trend toward higher early endothelial cell loss after DSEK, cell loss at 3 years after DSEK is lower than that after PK or DLEK. This difference might be related to graft diameter or to postoperative anatomic differences of the posterior corneal surface. Continued observation is required to determine the longer-term trend in cell loss after DSEK.

Keywords: 481 cornea: endothelium • 741 transplantation • 462 clinical (human) or epidemiologic studies: outcomes/complications  

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.