Reproducibility of Non-Invasive Endothelial Cell Loss Assessment of the Pre-Stripped DMEK Roll After Preparation and Storage.

PURPOSE

To present a novel, reproducible, and noninvasive method to quantify endothelial cell loss (ECL) of pre-stripped endothelial Descemet membrane lamellae (EDML) caused by its preparation and storage for 5 days.

DESIGN

Prospective laboratory investigation.

METHODS

Thirty EDML were stripped from corneoscleral discs and placed in a well plate containing organ culture medium 1 without dextran. An additional 5 corneoscleral discs were also placed in the same medium and served as a control group. Endothelial cell density (ECD) was measured without any additional manipulation by using spectral microscopy following an extensive protocol by which 3 clear images from the center and periphery were used for each measurement, and each measurement was repeated 5 times. ECD was measured before and directly after preparation and on days 1, 2, and 5 of storage.

RESULTS

The average ECD of the 30 corneoscleral discs, which later underwent stripping, was 2,292 ± 308 cells/mm vs 2,129 ± 222 cells/mm for the 5 corneoscleral discs of the control group. The ECL of the control group was significantly lower than that of the EDML group (P < .0001), reaching ±2% versus 11 ± 5%, respectively, on day 0; 3% ± 4% versus 19 ± 10%, respectively, on day 1; 2% ± 2% versus 22% ± 11%, respectively on day 2; and 4% ± 3% versus 23% ± 9%, respectively, on day 5. Reproducibility of the results on all measurement days was good, with Cronbach alpha values ranging from 0.85 to 0.98.

CONCLUSIONS

A highly reproducible, noninvasive method was presented for measuring the ECD of the EDML. Prestripped EDML lose a significant amount of cells, up to 11%, due to the preparation process and up to 23% after 5 days of storage. Therefore, shipping them after several days of storage to be used in a DMEK surgery cannot be recommended.