Donor death-surgery interval may affect DSAEK outcomes

Data analyzed from a total of 568 eyes included in a prospective study of Descemet's stripping automated endothelial keratoplasty indicate that the time from donor death to surgery affects endothelial cell density loss and risk of graft dislocation, but more data are needed to understand possible associations better.

Key Points

She presented results based on 568 consecutive eyes that underwent DSAEK and were categorized into two groups according to whether the death-to-surgery (DTS) time for donor cornea was <96 hours or longer; 96 hours was chosen as the cut-off because it corresponded to the median DTS time.

Pearson correlation coefficient analyses showed that DTS time of ≥96 hours correlated with greater endothelial cell density (ECD) loss at 6 months, she said. The correlation, however, was lost at 12 months, according to Dr. Shamie. Eyes with a longer DTS time also had a significantly higher rate of graft dislocation compared with those where the DTS time was <96 hours (11 of 292, 3.8% versus 2 of 276, 0.7%; p = 0.009). Mean DTS time among all 13 eyes with a graft dislocation was 96.51 hours. No primary graft failures were seen in the entire series.

"We believe other confounding factors that can affect ECD loss-such as presence of glaucoma filtering tubes, graft dislocation, or diagnoses such as iridocorneal endothelial syndrome-also must be considered to determine best the donor characteristics that may affect ECD loss rates," she added. "Longer-term prospective follow-up . . . will be necessary to decipher these correlations better."

The group that received the grafts with the DTS time <96 hours included 86 eyes with a surgeon-cut donor cornea implanted and 206 eyes that received a precut donor cornea. For the ≥96 hour group, 59 eyes received a surgeon-cut graft and 217 had a precut donor. Mean DTS time was 72 hours (range, 21 to 95) in the <96 hour group and 126 hours (range, 96 to 182) in the ≥96 hour group. No differences were found between the two DTS time groups in mean donor age, death to preservation time, and mean preoperative ECD measured with specular microscopy, she said.

At 6 months, percent ECD loss was higher in the ≥96 hours DTS group compared with eyes that received grafts with the shorter DTS interval, 31% versus 28%. Endothelial cell loss rates increased only slightly over the next 6 months in both groups and remained slightly higher in the group where DTS time was ≥96 hours compared with <96 hours, 33% versus 29%. At both 6 and 12 months, the difference between groups in the ECD loss rates showed a trend to statistical significance (p = 0.054 and p = 0.056, respectively).

Further study of a possible association between DTS time and ECD loss was based on a subset of eyes that included only those with a diagnosis of Fuchs' dystrophy and also removing eyes with graft dislocation, that required graft repositioning, or that had graft rejection. In this subgroup, the endothelial cell loss rate at 6 months was significantly higher in eyes in the DTS time =96 hours group compared with those that received a graft with a shorter DTS time, 31% versus 26% (p = 0.045), but the significant difference was lost at 12 months, 31% versus 28% (p = 0.24).

To see whether the interval from precutting to surgery influenced outcomes, eyes with precut tissue were subdivided into four groups based on DTS time (<96 versus ≥96 hours) and the time from the resection to surgery (≤24 hours versus >24 hours). No significant differences in ECD loss rates were observed. No interaction took place between DTS time and precutting-to-surgery time.