Primary corneal graft failure. A national reporting system. Medical Advisory Board of the Eye Bank Association of America.

OBJECTIVES: To describe a national eye banking registry and to assess the effects of donor age, cause of donor death, time from death to procurement, storage time, and distance between the points of recovery and transplantation on the reported occurrence of primary corneal graft failure.
DESIGN: We performed a retrospective case-control study to estimate the odds ratios of five donor factors for cases of primary graft failure voluntarily reported to a national registry using controls from selected eye banks. We also performed a nested case-control cohort study to compare cases of primary graft failure that occurred in both corneas from the same donor with those of nonmated corneas in which primary graft failure was reported to assess odds ratios for the same donor factors. PATIENTS: One hundred forty-seven patients developed primary graft failure in penetrating keratoplasty transplantations performed between January 1, 1991, and December 31, 1993. These cases were reported to the Adverse Reaction Registry of the Eye Bank Association of America, Washington, DC. Controls included 7240 donor corneas distributed by nine eye banks during 1992.
RESULTS: Of the 147 donor corneas that developed primary graft failure, 17 (12%) were obtained from donors who were aged 70 years and older, 39 (27%) came from donors who died of trauma, 13 (9%) had a cadaver time longer than 12 hours, 10 (7%) had a storage time longer than 7 days, and 38 (26%) were distributed outside the eye bank's region. Compared with controls, these donor corneas were more likely to have a storage time longer than 7 days (odds ratio, 2.4; 95% confidence limits, 1.2 and 4.6) and to come from donors aged 70 years and older (odds ratio, 2.4; 95% confidence limits, 1.4 and 4.0). The 22 corneas (15%) in which primary graft failure occurred in both recipients from the same donor were 8.5 times (95% confidence limits, 1.1 and 51.5) more likely to be preserved beyond 1 week than were nonmated corneas with primary failure but were not from significantly older donors. Logistic regression analysis showed that the association between prolonged storage time and primary graft failure in mated corneas remained significant even when the analysis was controlled for other donor factors.
CONCLUSIONS: No clearly defined donor or eye banking factor accounted for most cases of primary graft failure, although prolonged corneal storage and advanced donor age may increase its risk. Ophthalmologists are urged to report to their eye bank all cases of primary graft failure and other adverse events that might be attributable to donor eye tissue.