The Cornea Donor Study (CDS) was designed as a prospective cohort study with the following objectives:

To determine whether the graft-failure rate over a 5-year follow-up period following corneal transplantation is the same when using corneal tissue from donors older than 65 years of age compared with tissue from younger donors.

To assess the relationship between donor/recipient ABO blood type compatibility and graft failure due to rejection.

To assess corneal endothelial cell density as an indicator of the health of the cornea and as a alternate outcome measure (in an optional Specular Microscopy Ancillary Study).

The study enrolled 1101 subjects with a corneal disease considered to be at moderate risk for failure (principally Fuchs' dystrophy and pseudophakic corneal edema). A donor cornea meeting the following criteria was assigned to the subject by one of 43 participating eye banks:

• donor age 10 to 75 years
• endothelial cell count 2300 to 3300
• tissue quality very good to excellent
• death to preservation time <12 hrs if body refrigerated or eyes on ice and <8 hrs if not
• death to surgery time <5 days

The cornea surgeon (investigator) and patient were masked to donor age and characteristics of the donor cornea. Preoperative management, surgical technique, and postoperative care, including prescription of medications, were provided according to each investigator's customary routine. The follow-up visit schedule for the initial six months was left to each investigator's discretion and after this time the minimum follow-up visit schedule included a visit between six and 12 months and then one visit every 12 months through five years. The primary study outcome was graft survival at five years. The definition of graft failure, based on the definition used in Collaborative Corneal Transplantation Studies, was a regraft or, in the absence of regraft, a cloudy cornea in which there was loss of central graft clarity sufficient to compromise vision for a minimum of three consecutive months. Follow-up in the initial phase of the study continued for five years unless the patient had a regraft of the study eye.

For the ABO compatibility study, the ABO blood type of both the donor and recipient were determined in order to compare the rate of graft failure for ABO-compatible cases with the rate for ABO-incompatible cases.

For the Specular Microscopy Ancillary Study, endothelial cell counts were determined from specular images by a central reading center, and the relationship of the cell counts to donor age were assessed. In the initial phase of the study, specular images were obtained at 6 months, and then annually through five years post-transplant.

Five-year follow up was completed in November 2007. The 5-year cumulative probability of success was 86%: 86% in the <66.0 year donor age group and 86% in the >=66.0 year donor age group (difference = 0%, upper limit of one-sided 95% confidence interval = 4%). Adjusting for baseline endothelial cell density had no appreciable effect on these results. In a statistical model with donor age as a continuous variable, there was not a significant relationship between donor age and outcome (P=0.11). Three graft failures were due to primary donor failure, 8 to uncorrectable refractive error, 48 to graft rejection, 46 to endothelial decompensation, and 30 to other causes. At least one probable or definite graft rejection episode preceded graft failure in 23 of the 46 failures attributed to endothelial decompensation (4 definite and 19 probable) and in 18 of the 30 failures attributed to other causes (4 definite and 14 probable). The distribution of causes of failure between the donor age groups did not substantially differ.

Although the 5-year results indicated no difference in the success rate of moderate-risk transplants according to donor age, results from the SMAS indicated that among the successful cases, there was a slight association between donor age and endothelial cell loss, with the cell loss after 5 years being slightly lower in corneas from younger donors (r adjusted for baseline endothelial cell density = -0.19, 95% confidence interval -0.29 to -0.08). Whether this slight association between cell loss and donor age is of clinical importance is not known. Of perhaps even greater importance, however, was the finding that irrespective of donor age, endothelial cell loss was substantial over the first five years after transplant even when the graft had been successful. Half of the successful cases experienced a cell loss of 70% or more, and at five years more than half had an endothelial cell density <800 cells/mm2.

Patients in the CDS continued in annual follow up through 2012 in order to to determine the overall 10-year survival rate for moderate risk grafts and to determine whether the graft-failure rate is related to donor age. Additional objectives included determining the value of endothelial cell density in predicting graft failure and evaluating donor and recipient characteristics that may be predictive of late graft failure. All CDS subjects who were active at the 5-year exam were eligible for the extended follow-up phase.

As part of the Specular Microscopy Ancillary Study, follow-up images were obtained during the extended follow-up phase at 7-8 years and again at 10-years. The same procedures used during the first 5 years were followed for the grading of the 7-8 year and 10-year images.

Inclusion

1. Age range 40-80 years
2. Corneal disease: Presence of a condition associated with endothelial dysfunction, including pseudophakic/aphakic corneal edema, Fuchs' dystrophy, posterior polymorphous dystrophy, interstitial keratitis (nonherpetic) or perforating corneal injury
3. Willingness to be followed for 5 years and life expectancy at least 5 years (in investigator's judgement)

Exclusion

1. Patients are to be excluded when there is:

   • a high risk of graft failure, including failed prior penetrating keratoplasty in the eye to be transplanted, chemical burns or other significant cicatricial conjunctivitis (ocular surface disease), herpes simplex/zoster, temporary keratoprosthesis, iridocorneal endothelial syndrome or any corneal condition in which there are two or more quadrants of stromal neovascularization

     • A quadrant is considered vascularized if there is at least one patent stromal vessel >2.0 mm in length crossing the limbus within a quadrant. Two quadrants are considered vascularized if there are at least two stromal vessels 90º or more apart. Enrollment will be restricted to no more than 20% of the total cohort having stromal neovascularization.
     • Patients should be excluded if there are PAS or other abnormalities that, in the view of the investigator, place the patient at high risk for graft failure.

   • a very high probability of success including keratoconus, stromal dystrophies, stromal scars without edema or post-refractive surgery with healthy endothelium

2. Uncontrolled glaucoma or prior filtering surgery with placement of a shunt in the eye to be grafted (Note: eyes requiring a filtering procedure combined with the PK are considered to have uncontrolled glaucoma and are ineligible. Eyes which have undergone filtering surgery (without a shunt) in which glaucoma is currently considered under control are eligible)

3. Uncontrolled uveitis in the eye to be grafted

4. Fellow eye visual acuity <20/200

5. Fellow eye already included in the study

Public dataset access: http://cds.jaeb.org/Studies.aspx?RecID=185