ATC 2012 Report – Large donor-outcome studies are valuable but a major logistical challenge

by Thomas R. Collins – Tracking living kidney donor outcomes in a comprehensive way is fraught with difficulties and complications, from coordination challenges to being forced to rely on sources of data that are not always reliable, an expert said here at the 2012 American Transplant Congress.

Alan Leichtman, MD, Professor of Internal Medicine with an interest in transplantation outcomes, at the University of Michigan, is a lead investigator on RELIVE, or Renal and Lung Living Donors Evaluation, a series of both retrospective and prospective studies involving six research centres covering a broad geographic area of the U.S.

The studies involve looking retrospectively at outcomes from all donors at the centres through 2007, using chart reviews at the study sites and linking to national databases for vital status, cause of death and end-stage renal disease status.

The project also includes a “cross-sectional” component, in which researchers try to reach all donors with contact information who donated through 2005, and are still alive, attempting to determine renal and cardiovascular morbidities, quality of life and glomerular filtration rate.

The project also includes a prospective arm that is ongoing.

The study now encompasses 8,951 donors, with more than 1,000 variables having been collected, Dr Leichtman said.

There is value in these kinds of studies, providing a comprehensive look at medical, surgical, psychosocial and socioeconomic endpoints, cross-sectional components to capture mid-term morbidities, he said. And the size, scope and length of time the project covers gives a high probability of accurately estimating the frequency of common events that follow donation. Plus, these studies can identify uncommon or late events involved with donation.

But getting there involves “an enormous amount of work,” Dr Leichtman said.

The project has been going on for six years, with 17 manuscripts at various points in the publication process, and there are efforts now to keep funding going to extend the study’s scope.

A project of this type also involves finding agreement among a large group of investigators on methodology and interpretation of the results.

And, in the case of contacting donors to find out about morbidities, there may be a bias among those who respond — only 33 percent participated in that phase, he said.

Using information from the Organ Procurement and Transplantation Network is tricky, in part because, before 1994, demographic information was incomplete and reporting of adverse outcomes was sporadic.

Researchers must link clinical databases with administrative databases in order to match donors with records of people who have died, but that is a task in wading through imperfections, from middle names used as first names to last names that don’t match often because of a marriage.

Of 650 deaths obtained from the Social Security Death Master File — which includes the most U.S. deaths of those with a social security number — using an algorithm to make a match, 96 of those matched using one algorithm but not a second algorithm.

Dr Leichtman stressed the importance of wading patiently through this information to weed out inaccuracies.

“The quality of the results,” he said, “depends on the use of multiple data sources, the accuracy of the matching algorithm, and the care used to resolve imperfect matches.”