To the Editor: In response to the letter from Drs. Hariharan and Stablein, commenting on our manuscript entitled ˝Long term renal allograft survival: Have we made significant progress or do we need to rethink our analytic and therapeutic strate-gies?˝ (1Meier-Kriesche HU Schold JD Kaplan B Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.Am J Transplant. 2004; 4: 1289-1295Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar), we would like to reemphasize the point that, in this paper we presented actuarial survival data, as op-posed to projections that were presented in the analysis by Hariharan et al. (2Hariharan S Johnson CP Bresnahan BA Taranto SE McIntosh MJ Stablein D Improved graft survival after renal transplantation in the United States, 1988 to 1996..N Engl J Med. 2000; 342: 605-612Crossref PubMed Scopus (1628) Google Scholar). Thus, differences in methodology of half-life projections, as cited in Hariharan's letter, have no bearing on the results and conclusions from our analysis, because we did not use projections (1Meier-Kriesche HU Schold JD Kaplan B Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.Am J Transplant. 2004; 4: 1289-1295Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar). Our analysis is based on data provided by the Scientific Renal Transplant Registry (SRTR) as stated in our paper (1Meier-Kriesche HU Schold JD Kaplan B Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.Am J Transplant. 2004; 4: 1289-1295Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar), but not the USRDS database, as Hariharan has assumed in his letter. It is correct though, that we excluded multiorgan recipients from the analysis, and that accounts for the difference in number of patients cited. Since Hariha- ran implies that the different results of the papers are in part due to this difference, we reanalyzed the data including multiorgan recipients. As expected, the results were very similar to the numbers presented in our paper (1Meier-Kriesche HU Schold JD Kaplan B Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.Am J Transplant. 2004; 4: 1289-1295Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar), with actual transplant half-lives between 1988 and 1995, 6.0, 6.8, 7.0, 7.5, 7.4, 7.6, 7.8 and 8.0 years, respectively. However, when examining only primary transplants, even when including multiorgan transplants, we did not find any significant trend in graft survival half-lives between 1988 and 1995. Our paper described actual half-lives for the overall cohort of patients and separately for retransplants and first trans-plants (1Meier-Kriesche HU Schold JD Kaplan B Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies?.Am J Transplant. 2004; 4: 1289-1295Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar), and not only the overall cohort as implied in Hariharan's letter. However, it was specifically mentioned in the original paper by Hariharan et al. (2Hariharan S Johnson CP Bresnahan BA Taranto SE McIntosh MJ Stablein D Improved graft survival after renal transplantation in the United States, 1988 to 1996..N Engl J Med. 2000; 342: 605-612Crossref PubMed Scopus (1628) Google Scholar), that the projected deceased donor half-life projections included re transplant patients, and that is corroborated by the total number of patients analyzed. As to the controversy about how to calculate half-lives, we would like to reiterate how half-lives are commonly used. The Merriam Webster dictionary defines ‘half-life’ as ´the time required for half the amount of a substance in or introduced into a living system or ecosystem to be eliminated or disintegrated by natural processes´. In fact, half-life estimations in medicine are used to give life ex-pectancy estimations and are made from the moment the process under scrutiny starts, in this case, from the time of transplantation (3Parikh C Ellison DH Improved graft survival after renal transplantation in the United States, 1988 to 1996..N Engl J Med. 2000; 342: 1837Crossref PubMed Scopus (21) Google Scholar). If one wants to use a projection, in order to estimate the time to a 50% survival threshold, a stable part of the survival curve has to be used in order to obtain a reliable projection. Nevertheless, the projected slope needs to be grafted onto the end of the last portion of the available data, as emphasized by Dr. Wolfe in his editorial to our paper (4Wolfe RA Long-term renal allograft survival: a cup both half-full and half-empty..Am J Transplant. 2004; 4: 1215-1216Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar). Alternatively, if estimates are conditioned on 1-year sur-vival (i.e. a terminal half-life or beta half-life) this needs to be clearly specified and interpreted accordingly. In fact, we did this academic exercise, which could be more easily achieved by simply comparing slopes of survival curves, and calculated beta half-lives for patients who still had a functioning graft at 1 year after transplant. As opposed to the findings by Hariharan et al. reported earlier (2Hariharan S Johnson CP Bresnahan BA Taranto SE McIntosh MJ Stablein D Improved graft survival after renal transplantation in the United States, 1988 to 1996..N Engl J Med. 2000; 342: 605-612Crossref PubMed Scopus (1628) Google Scholar), half-lives following the first year of transplantation were actu-ally slightly declining (beta half-life 1988 and 1995 for first deceased donor transplants: 10.1, 10.0, 10.0, 9.6, 9.3, 9.9, 9.0 and 9.5 years). As to the potential follow-up reporting bias discussed in the Hariharan letter, it needs to be mentioned that the current SRTR database has an individualized follow-up censor for each patient. More importantly, the database events are enriched with data from the social security files to ascertain patient death, and from CMS to ascertain graft loss. This additional information results in a very complete reporting of these endpoints in the current SRTR database as opposed to a more than 20% under reporting of these events if this additional data are not used (5Dickinson DM Bryant PC Williams MC et al.Transplant data: Sources, collection, and caveats..Am J Transplant. 2004; 4: 13-26Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). This type of quality control was historically done by the USRDS, but not by UNOS, and the delayed graft loss and death reporting to UNOS may have contributed to the projected trend described in the publication by Hariharan et al. (2Hariharan S Johnson CP Bresnahan BA Taranto SE McIntosh MJ Stablein D Improved graft survival after renal transplantation in the United States, 1988 to 1996..N Engl J Med. 2000; 342: 605-612Crossref PubMed Scopus (1628) Google Scholar). Therefore, we would like to reemphasize that based on ac-tual data short-term renal allograft survival has improved significantly, while long-term survival has not improved from 1988 to 1995. As opposed to trying to reconcile historic projections, we, as the transplant community, should concentrate on addressing the pathophysiology and epidemiology of the constant long-term attrition of grafts.
No takes yet. Share an insight, caveat, or question.
Meier‐Kriesche et al. (2005) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: