To the Editor:
Re: AJT13: Pages 86–99, 2013 by M. Naesens et al. 
We read with interest this report indicating that chronic histologic damage in early indication biopsies is an independent risk factor for late allograft failure. The authors are to be commended for this exhaustive retrospective analysis of their renal transplant material.
There is no doubt that there is an element in late graft loss that is still not understood, and the paper by Naesens et al.  is useful in reminding us that atrophy and fibrosis have a robust association of progression to failure in kidney transplant populations and indeed in primary kidney disease populations. In all chronic renal diseases, atrophy-fibrosis is the sign of previous parenchymal injury, often due to an underlying progressive disease. Nevertheless, diseases are commonly missed due to sampling error, particularly in biopsies with atrophy-fibrosis.
Our main reservation is that their strategy actually does not identify the main progressive disease, antibody-mediated rejection (ABMR). The diagnosis of ABMR by histology always requires HLA antibody data. The use of microcirculation lesions with no HLA antibody data as a criterion for ABMR is unreliable, as they acknowledge in the discussion, and potentially misleading. From the first analyses of ABMR , it was always clear that the microcirculation lesions alone are common and nonspecific, and by themselves would overestimate the incidence of ABMR. Thus most of the cases in the survival analysis labeled “antibody-mediated changes” probably do not have it. By suggesting a diagnosis in many cases that do not have ABMR, this strategy dilutes the true importance of ABMR. This is why they find 40% of their biopsies have “changes suggestive of ABMR”—much higher than any other study of early biopsies.
It would be instructive to find sera from these people in their unique biopsy set and establish the HLA antibody phenotype with modern methods. It would also be of interest to attribute causes to the failures in the group with chronic histologic damage, compared to the other patients, as has been done in recent studies [3, 4]. What phenotype is over-represented in the failed kidneys in the group with early atrophy and fibrosis?
However, studies of previous eras in transplantation populations, with older immunosuppression practices, will always have limitations. We need new prospective studies in consented patients in the current transplant population to address the issue of the relationship between early chronic kidney injury and outcomes, and the Naesens group is well placed to contribute to this task.