THE REPORTED INCIDENCE of recurrent and de novo glomerulonephritis in transplants varies dramatically in different centres. This is probably a reflection of different levels of interest and different biopsy policies in relation to declining function in older grafts. A recent long-term study reported that about 10% of late graft losses were attributable to recurrent disease, a figure second only to chronic allograft nephropathy. Recurrent disease also predisposes to other forms of morbidity, particularly if there is heavy proteinuria. In 1999 we established a new register of recurrent and de novo glomerular disease developing in transplants. This is organised through “UK Transplant” (formerly UKTSSA), the United Kingdom (UK) national transplant support authority. A question about glomerular disease has been added to UK Transplant’s annual follow-up form, which is sent to the clinical staff in all transplant units. As a separate route of case identification, the UK’s transplant pathologists are encouraged to report all new cases of recurrent disease to the first two authors, who run the UK’s external quality assessment schemes in renal and transplant pathology. Uniquely, before a case is accepted to the register, we insist on the biopsy specimen being reviewed independently by two pathologists. Some conditions (eg, recurrent haemolytic uraemic syndrome) tend to recur soon after engraftment whereas others (eg, IgA nephropathy) tend to recur late. The UK Registry has not been in operation long enough to provide data on the influence of recurrent disease on graft survival, although others have already documented that even late recurrence has an adverse influence on graft survival. Preliminary results indicate that the most common recurrent glomerular disease is IgA nephropathy (30% of all cases). Second most common is focal segmental glomerulosclerosis (FSGS) at 10%. This is the reverse of other published reports. We attribute this to our insistence on biopsy specimen review. We have reviewed a number of biopsy specimens from established transplants where a small segment of glomerular sclerosis had led to a diagnosis of FSGS, despite the absence of significant proteinuria. It is well established that many pathological processes, including chronic cyclosporin toxicity, can cause late segmental glomerular scarring, so we excluded such cases and accepted as recurrent FSGS only cases with an appropriate combination of clinical, biochemical, and histological features. There are huge variations in the reported incidence of recurrent disease. In those centres with an aggressive biopsy policy, recurrent disease is found in about 10% of grafts. One centre reported recurrent disease in almost 30% of grafts, but the majority of these cases were not accepted by the Registry for reasons given previously. In the first year of operation of the Registry, no cases of recurrent disease were reported from one third of the UK’s transplant centres. This could in part be a problem of reporting, but in some centres it is clear that the diagnosis is not made because late graft biopsies are not performed. We conclude that recurrent glomerular disease in transplants is a problem that is more common than most centres realise because it is often undetected. At the moment, most types of glomerular disease have no specific treatment, but this is gradually changing. The diagnosis of glomerular disease may justify changes in therapy and aggressive treatment of recurrent FSGS is now moderately successful. Therefore, transplant units will need to consider a diagnosis of recurrent disease more often in the future.