Aims There has been much focus on building effective transition programmes for patients with chronic illness including chronic kidney disease. The structure of current transition programmes can be variable with financial as well as human resource restraints. We aimed to review factors, which influence graft function following the introduction of a transition programme for within our region to identify those at risk of poor outcome.

Methods Data was collected for two groups of patients: nontransition group (pre 2006) and transitioned group (post 2006). This included transplant details, donor type, age at transplant and diagnosis of intellectual disability (ID). In addition, further data was retrospectively collected at 6 monthly intervals for age, height, creatinine, blood pressure, tacrolimus levels, non-attendances and rejection episodes. GFR was estimated using the modified Schwartz formula and analysed using segmented linear regression analysis. These models contained two covariates which estimated overall GFR gradient and the magnitude of change in gradient post-transfer. Follow-up was truncated at 2 years pre- and 4 years post transfer to prevent patients with longer follow-up becoming influential outliers.

Results Following implementation of a transition programme rate of GRF decline was significantly lower for transitioned (n=30) compared with non-transitioned patients (p=0.028).

GFR declined significantly faster pre-transition in patients who received deceased donor organs (p=0.004), were transplanted at <13 years old (p=0.016), had a lower GFR at transfer (p<0.001), no ID (p=0002) had more than 0.5 non-attendances at adult unit/year or the tour (p<0.001).

After transition, patients that attended the tour (p=0.001) or attended with >0.5 attendances at the adult (p=0.022) unit did significantly better.

Subgroup analysis for transitioned patients was then performed (see table). Comparing those with GFR of <30 (n=8) and 30 (n=22) at transfer, we found that patients in the lower GFR group had a significantly faster decline pre-transition ( 11.9 vs 4.9 ml/min/1.73 m2/year, p<0.001). Post-transfer improvement in the GFR <30 group was significantly greater than in 30 group (p=0.003), resulting in gradients in two groups being similar after transfer ( 2.4 vs 3.4 ml/min/1.73 m2/year).

Conclusions Identifying those at risk of poor prognosis following transfer to adult care is crucial to help modify and personalise transition services.