RESUMEN
INTRODUCTION: We have previously shown that children who developed de novo donor-specific human leukocyte antigen (HLA) antibodies (DSA) had greater decline in allograft function. We hypothesised that patients with complement-activating DSA would have poorer renal allograft outcomes. METHODS: A total of 75 children developed DSA in the original study. The first positive DSA sample was subsequently tested for C1q and C3d fixing. The primary event was defined as 50% reduction from baseline estimated glomerular filtration rate and was analysed using the Kaplan-Meier estimator. RESULTS: Of 65 patients tested, 32 (49%) and 23 (35%) tested positive for C1q and C3d fixing, respectively. Of the 32 C1q-positive (c1q+) patients, 13 (41%) did not show concomitant C3d fixing. The mean fluorescence intensity values of the original immunoglobulin G DSA correlated poorly with complement-fixing positivity (C1q: adjusted R 2 0.072; C3d: adjusted R 2 0.11; p < 0.05). C1q+ antibodies were associated with acute tubulitis [0.75 ± 0.18 (C1q+) vs. 0.25 ± 0.08 (C1q-) episodes per patient (mean ± standard error of the mean; p < 0.05] but not with worse long-term renal allograft dysfunction (median time to primary event 5.9 (C1q+) vs. 6.4 (C1q-) years; hazard ratio (HR) 0.74; 95% confidence ratio (CI) 0.30-1.81; p = 0.58]. C3d-positive (C3d+) antibodies were associated with positive C4d histological staining [47% (C3d+) vs. 20% (C3d-); p = 0.04] and with significantly worse long-term allograft dysfunction [median time to primary event: 5.6 (C3d+) vs. 6.5 (C3d-) years; HR 0.38; 95% CI 0.15-0.97; p = 0.04]. CONCLUSION: Assessment of C3d fixing as part of prospective HLA monitoring can potentially aid stratification of patients at the highest risk of long-term renal allograft dysfunction.