ABSTRACT Background Recurrence of immunoglobulin A nephropathy (IgAN) limits graft survival in kidney transplantation. However, predictors of a worse outcome are poorly understood. Methods Among 442 kidney transplant recipients (KTRs) with IgAN, 83 (18.8%) KTRs exhibited biopsy-proven IgAN recurrence between 1994 and 2020 and were enrolled in the derivation cohort. A multivariable Cox model predicting allograft loss based on clinical data at the biopsy and a web-based nomogram were developed. The nomogram was externally validated using an independent cohort (n = 67). Results Patient age <43 years {hazard ratio [HR] 2.20 [95% confidence interval (CI) 1.41–3.43], P < .001}, female gender [HR 1.72 (95% CI 1.07–2.76), P = .026] and retransplantation status [HR 1.98 (95% CI 1.13–3.36), P = .016] were identified as independent risk factors for IgAN recurrence. Patient age <43 years [HR 2.77 (95% CI 1.17–6.56), P = .02], proteinuria >1 g/24 hours [HR 3.12 (95% CI 1.40–6.91), P = .005] and C4d positivity [HR 2.93 (95% CI 1.26–6.83), P = .013] were found to be associated with graft loss in patients with IgAN recurrence. A nomogram predicting graft loss was constructed based on clinical and histological variables, with a C statistic of 0.736 for the derivation cohort and 0.807 for the external validation cohort. Conclusions The established nomogram identified patients with recurrent IgAN at risk for premature graft loss with good predictive performance.