Patients who are candidates for a second kidney transplant (SKT) frequently have a higher level of panel reactive antibodies (PRA). We assessed the allosensitisation change after a first graft failure (GF), its predictors and impact on retransplantat outcomes. We retrospectively selected 140 adult patients who received a SKT. Recipient and donor characteristics were analyzed. We defined the delta PRA (dPRA) as the difference between peak PRA before the SKT and first one (cohort median value = +10%). Logistic regression analysis was used to determine risk factors for dPRA ≥ 10% and acute rejection (AR) in the SKT. Univariable and multivariable Cox analysis was applied to assess independent predictors of second GF. Risk factors for dPRA ≥ 10% at SKT were AR (OR = 2.57; P = 0.022), first graft survival <1 year (OR = 2.47; P = 0.030) and ABDR HLA mismatch (OR = 1.38 per each mismatch; P = 0.038). AR in the SKT was associated with dPRA ≥ 10% (OR = 2.79; P = 0.047). Induction with a lymphocyte-depleting agent had a protective effect (OR = 0.23; P = 0.010). SKT survival was lower (P = 0.008) in patients with a dPRA ≥ 10% (75.6%, 60.5% in dPRA ≥ 10%; 88.6%, 88.6% in dPRA < 10% patients at 5 and 10 years, post-transplant respectively). Multivariable Cox regression showed that dPRA ≥ 10% (HR = 2.38, P = 0.042), delayed graft function (HR = 2.82, P = 0.006) and AR (HR = 3.30, P = 0.001) in the SKT were independent predictors of retransplant failure. This study shows that an increased allosensitisation at retransplant was associated with the degree of HLA mismatch and led to poorer outcomes. De-emphasis of HLA matching in current allocation policies may be undesirable, particularly in patients with a higher chance of needing a SKT.