Elevated parathyroid hormone (PTH) concentrations were reported to be associated with chronic renal allograft failure. However, measurements of PTH are challenging, because PTH can occur either as non-oxidized (n-ox) or oxidized (ox) PTH. Only n-ox PTH is a PTH receptor agonist. The intact PTH (iPTH) concentrations measured routinely in clinical practice, however, equals non-oxidized PTH (n-oxPTH) plus oxidized PTH (oxPTH). In CKD patients, the majority of the circulating PTH is oxidized. We measured iPTH, oxPTH and n-oxPTH at study entry in 600 kidney transplant recipients (KTRs). They were followed for graft loss for 3 years. Graft loss was defined as need for initiation of renal replacement therapy. Thirty-eight patients had graft loss during the 3 years follow-up. OxPTH correlated very well with iPTH (R2 = 0.997, p < 0.0001), whereas the correlation between n-oxPTH and iPTH was much weaker (R2 = 0.762, p < 0.0001). Compared to KTRs without graft loss, KTRs with graft loss had significantly higher levels of iPTH, oxPTH, and n-oxPTH (p < 0.0001 in all cases). After adjusting for confounding factors in cox proportional hazards analysis, only n-oxPTH, but not oxPTH neither iPTH, was significantly associated with graft loss (Hazard ratio (HR): 1.02, 95% CI: 1.01–1.03, p = 1.84 × 10−3). The very close correlation between oxPTH and iPTH measurements suggests that conventional iPTH measurements most likely describe oxidative stress rather than PTH bioactivity. Only non-oxidized PTH but not oxidized PTH nor intact PTH is associated with graft loss in stable kidney transplant recipients.