The usefulness of glycated albumin for post-transplantation diabetes mellitus after kidney transplantation: A diagnostic accuracy study

Background: There is no study evaluating the use of glycated albumin (GA) for the detection of post-transplantation diabetes mellitus (PTDM) in kidney transplant recipients. We evaluated the overall accuracy of GA at four months after kidney transplantation. Methods: Diagnostic test accuracy study including 134 kidney transplant recipients without pre-existing diabetes. Receiver operator characteristic (ROC) curve was used to estimate sensitivity, specificity, likelihood ratios and area under the curve (AUC) for GA, considering oral glucose tolerance WA (OGTT) and/or glycated hemoglobin (HbA1c) as reference criteria. Results: Thirty-three patients were diagnosed with PTDM by OGTT and/or HbA1c >= 6.5%. GA showed moderate accuracy to detect PTDM [AUC 0.673 (95% CI 0.557-0.789, p < 0.01)]. The use of OGTT and/or HbA1c >= 6.2% increased the number of PTDM cases from 33 to 38, and AUC was 0.713 (95% CI 0.608-0.819, p < 0.01). GA >= 17% showed specificity close to 90% when OGTT and/or HbA1c 6.5% were used as reference tests. Conclusions: GA showed low diagnostic accuracy for the detection of PTDM at the fourth month after transplantation. The use of a single GA point is not enough for the screening and diagnosis of PTDM; however, GA >= 17% presented high specificity to rule in the disease after kidney transplantation.