A secondary peak (#C) included in the HbA1c-calculation by the HA-8140 HPLC (Menarini) shows a fairly good correlation with serum urea. The correlation with HbA1c and with serum glucose is at first glance significant but reveals at a closer look being biased by some incorrect assumptions. The difference between immunoturbidimetric determination (Tinaquant HbA1c II Roche) and HPLC shows a similar behaviour to urea as does the #C-peak of the chromatographic separation. This peak as well as the difference between both determinations of HbA1c could be attributed to carbamylated haemoglobin. The definitive identity has to be proven. This peak could be a monitoring tool for long-time urea. The integration of this peak into total HbA1c by the HA-8140 (Menarini) can lead to a false diagnosis of diabetes in non-diabetic patients with elevated urea.