Background: The determination of urinary neopterin using high performance liquid chromatography (HPLC) has routinely been performed in Innsbruck since the middle of the 1980s. The aim was to compare the HPLC method with a commercially available radioimmunoassay.
Methods: Neopterin to creatinine ratios were calculated in the urine of 33 healthy volunteers (33±7.3 years) collected at eight given time points. Neopterin and creatinine were simultaneously determined by HPLC and by radioimmunoassay (RIA; BRAHMS; neopterin) and an enzymatic method (Roche; creatinine), respectively. Assays were compared with Bland-Altman plots and Passing-Bablok regression analysis.
Results: Neopterin to creatinine ratios ranged from 62 to 490 μmol/mol (HPLC) and from 17 to 425 μmol/mol (RIA-enzymatic combined). The ratios were on average 17.3% lower using the combined RIA-enzymatic method compared to the HPLC method. Neopterin concentrations measured by RIA were only slightly lower (on average 4.9%) than by the HPLC and the regression analysis revealed a good accordance with linearity. In contrast, creatinine concentrations by the enzymatic method were on average 12.8% higher than by HPLC.
Conclusions: There is a marked bias between HPLC and the combined RIA-enzymatic method for measurement of urinary neopterin to creatinine ratios which has to be considered when methods are switched during the follow-up of patients.