Clin Chem Lab Med 2017; 55(9): e195–e196 Letter to the Editor Simona Ferraro*, Andrea Panzeri, Simona Borille, Dominika Szoke and Mauro Panteghini Estimation of the reference interval for serum folate measured with assays traceable to the WHO International Standard DOI 10.1515/cclm-2016-0941 Received October 17, 2016; accepted December 16, 2016; previously published online January 20, 2017 Keywords: immunoassay; reference limit; traceability. To the Editor, Investigating the harmonization of folate immunoassays, Kristensen et al. recently reported for most commercial assays acceptable measurement differences within the limit of desirable bias [1]. This is probably due to the assay recalibration to the WHO International Standard-National Institute for Biological Standards and Control (NIBSC) code 03/178 [2]. However, we have recently experienced with the recalibration of Roche Diagnostics Folate III assay, now traceable to NIBSC 03/178 material, a significant shift in the average folate measured values. Particularly, at serum folate concentrations around the lower reference limit (LRL) of the old Roche assay (which was calibrated by manufacturer) we observed a positive bias of ~ 50% vs. the new Roche recalibrated assay. The regional external quality assessment program confirmed these results where, by cumulating data generated by participants using the Roche system, a dichotomous distribution of folate results became evident. Taking into account this difference, the shift from 4.6 μg/L (Roche recommended LRL for old cali- bration) to 3.9 μg/L (Roche recommended LRL for recali- brated assay) appears to be inconsistent. Here we report data from apparently healthy individuals obtained with the WHO-recalibrated assay in order to accurately define the traceable reference interval for serum folate. Without an adequate reference interval, the recalibration changes can indeed impair the interpretation of the folate results and, paradoxically, worsen the patient’s outcome [3]. We derived the reference interval of serum folate in agreement with the Clinical and Laboratory Standards Institute C28-A3c standard [4], by enrolling a cohort of 322 healthy blood donors (50% males), signing an informed consent and filling in a nutritional questionnaire. Subjects were excluded if: a) hemoglobin (Hb) and erythrocyte mean corpuscolar volume (MCV) values exceeded reference limits for adults (females: 120–160 g/L, males: 130–175 g/L for Hb and 80–99 fL for MCV, respectively), b) supplemented with folic acid, or c) the hemolytic index of collected serum sample (estimated on a Abbott Architect c16000 system) was ≥ 25. After blood centrifugation, serum samples were immediately stored at −80 °C for a maximum of 3 months until folate determination using WHO-recalibrated Folate III assay (code 07559992190) on a Cobas 6000 analyzer (Roche Diagnostics). Measurements were performed in four runs using a single reagent lot (no. 123242) by checking the system alignment before and just after running samples by control materials provided by Roche (lot no. 18746200 and 18746300). The assay shows a total imprecision (as coef- ficient of variation) ≤ 7.2% at an average concentration of 6.5 μg/L (data from our internal quality control), a limit of detection of 0.6 μg/L and an upper limit of calibration range of 20 μg/L. Continuous variables with normal distri- bution according to the Shapiro-Wilk test were expressed as mean ( ± standard deviation), alternatively as median (25th–75th percentile), whereas dichotomous variables were expressed as absolute number and percentage. T-test or Mann–Whitney test when appropriate were used to compare groups and a multiple regression model (MRM) to estimate the influence of subject features (i.e. age, gender, Italian origin, smoking habit, portions of consumed fruit/ vegetables) on serum folate concentrations. All statistical analyses were performed by R software. *Corresponding author: Simona Ferraro, UOC Patologia Clinica, Ospedale ‘Luigi Sacco’, Via G.B. Grassi 74, 20157 Milan, Italy, Phone: + 39-02-3904-2766, Fax: + 39-02-5031-9835, E-mail: simona.ferraro@asst-fbf-sacco.it Andrea Panzeri: Research Centre for Metrological Traceability in Laboratory Medicine, University of Milan, Milan, Italy Simona Borille and Dominika Szoke: Clinical Pathology Unit, University Hospital ‘Luigi Sacco’, Milan, Italy Mauro Panteghini: Clinical Pathology Unit, University Hospital ‘Luigi Sacco’, Milan, Italy; and Research Centre for Metrological Traceability in Laboratory Medicine, University of Milan, Milan, Italy Brought to you by | Università degli Studi di Milano Authenticated Download Date | 8/9/17 5:02 PM