PB2280 LABORATORY PERFORMANCE OF SERUM FOLATE ASSAY IN THE UNITED KINGDOM (UK) AS ASSESSED BY THE UK NATIONAL EXTERNAL QUALITY ASSESSMENT SCHEME FOR HAEMATINICS: IMPLICATIONS FOR CLINICAL INTERPRETATION

Background: Serum Folate assay is the most commonly used routine test for assessing folate status in the body. In the United Kingdom, there are around 320 laboratories performing the assay using nine different platforms. External quality assessment of the assays is organised by UK NEQAS for Haematinics by sending a serum sample every 3 months. An ‘all participants’ consensus mean is calculated and used as the target value and the results analysed with respect to intra‐ and inter‐ group variation.The percentage bias from the target value is used to assess performance. Aims: We present data on one such assessment to demonstrate the problems of the serum folate assay (Survey number 248, April 2018) and also how it is interpreted by the laboratory for clinical use. Methods: A serum sample with a folate target value of approximately 2.84 μg/L was sent to participating laboratories for analysis. Laboratories were also asked for an interpretation of their result which would be reported to the requesting clinician, namely from low to high. Results: Fig 1 shows an individual laboratory's result (indicated by the arrow) in relation to all laboratories using the same technology (shaded histogram) or all methods (open histogram).There is a significant variation ranging from 1.6 to 4.0 μg/L depending on the assay used, with an overall standard deviation of 0.34 and co‐efficient of variation of 11.9 %.Fig 2 shows the distribution of results in the different methodologies used and how each laboratory interpreted its result. It demonstrates the bias of results obtained by the different methods.The assays using BO5 gave lower results than other methods, with CO 10 showing slightly higher results.Most strikingly there is no consistency both between methods and within methods for interpreting the sample as ‘normal’, ‘deficient’, or ‘indeterminate’ in relation to the numerical value obtained. For example, within the group using BO5, one laboratory reported its level of 3.6 μg/L as ‘low’ despite the result being the highest within its group. Within the full range of results obtained by all groups, there is variable reporting of ‘low’ and ‘normal. Summary/Conclusion: The numerical value obtained of any sample can vary considerably between the different methodologies used. In addition, the interpretation provided by the laboratory can also be highly variable. This may have profound implications for clinical assessment and management. In addition, this is compounded by the fact that it is not quite clear what the reference range is. Our data shows that the different reference ranges quoted by the kit manufacturers do not correspond to their assay bias, ie there is no simple direct correlation between the ranges provided against the bias of their particular assay.Conclusion:The UK NEQAS Haematinics Programme has demonstrated variable performance of the serum folate assay in terms of results obtained, bias and interpretation of results in a style that is unique across EQA/PT schemes. Laboratories and Clinicians need to be aware of the of the shortcomings in assessing serum folate levels in order to provide appropriate clinical advice.