Abstract
Recent restandardization of the Roche serum folate assay with traceability to the World Health Organization (WHO) International Standard 03/178 has resulted in a lowering of serum folate results determined in clinical laboratories, particularly at concentrations close to the lower reference limit (LRL). We observed a median (interquartile range [IQR]) negative bias of 21% (8–35%) compared with the preceding Roche assay, with a negative bias of 33–63% at concentrations <4.0 μg/L. The percentage of patients with biochemical deficiency increased from 3% to 18% following restandardization.
We estimated a local population reference interval using GP results obtained between 3 January and 31 May 2017. Subjects were excluded if: (a) <18 or >90 years old; (b) folate results were outside the quantitative range of the assay (<2.0 or >20 μg/L); or (c) haemoglobin (Hb), vitamin B12, or erythrocyte mean corpuscular volume (MCV) values were outside reference limits for adults (females: 115–160 g/L, males: 135–175 g/L for Hb; 197–771 pg/mL for vitamin B12; 80–100 fL for MCV). Duplicate results on individual patients were also excluded.
Folate was determined using the WHO-recalibrated Folate III assay (catalogue number 07559992190), a competitive folate-binding protein assay with electrochemiluminescent detection (Cobas 8000, Roche Diagnostics Ltd, Burgess Hill, UK). Statistical analyses were performed using the Analyse-It v4.51 add-in package for Microsoft Excel (Analyse-It Software Ltd, Leeds, UK).
Median (IQR) folate concentration (n = 4898) was 8.1 (6.3–10.7) μg/L prior to changing to the Folate III assay on 1 December 2016. Following restandardization (n = 9797), it fell to 6.4 (4.4–9.5) μg/L. Non-parametric statistics identified a 95% reference interval, defined by 2.5–97.5 percentile limits (90% confidence intervals), 1 of 2.4 (2.4)–17.5 (17.2–17.7) μg/L.
Median (IQR) age of subjects (77% female) was 54 (39–70) years. Median folate concentration did not differ between genders (P = 0.10) and was consistent across the study period (P ≥ 0.06), except in February when the median folate was higher (6.8 μg/L; P < 0.002).
The calculated 95% reference interval was found to be lower than that reported by the manufacturer (3.9–26.8 μg/L). However, it was higher than that reported by Ferraro et al., in an Italian study of apparently healthy individuals not taking folic acid supplementation (reported LRL 1.3 μg/L). 2
Using a folate LRL of 2.4 μg/L, we estimate that 4.3% of our population would be biochemically deficient. Such a statistically derived LRL may not be suitable as a treatment cut-off, since it is likely to reflect recent dietary intake. Additionally, there is no clear consensus on how low serum folate must be for adverse biochemical/haematological effects to be anticipated; these can occur in the absence of clinical symptoms. 3 Since the results reflect regional diet, there is a risk of normalizing deficiency, particularly in the UK where food fortification with folic acid is not mandatory. The sequelae associated with suboptimal folate intake outweigh the costs of supplementation. We therefore decided on a serum folate of <3.0 μg/L as indicative of deficiency, in line with the British Society of Haematology guidelines. 3
Footnotes
Acknowledgements
We would like to thank Stephen Czarnota for acquisition of patient data.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
Not applicable.
Guarantor
SH.
Contributorship
SH oversaw data analysis and interpretation, and wrote the first draft of the letter. TL and PJT contributed to the concept of the study and provided critical review. All authors approved the final version.