Biotin is a water-soluble vitamin that is an essential coenzyme for multiple metabolic processes in the human body [1]. In immunoassays, streptavidin–biotin binding is widely used due to its strong affinity. Therefore, increased serum biotin levels can interfere with immunoassays, thereby leading to serious medical problems. Indeed, the United States (US) Food and Drug Administration (FDA) released a safety communication about the potential for biotin interference with laboratory test results in 2017 and updated the documents in 2019 [2].

In general, biotin supplementation is rarely necessary, except for the treatment of metabolic disorders, such as propionic acidemia, biotinidase deficiency, and multiple sclerosis. Recently, in addition to medical indications, biotin intake increased for esthetic indications, such as hair growth, skin health, and nail strength [3]. In the US, the market size for vitamin and mineral supplements is over $30 billion annually [3]. In Korea, the consumption of dietary supplements is growing, and the most frequently consumed dietary supplements are multivitamin mineral supplements, with a prevalence rate of 89.6 per 1,000 persons [4]. According to the 2020 data from the Ministry of Food and Drug Safety (Korea), the top five companies earned approximately $1.1 million from biotin sales alone [5].

There are significant differences in biotin levels in the general population from country to country. According to some studies, 0%–7.4% of patients’ samples contain biotin levels above 10 ng/mL: the lowest threshold for biotin interference [6-10]. Nonetheless, serum biotin levels have never been evaluated in Koreans. In this study, we investigated serum biotin levels in the general Korean population and estimated the prevalence of biotin interference in the thyroid function test (TFT) and vitamin D assays.

From October 2020 to March 2021, 723 samples were collected from medical checkup examinees. After centrifugation, serum was separated and stored at 4°C. All stored serum samples were analyzed within 7 days on the basis of our preliminary study indicating that the change of biotin levels at 4°C is less than 6% after 7 days of storage. Biotin levels were quantified by the method described by Song [11] with modifications; the method is based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) on a Xevo TQD instrument (Waters Corporation, Milfold, MA, USA). Purified biotin (Sigma-Aldrich, St. Louis, MO, USA) served as a standard material, and Biotin-d4 (IsoSciences, Ambler, PA, USA) was used as an internal standard. The lower limit of quantitation was 5 ng/mL. Biotin interference in the TSH, free-T4, and vitamin D immunoassays that are based on biotin was assessed for samples with biotin levels >10 ng/mL. The threshold of biotin interference claimed by a manufacturer is 1,200 ng/mL for TSH, 30 ng/mL for vitamin D, and 100 ng/mL for free T4, respectively. To assess the interference, TSH, free-T4, and vitamin D results obtained by a streptavidin-biotin-based electrochemiluminescence immunoassay (ECLIA) using a Cobas c702 analyzer (Roche Diagnostics, Mannheim, Germany) were compared to those of a chemiluminescent microparticle immunoassay (CMIA) by means of ARCHITECT (Abbott Laboratories, Abbott Park, IL, USA) and LC-MS/MS, respectively. The following reagents were employed: Elecsys TSH (Roche Diagnostics) and ARCHTECT TSH (Abbott Laboratories) for TSH, Elecsys FT4 III (Roche Diagnostics) and ARCHITECT Free T4 (Abbott Laboratories) for free T4, and Elecsys Vitamin D total II (Roche Diagnostics) and the MSMS Vitamin D Kit (PerkinElmer, Waltham, MA, USA) for vitamin D. The study protocol was approved by the Kangbuk Samsung Hospital Institutional Review Board (approval number: KBSMC 2020-08-011). Because this study involved anonymized residual samples, the requirement to obtain informed consent was waived.

The median age of examinees was 48 years (range: 18–81 years). The proportions of males and females were 50.5% and 49.5%, respectively. Of all the samples analyzed, 99.6% (720/723) showed biotin levels lower than the lowest interference threshold. Two samples contained biotin levels of 5–10 ng/mL, and 718 samples contained biotin at concentrations lower than 5 ng/mL. Only three (0.4%) samples showed biotin levels >10 ng/mL: 12.0, 14.1, and 130.3 ng/mL. Among these three samples, only one featured a biotin level higher than the threshold of biotin interference with free-T4 (100 ng/mL) and vitamin D assays (30 ng/mL). There were no samples with a biotin concentration above the interference threshold for TSH (1,200 ng/mL). For the three samples with biotin levels >10 ng/mL, biotin interference in free-T4 and vitamin D immunoassays was evaluated (Table 1). For the sample with the biotin level of 130.3 ng/mL, vitamin D status was normal in ECLIA (30.6 ng/dL) but deficient in LC-MS/MS (29.3 ng/dL). For all three samples with biotin levels above 10 ng/mL, free-T4 values were slightly higher in ECLIA than in CMIA but were within the reference ranges of both methods. Interference with the TSH assay was not evaluated because no samples contained more biotin than the interference threshold for TSH (1,200 ng/mL).

Table 1 . Assessment of biotin interference with free-T4 and vitamin D immunoassays

Biotin (ng/mL)	FreeT4 (ng/dL)*		Vitamin D (ng/mL)†
	ECLIA (RI: 0.84-1.74)	CMIA (RI: 0.70-1.48)	ECLIA (RI: 30-39)	LC-MS/MS (RI: 30-39)
130.3	1.39 (N)	1.16 (N)	30.6 (N)	29.3 (D)
14.1	1.30 (N)	1.15 (N)	30.4 (N)	30.3 (N)
12.0	1.23 (N)	1.12 (N)	31.6 (N)	34.8 (N)

*The biotin threshold for interference with free-T4 assays is 100 ng/mL; ^†The biotin threshold for interference with vitamin D assays is 30 ng/mL.

Abbreviations: ECLIA, electrochemiluminescence immunoassay; CMIA, chemiluminescent microparticle immunoassay; LC-MS/MS, liquid chromatography coupled with tandem-mass spectrometry; RI, reference interval; N, normal; D, decreased.

Other studies on the prevalence of biotin interference have involved patients visiting clinics and emergency rooms. To the best of our knowledge, this is the first study on the prevalence of biotin interference in the general Korean population. The prevalence of biotin above the interference threshold varies among studies. Studies in the United Kingdom, the Netherlands, Australia, and the US point to prevalence rates of 0%, 0.2%, 0.8%, and 4.1%–7.4%, respectively [6-10]. In the present work, the prevalence was found to be 0.4%, which is much lower than that in the US and similar to that in the Netherlands. According to the observed biotin levels in the general population, the risk of biotin interference may be low in Korea. The prevalence may depend on population characteristics, such as age, socioeconomic status, and health management knowledge.

Most studies on biotin interference have been conducted via experiments such as spiking of serum samples with biotin or comparing results before and after biotin ingestion [12]. Therefore, the prevalence of this interference has not been estimated accurately. We evaluated biotin interference by comparing the results of two methods without a biotin intervention. Although the difference was not sufficient to be regarded as interference, vitamin D status was affected in the sample with biotin at 130.3 ng/mL. Although free T4 was slightly higher in biotin-based ECLIA than in CMIA, the interpretation of the TFT did not change. If more samples with free-T4 values near the reference limit had been included, then the interpretation could have been affected. Although biotin ingestion can pose a risk of interference, clinically significant interference is uncommon. There has been biotin interference in TFT and vitamin D assays in a few cases of high-dose biotin treatment [12, 13]. Two cases of misdiagnosed Graves’ disease due to biotin interference were reported in 2 years at a Swiss institution [14]. The two patients were treated with high-dose biotin for multiple sclerosis. Although their TFT results were compatible with hyperthyroidism, there were no symptoms or signs of hyperthyroidism.

The current study has some limitations. First, the study was conducted by a single institution. Biotin levels can vary from region to region even in the same country depending on population characteristics. Therefore, multicenter studies are necessary to confirm that the results of this report represent the biotin levels of the general Korean population. Second, evaluation of biotin interference was limited because only three samples showed biotin levels above the lowest threshold of biotin interference. Future studies focusing on individuals taking biotin supplements can help to evaluate biotin interference. Third, biotin interference was not evaluated in samples with biotin levels <10 ng/mL. Therefore, it is not clear whether the differences observed in samples with biotin concentration >10 ng/mL are due to the interference or a difference in analytical methods.

In conclusion, we investigated serum biotin levels in the general Korean population and biotin interference in TSH, free-T4, and vitamin D assays. According to these biotin interference assessments, the risk of biotin interference appears to be low in Korea. Studies on other analytes are needed to ascertain the impact of biotin interference.

None declared.