Letter to Editor in response to the article "Vitamin D insufficiency as a potential culprit in critical COVID-19 patients".

To The Editor,

We carefully read the meta‐analysis article on the association between serum vitamin D status and outcome of COVID‐19 by Munshi et al. and would like to make a technical critique about the pooled estimation of serum vitamin D level. Munshi et al. have enrolled six articles that are heterogeneous in vitamin D measurement methods as follows.

Article reference No.8: Chemiluminescence Immunoassay (CLIA); article reference No.16: Electrochemiluminescence (ECL); article reference No.19: Liquid Chromatography Tandem Mass‐Spectrometry (LC‐MS/MS); article reference No.20: is an author's reply to a letter to the editor and have no experimental data; article reference No.21: is a letter to editor and has no any information regarding the method of vitamin D assay; article reference No.22: Enzyme‐Linked Fluorescent Assay (ELFA). In this regard, we believe that all the aforementioned articles are not eligible for estimation of pooled serum vitamin D level.

Of note, two major metabolites of vitamin D, that is, 25OH‐Vit D3 and D2 are being considered as markers for evaluating vitamin D status in serum. To measure these metabolites in clinical settings, various methods are in use, most notably immunoassay‐based methods such as Enzyme‐Linked Immunosorbent Assay, ELFA, radioimmunoassay, CLIA, and ECL along with Competitive Protein Binding Assay, High‐Performance Liquid Chromatography, and LC‐MS/MS. The advantages and disadvantages of these methods are reviewed in Table 1. Considering the differences in principle of measurement and extraction method, inconsistency between the results is expectable which leads to remarkable difficulties in interpretation of the clinical data and decision making. Therefore, it can be inferred that various vitamin D measurement methods may affect the result outcomes for investigation in both clinical practice and basic research.

Table 1

Advantages and disadvantages of different 25 OH Vitamin D measurement methods

	Advantages	Disadvantages
CPBA	✓ Inexpensive ✓ Can be performed on a small sample size ✓ Co‐specific for 25OH‐Vit D2 and D3. 3	– Underestimates 25OH‐Vit D3 at low levels and overestimates it at high levels. – Poor reproducibility. – Sensitive to nonspecific interfering substances. – Instability of the binding proteins. 3 , 4
RIA	✓ Inexpensive ✓ Less susceptible to nonspecific interference ✓ Accurate. 3	– Requires the use of radionuclides. – Some RIAs are not able to detect both 25OH‐Vit D2 and D3 equally. 4
ELISA and ELFA	✓ Inexpensive ✓ Acceptable precision, accuracy, sensitivity, and specificity. 5	– Some ELISA/ELFA kits are not able to detect both 25OH‐Vit D2 and D3 equally and may underestimate the 25OH‐Vit D2. – Interferences due to matrix effect. 5
CLIA and ECL	✓ It is a sensitive and specific method. 6	– In some cases, it has acceptable performance in healthy individuals and in vitamin D3‐supplemented patients, but the performance is unacceptable in patients who receive vitamin D2 ‐supplements. – In some cases it overestimates the circulating 25OH‐Vit D3 concentrations. 6
HPLC	✓ Can detect 25OH‐Vit D2 and D3 separately. ✓ Much evidence for the precision and accuracy of the test. 4 , 7	– Time consuming and low throughput. – Needs an expert technician. – Sometimes assay is subject to interference.
LC‐MS/MS	✓ Can detect 25OH‐Vit D2 and D3 separately. ✓ Minimizes the interferences and matrix effects. ✓ It is considered as a gold‐standard method. ✓ High sensitivity, specificity and repeatability. 3 , 7 , 8	– Time‐consuming and low throughput. 3 , 9 – A common problem with LC‐MS/MS is its relative inability to discriminate between 25OH‐Vit D3 and its inactive isomer 3‐epi‐25OH‐Vit D3 which causes overestimation of total concentration of vitamin D. 3 , 7 , 9

Abbreviations: CLIA, chemiluminescence immunoassay; CPBA, competitive protein binding assay; ECL, electrochemiluminescence; ELFA, enzyme‐linked fluorescent assay; ELISA, enzyme‐linked immunosorbent assay; HPLC, high‐performance liquid chromatography; LC‐MS/MS, liquid chromatography tandem mass‐spectrometry; RIA, radioimmunoassay.

Co‐specific for 25OH‐Vit D2 and D3.

Instability of the binding proteins. ,

Accurate.

Some RIAs are not able to detect both 25OH‐Vit D2 and D3 equally.

Acceptable precision, accuracy, sensitivity, and specificity.

Interferences due to matrix effect.

It is a sensitive and specific method.

In some cases it overestimates the circulating 25OH‐Vit D3 concentrations.

Much evidence for the precision and accuracy of the test. ,

High sensitivity, specificity and repeatability. , ,

A common problem with LC‐MS/MS is its relative inability to discriminate between 25OH‐Vit D3 and its inactive isomer 3‐epi‐25OH‐Vit D3 which causes overestimation of total concentration of vitamin D. , ,

Obviously, choosing an inappropriate method for vitamin D measurement can lead to misclassification of the disease status and advising a wrong treatment strategy. As a consequence, clinicians should be aware of the limitations and possible differences in result interpretation between different methods before ordering the vitamin D test.

One suggestion for overcoming the vitamin D measurement challenges is to monitor vitamin D status in different stages of the disease in the same laboratory. Moreover, the variations in the results of vitamin D assays have been decreased using the vitamin D standardization‐certification program (VDSCP), under the authority of the Centers for Disease Control and Prevention (CDC) that evaluates the accuracy and reliability of vitamin D tests using well‐established methods. This program certifies those methods that have bias and CV equal to or less than 5% and 10%, respectively. Thus, it is recommended to use commercially available vitamin D assay kits that meet CDC's analytical performance criteria. The list of VDSCP certified participants is available from https://www.cdc.gov/labstandards/vdscp_participants.html.

In conclusion, there are some significant differences in the vitamin D measurement results obtained by various methods. It is suggested that Munshi et al. should consider the advantages and disadvantages of each method and re‐evaluate the interpretation of the results based on such differences.