Comparisons of metabolite profile from paired serum and ethylenediaminetetraacetic acid–plasma samples using dry chemistry technology: An emergency department perspective

Introduction

Conventionally, serum is the most commonly used matrix for biochemical assays in the clinical laboratory. The principle advantage of using serum as matrix is that reference ranges are readily available for all the different biochemical systems and platforms used for assay. However, serum separation requires at least 20–30 min standby time, which may be further increased in cooler temperatures. This leads to a significant increase in the turnaround time for an investigation (time taken from blood sample collection, processing, analysis, and the final report reaching the clinician). Although it may not be important for elective investigations, this increased turnover time can lead to a significant delay in the clinical decision-making by the treating doctor in the Emergency Department, which can have an adverse impact on patient morbidity and mortality. As a consensus, <60 min is considered to be an ideal acceptable turnover time for laboratory investigations (sample registration to result reporting) for common laboratory tests.[1] Advantages of ethylenediaminetetraacetic acid (EDTA) anticoagulated plasma for biochemical analysis include reduced time for sample processing (as no time lost for serum generation). Also for certain investigations such as hemogram, only EDTA anticoagulated blood sample is used. If all the biochemical investigations can be done from EDTA anticoagulated plasma, it will reduce the need for collection of blood samples in different vials at the time of sample collection, thus reducing the volume of blood sample needed for biochemical investigations. Sometimes, there may be a restriction to the volume of blood sample that can be collected from a patient based on the clinical status of the patient.

However, there are a few reports to suggest that significant difference may be observed when the same biochemical parameter is measured in serum and plasma samples based on the platform used for analysis.[23] Incorrect reporting may lead to diagnosis of nonexistent disorders (labomas), which can lead to significant patient morbidity and mortality.[45]

Dry chemistry technology is increasingly being used in large laboratory setups. Advantages of dry chemistry technology include the ability to process a large number of samples in a relatively shorter duration of time along with a good accuracy of testing. However, till date, no data is available evaluating the difference in serum versus plasma sample assay of commonly tested biochemical parameters in clinical practice. Hence, the aim of this study was to evaluate the differences in serum versus plasma sample estimation of commonly evaluated biochemical parameters using dry chemistry technology in a large tertiary care center of India.

Materials and Methods

Paired blood samples were collected from a single venipuncture of 405 patients admitted to the emergency department of a large tertiary care center in India. Blood samples were collected in plain as well as K2-EDTA vacutainer tubes (Becton Dickinson). K2-EDTA plasma and serum were separated by centrifugation using standard guidelines and processed for estimation of glucose, urea, creatinine, uric acid, total bilirubin (TBil), direct bilirubin (DBil), aspartate transaminase, alanine transaminase, total protein, albumin, total cholesterol (TC), direct high-density lipoprotein (HDL), triglycerides (TGs), sodium, chloride, phosphorus, creatine phosphokinase total, CPK-MB (CPKmb), lactate dehydrogenase (LDH), amylase, lipase, and C-reactive protein (CRP). Dry chemistry autoanalyzer (Vitros-350, Ortho Clinical Diagnostics) was used to process all the samples. The Institute Ethics Committee approved the study protocol. The study protocol was explained and informed written consent was taken from all patients before sample collection for this study.

Dry chemistry technology

The main principle of dry chemistry is based on the reflectance spectrophotometry in which the reflectance of the materials is measured to give a reference standard for the comparison of the color of the different samples. In dry chemistry technology, the slides are dry, multilayered analytical elements coated onto polyester support. A small amount of patient sample is deposited onto slide and evenly distributed to all the layers. The spreading layer consists of appropriate substance and other components needed for the chemical reaction to occur. The analyte in the sample catalyzes the reaction sequence to yield products which absorb light at wavelengths in various regions (340–680 nm), diffuses into underlying layer, and is monitored by reflectance spectrophotometry. The tests performed were colorimetric, enzymatic endpoint, two-point or multipoint rate, or potentiometric analysis depending on the analysate. The CRP was measured by immunoturbidimetric assay. Advantages of DCT include the extremely small amount of sample needed for analysis (10 μL), accuracy, and high volumes of patient samples being analyzed over a shorter period.

The methods used for analysis, the assay range, and intra- and inter-assay coefficient of variation for each of the biochemical parameter evaluated have been elaborated in Table 1.

Table 1 Details of the methods used for biochemical analysis of different evaluated parameters in this study with their assay range, minimal detectable concentration, and their inter- and intra-assay coefficient of variations

Results

For the collected samples from 405 patients, data from 401 patients were analyzed. Four pairs of samples were excluded from the analysis due to the presence of significant hemolysis. Percentage differences between serum versus anticoagulated plasma samples for all analytes ranged from 0.0% to 57.44% and were <±4% for a majority of parameters, except for uric acid (−6.25%), albumin (+11.90%), chloride (–5.05%), phosphorus (−6.06%), CPK total (−57.44%), amylase (−37.53%), lipase (−42.74%), LDH (−8.53%), and CRP (−7.44%) [Table 2]. A significant positive correlation was observed between serum and plasma values of all the biochemical parameters evaluated.

Table 2 Difference between serum and plasma values of selected biochemical tests using dry chemistry autoanalyzer (Vitros-350, Ortho Clinical Diagnostics) from blood samples obtained from emergency department (n=401)

When we compare the results of our study with the maximum allowable error from College of American Pathologists (CAP) Chemistry Survey reports,[3] we observe that albumin, CPK total, amylase, and lipase – the biochemical parameters where the difference between serum and plasma samples – were more than the accepted upper range recommended by CAP.

Correlation between serum and plasma values for all the parameters evaluated in this study was statistically significant (P < 0.001), with the correlation coefficient for glucose, urea, creatinine, uric acid, TBil, DBil, indirect bilirubin, serum glutamate-pyruvate transaminase (SGPT), total protein, albumin, TC, HDL-C, TGs, sodium, chloride, phosphorus, CPK total, CPK-MB, LDH, amylase, lipase, and CRP being 0.973, 0.979, 0.980, 0.960, 0.927, 0.806, 0.919, 0.996, 0.908, 0.928, 0.952, 0.867, 0.739, 0.350, 0.655, 0.404, 0.308, 0.263, 0.911, 0.885, 0.994, and 0.996, respectively.

Discussion

Significant data are available in literature highlighting the impact of the nature of blood sample used for a biochemical analysis on the final result. A study published in 1990 highlighted that when finger prick blood was used for serum cholesterol measurement, it resulted in increased measurement of serum cholesterol with an average positive bias +2.4% as compared to serum levels, resulting in a substantial larger number of patients being labeled with hypercholesterolemia.[6] In a small study of 24 volunteers in 1994 revealed significant variance in serum and plasma levels of TC, TGs, and HDL-C on different assay platforms.[7]

Studies have demonstrated that the nature of the blood sample collected (serum vs. plasma) may have an impact even on epigenetic studies, due to their potential interference with biological processes such as DNA methylation during the sample collection.[8] Matrix metalloproteinases (MMPs) have been used as useful diagnostic or prognostic markers in different malignancies. Studies have shown that there is a significant difference in MMP levels depending on when plasma or serum samples were used for analysis, and plasma samples, in general, are preferred for analysis due to the higher levels reported from plasma samples.[9]

This is the largest ever study reported, comparing the serum and plasma levels of commonly used anaylates in the biochemistry department. The study highlights that for most of the biochemical parameters which are evaluated in the emergency department, the testing can interchangeably be done on serum or plasma samples without any significant difference in the laboratory results. These parameters, which can be freely assayed on serum or plasma samples using dry chemistry technology, include glucose, urea, creatinine, bilirubin, total protein, SGPT, total cholesterol, HDL-C, TGs, sodium, and CPK-mb. This is because the percentage difference between serum and plasma samples was consistently <±4%. Hence, even in routine clinical practice, either serum or plasma can interchangeably be used for analysis, without any impact on the outcome result.

It must be highlighted that the parameters which should be assayed preferably on serum samples, and not plasma samples due to significant associated differences, include CPK total, amylase, and lipase, as the percentage differences for this parameters were more than the cutoffs provided by CAP. For parameters such as uric acid, chloride, phosphorous, and LDH, although the difference between serum and plasma samples was more than ±4%, it was less than the cutoffs provided by CAP. Hence, only in real emergency scenarios should these parameters be evaluated from plasma samples. In routine clinical practice, we should continue to evaluate these parameters from serum samples preferably.

Conclusion

It may be said that said that serum versus plasma samples should not be interchangeably be used in routine clinical practice unless specific data for the same are available depending on the assay platform used for analysis.