Assessment of the stability of 20 biochemical analytes in serum and whole blood samples after storage at nonstandard temperatures

Abstract

Background

Blood specimen transport conditions can have critical impacts on analyte stability and test result accuracy. This study evaluated the impact of multiday storage of specimens at 30 °C and 0 °C, approximating shipment under summer conditions and shipment in contact with melting ice, respectively.

Methods

Blood samples from 16 healthy subjects were processed as serum samples and as clotted whole blood (uncentrifuged) stored at 30 °C on separator gel, and as serum stored in microtainers at 0 °C, each for 24- and 72-h periods prior to analysis. Each sample was analyzed for 20 common analytes and compared to the values from a paired baseline control sample. Mean absolute and/or relative biases were calculated and compared to CLIA acceptance limits.

Results

Serum samples stored at 30 °C for 24- or 72-h met acceptance criteria for all assessed analytes, except for 72-h ALT, Total Bilirubin, Carbon Dioxide, Creatinine, and Sodium. Numerous analytes were unstable at 30 °C in uncentrifuged whole blood, with only Albumin, ALP, Total Bilirubin, Cholesterol, HDL, Total Protein, Triglycerides, Uric Acid, and hsCRP remaining stable up to 72 h. For serum samples stored at 0 °C, all 24- hour and 72-h analytes showed biases within limits.

Conclusion

Storage of serum samples at 0 °C for up to 72 h yielded valid results for all analytes studied. Storage at 30 °C for up to 72-h yielded valid results for most analytes when combined with centrifugation prior to storage. Compared to room temperature storage, ALT and Sodium in serum and whole blood showed signs of accelerated degradation at 30 °C.

Highlights

• •Blood specimens often require shipment to a central laboratory for testing.
• •Temperature excursions during storage/shipment can impact analyte stability.
• •Limited data is available on analyte stability at 30 °C and 0 °C conditions.
• •Storage of serum at 0 °C up to 72 h yields valid results for common analytes.
• •Centrifugation prior to storage at 30 °C increases the stability of several analytes.

1. Introduction

Collection of blood specimens can be performed in a variety of settings including hospitals, clinics, dedicated collection sites, nursing facilities, and even patient homes. While blood collection can be accomplished with limited staff, mobile phlebotomy services, and self-collection [1], the specimens are often shipped to a central laboratory for testing. In the absence of specialized packaging, the shipping process can result in samples being exposed to a range of temperatures for varying amounts of times [2]. Temperature excursions during storage and/or shipment can have critical impacts on analyte stability and the accuracy of patient results [[3], [4], [5]].

While the analytical impact of storage or transport of blood or serum at certain, easy-to-implement temperatures (e.g., −20 °C, 4 °C, 20–25 °C) has been well characterized, the impact of storage or transport at other temperatures has rarely been examined [4,[6], [7], [8], [9], [10], [11], [12], [13]]. This study aims to address this gap in the current literature by assessing the analytical impact of one- and three-day storage of serum and uncentrifuged whole blood samples at 30 °C (e.g. summer or elevated ambient conditions). Furthermore, this study evaluates blood specimens centrifuged by a field-portable centrifuge [14] and stored as serum in gel microvolume tubes at 0 °C (e.g., in direct contact with ice packs) as a specific potential configuration for shipment of microsamples collected at-home or in other non-clinical locations. These results could be helpful to researchers, clinicians, and central laboratories to better understand the impact that storage and/or shipping of blood at alternative temperatures may have on several analytes commonly measured in serum samples.

2. Methods

Venous blood was collected from each of sixteen reportedly healthy subjects under an IRB-approved protocol (Protocol #2; Salus IRB, Austin, TX) into four BD Vacutainer® Serum Separator Tubes (Tubes A-D; SKU 367986; Becton, Dickinson and Company Life Sciences, Franklin Lakes, NJ). All tubes were inverted and allowed to clot for 30–60 min at room temperature (20–25 °C). Tubes A and B were separated into serum by a swinging bucket centrifuge (Thermo Fisher Sorvall™ ST 40R; Thermo Fisher Scientific, Waltham, MA) for 10 min at 1500 g, while tubes C and D were left as uncentrifuged, clotted whole blood. A 500 μL serum aliquot was immediately removed from Tube A and pipetted into a screw top transport tube (Item Number 6101S; Globe Scientific, Mahwah, NJ) to serve as baseline control. This and all subsequently described test aliquots were frozen and stored at −20 °C until analysis could be performed. After removing the test aliquot, Tube A was recapped and stored under refrigerated conditions (2–8 °C). Tubes B, C, and D were transferred to a controlled incubator (Yamoto DX402 Oven) set to 30 °C and maintained within ±1 °C.

After a 24-h (±2 h) post-collection incubation, a test aliquot was removed from each of tubes A (refrigerated) and B (30 °C incubated). Tubes were recapped and returned to their respective storage temperatures. Tube C was separated into serum by centrifugation and an aliquot of serum was removed and frozen for later analysis. After a 72-h (±4 h) post-collection incubation, aliquots from tubes A and B were removed and frozen for later analysis. Tube D was separated into serum by centrifugation, then an aliquot was removed and frozen for later analysis. An overview of the workflow can be found in Fig. 1.

Fig. 1.

Flowchart depicting the handling and storage conditions of the 5 vacutainer tubes drawn from each of 16 subjects, along with points in the process where aliquots were obtained for the 9 conditions analyzed.

In order to simulate potential conditions for shipment of microsamples collected in non-clinical settings, an additional tube of venous blood was collected (concurrent with Tubes A-D) in a BD Vacutainer® Uncoated Tube (Tube E; SKU 366408; Becton, Dickinson and Company Life Sciences, Franklin Lakes, NJ) and immediately aliquoted into two SST Microtainer® blood collection tubes (600 μL volume each; SKU 395967; Becton, Dickinson and Company Life Sciences, Franklin Lakes, NJ). After inversion, Microtainers® were allowed to clot for 30–60 min. Microtainers® were then separated into serum by centrifugation for 10-min at 1500 g using the Labcorp TrueSpin™, a compact, battery-powered portable centrifuge system [14]. Microtainers® were put in biohazard bags and placed inside a 30 oz, custom, hard-shelled vacuum tumbler (the Labcorp TrueTherm™ TL0 packaging system [15]). Inside the TrueTherm™ tumbler were aqueous cold packs filled with a 250 mL mixture of water and 2.0 % superabsorbent polymer (sodium polyacrylate; Sigma Aldrich, St. Louis, MO) that had been stored overnight in a freezer. The TrueTherm™ was sealed and put in a refrigerator to maintain an internal temperature of 0 °C for 72-h. A representative 72-h thermal trace from the interior of a TrueTherm™ TL0 can be found in Supplemental Fig. 1. After 24 h (±2 h), one Microtainer® was removed from incubation and frozen at −20 °C for later analysis. After 72 h (±4 h), the remaining Microtainer® was removed from incubation and frozen at −20 °C for later analysis (Fig. 1). In total, the samples were kept at −20 °C for a maximum of 16 days.

Once all samples for the study were collected and processed, the frozen test aliquots were packed with dry ice in an insulated package and shipped overnight via FedEx to Labcorp (Burlington, NC) for analysis. Once received, samples were transferred to a −80 °C freezer for less than two weeks, until testing could be performed. On the day of analysis, the specimens were thawed unaided at room temperature (20–25 °C) for at least 1 h, then centrifuged at 6000 g for 2 min (regardless of prior centrifugation), and the resulting serum was transferred to an appropriate measurement vessel. A panel of 20 common chemistry analytes were then measured in the serum samples using Roche cobas® 8000 autoanalyzers (Roche Diagnostics, Indianapolis, IN): albumin, alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin, blood urea nitrogen (BUN), calcium, carbon dioxide, cholesterol, chloride, creatinine, glucose, high-density lipoprotein (HDL), phosphate, potassium, sodium, total protein, triglycerides, uric acid, and C-Reactive Protein (hs-CRP). These analytes were chosen because they represent some of the most commonly ordered tests in clinical diagnostics, as well as a variety of assay types. A list of the assays utilized can be found in Supplemental Table 1. Samples were run in singlicate; all samples from the same individual (i.e., baseline controls and other timepoints) were processed on the same day and instrument. All study samples were tested on the same day, in two batches, and all quality metrics were within the internal acceptance ranges utilized for testing of patient samples.

For each storage condition, the absolute or relative biases of measurements from samples were determined with respect to paired baseline control values, and biases were compared to CLIA acceptance limits [16]. For each analyte and storage condition, agreement was determined by dividing the number of stability sample results within acceptance limits by the total number of results. Acceptable agreements were defined as greater than two-thirds (i.e., 66.7 %) of results within acceptance limits.

3. Results

3.1. Analyte stability after storage at 30 °C

For serum specimens centrifuged following collection and stored at 30 °C for 24 h, mean biases observed for all 20 analytes were within CLIA acceptance limits. Mean biases observed for centrifuged serum samples stored at 30 °C for 72 h were within CLIA acceptance limits for 18 of 20 analytes; the two analytes out of range were carbon dioxide and creatinine. Additionally, three analytes (ALT, total bilirubin, and sodium) produced mean biases within limits, but agreement was less than two-thirds (Table 1). Bias plots for each analyte (which include a visual representation of the standard deviation for each analyte) can be found in Supplemental Fig. 2, and regression plots for each analyte over time can be found in Supplemental Fig. 3.

Table 1.

Summary of mean biases of serum and clotted whole blood (uncentrifuged) samples stored under 30 °C conditions on separator gel, and standard conditions (serum refrigerated at 4 °C) relative to paired baseline control samples (time 0, serum samples). A value in green indicates a condition in which the mean bias was within CLIA acceptance limits, and at least two-thirds of individual samples were in agreement; a value in yellow indicates a condition in which the mean bias was within CLIA acceptance limits, but less than two-thirds of individual samples were in agreement; a value in red indicates a condition in which the mean bias was outside CLIA acceptance limits. Note: The percent (%) difference was used for all biases when available in the guidance document, including when a second form of bias limit was provided.

ALP = alkaline phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase; BUN = blood urea nitrogen; HDL = high-density lipoprotein; CRPhs = Cardiac C-Reactive Protein (high sensitive).

For aliquots derived from uncentrifuged whole blood samples stored at 30 °C for 24 h, mean biases for each analyte were within range for 15 analytes; the 5 analytes out of range were: ALT, creatinine, glucose, phosphorus, and potassium. Additionally, chloride was found to have a mean bias within limits, but less than two-thirds of individual samples were in agreement. Mean bias values for aliquots derived from uncentrifuged whole blood stored at 30 °C for 72 h were within CLIA acceptance limits for 11 of 20 analytes; the nine analytes with mean biases out of range were: ALT, BUN, calcium, chloride, creatinine, glucose, phosphorus, potassium, and sodium. Additionally, for two analytes (AST and carbon dioxide), mean biases were within limits but less than two-third of individual samples were in agreement (Table 1). Bias plots for each analyte can be found in Supplemental Fig. 2, and regression plots for each analyte over time can be found in Supplemental Fig. 3.

3.2. Analyte stability after separation and storage in microvolume tubes at 0 °C

For specimens separated using the TrueSpin™ portable centrifuge following collection and stored as serum within Microtainers® at 0 °C for 24 h prior to freezing, all mean bias values were within CLIA acceptance limits, though for potassium, less than two-third of individual samples were in agreement. Mean bias values of specimens stored as centrifuged serum within Microtainers® at 0 °C, for 72 h were within CLIA acceptance limits for all analytes. Note that one of the measured CRPhs values for the 0 °C, 72hr condition was excluded from calculation of the mean bias as an extreme outlier (with a measured value 526 % of baseline), likely representing an analytical error (Table 2). Bias plots for each analyte can be found in Supplemental Fig. 2, and regression plots for each analyte over time can be found in Supplemental Fig. 3.

Table 2.

Summary of mean biases of serum and clotted whole blood (uncentrifuged) samples stored under 0 °C conditions on separator gel, and standard conditions (serum refrigerated at 4 °C) relative to paired baseline control samples (time 0, serum samples). A value in green indicates a condition in which the mean bias was within CLIA acceptance limits, and at least two-thirds of individual samples were in agreement; a value in yellow indicates a condition in which the mean bias was within CLIA acceptance limits, but less than two-thirds of individual samples were in agreement; a value in red indicates a condition in which the mean bias was outside CLIA acceptance limits. Note: The percent (%) difference was used for all biases when available in the guidance document, including when a second form of bias limit was provided.

∗One hs-CRP value was considered an extreme outlier (measured at 526 % of baseline), likely representing an analytical error, and was excluded from the calculation of mean bias.

3.3. Hemolysis levels

Hemolysis levels were measured for specimens under each storage condition and reported as the hemolysis index (HI). HI is a dimensionless measure generated by the Roche cobas® 8000 analyzer to estimate the approximate concentration of hemoglobin in a specimen, with one HI unit corresponding to approximately 1.25 mg/dL [17].

The maximum allowable HI for each analyte is shown in Supplemental Table 1; the lowest of these values is for AST (with a maximum allowable HI of 40). Using this value (HI = 40) as a benchmark, all control samples, refrigerated serum samples, and serum samples stored on separator gel at 30 °C had HI within allowable limits. One clotted whole blood sample (uncentrifuged) stored on separator gel at 30 °C, and four serum samples stored on separator gel at 0 °C, fell above the HI = 40 threshold (Fig. 2).

Fig. 2.

Hemolysis indices as measured by the Roche cobas® 8000 analyzer, stratified by storage condition. The maximum allowable hemolysis index (HI) for each analyte is available in Supplemental Table 1. The dotted line at HI = 40 indicates the maximum allowable HI for AST, which is the lowest maximum allowable value for the analytes studied. Samples with an HI above the maximum allowable value may yield anomalous results due to interference from hemolysis in the specimen but are still reported.

4. Discussion

Blood specimens are ideally stored and transported within a well-defined temperature range and reach the laboratory for analysis within a short period of time in order to maintain the integrity of the specimen. Analyte stability is typically validated within specific temperature ranges that are readily available in a clinical laboratory, including standard refrigerated, frozen, and ambient conditions. However, when specimens are collected at decentralized locations and shipped to regional laboratories, extremes in weather and shipping delays can result in the specimens being exposed to conditions outside of the standard temperatures used for validation. The heavy insulation and phase change materials required to control temperatures within narrow ranges for multiple days can add considerable cost to the processing of specimens shipped to central laboratories. Therefore, it is important to understand how analyte stability may be impacted by a broader range of storage and/or shipping conditions and whether higher or lower temperature ranges are tolerable for analytes of interest.

It is expected that non-insulated specimens shipped by commercial freight will commonly encounter summer temperatures around 30 °C [18]. On the other hand, the most common and cost-effective temperature controlled packaging comprises insulated packages containing ice packs. Specimens in direct contact with frozen icepacks experience temperatures around 0 °C as the ice slowly melts (or slowly re-freezes under sub-zero winter conditions). Although common packaging systems may seek to separate specimens from direct contact with packs, this strategy complicates the packing procedure, reduces cargo space, is prone to user error, and does not prevent excursions to 0 °C or slightly below under winter conditions. This study evaluated 20 analytes after storage at 0 °C and 30 °C – temperatures not commonly examined in prior studies.

In line with studies conducted at 20–25 °C, centrifugation played a role in maintaining the stability of several analytes stored at 30 °C. In general, separation of serum from cellular components shortly after draw mitigates the impact of continued cellular metabolism, and the exchange of analytes across cellular compartments [19]. All 20 analytes were relatively stable (within CLIA acceptance limits) after 24 h of storage at 30 °C in samples centrifuged immediately following draw. After 72 h of storage at 30 °C some analytes that are typically stable at 20-25 °C began to show signs of degradation in separated serum samples. Two analytes (carbon dioxide and creatinine) showed mean biases outside CLIA acceptance limits, and three additional analytes (ALT, total bilirubin, and sodium) showed more than one-third of individual samples outside bias limits (despite mean biases within acceptable limits). Two analytes: ALT and sodium showed signs of instability despite previously being found to be stable at 20-25 °C over similar time periods in serum samples [7,8,[10], [11], [12], [13]]. This may indicate that these analytes are sensitive to extended elevation above room temperature, even in separated serum specimens. ALT has been found to be unstable (decreasing) in serum over 7 days at room temperature, so it is possible that the increased temperature accelerates the degradation process [12,13,20].

The decreased carbon dioxide and increased creatinine levels observed at elevated temperatures are generally consistent with previous studies of these analytes [11]. Carbon dioxide levels are known to decline over time, especially with exposure to air [21]. Creatinine levels (measured by a Jaffee assay) may appear increased with time and temperature, with a prior study hypothesizing that this may be due to non-specific formation of pseudocreatinines or non-creatinine chromagens during storage [11,22]. With regard to bilirubin, prior studies suggest that bilirubin is relatively stable at various temperatures (including temperature elevations) but is highly sensitive to light. Therefore, the decrease observed in direct bilirubin samples could be related to photodegradation (from possible light exposure) during sample handling, rather than the elevated temperature [23].

Though 14 of 20 analytes from uncentrifuged samples were within CLIA acceptance limits after one day of storage at 30 °C, five analytes showed unacceptable mean bias deviations – specifically, elevations in ALT, creatinine, phosphorus, and potassium, and declines in glucose levels outside CLIA limits – and one analyte (chloride) showed more than one-third of individual samples outside bias limits. Stability of serum samples stored for 24 h at 30 °C was comparable to that found in prior studies for whole blood specimens held at room temperature, with the exception of ALT, which may show signs of accelerated degradation due to temperature [24]. Following three days of storage at 30 °C, only 9 of 20 analytes in uncentrifuged samples were within acceptable limits; four additional analytes showed unacceptable mean biases (including an elevation in BUN, and declines in calcium, chloride, and sodium), and one analyte (carbon dioxide) showed more than one-third of individual samples outside bias limits. Collectively, these findings are largely consistent with previous studies, which found that potassium, inorganic phosphate, calcium, glucose, creatinine, and BUN were among the analytes “most influenced by delayed centrifugation” [9]. In contrast to a previous study of whole blood stored at room temperature for 72 h, the current study found that sodium, AST, and ALT may become unstable at this timepoint when stored at 30 °C compared to 20–25 °C [10].

In unseparated samples, glycolysis (the metabolic conversion of glucose into pyruvate) continues to occur by red blood cells present in serum and occurs at a faster rate in the presence of elevated temperatures. Because of ongoing glycolysis in samples that are not centrifuged following draw, glucose levels decline rapidly. Centrifugation is crucial to halting glycolysis from occurring in serum. Within hours of draw, glycolysis in an unseparated sample slows and halts, at which point ATP production also ceases. ATP powers the sodium-potassium pump in cell membranes, so when this occurs, potassium passively diffuses out of cells, resulting in increasing levels in serum [25]. Similarly, sodium levels in serum may decrease at this point, as the concentration gradient drives passive diffusion of this ion into cells. Increases in ALT and phosphorus levels may be caused by changes in cell permeability and cell lysis with time [19,25,26].

The second arm of this study focused on analyte stability and hemolysis levels for samples stored at 0 °C. The 0 °C storage condition is relevant for the design of minimalist thermal protection systems, helping to determine whether cooldown to 0 °C or slightly below must be prevented. The Microtainers®, portable TrueSpin™ centrifuge, and TrueTherm™ packaging systems used in this arm were intended to simulate a potential collection workflow for microsample collection in a non-clinical setting. If analyte stability can be maintained for samples that reach 0 °C or slightly below, then packaging that incorporates simple, water-based ice packs can be used rather than expensive, winter protection PCMs and/or supplemental insulation to prevent direct contact between specimens and icepacks. The use of standard ice packs without engineered partitioning could streamline the workflow for at-home sample collection.

It should be noted that the 0 °C studies did not include uncentrifuged samples. This is because it is well established that ice formation causes rupture of red blood cells, leading to severely elevated levels of hemolysis [27] and rendering whole blood samples subject to temperatures near or below 0 °C largely untestable. However, hemolysis levels and analyte stability have not been well studied for separated serum in gel tubes (such as the Microtainers® used in this study) stored at 0 °C. Therefore, this study focused on separated samples, postulating that separation into serum would remove the majority of RBCs from the serum and would protect the sample from excessive hemolysis at 0 °C.

Although hemolysis was modestly elevated in specimens stored at 0 °C, the great majority of these were well within the allowable range for all the analytes studied. All mean bias values for 24- and 72-h analytes were within CLIA acceptance limits and only one analyte condition (Potassium at 24 h) showed a notable portion of individual specimens outside of CLIA limits. These results suggest that a minimalist thermal protection system that places specimens in direct contact with ice may be suitable for separated serum specimens in gel microtubes. Please note that CLIA acceptance criteria are relevant for clinical laboratories in the United States, but not necessarily applicable in other area of the world. An analysis of study data based on other criteria is outside the scope of this manuscript but represents an opportunity for future research.

One limitation of this study is that specimens from reportedly “healthy” subjects were used to evaluate analyte stability. Given the sample size of the study (n = 16), these healthy subjects are unlikely to cover the analytical measurement range (AMR) for all analytes. Additional studies may be helpful in determining whether storage at 30 °C and 0 °C could have more dramatic impacts on analyte concentrations in samples from individuals with various medical conditions and spanning the AMR of all analytes.

Another limitation of the study was that aliquots from the various timepoints and the Day 0 paired controls were frozen and stored (at −20 °C, and then −80 °C) prior to analysis; in total, these specimens were frozen for less than one month. Certain analytes (specifically, ALT, carbon dioxide, and glucose [4]) may be unstable in serum when stored under these conditions, which could have impacted results. Of note, ALT was within range for all baseline control (Day 0) samples, and all but one carbon dioxide level was slightly below normal range in these specimens. Additionally, all samples were centrifuged prior to being frozen, which is expected to mitigate the impact of storage on glucose levels. Therefore, any potential effect from frozen storage was likely to be minimal in this study.

Additionally, demographic information was not collected from the donor subjects for this study. Therefore, results could not be analyzed or stratified based on subject characteristics.

A final limitation was that sample separation was performed using a benchtop centrifuge and Vacutainers® for the 30 °C conditions, whereas a TrueSpin™ portable centrifuge and Microtainer® samples were used for the 0 °C conditions. Though both benchtop and TrueSpin™ centrifuges were spun at the same g-force (1500 g), the recommended g-force for Microtainers® is higher than that recommended for Vacutainers®. Therefore, the TrueSpin™ may have been underpowered for separation of the 0 °C specimens – though results suggest the level of separation achieved by TrueSpin™ was sufficient for most samples.

The findings from this study suggest that storage of blood at 0 °C for up to three days is a viable option for separated samples for the majority, if not all, of the analytes studied. Storage at 30 °C for up to three days appears to be a viable option for many common analytes when combined with centrifugation prior to storage. However, sodium and ALT were identified as analytes potentially sensitive to elevated temperature in serum or whole blood over 72 h. Increasing the breadth of acceptable storage temperatures and identifying analytes that are sensitive to deviations from standard temperatures are important steps for the decentralization of blood collection and may provide a wider range of testing that is available to patients, particularly in resource limited settings.

CRediT authorship contribution statement

Gabriella Iacovetti: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Investigation, Formal analysis, Data curation. Bradley B. Collier: Writing – review & editing, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jill M. Rafalko: Writing – review & editing, Writing – original draft, Visualization. Mitchell Peevler: Writing – review & editing, Visualization, Project administration, Investigation. Nicolas Tokunaga: Writing – review & editing, Visualization, Validation, Investigation, Data curation. Jason Ragar: Writing – review & editing, Investigation. Whitney C. Brandon: Writing – review & editing, Visualization, Validation, Investigation, Formal analysis, Data curation. Matthew R. Chappell: Writing – review & editing, Visualization, Validation, Investigation, Formal analysis, Data curation. Russell P. Grant: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Greg J. Sommer: Writing – review & editing, Visualization, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Ulrich Y. Schaff: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Funding sources

Funding support for this study was provided by Labcorp.

Declaration of competing interest

All authors are employees of Labcorp. GI, BC, JMR, MP, JR, MC, RG, GS, and US own stock in Labcorp. Labcorp has filed a U.S. Patent Application (No. 17/229,460) covering the TrueSpin™ portable centrifuge system discussed in this manuscript; GS and US are listed inventors on this patent. GI received funding from Labcorp to travel to the ADLM annual meeting to present a poster related to this work. RG received a Distinguished Contribution award from MSACL, which included a monetary award, as well as flight and travel expenses to accept the award. RG serves on the advisory board of ADA DX.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.