Automatic for the people: an introduction to self-collected capillary blood testing

Although CBT has been used for decades, most consumables are based on earlier pediatric sample-collection applications that have been available for many years, and many are still intended for professional use only. These consumables, however, have been successfully co-opted for self-collection. Recent innovations in the sample collection space include transdermal devices, which have improved both access and usability and are approved for self-collection [13]. Although they are more expensive than simple finger-prick consumables, they allow the layperson to produce a higher-quality sample [13] (Table 1).

Capillary blood sampling is routinely performed using the fingertips of the middle and ring fingers. A lancet with sufficient depth is used to open the finger capillary network; the first drop of blood is wiped away, and subsequent drops are collected into the blood tube. Although straightforward, the self-collection process may be challenging for some patients, particularly those with limited dexterity. To make collection easier, many kit suppliers now provide a sample holder integrated into the packaging; however, other options are also available. The recently released SelfSafeSure Blood Collection Assist Device (BCAD)™ includes both a sample funnel and a secure tube base [22]. The collection process has been successfully tested with self-collected samples across a range of analytes [23].

Becton Dickinson (BD) has also released the BD MiniDraw™ Capillary Blood Collection System, which has shown promising results [24]. This innovative collection device combines a finger sleeve and a new tube format. The sleeve acts as a tourniquet, a lancet guide, and a tube holder, whereas the new tube type is designed for upside-down centrifugation. The finger sleeve gently pools blood in the fingertip, helps the user position the lancet, and allows the blood tube to slide into place during collection. Unfortunately, each device is designed for single-use with one tube, and the finger sleeve must be sized to fit the patient's finger (available in small, medium, large, and extra-large). Likewise, the MiniDraw is approved for professional use only.

Although CE-marked for professional use only, both BD's Microtainer [18] and Greiner's MiniCollect series [20] are popular self-sampling solutions, and the BD Microtainer tubes have become the standard for use with transdermal devices (see Table 1).

Additionally, all three manufacturers offer pressure-activated lancets with sufficient depth (e.g., 2 mm) to obtain a high-quality sample. However, key differences between the tubes may affect their operational suitability. This is especially important when large sample volumes are required, as the Greiner MiniCollect tube enables the collection of larger blood volumes.

The Microvette by Sarstedt [21] offers a lower-volume capillary tube for serum and lithium heparin plasma, which may limit its use in some cases; however, the Microvette EDTA automated processing tube (APT) can be handled directly on some haematology analysers and, as such, can be treated in the same way as venous tubes (Table 2).

Recent advances in self-collection methods have led to the introduction of several new transdermal devices [13,14]. Some, however, are either not yet available for commercial use or, when going to press, did not have CE marking; these include the Onflow from Loop Medical (Lausanne, Switzerland), ezdraw™ from Preci-Health (Neuchâtel, Switzerland), ImPress from LetsGetChecked (Dublin, Ireland), and RedDrop from RedDrop Dx (Fort Collins, USA). Although the RedDrop device is commercially available and widely used in the USA [13]. Those that do have a CE mark include the TAP II (YourBio Health, Medford, USA) and Tasso+ (Tasso Inc., Seattle, USA). Given the additional validation steps required for using non-CE-marked devices, we will focus on those already approved for use in the EU/UK marketplace.

The Touch Activated Phlebotomy (TAP) device is a minimally invasive, transdermal blood-collection device that uses a gentle vacuum and microneedle HALO array technology to collect capillary blood samples.

The TAP device is placed on the upper arm, and a peel-off self-adhesive seal secures the device in place before activation. The device is then activated by pressing the plunger, which forces a microneedle array into the upper dermal layer. Negative pressure is created, increasing blood flow from the skin into the collection tube. TAP device utilises BD Microtainer tubes and is available in the following varieties: EDTA, serum, serum gel, heparin plasma and heparin plasma gel. This device has been evaluated in several studies, including immune-mediated rheumatic disease, Anti-Müllerian Hormone, and carcinoembryonic antigen (CEA) [25–27].

The Tasso+ device is also placed on the upper arm and secured with a self-adhesive seal. After activating the lancet by pressing the red button, negative pressure is created, drawing blood from the skin into a BD Microtainer. The Tasso+ has also been used in several studies, including assays for COVID-19 antibodies, autoantibodies, liver, kidney and lipid markers [28–31].

The pre-analytical phase of self-collected capillary blood is primarily outside the laboratory's control. It is therefore critical that collection kit IFUs are tested with the user cohort in mind; a patient may not be using a self-collection device for the first time and may only need a reminder, whereas a patient unfamiliar with the collection process will require more detailed instructions and/or additional guidance.

A delicate balance needs to be achieved between hard-copy instructions, in-person training and online content. The patient cohort will inevitably influence this. Engagement with patient-representative groups and caregivers would be advantageous here. For example, instructions in relevant languages or tailored for patients with learning disabilities or pediatric scenarios can be better designed when the end users are involved in their development.

Although self-collected CBT was designed for samples to be returned to the laboratory via the postal network, this need not be the route in every case. Delivering samples in person to a designated drop-off point, such as a primary care clinic, will ensure they are collected alongside samples taken within the clinical setting. In this case, sample stability may be within those listed in the assay IFU. For samples returned via the postal service, analyte stability in whole blood will need to be evaluated over expected timeframes and transit temperatures.

There is a growing body of published work on CBT using standard capillary tubes and finger-prick collection. Published data include comparisons between capillary and venous blood for full blood count [32], enzymatic creatinine [33], urea, liver function tests [7,34], lipids [31], C-reactive protein (CRP), HbA1c [7,8], enzymatic creatinine [33], prostate-specific antigen [35], Anti-Müllerian Hormone, luteinising hormone, follicle-stimulating hormone, oestradiol, testosterone, prolactin, dehydroepiandrosterone sulphate (DHEA-S), sex hormone-binding globulin, thyroid-stimulating hormone and free thyroxine (FT4) [36]. Some of these analytes are highly robust, whereas others can be affected by even low levels of degradation [37]. Given this, it is advised that the laboratory undertake whole-process verification, including sample stability experiments using whole blood stored at the expected temperatures over a set period.

In the UK, samples must be shipped via Royal Mail Tracked 24. It has been found that over 95% of samples are returned within 72 h using this service (own unpublished data). Therefore, it is advisable to conduct stability studies relevant to the observed transit time.

Ideally, each sample kit would include a temperature logger. Although temperature damage usually results in haemolysis and is therefore detected by the serum indices assay, this is not always the case, and some analytes are unstable at higher temperatures, e.g., HbA1c [37].

Unfortunately, temperature loggers can be challenging for laypeople to use. However, several options are available, including the Varcode Smart Tag [38], the Tive Tag [39] and the TimeStrip [40].

The Varcode Smart Tag uses a barcode made of thermolabile ink that changes in response to temperature and duration. The Tive Tag is a passive temperature data logger that the user activates and collects real-time data via an online application. Data are then available on the cloud once the parcel is re-scanned at the laboratory. Simpler and cheaper options include liquid-based irreversible temperature indicators such as the Timestrip. These are manually activated and read, but they only indicate temperature breach and do not provide data points. They are, however, easy to use and inexpensive, and have a long track record in blood product monitoring, although only limited temperature ranges are available.

Currently, CBT utilises repurposed paediatric tubes, which in most cases were not designed to be placed directly onto automated laboratory analysers. To protect the analyser probes from short samples or gel contamination, a secondary tube may be required. As yet, there are limited off-the-shelf total laboratory automation solutions for capillary blood samples, and modern pre- and post-analytical units struggle to process capillary tubes [14].

Adequate labelling of secondary tubes is also often problematic in a routine hospital laboratory. Despite this, incorporating self-sampling CBT into the routine diagnostic pathway is relatively straightforward, and laboratories require minimal additional resources to integrate these samples seamlessly into their standard workflow [14].

Depending on the requested tests, capillary samples may need to be manually checked. Although most haematology analysers can process capillary samples [41], the collection process itself raises concerns about clotted samples. In such cases, to prevent analyser downtime, it may be necessary to check each sample, which can hinder rapid throughput. Clotting is not an issue for most chemistry analyses, but the small sample volume obtained may be.

Analysers with fixed-sample probes risk blockage if the probes enter the gel of the serum separator tube or the cellular layer. When disposable pipette tips and sample reaction vessels are used, this is generally not a concern, and it may be possible to sample directly from the primary tube. If this is not the case, it may be necessary to aliquot the sample into a secondary analyser cup. Without automation, this process is likely to be manual, but it does allow for a physical sample check before analysis. At that point, samples that are insufficient in volume or grossly hemolyzed can be rejected.

Once the sample is placed in an analyser cup, it can be treated much like any other sample. However, the number of tests available to the clinician will depend on the sample volume obtained, the efficiency of serum/plasma recovery, and the laboratory's analyser methodology.

Although most chemistry analysers use similar sample volumes for basic chemistry profiles (e.g., liver and kidney function, lipids), much larger volumes are sometimes required for immunochemistry assays, e.g., the Roche e801 uses 12 ul for Total PSA (not including dead volume), while the Beckman Dxl uses 25 ul.

Due to the nature of this technology, the sample collection process is literally in the hands of a layperson and, as such, difficult to fully control. Result interpretation, therefore, must also account for sample contamination from various sources, including the collection site (e.g., skin). With CBT, anything on the skin or within the capillary network at the collection site is likely to be transferred to the blood tube. High concentrations of biotin (vitamin B7) have been detected in the capillary network following topical cream use [42], and although many assays have recently been reformulated to include biotin scavengers, biotin interference could still be an issue in some immunoassays [43]. Given this, other potential interfering substances are likely to emerge as CBT becomes more widely used across various patient cohorts. For example, creams containing large amounts of coconut oil would be expected to falsely increase triglyceride results, and patients using topical hormone replacement therapy may exhibit falsely high hormone levels. These issues may be overcome by including additional guidance in user instructions.

It has also been found that serum separator tubes may falsely lower some analyte results compared with plain serum tubes when testing is delayed, as seen with postal samples [36]. This has been reported for oestradiol and testosterone and is believed to result from these hormones binding to the gel, an effect exacerbated by the higher gel-to-blood ratio in serum separator capillary tubes than in venous tubes. The decrease may be more pronounced with extended contact times [36]; therefore, it is critical to verify comparability between venous and capillary samples using realistic stability timelines, and laboratory scientists must have an appreciation that interference from unknown substances with capillary blood may come from both the site and the collection process, as well as other well-known routes such as EDTA contamination.

The use of capillary blood self-sampling devices has the potential to benefit several patient cohorts. These include patients with learning disabilities, needle-phobia, those with poor venous access, patients in remote or underserved areas, and those requiring regular long-term monitoring (e.g., therapeutic drug and chronic disease monitoring).

In some instances, a business case for CBT applications may be required to implement the service. The savings from the introduction of CBT are spread across the whole patient journey, and it is appreciated that health services do not always allocate budgets in this manner. However, a business case should include the costs of devices, administration, transport, environmental impact, workflow changes (including pre- and post-analytical processing requirements), as well as the cost-benefit to the patient [3]. The early involvement of patient groups will be essential when introducing a CBT service.

Applications have been demonstrated for prostate disease monitoring [35], immunosuppressant and creatinine measurement in transplant recipients [44–48], therapeutic drug monitoring for inflammatory bowel disease [49], and anti-psychotic drug monitoring [50]. In addition, CBT has been used for HbA1c analysis [51], C-peptide testing for classification of diabetes subtype [52], chronic respiratory disease monitoring [4] and steroid monitoring [36,53]. For a more comprehensive list, see Schröder et al., who recently conducted a meta-analysis comparing capillary and venous samples with relevant stability data [54].

Most clinical laboratory tests are verified for use only with venous blood. Therefore, validation is necessary for non-conventional sample types, including CBT. The Medicines and Healthcare products Regulatory Agency (MHRA) has stated that unless the manufacturer has explicitly listed the sample type being used (e.g., capillary blood), the assay is being used off-label [55]. Therefore, concordance between venous and capillary blood must be evaluated using a suitable analytical method. The Clinical and Laboratory Standards Institute EP35 guideline (Assessment of Equivalence or Suitability of Specimen Types for Medical Laboratory Measurement Procedures) [56] offers recommendations for assessing clinically equivalent performance across similar-matrix specimen types. It is also noted that the LabMed UK Patient-Centred Testing and Sampling (PaCTS) Interest Group is in the process of publishing its recommendations on the validation and adoption of CBT in routine clinical laboratories, which will no doubt help laboratories navigate the complexities of validation versus verification.