Taking a step back from testing: Preanalytical considerations in molecular infectious disease diagnostics

Abstract

Recent studies evaluating the preanalytical factors that impact the outcome of nucleic-acid based methods for the confirmation of SARS-CoV-2 have illuminated the importance of identifying variables that promoted accurate testing, while using scarce resources efficiently. The majority of laboratory errors occur in the preanalytical phase. While there are many resources identifying and describing mechanisms for main laboratory testing on automated platforms, there are fewer comprehensive resources for understanding important preanalytical and environmental factors that affect accurate molecular diagnostic testing of infectious diseases. This review identifies evidence-based factors that have been documented to impact the outcome of nucleic acid-based molecular techniques for the diagnosis of infectious diseases.

1. Introduction

Laboratory testing is a complex multistep process susceptible to variability. Factors that can lead to inconsistent results are as simple as environmental variables such as time and temperature in addition to interindividual differences (regarding both patient and sample handling). Variability in these factors is unavoidable throughout the testing process. However, when these factors are not controlled or accounted for, they are capable of compromising specimen integrity, and even altering the result outcome. This influence is pronounced in the preanalytical phase, which includes steps like specimen ordering, collection, processing, storage, and transport (Fig. 1 ). Therefore, special considerations should be taken to minimize the influence of such factors. Manufacturers play a key role in mitigating the effect of preanalytical error by providing detailed instructions on sample processing. However, it is not unusual for laboratories to make modifications to manufacturer instructions to optimize workflow or reagent utilization. With a better understanding of how these general preanalytical factors can potentially influence testing, laboratories can make informed decisions when performing additional validation studies and defining sample acceptability criteria for a given assay.

Fig. 1.

Preanalytical processing variables that can impact the outcome of laboratory tests that rely on molecular techniques for the diagnosis of infectious diseases.

Preanalytical errors are quite common and can have significant impact on patient care and outcomes (Table 1 ). There are many reviews describing the impact of preanalytical errors on automated diagnostic assays in the clinical chemistry laboratory. Many of these assays rely on spectrophotometry, electrochemistry, and immunoassay technologies— and have distinct criteria for preanalytical processing from molecular diagnostics. Molecular diagnostics involve the amplification, detection, and often quantitation of nucleic acids. Current nucleic-acid based technologies used for diagnosis of infectious disease include: mono- and multiplex polymerase chain reaction (PCR), microarray panels, peptide nucleic acid fluorescent in situ hybridization (FISH), magnetic resonance-based detection, and next generation sequencing (NGS) [1].

Table 1.

Summary of preanalytical variables that can impact outcomes of molecular testing for the detection of infectious diseases.

Variable	Effect	Common Examples	Minimization	References
Specimen collection containers	Collection container additives can inhibit nucleic acid amplification.	Heparin	Follow manufacture recommendations for sample collection containers and/or perform validation studies to confirm sample integrity in the presence of additives.	[29], [49]
Specimen contamination	Risk for false positive results due to sample contamination.	Not changing gloves.	Appropriate work-flow and engineering controls. Transition to “closed” reaction platforms to minimize human manipulation.	[59], [60], [114]
Endogenous and exogenous inhibitors	Samples contain endogenous or exogenous compounds that inhibit enzymatic reactions for nucleic acid amplification. Presents an issue for direct sample analysis methods.	Endogenous inhibitors: IgG, hemoglobin, heme metabolites, lactoferrin, and antiviral substances (i.e. acyclovir) Exogenous inhibitors: proteases, nucleases	Proper extraction and purification can remove inhibitors. Appropriate sample collection can also minimize presence of exogenous inhibitors.	[66]
Time, temperature, and freeze–thaw cycles	Nucleic acid targets have variable stability under different temperature conditions depending on sample type and pathogen.	Samples being exposed to extreme cold, freezing and then thawing during transport or storage.	Perform validation studies to confirm sample integrity under standard and anticipated sample processing conditions.	[27], [29]
Humidity	High humidity can compromise dry oral/buccal swabs and dried blood spots stored under ambient conditions.	Samples being left in specimen lock box for pick up, but pick up is delayed and lock box not insulated.	Avoid these sample types if storage conditions cannot be controlled or perform validation studies to confirm sample integrity.	[94], [97], [99]
Light exposure	Potential temperature variation and UV exposure can degrade patient samples.	Sample forgotten in direct sun light.	Protect samples from unnecessary light exposure (i.e. covering).	[63], [64]
Timing of Collection	Potential false negative results due to insufficient genetic material of microorganism in specimen	Testing for a pathogen before the onset of symptoms or when symptoms have nearly resolved.	Consider pathogen’s incubation and latent period. Consider longitudinal performance of test being performed. Consider a specimen type that will yield high amounts of genetic material.	[27], [115], [116]
Antibiotic Treatment	Potential false negative results due to insufficient genetic material of microorganism in specimen as result from treatment.	Patient is empirically started on antimicrobials for C. difficile and is then tested for C. difficile with PCR toxin B genes, glutamate dehydrogenase.	Thorough patient history review. Combine Immunologic and molecular testing.	[22]
Antiretroviral therapy	Potential false negative results due to insufficient genetic material of microorganism in specimen as result from treatment.	Patient recently received ARV or PrEP and under goes HIV testing for HIV RNA.	Thorough patient history review. Combine immunologic and molecular testing.	[17], [18], [19]
Dialysis	Potential false negative results due to insufficient genetic material of microorganism in specimen as result from dialysis.	Patient with HCV is receiving dialysis and has HCV PCR performed after dialysis.	Thorough patient history review. Combine immunologic and molecular testing.	[23]
Apheresis	Potential false negative results due to insufficient genetic material of microorganism in specimen as result from apheresis.	Patient undergoes apheresis procedure experimentally for HIV management and afterwards has test performed for HIV NAAT test results.	Thorough patient history review. Combine immunologic and molecular testing.	[117]

While there are some well-cited general recommendations for the preanalytical treatment of samples for molecular diagnostics, these guidelines are often from studies or literature that predates current technologies used in laboratories [2]. With the recent world-wide diagnostic testing challenges experienced during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, there has been a dramatic increase in the number of studies evaluating preanalytical factors for confirming infection via nucleic acid amplification techniques [3], [4], [5]. Such studies shine a light on the importance of laboratory evaluations of commonly experienced preanalytical and environmental variables. This review will identify major variables during the preanalytical phase that impact diagnosis of infectious diseases.

2. Clinical decision making for ordering tests

2.1. Appropriate test ordering

Order entry is a commonly overlooked step of the preanalytical phase. This step initiates the testing process and can impact clinical outcomes. Therefore, healthcare providers should carefully assess which test can provide timely diagnostic information that can be acted upon or answers the clinical question at hand. The following discussion on test ordering will review where errors can occur and provide mitigation strategies.

Errors in test ordering can occur for many reasons and are often the result of very simple mistakes such as unnecessarily repeating a test order, forgetting to order tests, or simply ordering the wrong test. Approximately 70 % of test orders contain some sort of error [6]. Advances in electronic medical records systems have helped diminish errors associated with order entry. The implementation of test ordering governance programs has greatly decreased inappropriate test orders and order duplications, while providing significant cost savings to the healthcare system [7]. This is mostly achieved by integrating consultation services with pathologists and laboratory directors or the implementation of infection specific testing algorithms. Additionally, clinical decision support systems have proven helpful in minimizing test errors (repeat or unnecessary tests) [8]. The advent of reflex testing, providing test interpretation and test algorithms has also proven efficacious in minimizing test errors [9]. Behavioral interventions such as providing feedback to ordering physicians, changing test ordering options, and policies around panel ordering have also proven beneficial [7]. Diagnostic stewardship programs for infectious diseases have a track record of minimizing unnecessary testing, improving turn-around-time (TAT), mitigating unnecessary prescriptions, and contributing to institutional and patient cost saving [10], [11]. With the constant evolution of technology and increasing variety of test options available to clinicians, it is important that laboratory leaders work together with clinical teams to develop effective testing algorithms and stewardship initiatives to ensure appropriate test ordering.

2.2. Patient status: health, demographic and geographical history

Patient demographics play a role in infectious molecular disease testing. Consideration of the patient with respect to their region of origin, where they have traveled or lived, and which populations they have interacted with is necessary. The concept of patient demographics is best highlighted when considering the various types of HIV infection. In the United States, the most prevalent form of HIV is human immunodeficiency virus type 1 (HIV-1) while in West Africa human immunodeficiency virus type 2 (HIV-2) is the more common form [12]. However, when people travel between continents or interact with those of differing continents their risk profile for specific infections changes. It is imperative to avoid the assumption that a pathogen or risk of infection is only likely amongst the populations found in the region of prevalence.

When interpreting molecular results, the time between testing and treatment of a suspected pathogen must be considered. The goal of antiretrovirals (ARVs) is to prevent transmission of HIV as well as treat the infection. Rarely, individuals with an HIV infection on an antiretroviral therapy have demonstrated undetectable antibodies by third and fourth generation serologic assays [13], [14], [15]. More often, patients receiving ARVs are monitored for treatment efficacy by nucleic acid amplification tests (NAAT), where viral loads diminish to low or undetectable levels with treatment [16]. Individuals using preexposure prophylaxis (PrEP) also show diminished viral loads and delayed seroconversion, making detection of HIV RNA and antibodies challenging [17], [18]. When interpreting HIV RNA results it is important to consider prior or recent ARV or PrEP use, in addition to false reactivity, immunosuppression, or early window period [17], [18], [19].

The context of demographics and patient history is also paramount when considering potential differential diagnosis in an individual presenting with infectious diarrhea. When provided with a history of camping and drinking fresh water from a stream a parasitic disease, like that caused by Giardia lamblia comes to mind, while a viral or bacterial etiology maybe considered if consumption of improperly prepared food is discovered. Depending on sexual history and sexual practices reported, Entamoeba histolytica may crawl into the differential diagnosis for the cause of diarrhea [20].

3. Specimen collection and types: Specimen adequacy

3.1. Time of collection

Timing of sample collection for infectious disease testing is very important. Inappropriate timing can be the cause of false negative results. The dynamics of viral replication is a key factor to consider when testing and has been most recently highlighted throughout the coronavirus disease 2019 (COVID-19) pandemic caused by the emergence and global spread of SARS-CoV-2. Viruses may have incubation and latent periods that must be considered when providing guidelines for sample collection. Many viruses other than SARS-CoV-2 have latent periods, including hepatitis B virus (HBV), hepatitis C virus (HCV) and HIV. In an acute viral infection, the patient experiences overt symptoms and viral load is at detectable levels [21]. However, testing during the early phases of a latent period can lead to false negative results if there are low levels of viral particles. Hence it is importance to consider viral replication dynamics when performing molecular testing.

Molecular infectious disease testing and interpretation should not occur in isolation. When suspicion is high for bacterial infection, antibiotics are often initiated rapidly, prior to test ordering. In these scenarios it is possible that the initiated treatment may already have an impact on the patient’s molecular test results. This example is best highlighted with empiric antimicrobial treatment of C. difficile (CD) infections. Researchers have found patients empirically treated for CD infection can have false negative PCR results as soon as 1 day after treatment with % negativity increasing from 14 % to 35 % to 45 % after 1, 2 and 3 days of treatment [22].

Therapies besides medications can also impact infectious disease testing results. Research detecting HCV in dialysis patients have shown serologic assays may not be ideal for testing. Of the 73 dialysis patients who were found to be HCV negative by an ELISA assay, 17 were found to be HCV positive when assessed by PCR [23]. Research in the HIV realm using virus apheresis tags to remove HIV virus from circulation is ongoing. If a patient has undergone an experimental apheresis procedure for HIV management this could impact their HIV NAAT test results [23]. These examples highlight the importance of considering the patient treatment history prior to ordering or interpreting a molecular test result.

3.2. Specimen handling by specimen type

The nature of pathogen infectivity influences the quantity and location of pathogen nucleic acid shedding. This dictates which specimen types will contain the highest amount of nucleic acids. For example, when testing for viral respiratory pathogens (i.e. influenza or SARS-CoV-2) a nasopharyngeal specimen is preferred over oropharyngeal swabs as they contain higher viral loads. There has been some recent debate in the literature regarding the performance of saliva vs the traditional nasopharyngeal specimen for detecting SARS-CoV-2. Saliva is a very convenient specimen type that does not require special training of the technician nor put the technician at risk of SARS-CoV-2 exposure and possible infection. However, disadvantages to saliva-based testing include risk of contamination and lower yields of nucleic acids [24]. While there are contradictory publications, evidence suggests saliva performs similarly to the nasopharyngeal swabs for the detection of SARS-CoV-2 [25]. Although the recent pandemic has encouraged research into alternative specimen to maximize efficiency, there is a limit as to which specimens to investigate when considering the mechanism of infection. For example, nasopharyngeal specimen would not be used to test for enteric pathogens, rather stool samples would be collected [26], [27].

Along with selecting an appropriate specimen type for the infection being assessed, each specimen type can be degraded or contaminated by distinct variables, potentially impacting results. To prevent most issues with degradation or contamination, the laboratory should follow the manufacturer guidelines for specimen type and appropriate handling or perform internal validations for any modifications to these guidelines.

There is an abundance of specimen types which can be used in molecular infectious disease testing. The various specimen types, commonly associated pathogen class, variables for consideration, and general guidelines for specimen handling are summarized in Table 2 . Note that these guidelines are broad and do not take into account specific manufacturer recommendations, or unique considerations for distinct pathogens.

Table 2.

Summary of sample type and considerations for preanalytical processing.

Sample Type	Common class of pathogens detected	Specific Pathogen Examples	Important preanalytical considerations	General guidelines	References
Whole blood	Viruses	HIV, HCV, CMV, HBV	Time, temperature, collection tube additives, hematocrit	Storage: RT, 24 h (DNA) 2–8 °C, 72 h (DNA) 2–8 °C, 4 h (RNA) Additives: EDTA ACD	[2], [27], [29], [35]
Serum	Viruses	HIV, HBV	Time, temperature, hematocrit	2 –8°C, 2–7 days −20 – −80 °C, long-term	[27], [90]
Plasma	Viruses	HIV, CMV, BKV	Time, temperature, collection tube additives, hematocrit	Storage: 2 °C –8°C, 5 days −20 – −80 °C, long-term Additives: EDTA ACD	[29], [35]
Dried Blood Spot	Viruses (genomic data)	HCV, HIV	Time, temperature, humidity, light exposure	Must be adequately dried; 3–12 month stability under appropriate storage conditions	[118]
Bronchoalveolar Lavage (BAL)	Fungi, bacteria, viruses	PCP, Mycobacterium tuberculosis	Time, temperature	Transportation: On ice Storage: 4 °C, 72 h −20 °C – −80 °C, long-term	[27], [29]
Nasopharyngeal Swab	Viruses (respiratory)	SARS-CoV-2	Time, temperature, composition of VTM	Storage: VTM/UTM media: 2–8 °C, 4 days Dry swab: RT, 12 h −20 °C, 2 days	[29]
Saliva	Viruses	SARS-CoV-2	Time, temperature, use of oral hygiene products, consumption of food/beverages, stabilization additives	Storage: RT (stabilized), 7 days 2–8 °C, 7 days	[29]
Buccal Swab	Viruses	SARS-CoV-2	Time, temperature, use of oral hygiene products, consumption of food/beverages	Storage: Dried, RT, 7 days (DNA) Stabilized, 2–8 °C, 24 h (RNA)	[29]
Cerebral Spinal Fluid (CSF)	Fungi, bacteria, viruses	Cryptococcus neoformans, Neisseria meningitidis, Herpes simplex virus (HSV)	Time, transportation, and storage temperature	Transportation: Wet ice (RNA) 2–8 °C (DNA) Storage: 2–8 °C, 72 h (DNA) −20/-80 °C long-term	[27], [29]
Stool	Viruses, bacteria, parasites	Norovirus, Campylobacter jejuni, Giardia lamblia	Time, transportation, and storage temperature, stabilization media medications	Transportation: RT (stabilized) On ice Storage: 2–8 °C, 4 – 7 days RT (stabilized), 4 – 7 days	[29]
Urine	Bacteria, parasites, viruses	Trichomonas vaginalis, Chlamydia trachomatis, Neisseria gonorrhoeae, BKV	Time, transportation and storage temperature, additives	Storage: 2–8 °C, 7 days −20 – −80 °C, 7 days	[29]
Cervical/Vaginal/Urethral Swabs	Viruses, bacteria, parasites	Herpes simplex virus |(HSV), Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis	Time, temperature, transport media	Storage: Dried, at or < RT, 72 h	[29], [48]
Sputum	Bacteria	Streptococcus pneumoniae	Time, temperature, mucolytic agents, stabilization additive	Storage: 4–8 °C, 7 days <-70 °C 12 months Additives: N-acetyl l-cysteine, dithiothreitol	[29], [119]

3.3. Specimen collection containers: The effect of additives

An additional challenge to ensuring nucleic acid integrity of patient samples for molecular testing includes the use of proper specimen collection containers. Choosing the correct specimen type is further complicated by established confirmatory testing algorithms, which requires multiple samples and sample types to be evaluated by different technologies. These tests may be performed at a different facility, depending on testing availability. Often, these distinct testing platforms also have their own requirements for collection tubes and additives. With more laboratories incorporating automated liquid handlers, consideration must also be given to the dimensions of the original collection tube or aliquoted sample tubes, which may not be compatible with the downstream automated steps. This section will review the evidence for appropriate additives for molecular testing of infectious disease according to sample type including: whole blood, plasma/serum, buccal and nasopharyngeal swabs, and saliva.

Currently, the recommended blood anticoagulants for molecular diagnostic testing are ethylenediaminetetraacetic acid (K₂/K₃-EDTA) and acid citrate dextrose (ACD). Both EDTA and ACD are chelating agents commonly used in molecular techniques involving nucleic acid isolation for their efficient sequestering of Ca²⁺ ions, which serve as cofactors for nuclease responsible for the degradation of DNA [28]. However, polymerase enzyme inhibition has been suggested, presumably by presence the of EDTA post nucleic-acid extraction and the sequestration of Mg²⁺, in samples collected in tube types containing high concentrations of EDTA, such as those used for trace element analysis [29]. By contrast, RNAse activity does not depend on divalent cations, so chelating agents like EDTA and ACD do not directly inhibit RNAse activity by the same mechanism as DNAse activity [30]. While many commercial kits utilize RNAse inhibitors, such as guanidine salts, which inhibit RNAse activity via denaturation during the extraction and purification phases of sample processing [31], [32], [33], a few studies have shown that various viral RNA targets are stable for extended periods of time in EDTA plasma without the addition of RNAse inhibitors into the collection vials [34], [35].

There is conflicting evidence concerning the acceptability of plasma samples containing heparin for molecular diagnostic analysis. Studies provide evidence for PCR interference when comparing DNA amplification from samples containing the same concentrations of heparin, but low versus high concentrations of leukocytes [36], [37], [38]. Additionally, a study found that heparin inhibited DNA amplification in a dose-dependent fashion [36]. The degree of inhibition was also found to be dependent on the Taq DNA polymerase used, indicating that some commercial polymerases were more tolerant of heparin interference. Heparin has been shown to be recovered during both the DNA and RNA extraction processes [39]. Similar findings of heparin interference and suppression of RNA amplification was seen in the detection of HCV RNA by PCR [39].

While the consensus appears to be that heparinized plasma samples are not suitable for molecular diagnostic techniques for determining infectious diseases [40], a recent study has shown that heparin samples can be used to evaluate DNA or RNA from white blood cells (WBC), after appropriate WBC washing [41], which can be used for the detection of cytomegalovirus (CMV) DNA [42]. Despite the handful of studies that have provided evidence for a lack of PCR inhibition following the washing of leukocytes from heparin collection tubes, the current guidelines recommend that heparin-containing specimens be rejected [29].

Buccal and nasopharyngeal swabs, as well as saliva have distinct collection processes that may or may not require additives depending on the analyte being evaluated (DNA or RNA) and the time to analysis. VTM usually contain salts to provide an isotonic environment, proteins to stabilize the virus, buffers to control pH, and antimicrobials or antifungals to prevent microorganism contamination [43]. There are many recipes for VTM which can be bought commercially or prepared in-house [44], [45], [46]. As a result of the global supply shortage during the COVID-19 pandemic, many laboratories have undertaken studies to evaluate components of VTM that either inhibit or enhance molecular testing performance [47], [48], [49], [50], [51].

A recent evaluation of the impact of VTM on the detection of nucleic acids from SARS-CoV-2 highlights the importance of individual laboratory validation of appropriate VTM for nucleic acid targets assessed [49]. During the pandemic, a number of organizations, including the World Health Organization (WHO) and the US Center for Disease Control (CDC), published recommendations to assist in the global effort to develop and perform diagnostic testing for SARS-CoV-2. The recommendation for the formulation of VTM to be used for sample collection included the addition of protein or glycerol to stabilize the SARS-CoV-2 virus [46], [52]. Following these guidelines, commercial VTM solutions included serum-derived components such as bovine albumin for stability, which, as the study suggests, likely contain some amounts of nucleases and proteinases [49]. When compared with an in-house VTM preparation consisting of sterilized PBS solution containing gelatin (autoclaved), the use of commercial VTM solutions significantly compromised the detection of free SARS-CoV-2 RNA. After dilution with commercial VTM SARS-CoV-2 RNA was either not detected or very weakly reactive, approaching the assay limit of detection (LOD), representing a reduction in sensitivity of approximately 6 log₁₀ for free SARS-CoV-2 RNA as compared to samples diluted with the in-house VTM preparation [49]. Control experiments performed using an exogenous RNA internal control (XICP) provided evidence for efficient RNA extraction, and also insight into the likely cause of free RNA degradation. XICP was diluted in the various VTM as a solution that also included nuclease inhibitors, implying that the commercial VTM was not free of nucleases. Degradation of free RNA from Type A influenza virus was also demonstrated.

While the study revealed little to no degradation when high quality, intact SARS-CoV-2 virions were introduced to in-house and commercial VTM, as the intact nucleocapsid offers protection to viral RNA, the authors make the case that it cannot be assumed that the target nucleic acid material will always be protected by nucleoproteins [49]. Furthermore, degradation of encapsulated viruses can occur endogenously over the course of infection, as well as post-sample collection during transport and storage (See Section 3) and even during lengthy purification processes. As such it is not only important to evaluate appropriate VTM preparations for components that inadvertently degrade the nucleic acid target, but also choosing the most appropriate VTM considering the method of extraction and even downstream logistical processes such as transport, time-to-analysis and storage.

3.4. Specimen contamination

Molecular techniques have the power to synthesize millions of copies of DNA or RNA from just a few copies of target or template material. These technologies are inherently sensitive. A major concern with molecular diagnostics is the risk of obtaining false-positive test results due to sample contamination with other template material or post-amplification product (amplicons). Cross-contamination of a positive clinical sample to a negative sample can occur at multiple points during sample collection, processing, or analysis. More recently, attention has been paid to the contamination of reagents with synthetically derived amplicons, or carry-over contamination [53].

This type of carry-over contamination can occur locally, within a laboratory performing routine clinical testing, as well as regionally, at the level of the commercial manufacturer. During the pandemic, a number of laboratories reported delayed implementation of SARS-CoV-2 testing as a result of contamination of commercial reagents for the initial CDC-designed test kit [54], [55]. At the time, no commercial or state laboratory had approval to use their own, or alternative validated test and this resulted in a nation-wide lag in COVID-19 testing in the United States [56], [57]. Typically, manufacturers produce synthetic control target templates at independent sites to avoid such issues with cross-contamination. These reports imply that with an unprecedented global demand for SARS-CoV-2 diagnostic testing, scaled-up manufacturing processes can lead to issues with contamination, affirming the general need for in-house laboratory quality assurances to prevent false reporting.

There are several engineering controls and workflow strategies that can be implemented to minimize contamination in the molecular laboratory [58], [59]. As more laboratories implement the latest molecular methods, especially “closed” reaction platforms in which extraction, amplification, and detection of amplicons occurs without human manipulation or even opening of the reaction vessel, many of the engineering strategies that will be discussed are becoming unnecessary [60]. Most guidelines, however, still recommend that laboratories should designate separate pre- and post-amplification areas with negative pressure. Ideally, these areas would be physically separated and contained, however, if this is not possible, these areas should be as far apart as possible. Within the pre- amplification area, master-mix preparation can be physically separated from sample addition by using a Dead Air Box (DAB) with designated equipment. Likewise, sample addition and extraction processes can also be contained and physically separated into a separate DAB or biological safety cabinet (BSC) with designated equipment. Within the post-amplification area, additional DABs with designated equipment would be used for any processes that require manipulation of open sample tubes after amplification.

Creating a unidirectional workflow from pre- to post-amplification is essential for preventing contamination of patient samples and reagents with amplicons. This also requires laboratory staff to use separate and clean personal protection equipment in the pre- and post-amplification areas of the laboratory. Frequent changing of gloves is recommended.

In addition to ensuring dedicated equipment for each stage of sample processing, it is also important to use proper pipetting techniques that reduce the risk of aerosol formation and downstream contamination. Aerosol-barrier pipette tips are recommended to avoid contaminating designated pipettes. Additionally, proper pipetting training is essential to ensure that proper volumes are aspirated and added without cross-contamination. Likewise, it is essential that technologists are aware of aerosol formation upon opening and closing sample tubes.

Routine performance of aseptic cleaning techniques is critical. Surfaces, especially frequent touch-points, should be cleaned with a 10–15 % bleach (0.5 % sodium hypochloride) solution [59], [61]. Solutions of at least 10 % bleach were found to destroy DNA and RNA [62]. Likewise, decontamination of surfaces as well as reagent mixes can also be achieved by UV-irradiation, however, care must be taken when irradiating mixtures with UV-sensitive enzymes [63], [64]. It is essential that laboratories proactively monitor the laboratory environment for contamination by performing wipe-tests in addition to routine positivity rate monitoring [65].

3.5. Inhibitors: Endogenous and exogenous

Molecular diagnostic testing involves enzymatic reactions which are susceptible to inhibition by either endogenous or exogenous inhibitors present in the sample (see Section 2.3 for discussion of sample collection additives).

In whole blood, serum and plasma samples, endogenous inhibitors include immunoglobulin G (IgG), hemoglobin, heme metabolites, lactoferrin, and antiviral substances (i.e acyclovir) [66], [67]. IgG has been shown, via electrophoretic mobility shift assays (EMSA) and isothermal titration calorimetry (ITC) to bind single-stranded genomic DNA, inhibiting or delaying DNA amplification [68] Both hemoglobin and hematin (hydroxide ligated heme) were found to lower the enzymatic activity of DNA polymerase, as well as cause static fluorescent quenching of free fluorescent dye molecules (ROX, EvaGreen, SYBR Green I) [68], [69]. Additional metabolites of heme such as hemin (chloride ligated heme) and catabolism product bilirubin have been shown to inhibit DNA polymerase efficacy by competing with template DNA, preventing the appropriate polymerase-template complex from forming [70], [71]. Lactoferrin, a single polypeptide glycoprotein containing up to two Fe³⁺ ions, has been shown to interact with nucleic acids, potentially interfering with the efficient extraction and amplification of whole blood samples [70], [72].

Saliva contains a variety of endogenous enzymes and proteases, including nucleases which can degrade nucleic acid polymers, especially RNA [25], [73]. Similarly, nasal and nasopharyngeal swabs may contain enzymes that degrade nucleic acids in addition to polymerase inhibitors such as immunoglobulins and cytokines as well as electrolytes, minerals and medications [74]. Although not discussed in detail within this review, urine and fecal samples may contain endogenous inhibitors, primarily high concentrations of urea or polysaccharides, respectively. The highly variable composition of fecal samples can introduce additional sources of inhibition such as bile salts, hemoglobin, glycolipids and heparin [66].

Endogenous inhibitors are usually removed or inactivated during sample collection (transport media) and processing via nucleic acid extraction and purification steps [66]. Some endogenous inhibitors can co-extract with nucleic acid material, as such, it is recommended that laboratories perform validation studies of extraction methods for the determined sample type and target material [75]. Direct amplification of biological samples for infectious disease detection is highly desirable and many recent advances have been made to reduce time-consuming purification steps as well as provide point-of-care (POC) diagnostic tools [76].

Exogenous inhibitors, apart from those discussed in Section 2.3, can be introduced during sample preparation, specifically during nucleic acid extraction and purification processes [66]. More heterogeneous sample types such as saliva, swabs, and fecal samples can contain inhibiting agents introduced via ingestion [66]. For example, saliva and buccal samples are typically collected prior to ingestion of food or non-water beverages for a defined amount of time, use of oral hygiene products, or consumption of tobacco products [77]. However, it is important to follow the manufacturer’s specific instructions for proper collection. Food contains substances known to interfere with molecular techniques, primarily: polysaccharides, glycogen, enzymes such as plasmin in milk products, and minerals such as calcium [66], [78].

The inhibitory effects of food consumption on the evaluation of saliva and buccal swab specimens continues to be studied. A recent study investigated the effect of eating prior to providing a saliva sample for SARS-CoV-2 detection in five children by reverse-transcription polymerase chain reaction matrix assisted laser desorption ionization (Agena MassARRAY SARS-CoV-2 Panel/System, RT-PCR/MALDI-TOF) [79]. The study found that interference from consuming food was minimal when samples were collected 20 min after a meal, followed by nucleic acid extraction. With rapid advancements in molecular testing techniques and the number of acceptable sample types for the detection of infectious diseases, it is important to evaluate the likelihood and impact of exogenous inhibitors that can be present as a result of common lifestyle variables such as diet and medication.

Depending on the sample type, nucleic acid target, and molecular testing technology used, endogenous and exogenous inhibitors can be mitigated by nucleic acid extraction, additional purification, the use of more tolerant or robust enzymes, or sample dilution.

4. Specimen handling and integrity: From specimen collection to laboratory

Under ideal conditions, samples would be collected and immediately processed by the laboratory. Minimizing transport time after sample collection is the best way to ensure that the specimen has not degraded from exposure to environmental conditions, such as prolonged time at incorrect temperatures, extremes in humidity levels, exposure to light, or multiple freeze-thaw cycles. Of course, practically, achieving ideal conditions by immediately processing samples in the laboratory is not realistic for most samples. Even when a laboratory is processing in-patient samples, there can be major time delays between when the sample was collected and when the sample arrived at the laboratory. When considering less-centralized healthcare systems, in which samples may need to travel great distances before reaching a laboratory, an understanding of how these environmental factors can affect the specimen quality is essential for ensuring meaningful diagnostic testing. Knowing the realistic timeline and the variability of conditions that are encountered by a specimen can help a laboratory to select testing based on methods that accommodate the most stable specimen under difficult-to-control environmental conditions.

The laboratory usually assumes the responsibility for the validation of sample stability to define specimen acceptability criteria, as required by accreditation organizations such as College of American Pathologists (CAP). In general, most laboratories will follow specified sample collection and handling recommendations of FDA-approved commercial testing kits. However, many laboratories supplement or validate these recommendations by performing in-house stability studies, especially when the recommended guidelines are not feasible. Acceptable variations in environmental conditions are dependent on sample type, target nucleic acid, and pathogen. This section provides discussion of environmental conditions that are known to affect the integrity of specimens (blood, plasma/serum, buccal and nasopharyngeal swabs, and saliva) for infectious disease testing via molecular methods as a resource for laboratories when deciding which stability studies to perform.

4.1. Time, temperature and freeze-thaw cycles

In general, for whole blood samples, storage below 0° C is not recommended. If DNA or RNA extraction cannot occur within an acceptable timeframe, the erythrocytes should be removed, and the plasma or serum should be stored at −20 °C or colder [29]. Removal of the erythrocytes is an essential step since heme can be released upon thawing of frozen samples and inhibit polymerase activity in PCR mechanisms as well as quench fluorophores used for real-time detection of amplicons (Section 2.5) [29], [80], [81], [82]. If the target nucleic acid material is RNA, the general recommendation is that whole blood should be collected in a tube containing RNA stabilizing reagent. Depending on the RNA stabilizing reagent used, appropriate storage temperatures and duration prior to extraction may vary. While there is no absolute time limit for the storage of whole blood samples at room temperature or 2–8 °C, prolonged storage at these temperatures is not recommended due to increased risk of hemolysis and nucleic acid degradation because of white blood cell lysis [29], [41].

Serum and plasma samples are more conducive to longer-term storage, since these specimen can be frozen without the introduction of endogenous inhibitors. Allowable freeze-thaw cycles must be evaluated by the laboratory for each target pathogen. For example, one study evaluated the effect of multiple freeze-thaw cycles on serum samples evaluated for HBV DNA and HCV RNA by Quantiplex branched-DNA assays (Chiron Diagnostics Corporation, Walpole, MA). The report determined that the respective nucleic acids could withstand up to eight freeze-thaw cycles [83]. Yet, an earlier report observed a 16 % decrease in HCV RNA titers after 5 freeze-thaw cycles and recommended aliquoting sample to reduce loss [84], and a later study reported no significant decreases in viral load after 10 freeze-thaw cycles [85]. Similar findings were observed for HBV DNA [86].

Recommendations for the storage and handling of nasopharyngeal swabs typically defer to those specified by the commercial manufacturer [29]. With the ongoing global pandemic and limited supply chain, some laboratories have made modifications to the specified storage conditions for SARS-CoV-2 RNA specimens [87], [88], [89]. Evaluations of long-term room-temperature storage condition for nasopharyngeal swabs in viral transport media revealed that samples could be stored for up to 21 days at room temperature without significantly impacting the RT-PCR results [87]. Room temperature stability (22 °C, 1 week) of nasopharyngeal swabs in VTM has also been observed for the RT-PCR detection of four additional viral RNA targets (influenza, enterovirus, herpes simplex virus, adenovirus) [48]. Yet, there are other reports indicating significant (66–74 %) loss of viral RNA for influenza A, regardless of storage temperature (21 °C, 4 °C, −80 °C) after two weeks [90].

Oral rinse and saliva specimen for DNA extraction are relatively stable at room temperature [2], [29]. The traditional recommendation for oral rinse and saliva storage for RNA extraction requires immediate placement on ice or in a transport medium with stabilizing agent and storage at 2–8 °C followed by extraction within 24 h of collection [2], [29]. A more recent study of SARS-CoV-2 RNA stability in saliva specimens stored without nuclease inhibitors or transport medium and at a variety of temperatures (−80 °C to 30 °C) and processing conditions (freeze-thaw), revealed that SARS-CoV-2 RNA was stable for extended periods of time (>16 days) at room temperature in addition to more extreme temperature treatments (freeze-thaw, and 30 °C for 72 h) [91].

Urine specimens should be processed immediately or stored at 2–8 °C for several reasons. First, the generally low pH of urine (4.5–8) and the concentration of urea denature DNA at ambient temperatures (25 °C and greater) [92], [93]. Additionally, prolonged storage at low temperatures increases the likelihood of uric acid, calcium oxalate, and phosphorus precipitation, which can inhibit nucleic acid amplification [92].

4.2. Humidity

High-humidity environments can compromise certain sample-types, primarily those that are deposited on materials and stored under ambient conditions such as dry oral and buccal swabs and dried blood spots [94], [95], [96]. Viral or microbial nucleic acids have been shown to degrade during prolonged storage [94], [97]. A recent study of an energy-based mechanism of DNA degradation under ambient conditions and various levels of humidity and irradiation suggests that water contributes to the formation of reactive oxygen species, which ultimately leads to DNA degradation [98]. Additionally, under humid conditions, these specimens are prone to fungal and bacterial contamination [99]. Validation studies would be necessary to evaluate the impact of microbial contamination on the outcome of the target pathogen assay.

4.3. Light exposure

Apart from the variability in temperature that comes from exposing patient samples to light for prolonged periods of time, UV irradiation is also a concern [100]. Samples that are exposed for long periods of time to any source of light emitting UV irradiation can lead to oxidative degradation of nucleic acid polymers.

Sample stability is influenced by many environmental variables en route to- and at the laboratory. While there are recent reviews [27], [96], [101] and studies [41] on the effects of some of these variables on specific sample types and pathogen targets, it is still important to perform in-house validations that address the common scenarios experienced by the individual testing center. If limited resources prevent these types of extensive validations, laboratories should err on the side of caution and adopt the more conservative evidence-based stability criteria.

5. Specimen processing

In the chain of preanalytical events, specimen processing is the final stop. Variables affecting specimen processing include physical characteristics of the samples, procedures like sample mixing and vortexing, specific pretreatment or pooling methods, and technology for handling pipetting.

While most blood-derived samples, or samples requiring viral transport media that are used for molecular testing of infectious diseases, present with few complications for sample processing, other types such as sputum, saliva, and swabs can yield highly viscous samples that can complicate sample processing. Viscous samples are particularly cumbersome for laboratories which have integrated automated liquid handlers (ALH) [102], [103]. These samples can cause pipetting errors and lead to contamination. Often, laboratories implement pretreatment techniques to either liquify viscous mucous with additives such as acetylcysteine or Sputasol (ThermoFisher), or by physical homogenization with inert beads and sterile media such as PBS [102], [104].

Although the occasional viscous sample may cause issues for an ALH, there are many benefits to implementing automated pipetting over manual processing. Automation improves throughput, reproducibility, and maximizes laboratory technologist’s time for data analysis and prevents repetitive stress injuries such as carpal tunnel and tendonitis [105], [106]. In addition, ALHs can decrease the risk of sample contamination as well as environmental contamination of the laboratory with potentially infectious material [105]. No automation is perfect, and technologists would need to learn how to operate, program, and troubleshoot ALH instruments to resolve the inevitable errors that would occur.

Sample preparation for molecular testing requires many processes in which the original sample is mixed with additional reagents for stability, extraction, and amplification. Along the way, it is common practice to thoroughly mix each solution by vortexing for specified amounts of time. While the majority of manufacturers and studies do not specifically characterize the effect of vortexing conditions on the outcome of molecular assays, there has been one recent study that specifically examined the effect of not vortexing nasopharyngeal and throat swabs for the detection on SARS-CoV-2 [107]. The study found no significant difference on the qualitative sensitivity of SARS-CoV-2 rRT-PCR tests when samples were not vortexed prior to analysis [107]. However, vortexing did improve recovery of cellular material, which may be important for other types diagnostic testing [107].

During the COVID-19 pandemic, a lack of supplies led to the decision of many laboratories to pool samples for processing to conserve precious reagents. Although sample pooling is common for blood banks, this was an adaptive solution to meet the exponential demands for testing with unreliable access to resources. There are a number of downsides to sample pooling for a prevalent infectious disease: 1) the possibility of false negative results due to diluted samples [108]; and, 2) increased TAT when individual samples from a pool must be re-run to confirm presence of infection.

Additional pretreatment concerns that can affect sample processing and impact downstream molecular testing are those that deal with higher risk pathogens, such as biosafety level (BSL) 3 and 4 pathogens. Laboratories that handle these pathogens, especially BSL-4, have very strict engineering controls such as separate or controlled ventilation, separate work areas for sample transfer, extraction and analysis, authorized access only, and special personal protective equipment such as full body and air supplied suits [109]. Attention must be given to containment and thorough disinfection techniques during sample processing since it is possible that pathogens will not be inactivated prior to molecular testing [110], [111].

Ultimately, the measurement of a laboratory’s efficiency is TAT, which relies on efficient processing of samples. There are many considerations for improving sample processing for molecular testing, such as which sample type may require less manual pretreatment, whether or not to vortex or use ALHs, how to process samples when diseases are prevalent or the pathogen in question is BSL-3 or 4. Laboratories strive to improve TAT so that clinicians have fast access to results that are needed to make timely patient care related decisions.

6. Clinical implications of preanalytical and environmental factors in molecular infectious disease testing

The implications of false reporting due to missteps in the preanalytical phase of processing patient samples for molecular testing are potentially life changing for the patient. Errors that lead to reporting falsely negative results endanger patients by either delaying or denying appropriate and often life-saving treatment for the infection. For example, delaying antibiotic treatment in adult and pediatric patients with sepsis can lead to death [112], [113].

On the other hand, errors that lead to the false reporting of positive results place the patient at risk of receiving inappropriate and unnecessary treatment for an infection that the patient does not have. The potential side-effects and long-term effects of receiving unnecessary treatments can also be life-threatening in the case of developing resistance to antibiotics. In addition, depending on the disease, the laboratory may be required to participate in infectious disease reporting toa state health agency. Preanalytical errors that lead to false test results can have a significant impact on the accuracy of this reporting, and down-stream public health initiatives and decisions that may be based on this data.

As highlighted throughout this review, it is the laboratory that is ultimately responsible for defining and enforcing preanalytical work-flow and acceptability criteria to ensure quality testing. Given the challenges and incredible resources required to complete this task properly, perhaps, in the future, progress will be made toward the standardization of molecular diagnostic assays to 1) reduce preanalytical errors resulting from non-standardized in-house validations and 2) improve the diagnostic accuracy on a national and even global scale.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.