Clinical Implications of Multiplex Pathogen Panels for the Diagnosis of Acute Viral Gastroenteritis

ABSTRACT

Acute gastroenteritis remains a significant cause of morbidity and mortality in both high- and low-resource settings. The development of nucleic acid-based testing has demonstrated that viruses are a common, yet often undetected, cause of acute gastroenteritis. The development of multiplex pathogen PCR panels makes it possible to detect these viral pathogens with greater sensitivity and rapidity than with previous methods. At present, there is insufficient evidence to recommend the routine use of these panels for the average patient with acute gastroenteritis. However, there are specific scenarios and patient populations, such as epidemiology/outbreak surveillance, antimicrobial stewardship, and the care of immunocompromised patients, where these tests could be clinically useful today. Further research on the effect of these syndromic panels on provider antibiotic prescribing behavior and patient length of stay will be necessary to know their ultimate role in clinical practice.

INTRODUCTION

The last 2 decades have seen dramatic declines in the burden of acute gastroenteritis (AGE), yet AGE remains one of our most critical public health challenges globally. AGE is responsible for an estimated 1.3 million deaths worldwide annually, making it a leading cause of mortality across the age spectrum. This impact is felt disproportionally by infants and children under 5 years of age, with nearly 500,000 children dying each year from diarrheal disease (1). A reliable estimate of AGE morbidity on the global scale is complicated by the difficulty in collecting high-quality data from the underresourced settings where AGE prevalence is highest. However, it is well recognized that severe AGE is only a small fraction of the mild to moderate illness experienced more frequently. Viruses have been increasingly found to be associated with AGE since the discovery of norovirus and rotavirus as causes of gastroenteritis in the 1970s (2). In low-resource settings, viral AGE is the most common diarrheal disease etiology in both the first and second years of life, accounting for 64% and 50% of cases, respectively (3).

While low-resource countries bear the brunt of the diarrheal disease burden, AGE also results in substantial morbidity and mortality in high-resource settings, such as the United States. Public health interventions designed to tackle this disease burden are dependent upon accurate epidemiological data on the etiology and distribution of infectious diarrheal disease. These epidemiological data are, in turn, reliant on the ability of clinical and surveillance laboratories to accurately detect pathogens. Therefore, innovation in diagnostic technology is crucial to the advancement of public health measures for controlling infectious diarrhea.

Modern laboratory methods, such as quantitative PCR (qPCR), have improved our understanding of the prevalence of enteropathogenic viruses and dramatically shifted the burden of disease attributed to viral AGE. For instance, reanalysis of data from two pivotal studies of AGE among children in several countries, the Global Enteric Multicenter Study (GEMS) (4) and the Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED) study (3), led to important modifications in estimates of pathogen-specific diarrheal disease burdens. These reanalyses demonstrate the potential for multipathogen molecular tests to offer new clinical data compared to traditional antigen- or culture-based diagnostic methods. Therefore, clinicians must become familiar with the characteristics of these diagnostic tests in order to best apply results in clinical settings and as public health measures. In this review, we discuss common and emergent viral pathogens contributing to AGE globally and in the United States, advancements in laboratory detection of viral AGE pathogens, and the clinical significance of improved pathogen detection.

EPIDEMIOLOGY AND DIAGNOSTIC TESTING OF COMMON AND EMERGENT VIRAL PATHOGENS

RV.

Rotavirus (RV) is a double-stranded RNA virus that was first described in 1973 and later determined to be the most frequent cause of diarrheal disease among young children in the United States (5). It is grouped into seven different serogroups, lettered A to G, with group A being responsible for the majority of human disease. RV infection is a nearly universal phenomenon, with most children worldwide infected by 5 years of age. Imperfect immunity results in the potential for subsequent infections, usually caused by a different genotype, throughout life; however, the majority of morbidity and mortality due to RV infection is in children under 5 years of age (1). RV is responsible for approximately one-third of diarrhea-related deaths in this age group. Fortunately, there are two live attenuated oral vaccines administered during infancy available globally: GlaxoSmithKline’s monovalent vaccine, Rotarix, and Merck’s pentavalent vaccine, RotaTeq (5). In the United States, implementation of early childhood vaccination has resulted in significant reductions in RV-attributable AGE and changes in disease seasonality (6). Testing for RV in the clinical setting has primarily relied on stool-based enzyme immunoassays (EIAs) since the 1980s (7), and although qPCR-based testing is the most sensitive, it is not routinely commercially available outside multiplex pathogen panels.

NoV.

Norovirus (NoV) is a single-stranded RNA calicivirus that is highly transmissible and a prominent cause of gastroenteritis epidemics worldwide, accounting for up to 50% of outbreaks (8). In the United States alone, NoV causes up to 800 deaths, 71,000 hospitalizations, 2 million outpatient visits, and 21 million illnesses annually (9). Severe disease is most common at extremes of age, with young children at the most risk for needing medical attention for NoV infection and adults over 65 years of age at the greatest risk for NoV-related death (9). Because of its frequent association with epidemic AGE, NoV is known to cause outbreaks at long-term care facilities, health care facilities, restaurants, schools, and other institutional settings (10). While multiple genotypes of NoV are capable of causing human disease, the GII.4 genotype is most frequently associated with NoV outbreaks (11). Historic surveillance data suggest that GII.4 strains undergo genetic drift that allows for the evasion of population-level immunity and occasional seasons of increased disease activity (10). Strains of genotype I and other strains of genotype II also cause human disease but show less genetic variation and, consequently, no distinct epidemiological pattern (10). Amplification of the viral genome by reverse transcriptase PCR (RT-PCR) is the most common modality used for detection of NoV from different clinical specimens, such as feces, emesis, or rectal swabs, and can be performed individually or as part of a multipathogen panel.

Enteric HAdV.

Adenoviruses (HAdV) are a diverse family of double-stranded DNA viruses associated with a variety of clinical phenotypes, including gastroenteritis (12). Several serotypes of HAdV have been associated with gastroenteritis, but serotypes 40 and 41 are the most common. Incidence estimates for HAdV-associated gastroenteritis range from 1% to 8% in high-resource settings and from 2% to 31% in low-resource settings (2). Recent findings facilitated by advancements in molecular diagnostic techniques suggest the true prevalence of enteric HAdV among children is up to five times higher than previously understood (4). Given that HAdV infection is frequently more severe than other types of viral AGE and that HAdV can cause severe extraintestinal manifestations, particularly in the immunocompromised population, this higher prevalence could have meaningful clinical impact (12, 13). HAdV can be grown and isolated in a variety of cell lines, but it is most commonly diagnosed through antigen-detecting enzyme-linked immunosorbent assay due to its greater sensitivity and the difficulty of viral culture. PCR can also be used for HAdV detection, and primers and probes specific to Ad40 and Ad41 are available for use individually or as part of a multiplex PCR (12).

Sapovirus.

Sapovirus is a single-stranded RNA calicivirus that causes both sporadic and outbreak-related AGE among individuals of any age (14). In resource-limited settings, there is increased incidence of sapovirus disease in the first and second years of life (3). Like NoV, sapovirus requires only a small inoculum to cause infection and undergoes rapid evolution (14). Sapovirus can be detected by electron microscopy (due to its distinctive shape), immunoassay, or PCR; however, PCR is the most sensitive. Standalone sapovirus testing, particularly antigen testing, is not widely commercially available (14).

HAstV.

Astrovirus (HAstV) is another single-stranded RNA virus with multiple serotypes causing AGE (15). HAstV primarily causes AGE in children and is only rarely reported as a pathogen in healthy adults (2). Infection with HAstV is generally mild and self-limited, but severe and/or disseminated infections have been reported in severely immunocompromised patients. Like sapovirus, HAstV historically has been detected using electron microscopy and immunoassays, but new PCR-based approaches have proven to be more sensitive, resulting in most commercially available astrovirus testing to be in the form of multiplex pathogen panels (15).

CLINICAL CHALLENGES IN DIAGNOSING ACUTE GASTROENTERITIS

The approach to a patient with suspected gastroenteritis can be clinically challenging, particularly when attempting to differentiate potential causes of infectious diarrhea on history alone. Contemporary Infectious Disease Society of America (IDSA) guidelines (16) recommend diagnostic testing in diarrheal patients with “fever, bloody or mucoid stools, severe abdominal cramping or tenderness, or signs of sepsis” as well as when patients have specific exposures placing them at risk for a particular pathogen (i.e., consumption of raw oysters [Vibrio] and exposures to cruises or long-term care facilities [norovirus]). Unfortunately, patients are often unclear about the exact characteristics of their stools when providing a history, and the severity of abdominal cramping/tenderness is inherently subjective.

Perhaps unsurprisingly, there is substantial heterogeneity in clinicians’ approach to evaluate adult and pediatric patients with acute diarrhea (17). Even when the decision is made to test for a culprit agent, the diagnostic yield of stool cultures and other conventional stool testing for enteric pathogens is poor. One recent study comparing a multiplex PCR panel to conventional microbiological methods found that the conventional methods revealed a potential pathogen in only 6.0% of cases (18). In part due to the poor yield of bacterial culture, viruses are thought to cause the majority of cases of infectious diarrhea, and patients generally are given supportive care while awaiting resolution of symptoms. Empiric antibiotic therapy is reserved for patients who have severe illness and bloody diarrhea or who present with dysentery in the setting of recent international travel (16). The IDSA guidelines do comment that empirical antimicrobial treatment is also sometimes provided to immunocompromised patients or young infants with severe acute watery diarrhea.

This approach, while sometimes frustrating to the patient and provider, is generally adequate for most cases of uncomplicated infectious gastroenteritis due to the high frequency of the spontaneous remittance of symptoms. However, the low yield of conventional microbiological methods is challenging when infection is not the only potential cause and other etiologies (e.g., inflammatory and medication induced) are under consideration. In many clinical scenarios, the ability to reliably exclude infection would allow for more prompt consideration of alternate etiologies and earlier implementation of more invasive testing (e.g., colonoscopy). For these reasons, an improved diagnostic approach to infectious diarrhea was recently considered one of the top unmet diagnostic needs in a survey of U.S.-based infection disease specialists (19).

It is in this clinical context that multiple pathogen nucleic acid panels for acute gastroenteritis have become commercially available. We will discuss the variations in available panel testing as well as the clinical benefits and challenges posed by their implementation, with specific emphasis on viral gastroenteritis.

CURRENT STRATEGIES IN MULTIPLEX DETECTION

In this review, we highlight four multiple-pathogen panels currently available in the United States and Europe for the diagnosis of AGE (Table 1). The selected panels were chosen because they detect at least 3 viral pathogens, have regulatory approval, and are available for clinical use. Other panels (i.e., Luminex Verigene) are available but either detect fewer than 3 viral pathogens or are less widely available. Each of these platforms relies on the extraction of nucleic acids from a submitted diarrheal stool sample; however, there are some differences in sample processing and testing that have implications in clinical practice.

TABLE 1.

Characteristics of multiplex pathogen panels for the diagnosis of acute viral gastroenteritis^a

Characteristic	RIDA gene	xTAG GPP	BioFire FilmArray	BD Max enteric viral panel
Technology	Multiplex real-time RT-PCR with hydrolysis probe	Multiplex RT-PCR with bead hybridization	Nested multiplex PCR with melting curve analysis	Multiplex real-time RT-PCR with hydrolysis probe
Specimen prepn	RNA extraction required	RNA extraction required	None (stool in CB medium)	None (raw stool or in CB medium)
Batch size	96- or 384-well plate	PCR reaction separate readout in 96-well plates	1 reaction per instrument (up to 8 in parallel)	Up to 24 reactions
Turnaround time	∼2 h	∼5 h	∼1 h	∼3 h
Regulatory approval	CE-IVD	FDA	FDA	FDA
Viral pathogens
Adenovirus F 40/41	+	+	+	+
Astrovirus	+		+	+
Norovirus GI/GII	+	+	+	+
Rotavirus A	+	+	+	+
Sapovirus (genogroups I, II, IV, V)			+	+

^aFDA, Food and Drug Administration. CE-IVD, CE in vitro diagnostic.

Multiplex real-time PCR.

Real-time PCR, or quantitative PCR, is a widely used technique that allows for the measurement of PCR product while the reaction is ongoing. Several companies have designed panels using this technology with multiple primer pairs and fluorophores to allow for the detection of multiple viral pathogens in the same sample. The RIDA GENE system (20) is approved for use in Europe and requires the extraction of RNA from a stool specimen prior to analysis. The BD Max system was recently approved for use in the United States and uses a similar approach but has the additional benefit of using unaltered stool as the sample reagent (21). Target RNA is converted to cDNA by reverse transcriptase before being amplified in the presence of fluorescent hybridization probes, which only fluoresce when bound to their target DNA sequence. This causes sample fluorescence to increase with each cycle of PCR if a target sequence is present. By using a different fluorophore for each pathogen, a single sample can be used to test for multiple viral pathogens.

Multiplex RT-PCR with hybridization.

Multiplex panels such as the xTAG gastrointestinal pathogen panel (22) use multiple primer probes in the same PCR to test a single sample for the presence of multiple pathogens at once. The xTAG GPP system is an RT-PCR system, meaning that it uses a reverse transcriptase to convert mRNA from the sample into cDNA, which is then amplified. The PCR product is then mixed with proprietary fluorescent bead mixture in a hybridization reaction before analysis. Each pathogen corresponds to a different type of microbead, allowing the hybridization reactions to occur in the same well. Positive results are determined by fluorescent activity in a sample well above a predetermined cut point and/or relative to an internal control. It should be noted that stool specimens must be processed and nucleic acids extracted before they can be tested using the GPP system, leading overall time to test one sample to be approximately 5 h, although samples can be run in parallel.

Nested multiplex PCR.

Panels that use nested multiplex PCR (i.e., Biofire FilmArray GI panel) (23) begin with extraction of nucleic acid from the submitted stool sample. The sample is then amplified through two sequential PCR steps, where the second PCR uses primers internal to the endpoints of the first PCR product. This design increases the specificity of the coupled reactions for the target region of interest, because off-target PCR products produced by the first amplification reaction will not be further amplified by the second reaction. In the BioFire panel, the second PCR occurs in a multiwell system in the presence of an intercalating fluorescent DNA dye. This allows the products to be immediately subjected to melting curve analysis as an additional validation step. Each pathogen is tested in triplicate and only reported as positive if two out of three replicates are positive and have similar melting curves. One of the key advantages of the BioFire system is the ability to carry out the entire process from nucleic acid extraction to detection in a single-pouch closed system in about 1 h. This closed system reduces laboratory labor requirements and also minimizes opportunities for contamination.

Relative analytical performance.

As discussed above, non-PCR-based clinical testing for viral pathogens that cause acute gastroenteritis is frequently insensitive (antigen detection), technically difficult (viral culture), and/or impractical for use at clinical scale (electron microscopy). For this reason, the multiplex pathogen panels discussed in this review have generally been validated against nonclinical detection methods, such as PCR with a distinct set of primers followed by sequencing (21, 23). In general, the multiplex pathogen panels have performed well in these analyses, with reported sensitivities of >90%. One study, by Huang et al., compared the performance of the BioFire and xTAG panels in addition to a third panel (Verigene), which detects only two viral pathogens (rotavirus and norovirus), to a composite reference standard (24). The BioFire and xTAG panels performed highly similarly for the detection of norovirus and rotavirus. The Verigene assay was relatively less sensitive for the detection of rotavirus but was similar to the others for norovirus. Overall, the available evidence suggests that all of the multiplex pathogen panels discussed are highly sensitive both in absolute terms and relative to non-PCR-based diagnostic methods. This improvement in diagnostic accuracy is notable; however, its clinical impact requires additional discussion.

MULTIPLEX TESTING IN CLINICAL PRACTICE

Routine use.

As previously mentioned, the IDSA recommends against the routine use of diagnostic testing in patients presenting with suspected infectious diarrhea, as the majority of these patients will spontaneously recover and do not need more thorough investigation or treatment (16). This caution extends to the use of multiplex panel tests with the additional caveat that interpretation of multiplex pathogen panels is more complex. In particular, it is unclear whether or not the detection of nucleic acid from a potential pathogen in a stool sample is equivalent to isolating that pathogen from conventional culture. This diagnostic puzzle is not unique to gastroenteritis but rather is part of the larger conundrum of attempting to identify the presence of actively replicating pathogens responsible for causing disease with a culture-independent method.

The most comprehensive analysis of the impact of the wide-spread use of multiplex pathogen panels in the diagnosis of AGE is the review by Freeman et al. (25) on behalf of the National Health Service in the United Kingdom. They use the available literature as well as local expert opinion to model the clinical and economic impact of replacing the current NHS testing algorithm with widespread use of a multiplex pathogen panel (in particular the xTAG gastrointestinal pathogen panel or BioFire FilmArray GI panel). They emphasize the high degree of uncertainty in interpreting the meaning of additional positive results produced by multiplex pathogen panels in the absence of a sensitive reference gold standard. Their economic model assumes substantial improvement in time to diagnosis, with the use of MPPs resulting in a 50% decrease in length of stay and subsequent cost savings from MPP use despite the higher cost of the test itself relative to traditional testing. Hospital length of stay is the most impactful variable in their cost analysis, and they draw attention to the fact that smaller or larger changes in length of stay as a result of multiplex pathogen panel implementation can result in either substantial cost savings or cost increases. As a result of these substantial uncertainties, they do not make a summary judgment but call for additional research to determine the effect of these tests on patient outcomes and health care system costs.

Despite these cautions, there are several scenarios in which multiplex testing for gastrointestinal pathogens can be useful.

Outbreak assessment/epidemiological monitoring.

One of the major advantages of the MPPs is the faster turnaround time compared to conventional methods (hours versus often days, depending on the pathogen), although this benefit is only realized if laboratory staffing can process specimens as they are received. Nevertheless, faster turnaround opens the possibility for testing to be performed in real time as part of an outbreak assessment, in turn leading to more informed triage and treatment decisions. This was recently exemplified by the use of a multiplex panel to assist in managing a gastroenteritis outbreak among the workers in an Ebola virus disease unit in West Africa (26). Additionally, the ability to test for multiple pathogens with a single sample and to detect pathogens not easily cultured (i.e., viruses) makes multiplex panels a promising tool for surveillance or epidemiological studies looking to determine the burden of disease caused by different gastrointestinal pathogens (27). However, a major shortcoming of PCR-based methods for epidemiological purposes is the lack of isolated specimens for subtyping and antimicrobial susceptibility testing for bacterial causes of AGE. For this reason, the IDSA recommends that any specimen that tests positive for a reportable pathogen by a culture-independent diagnostic test (i.e., a multiplex PCR panel) should be submitted for routine culture as well (16).

Antibiotic stewardship.

As previously mentioned, it is not possible to reliably differentiate bacterial and viral gastroenteritis based on clinical features alone. This uncertainty contributes to the overprescribing of antibiotics. A recent paper by Cybulski et al. (18) examined, in part, whether the availability of faster results of diagnostic testing (BioFire FilmArray compared to conventional stool culture) affected antimicrobial prescribing in the outpatient setting. They noted a significant decline in empirical antimicrobial prescriptions and a shift toward targeted therapy of identified bacterial pathogens. This change in prescribing habits occurred in the context of a decrease in time to result from 60 to 75 h (conventional methods) to 18 h (BioFire FilmArray). This study additionally compared the clinical presentations of patients with infections detected by conventional means with those whose pathogens were detected only by the FilmArray panel. They concluded that these groups did not present differently and argued that the increased rates of positive tests with the FilmArray panel represented true infections missed by conventional testing.

Immunocompromised hosts.

Diarrhea in an immunocompromised individual presents a particular diagnostic challenge, as the lists of both infectious and noninfectious etiologies are often far broader. Additionally, there is often greater urgency in identifying and reversing the underlying cause or in being able to rule out a more serious etiology. For example, rapid exclusion of infectious etiologies in solid-organ transplant patients may allow for the timelier and more confident consideration of alternate etiologies, such as side effects of common immunosuppressive medications (28). Additionally, norovirus is recognized as causing significant morbidity in the postallogeneic stem cell transplant population and can be difficult to differentiate from gastrointestinal graft-versus-host disease (29). Large multiplex PCR panels that can rapidly rule in or out common infectious etiologies may be of particular benefit in these complex clinical scenarios. It should be noted that none of the panels discussed include cytomegalovirus (CMV), which is a key viral pathogen causing infectious diarrhea in immunosuppressed hosts (30).

CONCLUSIONS

In summary, multiplex PCR panels represent a significant advancement in the ability to diagnose viral gastroenteritis and will likely transform our understanding of the epidemiology of this clinical syndrome. Currently, the clinical utility of these tests in routine cases is not evident, and the preponderance of available data argues against the routine use of these panels to investigate suspected AGE. However, there are scenarios, such as outbreak assessment and the evaluation of immunocompromised individuals, where these tests may have a role (Table 2). Additionally, some studies suggest a role for multiplex panels in decreasing inappropriate antibiotic use and shortening hospital length of stay. If these tests are to be used in routine clinical practice, practitioners must be aware of their shortcomings (i.e., inability to detect many subtypes of adenovirus, possibility of false positives, and detection of nonpathogenic organisms) to ensure that results of multiplex pathogen panels are applied appropriately to patient care. Further research into these possible benefits and risks of molecular gastroenteritis diagnostics may expand the clinically useful scope of these tests.

TABLE 2.

Summary of potential uses of multiplex pathogen panels for the diagnosis of viral gastroenteritis

Clinical scenario	Potential utility of multiplex pathogen testing
Uncomplicated inpatient with watery diarrhea	Currently available evidence does not support routine use of syndromic testing
Hematologic stem cell transplant recipients	Positive identification of viral pathogen (i.e., norovirus) could prompt reduction in immunosuppression; conversely, a negative result may increase likelihood of alternate noninfectious etiologies (i.e., graft-vs-host disease)
Solid organ transplant recipients	Negative results may accelerate timeline to advanced diagnostic techniques (i.e., colonoscopy) and consideration of alternate noninfectious etiologies (i.e., mycophenylate toxicity)
Advanced HIV disease	Negative results may accelerate timeline to advanced diagnostic techniques (i.e., colonoscopy)
Epidemic response	Rapid identification of an alternate diagnosis can decrease suspicion for a novel/severe pathogen (i.e., SARS-Cov-2, Ebola)