Rapid Diagnostic Tests for Meningitis and Encephalitis - Biofire®

Abstract

Meningitis and Encephalitis (ME) are important causes of morbidity and mortality worldwide. Patients suspected of having ME are often hospitalized and started on empiric antimicrobial treatment, due to the potential adverse consequences of delaying the diagnosis or treatment. Multiplexed polymerase chain reaction (PCR) panels are one of several rapid diagnostic technologies that have the potential to overcome some of the limitations of conventional diagnostic methods for ME. The BioFire® FilmArray® Meningitis/Encephalitis Panel was the first FDA-cleared multiplex PCR for the evaluation of cerebrospinal fluid samples, able to identify 14 organisms in a single test reaction. This newer rapid diagnostic tool has an overall high sensitivity and specificity for the diagnosis of ME with a fast turnaround time, and has the potential to improve resource utilization for patients presenting with suspicion of ME. However, further research is needed to determine its optimal use in the evaluation of patients with suspected ME.

Background:

Meningitis and Encephalitis (ME) are important causes of morbidity and mortality worldwide,^1–4 and tens of thousands of adults and children are diagnosed with meningitis or encephalitis each year in the United States.^5,6 Although viral central nervous system (CNS) infections are more common in the U.S.,^5,6 acute bacterial meningitis is the most rapidly fatal.⁷ In other parts of the world, especially in developing countries, bacterial meningitis represents an even larger cause of morbidity and mortality.^3,8 Therefore, rapid diagnosis and treatment are essential. For diagnosis, cerebrospinal fluid (CSF) culture, Gram stain and other molecular and cellular analyses of the CSF are frequently employed.^3,7 However, some of these conventional methods can have limitations in sensitivity and specificity, while the culture can take several days to result.⁷ Due to the potential adverse consequences of delaying the diagnosis or treatment, patients suspected of having ME are often hospitalized and started on empiric antimicrobial treatment while awaiting CSF cultures.^5,6,9–11 Polymerase chain reaction (PCR)-based methods can improve the process of identifying viral or bacterial pathogens in the CSF by increasing the diagnostic yield and providing faster turnaround times.⁷ For example, viral PCR for enterovirus (EV) and herpes simplex virus (HSV) have become standard of care for diagnosis.¹² Rapid diagnostics have significant potential to improve care, optimize antibiotic utilization, decrease hospitalizations and lower costs for patients presenting with suspicion of ME.^13,14 These make multiplex molecular assays an attractive option to screen and detect potential pathogens. The BioFire® FilmArray® ME Panel is the only multiplexed PCR assay currently cleared by the Food and Drug Administration (FDA), and is capable of detecting 14 organisms in the CSF.¹⁵ To understand the potential benefits and limitations, and to inform the proper implementation in clinical practice, it is essential to review the scientific literature on the FilmArray® Panel.

Current Approach and Diagnostic Challenges:

A wide array of infectious and non-infectious etiologies can cause ME, which contributes to the challenge in diagnosing these conditions.⁵ CSF culture is considered the gold standard for diagnosis of bacterial meningitis and is essential to determine the antimicrobial susceptibilities of the causative organism.³ The sensitivity of CSF culture has been reported to range from 67~88% although the sensitivity varies depending on the organism and is lower in patients pretreated with antimicrobial agents.³ While viruses are the most common cause of ME in the U.S. in both children and adults, patients are often hospitalized and treated with antimicrobial and sometimes antiviral therapy pending identification of the pathogen, which can take up to 72 hours.^5,6 The short turnaround time of CSF Gram stain can facilitate more rapid identification of the pathogen in cases of bacterial meningitis. It is also inexpensive and well-validated.³ The sensitivity of Gram stain for bacterial meningitis has been estimated to be in the range of 35~90%,^3,7,16–19 however it also varies widely depending on the causative organism^3,7 and is lower in patients pretreated with antimicrobial agents.^3,7 Although CSF cell count, glucose and protein analysis can also be helpful when trying to differentiate between bacterial and viral meningitis, these tests can also be normal or not characteristic of the causative pathogen.³ Latex agglutination test has been used to aid in the more rapid of diagnosis of bacterial meningitis, however its sensitivity also varies depending on the organism and is lower in samples pretreated with antimicrobial agents, and ultimately may provide limited additional value compared with CSF culture.³

BioFire Diagnostics FilmArray Meningitis/Encephalitis (ME) Panel:

Description

Multiplexed PCR technologies allow for the simultaneous detection and identification of microorganisms in a single test reaction. Currently there are multiple FDA-cleared/approved multiplex PCR respiratory, gastrointestinal and blood panels.^20,21 Although for years there have been FDA-approved PCR technologies to test for either enteroviruses or herpes simplex virus-1/2, in October, 2015, the FilmArrary® Meningitis/Encephalitis Panel (ME) (BioFire Salt Lake City, UT) was the first FDA-cleared multiplex PCR panel for the evaluation of CSF samples.^20,21 The FilmArray® ME Panel (hereafter referred to as “ME panel”) is a multiplexed PCR able to identify 14 organisms, which include 7 viruses, 6 bacteria, and 1 fungus (Table 1).²² The ME Panel consists of automated nucleic acid extraction, purification, reverse transcription, PCR, DNA melting analysis and automatic results analysis.^21,22 A minimum of 0.2 mL of CSF volume is needed, the ME panel is capable of analyzing 12 samples at a time, and less than 2 minutes of hands-on technician time are required. Most important for clinicians is that results are obtained in approximately 1 hour.²¹ Notably, the ME panel is intended to be used jointly with additional clinical, epidemiological and laboratory data, including CSF culture.²² Additionally, the panel is not intended for CSF specimens collected from indwelling CNS medical devices^22,23 or from patients in whom there is concern for a nosocomial infection, as pathogens which often cause these infections are not included in the ME panel.²⁴

Table 1.

Targets of the BioFire® FilmArray® ME Panel

Viruses	Bacteria	Fungi
Cytomegalovirus (CMV)	Escherichia coli K1	Cryptococcus neoformans/Cryptococcus gattii
Enterovirus (EV)	Haemophilus influenzae
Herpes Simplex Virus-1 (HSV-1)	Listeria monocytogenes
Herpes Simplex Virus-2 (HSV-2)	Neisseria meningitides
Human Herpesvirus 6 (HHV-6)	Streptococcus agalactiae
Human Parechovirus (HPeV)	Streptococcus pneumoniae
Varicella-Zoster Virus (VZV)

Evidence

There have been multiple investigations that have evaluated the ME panel for the detection and identification of pathogens in pediatric and adult patients assessed for meningitis/encephalitis. The largest prospective study to date by Leber et al, evaluated 1,560 CSF samples, and the sensitivity of the ME panel ranged from 85.7% for HHV-6 to 100% for 9 of the 14 organisms. However, there were limited numbers of samples for many of the organisms and two (L. monocytogenes and N. meningitides) were not detected in the study. The specificity of the ME panel was reported to be 99.2% or greater for all 14 organisms.²⁵

A recent meta-analysis by Tansarli et al. identified 8 studies that provide data required to estimate the performance of the ME panel through the evaluation of discordant results between the ME panel and reference methods. Using the data from the meta-analysis, we calculated the overall performance characteristics of the ME panel from the 8 studies (Table 2). Additionally, we calculated the combined performance characteristics for the ME panel: pooled sensitivity 90.2% (95% CI: 86.2–93.1), specificity 97.7% (95% CI: 94.6–99.0), positive predictive value (PPV) 85.1% (95% CI: 81.6–88.2), negative predictive value (NPV) 98.5% (95% CI: 97.9–98.9), likelihood ratio positive (LR+) 38.8 (95% CI: 16.6–90.8), likelihood ratio negative (LR−) 0.10 (95% CI: 0.07–0.14). Due to the limited number of studies and sample sizes, the authors of the meta-analysis were not able to calculate sensitivities and specificities of the ME panel for individual organisms in the panel.¹²

Table 2.

Studies of the BioFire® FilmArray® ME Panel included in the 2019 Meta-Analysis

Study	Study Population	Total Number of Samples in the Study	Type of Study	Overall Sensitivity	Overall Specificity	PPV	NPV	Positive Likelihood Ratio (LR+)	Negative Likelihood Ratio (LR−)
Leber et al.25	Pediatrics and Adults	1560	Prospective	94.2%	97.7%	74.8%	99.6%	41.6	0.06
Arora et al.40	Pediatrics	62	Prospective	100.0%	93.4%	55.6%	100.0%	15.3	0
Lee et al.30	Pediatrics and Adults	42	Prospective	60.0%	100%	100.0%	86.7%	N/A	0.40
Radmard et al.24	Pediatrics and Adults	705	Retrospective	85.7%	98.3%	36.8%	99.9%	52.2	0.13
Hanson et al.53	Pediatrics and Adults	342	Retrospective	91.8%	88.3%	88.4%	91.7%	7.8	0.09
Messacar et al.35	Pediatrics	138	Retrospective	91.1%	97.9%	95.3%	95.9%	43.3	0.09
Graf et al.36	Pediatrics	133	Retrospective	92.5%	100.0%	100.0%	93.0%	N/A	0.07
Piccirilli et al.41	Pediatrics and Adults	63	Retrospective	85.7%	100.0%	100.0%	77.8%	N/A	0.14

From the data gathered by Tansarli et al. from these 8 studies, we also calculated that the overall sensitivity for 5 of the 6 bacterial organisms in the ME panel was 96.8% (95% CI: 92.7–99.0). As the 1 positive case of L. monocytogenes was not able to be corroborated, we were not able to include this organism in the calculation of overall sensitivity for the bacterial pathogens in the panel.¹²

Young infants <3 months of age are particularly susceptible to ME.^6,8,26 Although many studies evaluating the ME panel have included infants in the first 3 month of life^{24,25,27–40}, 2 of the investigations specifically focused on this age group.^38,40 Arora et al. evaluated the ME panel in 62 infants ≤3 months of age and reported a sensitivity of 100% and specificity of 93.4%.⁴⁰ Blaschke et al., however, reported mixed results for infants ≤60 days of age, including two false positive bacterial pathogens detected by the ME panel. Additionally, larger number of infants had viruses detected with the ME panel than with conventional methods, but it was unclear if these detections were true positive or false positive results.³⁸

Potential Benefits and Limitations

There are multiple potential benefits and limitations of the ME panel (Table 3). Potential clinical benefits include the detection of CSF viruses and bacteria with high sensitivity and specificity, as noted previously.¹² Additional clinical benefits reported in the literature include shorter time to pathogen identification,^33,35,41detection of organisms in CSF missed by other conventional studies such as Gram stain^42,43 or culture,^{31,40,44–46} identification of organisms in samples of patients pretreated with antimicrobial agents,^{31,40,42,43,47} and enhanced implementation of chemoprophylaxis for close contacts.⁴⁷

Table 3.

Potential Benefits and Limitations of the BioFire® FilmArray® ME Panel

Potential Benefits

Potential Limitations

Faster turnaround time, diagnosis and definitive treatment/treatment discontinuation30,31,34 Pathogen identification in culture-negative CSF samples from patients with suspected bacterial meningitis31,40,44–46 Detection of organisms in CSF obtained after antimicrobial pretreatment31,40,42,43,47 Enables simultaneous identification of co-infections on the same sample7,56 Ability to test for multiple organisms simultaneously22 Facilitates proper administration of chemoprophylaxis for close contacts47 Relatively small amount of CSF sample (minimum 0.2 mL) required22 Limited hands-on time and technical expertise necessary12,28

Concern for false positive and false negative tests12,15,25,50–52,60 Not all pathogens able to cause CNS infections are detected by the panel22,25,30,31,38,39,41,60 Unable to provide antimicrobial susceptibilities22,25 Not intended for CSF samples obtained from indwelling CNS medical devices22 Positive results do not exclude the possibility of a co-infection with an organism not in the panel22 Relatively high cost of purchase ($35,550–$50,000), service ($4,000/year) and per test ($~200)11,28,48 Lower ability to detect viruses when compared to some singleplex assays15,35,55,56 Positive results for herpesviruses may be due to latency or reactivation of the virus with or without disease15,25

There is also a possible economic impact of using the ME panel. Given the adverse consequences of delayed or misdiagnosed meningitis/encephalitis, patients suspected of ME are often hospitalized for empiric antimicrobial therapy while awaiting the results of CSF culture.^9–11 Fast turnaround times of the ME panel (approximately 1 hour) have the potential to optimize resource utilization by decreasing unnecessary hospitalizations,^9,10 number of other diagnostic tests,^9,10,13 length of stay,^{9–11,13,33,39} and length of empiric antimicrobial therapy.^9,10,13 Theoretical models in pediatric and adult patients have suggested that the ME panel can lead to cost savings when compared with current practice standards.^9,10 A study in adult patients found a significant difference between the median costs per treatment course of antimicrobials for patients who received standard of care testing compared with those in which the ME panel was used.⁴⁸ Another study estimated cost savings of approximately $1,750 per case with the use of the ME panel in patients with suspected CNS infections, due to faster turnaround times comparted with conventional methods.¹¹ These potential cost savings must be evaluated while also taking into account the cost of purchase of the ME panel, the testing itself, and the service of the equipment.^11,48 The elevated cost of performing each test has been considered a limitation for the implementation of the ME panel in low-income countries.²⁸

The duration of antimicrobial therapy with the implementation of the ME panel has also been evaluated. A study in a pediatric population found a shorter duration of antimicrobial therapy after the implementation of the ME panel (2 vs. 3 days).³³ Additionally, in a study focused on partially-treated bacterial meningitis in adult and pediatric patients, the total duration of antimicrobial treatment was shorter with implementation of the ME panel (9.5 days vs. 15.2 days).⁴⁷ A shorter time to narrowing antimicrobials and a decrease in the number of acyclovir doses has also been reported with use of the ME panel,³⁹ as has an increase in use of narrow-spectrum regimens.⁴⁷ Other studies, however, have provided conflicting results, including a study of adult patients that reported no difference in the duration of antimicrobial therapy with the use of the ME panel.⁴⁹ As a possible mechanism for these findings, some investigations found that significant proportion of patients with negative ME panel results were still continued on antimicrobial therapy.^32,49 The potential impact of the implementation of the ME panel on hospital length of stay has also been analyzed. While some studies have reported a shorter duration of hospitalization with use of the ME panel,^11,33,39 others have found no difference.^47,49

There are important limitations of the ME panel. Some investigators have raised concerns for false positives^12,15,25,50 and false negatives^{12,15,25,51,52} with use of the ME panel. Case reports have described how false positive and/or false negative ME panel results led to delayed diagnosis of the causative pathogen.^50,51 Studies have also suggested that the ME panel should not replace the cryptococcal antigen test^12,15,44,51 and culture^12,51 for patients with suspicion of C. neoformans/C. gattii ME. Specifically, some potential false negative results of C. neoformans/C. gattii on ME panel have occurred in patients with low burden of disease^51,53 and/or in patients on antifungal treatment.^12,22,25,51 Additionally, studies have suggested the possibility that positive antigen results after initiation of therapy may indicate persistence of antigen and not actual detection of live organisms.^22,25,54 False negative results for viruses may be due to specimens containing low viral loads⁴¹ and to the lower ability of the ME panel to detect viruses when compared to some singleplex assays.^15,35,55,56 With regards to false positive results, there are concerns for the potential of contamination during collection and processing of CSF samples.^12,25

It is also important to highlight that all herpesviruses in the ME panel (HSV-1, HSV-2, CMV, VZV, HHV-6) can establish latent infections. Therefore, a positive result in the ME panel may be due to a primary infection, or alternatively to a latent infection present in the cells retrieved in the specimen (either the CSF or from peripheral blood in a traumatic tap) or reactivation of the virus (with or without true disease).^15,25,57 This accentuates the importance of evaluating the full clinical scenario when interpreting ME panel results.^12,25,58,59 In addition, HHV-6 can be integrated into human chromosomes and transmitted vertically giving a positive ME panel result.^41,58

Promising Additional Rapid Diagnostic Technologies:

Multiplexed PCR panels, like the BioFire® FilmArray® ME Panel, are one of several rapid diagnostic approaches that have the potential to overcome some of the existing limitations in the diagnosis of CNS infections. Some of these diagnostics employ different PCR-based techniques to improve the diagnostic yield such as, the utilization of nested-PCR (as in the BioFire® FilmArray®), loop-mediated isothermal amplification (LAMP),⁶¹ and 16s ribosomal RNA sequencing (broad-range PCR).⁶² Others employ different approaches for the identification of microorganisms such as matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF).⁶³ and metagenomics next-generation sequencing.⁶⁴ All of these offer promising avenues to improve our current strategies to diagnose CNS infections, but require further research.

Conclusions:

The BioFire® FilmArray® Meningitis/Encephalitis Panel is the first FDA-cleared multiplexed PCR capable of simultaneously detecting and identifying 14 organisms in CSF samples. This newer rapid diagnostic tool has an overall high sensitivity and specificity for CNS infections and has the potential to improve diagnosis and optimize utilization of healthcare resources for patients undergoing evaluation for ME. However, the ME panel should not be used as the sole diagnostic tool in patients with suspected bacterial meningitis, and clinicians should interpret ME panel results in combination with clinical, epidemiological and laboratory data. Additionally, both false positive and false negative results have been reported. A negative ME panel test does not indicate the absence of infection, as only 14 organisms are included in the panel, and a positive test may not necessarily reflect the true disease-causing organism, such as with latent viral infections. More research is needed to guide laboratories and clinicians in determining the optimal use of the ME panel panel in clinical decision-making for pediatric and adult patients undergoing evaluation for CNS infections.

CME Questions:

Which of the following viruses are part of the BioFire® FilmArray® ME Panel and are known to cause latent infections?

1. HSV-1/2
2. CMV
3. VZV
4. HHV-6
5. All of the above

e) All of the above

Which of the following subgroup of organisms are all targets of the FDA-approved version of the BioFire® FilmArray® ME Panel?

1. CMV, EBV, S. Aureus, S. agalactiae, C. neoformans
2. VZV, HHV-6, M. Tuberculosis, S. pneumoniae, S. Aureus
3. HPeV, EV, S. pneumoniae, L. monocytogenes, C. neoformans/C. gattii
4. HIV, HPeV, E. coli K1, M. Tuberculosis, C. neoformans/C. gattii
5. HSV-1/2, VZV, S. Aureus, N. meningitides, M. Tuberculosis

c) HPeV, EV, S. pneumoniae, L. monocytogenes, C. neoformans/C. gattii

Per the manufacturer, what is the minimum sample volume needed for the BioFire® FilmArray® ME Panel?

1. 0.2 mL
2. 1.0 mL
3. 1.5 mL
4. 3.0 mL
5. 5.0 mL

a) 0.2 mL

Which of the following has been highlighted as a significant concern in studies evaluating the BioFire® FilmArray® ME Panel?

1. High technical expertise required for the processing of samples
2. False positive and false negative results
3. Turnaround time of the assay
4. Amount of CSF volume required
5. Number of samples that can be run at the same time

b) False positive and false negative results

One of the reasons CSF cultures continue to be essential for the evaluation of patients for CNS infections is:

1. Higher sensitivity than any other method
2. Higher specificity than any other method
3. Faster turnaround time than most other methods
4. Importance for determining antibiotic susceptibilities
5. Diagnostic yield unchanged on antibiotic pretreated samples

d) Importance for determining antibiotic susceptibilities