Surveillance of upper respiratory infections using a new multiplex PCR assay compared to conventional methods during the influenza season in Taiwan

Developing and validating clinical features-based machine learning algorithms to predict influenza infection in influenza-like illness patients

BACKGROUND

Seasonal influenza poses a significant risk, and patients can benefit from early diagnosis and treatment. However, underdiagnosis and undertreatment remain widespread. We developed and compared clinical feature-based machine learning (ML) algorithms that can accurately predict influenza infection in emergency departments (EDs) among patients with influenza-like illness (ILI).

MATERIAL AND METHODS

We conducted a prospective cohort study in five EDs in the US and Taiwan from 2015 to 2020. Adult patients visiting the EDs with symptoms of ILI were recruited and tested by real-time RT-PCR for influenza. We evaluated seven ML algorithms and compared their results with previously developed clinical prediction models.

RESULTS

Out of the 2189 enrolled patients, 1104 tested positive for influenza. The eXtreme Gradient Boosting achieved superior performance with an area under the receiver operating characteristic curve of 0.82 (95% confidence interval [CI] = 0.79-0.85), with a sensitivity of 0.92 (95% CI = 0.88-0.95), specificity of 0.89 (95% CI = 0.86-0.92), and accuracy of 0.72 (95% CI = 0.69-0.76) in the testing set over cut-offs of 0.4, 0.6 and 0.5, respectively. These results were superior to those of previously proposed clinical prediction models. The model interpretation revealed that body temperature, cough, rhinorrhea, and exposure history were positively associated with and the days of illness and influenza vaccine were negatively associated with influenza infection. We also found the week of the influenza season, pulse rate, and oxygen saturation to be associated with influenza infection.

CONCLUSIONS

The clinical feature-based ML model outperformed conventional models for predicting influenza infection.

Applying symptom dynamics to accurately predict influenza virus infection: An international multicenter influenza-like illness surveillance study

BACKGROUND

Public health organizations have recommended various definitions of influenza-like illnesses under the assumption that the symptoms do not change during influenza virus infection. To explore the relationship between symptoms and influenza over time, we analyzed a dataset from an international multicenter prospective emergency department (ED)-based influenza-like illness cohort study.

METHODS

We recruited patients in the US and Taiwan between 2015 and 2020 with: (1) flu-like symptoms (fever and cough, headache, or sore throat), (2) absence of any of the respiratory infection symptoms, or (3) positive laboratory test results for influenza from the current ED visit. We evaluated the association between the symptoms and influenza virus infection on different days of illness. The association was evaluated among different subgroups, including different study countries, influenza subtypes, and only patients with influenza.

RESULTS

Among the 2471 recruited patients, 45.7% tested positive for influenza virus. Cough was the most predictive symptom throughout the week (odds ratios [OR]: 7.08-11.15). In general, all symptoms were more predictive during the first 2 days (OR: 1.55-10.28). Upper respiratory symptoms, such as sore throat and productive cough, and general symptoms, such as body ache and fatigue, were more predictive in the first half of the week (OR: 1.51-3.25). Lower respiratory symptoms, such as shortness of breath and wheezing, were more predictive in the second half of the week (OR: 1.52-2.52). Similar trends were observed for most symptoms in the different subgroups.

CONCLUSIONS

The time course is an important factor to be considered when evaluating the symptoms of influenza virus infection.

A retrospective study examining the clinical significance of testing respiratory panels in children who presented to a tertiary hospital in 2019

Background. Respiratory tract infections are a leading cause of hospital visits in the paediatric population and carry significant associated morbidity and mortality in this population. The introduction of respiratory panel testing has been said to guide clinicians in the overall management of patients.

Methods. We conducted a retrospective study examining all respiratory panels carried out in our hospital during 2019 on paediatric patients. Patients included were those who had symptoms indicative of respiratory infections who presented acutely, including those with chronic respiratory conditions. A total of 188 respiratory panel results were obtained along with collected patient data. These were analysed using SPSS V. 25.0 to get the below mentioned results.

Results. The majority (76.6 %) of patients were less than 3 years with 59 % of total population being males. The majority (80.9 %) had mild clinical severity score. The most common pathogen that was detected on the respiratory panel was Enterovirus Human Rhinovirus spp, followed by the influenza viruses. Only four cases were positive for bacterial pathogens (two Mycoplasma pneumoniae, one Bordetella pertussis and one Chlamydia pneumoniae), which accounts for 2.1 % of all panels analysed. The significance of respiratory panels in influencing treatment were analysed in the forms of change of management plans before and after results of respiratory panels. This was observed in 14.4 % of patients who were not on any empiric medication and then based on panel results were started on medications, as well as 11.7 % who were on medications already, and the medications were altered based on the result of the panel (Chi square P=0.057). This was mainly seen with cases of influenza A H1N1 patients and to a lesser extent, Mycoplasma pneumonia.

Conclusion. The use of respiratory panels in our hospital had little impact on patient care and management. The main organisms that influenced clinician decision in treatment were influenza A viruses and bacterial organisms (Mycoplasma pneumoniae, Chlamydia pneumoniae and Bordetella pertussis). Other than that, the use of clinical judgement proved more beneficial. We recommend use of specific testing for these organisms rather than the whole panel as case to case bases, which would be more cost-effective and consistent with patient management.

Repeat Molecular Testing for Respiratory Pathogens: Diagnostic Gain or Diminishing Returns?

BACKGROUND

Upper respiratory tract infections are common, and the ability to accurately and rapidly diagnose the causative pathogen has important implications for patient management.

METHODS

We evaluated the test-ordering practices for 2 commonly utilized nucleic acid amplification tests (NAATs) for the detection of respiratory pathogens: the Xpert Flu Assay for influenza A/B (Flu assay) and the Biofire FilmArray respiratory panel assay (RP assay), which detects 20 different targets. Our study examined repeat testing; that is, testing within 7 days from an initial test.

RESULTS

Our study found that repeat testing is common for each of the individual assays: 3.0% of all Flu assays and 10.0% of all RP assays were repeat testing. Of repeat testing, 8/293 (2.7%) of repeat Flu assays and 75/1257 (6.0%) of RP assays resulted diagnostic gains, i.e., new detections. However, for the RP assay, these new detections were not always clinically actionable. The most frequently discrepant organisms were rhinovirus/enterovirus (28/102, 27.5%), followed by respiratory syncytial virus (12/102, 11.8%) and coronavirus OC43 (11/102, 10.8%). Furthermore, there were 3,336 instances in which a patient was tested using both a Flu assay and RP assay, of which only 44 (1.3%) had discrepant influenza results.

CONCLUSIONS

Our findings suggest opportunities exist to better guide ordering practices for respiratory pathogen testing, including limiting repeat testing, with the goal of optimization of clinical yield, and diagnostic stewardship.

Advances in laboratory assays for detecting human metapneumovirus

Human metapneumovirus (HMPV) is one of the major causes of acute respiratory tract infection (ARI) and shows high morbidity and mortality, particularly in children and immunocompromised patients. Various methods for detecting HMPV have been developed and applied in clinical laboratories. When reviewing the literature, we found that polymerase chain reaction (PCR)-based assays have been most frequently and consistently used to detect HMPV. The most commonly used method was multiplex reverse transcriptase-PCR (RT-PCR; 57.4%), followed by real-time RT-PCR (38.3%). Multiplex RT-PCR became the more popular method in 2011-2019 (69.7%), in contrast to 2001-2009 (28.6%). The advent of multiplex PCR in detecting broader viral pathogens in one run and coinfected viruses influenced the change in user preference. Further, newly developed microarray technologies and ionization mass spectrometry were introduced in 2011-2019. Viral culture (including shell vial assays) and fluorescent immunoassays (with or without culture) were once the mainstays. However, the percentage of studies employing culture and fluorescent immunoassays decreased from 21.4% in 2001-2010 to 15.2% in 2011-2019. Meanwhile, the use of PCR-based methods of HMPV detection increased from 78.6% in 2001-2010 to 84.8% in 2011-2019. The increase in PCR-based methods might have occurred because PCR methods demonstrated better diagnostic performance, shorter hands-on and run times, less hazards to laboratory personnel, and more reliable results than traditional methods. When using these assays, it is important to acquire a comprehensive understanding of the principles, advantages, disadvantages, and precautions for data interpretation. In the future, the combination of nanotechnology and advanced genetic platforms such as next-generation sequencing will benefit patients with HMPV infection by facilitating efficient therapeutic intervention. Analytical and clinical validation are required before using new techniques in clinical laboratories.

Rapid detection of respiratory organisms with FilmArray respiratory panel and its impact on clinical decisions in Shanghai, China, 2016-2018

BACKGROUND

In this study, we evaluated the diagnostic potential and clinical impact of an automated multiplex PCR platform (the FilmArray Respiratory Panel; FA-RP), specially designed for pathogen detection in respiratory tract infections in adults with unexplained pneumonia (UP).

METHODS

A total of 112 UP patients in Shanghai, China, were enrolled prospectively and assessed using the FA-RP from October 2016 to March 2018. We examined the test results and their influence on clinical decisions. Furthermore, as a control group, we retrospectively obtained the clinical data of 70 UP patients between October 2014 and March 2016 (before the FA-RP was available). The two patient groups were compared with respect to factors, including general antimicrobial use and defined daily dose (DDD) numbers.

RESULTS

Between October 2016 and March 2018, the positive rate obtained using FA-RP for UP was 76.8%. The primary pathogens in adults with UP were Influenza A/B (47.3%, 53/112). Compared with the patients before FA-RP was available, patients who underwent FA-RP testing had higher rates of antiviral drug use and antibiotic de-escalation during clinical treatment. FA-RP significantly decreased the total DDDs of antibiotic or antifungal drugs DDDs by 7 days after admission (10.6 ± 2.5 vs 14.1 ± 8.8, P < .01).

CONCLUSIONS

The FA-RP is a rapid and sensitive nucleic acid amplification test method for UP diagnosis in adults. The application of FA-RP may lead to a more accurately targeted antimicrobial treatment and reduced use of antibiotic/antifungal drugs.

Evaluation of FilmArray respiratory panel multiplex polymerase chain reaction assay for detection of pathogens in adult outpatients with acute respiratory tract infection

Although viruses are the major pathogen that causes upper respiratory tract infection (URTI) and acute bronchitis, antibiotics have been prescribed. This was a prospective observational study in influenza epidemics that enrolled adult outpatients who visited a hospital with respiratory tract infection symptoms. In this study, we evaluated the usefulness of FilmArray respiratory panel (RP). Fifty patients were enrolled. FilmArray RP detected the pathogens in 28 patients. The common pathogens were influenza virus (n = 14), respiratory syncytial virus (n = 6), and human rhinovirus (n = 6). Of the 14 patients with influenza virus, 6 were negative for the antigen test. The physicians diagnosed and treated the patients without the result of FilmArray in this study. Of the patients with positive FilmArray RP, 9 were treated with antibiotics; however, bacteria were detected in only 3 patients. By implementing FilmArray RP, URTI and acute bronchitis would be precisely diagnosed, and inappropriate use of antibiotics can be reduced.

Comparison of Respiratory Pathogen Detection in Upper versus Lower Respiratory Tract Samples Using the BioFire FilmArray Respiratory Panel in the Immunocompromised Host

Background

The FilmArray Respiratory Panel (FARP) (BioFire Diagnostics, Inc.) is a multiplex, polymerase chain reaction (PCR) technique that can detect 17 respiratory viruses and 3 bacterial targets in a single reaction. Immunocompromised hosts (ICH) with respiratory illnesses often undergo bronchoscopy with bronchoalveolar lavage (BAL). This prospective study aimed to evaluate the yield and concordance of NP and BAL FARP testing when performed on the same patient concurrently.

Methods

From February to December 2016, 125 patients (100 ICH and 25 non-ICH) were enrolled. NP swabs and BAL samples were sent for FARP testing.

Results

The yield of the BAL FARP among ICH and non-ICH was 24% (24/100) and 8% (2/25), respectively. The yield of positive NP swabs in ICH was 27% (27/100) versus 4% (1/25) in non-ICH. The majority of patients (89%; 111/125) had concordant results between NP and BAL specimens. Of the 24 ICH patients who had a positive BAL FARP, the majority (79%) had the same pathogen detected from the NP swab.

Conclusion

The FARP may be useful in the ICH. Given the high concordance, in patients whom a pathogen is identified on the NP FARP, a FARP performed on BAL will likely yield the same result. However, if the NP FARP is negative, performing the test on a BAL sample may have an incremental yield.

Multiplex PCR system for the rapid diagnosis of respiratory virus infection: systematic review and meta-analysis

OBJECTIVES

To provide a summary of evidence for the diagnostic accuracies of three multiplex PCR systems (mPCRs)-BioFire FilmArray RP (FilmArray), Nanosphere Verigene RV+ test (Verigene RV+) and Hologic Gen-Probe Prodesse assays-on the detection of viral respiratory infections.

METHODS

A comprehensive search up to 1 July 2017 was conducted on Medline and Embase for studies that utilized FilmArray, Verigene RV+ and Prodesse for diagnosis of viral respiratory infections. A summary of diagnostic accuracies for the following five viruses were calculated: influenza A virus (FluA), influenza B virus, respiratory syncytial virus, human metapneumovirus and adenovirus. Hierarchical summary receiver operating curves were used for estimating the viral detection performance per assay.

RESULTS

Twenty studies of 5510 patient samples were eligible for analysis. Multiplex PCRs demonstrated high diagnostic accuracy, with area under the receiver operating characteristic curve (AUROC) equal to or more than 0.98 for all the above viruses except for adenovirus (AUROC 0.89). FilmArray, Verigene RV+ and ProFlu+ (the only Prodesse assay with enough data) demonstrated a summary sensitivity for FluA of 0.911 (95% confidence interval, 0.848-0.949), 0.949 (95% confidence interval, 0.882-0.979) and 0.954 (95% confidence interval, 0.871-0.985), respectively. The three mPCRs were comparable in terms of detection of FluA.

CONCLUSIONS

Point estimates calculated from eligible studies showed that the three mPCRs (FilmArray, Verigene RV+ and ProFlu+) are highly accurate and may provide important diagnostic information for early identification of respiratory virus infections. In patients with low pretest probability for FluA, these three mPCRs can predict a low possibility of infection and may justify withholding empirical antiviral treatments.