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

Abstract 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.


| INTRODUC TI ON
After the severe acute respiratory syndrome outbreak, in order to track similarly highly contagious and severe lower respiratory illness with pneumonia symptoms, the Chinese government has paid great attention to unexplained pneumonia (UP), a phrase coined by the Chinese Human Unexplained Pneumonia Surveillance Network (CHUPSN) and published in 2004. UP refers to pneumonia, which could be life-threatening, characterized by rapid progression, a normal white blood cell count range, and poor response to antibiotics. 1 Although the common pathogens causing UP are viruses or atypical pathogens, their overlapping clinical manifestations can impede the ability of clinicians to directly diagnose the causative pathogens, which may lead to unnecessary antimicrobial usage. Thus, rapid and precise diagnosis of the causative agents is critical for the prompt management of UP.
To date, culture-based and other traditional methods have been used to help the clinical approach to diagnose UP; however, these methods have certain disadvantages. For example, rapid antigen testing has only a 40%-100% sensitivity for adult influenza and is dependent on the use of specific scientific techniques, whereas culturing viruses can take up to 10 days. 2 In recent years, there has been an increase in the use of nucleic acid amplification test (NAAT)-based methods for the detection of viral pathogens due to their excellent diagnostic ability in identifying a broad spectrum of pathogens. 3

| Study design and participants
The prospective cohort was enrolled in Shanghai, China, between

| Sampling
Within 24 hours of patient admission, respiratory tract specimens were collected, including bronchoalveolar lavage fluid (BALF: among those patients who gave consent to undergo bronchoscopy), sputum (if expectoration was evident), and nasopharyngeal swab (NPS, if no bronchoscopy was performed or no expectoration sample was collected). Notably, lower respiratory tract specimens were given higher priority (BALF > sputum > NPS) during the study. As soon as possible after sampling, the specimens were transferred to the Laboratory of Infectious Diseases Research Institute, Huashan Hospital, Shanghai.
Professional pulmonary physicians oversaw BALF sampling, while the collection of NPS and sputum was performed by clinical physicians per the clinical standard operating procedure. Notably, those samples of NPS and BALF for which the viscosity was low required no further sample preparation before FilmArray panel testing, whereas viscous samples, such as sputum, were pre-treated using the previously described dunk and swirl method. 5 The remaining specimens were stored in a −80°C freezer for further use.

| The FilmArray respiratory panel
In this study, we used version 1.7 of the FA-RP. It is an automated multiplex PCR detection platform designed to detect 20 common pathogens causing respiratory infection and has a single specimen turnaround time of <2 hours. Tests using the panel were performed following the manufacturer's instructions.

| Real-time RT-PCR
For specimens that the FA-RP reported as "Influenza A, not sub-

| Traditional laboratory tests
For both patient groups, traditional standard-of-care laboratory diagnostic tests for respiratory infections were performed according to the requests of physicians, including smears, cultures, and serological tests. The serological test in our study was the PNEUMOSLIDE IgM (Vircell) commercial kit based on indirect immunofluorescence.

| Evaluation of clinical antimicrobial use
In this study, clinical data, including antimicrobial use, and adjustment or de-escalation of antimicrobial therapy, were compared between the FA-RP group and retrospective control group. Turnaround time was defined as the length of time from sample collection to receiving the FA-RP results. Antimicrobial use was quantified as the defined daily dose (DDD)) number, calculated as ∑ (Drug Total Dosage (g)/Drug DDD index). 6 Adjustment of antimicrobial therapy was defined as a change in antimicrobial regimen immediately after receiving the FA-RP results.

F I G U R E 1
Overview of Patient Enrollment Workflow. A total of 112 patients selected prospectively were submitted to FA-RP analyses and are referred to as the "FA-RP group." Those who were not lost to follow-up and had complete data are referred to as the "intervention group." Comparison was made between it and the historical control group

| Summary of pathogens detected in prospective group patients using the FA-RP
For 13 samples, the FA-RP results were "Influenza A/not subtyped." In these cases, further subtyping was performed using real-time RT-PCR. As a consequence, Influenza A virus subtype H7N9 was detected in 11 samples and H1N1-2009 in the remaining two samples. As a consequence of the findings of the FA-FP regarding Influenza A, 15 patients were isolated in a negative pressure ward or single ward, including seven highly pathogenic H7N9 cases. In contrast, in the control group, no comparable isolation measures were undertaken.

| Impact on treatment decisions compared with the control group
When we compared antimicrobial use between the intervention and control groups, we found that the two groups were comparable for the most baseline characteristics (Table 3). Compared with the control group, the intervention group showed no significant decrease of antimicrobial DDDs (control group 14.3 ± 9.6 vs intervention group 13.8 ± 4.6, P > .05). However, after excluding antiviral drugs, we detected significantly lower antibiotic/antifungal DDDs

| D ISCUSS I ON
In this study, we applied FA-RP on diagnosis of UP, which is a certain type of pneumonia causing mainly by viruses and atypical pathogens.
We found that the FA-RP detected pathogens in 76.8% (86/112) of all specimens, which is higher than that the detection rate obtained when this technique was applied for general community-acquired pneumonia (38.6%, 49/127). 7 These results, therefore, show that when applied clinically for UP, the FA-RP would provide a higher probability of identifying causal pathogenic microorganisms.  In the present study, we found that 53.8% (60/112) of the patients in the FA-RP group received antiviral treatment, which is significantly higher than that in the control group (14.3%, 10/70;  randomized controlled trial discovered that use of the FA-RP was not associated with a reduction in the overall duration of antibiotic usage. 18 Nevertheless, a higher number of patients in the FA-RP group received single doses or brief courses of antibiotics than those in the control group. Given these inconclusive findings, further studies are required concerning the impact of FA-RP on the clinical decisions regarding pneumonia.
We compared the duration of hospital stay between the two groups. As a result, intervention group unexpectedly showed longer duration of hospital stay than the control group (12.1 vs 7.6 days, P = .001). One reasonable explanation is that subjects in the intervention group had more severe cases of pneumonia than those in the control group (higher PSI, This present study is one of the several that have examined the application of FA-RP in Asia. [19][20][21][22][23][24] Compared with other published studies, our study focused on adults with more severe pneumonia conditions, and to the best of our knowledge, it is the only study that Our study does, however, have a few limitations. Notably, the sample size was limited, and the study was conducted only at a single location, which may thus limit the general applicability of the findings. In addition, the study included two cohorts from different time periods; hence, there may have been differences in factors that influenced clinician behavior between the two periods that were not accounted in our analysis. Furthermore, the incidence of various respiratory viruses may have varied across the two periods that may thus limit the validity of a direct comparison of clinical outcomes.
In conclusion, the FA-RP is a rapid and sensitive NAAT method that can be used for the diagnosis of UP in adults. In the two recent years in Shanghai, the most common pathogen causing UP was identified as the Influenza virus. However, in this regard, it should be taken into consideration that the current version of the FA-RP is unable to directly detect H7N9. Nevertheless, the results obtained in this study show that the FA-RP may make a valuable contribution to clinical decision making and facilitate the reasonable use of antibiotics.

ACK N OWLED G EM ENT
We would like to thank Yinlei Zhang, Tracy Wang, Xuyang Wang, and Xinchang Chen for their substantial contributions to this study.