key: cord-0866257-5lcfxjkc authors: Thi Ty Hang, Vu; Thi Han Ny, Nguyen; My Phuc, Tran; Thi Thanh Tam, Pham; Thao Huong, Dang; Dang Trung Nghia, Ho; Tran Anh Vu, Nguyen; Thi Hong Phuong, Pham; Van Xang, Nguyen; Dong, Nguyen; Nhu Hiep, Pham; Van Hung, Nguyen; Tinh Hien, Tran; Rabaa, Maia; Thwaites, Guy E.; Baker, Stephen; Van Tan, Le; van Doorn, H.Rogier title: Evaluation of the Luminex xTAG Respiratory Viral Panel FAST v2 assay for detection of multiple respiratory viral pathogens in nasal and throat swabs in Vietnam date: 2018-04-30 journal: Wellcome Open Res DOI: 10.12688/wellcomeopenres.12429.2 sha: 04d638b3c1d8b8fcd771e569e439b364d4ac2c1a doc_id: 866257 cord_uid: 5lcfxjkc Background: Acute respiratory infections (ARI) are among the leading causes of hospitalization in children ≤5 years old. Rapid diagnostics of viral pathogens is essential to avoid unnecessary antibiotic treatment, thereby slowing down antibiotic-resistance. We evaluated the diagnostic performance of the Luminex xTAG Respiratory Viral Panel FAST v2 against viral specific PCR as reference assays for ARI in Vietnam. Methods: Four hundred and forty two nose and throat swabs were collected in viral transport medium, and were tested with Luminex xTAG Respiratory Viral Panel FAST v2. Multiplex RT-PCR and single RT-PCR were used as references. Results: Overall, sensitivity of the Luminex against reference assays was 91.8%, 95% CI 88.1-94.7 (270/294), whilst 112/6336 (1.8%, 95% CI, 1.4-2.1) of pathogens were detected by the Luminex, but not by reference assays. Frequency of pathogens detected by Luminex and reference assays was 379 and 292, respectively. The diagnostic yield was 66.7% (295/442, 95%CI 62.1-71.1%) for the Luminex assay and 54.1% (239/442, 95% CI, 49.3-58.8%) for reference assays. The Luminex kit had higher yields for all viruses except influenza B virus, respiratory syncytial virus, and human bocavirus. High agreements between both methods [mean (range): 0.91 (0.83-1.00)] were found for 10/15 viral agents. Conclusions: The Luminex assay is a high throughput multiplex platform for rapid detection of common viral pathogens causing ARI. Although the current high cost may prevent Luminex assays from being widely used, especially in limited resource settings where ARI are felt most, its introduction in clinical diagnostics may help reduce unnecessary use of antibiotic prescription. (0) Comments agreements between both methods [mean (range): 0.91 (0.83-1.00)] were found for 10/15 viral agents. : The Luminex assay is a high throughput multiplex platform for Conclusions rapid detection of common viral pathogens causing ARI. Although the current high cost may prevent Luminex assays from being widely used, especially in limited resource settings where ARI are felt most, its introduction in clinical diagnostics may help reduce unnecessary use of antibiotic prescription. Acute respiratory infections (ARI) are the leading cause of morbidity and mortality in infants and children worldwide, especially in Southeast Asia (including Vietnam) and Africa 1,2 . In Vietnam, there is a high burden of ARI in children in the first year of life, who are more likely to be admitted to intensive care and have a longer hospital stay than children with other infectious diagnoses 3 . Viruses are the most common causes of ARI 4-8 . Rapid identification of causative agents is therefore of clinical and public health significance, and may reduce the widespread inappropriate use of antibiotics. The advances in molecular diagnostics have provided powerful means for detection of viruses in terms of sensitivity, specificity and turnaround time 5, [9] [10] [11] [12] Equal volumes of VTM from nose and throat swabs were pooled and subjected to total nucleic acid extraction after addition of internal control (EAV -equine arteritis virus for reference assays and bacteriophage MS2 for the Luminex assay) using the MagNApure 96 platform (Roche Diagnostics, Mannheim, Germany), according to the manufacturer's instructions. Extracted nucleic acids were eluted in 50 ul of elution buffer and stored at -80°C for further analyses. The kits were provided for free by the Luminex company. ; and the 4-tube real time multiplex RT-PCR 22 . Because the influenza A viruses, InfB and RVS targets were tested in the above specific RT-PCRs, they were not tested in the 4-tube assay. Detection of Influenza A virus H1N1pdm09 was performed by an in-house RT-PCR assay using primer set targeting HA (haemagglutinin) gene segment 4 (forward primer, 5'-GTTACCCAGGAGATTTCATCGA-3'; reverse primer, 5'-CATGCTGCCGTTACACCTTTG-3'; and probe, 5'-FAM-AAGT-TCATGGCCCAATCATGACTCGA-BHQ1-3' [FAM, 6-carboxyfluorescein, BHQ1, black hole quencher 1]). The reference was considered to be positive if any one of the reference assays was positive. Performance of the Luminex assay was evaluated as diagnostic yields and sensitivity against reference assays with 95% confidence interval using 2x2 tables. The calculation of these was performed with Intercooled Stata 9.2 (Stata, College station, TX, USA). Agreement between the Luminex and reference assays was determined by Kappa statistic using SPSS version 23 (IBM Corp. SPSS Statistic, NY, USA). McNemar test (SPSS version 23) was used to examine whether there was any difference of diagnostic rates of individual viruses between the Luminex and reference assays. Statistical significance was set at P<0.05. A total of 442 samples collected between November 2012 and April 2014 were analysed for the evaluation of the Luminex assay. Three hundred forty-eight samples were from children (≤15 years old) and 94 samples from adults. The male/female ratio was 0.62 (273): 0.38 (169). The median age of children was 1 year (Interquartile range, IQR: 1, 2) and of adults 46 years (IQR: 34, 72). Overall, 302 specimens (68.3%) were positive by either Luminex or reference assays or both. One hundred forty samples (31.7%) remained undiagnosed. ENT/Rhi was the most frequently detected pathogen by the two techniques (159 over 403 total count of all pathogens, 40%), followed by hBoV (n = 45, 11%) and PIV3 (n = 42, 10%). Less frequently detected were ADV (n = 32, 8%), hMPV (n = 29, 7%), InFA (n = 22, 6%), hCoV (n = 22, 6%), RSV A and B (n = 21, 5%), PIV4 (n = 12, 3%), InFB (n = 10, 2.5%), PIV1 (n = 6, 1%) and PIV2 (n = 3, 0.5%). Regarding mixed-infection, the Luminex assay also detected more co-infections compared to the reference assays: 68 versus 47. The maximum number of pathogens detected in a single patient was 4. Parechoviruses were detected in four samples by the reference assays, but were not included in the Luminex assay. HCoV were not subtyped by reference assays. also detected by Luminex assay (true positive rate or "sensitivity" against reference assays as gold standard). There were 112 pathogens detected by Luminex but not detected by references, corresponding to an additional 1.8% (95%CI: 1.4-2.1) positives. In addition, there were 24 pathogens detected by reference assays but not detected by the Luminex, corresponding to an additional 0.3% (95% CI: 0.2-0.6) positives. For individual targets, detection rate by Luminex was higher for 12 targets, ranging from 88.8% to 100% (ENT/Rhi, PIV3, ADV, hMPV, InFA matrix, H1N1pdm09, H3N2, hCoV, RSVA, PIV4, PIV1, PIV2), whilst it was less often positive for hBoV and RSVB and InFB (35 vs 43, 10 vs 13 and 9 vs 10, respectively). Remarkably, 43.4% (69/159) of ENT/Rhi were detected by Luminex but were negative by reference assays. Among the 442 clinical swabs, concordance between the two techniques was noted in 372 samples and discordance was recorded in 70 samples, showing substantial agreement (overall kappa 0.67, 95%CI 0.61-0.73). Table 2 also shows the test agreement of all viral pathogens. Almost perfect agreement was recorded in 10 pathogens, hBoV, PIV3, InFA matrix, H1N1pdm09, H3N2, hCoV, RSVA, RSVB, InFB and PIV2 (kappa 0.83 -1.00). Substantial agreement (kappa 0.62-0.80) was seen in hMPV, ENT/Rhi, ADV and PIV1, while agreement in PIV4 was moderate (kappa 0.58). Fast and reliable diagnostic tests are a practical need in helping physicians to make appropriate treatment decisions and are a useful tool in research and surveillance. We report here the evaluation of a Luminex assay on respiratory swabs collected from patients admitted to four provincial hospitals in Vietnam. Lacking a true "gold standard", we combined a number of published PCR assays as reference tests for evaluation. Overall positivity of Luminex among reference assays was high (91.8%, CI 95% 88. 1-94.7) . Besides that 5 targets (ENT/ Rhi , ADV, hMPV, PIV4 and PIV1) which were all more often detected by Luminex than by reference assays and were confirmed by single-RT PCR (number of +L/-ref cases ranged from 3 -69; Table 2 ). This suggests that detection using Luminex is superior to reference assays for these targets. Compared to reference assays, our study found an increased detection rate by the Luminex for most targets, and significant difference was seen in Ent/Rhi and PIV4 (+L/-ref= 69, p<0.001; and +L/-ref= 7, p=0.01, respectively; Figure 1 ). Especially a high number of +L/-ref for Ent/Rhi agent shows a considerable difference with other studies in previous xTAG Luminex studies 15, 18, [23] [24] [25] ; it probably again reveals that Luminex is a strong assay in detection of this viral agent. Meanwhile, detection rate for hBoV was significantly higher in reference assays than in the Luminex (10 versus 2, p=0.04; Figure 1 ), which is similar to other previous studies 15,18 (sensitivity of Luminex for this viral agent was rather low, 41.4% and 20.0% in these studies, respectively). Though the Luminex assay may be of benefit to diagnostics and cost of treatment 12, 13, 15, 17, 26 , it requires a specific instrument for detection and data acquisition. Therefore, it may not be appropriate to laboratories with limited equipment. However, with a highly automated system and the capacity to test up to 94 specimens within two hours (not including nucleic acid extraction and hands-on time), this high throughput Luminex RVP FAST method would be useful in large hospitals where they could have high input of respiratory samples to run by batch. Globally, the fear of clinical worsening for patients with ARI usually results in empiric antibiotic prescription, even though doctors are aware that most ARI are caused by viruses. One of the factors contributing to this is the long interval between sampling and reporting of test results. The Luminex assay may be part of the solution to this with its fast turnaround time. The Luminex has a number of weaknesses: it cannot distinguish enterovirus and rhinovirus, it is not quantitative, it comprises a two-tube step for RT-PCR and DNA hybridization, including an open-tube step for transferring PCR product from RT-PCR tube to hybridization tube, which brings a risk for contamination, and it is expensive. A significantly lower positivity rate of the Luminex assay for hBoV found in this study (P=0.04) is also a weakness of this kit, which may need further clinical evaluation. Limitations of this study are the lack of a true gold standard as is commonly seen when evaluating diagnostic assays and low numbers of positive samples for several 'uncommon' viruses (such as PIV1, PIV2). This low number of positive samples may conceal true diagnostic rates for these viruses. In conclusion, our study shows the Luminex RVP FAST has a good diagnostic performance for detection of multiple respiratory viruses. Results from this study provided an additional evaluation on the utility of this commercial test. Though the cost of Luminex assay is rather high, Luminex RVP FAST platform could become affordable in large hospitals where samples are high input, reducing cost of the test by batch run. Once the per assay cost of this assay become more affordable, the above advantages and the short turnaround time could contribute to improving patient management and changing the prescription culture in countries like Vietnam. No competing interests were disclosed. This work was supported by the Wellcome Trust [093724]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Epidemiology and etiology of childhood pneumonia Although more expensive than in-house diagnostic assays, Luminex performed well and it could be of real benefit in laboratories with a high throughput of samples. The paper is well written, sample size is appropriate and conclusions are supported by the results. I only have minor suggestions. I think Table 2 is a bit difficult to follow. When I read "positive rate" I think about N+/total number of test (while here it is over the N of positives or the N of negatives with the reference methodcolumns 6 and 7). I think rephrasing the captions for columns 6 and 7 and/or including a note with the used formulas will greatly increase the readability of the table (e.g. something like "Luminex sensitivity Abstract, results: Similarly, the sentence "viral pathogens were detected in a total count of 270/294 by the Luminex among reference assays third paragraph: I would rephrase the sentence "…corresponding to a 1.8% detection rate Methods: Was the used VTM commercially available or prepared in house? A reference to the provider or a list of component used for its preparation should be included Shouldn't it be <=15, instead of >=15? Was the higher number of co-infections detected by the Luminex due to the higher number of Entero/Rhinoviruses identified? For individual targets": I think a word is missing after "more often Is the work clearly and accurately presented and does it cite the current literature? Is the work clearly and accurately presented and does it cite the current literature? This work was done as part of the Vietnam initiative on Zoonotic infections study (VIZIONS).We would like to thank the study teams in VIZIONS consortium members at the all collaborators at Dong Thap General Hospital, Are all the source data underlying the results available to ensure full reproducibility? Yes No competing interests were disclosed. I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Was any other reference method used to assess whether additional positive by Luminex were false positive or true positive to truly evaluate the performance of luminex / label it as more sensitive Are all the source data underlying the results available to ensure full reproducibility?