key: cord-0297688-t08zpt7f authors: Steppert, C.; Steppert, I.; Bollinger, T.; Sterlacci, W. title: Rapid non- invasive detection of Influenza- A- infection by multicapillary column coupled ion mobility spectrometry date: 2020-06-05 journal: nan DOI: 10.1101/2020.06.04.20099259 sha: ccb17d87aeddad5c904f89280ef276a499f24d79 doc_id: 297688 cord_uid: t08zpt7f Infectious pathogens are a global issue. Global air travelling offers an easy and fast opportunity not only for people but also for infectious diseases to spread around the world within a few days. Therefore a rapid on-site screening of infected people is urgently needed. Due to the small size and easy handling, the ion mobility spectrometry coupled with a multicapillary column (MCC-IMS) is a very promising, sensitive method for the on-site identification of infectious pathogens based on scents, volatile organic compounds (VOCs). The purpose of this study was to prospectively assess whether differentiation between patients suffering from common cold and those from Influenza-A-infection based on VOCs with MCC-IMS is possible. Nasal breath was investigated in 24 consecutive patients suspected of having Influenza-A-infection by MCC-IMS. In 14 Influenza-A-infected patients, infection was proven by PCR of nasopharyngeal swabs. Ten patients with negative PCR diagnosis served as controls. For picking up relevant VOCs in MCC-IMS spectra, software based on cluster analysis followed by multivariate statistical analysis was applied. With only four VOCs canonical discriminant analysis was able to distinguish Influenza-A-infected patients from PCR- negative patients with 100% sensitivity and 100% specificity. This present proof-of-concept- study yields encouraging results showing a rapid diagnosis of viral infections in nasal breath within 5 minutes by MCC-IMS. The next step is to validate the results with a greater number of patients with Influenza-A-infection as well as other viral diseases. Registration number at ClinicalTrials.gov NCT04282135. Air travel not only leads to global connection of people and goods but also to an easy spread of infectious diseases [1] . During the SARS epidemic non-contact temperature measurements were used to recognize infected passengers before boarding but even for the detection of fever there is a lack of specificity and sensitivity [2] . A Chinese colleague who became symptomatic on her flight back to China infected the first German COVID-19 patient. So a rapid detection of communicable viral diseases is urgently needed not only to avoid spread by air travel but also to detect and cut infection chains. Ancient Greek physicians already used their sense of smell to detect diseases, especially infectious diseases [3] . These scents are volatile organic compounds (VOC) released during the metabolism of the bacteria or the host. Different analytic procedures including multicapillary column coupled ion mobility spectrometry (MCC-IMS) can be used to differentiate between bacterial species in vitro [4, 5] . A detection of bacterial infection could also be achieved in the breath of infected animals as well as in tuberculosis-infected humans [6, 7] . Unlike bacteria, viruses do not have a metabolism of their own but VOCs can also be detected when produced by the infected host cells [8] . The present study was designed as a proof-of-concept to show whether it is possible to detect Influenza-A-infections by MCC-IMS of exhaled breath. Eligible patients included male of female adults (aged ≥ 18 years) that were admitted to our hospital with suspected Influenza-A-infection. According to hospital policies in every patient a nasopharyngeal swab was taken for the PCR-analysis of virus RNA. Patients were consecutively asked to participate in the study. The study was approved by the ethics committee of the university of Erlangen (Nr 426_18 B) and conducted according to the principles of the Declaration of Helsinki, good clinical practice and the European General data protection regulation. All patients gave written informed consent prior to participation. The study is registered at ClinicalTrials.gov (NCT04282135). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. Influenza-A was tested by taking a deep nasopharyngeal swab applying the "Xpert ® Nasopharyngeal Sample Collection Kit for Viruses" (Cepheid, Maurens-Scopont, France) and performing Influenza-A PCR with the "Xpert ® Xpress Flu/RSV" (Cepheid) on the "Infinity" (Cepheid). As the primary replication site of the Influenza-A virus is the nasopharyngeal mucosa, nasal breath was aspirated for 10 seconds during normal respiration by foam cuffed oxygen catheter For the VOC analysis a MCC-IMS-device from STEP Sensortechnik und Elektronik, Pockau, Germany (STEP IMS NOO) was used. The breath sample was drawn into the sampling loop using an internal pump with a flow rate of 200 milliliter per minute. In the sampling loop the collected sample was pre-separated by isothermally heated multicapillary gas chromatography column (60°C) into single analytes, which finally entered the IMS unit based on their retention times. In the IMS unit the analytes were ionized by beta radiation with a tritium source (99 MBq). Afterwards the generated ions were accelerated in a 50-mm-long driftube under the influence of an electric field (400 V/cm) towards the detector which is also tempered to 60°C. On their drift way they collide with air molecules from the drift gas (400 ml/min) flowing in the opposite direction getting separated depending on their ion mobilities and detected by the collector electrode sampled every 10µs. IMS measurements were performed in positive ionization mode. The received IMS spectra were stored internally in the device and later analyzed offline. The used IMS device is equipped with a circulation filter and internal gas circulation. Using a circulation pump, ambient air filtered by activated carbon was provided as drift gas and analysis gas (20 ml /min) to the device. For the IMS-settings see web-only Supplementary Table S1. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. Depending on these parameters the clusters are numbered. see web-only Supplementary Table S2 Statistical analysis. For the demographics and laboratory findings descriptive statistics were used. Due to the high cross-correlation of the clusters multivariate analysis had to be employed. We decided to perform a stepwise canonical discriminant analysis for the classifier. The discriminant analysis was aimed for a maximal reduction of Wilks-Lambda with maximum significance of F to enter a variable of 0.05 and a minimum significance to remove of 0.1. Cross-validation was performed by the leave-one-out method. For the statistical analysis we used IBM SPSS 26.0 (IBM, Armonk, NY). Between February 15 th 2020 and March 5 th 2020 24 patients with suspected Influenza-Ainfection and PCR confirmed Influenza-A status were included in our study. For details please refer to table 1. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. Fig. S2 ). Using only 3 clusters there was only 1 false positive and 1 false negative (Tab. 2) leading to a sensitivity of 92.9% and a specificity of 90%. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. a. 100,0% of original grouped cases correctly classified. b. Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case. c. 91,7% of cross-validated grouped cases correctly classified. Table 2 Results of linear canonical discrimination analysis using clusters c_521 (164s, 4.41ms), c_378 (119s, 4,56ms) and c_116 (44s, 2,69ms). Neg = Influenza-A negative, pos = Influenza-A positive. Fig. 1 shows the separation of the positive and negative patients in a 3D-plot of the signal intensities of these cluster c_116, c_378 and c_521. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. . https://doi.org/10.1101/2020.06.04.20099259 doi: medRxiv preprint To the best of our knowledge this is the first study indicating that Influenza-A-infections can be detected by aspirating nasal breath in infected patients. There is only little known about breath analysis in viral diseases. Most of the studies either have analyzed headspace over cell cultures [10, 11] or the influence of vaccination on breath VOCs [12] . In another study dogs were used to detect the smell of virus infected cell cultures [13] . There are studies using "breath biopsy" for the detection of cancer by differential ion mobility spectrometry but the device used in this study needs pre-analytic procedures with absorption/desorption tubes while using the STEP device the breath is aspirated during normal breathing and analyzed instantaneously [14] . This can also be performed by less trained personal and is well tolerated by the patients. Another advantage is the speed of the analysis. Though using a retention time of 240 seconds most relevant VOCs could be found in the first 180 seconds. So a reduction of the analysis time to 3 minutes seems to be possible. Compared to PCR as the gold-standard, MCC-IMS is not only faster but does not need reagents and other supplies where shortages of swabs especially flocked nasopharyngeal swabs and of less sensitive swabs have emerged in the SARS-CoV-2 pandemics. Compared to gas chromatography coupled mass spectrometry the device is quite small, easily portable and can be used as a point-of-care diagnostic. PCR-analysis of nasopharyngeal swabs used for the detection of viral disease may be false negative if the replication site is in the distal airways or the lung parenchyma [15] . Using a deep exhalation procedure all respiratory compartments are accessible by breath analysis. One shortcoming of our study is the sample size. Due to the late start of the study and the weak Influenza season 2020 only few patients could be analyzed and no validation cohort was available. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. After demonstrating the proof-of-principle for the MCC-IMS diagnostic in respiratory infections in the current study, future studies shall address the optimization of sampling procedure and analysis algorithms as well as extension to other viral infections. Besides statistical methods we are planning to implement methods of artificial intelligence. To cut infection chains effective screening for Influenza-A and other communicable virus diseases is urgently needed. In this proof-of-concept study MCC-IMS was able to discriminate Influenza-A-infections by simple aspiration of nasal breath during normal breathing. As the whole analysis takes less than 5 minutes MCC-IMS allows for rapid screening. Further investigation of this method is needed as well as application to other viral diseases. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. . https://doi.org/10.1101/2020.06.04.20099259 doi: medRxiv preprint Germs on a Plane: Aircraft, International Travel, and the Global Spread of Disease Non-contact infrared thermometry temperature measurement for screening fever in children The scent of disease: volatile organic compounds of the human body related to disease and disorder Ion mobility spectrometry for microbial volatile organic compounds: a new identification tool for human pathogenic bacteria Rapid in vitro detection of resistant strains by MCC-IMS Chronic intestinal Mycobacteria infection: discrimination via VOC analysis in exhaled breath and headspace of feces using differential ion mobility spectrometry A simple breath test for tuberculosis using ion mobility: A pilot study VOC breath profile in spontaneously breathing awake swine during Influenza-A-infection doi: medRxiv preprint 9. Purkhard R Klassifikation von Ausatemluft anhand ihrer differenziellen Volatile emanations from in vitro airway cells infected with human rhinovirus Cellular Scent of Influenza Virus Infection Effect of influenza vaccination on oxidative stress products in breath Real-Time Detection of a Virus Using Detection Dogs. Front Vet Sci Breath biopsy for early detection and precision medicine in cancer. Ecancermedicalscience Negative Nasopharyngeal and Oropharyngeal Swabs Do Not Rule Out COVID-19 The authors state that there is no conflict of interests The paper has been submitted to medrxiv.org . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. . https://doi.org/10.1101/2020.06.04.20099259 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 5, 2020. . https://doi.org/10.1101/2020.06.04.20099259 doi: medRxiv preprint