Dataset: 11.1K articles from the COVID-19 Open Research Dataset (PMC Open Access subset)
All articles are made available under a Creative Commons or similar license. Specific licensing information for individual articles can be found in the PMC source and CORD-19 metadata.
More datasets: Wikipedia | CORD-19
Deep Learning Technology: Sebastian Arnold, Betty van Aken, Paul Grundmann, Felix A. Gers and Alexander Löser. Learning Contextualized Document Representations for Healthcare Answer Retrieval. The Web Conference 2020 (WWW'20)
Funded by The Federal Ministry for Economic Affairs and Energy; Grant: 01MD19013D, Smart-MD Project, Digital Technologies
There is currently no specific approved treatment for HBoV infection. Symptomatic treatment may be required in severe cases and is analogous to the treatment of other respiratory tract infections. To date, there has been only a single case report in which antiviral treatment was associated with the elimination of HBoV in a patient co-infected with HBoV and HHV-6. The boy suffered from an immunodeficiency and had lost the ability to mount an antibody-based immune response. During treatment with cidofovir, which is specific for herpesviruses, HHV-6 viremia decreased, and the HBoV infection was successfully eliminated. The outcome of this case is consistent with the outcome of another clinical case that we observed, in which HBoV viremia decreased during treatment, although this case had a fatal outcome for other reasons.
It must also be mentioned that, due to the lack of an animal model and a versatile cell culture system, the development of a vaccine is extremely difficult, and therefore, no results of such efforts have been published. The prevention methods for HBoV infection are analogous to those of other respiratory virus infections, and for laboratory research, this virus is “treated” in the same manner as other parvoviruses. The German National Committee for Biologic Safety (Zentralkommission für Biologische Sicherheit, ZKBS) has consequently classified this virus as a biosafety level 2 agent.
From December 2014 to May 2015, adenoid tissue samples were obtained from the 70 patients consecutively admitted for adenoidectomy at the Bonn University Medical Centre, Department of Otorhinolaryngology. Ear, nose and throat specialists determined the indication for surgery. From 45 of these patients, a throat swab was taken just before the surgical procedure. All patients had clinical symptoms caused by hypertrophy of adenoids. At the time of surgery and the 2 weeks before, no children displayed symptoms of acute upper or lower airway infection.
Due to the lack of a cell culture system, the diagnosis of HBoV infection is exclusively based on molecular detection methods. Most laboratories currently use in-house PCR and real-time PCR assays targeting the NP-1, NS-1 or VP1/2 gene, but other nucleic acid-based detection methods for the diagnosis of HBoV have been described. A number of commercially available approved multiplexing assays have been developed and brought to the market. Some of these assays also detect human bocavirus, including the Luminex RVP assay (Luminex, USA) and the RespiFinder assay (Pathofinder, the Netherlands). It appears likely that other assays to detect HBoV will be developed, although in many currently available assays, such as the FilmArray (Idaho Diagnostics, USA) and the RespID assay (Luminex, USA), this pathogen is neglected. Of note, for those laboratories that follow FDA rules, it is important to keep in mind that the Luminex RVP xTAG fast assay has received FDA clearance, whereas the other assays with FDA approval do not detect human bocavirus.
To correctly diagnose an HBoV infection, it is necessary to screen clinical samples from the respiratory tract or stool samples (depending on whether the primary symptoms are respiratory or gastrointestinal) and a corresponding serum sample. The latter is of great importance, as viremia is observed only during active infection, whereas HBoV can be shed by otherwise healthy patients, most likely due to persistent infection without viremia.
This study reported a high prevalence of respiratory viruses in children ≤ 5 years using a custom assay and an FTD assay. Good concordance was observed for all the viruses between both assays except for RSVA/B. However larger numbers of positive samples need to be tested for thorough evaluation of less prevalent viruses. The custom primer and probe mix was much more economical than the commercial FTD kit. Our study suggests that this custom multiplex real-time RT-PCR can be used for simultaneous and rapid detection of multiple viruses in resource limited settings. This will help prevent unnecessary use of antibiotics and permit timely initiation of supportive therapy/antiviral drugs if available.
As HBoV has been first identified in respiratory samples, it has been suggested as a respiratory tract infection agent. The majority of the following studies in fact detected HBoV in children with respiratory tract infections. Clinical symptoms mostly described in conjunction with an HBoV infection are wheezing, fever, bronchiolitis and pneumonia. Studies including asymptomatic controls showed that HBoV is also detectable in these controls but with a lower incidence. For example, HBoV was detected in 17 % of children hospitalized because of respiratory infection, while only 5 % of the surveyed asymptomatic children were HBoV positive. This supports the assumption that HBoV in fact could be assigned to the respiratory viruses.
In contrast to other studies, in the study of Longtin et al. 43 % of asymptomatic children tested positive for HBoV. Most of those children underwent myringotomies, adenoidectomies or tonsillectomies. Thus, Lu et al. suggested that HBoV may be present in tonsillar lymphocytes. They tested DNA extracts of lymphocytes from nasopharyngeal tonsils or adenoids and palatine/lingual tonsils. 32.3 % of the tested extracts were HBoV positive, indicating that HBoV establishes latent or persistent infection.
Coinfections with other viruses are frequently observed in HBoV infections and often occur in more than 50 % of the tested samples. Two recent studies report that the viral load of HBoV was significantly higher in children with monoinfections than in children with coinfections. The high rate of coinfections with other viruses may then be explained by the persistence of HBoV in the respiratory tract. DNA quantification in HBoV positive samples revealed that the viral load of 42.5 % of the positive patients was > 1.0 × 105 DNA copies/mL, suggesting that below this cut-off HBoV may be a persistant virus or a bystander.
Extirpated adenoids were picked up in the surgical room and transported on ice to the laboratory for immediate preparation. For nucleic acid preparation, approximately 25 mg of adenoid tissue was crushed mechanically with a scalpel followed by incubation with 600 μL RLT buffer (Qiagen Hilden, Germany) and 1% β-mercaptoethanol (Sigma-Aldrich/Merck, Munich, Germany). The lysate was homogenized by using QIAshredder homogenizer spin columns (Qiagen) according to manufacturer´s instructions. After addition of 1 volume 70% ethanol to the homogenized lysate, RNA was extracted from the sample with the RNeasy Mini Kit (Qiagen) according to the manufacturer´s protocol. All precautions to avoid contamination were strictly adhered to.
SAFV RNA was detected by real-time reverse transcription (RT-)PCR using the primers and probe as previously described, with sequences as follows: CF723: TGTAGCGACCTCACAGTAGCA; CR888: CAGGACATTCTTGGCTTCTCTA; CP797: FAM-AGATCCACTGCTGTGAGCGGTGCAA-BHQ1. RT-PCR was performed in a volume of 25 μL containing 5 μL RNA preparation (approximately 1.25 mg) and by using SuperScriptIII One-Step RT-PCR System with Platinum Taq DNA Polymerase (Invitrogen/ThermoFisher Scientific, Schwerte, Germany) and 1 μg bovine serum albumin (VWR International, Langenfeld, Germany). RT and cycling conditions were 52°C for 20 min, denaturation at 94°C for 3 min, followed by 45 PCR cycles, each consisting of 95°C for 15 sec and 58°C for 30 sec. The PCR amplified a 187-bp fragment of the SAFV genome within the 5´ untranslated region (5´UTR). The limit of detection 95% (LOD95) was 9 copies per reaction.
Samples testing positive by RT-PCR were subjected to nested RT-PCR for amplifying a larger genomic stretch, with an inner fragment of approximately 592 bp within the 5´ untranslated region (nucleotide positions [nts] 204–795, according to GenBank number EF165067; without primers, nts 224–775; please note the small SAFV strain-specific differences in fragment length) followed by nucleotide sequencing. SuperScript III One-Step RT-PCR System with Platinum Taq DNA Polymerase and 1 μg BSA was used for first round of amplification and Invitrogen Platinum Taq DNA Polymerase was used for second round amplification. Reaction volume was 25 and 50 μL in the first and second round, respectively, each with 5 μL target. RT and amplification primers were as follows: first round, Cardio-Universal-F1, 5´-GCTAATCAGAGGAAAGTCAGCATT-3´; Cardio-Universal-R1, 5´- GACCACTTGGTTTGGAGAAGCT-3´; second round, Cardio-Universal-F2, 5´-CAGCATTTTCCGGCCCAGGC-3´, Cardio-Universal-R2, 5´-ATCCACGGGGCTTTTGGCCG-3´. RT and cycling conditions were as follows: RT and first round, 48°C for 30 min, 95°C for 5 min, 35 cycles each consisting of 95°C for 1 min, 50°C for 1 min, 72°C for 1 min, followed by 72°C for 5 min; nested PCR, 95°C for 3 min, 35 cycles (94°C for 1 min, 60°C for 1 min, 72°C for 1 min), 72°C for 5 min. Amplified products were visualized on agarose gels and were subjected to DNA cycle sequencing using BigDye Terminator technology (3130XL Genetic Analyzer, Applied Biosystems, Foster City, USA). Sequencing was done in both directions. Sequences were manually reviewed and compared with genome sequences in GenBank.
Viral RNA was quantified by use of an in vitro transcript of a plasmid-based standard (pCR4.0 TOPO-TA vector, ThermoFisher) derived from an 803-bp PCR amplicon encompassing the screening real-time PCR target region.
Throat swabs were taken by flocked swabs (Copan) and were dissolved in 500 μL phosphate-buffered saline. Viral nucleic acid was prepared by use of the QIAamp Viral RNA Mini Kit (Qiagen) and eluted in 100 μL. Testing for SAFV RNA was performed by real-time RT-PCR as mentioned above.
Testing for typical respiratory viruses was performed by RT-PCR as described previously. Tested viruses were Influenza A and B viruses, Human parainfluenza viruses 1–4 (now termed Human respiroviruses 1 and 3, Human rubulaviruses 2 and 4), Human Rhinovirus, Human respiratory syncytial virus, Human metapneumovirus, Enterovirus, Human parechovirus and Human coronaviruses 229E, NL63, HKU-1, and OC43, and Human adenovirus.
To strengthen our findings, we subsequently tested our SAFV-positive tissues and swabs for the non-respiratory viruses Norovirus and Zika virus by use of the RealStar Norovirus RT-PCR Kit 1.0 (Altona Diagnostics) and the RealStar Zika Virus RT-PCR Kit 3.0 (Altona Diagnostics) according to the manufacturer´s protocols.
This study was approved by Nepal Health Research Council (reg. no. 180/2015).
The tracheal aspirate specimens were first treated with an equal volume of Sputasol solution (Oxoid, Basingstok, UK) in a 37 °C water bath for liquefaction. The total RNA/DNA was then extracted from the liquefaction sample using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) per the manufacturer's instructions. The nucleic acids was stored in aliquots at −80 °C until use.
It remains unclear how far HBoV contributes to respiratory and/or gastrointestinal disease. More and more evidence supports the assumption that HBoV is indeed an infectious and contagious agent, but a chance remains that it solely synergistically increases the clinical severity of other infections. Consequently, well planned and designed clinical studies with sophisticated case controls need to be performed in order to finally rule out the role of bocavirus, unless animal or at least in vitro models demonstrate its pathogenicity.
Eighty percent of influenza virus infected patients were treated with oseltamivir (n = 247), while the other 20% of these patients were not prescribed oseltamivir (n = 65). They were less likely to be treated with oseltamivir if diabetic, otherwise similar groups (Table 3). Oseltamivir-treated patients had no significant difference in mortality than the patients not treated with oseltamivir did; the 20-day all-cause mortality rates were 6.4% and 7.6%, respectively (P = 0.72, Fig. 4).
A total of 356 nasopharyngeal aspirate and throat swab samples were collected from patients with SARI by a trained technician using a sterile nylon flocked swab and placed in viral transport medium (VTM), labelled and transported on ice at the earliest to Advanced research lab (ICMR Grade-1 Virology Lab) of SMS Medical college Jaipur for further processing and storage of the samples. The study was approved by the institutional ethics committee.
HBoV-IgM antibodies were determined by a commercially available immunoglobulin M (IgM) enzyme-linked immunosorbent assay supplied by (Dako, Glostrup, Denmark) for the quantitative determination of IgM antibodies to HBoV in serum. Briefly: serum samples diluted 1:200 in phosphate buffer saline (PBS) and 0.05% Tween (PBST) were applied in duplicate into wells of plates coated with goat anti-human IgM for 60 min at room temperature. After being rinsed 5 times with PBST, biotinylated HBoV viral like particles (VLPs) were applied at a concentration of 25 ng/well and incubated for 45 min at 37°C. Bound antigen was visualized by using horseradish peroxidase-conjugated streptavidin at 1:12,000 in PBST plus 0.5% bovine serum albumin for 45 min at 37°C, followed by o-phenylenediamine dihydrochloride and H2O2 for 15 min at 37°C. The reaction was stopped after 10 min with 0.5 M H2SO4, and the absorbance at 492 nm were recorded. Cut off absorbance for negative and positive IgM ELISA results were 0.136 and 0.167, respectively.
Nasopharyngeal aspirates were collected from patients and control group according to Svensson et al.,. Sterile normal saline solution was instilled in one nostril while occluding the other nostril, using a sterile blunt-tipped disposable syringe. Then the patient was instructed to forcibly exhale through the lavaged side into a sterile specimen cup. The sequence was then repeated in the other side of the nose. NPA specimens were examined microbiologically immediately and part of the specimens was stored in aliquots at -70°C for PCR. Two ml blood samples from both patients and control were collected into vacutainer, centrifuged and serum was separated and stored at - 20°C for HBoV-IgM antibodies by ELISA.
Taqman primers and probes listed in Table 3, were designed in-house using Primer Express software (Applied Biosystems, USA), with the exception of those for the influenza virus A matrix gene. The letter in the final column of the table indicates which of the enrichment PCR mixes contained the primers, and the number indicates which of the real-time PCR mixes contained the primers and probes. At present, not all of the real-time PCR mixes contain uniquely specific fluorophore-labelled probes, so further confirmation of some positive results is required. For example, in the parainfluenza mixes the types 1 and 2 probes both have FAM labels and the type 4A and 4B probes both have VIC labels, so individual real-time PCRs may be performed from the original enrichment PCR material if further typing is required. The mixes were prepared in large volumes, stored in single use aliquots at −80°C, and batches were quality tested prior to use.
Bacterial and viral RNA and DNA was extracted from respiratory specimens including nose swabs, throat swabs, pernasal aspirates and sputum samples, using a modified liquid sample protocol with the X-tractor Gene instrument (Corbett Life Science, Australia). Standardized doses of equine herpesvirus type 4 (EHV4) and MS2 RNA coliphage (MS2) were added to the lysis buffer supplied with the kit to monitor the efficiency of sample extraction, removal of reverse transcription and PCR inhibitors, and cDNA production [42, 43]. The CT values expected for the EHV4 and MS2 assays were between 15 and 20, and samples with CT values greater than 27 were retested.
In this study, the results showed that NxTAG RPP, xTAG RVP, and FilmArray RP assays had high sensitivity and specificity for diagnosis of respiratory diseases. They are suitable for use in clinical diagnostic laboratories for detection of respiratory pathogens in patients with RTIs. However, awareness should be raised that H3N2 could not be distinguished from H3N2v by NxTAG RPP. Further investigation should be performed if H3N2v is suspected to be the cause of disease.
Bocavirus is a single-stranded DNA virus belonging to the family Parvoviridae, subfamily Parvovirinae, genus Bocavirus. Bocaviruses are unique among parvoviruses because they contain a third ORF between the non-structural and structural coding regions–. The genus bocavirus includes viruses that infect bovine, canine, feline, porcine and some simian species as well as sea lions–. Human bocavirus (HBoV) was first found in children with acute respiratory tract infections in 2005. It was then detected in children with respiratory tract infections in addition to gastroenteritis worldwide–. The virus exists in four different serotypes HBoV1-4–,–. Although HBoV 1 and 2 were reported in respiratory samples, all the 4 genotypes of HBoV have been identified in children with acute gastroenteritis.
HBoV has been reported in various countries, indicating its worldwide endemic nature. The virus has been identified in Europe–, America–, Asia,, Australia–, Africa, and the Middle East. The prevalence of HBoV ranges between 1.5 to 19.3%,. Primary infection with HBoV seems to occur early in life and children between the ages of 6–24 months seem to be mostly affected–, but older children can also be infected. Newborn children may become protected by maternally derived antibodies. HBoV infections are rarely found in adults–. Lindner et al. detected anti-HBoV antibodies in 94% of healthy blood donors >19 years of age.
HBoV detection has been mostly performed on nasopharyngeal aspirates and swabs and relies mostly on classical,,,,, and real-time PCR,,. Real-time PCR possesses many advantages over conventional PCR, as it offers greater sensitivity, specificity, and reduced expenditure of time.
The current study aims to screen the epidemiological status and molecular phylogeny of HBoV isolates prevailing in pediatric patients with respiratory infection in Saudi Arabia.
A 17-month-old Latvian boy was admitted to the Children’s Clinical University Hospital of Riga, Latvia, on the seventh day of illness in January 2015. He presented with a history of rhinorrhea and cough for 6 days and fever (axillar temperature 39.0 °C) for the last 2 days prior to admission. Due to severe respiratory distress, he was immediately transferred from the regional hospital to our intensive care unit.
On admission, his respiratory rate was 44 breaths/minute (reference 20–30), heart rate 146 beats/minute (reference 80–130), oxygen saturation 99% (with an oxygen flow of 5 liters/minute via face mask), and axillary temperature 38.7 °C. Auscultation of his lungs revealed bilateral wheezing and crepitation with severe intercostal and subcostal recessions. The other organ systems were without pathology. Due to the severe respiratory distress, tracheal intubation was performed.
The child had been born full term as the seventh in the family. He had no known underlying illness, history of previous hospitalizations, or severe acute illnesses. He had been fully immunized according to the national immunization scheme.
On admission, his white blood cell (WBC) count was 30.6 × 103/μL with 66.9% of granulocytes (in absolute numbers 20.6 × 103/μL), hemoglobin 12.4 g/dL, and platelet count 321 × 103/μL. His C-reactive protein (CRP) was 5.09 mg/L. A chest radiograph showed infiltration of the upper lobe of his right lung (Fig. 1).
At the time of admission, a nasopharyngeal swab (NPS) tested negative by direct immunofluorescence (IMAGEN™ OXOID, UK) for antigens of RSV, influenza virus A and B, parainfluenza virus types 1–3, and adenovirus. Bacterial blood cultures were negative. NPA, blood, and stool samples were collected for HBoV1 molecular diagnostics and serology.
NPA tested by qualitative multiplex PCR (Seegene Respiratory Panel, South Korea) was negative for: influenza virus A and B; RSV A and B; flu A types H1, H1pdm09, and H3; adenovirus; enterovirus; parainfluenza virus types 1–4; metapneumovirus; rhinovirus; and coronavirus types NL63, 229E, and OC43. However, the NPA tested by qualitative multiplex PCR was positive for HBoV1. NPA, whole blood with corresponding cell-free blood plasma, and stool samples underwent qualitative PCR for HBoV1 NS1 DNA, as described. An HBoV1-containing plasmid was used as a positive control in PCR. All these samples were HBoV1 DNA positive. Upon re-examination by quantitative PCR (qPCR) (Human bocavirus genomes, Standard kit, Genesig, Primerdesign Ltd., UK), the copy numbers in NPA and stool were high, 5.7 × 105 per μg DNA in NPA and 1.4 × 108 per μg DNA in stool. The viral load in blood was 21 copies/μg DNA, but in cell-free blood plasma the viral load was under detection level.
To prove that the HBoV1 infection was actively ongoing, HBoV1 transcription in PBMCs was applied. Total ribonucleic acid (RNA) was extracted from PBMCs using TRI Reagent® solution according to the manufacturer’s instructions (Thermo Fisher Scientific, USA). The extracted RNA was quantified spectrophotometrically and analyzed by electrophoresis in a 1% agarose gel. RNA was treated with DNase (TURBO DNA-free™ Kit, Thermo Fisher Scientific, USA) before the synthesis of complementary DNA (cDNA) by the reverse transcriptase (RT) using RevertAid™ First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, USA). The β-actin gene sequence was detected by PCR to assess the quality of synthesized cDNA (Fig. 2).
HBoV1-specific cDNA was detected by PCR targeting the HBoV1 NS1 gene as described by Sloots et al., in 2006, followed by electrophoretic visualization of the amplification products in a 1.7% agarose gel (Fig. 3). The same DNase-treated RNA sample but without the RT step, served as a negative control in both the β-globin and HBoV1 PCRs to make sure that there was no contamination with DNA.
Biotinylated virus-like particles (VLPs) of the recombinant major capsid protein VP3 were used as antigen in enzyme immunoassays (EIAs) for detection of HBoV1-specific immunoglobulin M (IgM) and immunoglobulin G (IgG) in our patient’s plasma sample [23, 24]. For removal of possible cross-reacting heterologous human bocavirus 2 (HBoV2) and human bocavirus 3 (HBoV3) IgG, non-biotinylated VLPs in competition assays were used as described. Our patient’s plasma sample was positive for both HBoV1-specific IgM and IgG antibodies.
Because of the right lung upper lobe infiltration and increased WBC initially, the child was treated with intravenously administered ceftriaxone 350 mg twice a day for 7 days and per-oral oseltamivir 30 mg twice a day (due to influenza season). Oseltamivir was discontinued after 3 days due to the negative influenza virus A and B antigen findings. Extubated on day 3, our patient was brought to the Department of Paediatrics, where intravenously administered ceftriaxone was continued, inhalations via nebulizer with salbutamol and budesonide were begun and pulmonary rehabilitation started. During the next 10 days, the child’s general condition improved, his body temperature was normal, lung sounds were without the pathology, and no additional oxygen was needed. During the hospitalization, poor weight gain was observed for our patient; therefore, additional diagnostic tests were done and his hospitalization length increased. On day 17 of hospitalization, he developed a new episode of fever for 2 days. The second NPS tested negative for RSV, and influenza virus A and B; however, self-limiting viral upper respiratory tract infection was suspected and he was treated with intravenously administered rehydration and ibuprofen 70 mg for these 2 days. Due to the very low socioeconomic status of the family, he was kept in the hospital mainly for observation, although his general condition was good. On day 30 he developed a new episode of fever, cough, and wheezing lasting 6 days. In this episode, LRTI was diagnosed based on the clinical symptoms and he was treated with nebulized salbutamol and budesonide.
After 46 days of hospitalization he recovered completely from HBoV1-associated acute bilateral bronchiolitis with right-side pneumonia and a subsequent hospital-acquired upper and LRTI and was discharged.
We found that viral infections were often neglected during this population-based “real life” study of suspected septic patients. The study was performed during the winter period, referred to as “the flu season”, when respiratory viral infections are most prevalent. This should have resulted in increased clinical suspicion. Yet, during the first four weeks of the influenza epidemic, very few clinical samples are requested. In this material, a viral respiratory infection was initially suspected by clinicians in only 30% of patients with viral findings by multiplex PCR. This was especially true when CRP was over 100 mg/L, or if there was a new infiltrate on the chest X-ray indicating pneumonia. This underestimation may lead to nosocomial spread or outbreaks of viral respiratory infections, as we have previously experienced in our own hospital. It may also lead to overuse of antibiotics, as well as underuse of antivirals, especially in risk groups that might benefit from such treatment.
As in a comprehensive study on bacterial–viral respiratory tract illness over three winter seasons by Falsey et al. in 2013, influenza A virus was the most common viral finding, appearing in study samples almost two weeks earlier than in clinical samples. In only 35 of 96 cases of influenza A virus infection (36%) was influenza virus initially suspected as sole cause or contributing factor to the acute illness.
Respiratory syncytial virus and human metapneumovirus may cause critical respiratory illness and pneumonia, not only in children, but also in elderly. For example, human metapneumovirus was found to be the causative agent in an outbreak of pneumonia among elderly at an institution in the Netherlands. In this study, human metapneumovirus was a slightly more common finding than respiratory syncytial virus, especially in patients with long history of fever and respiratory tract congestion, combined with radiological signs of pneumonia.
Nasopharyngeal culture is generally discouraged or not recommended for etiological diagnosis of pneumonia. However, in a Swedish study by Strålin et al. in 2006, there was a good correlation between nasopharyngeal findings of these bacteria and the etiology of pneumonia, as has been seen in previous Swedish studies. In our study of patients in the emergency department suspected to be septic, there was a strong correlation between nasopharyngeal findings of S. pneumoniae or H. influenzae and X-ray findings of a new infiltrate, indicative of pneumonia. More so, these bacteria were not found in the nasopharynx of any of the 210 patients with a non-respiratory infection or no infection, and they were rarely found in patients with a respiratory tract infection but not pneumonia. The study results imply that nasopharyngeal findings of S. pneumoniae and H. influenzae in sepsis patients should be considered carefully for patient treatment. A recently published paper by Bjarnason et al. in 2017 demonstrates a good correlation between real-time PCR findings of S. pneumoniae or H. influenzae to pneumonia diagnosis in adults, which also builds support for a clinical relevance of these upper respiratory bacterial findings.
Co-infections of bacteria and respiratory viruses, mainly S. pneumoniae and influenza A or respiratory syncytial virus, are found in 3–40% of patients with CAP, depending on diagnostic methods used, with the higher end reflecting studies in which nasopharyngeal culture is included for etiological diagnosis [1, 2]. Using nasopharyngeal sampling only, we found indications of viral-bacterial co-infections in 28 of 137 (20%), a proportion we believe to be an underestimation. As in the study by Falsey et al. in 2013, S. pneumoniae was the bacteria most often associated with pneumonia and a viral co-infection. As many as 75% of patients with pneumonia and S. pneumoniae in the nasopharynx were positive for a respiratory virus, mainly influenza A virus, but also human metapneumovirus. The two youngest patients, with pneumonia and severe sepsis, aged 37 and 42 years respectively, were both positive for S. pneumoniae and human metapneumovirus in the nasopharynx. No other pathogens could be demonstrated by routine cultures.
In the clinical setting it is often difficult to determine whether a patient with respiratory symptoms has a viral infection, a bacterial infection, or a mixed viral-bacterial infection. No constellation of clinical symptoms, vital signs, biomarkers (such as white blood cell count, C-reactive protein, or procalcitonin) have adequate sensitivity and specificity. New tools to improve predictions of patient benefit from antibiotic treatment are urgently needed. Recently, whole blood analysis for the identification of host gene activation profiles has been able to discriminate viral infections from bacterial infections with high accuracy in severely ill infants, as described by Herberg et al. in 2016. In adults with lower respiratory tract infections, a similar technique seems able to discriminate viral from bacterial infections much better than procalcitonin, as shown by Suarez et al. in 2015. In the same study mixed viral-bacterial infections also elicited a characteristic gene activation profile.
Our study supports increased testing for respiratory viruses in patients believed to be septic, especially those presenting with respiratory tract symptoms. With current technology, results can be obtained within a few hours and have an impact on clinical decisions and patient logistics in the emergency department. Cost effectiveness should be investigated. A viral diagnosis may not only lead to fewer admissions and less antibiotic treatment if bacterial pneumonia is suspected or demonstrated, but may also decrease viral exposures for admitted patients. Patients with influenza A or B may benefit from antiviral treatment alone or in conjunction with antibacterial treatment, if bacterial pneumonia is suspected or demonstrated, perhaps even reducing viral contagiousness. Even in neutropenic patients, a viral finding and a favorable outcome in the first few days may safely allow for discontinuation of antibiotic treatment.
This study has several limitations. It is a single-center study performed during one winter period only. The study was primarily not an etiological study of respiratory tract infections. We did not take ongoing antibiotic treatment into account. Only routine sampling from the nasopharynx was performed, albeit using a flocked swab for better yield. If sputum or nasopharyngeal aspirates had been analyzed with molecular techniques, we would have expected a higher yield for both bacteria and viruses [2, 30]. A further weakness is the open inclusion based on clinical suspicion of sepsis only, without specific criteria. Yet another was the subjectivity involved in deciding the relevance of the findings. What role do the viral findings play, both alone or in conjunction with bacterial findings? We can only show correlations between findings and clinical entities, yet the majority of findings do seem to correlate well to respiratory tract infections. Therefore, we believe that our conclusion, that significant viral disease in severely ill patients is underdiagnosed by clinicians, is warranted. Diagnosing these infections early may be of help for the clinical decision making process and thereby for the patients.
The study protocol was approved by the medical ethics review board of the College of Medicine, Taif University and by the pediatric hospital ethics committee in accordance with the guidelines for the protection of human subjects. Informed written consents from the next of kin of the participants involved in the study were taken.
Detection of respiratory viruses, including HPIV 1–4, HRSV, HMPV, HCoV-OC43, HCoV-229E, HCoVNL63, HCoV-HKU1, HRV, HAdV, and HBoV was performed using a RT-PCR Kit «AmpliSens ARVI-screen-FL» (Interlabservice, Russia), and IFVA and IFVB virus detection was performed using a RT-PCR Kit « AmpliSens Influenza virus A/B-FL» (Interlabservice, Russia) according to the manufacturer's instructions. Positive and negative controls were included in each run.
FL planned the study, collected clinical samples in conformation with all the human subject consents and participated in work of ELISA, DNA extraction and Q-PCR performance. AZ and HL performed ELISA, extracted DNA and Q-PCR. NY, EY and SW developed and standardized the Q-PCR method. DP and JQ coordinated the study, constructed plasmid, designed primers and probe, initiated Q-PCR and wrote the manuscript.
With regards to complications, pleural effusion developed in 12.5% of patients without viral identification and 19% of patients with viral identification; closed thoracotomy was required in 4.2% of patients without viral identification and 3.2% of patients with viral identification; inpatient infection was present in 8.3% and 3.2% of patients without and with viral identification, respectively; and underlying disease complications were found in 25% of patients without viral identification and 38.1% of patients with viral identification. These differences were not statistically significant. Patients with chronic pneumopathy showed more complications in their underlying pathologies (77.3% vs. 20.3%, OR: 13.5; CI 95%: 3.5–52.7).
A total of 14 (15.4%) patients died. However, two cases included either an inadequate sample for PCR or no PCR amplification; thus, the outcome analysis was performed for 12 patients. Mortality was reported for 4 cases (16.7%) in the group without viral identification and 8 cases (12.7%) in the group with viral identification (P = 0.23, OR: 0.72; 95% CI: 0.17–3.68) (Table 4). Mechanical ventilation was required for 33 patients (36.3%); this intervention was invasive in 24 (26.4%) patients, non-invasive in 13 (14.3%) patients and invasive and non-invasive in 4 (3.6%) patients. This outcome occurred among 7 patients (29.2%) without viral identification and 23 patients (36.5%) with viral identification (P = 0.41, OR: 1.4; 95% CI: 0.46–4.59). The average length of mechanical ventilation was 7 days, with a minimum of 1 day and a maximum of 20 days. Thirty-nine (42.9%) patients were admitted to the intensive care unit (ICU); 3 had unsuitable samples or non-amplified PCR results, 9 (37.5%) were from the group without an isolated virus, and 27 (42.9%) were from the group with an isolated virus (P = 0.21, OR 1.25; 95% CI: 0.43–3.75). One case with viral identification was readmitted to the unit with an identification of Influenza subtype AH1N1. The average length of ICU stay was 8.4 days, with a range between 1 and 21 days. The group with an isolated virus had an average ICU stay of 3.8 days, with a range between 1 and 21 days, and the group without an isolated virus had an average stay of 2.8 days, with a range between 1 and 13 days (P > 0.05). The average hospital stay in the SARI group was 9.9 days, with a range between 1 and 45 days, with average stays of 10.6 and 9.8 days for patients with and without a virus, respectively (P = 0.6). Among patients without a virus, the hospital stay varied between 1 and 45 days, while patients with a virus had hospital stays that varied between 1 and 28 days.
Of the 8 patients from the viral infection group who died, Bocavirus was isolated in 5, Influenza was isolated in 4 (all cases of Influenza A), Metapneumovirus was isolated in 3, and RSV was isolated in 1. Of the patients with Bocavirus isolation, 19.2% died. Death occurred in 14.3% of those in whom Influenza was identified, 50% of those in whom Metapneumovirus was identified and 20% of those in whom RSV was identified. Among patients with viral identification who died, 5 (62.5%) had viral infections of 2 or more viruses (P = 1.7, OR: 2.4; 95% CI: 0.46–18.89), and fewer had only one virus detected. No statistically significant differences were observed in the outcomes of the remaining patients with mixed viral infection, which does not confirm higher morbidity in patients with more viruses isolated.
Two sets of blood cultures were drawn before initiation of intravenous antibiotic treatment in all 432 patients. Of those, 35 patients (8%) were positive for a true bacterial pathogen. The bacterial species were S. aureus (n = 12), E. coli (n = 8), S. pneumoniae (n = 5), Klebsiella pneumoniae (n = 4), S. pyogenes (n = 2), and others species in four patients. In the overall nine-month epidemiological study, 313 of 2,472 (12.6%) blood cultures were positive for a true pathogen.