Trial of Respiratory Infections in Children for Enhanced Diagnostics

Overview

The overall aim of the TREND study is to improve the differential diagnosis of bacterial and viral etiology in children below 5 years of age with clinical community acquired pneumonia. Specific objectives: * To assess the diagnostic accuracy of MxA for viral CAP (sub-study I) * To study etiologies in children with CAP (sub-study II) * To evaluate sensitivity and specificity for MariPOC® Respi test versus PCR for detection of respiratory viruses (sub-study III) * To assess sensitivity and specificity for a novel RPA-based point-of-care test versus PCR for detection of respiratory viruses (sub-study IV) * To assess long-term complications in children with CAP (sub-study V The study takes place at Sachs' Children and Youth hospital in Stockholm.

Background: Respiratory viral and bacterial infections are hard to distinguish clinically, and many children with viral infections receive unnecessary treatment with antibiotics, which contributes to development and spread of antibiotic resistance. Therefore, there is a need for new point-of-care diagnostic tests that better distinguish viral from antibiotic-requiring bacterial infections, particularly in children presenting with suspected clinical community-acquired pneumonia (CAP), and thus assist health-care workers decision making and improve rational use of antibiotics. Myxovirus resistance protein A (MxA) is a promising biomarker for viral infection, but no studies have investigated MxA in children with CAP. Procalcitonin (PCT) is used as a biomarker for severe bacterial infection as the protein rapidly increases in plasma levels in response to stress and systemic infection. PCT has been reported to be more specific for bacterial infection as compared to CRP, but there are conflicting data on the clinical utility of PCT in children with CAP. The role of viruses and atypical bacteria in childhood CAP Is increasingly recognized. Recent studies have reported an increasing incidence of B. pertussis and there have been several deaths in previously healthy infants associated with whooping cough in Sweden over the last ten years. Consequently, there is a need for new etiologic studies in childhood CAP. Real-time PCR is currently considered gold-standard for detection of respiratory viruses in children with respiratory tract infection. Nevertheless, the turn-around time is usually long and the test results can rarely be used for decision making regarding treatment. There are currently several new antigen-based point-of-care tests of respiratory infections on the market, one is Multianalyte Point-of-care Antigen Detection Test System (MariPOC®) Respi. The sensitivity for respiratory syncytial virus (RSV) and influenza virus is as high as 90% as compared to PCR, the current gold-standard PCR, but, the sensitivity for less common respiratory viruses such as metapneumovirus (hMPV), parainfluenza virus (PIV), coronavirus and bocavirus have been insufficiently investigated. Recombinase polymerase amplification (RPA) is a nucleic acid amplification method that doesn't require thermal cycling. As the test reaction can be carried out at room temperature it is a particularly interesting method for resource-limited settings where the need for new diagnostic tests is high. Studies on long-term outcomes of radiologically confirmed bacterial CAP have indicated that the disease is associated with later development of asthma and decreased lung-function. Given the ongoing change in etiology of pediatric CAP, there is a need for new studies of long-term complications in pediatric CAP. Overall Aim: The overall aim of the TREND study is to improve the differential diagnosis of bacterial and viral etiology in children below 5 years of age with clinical CAP. Specific objectives: * To assess the diagnostic accuracy of MxA for viral CAP (sub-study I) * To study etiologies in children with CAP (sub-study II) * To evaluate sensitivity and specificity for MariPOC® Respi test versus PCR for detection of respiratory viruses (sub-study III) * To assess sensitivity and specificity for a novel RPA-based point-of-care test versus PCR for detection of respiratory viruses (sub-study IV) * To assess long-term complications in children with CAP (sub-study V) Study Design: The TREND study is a hospital-based prospective observational study of children with clinical CAP with asymptomatic controls at the emergency department at Sachs' Children and Youth hospital, Stockholm. Case definition Children 1-59 months with clinical CAP according to WHO-criteria. Bronchodilator challenge: Inhalation with a rapid acting bronchodilator will be administered to children with wheezing and in-drawings to improve the specificity of the WHO clinical CAP criteria as suggested by the PERCH study team. Resolved in-drawings after bronchodilator challenge will be recorded but not considered an exclusion criteria to be able to exclude these patients in a sub-analysis. Control definition: Children 1-59 months at Sachs' Children and Youth Hospital treated for a minor orthopedic or minor surgical disease. The parents to the controls will be contacted via email/telephone 1-2 weeks after enrollment and be asked whether the child has developed respiratory symptoms or not. No matching will be performed but adjustments for age and season will be performed in the analyses. Sampling: A capillary blood sample and nasopharyngeal swabs/aspirates will be collected from all study subjects. Microbiological and Biochemistry Analyses: MariPOC® Respi as well as real-time PCR analysis (detecting: 16 respiratory viruses as well as Streptococcus pneumoniae, Bordetella pertussis, B. parapertussis and Mycoplasma pneumoniae) will be performed. Biochemistry Analyses: Serum MxA, procalcitonin and CRP levels will be analysed. Study Variables: Information regarding the study subject, number of siblings, days of illness, current symptoms, vaccinations, antibiotic treatment, medication, underlying diseases, heredity for asthma, previous hospitalization, recent stay abroad, allergies, smoking, recent travel abroad, recent contact with unwell individual, breastfeeding, preschool, origin of parents and socio-economic status will be collected through a standardized questionnaire based on previous studies. Clinical parameters will be registered by the study doctor responsible for patient screening/enrolment in line with the study protocol of PERCH. Some of the clinical parameters included in the PERCH protocol are very rare in a Swedish context and to avoid overloading of the case report form, these will not be systematically registered at inclusion. However, information about these symptoms will be retrospectively collected from the medical records. Some clinical parameters are routinely recorded multiple times at the emergency unit. In these cases, the most extreme value (highest pulse/respiratory rate/body temperature/etc and lowest peripheral oxygen saturation) during the visit at the emergency unit until enrolment will be recorded. Information regarding admission, length of stay, radiological, routine clinical examination, microbiological and chemistry analyses, treatment, discharge diagnosis and complications will be retrospectively collected from the medical records. All study subjects will be linked to the National Vaccination register to collect information regarding previous immunizations. To allow assessment of long-term complication, study subjects will also be linked to the National Patient Register, the National Death Register and the National Prescribed Drug Register for collection of discharge diagnoses according to ICD-10 as well as prescribed drugs. Classification of Disease: Etiology will be classified as probable or definitive based on clinical significance of the different microbiological test in the studies above. In TREND, the combination of probable and definitive etiology will be used in the main analysis, whereas children with definitive etiology will be assessed separately in a sub-analysis. Definitive Viral Infection: • PCR positive for influenza, RS virus, metapneumovirus or parainfluenza virus Probable Viral Infection: * PCR positive for adenovirus * PCR positive for coronavirus, rhinovirus, bocavirus or enterovirus AND CRP \<20 AND reported fever \>24h. Definitive Bacterial Infection: * positive bacterial blood culture in blood or pleura fluid * positive pneumococcal antigen test in pleura fluid Probable Bacterial Infection: * CRP \>80 (children ≤2 years) / \>120 (children 2-5 years) AND/OR * Radiographic evidence of empyema on X-ray or ultrasound AND/OR * Large dense infiltrate or lobar consolidation on chest x-ray Definitive Atypical Bacterial Infection: • Positive PCR B. pertussis or B. parapertussis Probable Atypical Bacterial Infection: • Positive PCR M. pneumoniae Undetermined: • Cases not fulfilling any criteria above Mixed Viral-bacterial Infection: • Children fulfilling criteria for both viral and bacterial infection Classification of Long-term Complications: Long-term complications (asthma and number of hospital-requiring respiratory infections) will be assessed after 3, 7 and 10 years following study completion by linking to the National Patient Register. Asthma will be classified as ICD-10 diagnosis of J45 or ≥3 prescriptions of inhalation steroids, beta-2-agonists or leukotriene antagonists according to the Prescribed Drug Register. Power Calculation: For the sample size calculation, the investigators focused on the assessment of MxA-levels in cases with viral CAP as compared to cases with bacterial CAP/controls (study I). Two power calculations were made, one for viral CAP versus bacterial CAP and one for viral CAP versus controls. The following assumptions were made: A difference in MxA-level of 500µg/l between the groups was considered clinically relevant. A standard deviation of 1000 and 300 was assumed in cases with viral CAP and bacterial CAP/controls respectively based on previous studies on MxA. Using an alpha-level of 0.05 (two-sided) at an 80% power, with an additional 20% addition to account for non-parametric testing and multivariate analyses, 42 children in each group (viral CAP, bacterial CAP and controls) would be needed. To ensure that enough of the included cases would fulfill the TREND definition of viral and bacterial CAP, the proportion of viral respectively bacterial CAP (TREND definition) was calculated in our previous study that assessed Swedish children with x-ray verified CAP. By doing this, the prevalence of viral and bacterial CAP was estimated to 45% and 14% respectively. Hence 300 cases and 42 controls would be needed to ensure sufficient collection of cases with viral and bacterial CAP respectively. The investigators also would like to compare cases with viral CAP with controls testing positive for one or more virus by PCR. In our previous study, 35.4% of asymptomatic children tested positive for one or more virus. To include a sufficient number of virus-positive controls, the investigators hence aim at including 300 cases and 119 controls (42/0.354=119) in the TREND-study. Ethical Considerations: The study will be conducted in accordance with the latest version of The Declaration of Helsinki and the fundamental principles of respect for the individual (Article 8), their right to self-determination and the right to make informed decisions (Articles 20, 21 and 22) regarding participation in research, both initially and during the course of the research. Significance: The findings from the TREND project can be an important step to improve the care of children with clinical CAP. Improved near-patient differential diagnosis is a prerequisite for rational antibiotic use and decreasing unnecessary antibiotic treatment. Further better diagnosis of the pathogens causing acute respiratory infections makes it easier to give advice to parents on how their children should be cared for and better surveillance in the society.