CRP, PCT, TRAIL and IP‑10 as Host‑Response Biomarkers for Differentiating Bacterial and Non‑Bacterial Community‑Acquired Pneumonia in Children

CRP, PCT, TRAIL and IP‑10 as Host‑Response Biomarkers for Differentiating Bacterial and Non‑Bacterial Community‑Acquired Pneumonia in Children | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article CRP, PCT, TRAIL and IP‑10 as Host‑Response Biomarkers for Differentiating Bacterial and Non‑Bacterial Community‑Acquired Pneumonia in Children Anastasios Smyrnaios, Turid Follestad, Andreas Christensen, Kari Risnes, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9436832/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Distinguishing bacterial from non-bacterial community-acquired pneumonia (CAP) in children is challenging and often leads to unnecessary antibiotic prescribing. Host‑response biomarkers may improve diagnostic assessment. We evaluated the performance of C‑reactive protein (CRP), procalcitonin (PCT), tumour necrosis factor‑related apoptosis‑inducing ligand (TRAIL) and interferon gamma‑induced protein‑10 (IP‑10), individually and in combination, in differentiating bacterial from viral and atypical CAP. Methods Banked serum samples from a prospective cohort of Norwegian children with radiologically confirmed CAP (2012–2014) were analysed. Extensive microbiological testing classified cases as bacterial, viral, atypical or indeterminate. Of 265 eligible participants, 138 had measurements for all four biomarkers. Diagnostic performance was assessed using logistic regression and receiver operating characteristic (ROC) curves. Results Children with bacterial CAP generally had higher CRP and PCT and lower TRAIL concentrations than those with viral or atypical CAP. IP‑10 levels were broadly similar across aetiological groups. Among individual biomarkers, PCT (AUC 0.781, 95% CI 0.647–0.916) and TRAIL (AUC 0.776, 95% CI 0.655–0.898) showed slightly better discrimination than CRP or IP‑10, but confidence intervals were wide and largely overlapping. Biomarker combinations yielded numerically higher AUCs than individual markers and the combination of CRP, PCT and TRAIL had the highest AUC (0.848, 95% CI 0.766–0.930), but precision was limited by sample size. Conclusion In children with radiologically confirmed CAP, the combination of CRP, PCT and TRAIL showed the most favourable AUC, though differences between biomarkers and panels were modest. Multi‑biomarker approaches may help support more judicious antibiotic use, but validation in larger, adequately powered cohorts is required. community‑acquired pneumonia biomarkers paediatrics CRP PCT TRAIL IP‑10 Figures Figure 1 Figure 2 1. Introduction Paediatric community-acquired pneumonia (CAP) is one of the most common reasons for evaluation and antibiotic use in both primary care and hospital settings (1). In recent years, several studies have shown that in countries with high vaccination coverage against Haemophilus influenzae and Streptococcus pneumoniae , viruses are the predominant cause of paediatric CAP (2–5). Despite this, the majority of paediatric patients continue to be treated with antibiotics (2). The main reason is that distinguishing between viral and bacterial pneumonia at the individual level is challenging. Clinical features and chest radiography are often insufficient to differentiate between viral and bacterial aetiology (6–8). Moreover, microbiological samples from the lungs are rarely available, while other specimens—such as nasopharyngeal aspirates and blood cultures—have low specificity and/or sensitivity and do not exclude the possibility of co-infection (9). Excessive use of antibiotics due to diagnostic uncertainty leads to increased healthcare costs (10) and contributes to antimicrobial resistance (11). On the other hand, withholding antibiotics in patients with bacterial infection may result in increased morbidity and mortality (12). Routinely used biomarkers, such as white blood cell count (WBC), neutrophils, C-reactive protein (CRP), and procalcitonin (PCT), have suboptimal sensitivity and specificity in differentiating viral from bacterial infections (13). Recently, several studies have demonstrated that a host-response-based assay comprising CRP, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), and interferon gamma-induced protein 10 (IP-10) can accurately distinguish between viral and bacterial aetiology in children presenting with either lower respiratory tract infection or fever without a source (14–19). Furthermore, a study in adult patients showed that the addition of PCT to the three-protein assay further improved its accuracy (20). The aim of this study is to assess the performance of CRP, PCT, TRAIL, and IP-10 individually and in combination in distinguishing bacterial from viral and atypical paediatric CAP. 2. Methods Study population We used banked, frozen serum samples from a prospective, observational study of childhood CAP aetiology, conducted in Norway between 2012 and 2014 (4). A detailed description of the material and methods used have previously been published (4, 21). In brief, the study population consisted of patients under 18 years of age who were referred to the Paediatric Department, Akershus University Hospital with suspected CAP and treated either ambulatory or in hospital. For each participant, medical history and clinical findings were recorded. In addition, chest radiography was performed at inclusion and independently assessed by two radiologists who were blinded to the clinical data. Patients were diagnosed with pneumonia and included in the study if they met the following criteria: (a) fever or a history of fever; (b) one or more signs of acute lower respiratory tract infection; and (c) chest radiography with localised or interstitial infiltrates. Perihilar changes alone were not considered evidence of pneumonia. Exclusion criteria were hospital-acquired pneumonia, infection acquired during travel abroad, and underlying conditions predisposing to pneumonia. The aetiology of CAP was established using a combination of paired serum samples, polymerase chain reaction (PCR) testing of nasopharyngeal specimens, blood and pleural fluid cultures, and pneumococcal antigen testing in pleural fluid. Serological evidence of pneumococcal infection was assessed using two methods: (a) an in-house enzyme-linked immunosorbent assay (ELISA) detecting IgG against pneumolysin, and (b) flow cytometry to detect antibody binding to live pneumococci. A two-fold rise in antibody titers between paired serum samples by either method was considered indicative of pneumococcal infection. A more detailed description of these methods has been published previously (4). Bacterial CAP, with or without viral co-infection, was defined as the detection of a bacterium in blood culture, pleural fluid, or a positive serological test for S. pneumoniae . Atypical CAP, with or without viral co-infection, was defined as the detection of Mycoplasma pneumoniae or Chlamydia pneumoniae by nasopharyngeal PCR or serology. Viral CAP was defined as the detection of one or more viruses by nasopharyngeal PCR or serology, without evidence of bacterial co-infection. Serological test was performed for the following viruses: respiratory syncytial virus, adenovirus, influenza virus A/B and parainfluenza virus 1–3. CAP of indeterminate aetiology was defined as pneumonia for which no microbiological diagnosis could be established. In Norway, vaccination against Streptococcus pneumoniae and Haemophilus influenzae is part of the publicly funded national immunisation programme with 3 doses given at 3, 5 and 12 months of age. A seven-valent pneumococcal conjugate vaccine was introduced in 2006 and was replaced by a thirteen-valent vaccine in 2011. The Haemophilus influenzae type b vaccine was introduced in 1992. Vaccination coverage exceeds 95% for both vaccines (22). Biomarker Measurements Blood samples from all enrolled subjects were collected at the time of inclusion. CRP was measured as part of the initial study. PCT was measured later at the Research and Development Unit, Department of Paediatrics and Adolescent Medicine, Akershus University Hospital. PCT was analysed in 30 µL serum using the automated immunofluorescent assay Kryptor PCT (B.R.A.H.M.S., Hennigsdorf, Germany), which has a lower detection limit of 0.02 µg/L and a sensitivity of 0.06 µg/L. The assay principle is based on TRACE technology (time-resolved amplification of cryptate emission). Eight subjects had CRP values of 0 and five had PCT values of 0; for statistical analyses, these were replaced with 1 and 0.01, respectively. For TRAIL and IP-10 measurements, frozen serum remnants from the enrolment samples were used. Analyses were performed by trained personnel at the Department of Immunology and Transfusion Medicine, St. Olavs University Hospital of Trondheim, Norway. Laboratory technicians were blinded to all clinical data. Subjects without available samples for TRAIL and IP-10 were excluded. IP-10 and TRAIL concentrations were determined using Quantikine ELISA kits (Human CXCL10/IP-10 and Human TRAIL/TNFSF10; R&D Systems), following the procedures described in the package inserts. For IP-10 measurement, samples with concentrations above 500 pg/mL were diluted 1:5 using calibrator diluent. Seven samples required additional dilution (1:10) to obtain the final value. Data analysis/statistics Statistical analyses were conducted using IBM SPSS Statistics version 30 and R version 4.4.0. As the distributions of the four biomarkers were non-normal, values are presented as medians with interquartile ranges (IQR). Univariable and multivariable binary logistic regression models were used to assess the associations between a set of biomarkers, separately or in combination, and the presence of bacterial CAP. The models included from one to four of the analytes CRP, PCT, IP-10, and TRAIL. Subjects with indeterminate aetiology were excluded from the analyses. Receiver operating characteristic (ROC) curves were generated to evaluate the diagnostic performance of individual biomarkers and combinations of biomarkers in differentiating bacterial from non-bacterial CAP (viral or atypical). Due to a limited number of bacterial CAP cases, only in-sample validation was carried out. Results are reported as area under the ROC curve (AUC) with 95% confidence intervals (CI), and p-values for comparisons of AUCs between models. Comparisons of individual biomarker levels (CRP, TRAIL, IP-10, and PCT) across CAP types were performed using the Kruskal–Wallis test. A p-value < 0.05 was considered to indicate a statistically significant result. Ethics The study was in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The project was approved by the regional ethics committee (nr: 2011/2188). Informed, written consent was attained from parents or participants > 15 years of age prior to inclusion in 2012-14. Information about ongoing studies in biobanked samples was sent to all participants, or guardians of participants still < 16 years, in 2020, and participants were given the opportunity to withdraw consent. 3. Results Patient characteristics During the study period (2012–2014), 394 children with suspected CAP were recruited. Of these, 265 had radiologically confirmed CAP. A total of 127 children were excluded because either no blood sample was collected at inclusion (n = 41) or the available material was insufficient for measurement of TRAIL and IP-10 (n = 86). Consequently, 138 subjects in whom all four biomarkers were measured were included in the analyses. The two groups had similar characteristics, except that the median age of included subjects was statistically significantly higher than that of excluded subjects (Supplementary Table 1). Among the included children, 80 were diagnosed with viral CAP, 12 with atypical CAP (with or without viral co-infection), 16 with bacterial CAP (with or without viral co-infection), and in 30 cases the aetiology remained indeterminate (Fig. 1 ). Among all cases of atypical CAP, 11 were caused by Mycoplasma pneumoniae and one by Chlamydia pneumoniae . Among bacterial CAP, 14 were caused by Streptococcus pneumoniae and two by Streptococcus pyogenes . There were three cases with positive blood cultures: one with Streptococcus pneumoniae and two with Streptococcus pyogenes . The median (interquartile range [IQR]) age of children with viral CAP was 2.1 years (1.5–3.4), 10.6 years (5.4–14.6) for atypical CAP, 2.3 years (1.6–4.3) for bacterial CAP, and 3.6 years (1.8–7.9) for indeterminate cases. A majority of children received a full course of antibiotics, including 41% of those with viral CAP and three quarters of those with indeterminate cause. Hospitalisation rates ranged from 50% in atypical CAP to 75% in bacterial CAP, while the median length of stay varied from 2 days in atypical CAP to 4 days in viral CAP (Table 1 ). Table 1 Age in years, median (IQR) Viral (n = 80) Atypical (n = 12) Bacterial (n = 16) Indeterminate (n = 30) 2.1 (1.5–3.4) 10.6 (5.4–14.6) 2.3 (1.6–4.3) 3.6 (1.8–7.9) Male sex (%) 44 (55) 8 (66) 9 (56) 17 (56) Hospitalized (%) 48 (60) 6 (50) 12 (75) 20 (66) Length of admission, median in days (IQR) 4.0 (2.0–6.0) 2.0 (1.0-5.7) 3.0 (2.0-6.5) 2.5 (1.2-3.0) Antibiotic treatment No antibiotics (%) 21 (26) 2 (17) 4 (25) 6 (20) Antibiotics before inclusion (%) 28 (35) 5 (41) 1 (6) 9 (30) Treatment with full course of antibiotics (%) 33 (41) 10 (83) 12 (75) 23 (76) CRP, median (IQR) 70.0 (27.2-147.5) 47.0 (26.2–90.0) 275.0 (37.0-395.0) 195.0 (90.0-312.5) PCT (ug/L), median (IQR) 0.38 (0.08–1.69) 0.05 (0.04–0.09) 5.23 (0.59–15.7) 0.33 (0.15–1.71) TRAIL (pg/ml), median (IQR) 57.0 (32.5–88.7) 79.3 (58.5–108.0) 24.0 (16.2–49.0) 34.5 (21.0–54.0) IP10 (pg/ml), median (IQR) 620.6 (174.5-1194.9) 718.2 (195.0-1133.5) 770.8 (404.7–1283.0) 325.6 (93.0-747.7) Baseline characteristics of study participants according to type of pneumonia. Variables are presented as medians with interquartile ranges (IQR) or as percentages of the subgroup, as indicated. IQR, interquartile range; CRP, C‑reactive protein; PCT, procalcitonin; IP‑10, interferon gamma‑induced protein 10; TRAIL, tumour necrosis factor‑related apoptosis‑inducing ligand. Serum levels of biomarkers The levels of TRAIL were significantly lower in children with bacterial CAP compared with those with viral (p= .004) or atypical CAP (p= .001), with median (IQR) values of 24.0 (16.2–49.0) pg/mL versus 57.0 (32.5–88.7) pg/mL and 79.3 (58.5–108.0) pg/mL, respectively (Fig. 2 ). CRP and PCT showed the opposite pattern; median (IQR) CRP values were 275 (37–395) mg/L for bacterial CAP, compared with 70.0 (27.2–147.5) mg/L for viral CAP (p= .013) and 47.0 (26.2–90.0) mg/L for atypical CAP (p= .039); PCT values were 5.23 (0.59–15.7) µg/L versus 0.38 (0.08–1.69) µg/L (p= .011) and 0.05 (0.04–0.09) µg/L (p = < .001), respectively (Fig. 2 ). We found no statistically significant differences in IP10 levels among bacterial, viral, and atypical CAP (Fig. 2 ). Diagnostic performance of biomarkers We performed multivariable logistic regression analysis, and the odds ratios (ORs) of the biomarkers included in each model are presented in Supplementary Table 2. The biomarkers included in these models were selected based on the performance of the individual biomarkers. The AUC of CRP in differentiating bacterial from non-bacterial CAP was 0.727 (95% CI 0.536–0.919); for PCT, 0.781 (95% CI 0.647–0.916); for IP‑10, 0.606 (95% CI 0.474–0.739); and for TRAIL, 0.776 (95% CI 0.655–0.898). The ROC curves are presented in Fig. 3 . The AUCs and 95% CIs for the different combinations of biomarkers in differentiating bacterial from non-bacterial CAP are presented in Table 2 . The ROC curves for these combinations are presented in Fig. 3 . We found no statistically significant difference when comparing the AUC of CRP with those of the different biomarker combinations (Table 2 ). When the indeterminate cases were included either as bacterial CAP or as viral/atypical CAP, diagnostic performance metrics for most biomarkers and biomarker combinations remained largely unchanged (data not presented). Table 2 CRP AUC 95% CI p-value 0.727 0.536–0.919 NA PCT 0.781 0.647–0.916 NA IP10 0.606 0.474–0.739 NA TRAIL 0.776 0.655–0.898 NA CRP + PCT 0.806 0.708–0.905 .295 PCT+TRAIL 0.817 0.691–0.942 .106 CRP+TRAIL 0.784 0.655–0.913 .396 CRP+IP10 + TRAIL 0.821 0.707–0.934 .287 PCT+IP10 + TRAIL 0.806 0.677–0.935 .196 CRP + PCT+TRAIL 0.848 0.766–0.930 .141 CRP + PCT+IP10 + TRAIL 0.846 0.750–0.941 .175 Comparison of the diagnostic performance of biomarker combinations with the diagnostic performance of CRP. P‑values below 0.05 were considered statistically significant. AUC= Area Under the Curve, CI= Confidence Interval, NA = not applicable; CRP = C‑reactive protein; PCT = procalcitonin; Interferon gamma‑induced Protein 10; TRAIL = Tumour Necrosis Factor‑Related Apoptosis‑Inducing Ligand. Using Youden’s index, we calculated that a CRP concentration of 20 mg/L yielded a sensitivity of 87.5% and specificity of 18.5% in differentiating bacterial from viral or atypical CAP. The combination of CRP, PCT and TRAIL, which had the highest AUC in our study, achieved a sensitivity of 87.5% with a specificity of 65.2%. 4. Discussion This study indicated that, in children with radiologically confirmed CAP, the combination of CRP, PCT and TRAIL showed the highest numerical ability to distinguish bacterial from non‑bacterial CAP. Although this panel performed better than CRP alone and the combinations of TRAIL or IP‑10 with either CRP or PCT, the limited sample size means that these differences should be interpreted cautiously, as they did not reach statistical significance. When only two biomarkers were considered, the combination of PCT and TRAIL demonstrated the best overall diagnostic performance. In general, biomarker panels that integrated both viral and bacterial host‑response markers tended to perform better than individual biomarkers. A major strength of this study is the comprehensive microbiological evaluation used to classify viral, atypical and bacterial CAP. Because obtaining samples from the primary site of infection is rarely feasible in children, the assessment of diagnostic tests for CAP remains challenging. Several previous studies on biomarkers used to differentiate bacterial from non-bacterial infection, have used expert adjudication which facilitates inclusion of diagnostically complex cases but carries risks of misclassification and incorporation bias, particularly when adjudicators have access to biomarkers such as CRP (14–19). By applying broad microbiological testing, our study minimised these risks and strengthened the validity of case classification. Serological testing is useful in epidemiological studies (23, 24), although it is of limited value in routine clinical practice owing to restricted availability and the need for paired samples several weeks apart. In our study, two serological methods for S.pneumoniae were applied for each case, and the resulting measurements were strongly correlated, thereby increasing confidence in the resulting classification (4). Another strength is the inclusion of a homogeneous paediatric cohort, all with radiologically confirmed CAP, recruited from a range of clinical settings, including primary care, the emergency department, and inpatient wards. This diversity increases the generalisability of the findings. Treating atypical CAP as a separate aetiological category from bacterial CAP is also a strength, as the biomarker profile of atypical CAP more closely resembles that of viral infections. For this reason, viral and atypical CAP were evaluated together in our analyses of biomarker combinations. The most important limitation of the study is the small sample size, which reduced statistical power to detect modest differences in diagnostic accuracy and precluded external validation of the logistic regression models. Only about 40% of the initial cohort of children with radiologically confirmed CAP was included in the final analysis; the remainder were excluded because no sample was available for biomarker testing or because the reference standard was indeterminate. There were, however, no indications of major differences between the included and excluded children, suggesting that the generalizability of the findings is high. Several studies over the past decade (14–19) have reported high diagnostic accuracy for a three‑protein host‑response assay comprising CRP, TRAIL and IP‑10 in distinguishing viral from bacterial infections in children. These studies have generally shown sensitivity and specificity values close to 90%, along with a negative predictive value for bacterial infection of approximately 95%. In most of these investigations, the reference standard was determined by an expert adjudication panel, and patients with an indeterminate reference standard or an equivocal assay result were excluded from analysis. To the best of our knowledge, no previous studies have evaluated the performance of the combined biomarkers CRP, PCT, TRAIL and IP‑10 specifically in paediatric CAP. In adults presenting to the Emergency Department with fever, however, one study demonstrated that combining CRP, PCT, TRAIL and IP‑10 resulted in greater diagnostic accuracy than CRP, TRAIL and IP‑10 or PCT, TRAIL and IP‑10 alone (20). Our findings support the notion that viral and bacterial biomarkers may provide complementary diagnostic information. TRAIL provided an additive benefit when combined with CRP, PCT, or both. This observation is consistent with established biological patterns: TRAIL typically increases during viral infections and decreases during bacterial infections (25), thereby enhancing the diagnostic value of CRP and PCT, which primarily increase during bacterial infections. Although studies have shown that IP‑10 increases predominantly during viral infections, this was not the case in our cohort. In our study, median IP‑10 value for viral CAP was lower than that observed in bacterial and atypical CAP; however, these differences were not statistically significant (Table 1 , Fig. 2 ). Other studies have also reported only modest differences in IP‑10 concentrations between viral, atypical and bacterial infections in paediatric populations (26, 27). In our cohort 10 cases with confirmed bacterial CAP had a viral coinfection and this might have led to intermediate or blunted IP‑10 responses, thereby diminishing its discriminative ability. A novel aspect of this study, compared with many previous investigations, is that we assessed the diagnostic performance of these biomarkers specifically in atypical CAP. We confirmed the findings of an earlier study demonstrating that Mycoplasma-related CAP produced biomarker levels within the range typically seen in viral CAP (27). In our cohort, CRP and PCT levels were somewhat higher in viral than in atypical CAP, whereas TRAIL and IP‑10 levels were slightly higher in atypical than in viral CAP. Except for PCT, for which the difference was statistically significant, no statistically significant differences were observed for the other three biomarkers. The limited ability of traditional biomarkers and clinical features to distinguish viral from bacterial infection is illustrated by the high rate of antibiotic prescribing among children with confirmed viral CAP in our cohort: 74% received antibiotics and 41% completed a full treatment course. Other studies have shown that biomarker combinations with improved diagnostic accuracy can help reduce inappropriate antibiotic use (16, 18). Almost one fifth of our cohort had an indeterminate reference standard and were therefore excluded from the analysis. Interestingly, the values of TRAIL, PCT, and CRP in this group were not statistically different from those in children with bacterial CAP. Compared with children who had viral CAP, there were statistically significant differences in CRP and TRAIL, but not in PCT or IP‑10. This group likely included both bacterial and non‑bacterial cases, but our data does not allow firm conclusions. In conclusion, our study suggests that, in a cohort of children with radiologically confirmed CAP, the combination of CRP, PCT, and TRAIL showed the most promising diagnostic performance among the biomarker panels evaluated, indicating a potential role in supporting more judicious and restrictive antibiotic use. These findings should be confirmed in a larger study of radiological confirmed CAP classified on the basis of both host response (serology) and broad microbiological examinations. Declarations Competing Interests: The authors declare no competing interests. Funding: This study was supported by The Joint Research Committee between St. Olavs hospital and the Faculty of Medicine and Health Sciences, NTNU (grant number 2018/42795) Conflict of interest: The authors declare no conflicts of interest. Ethics approval: The study was in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Regional Committee for Medical and Health Research Ethics (reference no. 2011/2188). Acknowledgements: We would like to thank Tonje Stiansen‑Sonerud, molecular biologist, for performing the procalcitonin analyses; Hedda Trømborg Jalving, MD, for her contribution to creating Fig. 2; and Vibeke Stenhaug Langaas, MD, for her input regarding the analyses of IP‑10 and TRAIL. Consent: Written informed consent was obtained from parents or participants > 15 years of age at inclusion (2012–2014). Data availability: Data are available from the corresponding author upon reasonable request. Author Contribution AC, HD and CSI made substantial contributions to the conception of the work. AS, TF, KR, MKA, TH and HD made significant contributions to the data analysis and interpretation. AS drafted the original manuscript. All authors edited, revised, and approved the final version of the manuscript. Henrik Døllner and Christopher Stephen Inchley are shared last authors, contributed equally. Acknowledgement We would like to thank Tonje Stiansen‑Sonerud, molecular biologist, for performing the procalcitonin analyses; Hedda Trømborg Jalving, MD, for her contribution to creating Figure 2; and Vibeke Stenhaug Langaas, MD, for her input regarding the analyses of IP‑10 and TRAIL. References Florin TA. Differentiating Bacterial From Viral Etiologies in Pediatric Community-Acquired Pneumonia: The Quest for the Holy Grail Continues. J Pediatric Infect Dis Soc. 2021;10(12):1047–50. Jain S, Williams DJ, Arnold SR, Ampofo K, Bramley AM, Reed C, et al. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372(9):835–45. Eklundh A, Rhedin S, Ryd-Rinder M, Andersson M, Gantelius J, Gaudenzi G, et al. Etiology of Clinical Community-Acquired Pneumonia in Swedish Children Aged 1-59 Months with High Pneumococcal Vaccine Coverage-The TREND Study. Vaccines (Basel). 2021;9(4). Berg AS, Inchley CS, Aase A, Fjaerli HO, Bull R, Aaberge I, et al. Etiology of Pneumonia in a Pediatric Population with High Pneumococcal Vaccine Coverage: A Prospective Study. Pediatr Infect Dis J. 2016;35(3):e69–75. Smyrnaios A, Risnes K, Krokstad S, Nordbø SA, Heimdal I, Christensen A, et al. The Contribution of Viruses and Bacteria to Childhood Community-acquired Pneumonia: 11-Year Observational Study From Norway. Pediatr Infect Dis J. 2023;42(6):456–60. Andronikou S, Lambert E, Halton J, Hilder L, Crumley I, Lyttle MD, et al. Guidelines for the use of chest radiographs in community-acquired pneumonia in children and adolescents. Pediatr Radiol. 2017;47(11):1405–11. Virkki R, Juven T, Rikalainen H, Svedström E, Mertsola J, Ruuskanen O. Differentiation of bacterial and viral pneumonia in children. Thorax. 2002;57(5):438–41. Bhuiyan MU, Blyth CC, West R, Lang J, Rahman T, Granland C, et al. Combination of clinical symptoms and blood biomarkers can improve discrimination between bacterial or viral community-acquired pneumonia in children. BMC Pulm Med. 2019;19(1):71. Rodrigues CMC, Groves H. Community-Acquired Pneumonia in Children: the Challenges of Microbiological Diagnosis. J Clin Microbiol. 2018;56(3). Leigh S, Grant A, Murray N, Faragher B, Desai H, Dolan S, et al. The cost of diagnostic uncertainty: a prospective economic analysis of febrile children attending an NHS emergency department. BMC Med. 2019;17(1):48. Wenzel RP, Edmond MB. Managing antibiotic resistance. N Engl J Med. 2000;343(26):1961–3. Craig JC, Williams GJ, Jones M, Codarini M, Macaskill P, Hayen A, et al. The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses. Bmj. 2010;340:c1594. Gunaratnam LC, Robinson JL, Hawkes MT. Systematic Review and Meta-Analysis of Diagnostic Biomarkers for Pediatric Pneumonia. J Pediatric Infect Dis Soc. 2021;10(9):891–900. van Houten CB, de Groot JAH, Klein A, Srugo I, Chistyakov I, de Waal W, et al. A host-protein based assay to differentiate between bacterial and viral infections in preschool children (OPPORTUNITY): a double-blind, multicentre, validation study. Lancet Infect Dis. 2017;17(4):431–40. Srugo I, Klein A, Stein M, Golan-Shany O, Kerem N, Chistyakov I, et al. Validation of a Novel Assay to Distinguish Bacterial and Viral Infections. Pediatrics. 2017;140(4). Ashkenazi-Hoffnung L, Oved K, Navon R, Friedman T, Boico O, Paz M, et al. A host-protein signature is superior to other biomarkers for differentiating between bacterial and viral disease in patients with respiratory infection and fever without source: a prospective observational study. Eur J Clin Microbiol Infect Dis. 2018;37(7):1361–71. Stein M, Lipman-Arens S, Oved K, Cohen A, Bamberger E, Navon R, et al. A novel host-protein assay outperforms routine parameters for distinguishing between bacterial and viral lower respiratory tract infections. Diagn Microbiol Infect Dis. 2018;90(3):206–13. Papan C, Argentiero A, Porwoll M, Hakim U, Farinelli E, Testa I, et al. A host signature based on TRAIL, IP-10, and CRP for reducing antibiotic overuse in children by differentiating bacterial from viral infections: a prospective, multicentre cohort study. Clin Microbiol Infect. 2022;28(5):723–30. Mor M, Paz M, Amir L, Levy I, Scheuerman O, Livni G, et al. Bacterial vs viral etiology of fever: A prospective study of a host score for supporting etiologic accuracy of emergency department physicians. PLoS One. 2023;18(1):1–11. van der Does Y, Rood PPM, Ramakers C, Schuit SCE, Patka P, van Gorp ECM, et al. Identifying patients with bacterial infections using a combination of C-reactive protein, procalcitonin, TRAIL, and IP-10 in the emergency department: a prospective observational cohort study. Clin Microbiol Infect. 2018;24(12):1297–304. Berg AS, Inchley CS, Fjaerli HO, Leegaard TM, Lindbaek M, Nakstad B. Clinical features and inflammatory markers in pediatric pneumonia: a prospective study. Eur J Pediatr. 2017;176(5):629–38. FHI. Statistikk for barnevaksinasjonhttps://www.fhi.no/va/sysvak/barnevaksinasjon---statistikk/#dekningsstatistikk-2024 2024 [Available from: https://www.fhi.no/va/sysvak/barnevaksinasjon---statistikk/#dekningsstatistikk-2024. Andrade DC, Borges IC, Ivaska L, Peltola V, Meinke A, Barral A, et al. Serological diagnosis of pneumococcal infection in children with pneumonia using protein antigens: A study of cut-offs with positive and negative controls. J Immunol Methods. 2016;433:31–7. Korppi M, Leinonen M, Ruuskanen O. Pneumococcal serology in children's respiratory infections. Eur J Clin Microbiol Infect Dis. 2008;27(3):167–75. Oved K, Cohen A, Boico O, Navon R, Friedman T, Etshtein L, et al. A novel host-proteome signature for distinguishing between acute bacterial and viral infections. PLoS One. 2015;10(3):e0120012. Chokkalla AK, Tam E, Liang R, Cruz AT, Devaraj S. Validation of a multi-analyte immunoassay for distinguishing bacterial vs. viral infections in a pediatric cohort. Clin Chim Acta. 2023;546:117387. Papan C, Sidorov S, Greiter B, Bühler N, Berger C, Becker SL, et al. Combinatorial Host-Response Biomarker Signature (BV Score) and Its Subanalytes TRAIL, IP-10, and C-Reactive Protein in Children With Mycoplasma pneumoniae Community-Acquired Pneumonia. J Infect Dis. 2024;230(2):e247–e53. Additional Declarations No competing interests reported. Supplementary Files SupplementaryTable1.docx SupplementaryTable2.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9436832","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":629861818,"identity":"ae08fc1c-39f7-432b-8f68-522455949da5","order_by":0,"name":"Anastasios Smyrnaios","email":"data:image/png;base64,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","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Anastasios","middleName":"","lastName":"Smyrnaios","suffix":""},{"id":629861819,"identity":"5a57c304-f0fa-443d-ae46-1ac76668939d","order_by":1,"name":"Turid Follestad","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Turid","middleName":"","lastName":"Follestad","suffix":""},{"id":629861820,"identity":"77888a3f-6185-45a0-b15d-f98cb3c86a00","order_by":2,"name":"Andreas Christensen","email":"","orcid":"","institution":"St Olavs University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Christensen","suffix":""},{"id":629861829,"identity":"59ebd172-de58-4bc5-8858-f8efabedb988","order_by":3,"name":"Kari Risnes","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Kari","middleName":"","lastName":"Risnes","suffix":""},{"id":629861832,"identity":"c82a05fb-9076-4c1d-be90-d86ef5cc4b1b","order_by":4,"name":"Marit Kristin Aarhaug","email":"","orcid":"","institution":"St Olavs University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Marit","middleName":"Kristin","lastName":"Aarhaug","suffix":""},{"id":629861835,"identity":"79885838-e04c-42bc-8adf-70317f20d661","order_by":5,"name":"Toril Holien","email":"","orcid":"","institution":"St Olavs University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Toril","middleName":"","lastName":"Holien","suffix":""},{"id":629861839,"identity":"ec8382e4-ddb2-4e33-96f1-ab26e0d3638f","order_by":6,"name":"Henrik Døllner","email":"","orcid":"","institution":"Norwegian University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Henrik","middleName":"","lastName":"Døllner","suffix":""},{"id":629861842,"identity":"82a82b6e-79af-4542-9dcd-f66ab354b982","order_by":7,"name":"Christopher Stephen Inchley","email":"","orcid":"","institution":"Akershus University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"Stephen","lastName":"Inchley","suffix":""}],"badges":[],"createdAt":"2026-04-16 10:23:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9436832/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9436832/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108398008,"identity":"d41bc4e5-56c8-4b02-a7c7-2798a7e0f433","added_by":"auto","created_at":"2026-05-04 08:25:51","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":39702,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of participants with suspected CAP.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9436832/v1/3c17bfb6e8c332175d25623e.jpg"},{"id":108398018,"identity":"0b205c2e-f252-4f20-9b79-f7815cf56563","added_by":"auto","created_at":"2026-05-04 08:25:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":71481,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots of individual biomarker concentrations according to the type of community‑acquired pneumonia. Biomarkers are presented in their logarithmic/normalised expression. Boxes represent the 25th–75th percentiles, with the median indicated by the line within each box. Circles denote outliers. Statistically significant differences between pneumonia groups for each biomarker are displayed in the upper part of each panel; non‑significant comparisons are shown in the lower part of each panel.\u003c/p\u003e\n\u003cp\u003eCRP = C‑reactive protein; PCT = procalcitonin; Interferon gamma‑induced Protein 10; TRAIL = Tumour Necrosis Factor‑Related Apoptosis‑Inducing Ligand.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9436832/v1/7799f3b78a2749086f3a1046.png"},{"id":109257759,"identity":"d0dce3f6-b7fc-472b-956d-57130e05c429","added_by":"auto","created_at":"2026-05-14 10:25:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":323423,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9436832/v1/a332ccd1-ac34-41b7-b6b6-b6ca43c35cf5.pdf"},{"id":108398017,"identity":"1d91f91e-98a3-4e85-a724-76278a30a700","added_by":"auto","created_at":"2026-05-04 08:25:56","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":29691,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9436832/v1/4d2744f7d87fac0e9b7594c7.docx"},{"id":108398068,"identity":"a381b363-2c22-47f3-b567-8edbd77ee0be","added_by":"auto","created_at":"2026-05-04 08:26:11","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":33439,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable2.docx","url":"https://assets-eu.researchsquare.com/files/rs-9436832/v1/b25a1fa4996cebc59258c239.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"CRP, PCT, TRAIL and IP‑10 as Host‑Response Biomarkers for Differentiating Bacterial and Non‑Bacterial Community‑Acquired Pneumonia in Children","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003ePaediatric community-acquired pneumonia (CAP) is one of the most common reasons for evaluation and antibiotic use in both primary care and hospital settings (1). In recent years, several studies have shown that in countries with high vaccination coverage against \u003cem\u003eHaemophilus influenzae\u003c/em\u003e and \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e, viruses are the predominant cause of paediatric CAP (2\u0026ndash;5). Despite this, the majority of paediatric patients continue to be treated with antibiotics (2).\u003c/p\u003e \u003cp\u003eThe main reason is that distinguishing between viral and bacterial pneumonia at the individual level is challenging. Clinical features and chest radiography are often insufficient to differentiate between viral and bacterial aetiology (6\u0026ndash;8). Moreover, microbiological samples from the lungs are rarely available, while other specimens\u0026mdash;such as nasopharyngeal aspirates and blood cultures\u0026mdash;have low specificity and/or sensitivity and do not exclude the possibility of co-infection (9).\u003c/p\u003e \u003cp\u003eExcessive use of antibiotics due to diagnostic uncertainty leads to increased healthcare costs (10) and contributes to antimicrobial resistance (11). On the other hand, withholding antibiotics in patients with bacterial infection may result in increased morbidity and mortality (12).\u003c/p\u003e \u003cp\u003eRoutinely used biomarkers, such as white blood cell count (WBC), neutrophils, C-reactive protein (CRP), and procalcitonin (PCT), have suboptimal sensitivity and specificity in differentiating viral from bacterial infections (13). Recently, several studies have demonstrated that a host-response-based assay comprising CRP, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), and interferon gamma-induced protein 10 (IP-10) can accurately distinguish between viral and bacterial aetiology in children presenting with either lower respiratory tract infection or fever without a source (14\u0026ndash;19). Furthermore, a study in adult patients showed that the addition of PCT to the three-protein assay further improved its accuracy (20).\u003c/p\u003e \u003cp\u003eThe aim of this study is to assess the performance of CRP, PCT, TRAIL, and IP-10 individually and in combination in distinguishing bacterial from viral and atypical paediatric CAP.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cp\u003eStudy population\u003c/p\u003e \u003cp\u003eWe used banked, frozen serum samples from a prospective, observational study of childhood CAP aetiology, conducted in Norway between 2012 and 2014 (4). A detailed description of the material and methods used have previously been published (4, 21). In brief, the study population consisted of patients under 18 years of age who were referred to the Paediatric Department, Akershus University Hospital with suspected CAP and treated either ambulatory or in hospital. For each participant, medical history and clinical findings were recorded. In addition, chest radiography was performed at inclusion and independently assessed by two radiologists who were blinded to the clinical data. Patients were diagnosed with pneumonia and included in the study if they met the following criteria: (a) fever or a history of fever; (b) one or more signs of acute lower respiratory tract infection; and (c) chest radiography with localised or interstitial infiltrates. Perihilar changes alone were not considered evidence of pneumonia. Exclusion criteria were hospital-acquired pneumonia, infection acquired during travel abroad, and underlying conditions predisposing to pneumonia.\u003c/p\u003e \u003cp\u003eThe aetiology of CAP was established using a combination of paired serum samples, polymerase chain reaction (PCR) testing of nasopharyngeal specimens, blood and pleural fluid cultures, and pneumococcal antigen testing in pleural fluid. Serological evidence of pneumococcal infection was assessed using two methods: (a) an in-house enzyme-linked immunosorbent assay (ELISA) detecting IgG against pneumolysin, and (b) flow cytometry to detect antibody binding to live pneumococci. A two-fold rise in antibody titers between paired serum samples by either method was considered indicative of pneumococcal infection. A more detailed description of these methods has been published previously (4).\u003c/p\u003e \u003cp\u003eBacterial CAP, with or without viral co-infection, was defined as the detection of a bacterium in blood culture, pleural fluid, or a positive serological test for \u003cem\u003eS. pneumoniae\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eAtypical CAP, with or without viral co-infection, was defined as the detection of \u003cem\u003eMycoplasma pneumoniae\u003c/em\u003e or \u003cem\u003eChlamydia pneumoniae\u003c/em\u003e by nasopharyngeal PCR or serology.\u003c/p\u003e \u003cp\u003eViral CAP was defined as the detection of one or more viruses by nasopharyngeal PCR or serology, without evidence of bacterial co-infection. Serological test was performed for the following viruses: respiratory syncytial virus, adenovirus, influenza virus A/B and parainfluenza virus 1\u0026ndash;3.\u003c/p\u003e \u003cp\u003eCAP of indeterminate aetiology was defined as pneumonia for which no microbiological diagnosis could be established.\u003c/p\u003e \u003cp\u003eIn Norway, vaccination against \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e and \u003cem\u003eHaemophilus influenzae\u003c/em\u003e is part of the publicly funded national immunisation programme with 3 doses given at 3, 5 and 12 months of age. A seven-valent pneumococcal conjugate vaccine was introduced in 2006 and was replaced by a thirteen-valent vaccine in 2011. The \u003cem\u003eHaemophilus influenzae type b\u003c/em\u003e vaccine was introduced in 1992. Vaccination coverage exceeds 95% for both vaccines (22).\u003c/p\u003e \u003cp\u003eBiomarker Measurements\u003c/p\u003e \u003cp\u003eBlood samples from all enrolled subjects were collected at the time of inclusion. CRP was measured as part of the initial study. PCT was measured later at the Research and Development Unit, Department of Paediatrics and Adolescent Medicine, Akershus University Hospital. PCT was analysed in 30 \u0026micro;L serum using the automated immunofluorescent assay Kryptor PCT (B.R.A.H.M.S., Hennigsdorf, Germany), which has a lower detection limit of 0.02 \u0026micro;g/L and a sensitivity of 0.06 \u0026micro;g/L. The assay principle is based on TRACE technology (time-resolved amplification of cryptate emission).\u003c/p\u003e \u003cp\u003eEight subjects had CRP values of 0 and five had PCT values of 0; for statistical analyses, these were replaced with 1 and 0.01, respectively. For TRAIL and IP-10 measurements, frozen serum remnants from the enrolment samples were used. Analyses were performed by trained personnel at the Department of Immunology and Transfusion Medicine, St. Olavs University Hospital of Trondheim, Norway. Laboratory technicians were blinded to all clinical data. Subjects without available samples for TRAIL and IP-10 were excluded.\u003c/p\u003e \u003cp\u003eIP-10 and TRAIL concentrations were determined using Quantikine ELISA kits (Human CXCL10/IP-10 and Human TRAIL/TNFSF10; R\u0026amp;D Systems), following the procedures described in the package inserts. For IP-10 measurement, samples with concentrations above 500 pg/mL were diluted 1:5 using calibrator diluent. Seven samples required additional dilution (1:10) to obtain the final value.\u003c/p\u003e \u003cp\u003eData analysis/statistics\u003c/p\u003e \u003cp\u003eStatistical analyses were conducted using IBM SPSS Statistics version 30 and R version 4.4.0. As the distributions of the four biomarkers were non-normal, values are presented as medians with interquartile ranges (IQR). Univariable and multivariable binary logistic regression models were used to assess the associations between a set of biomarkers, separately or in combination, and the presence of bacterial CAP. The models included from one to four of the analytes CRP, PCT, IP-10, and TRAIL. Subjects with indeterminate aetiology were excluded from the analyses. Receiver operating characteristic (ROC) curves were generated to evaluate the diagnostic performance of individual biomarkers and combinations of biomarkers in differentiating bacterial from non-bacterial CAP (viral or atypical). Due to a limited number of bacterial CAP cases, only in-sample validation was carried out. Results are reported as area under the ROC curve (AUC) with 95% confidence intervals (CI), and p-values for comparisons of AUCs between models. Comparisons of individual biomarker levels (CRP, TRAIL, IP-10, and PCT) across CAP types were performed using the Kruskal\u0026ndash;Wallis test. A p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered to indicate a statistically significant result.\u003c/p\u003e \u003cp\u003eEthics\u003c/p\u003e \u003cp\u003e The study was in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The project was approved by the regional ethics committee (nr: 2011/2188). Informed, written consent was attained from parents or participants\u0026thinsp;\u0026gt;\u0026thinsp;15 years of age prior to inclusion in 2012-14. Information about ongoing studies in biobanked samples was sent to all participants, or guardians of participants still\u0026thinsp;\u0026lt;\u0026thinsp;16 years, in 2020, and participants were given the opportunity to withdraw consent.\u003c/p\u003e"},{"header":"3. Results","content":"\u003cp\u003ePatient characteristics\u003c/p\u003e \u003cp\u003eDuring the study period (2012\u0026ndash;2014), 394 children with suspected CAP were recruited. Of these, 265 had radiologically confirmed CAP. A total of 127 children were excluded because either no blood sample was collected at inclusion (n\u0026thinsp;=\u0026thinsp;41) or the available material was insufficient for measurement of TRAIL and IP-10 (n\u0026thinsp;=\u0026thinsp;86). Consequently, 138 subjects in whom all four biomarkers were measured were included in the analyses. The two groups had similar characteristics, except that the median age of included subjects was statistically significantly higher than that of excluded subjects (Supplementary Table\u0026nbsp;1). Among the included children, 80 were diagnosed with viral CAP, 12 with atypical CAP (with or without viral co-infection), 16 with bacterial CAP (with or without viral co-infection), and in 30 cases the aetiology remained indeterminate (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Among all cases of atypical CAP, 11 were caused by \u003cem\u003eMycoplasma pneumoniae\u003c/em\u003e and one by \u003cem\u003eChlamydia pneumoniae\u003c/em\u003e. Among bacterial CAP, 14 were caused by \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e and two by \u003cem\u003eStreptococcus pyogenes\u003c/em\u003e. There were three cases with positive blood cultures: one with \u003cem\u003eStreptococcus pneumoniae\u003c/em\u003e and two with \u003cem\u003eStreptococcus pyogenes\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eThe median (interquartile range [IQR]) age of children with viral CAP was 2.1 years (1.5\u0026ndash;3.4), 10.6 years (5.4\u0026ndash;14.6) for atypical CAP, 2.3 years (1.6\u0026ndash;4.3) for bacterial CAP, and 3.6 years (1.8\u0026ndash;7.9) for indeterminate cases. A majority of children received a full course of antibiotics, including 41% of those with viral CAP and three quarters of those with indeterminate cause. Hospitalisation rates ranged from 50% in atypical CAP to 75% in bacterial CAP, while the median length of stay varied from 2 days in atypical CAP to 4 days in viral CAP (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e\u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAge in years, median (IQR)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eViral\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;80)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAtypical\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;12)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBacterial\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;16)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIndeterminate (n\u0026thinsp;=\u0026thinsp;30)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.1 (1.5\u0026ndash;3.4)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.6 (5.4\u0026ndash;14.6)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.3 (1.6\u0026ndash;4.3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.6 (1.8\u0026ndash;7.9)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale sex (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44 (55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (66)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e17 (56)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospitalized (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48 (60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (50)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20 (66)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLength of admission, median in days (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.0 (2.0\u0026ndash;6.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.0 (1.0-5.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.0 (2.0-6.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.5 (1.2-3.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAntibiotic treatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo antibiotics (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21 (26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (20)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAntibiotics before inclusion (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (30)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTreatment with full course of antibiotics (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33 (41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (83)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e23 (76)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70.0 (27.2-147.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47.0 (26.2\u0026ndash;90.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e275.0 (37.0-395.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e195.0 (90.0-312.5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCT (ug/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.38 (0.08\u0026ndash;1.69)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.05 (0.04\u0026ndash;0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.23 (0.59\u0026ndash;15.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.33 (0.15\u0026ndash;1.71)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTRAIL (pg/ml), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e57.0 (32.5\u0026ndash;88.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e79.3 (58.5\u0026ndash;108.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24.0 (16.2\u0026ndash;49.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e34.5 (21.0\u0026ndash;54.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIP10 (pg/ml), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e620.6 (174.5-1194.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e718.2 (195.0-1133.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e770.8 (404.7\u0026ndash;1283.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e325.6 (93.0-747.7)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eBaseline characteristics of study participants according to type of pneumonia. Variables are presented as medians with interquartile ranges (IQR) or as percentages of the subgroup, as indicated.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eIQR, interquartile range; CRP, C‑reactive protein; PCT, procalcitonin; IP‑10, interferon gamma‑induced protein 10; TRAIL, tumour necrosis factor‑related apoptosis‑inducing ligand.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eSerum levels of biomarkers\u003c/p\u003e \u003cp\u003eThe levels of TRAIL were significantly lower in children with bacterial CAP compared with those with viral (p= .004) or atypical CAP (p= .001), with median (IQR) values of 24.0 (16.2\u0026ndash;49.0) pg/mL versus 57.0 (32.5\u0026ndash;88.7) pg/mL and 79.3 (58.5\u0026ndash;108.0) pg/mL, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). CRP and PCT showed the opposite pattern; median (IQR) CRP values were 275 (37\u0026ndash;395) mg/L for bacterial CAP, compared with 70.0 (27.2\u0026ndash;147.5) mg/L for viral CAP (p= .013) and 47.0 (26.2\u0026ndash;90.0) mg/L for atypical CAP (p= .039); PCT values were 5.23 (0.59\u0026ndash;15.7) \u0026micro;g/L versus 0.38 (0.08\u0026ndash;1.69) \u0026micro;g/L (p= .011) and 0.05 (0.04\u0026ndash;0.09) \u0026micro;g/L (p\u0026thinsp;=\u0026thinsp;\u0026lt;\u0026thinsp;.001), respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). We found no statistically significant differences in IP10 levels among bacterial, viral, and atypical CAP (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDiagnostic performance of biomarkers\u003c/p\u003e \u003cp\u003eWe performed multivariable logistic regression analysis, and the odds ratios (ORs) of the biomarkers included in each model are presented in Supplementary Table\u0026nbsp;2. The biomarkers included in these models were selected based on the performance of the individual biomarkers.\u003c/p\u003e \u003cp\u003eThe AUC of CRP in differentiating bacterial from non-bacterial CAP was 0.727 (95% CI 0.536\u0026ndash;0.919); for PCT, 0.781 (95% CI 0.647\u0026ndash;0.916); for IP‑10, 0.606 (95% CI 0.474\u0026ndash;0.739); and for TRAIL, 0.776 (95% CI 0.655\u0026ndash;0.898). The ROC curves are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eThe AUCs and 95% CIs for the different combinations of biomarkers in differentiating bacterial from non-bacterial CAP are presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The ROC curves for these combinations are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. We found no statistically significant difference when comparing the AUC of CRP with those of the different biomarker combinations (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). When the indeterminate cases were included either as bacterial CAP or as viral/atypical CAP, diagnostic performance metrics for most biomarkers and biomarker combinations remained largely unchanged (data not presented).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e\u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eCRP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.727\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.536\u0026ndash;0.919\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.647\u0026ndash;0.916\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIP10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.606\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.474\u0026ndash;0.739\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.776\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.655\u0026ndash;0.898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP\u0026thinsp;+\u0026thinsp;PCT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.708\u0026ndash;0.905\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.295\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCT+TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.817\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.691\u0026ndash;0.942\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.106\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP+TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.655\u0026ndash;0.913\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.396\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP+IP10\u0026thinsp;+\u0026thinsp;TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.821\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.707\u0026ndash;0.934\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCT+IP10\u0026thinsp;+\u0026thinsp;TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.677\u0026ndash;0.935\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.196\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP\u0026thinsp;+\u0026thinsp;PCT+TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.766\u0026ndash;0.930\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.141\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP\u0026thinsp;+\u0026thinsp;PCT+IP10\u0026thinsp;+\u0026thinsp;TRAIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.846\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.750\u0026ndash;0.941\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e.175\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eComparison of the diagnostic performance of biomarker combinations with the diagnostic performance of CRP. P‑values below 0.05 were considered statistically significant.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eAUC= Area Under the Curve, CI= Confidence Interval, NA\u0026thinsp;=\u0026thinsp;not applicable; CRP\u0026thinsp;=\u0026thinsp;C‑reactive protein; PCT\u0026thinsp;=\u0026thinsp;procalcitonin; Interferon gamma‑induced Protein 10; TRAIL\u0026thinsp;=\u0026thinsp;Tumour Necrosis Factor‑Related Apoptosis‑Inducing Ligand.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eUsing Youden\u0026rsquo;s index, we calculated that a CRP concentration of 20 mg/L yielded a sensitivity of 87.5% and specificity of 18.5% in differentiating bacterial from viral or atypical CAP. The combination of CRP, PCT and TRAIL, which had the highest AUC in our study, achieved a sensitivity of 87.5% with a specificity of 65.2%.\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eThis study indicated that, in children with radiologically confirmed CAP, the combination of CRP, PCT and TRAIL showed the highest numerical ability to distinguish bacterial from non‑bacterial CAP. Although this panel performed better than CRP alone and the combinations of TRAIL or IP‑10 with either CRP or PCT, the limited sample size means that these differences should be interpreted cautiously, as they did not reach statistical significance. When only two biomarkers were considered, the combination of PCT and TRAIL demonstrated the best overall diagnostic performance. In general, biomarker panels that integrated both viral and bacterial host‑response markers tended to perform better than individual biomarkers.\u003c/p\u003e \u003cp\u003eA major strength of this study is the comprehensive microbiological evaluation used to classify viral, atypical and bacterial CAP. Because obtaining samples from the primary site of infection is rarely feasible in children, the assessment of diagnostic tests for CAP remains challenging. Several previous studies on biomarkers used to differentiate bacterial from non-bacterial infection, have used expert adjudication which facilitates inclusion of diagnostically complex cases but carries risks of misclassification and incorporation bias, particularly when adjudicators have access to biomarkers such as CRP (14\u0026ndash;19). By applying broad microbiological testing, our study minimised these risks and strengthened the validity of case classification. Serological testing is useful in epidemiological studies (23, 24), although it is of limited value in routine clinical practice owing to restricted availability and the need for paired samples several weeks apart. In our study, two serological methods for \u003cem\u003eS.pneumoniae\u003c/em\u003e were applied for each case, and the resulting measurements were strongly correlated, thereby increasing confidence in the resulting classification (4).\u003c/p\u003e \u003cp\u003eAnother strength is the inclusion of a homogeneous paediatric cohort, all with radiologically confirmed CAP, recruited from a range of clinical settings, including primary care, the emergency department, and inpatient wards. This diversity increases the generalisability of the findings. Treating atypical CAP as a separate aetiological category from bacterial CAP is also a strength, as the biomarker profile of atypical CAP more closely resembles that of viral infections. For this reason, viral and atypical CAP were evaluated together in our analyses of biomarker combinations.\u003c/p\u003e \u003cp\u003eThe most important limitation of the study is the small sample size, which reduced statistical power to detect modest differences in diagnostic accuracy and precluded external validation of the logistic regression models. Only about 40% of the initial cohort of children with radiologically confirmed CAP was included in the final analysis; the remainder were excluded because no sample was available for biomarker testing or because the reference standard was indeterminate. There were, however, no indications of major differences between the included and excluded children, suggesting that the generalizability of the findings is high.\u003c/p\u003e \u003cp\u003eSeveral studies over the past decade (14\u0026ndash;19) have reported high diagnostic accuracy for a three‑protein host‑response assay comprising CRP, TRAIL and IP‑10 in distinguishing viral from bacterial infections in children. These studies have generally shown sensitivity and specificity values close to 90%, along with a negative predictive value for bacterial infection of approximately 95%. In most of these investigations, the reference standard was determined by an expert adjudication panel, and patients with an indeterminate reference standard or an equivocal assay result were excluded from analysis. To the best of our knowledge, no previous studies have evaluated the performance of the combined biomarkers CRP, PCT, TRAIL and IP‑10 specifically in paediatric CAP. In adults presenting to the Emergency Department with fever, however, one study demonstrated that combining CRP, PCT, TRAIL and IP‑10 resulted in greater diagnostic accuracy than CRP, TRAIL and IP‑10 or PCT, TRAIL and IP‑10 alone (20).\u003c/p\u003e \u003cp\u003eOur findings support the notion that viral and bacterial biomarkers may provide complementary diagnostic information. TRAIL provided an additive benefit when combined with CRP, PCT, or both. This observation is consistent with established biological patterns: TRAIL typically increases during viral infections and decreases during bacterial infections (25), thereby enhancing the diagnostic value of CRP and PCT, which primarily increase during bacterial infections. Although studies have shown that IP‑10 increases predominantly during viral infections, this was not the case in our cohort. In our study, median IP‑10 value for viral CAP was lower than that observed in bacterial and atypical CAP; however, these differences were not statistically significant (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Other studies have also reported only modest differences in IP‑10 concentrations between viral, atypical and bacterial infections in paediatric populations (26, 27). In our cohort 10 cases with confirmed bacterial CAP had a viral coinfection and this might have led to intermediate or blunted IP‑10 responses, thereby diminishing its discriminative ability.\u003c/p\u003e \u003cp\u003eA novel aspect of this study, compared with many previous investigations, is that we assessed the diagnostic performance of these biomarkers specifically in atypical CAP. We confirmed the findings of an earlier study demonstrating that Mycoplasma-related CAP produced biomarker levels within the range typically seen in viral CAP (27). In our cohort, CRP and PCT levels were somewhat higher in viral than in atypical CAP, whereas TRAIL and IP‑10 levels were slightly higher in atypical than in viral CAP. Except for PCT, for which the difference was statistically significant, no statistically significant differences were observed for the other three biomarkers.\u003c/p\u003e \u003cp\u003eThe limited ability of traditional biomarkers and clinical features to distinguish viral from bacterial infection is illustrated by the high rate of antibiotic prescribing among children with confirmed viral CAP in our cohort: 74% received antibiotics and 41% completed a full treatment course. Other studies have shown that biomarker combinations with improved diagnostic accuracy can help reduce inappropriate antibiotic use (16, 18).\u003c/p\u003e \u003cp\u003eAlmost one fifth of our cohort had an indeterminate reference standard and were therefore excluded from the analysis. Interestingly, the values of TRAIL, PCT, and CRP in this group were not statistically different from those in children with bacterial CAP. Compared with children who had viral CAP, there were statistically significant differences in CRP and TRAIL, but not in PCT or IP‑10. This group likely included both bacterial and non‑bacterial cases, but our data does not allow firm conclusions.\u003c/p\u003e \u003cp\u003eIn conclusion, our study suggests that, in a cohort of children with radiologically confirmed CAP, the combination of CRP, PCT, and TRAIL showed the most promising diagnostic performance among the biomarker panels evaluated, indicating a potential role in supporting more judicious and restrictive antibiotic use. These findings should be confirmed in a larger study of radiological confirmed CAP classified on the basis of both host response (serology) and broad microbiological examinations.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting Interests:\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding:\u003c/h2\u003e\n\u003cp\u003eThis study was supported by The Joint Research Committee between St. Olavs hospital and the Faculty of Medicine and Health Sciences, NTNU (grant number 2018/42795)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the Regional Committee for Medical and Health Research Ethics (reference no. 2011/2188).\u003c/p\u003e\n\u003cp\u003eAcknowledgements: We would like to thank Tonje Stiansen‑Sonerud, molecular biologist, for performing the procalcitonin analyses; Hedda Tr\u0026oslash;mborg Jalving, MD, for her contribution to creating Fig.\u0026nbsp;2; and Vibeke Stenhaug Langaas, MD, for her input regarding the analyses of IP‑10 and TRAIL.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWritten informed consent was obtained from parents or participants\u0026thinsp;\u0026gt;\u0026thinsp;15 years of age at inclusion (2012\u0026ndash;2014).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eAC, HD and CSI made substantial contributions to the conception of the work. AS, TF, KR, MKA, TH and HD made significant contributions to the data analysis and interpretation. AS drafted the original manuscript. All authors edited, revised, and approved the final version of the manuscript. Henrik D\u0026oslash;llner and Christopher Stephen Inchley are shared last authors, contributed equally.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eWe would like to thank Tonje Stiansen‑Sonerud, molecular biologist, for performing the procalcitonin analyses; Hedda Tr\u0026oslash;mborg Jalving, MD, for her contribution to creating Figure 2; and Vibeke Stenhaug Langaas, MD, for her input regarding the analyses of IP‑10 and TRAIL.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eFlorin TA. Differentiating Bacterial From Viral Etiologies in Pediatric Community-Acquired Pneumonia: The Quest for the Holy Grail Continues. J Pediatric Infect Dis Soc. 2021;10(12):1047\u0026ndash;50.\u003c/li\u003e\n\u003cli\u003eJain S, Williams DJ, Arnold SR, Ampofo K, Bramley AM, Reed C, et al. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372(9):835\u0026ndash;45.\u003c/li\u003e\n\u003cli\u003eEklundh A, Rhedin S, Ryd-Rinder M, Andersson M, Gantelius J, Gaudenzi G, et al. Etiology of Clinical Community-Acquired Pneumonia in Swedish Children Aged 1-59 Months with High Pneumococcal Vaccine Coverage-The TREND Study. Vaccines (Basel). 2021;9(4).\u003c/li\u003e\n\u003cli\u003eBerg AS, Inchley CS, Aase A, Fjaerli HO, Bull R, Aaberge I, et al. Etiology of Pneumonia in a Pediatric Population with High Pneumococcal Vaccine Coverage: A Prospective Study. Pediatr Infect Dis J. 2016;35(3):e69\u0026ndash;75.\u003c/li\u003e\n\u003cli\u003eSmyrnaios A, Risnes K, Krokstad S, Nordb\u0026oslash; SA, Heimdal I, Christensen A, et al. The Contribution of Viruses and Bacteria to Childhood Community-acquired Pneumonia: 11-Year Observational Study From Norway. Pediatr Infect Dis J. 2023;42(6):456\u0026ndash;60.\u003c/li\u003e\n\u003cli\u003eAndronikou S, Lambert E, Halton J, Hilder L, Crumley I, Lyttle MD, et al. Guidelines for the use of chest radiographs in community-acquired pneumonia in children and adolescents. Pediatr Radiol. 2017;47(11):1405\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003eVirkki R, Juven T, Rikalainen H, Svedstr\u0026ouml;m E, Mertsola J, Ruuskanen O. Differentiation of bacterial and viral pneumonia in children. Thorax. 2002;57(5):438\u0026ndash;41.\u003c/li\u003e\n\u003cli\u003eBhuiyan MU, Blyth CC, West R, Lang J, Rahman T, Granland C, et al. Combination of clinical symptoms and blood biomarkers can improve discrimination between bacterial or viral community-acquired pneumonia in children. BMC Pulm Med. 2019;19(1):71.\u003c/li\u003e\n\u003cli\u003eRodrigues CMC, Groves H. Community-Acquired Pneumonia in Children: the Challenges of Microbiological Diagnosis. J Clin Microbiol. 2018;56(3).\u003c/li\u003e\n\u003cli\u003eLeigh S, Grant A, Murray N, Faragher B, Desai H, Dolan S, et al. The cost of diagnostic uncertainty: a prospective economic analysis of febrile children attending an NHS emergency department. BMC Med. 2019;17(1):48.\u003c/li\u003e\n\u003cli\u003eWenzel RP, Edmond MB. Managing antibiotic resistance. N Engl J Med. 2000;343(26):1961\u0026ndash;3.\u003c/li\u003e\n\u003cli\u003eCraig JC, Williams GJ, Jones M, Codarini M, Macaskill P, Hayen A, et al. The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses. Bmj. 2010;340:c1594.\u003c/li\u003e\n\u003cli\u003eGunaratnam LC, Robinson JL, Hawkes MT. Systematic Review and Meta-Analysis of Diagnostic Biomarkers for Pediatric Pneumonia. J Pediatric Infect Dis Soc. 2021;10(9):891\u0026ndash;900.\u003c/li\u003e\n\u003cli\u003evan Houten CB, de Groot JAH, Klein A, Srugo I, Chistyakov I, de Waal W, et al. A host-protein based assay to differentiate between bacterial and viral infections in preschool children (OPPORTUNITY): a double-blind, multicentre, validation study. Lancet Infect Dis. 2017;17(4):431\u0026ndash;40.\u003c/li\u003e\n\u003cli\u003eSrugo I, Klein A, Stein M, Golan-Shany O, Kerem N, Chistyakov I, et al. Validation of a Novel Assay to Distinguish Bacterial and Viral Infections. Pediatrics. 2017;140(4).\u003c/li\u003e\n\u003cli\u003eAshkenazi-Hoffnung L, Oved K, Navon R, Friedman T, Boico O, Paz M, et al. A host-protein signature is superior to other biomarkers for differentiating between bacterial and viral disease in patients with respiratory infection and fever without source: a prospective observational study. Eur J Clin Microbiol Infect Dis. 2018;37(7):1361\u0026ndash;71.\u003c/li\u003e\n\u003cli\u003eStein M, Lipman-Arens S, Oved K, Cohen A, Bamberger E, Navon R, et al. A novel host-protein assay outperforms routine parameters for distinguishing between bacterial and viral lower respiratory tract infections. Diagn Microbiol Infect Dis. 2018;90(3):206\u0026ndash;13.\u003c/li\u003e\n\u003cli\u003ePapan C, Argentiero A, Porwoll M, Hakim U, Farinelli E, Testa I, et al. A host signature based on TRAIL, IP-10, and CRP for reducing antibiotic overuse in children by differentiating bacterial from viral infections: a prospective, multicentre cohort study. Clin Microbiol Infect. 2022;28(5):723\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eMor M, Paz M, Amir L, Levy I, Scheuerman O, Livni G, et al. Bacterial vs viral etiology of fever: A prospective study of a host score for supporting etiologic accuracy of emergency department physicians. PLoS One. 2023;18(1):1\u0026ndash;11.\u003c/li\u003e\n\u003cli\u003evan der Does Y, Rood PPM, Ramakers C, Schuit SCE, Patka P, van Gorp ECM, et al. Identifying patients with bacterial infections using a combination of C-reactive protein, procalcitonin, TRAIL, and IP-10 in the emergency department: a prospective observational cohort study. Clin Microbiol Infect. 2018;24(12):1297\u0026ndash;304.\u003c/li\u003e\n\u003cli\u003eBerg AS, Inchley CS, Fjaerli HO, Leegaard TM, Lindbaek M, Nakstad B. Clinical features and inflammatory markers in pediatric pneumonia: a prospective study. Eur J Pediatr. 2017;176(5):629\u0026ndash;38.\u003c/li\u003e\n\u003cli\u003eFHI. Statistikk for barnevaksinasjonhttps://www.fhi.no/va/sysvak/barnevaksinasjon---statistikk/#dekningsstatistikk-2024 2024 [Available from: https://www.fhi.no/va/sysvak/barnevaksinasjon---statistikk/#dekningsstatistikk-2024.\u003c/li\u003e\n\u003cli\u003eAndrade DC, Borges IC, Ivaska L, Peltola V, Meinke A, Barral A, et al. Serological diagnosis of pneumococcal infection in children with pneumonia using protein antigens: A study of cut-offs with positive and negative controls. J Immunol Methods. 2016;433:31\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eKorppi M, Leinonen M, Ruuskanen O. Pneumococcal serology in children\u0026apos;s respiratory infections. Eur J Clin Microbiol Infect Dis. 2008;27(3):167\u0026ndash;75.\u003c/li\u003e\n\u003cli\u003eOved K, Cohen A, Boico O, Navon R, Friedman T, Etshtein L, et al. A novel host-proteome signature for distinguishing between acute bacterial and viral infections. PLoS One. 2015;10(3):e0120012.\u003c/li\u003e\n\u003cli\u003eChokkalla AK, Tam E, Liang R, Cruz AT, Devaraj S. Validation of a multi-analyte immunoassay for distinguishing bacterial vs. viral infections in a pediatric cohort. Clin Chim Acta. 2023;546:117387.\u003c/li\u003e\n\u003cli\u003ePapan C, Sidorov S, Greiter B, B\u0026uuml;hler N, Berger C, Becker SL, et al. Combinatorial Host-Response Biomarker Signature (BV Score) and Its Subanalytes TRAIL, IP-10, and C-Reactive Protein in Children With Mycoplasma pneumoniae Community-Acquired Pneumonia. J Infect Dis. 2024;230(2):e247\u0026ndash;e53.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"community‑acquired pneumonia, biomarkers, paediatrics, CRP, PCT, TRAIL, IP‑10","lastPublishedDoi":"10.21203/rs.3.rs-9436832/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9436832/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDistinguishing bacterial from non-bacterial community-acquired pneumonia (CAP) in children is challenging and often leads to unnecessary antibiotic prescribing. Host‑response biomarkers may improve diagnostic assessment. We evaluated the performance of C‑reactive protein (CRP), procalcitonin (PCT), tumour necrosis factor‑related apoptosis‑inducing ligand (TRAIL) and interferon gamma‑induced protein‑10 (IP‑10), individually and in combination, in differentiating bacterial from viral and atypical CAP.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBanked serum samples from a prospective cohort of Norwegian children with radiologically confirmed CAP (2012–2014) were analysed. Extensive microbiological testing classified cases as bacterial, viral, atypical or indeterminate. Of 265 eligible participants, 138 had measurements for all four biomarkers. Diagnostic performance was assessed using logistic regression and receiver operating characteristic (ROC) curves.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eChildren with bacterial CAP generally had higher CRP and PCT and lower TRAIL concentrations than those with viral or atypical CAP. IP‑10 levels were broadly similar across aetiological groups. Among individual biomarkers, PCT (AUC 0.781, 95% CI 0.647–0.916) and TRAIL (AUC 0.776, 95% CI 0.655–0.898) showed slightly better discrimination than CRP or IP‑10, but confidence intervals were wide and largely overlapping.\u003c/p\u003e\n\u003cp\u003eBiomarker combinations yielded numerically higher AUCs than individual markers and the combination of CRP, PCT and TRAIL had the highest AUC (0.848, 95% CI 0.766–0.930), but precision was limited by sample size.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn children with radiologically confirmed CAP, the combination of CRP, PCT and TRAIL showed the most favourable AUC, though differences between biomarkers and panels were modest. Multi‑biomarker approaches may help support more judicious antibiotic use, but validation in larger, adequately powered cohorts is required.\u003c/p\u003e","manuscriptTitle":"CRP, PCT, TRAIL and IP‑10 as Host‑Response Biomarkers for Differentiating Bacterial and Non‑Bacterial Community‑Acquired Pneumonia in Children","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 08:23:57","doi":"10.21203/rs.3.rs-9436832/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a862f03a-0eed-419f-88b9-96df87195673","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-14T10:15:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-12T20:36:30+00:00","index":21,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T14:53:20+00:00","index":20,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-14T10:25:35+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 08:23:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9436832","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9436832","identity":"rs-9436832","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}