BACKGROUND:Since the introduction of pediatric pneumococcal conjugate vaccines were introduced in the US in 2000, acute respiratory illnesses, including acute otitis media (AOM) due to Streptococcus pneumoniae (Spn), have declined. Pathogens associated with AOM have changed over time. METHODS:Children aged 6 through 35 months diagnosed with either AOM (Cohort 1-AOM with tympanocentesis & Cohort 2-AOM without tympanocentesis) or upper respiratory infection (URI) without AOM (Cohort 3) were concurrently enrolled in Pittsburgh, PA. Nasal specimens were collected for all children. Middle ear fluid (MEF) specimens were obtained by tympanocentesis for children in Cohort 1. MEF and nasal specimens were tested for Spn, Haemophilus influenzae (Hflu), and Moraxella catarrhalis (Mcat). Serotypes (Spn) and susceptibility (Spn and Hflu) were determined. RESULTS:From October 2019 through August 2023, 451 children were enrolled (Cohort 1: n = 57; Cohort 2: n = 262; Cohort 3: n = 132). For bacterial testing of MEF, 45/57 (79%) had a single pathogen detected. Pathogens detected, alone or in combination in the MEF were Hflu 32/57 (56%), Mcat 13/57 (23%), and Spn 12/57 (21%). Of participants with bacterial pathogens detected in the MEF, there was only modest concordance with the pathogens detected in the nose. Spn serotypes observed in MEF were 3, 11A, 15B, and 19A. Among nasal specimens from children with AOM (n = 319), Spn, Hflu and Mcat were detected in 46%, 43%, and 70%, respectively. The majority (134/194; 69%) of Spn isolates from the nasal specimens were penicillin- susceptible; frequently observed serotypes were 3,15B, 15C, 23B/B1, 23A, and 35B. Almost 40% (63/162) of children colonized with Hflu had a β-lactamase producing isolate. CONCLUSIONS:Hflu was most frequently identified in MEF, although the sample size was small. In nasal specimens, antibiotic-resistant Hflu were common, while the majority of Spn were penicillin-susceptible. Continued surveillance could provide valuable data to guide vaccine development and therapeutic decision-making.

The human nasopharynx is colonized by a diverse community of commensal microbiota linked to many respiratory diseases, yet their associations with the host remain unclear. In this study, we introduced a dual-transcriptomics analysis strategy, which can characterize the host transcriptome and microbiome from nasal samples simultaneously. We applied this workflow to a local SARS-CoV-2 cohort with 76 asymptomatic infected patients, among whom 52 (68.42

Importance:Diagnosing urinary tract infection (UTI) in preverbal, pre-toilet trained children is challenging and may lead to unnecessary testing and treatment. UTICalc estimates UTI risk using clinical and laboratory data but has not been prospectively validated. Objectives:To prospectively validate UTICalc version 3.0 for predicting UTI and to evaluate its utility in guiding clinical decisions. Design, Setting, and Participants:This prospective diagnostic study was conducted at 2 tertiary care pediatric emergency departments (EDs) in Canada from November 2022 to January 2025. Children aged 2 to 24 months presenting with measured fever (≥38.0 °C) were enrolled. Exclusions included congenital urinary tract abnormalities, immunosuppression, current antimicrobials, or prior enrollment. Participants were followed up to identify UTIs; urine culture results were reviewed within 48 hours, and phone or email follow-up was completed within 2 weeks for those without testing at the index visit. Exposure:UTICalc-predicted probability of UTI. Main Outcomes and Measures:The primary outcome was UTI, defined as positive urinalysis and significant uropathogen culture growth. The primary analysis examined discriminative performance using area under receiver operating characteristic curve (AUROC). Secondary analyses included calibration (calibration plots, slope, Brier score) and clinical utility (net benefit analysis, comparison with clinician practice). Results:Among 2561 included participants (1212 [47%] female; 1326 [64%] younger than 12 months), 111 children (4%) were classified as having a UTI. In the full sample, 2256 (88.1%) had urine testing and/or provided follow-up data. The clinical model AUROC was 84.1% (95% CI, 80.4%-87.9%). At a 2% UTI risk threshold, sensitivity was 96.4%, and specificity was 34.1%; at a 5% risk threshold, sensitivity was 82.0%, and specificity was 73.8%. The clinical and dipstick model AUROC was 95.3% (95% CI, 93.3%-97.4%), with 94.0% sensitivity and 86.9% specificity at 5% risk. The sensitivity and specificity for clinicians were 98.2% and 57.3%, respectively. Calibration slopes were 0.11 (95% CI, 0.09-0.13) and 0.06 (95% CI, 0.05-0.07) and Brier scores were 0.04 (95% CI, 0.03-0.05) and 0.05 (95% CI, 0.04-0.06) for the clinical and clinical and dipstick models, respectively. Decision curve analysis showed both models were associated with positive net benefit across a range of risk thresholds. Conclusions and Relevance:In this cohort study of 2561 febrile children aged 2 to 24 months presenting to the ED, UTICalc showed strong diagnostic performance, especially with dipstick results. While it did not outperform experienced clinicians, it offered a useful, evidence-based adjunct for guiding urine testing decisions, supporting efficient care in the pediatric ED.

BACKGROUND:Limited data exist on whether the timing or duration of treatment for acute otitis media (AOM) affects nasopharyngeal carriage. We studied the eradication of pathogens (Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis) in relation to timing (immediate vs. delayed) and duration (5 vs. 10 days) of antibiotic treatment. METHODS:A secondary analysis of 3 randomized clinical trials: the Turku antibiotic versus placebo trial, the Pittsburgh antibiotic versus placebo trial and the Pittsburgh 5- versus 10-day trial (ClinicalTrials.gov registration numbers NCT00299455, NCT00377260 and NCT01511107, respectively). We investigated the change in AOM pathogen colonization rate in relation to timing (immediate vs. delayed antibiotic initiation) and duration (5 vs. 10 days) of antibiotic treatment for AOM. We included children aged 6-35 months with AOM who had nasopharyngeal samples collected both before and after treatment. RESULTS:Antibiotic treatment with amoxicillin-clavulanate significantly reduced the carriage of S. pneumoniae and M. catarrhalis but had very little effect on the carriage of H. influenzae. The timing of treatment did not affect the eradication rates of nasopharyngeal pathogens. Antibiotic treatment eradicated S. pneumoniae and M. catarrhalis in 78% (95% confidence interval [CI], 72-83) and 85% (95% CI, 72-95) of cases, while H. influenzae was eradicated in only 20% (95% CI, 12-29) of cases. CONCLUSIONS:In most cases, H. influenzae carriage in the nasopharynx persisted despite antibiotic treatment with amoxicillin-clavulanate. This should be considered when choosing antibiotic treatment for children with a relapse of AOM.

Bacterial urinary tract infections (UTIs) are prevalent in childhood and adolescence. Paediatric UTIs present unique challenges with respect to diagnosis, prevention and management, and the potential for adverse sequelae. Uropathogenic Escherichia coli (UPEC) accounts for the majority of UTIs and is the best studied uropathogen. Novel discoveries have advanced our understanding of host–pathogen interactions, cellular and molecular mechanisms of host defence, and risk factors for UTI recurrence. Emerging evidence also highlights an association of the gut, vaginal and urinary microbiota in influencing UTI risk and recurrence. Yet, key knowledge gaps persist regarding UTI pathogenesis, host susceptibility, optimal diagnostic and management strategies and prevention of UTI recurrence and sequelae, especially in paediatric populations. The development of standardized clinical pathways offers an opportunity to improve care consistency and outcomes by integrating evidence-based practices into routine management. As technologies evolve and understanding deepens, future efforts must integrate host, microbial and clinical insights to optimize UTI prevention and treatment in paediatric populations. In this Review, the authors describe and discuss the unique challenges of paediatric urinary tract infections including diagnosis, prevention and management, potential for adverse sequelae and key knowledge gaps.

Introduction Few qualitative studies have examined parents’ experiences in the days following their child’s acute otitis media (AOM) diagnosis. This qualitative study explored challenges faced by parents and probed parents about how these challenges could be mitigated. Methods Interviews took place from October 2022 to July 2024 with parents from four pediatric practices. Recruitment was limited to parents of children 3-24 months of age who were recently diagnosed with AOM. Transcripts were analyzed using template analysis. Results We conducted 19 interviews. Major concerns identified by parents were failure of symptom improvement, missed sleep/work, and issues related to antimicrobial treatment. The experience of parents of children with multiple AOM episodes differed from that of less experienced parents. Parents identified areas physicians could improve parental experiences such as setting expectations for the recovery period and offering optional follow-up visits at the end of the treatment period. Discussion and Conclusions This study provides insights into challenges parents face in the days following diagnosis of AOM. Our findings suggest parents may benefit from additional physician counseling at the time of diagnosis, including a detailed discussion of what to expect during the recovery period and review of indications for follow-up.

Bacterial urinary tract infections (UTIs) are prevalent in childhood and adolescence. Paediatric UTIs present unique challenges with respect to diagnosis, prevention and management, and the potential for adverse sequelae. Uropathogenic Escherichia coli (UPEC) accounts for the majority of UTIs and is the best studied uropathogen. Novel discoveries have advanced our understanding of host-pathogen interactions, cellular and molecular mechanisms of host defence, and risk factors for UTI recurrence. Emerging evidence also highlights an association of the gut, vaginal and urinary microbiota in influencing UTI risk and recurrence. Yet, key knowledge gaps persist regarding UTI pathogenesis, host susceptibility, optimal diagnostic and management strategies and prevention of UTI recurrence and sequelae, especially in paediatric populations. The development of standardized clinical pathways offers an opportunity to improve care consistency and outcomes by integrating evidence-based practices into routine management. As technologies evolve and understanding deepens, future efforts must integrate host, microbial and clinical insights to optimize UTI prevention and treatment in paediatric populations.

Background Urinary tract infection (UTI) is a common reason for hospitalization in young children. American and Canadian guidelines recommend universal renal ultrasound after a first febrile UTI to identify genitourinary abnormalities that require further management, despite limited supporting evidence. The Renal ultrasOund after first febrile Urinary Tract INfEction (ROUTINE) study aims to (a) determine the incidence of clinically important and insignificant genitourinary abnormalities on renal ultrasound, (b) describe the cascade of healthcare interventions following the detection of abnormalities, (c) explore parents' experiences, and (d) identify clinical factors associated with clinically important abnormalities. Method This multicentre, prospective, observational study with an embedded qualitative component will enroll children <24 months of age hospitalized at 17 children's and community hospitals in Canada and the United States after a first febrile UTI, with 1-year follow-up. The primary outcome is detection of genitourinary abnormalities on renal ultrasound, classified by clinical importance using a standardized system. Secondary outcomes include clinical (recurrent febrile UTI), radiologic (vesicoureteral reflux), healthcare interventions (surgery, antibiotic prophylaxis, imaging, specialist follow-up), and parent-reported outcomes (anxiety, involvement in decision making). Parent experiences will be explored through semistructured interviews. Primary analysis will estimate the incidence of clinically important abnormalities and potential overdiagnosis with 95% confidence intervals (CIs). Secondary analysis will summarize radiologic, clinical, healthcare intervention, and parent-reported outcomes and assess predictors of clinically important abnormalities using multivariable logistic regression. Discussion This multicentre prospective study with longitudinal follow-up will generate evidence on the utility of universal renal ultrasound after the first febrile UTI. Findings will inform evidence-based management, shared decision-making, and clinical practice guidelines.

OBJECTIVE:To assess the predictive accuracy of urine neutrophil gelatinase-associated lipocalin (NGAL) and the Shannon Diversity Index in diagnosing urinary tract infection (UTI) in children with spina bifida. STUDY DESIGN:This was a cross-sectional study of children with spina bifida who were evaluated for UTI in the emergency department. We included patients younger than the age of 22 years with spina bifida who used clean intermittent catheterization and had a urinalysis and urine culture sent. We measured urine NGAL and completed 16SrRNA sequencing. We generated receiver operating characteristic curves to assess the predictive accuracy of urinary NGAL and Shannon Diversity Index. RESULTS:There were 51 children, 13 with UTI, in this study. The median Shannon Diversity Index was lower in children with UTI compared with those without, whereas median NGAL was significantly greater in the UTI vs the no-UTI group. The area under the curve for the combination of NGAL and Shannon Diversity was 0.88 (0.78-0.99), which is not significantly different from either the under the curve for NGAL or Shannon Diversity Index alone. NGAL was more specific, whereas the Shannon Diversity Index was more sensitive for the diagnosis of UTI. CONCLUSIONS:NGAL is specific, and Shannon Diversity Index is sensitive, for diagnosing UTI in children with spina bifida who use clean intermittent catheterization. However, the combination of the 2 markers does not create improved predictive accuracy over either alone.

Prediction of outcomes of SARS-CoV-2 infection remains challenging, particularly in the early days following exposure. To better understand the heterogeneity of disease progression, we investigated the early immune response in the upper respiratory tract using transcriptomic analysis, comparing individuals who remained asymptomatic to those who developed symptoms of COVID-19. We conducted a study of 74 individuals (43 children, 31 adults) with confirmed SARS-CoV-2 infection. Mid-turbinate nasal swabs were collected during the first few days of infection and again one week later. We performed a paired analysis comparing baseline and follow-up nasal human gene expression. Additionally, we conducted a predictive analysis to identify transcripts associated with the development of symptomatic infection. From the differentially expressed transcripts in both analyses, we developed a predictive model to assess the likelihood of symptomatic disease. We also compared gene expression patterns between children and adults. A robust interferon response in the upper respiratory tract was strongly associated with the development of symptomatic infection. A panel of five interferon-stimulated genes (TAP2, DDX60, IFIT5, APOL6, and IFI6) predicted symptomatic infection with reasonable accuracy (area under the curve = 0.84). Notably, adaptive immune responses, including T-cell activation and cytokine signaling, differed substantially between children and adults. Our findings suggest that early measurement of interferon-stimulated gene expression may help identify individuals at risk of developing symptomatic COVID-19.

Abstract Background Despite the success of pediatric pneumococcal conjugate vaccine (PCV) programs in the US, acute respiratory illness, including acute otitis media (AOM) due to Streptococcus pneumoniae (SPN) persists. PCV13 has been recommended for pediatric routine use since 2010. With recent adoption of higher-valency PCVs, there is a continued need to survey pathogens causing AOM in children.Table 1:Bacterial Results of Nasal Specimens in Children 6-35 Months Diagnosed with Acute Otitis Media (Cohorts 1&2) or Upper Respiratory Infection (Cohort 3) Methods Children aged 6 to 35 months diagnosed with AOM (Cohort 1 & 2) and upper respiratory infection (URI, Cohort 3) were enrolled from primary care and ENT practices in Pittsburgh, PA. Nasal specimens (nasopharyngeal or mid-turbinate) were collected for all children. Middle ear fluid (MEF) specimens were collected by tympanocentesis for children enrolled in Cohort 1. MEF and nasal specimens were cultured for SPN, Hemophilus influenzae (Hflu), and Moraxella catarrhalis (Mcat). Serotypes (ST) and susceptibility testing were determined for SPN isolates.Table 2:Correlation of Bacterial Cultures of Nasal and Middle Ear Fluid (MEF) Specimens (n=72 pairs) in Children 6-35 Months Diagnosed with Acute Otitis Media (Cohort 1) Results From October 2019 to January 2024, 545 children were enrolled (Cohort 1: n=72; Cohort 2: n=322; Cohort 3: n=151). Among nasal specimens from children with AOM (n=394), SPN, Hflu and Mcat were detected in 47%, 44% and 70%, respectively. In MEF specimens from children with AOM (n=72), SPN, Hflu and Mcat were detected in 15%, 54% and 17% respectively. Nasal-MEF pairs were concordant for 79% (57/72). Children with Hflu had higher rates of nasal-MEF concordance than children with SPN or Mcat. In nasal specimens from children with URI (n=151), SPN, Hflu and Mcat were detected in 36%, 18% and 54%, respectively. In nasal specimens, co-detection of bacterial pathogens occurred more often in those with AOM (220/394, 56%) than URI (49/151, 32%). Of the SPN ST results available (nasal: n=174; MEF: n=8), 35B, 15B, 3, and 15C were most frequent; 8% of nasal and 50% of MEF were PCV13 STs. SPN was found simultaneously in the nasal-MEF pairs of 10/72 children, of which STs are available for 7 pairs; 6/7 pairs agreed, 1 disagreed (nasal:35B and MEF:3). SPN nonsusceptible to penicillin were uncommon (2% of nasal and 9% of MEF isolates).Figure 1:Distribution of S. pneumoniae Serotypes by Isolate Source Conclusion STs 3 and 19A, included in PCV13, were responsible for half of SPN isolates found in MEF of children with AOM. Continued surveillance is still warranted to guide vaccine development and therapeutic decision-making for children. Disclosures Judith M. Martin, MD, Centers for Disease Control and Prevention: funding to the institution to support this work|Centers for Disease Control and Prevention: funding to the institution for unrelated work|Merck, Sharp and Dhome: funding to the institution to support this work|Moderna: funding to the institution for unrelated work|NIH: funding to the institution for unrelated work Matthew C. Lee, BA, Centers for Disease Control and Prevention (CDC): Grant paid to the institution|Merck Sharpe & Dohme: Grant paid to the institution Jessica Weaver, PhD, MPH, Merck & Co. Inc.: Employment|Merck & Co. Inc.: Stocks/Bonds (Public Company) Meghan White, PharmD, Merck Sharp & Dohme, LLC: Employee|Merck Sharp & Dohme, LLC: Stocks/Bonds (Public Company) Alejandro Hoberman, MD, Kaizen Biosciences: Advisor/Consultant|Kaizen Biosciences: Pediatric oral suspension formulation of amoxicillin-clavulanate potassium and the method for use.|Kaizen Biosciences: Ownership Interest|Not licensed: Method and apparatus for aiding diagnosis of otitis media by classifying tympanic membrane images.

The upper respiratory tract (URT) microbiome has emerged as a key component of acute otitis media (AOM) pathophysiology; however, few studies conducted to date have evaluated URT microbiome composition in children with recurrent AOM (rAOM). We collected serial nasopharyngeal samples from a cohort of 58 children, 6 to 35 months of age, over a one-year period. Samples were analyzed using 16S rRNA sequencing and PCR-based assays for common otopathogens and respiratory viruses. Age was strongly associated with differential abundance of specific genera, including increased abundance of genera associated with respiratory health (e.g., Dolosigranulum, Corynebacterium). In contrast, samples collected during AOM episodes or within 30 days of receipt of an antibiotic had a lower relative abundance of these genera. Further, the number of antibiotic-free days prior to sample collection was associated with global changes in microbiome composition. Unsupervised clustering identified three microbiome profiles that differed by incidence of AOM, bacterial otopathogen burden, symptom score, and number of antibiotic-free days prior to sample collection. Increasing age was associated with transition to profiles characterized by lower incidence of AOM and bacterial otopathogen burden, while antibiotic use was associated with transition to a profile associated with greater incidence of AOM. Our findings indicate that alterations of the microbiome associated with aging may contribute to decreased incidence of AOM as children age, while systemic antibiotic use may induce dysbiosis, thereby enhancing AOM susceptibility. IMPORTANCE:Ear infections are the most common bacterial infection among young children and the leading cause of healthcare visits and antibiotic prescriptions. This study explores the connection between the microbiome of the nose-the community of microorganisms that live in different areas of the human body-and recurrent ear infections in young children. An analysis of nasal swabs collected from 58 children over a year showed that as children age, they tend to have fewer bacterial pathogens and more species that are associated with a healthy state in their microbiomes. These more mature microbiomes were associated with fewer ear infections. In contrast, recent use of antibiotics was associated with microbiomes that had more bacterial pathogens and that were associated with greater ear infection incidence. Overall, these findings indicate that the microbiome may be a key factor in reduced ear infections as children age.

BACKGROUND:Acute sinusitis (AS) is a frequent cause of antibiotic prescriptions in children. Distinguishing bacterial AS from common viral upper respiratory infections (URIs) is crucial to prevent unnecessary antibiotic use but is challenging with current diagnostic methods. Despite its speed and cost, untargeted RNA sequencing of clinical samples from children with suspected AS has the potential to overcome several limitations of other methods. In addition, RNA-seq may reveal novel host-response biomarkers for development of future diagnostic assays that distinguish bacterial from viral infections. There are however no available RNA-seq datasets of pediatric AS that provide a comprehensive view of both pathogen etiology and host immune response. METHODS:Here, we performed untargeted RNA-seq (metatranscriptomics) of nasopharyngeal samples from 221 children with AS and performed a comprehensive analysis of pathogen etiology and the impact of bacterial and viral infections on host immune responses. Accuracy of RNA-seq-based pathogen detection was evaluated by comparison with culture tests for three common bacterial pathogens and qRT-PCR tests for 12 respiratory viruses. Host gene expression patterns were explored to identify potential host responses that distinguish bacterial from viral infections. RESULTS:RNA-seq-based pathogen detection showed high concordance with culture or qRT-PCR, showing 87%/81% sensitivity (sens) / specificity (spec) for detecting three AS-associated bacterial pathogens, and 86%/92% (sens/spec) for detecting 12 URI-associated viruses, respectively. RNA-seq also detected an additional 22 pathogens not tested for clinically and identified plausible pathogens in 11/19 (58%) of cases where no organism was detected by culture or qRT-PCR. We reconstructed genomes of 196 viruses across the samples including novel strains of coronaviruses, respiratory syncytial virus, and enterovirus D68, which provide useful genomic data for ongoing pathogen surveillance programs. By analyzing host gene expression, we identified host-response signatures that differentiate bacterial and viral infections, revealing hundreds of candidate gene biomarkers for future diagnostic assays. CONCLUSIONS:Our study provides a one-of-kind dataset that profiles the interplay between pathogen infection and host responses in pediatric AS and URI. It reveals bacterial and viral-specific host responses that could enable new diagnostic approaches and demonstrates the potential of untargeted RNA-seq in diagnostic analysis of AS and URI.

OBJECTIVES:Urinary tract infections (UTIs) are common, but overdiagnosed, in children with spina bifida. We sought to evaluate the diagnostic test characteristics of urinalysis (UA) findings for symptomatic UTI in children with spina bifida. METHODS:Retrospective cross-sectional study using data from 2 centers from January 1, 2016, to December 31, 2021. Children with myelomeningocele aged <19 years who had paired UA (and microscopy, when available) and urine culture were included. The primary outcome was symptomatic UTI. We used generalized estimating equations to control for multiple encounters per child and calculated area under the receiver operating characteristics curve, sensitivity, and specificity for positive nitrites, pyuria (≥10 white blood cells/high-powered field), and leukocyte esterase (more than trace) for a symptomatic UTI. RESULTS:We included 974 encounters from 319 unique children, of which 120 (12.3%) met our criteria for UTI. Pyuria had the highest sensitivity while nitrites were the most specific. Comparatively, nitrites were the least sensitive and pyuria was the least specific. When the cohort was limited to children with symptoms of a UTI, pyuria remained the most sensitive parameter, whereas nitrites remained the least sensitive. Nitrites continued to be the most specific, whereas pyuria was the least specific. Among all encounters, the overall area under the receiver operating characteristics curve for all components of the UA was lower in children who use clean intermittent catheterizations compared with all others. CONCLUSIONS:Individual UA findings have moderate sensitivity (leukocyte esterase or pyuria) or specificity (nitrites) but overall poor diagnostic accuracy for symptomatic UTIs in children with spina bifida.

Acute sinusitis (AS) is the fifth leading cause of antibiotic prescriptions in children. Distinguishing bacterial AS from common viral upper respiratory infections in children is crucial to prevent unnecessary antibiotic use but is challenging with current diagnostic methods. Despite its speed and cost, untargeted RNA sequencing of clinical samples from children with suspected AS has the potential to overcome several limitations of other methods. However, the utility of sequencing-based approaches in analysis of AS has not been fully explored. Here, we performed RNA-seq of nasopharyngeal samples from 221 children with clinically diagnosed AS to characterize their pathogen and host-response profiles. Results from RNA-seq were compared with those obtained using culture for three common bacterial pathogens and qRT-PCR for 12 respiratory viruses. Metatranscriptomic pathogen detection showed high concordance with culture or qRT-PCR, showing 87%/81% sensitivity (sens) / specificity (spec) for detecting bacteria, and 86%/92% (sens/spec) for viruses, respectively. We also detected an additional 22 pathogens not tested for in the clinical panel, and identified plausible pathogens in 11/19 (58%) of cases where no organism was detected by culture or qRT-PCR. We assembled genomes of 205 viruses across the samples including novel strains of coronaviruses, respiratory syncytial virus (RSV), and enterovirus D68. By analyzing host gene expression, we identified host-response signatures that distinguished bacterial and viral infections and correlated with pathogen abundance. Ultimately, our study demonstrates the potential of untargeted metatranscriptomics for in depth analysis of the etiology of AS, comprehensive host-response profiling, and using these together to work towards optimized patient care.

Importance Acute otitis media (AOM) is a frequently diagnosed illness in children, yet the accuracy of diagnosis has been consistently low. Multiple neural networks have been developed to recognize the presence of AOM with limited clinical application. Objective To develop and internally validate an artificial intelligence decision-support tool to interpret videos of the tympanic membrane and enhance accuracy in the diagnosis of AOM. Design, Setting, and Participants This diagnostic study analyzed otoscopic videos of the tympanic membrane captured using a smartphone during outpatient clinic visits at 2 sites in Pennsylvania between 2018 and 2023. Eligible participants included children who presented for sick visits or wellness visits. Exposure Otoscopic examination. Main Outcomes and Measures Using the otoscopic videos that were annotated by validated otoscopists, a deep residual-recurrent neural network was trained to predict both features of the tympanic membrane and the diagnosis of AOM vs no AOM. The accuracy of this network was compared with a second network trained using a decision tree approach. A noise quality filter was also trained to prompt users that the video segment acquired may not be adequate for diagnostic purposes. Results Using 1151 videos from 635 children (majority younger than 3 years of age), the deep residual-recurrent neural network had almost identical diagnostic accuracy as the decision tree network. The finalized deep residual-recurrent neural network algorithm classified tympanic membrane videos into AOM vs no AOM categories with a sensitivity of 93.8% (95% CI, 92.6%-95.0%) and specificity of 93.5% (95% CI, 92.8%-94.3%) and the decision tree model had a sensitivity of 93.7% (95% CI, 92.4%-94.9%) and specificity of 93.3% (92.5%-94.1%). Of the tympanic membrane features outputted, bulging of the TM most closely aligned with the predicted diagnosis; bulging was present in 230 of 230 cases (100%) in which the diagnosis was predicted to be AOM in the test set. Conclusions and Relevance These findings suggest that given its high accuracy, the algorithm and medical-grade application that facilitates image acquisition and quality filtering could reasonably be used in primary care or acute care settings to aid with automated diagnosis of AOM and decisions regarding treatment.