Background The rapid development in big data analytics and the data-rich environment of intensive care units together provide unprecedented opportunities for medical breakthroughs in the field of critical care. We developed and validated a machine learning-based model, the Pediatric Risk of Mortality Prediction Tool (PROMPT), for real-time prediction of all-cause mortality in pediatric intensive care units. Methods Utilizing two separate retrospective observational cohorts, we conducted model development and validation using a machine learning algorithm with a convolutional neural network. The development cohort comprised 1445 pediatric patients with 1977 medical encounters admitted to intensive care units from January 2011 to December 2017 at Severance Hospital (Seoul, Korea). The validation cohort included 278 patients with 364 medical encounters admitted to the pediatric intensive care unit from January 2016 to November 2017 at Samsung Medical Center. Results Using seven vital signs, along with patient age and body weight on intensive care unit admission, PROMPT achieved an area under the receiver operating characteristic curve in the range of 0.89–0.97 for mortality prediction 6 to 60 h prior to death. Our results demonstrated that PROMPT provided high sensitivity with specificity and outperformed the conventional severity scoring system, the Pediatric Index of Mortality, in predictive ability. Model performance was indistinguishable between the development and validation cohorts. Conclusions PROMPT is a deep model-based, data-driven early warning score tool that can predict mortality in critically ill children and may be useful for the timely identification of deteriorating patients. Electronic supplementary material The online version of this article (10.1186/s13054-019-2561-z) contains supplementary material, which is available to authorized users.

Background: Community-acquired pneumonia (CAP) is one of the leading worldwide causes of childhood morbidity and mortality. Its disease burden varies by age and etiology and is time dependent. We aimed to investigate the annual and seasonal patterns in etiologies of pediatric CAP requiring hospitalization. Methods: We conducted a retrospective study in 30,994 children (aged 0-18 years) with CAP between 2010 and 2015 at 23 nationwide hospitals in South Korea. Mycoplasma pneumoniae (MP) pneumonia was clinically classified as macrolide-sensitive MP, macrolide-less effective MP (MLEP), and macrolide-refractory MP (MRMP) based on fever duration after initiation of macrolide treatment, regardless of the results of in vitro macrolide sensitivity tests. Results: MP and respiratory syncytial virus (RSV) were the two most commonly identified pathogens of CAP. With the two epidemics of MP pneumonia (2011 and 2015), the rates of clinical MLEP and MRMP pneumonia showed increasing trends of 36.4% of the total MP pneumonia. In children < 2 years of age, RSV (34.0%) was the most common cause of CAP, followed by MP (9.4%); however, MP was the most common cause of CAP in children aged 2-18 years of age (45.3%). Systemic corticosteroid was most commonly administered for MP pneumonia. The rate of hospitalization in intensive care units was the highest for RSV pneumonia, and ventilator care was most commonly needed in cases of adenovirus pneumonia.

Background: Bronchiectasis is a chronic pulmonary disease characterized by progressive and irreversible bronchial dilatation. The aim of the present study was to investigate the etiologies and clinical features of bronchiectasis in Korean children. Methods: We performed a retrospective review of the medical records for children diagnosed with bronchiectasis between 2000 and 2017 at 28 secondary or tertiary hospitals in South Korea. Results: A total of 387 cases were enrolled. The mean age at diagnosis was 9.2 ± 5.1 years and 53.5% of the patients were boys. The most common underlying cause of bronchiectasis was preexisting respiratory infection (55.3%), post-infectious bronchiolitis obliterans (14.3%), pulmonary tuberculosis (12.3%), and heart diseases (5.6%). Common initial presenting symptoms included chronic cough (68.0%), recurrent pneumonia (36.4%), fever (31.1%), and dyspnea (19.7%). The most predominantly involved lesions were left lower lobe (53.9%), right lower lobe (47.1%) and right middle lobe (40.2%). No significant difference was observed in the distribution of these involved lesions by etiology. The forced expiratory volume in 1 s (FEV 1 ) levels were lowest in cases with interstitial lung disease-associated bronchiectasis, followed by those with recurrent aspiration and primary immunodeficiency. Conclusions: Bronchiectasis should be strongly considered in children with chronic cough and recurrent pneumonia. Long-term follow-up studies on pediatric bronchiectasis are needed to further clarify the prognosis and reduce the disease burden in these patients.

Since mid-April 2020, cases of multisystem inflammatory syndrome in children (MIS-C) associated with coronavirus disease 2019 that mimics Kawasaki disease (KD) have been reported in Europe and North America. However, no cases have been reported in Korea. We describe an 11-year old boy with fever, abdominal pain, and diarrhea who developed hypotension requiring inotropes in intensive care unit. His blood test revealed elevated inflammatory markers, thrombocytopenia, hypoalbuminemia, and coagulopathy. Afterward, he developed signs of KD such as conjunctival injection, strawberry tongue, cracked lip, and coronary artery dilatation, and parenchymal consolidation without respiratory symptoms. Microbiological tests were all negative including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reverse transcription polymerase chain reaction. However, serum immunoglobulin G against SARS-CoV-2 was positive in repeated tests using enzyme-linked immunosorbent assay and fluorescent immunoassay. He was recovered well after intravenous immunoglobulin administration and discharged without complication on hospital day 13. We report the first Korean child who met all the criteria of MIS-C with features of incomplete KD or KD shock syndrome.