Abstract: Background：Pulmonary nodules are increasingly detected in healthy children while the relevant research are limited. There are still problems and challenges in diagnosis and treatment. Objective：To retrospectively review the clinical characteristics, imaging features, and prognosis of pulmonary nodules in children, and to provide a reference for the clinical diagnosis and treatment. Design：A case series report. Methods：Consecutive cases of children aged <18 years old, diagnosed with pulmonary nodules, and followed up for ≥6 months in the Department of Respiratory Medicine from January 1, 2015, to December 30, 2021, at Children's Hospital of Fudan University, were retrospectively collected. Children with malignancy, immunodeficiency disease, tuberculosis, and congenital pulmonary airway malformation were excluded. Data were extracted from outpatient medical records, including gender, age at the first identification, visit date, indication for imaging, the type of imaging leading to the diagnosis, combining or accompanying diseases, final diagnosis, and changes of repeated chest imaging. Main outcome measures：The detection rate of pulmonary nodules and followup outcomes. Results：A total of 74 children with pulmonary nodules were included in the analysis. The average age was (8.7±3.9) years old, 44 cases (59.5%) were older than 8 years, 49 cases (66.2%) were male, and 67 cases (90.5%) were detected by chest CT. Fiftyone patients had complete image data. Nodule size was <5 mm in 32 patients, 510 mm in 10 patients, and >10mm in 9 patients. Eighteen cases had bilaterally distributed nodules and 33 cases had unilaterally distributed nodules. There were 24 cases with a single nodule, 22 cases with 2 to 10 nodules, and 5 cases with over 10 nodules. Nodule shapes were mainly round/quasi round (29 cases) and mixed shaped (16 cases), and the edge was mainly smooth (41 cases). Solid nodules were found in 37 cases, mixed in 7 cases, ground glass in 5 cases, and partially solid in 2 cases. A total of 39 cases were followedup with CT. The median time of the first followup was 3.3 (1, 6) months. The nodules were shrunk or decreased in 14 cases (35.9%), remained unchanged in 9 cases (23.1%), disappeared in 8 cases (20.5%), and enlarged or increased in 8 cases (20.5%). For the 8 patients with enlarged or increased nodules, the mean age at the first identification was (8.1±3.0) years old, with the median nodule size of 5 (2.5, 15) mm. In 7 cases with image data in our hospital, 3 cases with a single nodule, 1 case with 2 to 10 nodules, 1 case with 10 to 20 nodules, and 2 cases with >20 nodules. Three had round/quasi round nodules, and 4 had mixedshaped nodules. Nodules with smooth margin were seen in 4 cases, with poorly defined, halo sign, and spiculated margin were seen in one case each. Nodules were solid in 6 cases and mixed in one case. Five patients underwent lung biopsies (all nodules size >10 mm). Eight patients were finally diagnosed, 4 cases were confirmed by lung biopsy (2 pulmonary sarcoidosis, 1 anaerobic infection, and 1 inflammatory myofibroblastic tumor), and the other 4 cases were clinically diagnosed (3 pulmonary infection, 1 caused by trauma). Sixtysix patients were undiagnosed, 8 of which had a clinical proposed diagnosis including pulmonary infection in 4 cases, granulomatous disease, pulmonary spaceoccupying lesions, connective tissue disease, and vascular disease each in one case. Fiftyeight cases did not have a definitive diagnosis. Conclusion：The potential detection rate of pulmonary nodules in children deserves attention. The overall diagnosis rate is low, and dynamic imaging followup is required. Biopsy should be considered in patients whose nodules are >10mm and without a definitive diagnosis by routine investigation.

Key words: Pulmonary nodule, Pediatric, Computed tomography