Pulmonary nodules are defined as round, oval or irregular lesions, surrounded by lung parenchyma, with a diameter of ≤3cm (1), and can be ground glass, solid, or contain both solid and ground glass components (2). Pulmonary nodules are more common in adults because primary lung cancer is one of the most common cancers in this population. The etiology of pediatric pulmonary nodules differs from that in adults, and often lack specificity. Common causes in pediatric pulmonary nodules include infectious lung diseases, pulmonary metastases, connective tissue diseases and other less common etiologies (such as chronic granulomatous disease, pulmonary nodulosis and multiple pulmonary arteriovenous fistulas) (3–5).

In recent years, the widespread use of chest computed tomography (CT) has made an increase in the detection of pediatric pulmonary nodules. The prevalence of incidental lung nodules in healthy children is not low. Alves et al. (6) evaluated preoperative thoracic CT scans of 99 children who were scheduled for surgery for pectus carinatum or excavatum and found the detection rate of nodule was 75%. In 2017, Samim and colleagues found a detection rate of pulmonary nodules on CT scan was 38.0% in pediatric patients after trauma (7). Similarly, researchers (8) retrospectively reviewed 259 trauma chest CT scans in pediatric patients and demonstrated a total of 86 patients (33%) with incidental pulmonary nodules. A large proportion of the studies revealed that pulmonary nodules are also not uncommonly identified in the pediatric population with extra thoracic malignancies. Silva and colleagues (9) reviewed 488 patients with a noncentral nervous system solid tumor or lymphoma and identified 111 patients (23%) with pulmonary nodules on the presentation CT scan. Recently, a study (10) reported more than one-fifth had pulmonary nodules in 316 pediatric patients with rhabdomyosarcoma. Another study discovered that 57.0% of the nodules detected by CT were malignant in children with osteosarcoma (11). These existing studies evaluated features of pulmonary nodules in children, but there is still no consensus on reliable features to determine a benign incidental pulmonary nodule versus metastatic disease. Therefore, the pulmonary nodule in pediatric population is not uncommonly encountered and becomes a challenging situation.

There were few accepted criteria for pediatric pulmonary nodule evaluation. In 2014, an American study found that 40% of pediatric pulmonologists referred to the Fleischner guidelines (12). This guideline aims at patients ≥35 years old, indicating a tendency towards overtreatment of pediatric pulmonary nodules (13). The development of standard guidelines for pediatric pulmonary nodules is an urgent demand (14). Although lung biopsy is the gold standard for nodule diagnosis, it is not the first choice due to its invasiveness. Chest CT is the first line guide for determining benignity or malignancy (6, 8, 15, 16). Our study is designed to provide a reference for clinical diagnosis, treatment, and management of pulmonary nodules for children.

This study retrospectively analyzed the clinical and radiological characteristics of pulmonary nodules in 1341 children. The detection rate of pediatric pulmonary nodules was low. Children with the history of malignancies, nodule diameter ≥5mm, and specific nodule distribution (multiple in both lungs, multiple in the right lung, solitary in the upper or middle right lobe) were important predictable factors for malignity. Clinical and radiological dynamic follow-up was important for those patients.

The etiological spectrum of pulmonary nodules in children is different from that in adults. Clarifying the causes of pulmonary nodules is a challenge in their diagnosis and treatment. Previous studies have shown that many causes were associated with pediatric pulmonary nodules, such as infectious diseases, hematological disorders, autoimmune diseases, granulomatous diseases, obliterative bronchiolitis, pulmonary nodulosis, Crohn’s disease (17–19). In our finding, the primary clinical diagnoses before examination were malignant tumors, including hepatoblastoma, neuroblastoma, nephroblastoma, osteosarcoma, leukemia, lymphoma, which was consistent with previous study (3). We found The most common non-tumor cause was infection (including respiratory infections and other systemic infectious diseases). We also revealed that most pulmonary nodules in children are incidental. Only 17.3% of all children showed respiratory symptoms such as coughing, wheezing, shortness of breath or chest pain before CT scanning. In addition, children with a history of malignant tumors were younger, showed larger nodule diameters, and presented with multiple nodules often accompanied by osteolytic lesions and mass effect, which is consistent with previous findings (7, 8, 20).

The differentiation and etiological diagnosis of benign versus malignant pulmonary nodules is crucial in clinical management. Guidelines for adult pulmonary nodules offer detailed assessment and clinical management plans for physicians. Generally, nodules<6 mm are considered as low probability of malignancy (18), and nodules>5mm are recommended for clinical follow-up, those with solid nodules should undergo annual follow-up (21). Lung biopsy is the golden standard method for defining the nature of nodules and widely used in adult (22). One study (23) found that lung biopsy results led to changes in the management of 63% of children, particularly affecting the use of steroids. But biopsy is not frequently performed in children due to its invasiveness and parents’ worries. Therefore, the clinical features and radiological presentations of children are important clue to nodule assessment. In 2015, American Society for Pediatric Radiology’s Imaging Committee published the first guide for pediatric pulmonary nodules offering individualized management suggestions based on the CT characteristics and clinical history features (24, 25), but it lacks clear follow-up guidelines and criteria for judging benignity or malignancy. Several studies show that a history of malignant tumors, nodule diameter size, distribution, and margin can help in judging the nature of the nodule (8, 9, 15, 17, 26). Similarly, our decision tree model suggested that a nodule diameter≥5mm, a history of malignancies and nodule distribution are important predictors of malignancy. Multiple nodules also have predictable effect on malignancy (27). In our findings, not only multiple lesions (in the right lung or both lungs), solitary lesions in the upper or middle right lobe were also suggestive of malignancy.

Our study has some limitations. The selection bias exists due to the single-center study. Multicenter studies are highly recommended in the future with larger sample size to reduce bias. In addition, we failed to collect lung biopsy report of nodules in these cases, this reduced the reliability of the model. Biopsy is invasive with potential complications. Pathological evidence is hard to obtain when patient shows no indication for biopsy or surgery. We believe that further studies or meta-analysis are necessary to confirm our preliminary findings in the future.

In conclusions, our study demonstrated a history of malignancies, nodule diameter ≥5cm in size and distribution were potential predictors for malignant nodules. We suggested that children with potential malignant tumors have a higher probability of malignancy in pulmonary nodules. These patients require frequent follow-up and an early pulmonary nodule diagnosis.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

The studies involving humans were approved by Institutional Review Board of the Shanghai Children’s Medical Center, Shanghai Jiao Tong University (SCMCIRB-K2022085-2). The studies were conducted in accordance with the local legislation and institutional requirements.

MD: Conceptualization, Data curation, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. SY: Data curation, Formal Analysis, Methodology, Writing – review & editing. HJ: Data curation, Supervision, Writing – original draft. YW: Data curation, Supervision, Writing – original draft. HM: Data curation, Supervision, Writing – original draft. SS: Data curation, Software, Supervision, Writing – original draft. JL: Conceptualization, Data curation, Formal Analysis, Methodology, Software, Supervision, Validation, Writing – review & editing. JC: Conceptualization, Methodology, Software, Supervision, Validation, Writing – review & editing. YY: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – review & editing, Writing – original draft.