Preliminary investigation of the clinical value of thoracic ultrasound in pediatric pleuropulmonary blastoma: a comparative analysis with computed tomography and pathology

Abstract

Objective

This study aimed to preliminarily characterize the ultrasonographic features of pediatric pleuropulmonary blastoma (PPB). With surgical–pathological findings serving as the gold standard, the imaging manifestations of thoracic ultrasound (US) and computed tomography (CT) in PPB were compared, and correlations between ultrasonic and pathological characteristics of this rare pediatric malignancy were explored. The diagnostic consistency and complementary value of the two imaging modalities were further evaluated, their clinical application scopes were clarified, and the potential value of thoracic US in the imaging evaluation of PPB was assessed, This study provided a preliminary reference for the optimization of clinical diagnostic strategies for PPB.

Methods

A retrospective analysis was performed on the clinical data of 12 pediatric patients with surgically–pathologically confirmed PPB who underwent preoperative thoracic US and CT at the Children’s Hospital Affiliated to Shandong University between October 2017 and November 2024. The sonographic and CT imaging manifestations of the lesions were summarized and documented in accordance with unified diagnostic criteria. The diagnostic consistency and differences between the two imaging modalities were further explored. A standardized ultrasound–pathology double-blind design was adopted to observe the correspondence between ultrasonic features and pathological findings.

Results

Among the 12 pediatric patients with PPB, thoracic US identified intrathoracic masses (12/12), atelectasis (10/12), pleural effusion (11/12), and pleural thickening (5/12), while CT detected intrathoracic masses (12/12), atelectasis (11/12), pleural effusion (11/12), and pleural thickening (6/12). No substantial difference was found between the two modalities for these findings. In this study cohort, the two modalities exhibited potential complementary strengths for PPB lesion assessment: CT tended to show relative advantages in detecting pneumatocele, microbubble-containing pneumothorax lesions involving tracheal carina and main bronchi, deep mediastinal invasion, and supradiaphragmatic lymph node metastasis; which are less optimally visualized by US. In comparison, US appeared to offer clearer delineation of pericardial effusion, tumor invasion into the left cardiac cavity, tumor attachment site, and its anatomical relationship with cardiac valve structures. A certain correlation was observed between the ultrasonic manifestations of PPB and its pathological features. Follow-up confirmed that US was suitable for repeated dynamic monitoring during long-term clinical follow-up.

Conclusion

Thoracic US can act as a feasible imaging modality for initial screening, dynamic monitoring and long-term postoperative follow-up of pediatric PPB. Combined US and CT may partially compensate for the limitations of single-modality imaging and provide complementary diagnostic value, potentially improving overall evaluation accuracy. Proper use of thoracic US may reduce unnecessary radiation exposure in children and provide reliable imaging evidence for individualized clinical diagnosis and management. Pathological examination is regarded as the reference standard for the definitive diagnosis of pediatric PPB.

Introduction

Pleuropulmonary blastoma (PPB) is a rare, invasive, and malignant thoracic tumor exclusive to children, primarily arising from the lung and/or pleural tissues [1, 2]. PPB is believed to arise from the visceral pleura (visceral pleural precursor) or somatic pleura (parietal pleural precursor) during early embryonic development [3]. Recent studies have classified this tumor as part of the DICER1 syndrome or the familial PPB tumor and dysplasia syndrome [4]. Approximately 66% of affected patients harbor heterozygous mutations in the DICER1 gene located at chromosome 14q13.2 [5].

In 1961, Spencer first coined the term “pleuropulmonary blastoma” and proposed that the tumor originates from the mesodermal blastema [6]. In 1988, Manivel et al. demonstrated that PPB exhibits a distinct biphasic epithelial-stromal morphology, which differs from that in adult [7]. In the 2015 World Health Organization (WHO) classification of lung tumors, PPB is categorized under mesenchymal tumors [8]. Based on histological features, PPB is classified into type I (cystic), type II (cystic-solid), and type III (solid) [4], with notable prognostic differences among these subtypes, which provides a reliable basis for clinical evaluation and prognostic stratification. The 5-year survival rates are approximately 91%, 71%, and 53%, respectively [5, 9]. The subtype Ir, introduced in 2006, is regarded as a regressive form or arrested progression of the previously defined type I PPB [10].

PPB is characterized by nonspecific clinical manifestations, with no typical symptoms in the early stage. Some children only present with mild cough, chest tightness, and exertional dyspnea, which are easily misdiagnosed as common pediatric pulmonary conditions such as pneumonia and pleurisy. Misdiagnosis and missed diagnosis are common in clinical practice, leading to delayed optimal treatment. Thus, early accurate screening and timely definitive diagnosis are important for improving the prognosis of affected children.

Currently, imaging examination is the core modality for the early screening, disease assessment, and follow-up monitoring of pediatric PPB. Among these imaging modalities, computed tomography (CT) has long been the primary method for PPB diagnosis due to its superior spatial resolution. Nevertheless, CT has several limitations, including immobility, ionizing radiation exposure, and relatively high examination costs. Furthermore, sedation is required for young and uncooperative children, which further increases the examination-related risks, especially for long-term follow-up. In recent years, with the advancement and refinement of ultrasound(US) technology, thoracic US has gradually become one of the commonly utilized modalities for pediatric pulmonary diseases, owing to its favorable attributes of reasonable cost, portability, real-time dynamic imaging, and repeatability without the need for sedation. At our institution (Children’s Hospital Affiliated to Shandong University), thoracic US has been routinely adopted as a primary screening and supplementary imaging tool for neonates and pediatric patients with pulmonary diseases. This technique is applied in the assessment of various thoracic conditions, including pneumonia, lung abscess, pneumothorax, atelectasis, interstitial pulmonary edema, respiratory distress syndrome, pleural lesions, and pleural effusion. It is also suitable for the initial evaluation and long-term follow-up of congenital malformations, as well as benign and malignant thoracic lesions.

However, systematic studies on the diagnostic performance, characteristic ultrasonic findings, and clinical utility of US in PPB remain scarce. While thoracic US holds potential as a low-risk imaging modality for children, several unresolved questions in this area warrant further exploration: first, the consistency and discrepancies between thoracic US and CT in identifying major PPB-associated lesions (e.g., intrathoracic mass, pleural effusion, atelectasis, vesicular pneumothorax, carina involvement, cardiac cavity invasion, pleural thickening, and metastases) have not been elucidated; second, the association between specific ultrasonic features (e.g., cystic-solid composition, echogenicity, calcification, and margin characteristics) and the pathological attributes of PPB (e.g., tissue components, hemorrhage and necrosis, mural nodules, and capsular formation) lacks clinical validation; third, the practical utility of thoracic US in specific clinical scenarios for PPB (e.g., initial screening, dynamic monitoring, and postoperative follow-up) has not been supported by empirical evidence.

To address these unresolved questions, the present study conducted a retrospective analysis of 12 children with pathologically confirmed PPB at our hospital. The main analytical components included summarizing the characteristic ultrasonic features of PPB, comparing the detection performance of thoracic US and CT for major PPB-associated lesions, delineating the respective strengths and limitations of the two imaging modalities, clarifying their clinical application scopes, and analyzing the association between ultrasonic features and pathological characteristics of PPB. This study aimed to preliminarily explore the clinical utility of thoracic US in the screening, disease evaluation, and long-term follow-up of pediatric PPB, thereby providing a theoretical basis and practical reference for the clinical application of safe and appropriate imaging approaches in the assessment of PPB.

Materials and methods

General data

This study adopted a retrospective design. The study subjects were selected based on strict inclusion and exclusion criteria, which are detailed as follows:

• Inclusion Criteria: (1) Definite diagnosis of PPB confirmed by postoperative histopathological examination; (2) Complete preoperative thoracic US and chest CT imaging data sufficient to support paired imaging-pathology correlation analysis; (3) Complete clinical medical records, surgical records, and pathological reports with no missing critical data; (4) Pediatric patients aged ≤ 8 years with thoracic space-occupying lesions who were admitted to Children’s Hospital Affiliated to Shandong University.

• Exclusion Criteria: (1) Unclear pathological diagnosis or failure to confirm PPB via histopathological examination; (2) Missing thoracic US or CT imaging data, or poor image quality that precludes reliable radiological evaluation; (3) Incomplete clinical medical records, surgical records, or pathological reports that hinder comparative analysis of study results; (4) Recurrent PPB cases referred from other hospitals without available baseline imaging and clinical data at our center.

Based on the above inclusion and exclusion criteria, 12 children who were definitively diagnosed with PPB by histopathological examination at our hospital between October 2017 and November 2024 were retrospectively enrolled. Their thoracic US features were then analyzed and compared with the corresponding thoracic CT findings and pathological characteristics.Complete clinical records, ultrasound and CT imaging data, surgical pathology results were available for all patients.

Clinical characteristics

Among the 12 children with pathologically confirmed PPB, 6 were male and 6 were female, ranging in age from 28 to 84 months, with a median age of 44.5 months. Patients presented with various respiratory infection-related symptoms, including fever, cough, chest tightness, chest pain, back pain, shortness of breath, and wheezing. Thoracic US and CT scans revealed intrathoracic masses in all cases, with 7 lesions located in the left hemithorax and 5 in the right hemithorax.

Examination procedures and pathological evaluation

Thoracic US

All ultrasound scanning and image acquisition procedures were performed by physicians with a minimum of 2 years of clinical experience in pediatric thoracic ultrasound. All acquired images were further reviewed by associate chief physicians specialized in pediatric ultrasound to ensure adequate image quality for standardized interpretation.

A Philips EPIQ7 color Doppler US diagnostic system (Philips Ultrasound, Inc., Bothell, WA, USA) equipped with a 5–12 MHz linear array probe and a 3–5 MHz convex array probe was used for all examinations. According to each child’s clinical condition, patients were placed in supine, sitting, or prone positions with full exposure of the anterior chest, axillae, lateral chest wall, and posterior thorax. Sequential, uninterrupted scanning was performed on all 12 bilateral lung zones from multiple planes and angles.

Predefined standardized criteria were applied for assessment of ultrasonic lesion features. (1) Morphology: lesions were categorized as regular (smooth contour, well-defined margin, and uniform configuration) or irregular (asymmetric structure, irregular contour, and ill-defined margin). (2) Size: three-dimensional dimensions, including maximal longitudinal, transverse, and anteroposterior diameters, were measured using the built-in electronic caliper of the ultrasound device. All measurements were recorded in centimeters and presented as longitudinal × transverse × anteroposterior diameter. (3) Margin: classified as clear (distinct demarcation between the lesion and adjacent normal parenchyma) or ill-defined (blurred boundary between the lesion and surrounding tissues). (4) Echogenicity: grouped into five categories: hypoechoic, isoechoic, hyperechoic, mixed echo, and anechoic. (5) Adjacent tissue involvement: defined as lesion-associated compression, displacement, or infiltration of surrounding tissues and organs, with possible secondary pleural thickening. (6) Pleural effusion: typically presented as anechoic or hypoechoic fluid accumulation within the pleural space; effusions containing hemorrhagic or necrotic components may demonstrate floating punctate or flocculent echoes with positional mobility. (7) Atelectasis: typically presented as wedge-shaped or strip-like homogeneous hypoechoic or solid echoes, with reduced or absent pulmonary aeration, indistinct or absent air bronchograms, and restricted respiratory mobility of the involved lung parenchyma.

Color Doppler flow imaging (CDFI) was used to evaluate intralesional vascular signals and lesion-related vascular changes, including vascular compression (extrinsic vessel compression by the lesion), vascular encasement (circumferential wrapping of major vessels by lesion tissue), and vascular penetration (major vessel traversal through the lesion parenchyma). All valid imaging data were archived in the hospital Picture Archiving and Communication System (PACS) for subsequent offline centralized analysis.

All ultrasound images were independently evaluated by two senior ultrasonologists with over 5 years of specialized experience in pediatric thoracic ultrasound, following the aforementioned standardized criteria. Any inter-observer discrepancies were resolved via consensus discussion with an additional ultrasound chief physician, to enhance the reliability and reproducibility of image evaluation.

CT examination

CT examinations were performed using a Philips Brilliance 256-slice spiral CT scanner (Philips Medical Systems (Cleveland), Inc., Cleveland, OH, USA) with children in the supine position. Midazolam or propofol sedation was administered to uncooperative children. The scanning range extended from the thoracic inlet to the inferior margin of the 12th thoracic vertebra (T12).

Plain CT scanning was performed prior to contrast-enhanced acquisition for preliminary assessment of lesion size, morphology, and density. Intravenous contrast-enhanced CT was performed using Iohexol as the contrast medium. Triple-phase (arterial, venous, and delayed-phase) enhanced scanning was performed when clinically appropriate. Multi-planar reconstruction (MPR) was processed on a dedicated post-processing workstation to facilitate comprehensive lesion evaluation.

Standardized CT interpretation criteria were predefined as follows. (1) Size: three-dimensional diameters were measured on axial, coronal, and sagittal reconstructed images and recorded in centimeters. (2) Morphology: classified as regular or irregular. (3) Density: defined as homogeneous with consistent attenuation or heterogeneous with mixed high and low attenuation, potentially accompanied by calcification, cystic change, or necrosis. (4) Enhancement pattern: categorized as non-enhancement, homogeneous enhancement, or heterogeneous enhancement. (5) Adjacent structural involvement: included lesion-induced displacement or invasion of the heart, mediastinum, diaphragm, major vessels, and bronchi, as well as secondary mediastinal lymphadenopathy and pleural thickening. (6) Pleural effusion: typically appeared as well-demarcated arcuate, band-like, or crescentic hypoattenuating lesions within the pleural cavity on plain CT. (7) Atelectasis: characterized by reduced local lung volume, wedge-shaped or fan-shaped soft-tissue attenuation, aggregated and disordered pulmonary markings, and homogeneous enhancement following contrast administration.

All CT images were independently reviewed by two senior radiologists with over 5 years of specialized experience in pediatric thoracic imaging. Any inter-observer interpretive differences were resolved through joint review and collective discussion with a chief physician specialized in pediatric thoracic imaging to reach a unified conclusion.

Pathological specimen evaluation

Surgically resected specimens were fixed in 4% formalin, embedded in paraffin, serially sectioned at 4 μm thickness, and stained with hematoxylin–eosin (HE) for routine histological examination. All pathological assessments were performed by a specialized pediatric pathologist with more than 5 years of diagnostic experience. The final pathological diagnosis was established based on the 2021 World Health Organization (WHO) classification criteria for pediatric lung tumors.

Microscopic evaluation was performed to assess tumor histological subtype, cellular morphology, and tissue composition, as well as secondary changes including hemorrhage, necrosis, calcification, and cystic degeneration. Tumor invasive features were also evaluated systematically, with particular attention paid to vascular invasion and involvement of the diaphragm, pericardium, and other adjacent structures.

Imaging–pathology correlation (standardized double-blind design)

A rigorous standardized double-blind design was adopted for imaging–pathology correlation analysis to minimize potential assessment bias. An independent third party, who was not involved in image interpretation or pathological diagnosis, performed de-identification of all imaging and pathological data by removing all personal identifiable information of enrolled patients. Each paired ultrasound image and corresponding pathological section was assigned a unique matched code (e.g., US-001/Path-001, US-002/Path-002) to ensure all evaluators were blinded to patient clinical information and mutual assessment outcomes.

Blinded to the pathological results, two specialized ultrasonologists independently recorded ultrasonic features of each lesion and marked the lesion location according to the unified ultrasound evaluation criteria. Meanwhile, the attending pediatric pathologist, who was blinded to all imaging findings, evaluated the matched pathological sections, documented histological manifestations, and annotated the corresponding lesion sites on the specimen slides.

After the completion of all independent evaluations, unblinding was uniformly performed by the third party. A one-to-one correlative analysis was performed between individual ultrasonic features and the corresponding histological manifestations. Any inter-evaluator discrepancies were resolved via joint review of original imaging data, pathological sections, and surgical records to achieve a consistent consensus. Finally, systematic analysis was performed to explore potential correlations between ultrasonic manifestations and pathological components, so as to preliminarily explore the potential predictive value of sonographic features for underlying histological changes.

Results

Pathological classification of PPB

Clinically, PPB is classified into three types (I, II, III).In the present study, there were 3 cases of type II and 4 cases of type III. Classification of the remaining 5 cases was not possible because tissue specimens were unavailable prior to the initiation of chemotherapy; however, all these cases exhibited pathologic features of PPB after chemotherapy.The clinicopathological and sonographic characteristics of the 12 patients with PPB are shown in Table 1.

Table 1.

The clinicopathological and sonographic characteristics of 12 pediatric patients with PPB

Case	Gender	Age (months)	Pathological Classification	Lesion Location	Lesion size (cm × cm × cm)	Intratumoral cystic degeneration	Calcification	Pleural effusion	Metastasis
1	Female	53	Ⅱ	LLL	14 × 11.7 × 9.2	+	-	+	+
2	Female	34	—	LLL	14.7 × 10.5 × 7.3	+	-	+	-
3	Male	28	—	RUL	10.9 × 10.5 × 9.5	+	+	+	+
4	Male	46	Ⅲ	LUL	3.7 × 3.0 × 2.7	-	-	+	-
5	Male	36	Ⅲ	RML	10.3 × 7.9 × 7.2	-	-	+	-
6	Female	31	—	LLL	5.3 × 4.8 × 4.6	+	-	-	+
7	Female	48	—	RUL	13.6 × 8.9 × 8.2	+	-	+	-
8	Male	84	Ⅱ	LLL	15.3 × 13.7 × 10.6	+	+	+	-
9	Female	42	Ⅱ	RUL	10.1 × 8.3 × 6.8	+	-	+	+
10	Male	47	Ⅲ	LUL	13.5 × 11.9 × 8.4	+	-	+	-
11	Female	56	—	RUL	10.7 × 10.4 × 6.6	+	-	+	+
12	Male	43	Ⅲ	LLL	12.9 × 12.4 × 12.3	+	-	+	+

LUL Left upper lobe, LLL Left lower lobe, RUL Right upper lobe, RML Right middle lobe.+, present; –, absent

Thoracic US findings

US detected solid or cystic-solid masses of varying sizes in the thoracic cavity in all 12 cases. The maximum diameter was 10–15 cm in 9 cases and ≤ 10 cm in 3 cases.The masses had clear or relatively clear margins and heterogeneous echogenicity, characterized by mixed medium-to-low echoes, and some exhibited a multinodular pattern. Scattered cystic anechoic areas or patchy fluid collections of various sizes were visible within the lesions, with fair or poor internal sound transmission; some areas showed septations or a reticular pattern. Small patchy areas of increased echogenity and punctate calcific foci were occasionally observed, with a low incidence (2/12 cases). CDFI demonstrated sparse short linear or patchy blood flow signals within the masses; no lesions showed abundant vascularity.

The masses were closely associated with the pleura, and 5 cases showed varying degrees of pleural thickening on US. Compressed and consolidated lung tissue echoes were seen around the masses in 10 cases, and air bronchograms were noted in some areas. Pleural effusion was present in 11 cases, appearing as fluid collections within the thoracic cavity, between the mass and consolidated lung, and in the costophrenic sulci.

Lesion boundaries with the chest wall and mediastinum were often indistinct. The lesions could compress or invade adjacent structures, causing displacement of the trachea, bronchi, mediastinum, diaphragm, heart, and great vessels. Vascular wall involvement was also observed. One case showed compression and narrowing of the superior vena cava, one case involved encasement of the aortic arch with luminal narrowing, and one case had the brachiocephalic vein coursing through the lesion with resulting compression and narrowing. One case each demonstrated gastric, diaphragmatic, hepatic, and splenic compression. Two cases were complicated with pericardial effusion.

US identified metastatic lesions in 5 cases: 1 case with tumor extension into the left heart (pathologically confirmed by tumor cells in the pericardium), 1 case of diaphragmatic involvement with pathologically confirmed tumor cells,1 case with mediastinal lymph node metastasis, and 2 cases of cervical lymph node metastasis.

CT findings

CT demonstrated solid or cystic-solid masses in the thoracic cavity with relatively well-defined borders and relatively regular morphology. On unenhanced CT, the lesions showed slight or heterogeneous hypoattenuation, with patchy and cystic areas of slightly lower attenuation, as well as multiple linear and punctate slightly hyperdense foci. Following contrast administration, the solid components of the lesions exhibited flocculent, progressive heterogeneous enhancement, with focal ring-like enhancement in some regions, while the cystic components showed no significant enhancement.

The cardiac silhouette, mediastinal great vessels, diaphragm, spleen, and left hepatic lobe were variably displaced by the masses. In one case, the lesion encased the distal aortic arch and a portion of the thoracic aorta, resulting in luminal narrowing. Adjacent lung parenchyma was compressed and atelectatic or partially atelectatic in 11 cases, manifesting as soft-tissue density. Pleural effusion was present in 11 cases, appearing as arcuate, patchy, or band-like fluid density around the lesions; microbubble-like gas foci were identified within the effusion in 2 cases. The lesions were closely associated with the chest wall, and localized pleural thickening was observed in 6 cases.

Metastatic lesions were detected in 6 cases: 2 cases of diaphragmatic involvement with pathologically confirmed tumor cells, 1 case of carinal metastasis, 1 case of metastasis within the left main bronchus, 1 case of mediastinal lymph node metastasis, and 1 case of multiple supradiaphragmatic lymph node metastases.

Comparison of differences in thoracic US and CT findings

Among 12 children with PPB, thoracic US identified intrathoracic masses (12/12), atelectasis (10/12), pleural effusion (11/12), and pleural thickening (5/12), whereas CT detected intrathoracic masses (12/12), atelectasis (11/12), pleural effusion (11/12), and pleural thickening (6/12). No substantial difference was found between the two modalities for these findings.Both modalities could effectively demonstrate the location, size, and general morphology of the tumors.

Regarding tumor metastasis, US identified diaphragmatic involvement (1/12), while CT demonstrated diaphragmatic involvement (2/12). Both US and CT detected mediastinal lymph node metastasis in 1 case (1/12).CT showed certain advantages in identifying pneumatocele or microbubble-containing pneumothorax(detected in 2/2 cases by CT vs. 0/2 by US), lesions involving the carina (1/1 case) and left main bronchus (1/1 case), as well as tumor encasement of the distal aortic arch and thoracic aorta (1/1 case). CT also performed better in depicting deep lymph node metastases such as supradiaphragmatic lymph node involvement (1/1 case), whereas US was relatively limited in evaluating these findings.In contrast, US allowed direct, real-time visualization of tumor extension into the left cardiac cavity (1/1 case) without contrast administration, and clearly illustrated the vascular relationship between the tumor and adjacent cardiac vessels on CDFI. In addition, US initially detected cervical lymph node metastases in 2 cases, providing valuable information for the clinical assessment of superficial metastatic disease.

Pathological findings

Macroscopic appearance

Most tumors were encapsulated or appeared separated from the surrounding tissues. Cross-sections showed a grayish-white, soft, and friable consistency, which was either solid or cystic-solid. Some tumor tissues exhibited hemorrhage and necrosis (appearing purplish-black), while others were mucinous, with necrotic tissue lining the inner wall. In type Ⅱ cases, multiple nodules were attached to the cyst wall, and in individual cases, cartilage tissue was observed in localized areas of the cyst wall.

Pathological diagnosis

PPB was confirmed as a mesenchymal-derived malignant tumor. Tumor cells were oval or short spindle-shaped, with foamy cells present in some areas. Focally, a small number of poorly differentiated embryonic mesenchymal cells and rhabdomyoblasts were identified. Most tumor tissues showed old hemorrhage and necrosis, with proliferating fibrous granulation tissue and collagenization around the necrotic areas, and occasional calcification. Residual tumor tissue after chemotherapy consisted of pleomorphic rhabdomyosarcoma with marked atypia and scattered cartilage, showing chondrosarcoma-like differentiation. Hilar lymph nodes showed reactive hyperplasia, and in individual cases, tumor invasion of vascular walls was evident. In metastatic cases, tumor cells were detectable at the diaphragmatic margin and within the pericardium.

Thoracic US findings and pathological correlation analysis

A certain degree of correspondence was observed between the ultrasonic features of PPB and its pathological characteristics, suggesting that US may help reflect the internal pathological structure of the tumor to a certain extent.

Echogenicity patterns and pathological components

PPB exhibited heterogeneous echogenicity on US, which may be related to the histological composition of the tumor, including primitive mesenchymal and epithelial components, as well as frequent intratumoral hemorrhage and necrosis. Anechoic areas with poor sound transmission and a reticular appearance on US were mostly associated with myxoid degeneration. Patchy hyperechoic areas were related to pathological calcifications secondary to collagenization within tumor necrotic regions.

Tumor margins and pathological capsule

Most lesions showed distinct or relatively distinct margins on US, likely attributable to the presence of a complete capsule in most tumors. Some lesions without a complete capsule were demarcated from surrounding tissues in a dissecting manner with relatively clear borders, which accorded with gross pathological findings.

Multinodular pattern and pathological nodules

The multinodular pattern identified on US largely corresponded with multiple nodular protrusions attached to the cyst wall observed pathologically, which likely arose from localized abnormal proliferation of tumor cells along the cyst wall.

Other imaging-pathological correlations

In several individual cases, slightly hyperechoic areas on US corresponded to small amounts of cartilage tissue in pathological sections. When US suggested possible tumor invasion of blood vessels, diaphragm, or pericardium, the findings largely accorded with pathological results, indicating that US may allow preliminary assessment of the local invasive extent of PPB to a certain degree.

Clinical course and follow-up information

This retrospective study enrolled 12 pediatric patients with pathologically confirmed PPB. All patients were followed up at our hospital, with the follow-up cutoff date being November 2025. Metastasis occurred in 6 pediatric patients.The duration and frequency of follow-up varied among individuals.

• i.Follow-up duration:

• Range: 4–105 months

• Mean follow-up duration: 31.67 months

• Median follow-up duration: 23.5 months

• ii.Number of US examinations during follow-up

• Range: 3–28

• Mean number of US examinations: 13.83

• Median number of US examinations: 15.5

• iii.Recurrence

Among the 12 cases, 2 patients experienced tumor recurrence(16.67%).

One patient developed recurrence 6 years after surgery, with the primary lesion in the left upper lobe (LUL) and the recurrent lesion in the left lower lobe (LLL). The other patient developed recurrence 11 months after surgery, with the primary lesion in the right upper lobe (RUL) and the recurrent lesion in the right middle lobe (RML). The follow-up characteristics of the 12 patients with PPB are shown in Table 2.

Table 2.

The follow-up characteristics of 12 pediatric patients with PPB

Case	Follow-up duration (months)	Number of follow-up US Examinations	Recurrence during follow-up
1	4	3	None
2	80	18	None
3	6	5	None
4	5	4	None
5	40	22	None
6	24	14	None
7	39	3	None
8	105	15	Recurrence after 6 years
9	24	17	None
10	23	21	None
11	21	28	Recurrence after 11 months
12	9	16	None

Discussion

PPB is a relatively rare malignant pulmonary tumor in children, accounting for approximately 0.25%–0.5% of all primary pediatric lung tumors. It predominantly occurs in children under 6 years of age, with no significant sex predilection [4]. Among the 12 children enrolled in this study, 11 were younger than 6 years and the male-to-female ratio was 1:1, which was generally consistent with the clinical features reported in previous studies, suggesting that the present case series reflects the clinical characteristics of pediatric PPB to a certain extent. Approximately 10% of type I PPB cases may eventually progress to type II or III disease [5]. In our case series, patients ranged from 2 to 7 years old, and no type I PPB was identified. This may be explained by the relatively long disease course, atypical early clinical manifestations, and delayed clinical presentation, with tumors having already progressed to type II or III at diagnosis. To some extent, this finding also indicates the existing challenges in the early recognition and diagnosis of pediatric PPB.

This study’s imaging-pathological correlation analysis revealed that type Ⅱ and type Ⅲ PPB exhibit certain differences in ultrasonographic morphological characteristics between subtypes, and their ultrasonic discrepancies are associated with the diverse histopathological growth patterns of the tumors. Type II PPB mostly presented as cystic-solid masses, often with a multinodular pattern and internal cystic areas, which corresponded pathologically to intralesional hemorrhage, necrosis, and nodular protrusions along the cyst walls, with US, CT, and pathological findings of type II PPB illustrated in Fig. 1. In comparison, type III PPB predominantly presented as solid masses with heterogeneous internal echotexture, and focal punctate calcifications were detected in some lesions. Pathological evaluation of type III PPB lesions in the present cohort identified atypical rhabdoid components, accompanied by focal cartilaginous tissue and microcalcifications, which appeared to be in general agreement with previous literature [11]. Figure 2 depicts the US, CT, and pathological manifestations of type III PPB. In addition, ultrasonographic assessment of tumour boundary, morphology and adjacent structural invasion tended to align with pathological findings. Overall, US may reflect the pathological structure and local growth characteristics of PPB to a certain extent.

Fig. 1.

US, CT and pathological features of case 1 (type II PPB). Type II. a–c US revealed a large cystic-solid mass in the left thoracic cavity. Some cystic spaces showed poor acoustic transmission with dense punctate echoes, while the solid component displayed heterogeneous echogenicity with patchy hyperechoic areas. d CT demonstrated a large mass-like soft-tissue density lesion in the left thoracic cavity with heterogeneous density, featuring patchy areas of relatively low density and punctate high-density foci. Contrast-enhanced CT showed heterogeneous enhancement. e Spindle tumor cells were arranged in fascicles, exhibiting a fibrosarcoma-like morphology. f Immunohistochemical staining revealed nuclear positivity for Myogenin (original magnification, medium power)

Fig. 2.

US, CT and pathological features of case 2 (type III pleuropulmonaryblastoma). Type III. a US revealed a solid, moderately hyperechoic multinodular mass in the left thoracic cavity with relatively well-defined margins and heterogeneous echogenicity. b CDFI demonstrated linear vascular signals within the mass. c Compressed and consolidated lung parenchyma was seen adjacent to the pulmonary hilum, accompanied by an anechoic collection with poor acoustic transmission in the left costophrenic sinus. d CT showed a large mass in the left thoracic cavity with heterogeneous attenuation and focal patchy hypodense areas. e Histopathologically, the epithelial cells showed no atypia, with primitive mesenchymal cells observed beneath the epithelial layer. f Immunohistochemical staining revealed nuclear positivity for Myogenin (original magnification, medium power)

Although this study summarized the characteristic ultrasonic manifestations of PPB, these signs are not unique to PPB, and the US appearances of various pediatric thoracic tumors overlap considerably in clinical practice, rendering preoperative differentiation by US alone challenging; definitive diagnosis still relies on pathological examination. Therefore, clinical differential diagnosis should include the following entities: neuroblastoma usually presents as a hypervascular solid mass often accompanied by variable calcifications and is closely associated with paraspinal nerve roots, whereas in this study, PPB showed poor vascularity and rare calcification, with lesions arising from the lung and/or pleura; yolk sac tumor predominantly occurs in the anterosuperior mediastinum and is usually accompanied by significantly elevated serum alpha-fetoprotein (AFP); metastatic Wilms’ tumor can often be traced to a primary renal lesion, with approximately 80% of cases presenting with pulmonary metastases [12], while individuals with DICER1 mutations carry a relatively higher risk of developing tumors such as Wilms’ tumor [13]; lymphoma typically manifests as homogeneous hypoechoic lesions with relatively rich blood supply, usually without obvious cystic degeneration. Thus, the differential diagnosis of PPB requires comprehensive evaluation combining US features, clinical history, laboratory examinations, and pathological results.

In the present study, CT well delineated tumor location, size, morphology, and extent of invasion, and exhibited higher sensitivity for detecting central airway involvement, deep mediastinal invasion, pneumothorax, and deep supradiaphragmatic lymph node metastases. Computed tomographic angiography (CTA) also helped visualize tumor invasion into the distal aortic arch and thoracic aorta. However, the use of CT in children is associated with inherent limitations: as pediatric tissues are immature and considerably more radiosensitive than adult tissues, repeated CT examinations, including both unenhanced and contrast‑enhanced scans, increase cumulative radiation exposure and potential long-term carcinogenic risks; additional limitations include the requirement for sedation in young or uncooperative children, which increases the risk of sedation-related adverse events, the risk of hypersensitivity reactions to iodinated contrast agents including life-threatening anaphylactic shock, the fixed nature of CT scanners precluding bedside use in critically ill or intraoperative patients, as well as elevated medical costs associated with long-term frequent surveillance. Collectively, these factors may be taken into account in the long-term surveillance of children with PPB.

Based on the imaging and follow-up results of the 12 children in this study, chest US may provide auxiliary support in the diagnosis and management of pediatric PPB.The main clinical indications for thoracic ultrasound in children are as follows: (1) Initial screening for suspected thoracic space-occupying lesions to reduce unnecessary ionizing radiation exposure; (2) Dynamic surveillance of tumor size and internal structural alterations during chemotherapy, facilitating safe and repeat radiation-free follow-up; (3) Long-term postoperative surveillance for the early identification of tumor recurrence and metastasis; In our cohort, the number of follow-up US examinations ranged from 3 to 28, supporting its feasibility for long-term dynamic monitoring. (4) Bedside clinical assessment of critically ill pediatric patients who are unable to undergo conventional CT examination; (5)Intraoperative real-time ultrasonographic evaluation of local thoracic lesions according to clinical needs. US also demonstrates utility in assessing cardiac function and intracardiac anatomy, as well as identifying pericardial effusion, which was detected by echocardiography in 2 cases in the present study. Notably, US can delineate the relationship between tumors and cardiovascular structures without the use of contrast agents, offering certain potential advantages in identifying intracardiac invasion and tumor attachment to cardiac valves. Previous reports indicate that vascular extension is rare, occurring as a severe complication in 3% of type II and III PPB cases. Associated complications include occluded pulmonary veins, left atrial tumor thrombus, great vessel occlusion, paraplegia, and thromboembolism. Tumor extension into the left heart and pulmonary veins is extremely rare [14–17]. Our cohort included one such case, in which the lesion extended through the right superior pulmonary vein into the left atrium, with one mass adherent to the interatrial septum and another attached to the anterior mitral valve chordae tendineae within the left ventricle.The US, CT, and pathological findings of this case are illustrated in Fig. 3.

Fig. 3.

US, CT and pathological features of case 3 (type III pleuropulmonary blastoma with tumor extension into the left heart and pulmonary veins). Type III. a US revealed a large solid hypoechoic mass in the right thoracic cavity with well-defined margins and heterogeneous echogenicity. b The lesion extended along the right superior pulmonary vein into the left atrium, where a mass was visualized attached to the interatrial septum. c A mass was detected within the left ventricular cavity, attached to the chordae tendineae of the anterior mitral leaflet. d CT showed a large mass-like soft-tissue density lesion in the right thoracic cavity with heterogeneous internal density and partially ill-defined borders. Contrast-enhanced CT demonstrated heterogeneous enhancement. e Microscopically, immature tumor cells were distributed in diffuse sheets with marked atypia and focal necrosis (original magnification, low power). f Immunohistochemical staining revealed nuclear positivity for Myogenin (original magnification, medium power)

The results of this study demonstrated that both chest US and CT effectively identified pleural effusion, atelectasis, and pleural thickening. According to previous reports, approximately 10% of patients with PPB present with metastatic disease at initial diagnosis, most commonly involving the brain and bone [18]; other metastatic sites include lymph nodes, liver, pancreas, adrenal gland, heart, and great vessels. For the detection of metastatic lesions, the two modalities show distinct advantages and play complementary roles. Specifically, cervical lymph node metastasis was first detected by US, whereas diaphragmatic and mediastinal lymph node involvement was identified by both modalities and pathologically confirmed. CT detected metastatic lesions in the carina, left main bronchus, and supradiaphragmatic lymph nodes, while US identified intracardiac invasion. CT possesses irreplaceable advantages in the evaluation of deep mediastinal lesions, central airway involvement, microbubble pneumothorax, invasion of the aortic arch and thoracic aorta, and minute pulmonary parenchymal lesions. Such abnormalities are poorly visualized on ultrasound due to sternal shielding and interference from pulmonary gas. Conversely, ultrasound is superior in delineating pericardial effusion, intracardiac tumor extension, tumor attachment sites, and the anatomical relationship between lesions and cardiac valves, thereby serving as an effective complementary modality to CT in these respects.Thus, US and CT are not competitive but highly complementary, and their combined use can substantially improve the diagnostic accuracy of PPB. The clinical application of these two modalities can be optimized based on pediatric imaging principles and local resource availability: in resource-limited settings, US may provide valuable support for the diagnosis and follow-up of PPB; in centers with access to CT, Magnetic Resonance Imaging(MRI), and other advanced imaging, US and CT can be used alternately for monitoring to reduce radiation exposure and sedation requirements. Nevertheless, relying solely on US when CT is available may place patients at risk of delayed diagnosis of disease recurrence. Notably, neither US nor CT alone can establish a definitive diagnosis of PPB. Imaging findings must be combined with histopathological examination, which remains the gold standard for the diagnosis of PPB.

This study has several limitations that warrant acknowledgment, which also provide directions for future research. First, this was a single-center retrospective study with a small sample size. Statistical analysis was not performed owing to the limited number of cases, and type I PPB cases were not included, which may have introduced selection bias and limited the generalizability of the findings. Second, no control group of other thoracic space-occupying lesions was included, resulting in a less systematic analysis of US features for the differential diagnosis of PPB. Third, the association between US findings and pathological results was only evaluated by preliminary qualitative analysis, without further quantitative investigation. Future multicenter, large-sample prospective studies are warranted, which should include all subtypes of PPB and appropriate control groups, establish standardized US diagnostic and differential diagnostic criteria, quantitatively analyze imaging-pathological correlations, and further validate the value of combined US and CT in long-term follow-up. Such efforts may provide evidence for developing more individualized imaging management strategies for pediatric patients of different ages and pathological subtypes.

Conclusions

In conclusion, thoracic US demonstrates useful clinical utility in the initial evaluation, dynamic monitoring, and long-term follow-up of children with pathologically confirmed PPB. As a radiation-free, repeatable, and bedside-applicable modality, US may act as a reasonable complementary imaging method to CT. For pediatric patients, an optimized strategy combining US and CT in appropriate clinical settings may help reduce cumulative radiation exposure and sedation-related risks, thereby supporting comprehensive clinical management.

Abbreviations

PPB

Pleuropulmonary blastoma

LUL

Upper lobe of the left lung

LLL

Lower lobe of the left lung

RUL

Upper lobe of the right lung

RML

Middle lobe of the right lung

US

Ultrasound

CT

Computed Tomography

CDFI

Color Doppler Flow Imaging

CTA

Computed tomographic angiography

PACS

Picture archiving and communication system

MPR

Multi-planar reconstruction

MRI

Magnetic Resonance Imaging

Authors’ contributions

Shuhua LI contributed to design of the work, data curation, methodology, formal analysis, writing-original draft, and writing-review and editing. Yuxi Yan contributed to methodology. Lili Miao contributed to investigation. Cuihua Yang contributed to formal analysis. Yuexia Bai contributed to data collection and analysis. Xiaofang Liu contributed to writing—review and supervision. Qinghua Liu contributed to writing—review, supervision, project administration. Wei Zhang contributed to data curation, supervision.

Funding

The author did not receive support from any organization for the submitted work.

Data availability

The data analyzed during the current study are not publicly available due to the protection of patient privacy and the restrictions of the local ethics committee, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was performed in compliance with the Declaration of Helsinki formulated by the World Medical Association.It was approved by the local ethics committee (Medical Ethics Committee of Children’s Hospital Affiliated to Shandong University, file lD: SDFE-IRB/T-2025142).The studies were conducted in accordance with the local legislation and institutional requirements. The requirement for informed consent was waived due to the retrospective nature of the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.