Title

Accuracy of Chest Computed Tomography in Distinguishing Cystic Pleuropulmonary Blastoma From Benign Congenital Lung Malformations in Children.

Document Type

Article

Publication Date

6-1-2022

Identifier

DOI: 10.1001/jamanetworkopen.2022.19814

Abstract

Importance: The ability of computed tomography (CT) to distinguish between benign congenital lung malformations and malignant cystic pleuropulmonary blastomas (PPBs) is unclear.

Objective: To assess whether chest CT can detect malignant tumors among postnatally detected lung lesions in children.

Design, Setting, and Participants: This retrospective multicenter case-control study used a consortium database of 521 pathologically confirmed primary lung lesions from January 1, 2009, through December 31, 2015, to assess diagnostic accuracy. Preoperative CT scans of children with cystic PPB (cases) were selected and age-matched with CT scans from patients with postnatally detected congenital lung malformations (controls). Statistical analysis was performed from January 18 to September 6, 2020. Preoperative CT scans were interpreted independently by 9 experienced pediatric radiologists in a blinded fashion and analyzed from January 24, 2019, to September 6, 2020.

Main Outcomes and Measures: Accuracy, sensitivity, and specificity of CT in correctly identifying children with malignant tumors.

Results: Among 477 CT scans identified (282 boys [59%]; median age at CT, 3.6 months [IQR, 1.2-7.2 months]; median age at resection, 6.9 months [IQR, 4.2-12.8 months]), 40 cases were extensively reviewed; 9 cases (23%) had pathologically confirmed cystic PPB. The median age at CT was 7.3 months (IQR, 2.9-22.4 months), and median age at resection was 8.7 months (IQR, 5.0-24.4 months). The sensitivity of CT for detecting PPB was 58%, and the specificity was 83%. High suspicion for malignancy correlated with PPB pathology (odds ratio, 13.5; 95% CI, 2.7-67.3; P = .002). There was poor interrater reliability (κ = 0.36 [range, 0.06-0.64]; P < .001) and no significant difference in specific imaging characteristics between PPB and benign cystic lesions. The overall accuracy rate for distinguishing benign vs malignant lesions was 81%.

Conclusions and Relevance: This study suggests that chest CT, the current criterion standard imaging modality to assess the lung parenchyma, may not accurately and reliably distinguish PPB from benign congenital lung malformations in children. In any cystic lung lesion without a prenatal diagnosis, operative management to confirm pathologic diagnosis is warranted.

Journal Title

JAMA Netw Open

Volume

5

Issue

6

First Page

2219814

Last Page

2219814

MeSH Keywords

Case-Control Studies; Child; Female; Humans; Lung; Lung Diseases; Lung Neoplasms; Male; Pregnancy; Pulmonary Blastoma; Reproducibility of Results; Tomography, X-Ray Computed

Keywords

Case-Control Studies; Lung Diseases; Lung Neoplasms; Pregnancy; Pulmonary Blastoma; Reproducibility of Results; Tomography, X-Ray Computed

Library Record

Share

COinS