Send to

Choose Destination
Chest. 2006 Oct;130(4):1191-7.

Differentiating peripheral pulmonary lesions based on images of endobronchial ultrasonography.

Author information

Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Chang Gung Memorial Hospital, Kaohsiung Medical Center, Taiwan.



To attempt to develop a simple method to discriminate between neoplasm and nonneoplasm peripheral pulmonary lesions based on images of endobronchial ultrasonography (EBUS).


Between June 2004 and June 2005, 151 patients with bronchoscopic peripheral lesions that could not be detected via a conventional bronchoscope underwent EBUS for advanced localization with a 20-MHz miniature radial probe in a tertiary-referral teaching hospital. The image characteristics were applied subsequently to correlate definite histopathologic results in studied patients.


Based on an initial 20 consecutive patients with a definite diagnosis, four image characteristics were issued: (1) continuous hyperechoic margin outside the lesion, (2) homogeneous, or heterogeneous internal echoes, (3) hyperechoic dots in the lesion, and (4) concentric circles along the echo probe. In the following 131 patients, excluding five cases due to inconsistent typing, 93 patients (73.8%) established a diagnosis later. Most cases involving the image characteristics of homogenous internal echoes and concentric circles had nonneoplasm lesions (18 of 19 cases, 94.7%, and 14 of 16 cases, 87.5%, respectively). The difference shown in these two respects with neoplasm lesions was significant by univariate analysis (p < 0.001), although only concentric circles had a significant p value after multivariate analysis. Another two image patterns (continuous hyperechoic margins and hyperechoic dots) did not yield a significant difference (p = 0.090 and p = 0.079, respectively). The average additional time for EBUS was 3.94 min (1.5 to 10 min).


EBUS can provide characteristic information to differentiate the nature of a peripheral pulmonary lesion from the image characteristics of concentric circles.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center