ABSTRACT

We aimed to evaluate the feasibility, accuracy, and complications of computed tomography (CT)-guided percutaneous transthoracic needle biopsy (PTNB) of cavitary lesions.

Consecutive PTNB procedures in an academic institution over a 4-year period were reviewed, 53 of which were performed on patients with cavitary lesions. The demographic data of patients, lesion characteristics, biopsy technique and complications, initial pathologic results, and final diagnosis were reviewed. A final diagnosis was established through surgical correlation, microbiology or clinico-radiologic follow-up for at least 18 months after biopsy.

The overall accuracy of PTNB was 81%. In 33 patients (62%) the cavitary lesion was found to be malignant (23 lung cancers and 10 metastases). The sensitivity and specificity for malignancy was 91% and 100%, respectively. In 20 patients (38%) a benign etiology was established (16 infections and 4 noninfectious etiologies), with PTNB demonstrating a sensitivity of 81% and specificity of 100% for infection. Wall thickness at the biopsy site, lesion in lower lobe, and malignancy were significant independent risk factors for diagnostic success. Minor complications occurred in 28% of cases: 13 pneumothoraces (5 requiring chest tube), 1 small hemothorax, and 1 mild hemoptysis. A nonsignificant higher chest tube insertion rate was seen in cavities with a thinner wall.

PTNB of cavitary lesions provides high accuracy, sensitivity, and specificity for both malignancy and infection and has an acceptable complication rate. Wall thickness at the biopsy site, lesion in lower lobe, and malignancy were significant independent risk factors for diagnostic success. Samples for microbiology should be obtained in all patients, especially in the absence of on-site cytology, due to the high prevalence of infection in cavitary lesions.