Non-malignant pathological results on transthoracic CT guided core-needle biopsy: when is benign really benign?

Aim To determine true negatives and characterise the variables associated with false-negative results when interpreting non-malignant results of computed tomography (CT)-guided lung biopsy. Materials and methods Nine hundred and fifty patients with initial non-malignant findings on their first transthoracic CT-guided core-needle biopsy (TTNB) were included in the study. Initial biopsy results were compared to definitive diagnoses established later. Results The negative predictive value (NPV) of non-malignant diseases upon initial TTNB was 83.6%. When the biopsy results indicated specific infection or benign tumour ( n =225, 26.1%), they all were confirmed true negative for malignancy later. Only one inconclusive “granuloma” diagnosis was false negative. All 141 patients (141/861, 16.4%) who were false negative for malignancy were from the “infection not otherwise specified (NOS)”, “inflammatory diseases”, or “inconclusive” groups. Age ( p =0.002), cancer history ( p p =0.003), and pneumothorax during lung biopsy ( p =0.003) were found to be significant predictors of false-negative results; 47.6% (410/861) of patients underwent additional invasive examinations to reach a final diagnosis. Ultimately, 52.7% (216/410) were successfully diagnosed. Conclusion Specific infection, benign tumour, and granulomatous inflammation of first TTNBs were mostly true negative. Older age, history of cancer, larger target size, and pneumothorax were highly predictive of false-negative results for malignancies. In such cases, additional invasive examinations were frequently necessary to obtain final diagnoses.