ObjectiveTo systematically review the diagnostic accuracy of C-arm cone-beam CT (CBCT)-guided percutaneous transthoracic needle biopsy (PTNB) for lung nodules.
MethodsWe electronically searched databases including PubMed, EMbase, EBSCO, Ovid, CBM, VIP, WanFang Data and CNKI from inception to Feb 28th, 2015, to collect diagnostic studies of CBCT-guided PTNB for lung nodules. Two reviewers independently screened literature, extracted data and assessed the methodological quality of included studies by QUADAS-1 tool. Then, meta-analysis was performed by Stata 12.0 and Meta-DiSc 1.4 softwares for calculating pooled sensitivity (Sen), specificity (Spe), positive likelihood ration (+LR), negative likelihood ration (-LR), and diagnostic odds ratio (DOR), drawing summary receiver operating characteristic (SROC) curve and estimating area under the curve (AUC).
ResultsA total of 9 studies involving 1 815 patients were included. The results of meta-analysis showed that the pooled Sen, Spe, +LR,-LR, and DOR were 0.95 (95%CI 0.92 to 0.96), 1.00 (95%CI 0.66 to 1.00), 2 076.58 (95%CI 1.8 to 2.3e+0.6), 0.05 (95%CI 0.04 to 0.08), and 39 443.88 (95%CI 30.53 to 5.1e+0.7), respectively. The AUC of SROC was 0.97 (95%CI 0.95 to 0.98).
ConclusionCBCT-guided PTNB can be used as one of the primary examination approaches for lung nodules with relatively high diagnostic accuracy. Due to limited quality and quantity of the included studies, more high quality studies are needed to verify the above conclusion.
Percutaneous transthoracic needle biopsy (PTNB) is an important modality for the qualitative diagnosis of lung cancer, which relies on medical-image guidance. X-ray equipment features wide availability and low radiation dose; nevertheless, X-ray-guided PTNB requires repeated calibration using physical calibrators. Accordingly, this paper proposes an X-ray-guided lung puncture navigation method integrating the direct linear transformation-Gauss-Newton method (DLT-GNM) and marker-free registration. Preoperatively, initial imaging parameters are acquired via fluoroscopic calibration. Intraoperatively, a binocular camera tracks the spatial pose of the C-arm in real time to virtually reconstruct calibration points. DLT-GNM is employed to solve the projection model and realize marker-free real-time navigation. Respiratory phantom experiments show that the overall localization error is (0.38 ± 0.25) mm under matched respiratory phases. It is verified that this method delivers high accuracy and simple operation, and possesses favorable clinical application potential for cases with large lesions or identifiable targets clearly visualized under X-ray fluoroscopy.
