ferent sampling instruments, the presence of a bronchus sign, the procedure performance under general anesthesia, the use of ROSE, the prevalence of malignancy [43]. Furthermore, the frequent additional use of other guidance systems may have a role, as demonstrated by Eberhardt et al. [16], that compared three different guidance modalities (EMN alone, EMN +rEBUS, rEBUS alone) in a randomized trial on 120 patients. The best diagnostic yield (88%) was obtained when EMN was combined with rEBUS, in comparison to rEBUS alone (69%) or EMN alone (59%).

The discrepancy in results between studies is well evident comparing the data of the AQuIRE registry (data from fteen Centers in the United States on 581 patients), where the EMN diagnostic yield was very low (38.5%) [47] with the data of a large prospective multicenter study (NAVIGATE) [48] that involved 29 centers and 1215 patients, with a diagnostic yield of 73%. However, it must be observed that in the NAVIGATE trial, fuoroscopy was used with EMN in 91% and rEBUS in 57% of cases.

The major limit of EMN is that it is not a real time guided procedure and that a mismatch between the lesion location on pre-procedural CT scan and its real position during procedure (the so-called “CT-to-body” divergence) may occur, due to movement of the lung with respiratory variation during bronchoscopy. The CT-to-body divergence may explain the difference between the very high reported navigation success (97.4%) and the lower diagnostic yield. Recently, to overcome this limit, a novel technology that use digital tomosynthesis via a conventional fuoroscopy C-arm has been introduced (fuoroscopic EMN, Medtronic, Minneapolis, MN). This system allows visualization of the target nodule on near real-time imaging. In a retrospective study on 67 lesions (mean diameter = 16 mm, range 12–24)

approached by fuoroscopic EMN, diagnostic yield was 79.1% [49].

Trans-Parenchymal Access

To overcome the limit due to the position of the PPL outside the bronchial tree (no bronchus sign), the technique called bronchoscopic trans-­ parenchymal nodule access (BTPNA) was proposed [17]. This method is based on the creation of a direct pathway from the bronchial wall to the lesion. Underlying the procedure, it is necessary to have a 3D model of airway realized by a virtual navigation system (Archimedes Bronchus Medical, Mountain View, CA) based on CT scan, that visualizes the lesion, the airways and the vascular structures. This system is able to identify the optimal airway point of entry and an avascular path through lung tissue from the point of entry and the PPL. After this pre-procedure evaluation, a coring needle is introduced at the de ned point of entry and a balloon dilator is used to enlarge the hole. Then, a radiopaque sheath with a blunt dissection stylet is inserted through the hole and pushed in the lung parenchyma toward the lesion under fuoroscopic guidance. Once the lesion is reached, the stylet is removed and biopsy forceps are inserted through the sheath.

In the rst feasibility study [17], 12 patients were recruited (average size of PPL= 25 mm, range 17–40), tunnel pathway was created in ten and diagnostic yield was 83%, without complications. Another study was performed using this system on a small number of patients (6 subjects) [50], con rming a high diagnostic yield (83%). The major limitation of the technique is the impossibility of realizing a tunnel pathway in some patients, due to the position of the PPL and

to the presence of vascular structures. This limit is particularly evident in PPL located in the apex of the left upper lobe, for the presence of aorta and pulmonary artery [17].

Another trans-parenchymal access system (Bronchoscopic Transbronchial Access Tool— TBAT) was evaluated in a pilot study on 22 patients [13], using EMN and cone beam CT. Seven patients without a de nitive airway leading to the lesion underwent TBAT. The overall diagnostic yield was 77.2% (17/22) and 100% (7/7) when TBAT was used.

Further studies on a larger number of patients are needed to de ne the real advantage of the trans-parenchymal approach in terms of safety and cost/bene t ratio.

Cone Beam CT (CBCT)

Cone beam CT is a variant of computed tomography that uses a cone-shaped X-ray beam instead of a fan-shaped X-ray beam. CBCT uses a rotating C-arm acquiring 2D images that are then reconstructed with a dedicated algorithm to provide 3D images analogous to conventional multi-­ slice CT. The lesion and the bronchial path visualized by CBCT can be overlaid on live fuoroscopy (augmented fuoroscopy), providing

real-time intra-procedural images and allowing also the simultaneous visualization of the lesion and of the position of the sampling instrument in a full 3D view, with the accuracy and the quality of CT view.

CBCT was used rst in other elds of medicine (dentistry, interventional radiology, interventional cardiology, neurosurgery, vascular surgery), but in the recent years several papers demonstrated the possibility to use it as guidance system for the transbronchial approach to PPLs. In the largest study performed with CBCT used with EMN on 93 PPLs (median nodule size = 20 mm), overall diagnostic yield was 83% [51]. Other studies on a smaller number of patients reported a diagnostic yield ranging from 70% to 90% [12, 29]. However, in all the studies CBCT was used in combination with EMN and/or rEBUS and/or ultrathin bronchoscopy.

Diagnostic yields reported using CBCT are summarized in Table 20.4.

Lung Vision

The LungVision system (Body Vision Medical Inc., New York, NY) is a novel method of navigation that provides image fusion of preoperative CT and intraoperative fuoroscopy to create

real-­time augmented fuoroscopic guidance, utilizing arti cial intelligence techniques and dedicated algorithms [15]. The CT scan of the patient is imported into the LungVision planning software and during the procedure an augmented fuoroscopic view of the instrument and of the lesion is displayed on the screen together with the navigation pathway. A study that assessed the distance between lesion location as shown by LungVision augmented fuoroscopy and actual location measured by cone beam CT (CBCT) reported an average distance of only 5.9 mm, demonstrating the reliability of the system [15]. In this trial, performed on 51 patients with a PPL (median size = 18 mm, range 7–48 mm), the diagnostic yield was

78.4%. In another study on 63 patients (median lesion size = 25.0 mm, range: 18–28 mm), using LungVision and rEBUS to con rm the correct location, the overall diagnostic yield was 81.8% and 72.2% for lesions smaller than 2 cm [14] (Table 20.4).

The major advantage of LungVision system is to provide an almost real time vision and an augmented fuoroscopy using a standard fuoroscopy C-arm, in this way allowing visualization of fuoroscopically invisible lesions and reducing the cost in comparison to the more expensive cone beam CT.

Table 20.5 summarizes the advantages and disadvantages of all the above-described guidance systems.