Despite the abundant evidence in the literature with regards to the development of atelectasis in patients having surgery under general anesthesia (GA), the same phenomenon was not recognized—or at least not reported—in patients undergoing bronchoscopy until very recently.1 The growing use of GA for bronchoscopy and the very recent addition of cone-beam computed tomography (CBCT) imaging guidance led to this quite relevant discovery.2 First described as an incidental finding of CBCT imaging during peripheral bronchoscopy, this phenomenon was then studied in more detail in the I-LOCATE Trial (Incidence and Location of Atelectasis).3 In this prospective trial, we enrolled patients undergoing peripheral bronchoscopy under GA, and we performed a survey for atelectasis with radial-probe endobronchial ultrasound (RP-EBUS) in dependent bronchial segments (B2, B6, B9, and B10). More than half of the patients developed atelectasis in all dependent lower lobe segments after 30 minutes of GA, and 89% of patients had at least 1 atelectatic segment. Body mass index and time under GA were associated with a greater number of bronchial segments developing atelectasis.
Why should we be concerned about atelectasis? Atelectasis can be highly detrimental to our outcomes in peripheral bronchoscopy. For those bronchoscopists without access to CBCT—the vast majority—RP-EBUS is the only real-time imaging modality to confirm that the target has been reached. Atelectactic parenchyma, even when recognized by bronchoscopists, can completely obscure our lung targets. Moreover, when atelectasis are not recognized, they can easily mimic a solid lung lesion (false-positive RP-EBUS image), leading bronchoscopists to perform unnecessary biopsies resulting in nondiagnostic samples. Atelectasis, along with respiratory movement, are partially responsible for computed tomography-body divergence which is an enormous barrier to accurate navigation with our current navigational technologies. For patients in need of both diagnosis and nodal staging of suspected lung cancer, when rapid on-site cytology examination is available, staging is supposed to occur first (evidence of malignancy in a lymph node will prevent the need for peripheral bronchoscopy, which has a higher rate of complications). However, the discovery of intrabronchoscopy development of atelectasis and the increase incidence with greater time under GA has led to questioning of this practice. In fact, multiple recent navigational trials have demanded navigation to occur first, or excluded patients with suspicion for lymph node metastases (ie, NCT0372745, NCT04182815). There is no doubt that strategies to prevent atelectasis during bronchoscopies are much needed.
In the current issue of this journal, Bhadra and colleagues retrospectively report on their experience with a dedicated lung navigation ventilation protocol (LNVP) utilized during peripheral bronchoscopy with CBCT guidance for diagnosis of peripheral lung nodules (<30 mm).4 They compare 2 groups of 25 patients each, the first one undergoing conventional ventilation and the second one—2 years later—with LNVP. Their primary endpoint was the development of atelectasis and the proportion of lesions obscured by them. The secondary endpoint was diagnostic yield. Two independent readers evaluated the computed tomography images. The conventional ventilation group received spontaneous intermittent or continuous mechanical ventilation with varying respiratory modes, tidal volumes, fraction of inspired oxygen, and positive end-expiratory pressure. In the LNVP group, all patients were intubated with at least a size 8.5 endotracheal tube, they were all paralyzed, received pressure control/volume guaranteed ventilation with tidal volumes of 10 to 12 mL/kg of ideal body weight, lowest tolerable fraction of inspired oxygen, positive end-expiratory pressure of 10 to 25 cm H2O (upper and middle lobe lesions) and 15 to 20 cm H2O (lower lobe lesions). In addition, patients received 4 alveolar recruitment maneuvers hand-delivered via bagging with 30 to 40 cm H2O over 30 to 40 seconds, with some variability dependent on the anesthesia team. Atelectasis were found in 64% and 68% of the patient (by readers 1 and 2, respectively) in the conventional group, and 36% and 16% (by readers 1 and 2, respectively) in the LNVP group (P=0.00014). The target lesion was obscured by atelectasis in 36% of patients in the conventional group (by both readers 1 and 2) and 4% (reader 1), and 8% (reader 2) for the LNVP group (P=0.01). No hemodynamic complications were reported, and there was only 1 pneumothorax in the LNVP group.
Multiple factors can explain atelectasis during bronchoscopy: wedging bronchoscopes, bleeding and clotting, excessive suctioning, or the mode of ventilation and oxygenation. While some of these factors (and those that are patient-related such as high body mass index) cannot be modified or avoided, a ventilatory strategy to avoid atelectasis is key, and Bhadra and colleagues are leading the way. Their ventilatory strategy showed a significant decrease in the development of atelectasis with respect to conventional ventilation. A few limitations are noticeable and due to the retrospective nature of the study. Unfortunately, ventilation in the conventional ventilation group was not protocol-driven, and also, according to the authors, LNVP may have slightly varied as well in between patients (in particular recruitment maneuvers). Another limitation is that the time from anesthesia induction to CBCT (proven risk factor for atelectasis) was not reported in either group. Peripheral bronchoscopy was performed before nodal staging, and even though the last CBCT utilized for navigation was used to detect atelectasis, navigation time is typically shorter than a complete systematic staging. Surprisingly, no hemodynamic complications were reported despite the high airway pressures. And fortunately, barotrauma was not seen, and there was only one report of pneumothorax, suggesting their technique was safe from that standpoint. While we all understand how atelectasis can negatively influence peripheral bronchoscopy, their impact in diagnostic yield has never been evaluated before. Bhadra and colleagues are the first to investigate this important matter. The diagnostic yield was greater in the LVNP group (92% for LVNP vs. 70% for conventional ventilation; P=0.08), but the difference did not reach statistical significance. The retrospective nature of the study, the small sample size, a single operator, and the difference in operator experience in CBCT-guided bronchoscopy at the time of performing bronchoscopies in each group (2 y apart) prevent us from drawing any valid conclusions with regards to diagnostic yield.
Atelectasis are a hidden enemy of peripheral bronchoscopy that must be defeated. The work of Bhadra and colleagues is pointing us in the right direction. Prospective randomized trials to establish a ventilatory strategy that can safely prevent atelectasis and hopefully provide us enough time to begin our bronchoscopies with mediastinal staging when indicated are underway (NCT04311723).
1. Tusman G, Böhm SH, Warner DO, et al. Atelectasis and perioperative pulmonary complications in high-risk patients. Curr Opin Anaesthesiol. 2012;25:1–10.
2. Casal RF, Sarkiss M, Jones AK, et al. Cone beam computed tomography-guided thin/ultrathin bronchoscopy for diagnosis of peripheral lung nodules: a prospective pilot study. J Thorac Dis. 2018;10:6950–6959.
3. Sagar AS, Sabath BF, Eapen GA, et al. Incidence and location of atelectasis developed during bronchoscopy under general anesthesia: The I-LOCATE Trial. Chest. 2020;158:2658–2666.
4. Bhadra K, Setser R, Condra W, et al. Lung navigation ventilation protocol to optimize biopsy of peripheral lung lesions. J Bronchology Interv Pulmonol. 2022;79:7–17.