Keywords

Safety profile of transbronchial lung cryobiopsy has been recently compared to that of the conventional forceps or surgical lung biopsy; however, literature shows a variable incidence of complications, mostly bleeding and pneumothorax (Table 6.1). The significant variability of complications is related to the rapid spread of the technique around the world with variable competency and safety standards in different centers and the consequent variability of the procedure itself, in terms of airway access, ventilation, sedation, use of balloon/endobronchial blockers, probe size and freezing time, distance from the pleura, or sampling strategy.

Table 6.1 Comparison between transbronchial forceps biopsy, transbronchial cryobiopsy, and surgical lung biopsy (Modified by Hetzel J et al. [1])
Full size table

Unlike conventional transbronchial biopsy, which is more rarely complicated by pneumothorax [39], pneumothorax seems to be the most common complication occurring after transbronchial lung cryobiopsy (TLCB), with a rate that varies considerably between different studies: from less than 1% to almost 30% [2, 3, 5, 6, 9, 11, 12, 15, 40,41,42,43]. In a meta-analysis that included 15 studies comprising 994 patients, the average rate was 10% [15], and similar results were confirmed by a more recent meta-analysis of 13 studies with an incidence of post-procedural pneumothorax of 9.5% (5.9–14.9%) [41]. There are very few data regarding chest drainage; however Ravaglia et al. showed that out of the pneumothoraces reported, 70% required chest tube drainage, with an overall probability of developing a pneumothorax requiring chest tube drainage of 0.03 (95% CI 0.01–0.08); furthermore, when chest drainage is necessary, time of drainage is usually similar to that of drainage after video-assisted thoracoscopy (VATS) [15]. The risk of pneumothorax can be influenced by patient-related factors (radiological fibrotic score and UIP pattern) or procedure-related factors (type of sedation/airway control, distance from the pleura, size of the probe and freezing time, sampling strategy, skill level of operator) (Table 6.2). Ravaglia et al. showed a higher proportion of events among intubated patients undergoing the procedure under deep sedation compared to those under conscious sedation [15]; this difference could be mainly due to the type of ventilation used, as patients undergoing the procedure under general anesthesia with invasive jet ventilation may develop pneumothorax much more frequently than under conditions of sedation and spontaneous breathing over a bronchoscopy tube [11]. In a large cohort of 699 patients who underwent transbronchial lung cryobiopsy for suspected diffuse parenchymal lung diseases at the Pulmonology Unit of Morgagni Hospital in Forlì (Italy), the risk of pneumothorax appears to be significantly reduced when a 1.9 mm probe is used compared to the 2.4 mm probe (2.7% vs 21.1%, respectively, p 0.00010). Furthermore, the risk of pneumothorax increases when samples are taken from different sites instead of a unique site (p 0.0005) [21] and if they are taken close to the pleura [15, 25]. Finally, the risk of pneumothorax seems to correlate significantly with the presence of histological UIP pattern on biopsy, high-resolution computed tomography (HRCT) fibrosis score of the lower lung zones, and the bronchoscopist’s learning curve [15, 25]. Chest computed tomography (CT) represents currently the gold standard for the diagnosis of pneumothorax; however it is not routinely used, to avoid excess of radiation [37], and it is applicable only to uncertain cases; chest X-ray is used routinely for pneumothorax diagnosis, but with a sensitivity of 46% [44]. In recent years, the use of chest ultrasonography (US) has spread for the diagnosis of pneumothorax as it can reach a pooled sensitivity of 87% (95% CI 81–92%) and a specificity of 99% (95% CI 98–99%) [45] according to a standardized method searching for specific pathognomonic signs [46]. Accuracy of US for the detection of pneumothorax is higher than that of chest X-ray with reference to CT scan as a gold standard, and its use would have some advantages, avoiding exposure to radiation and reducing costs of health care and hospital stay. Chest US in pneumothorax diagnosis after transbronchial lung cryobiopsy has been evaluated for the first time by Viglietta et al. [47]: the analysis showed a sensitivity and a specificity of 90% and 94%, respectively. This approach exploits the ready availability of US that allows the pulmonologist who perform cryobiopsy to detect post-procedural pneumothorax within a short time and optimizes the use of ionizing radiation. A post-procedural chest X-ray or ultrasound examination should be performed to assess for the occurrence of pneumothorax either immediately (if desaturation, persistent cough, and/or thoracic pain are present) or 2–3 h after the end of the procedure if the patient is asymptomatic [47]; this is particularly relevant in the outpatient setting. Patients should be observed in the recovery area as per local institutional guidelines.

Table 6.2 Factors that can influence the risk of pneumothorax
Full size table

Another common complication of cryobiopsy is bleeding [2, 3, 5, 6, 9, 11, 14, 43, 48,49,50,51], although is generally readily controlled endoscopically, e.g., by the use of bronchial blockers (Fogarty balloon or other tools) and/or use of rigid bronchoscopy [5, 14, 15, 25, 51, 52]. There is no generally accepted bleeding severity scale, and therefore comparability of different papers is difficult. However, most papers grade on a scale of four steps: no bleeding, mild bleeding (e.g., requiring suction to clear but no other endoscopic procedures), moderate bleeding (e.g., requiring endoscopic procedures like bronchial occlusion-collapse and/or instillation of ice-cold saline), and severe bleeding (e.g., causing hemodynamic or respiratory instability, requiring tamponade or other surgical interventions, transfusions, or admission to the intensive care unit) [53]. In a previous meta-analysis, moderate bleeding after cryobiopsy was observed in 65 cases among 383 patients from 12 studies (16.9%), with an overall pooled probability of developing a moderate bleeding of about 0.12 (CI 0.02–0.25) [15]. No episodes of severe bleeding, as defined above, are reported in literature (in some papers bleeding has been reported as severe, but it was controlled by placement of bronchial blocker or catheter) [49], and no bleeding-related deaths have been reported after cryobiopsy. A recently published report highlights the risk of potentially life-threatening complications when these precautions are not taken [17]. Abnormal coagulation parameters and the use of clopidogrel or other new antiplatelet drugs are considered contraindications; treatment with aspirin is regarded as a relative contraindication. In the absence of more definitive data and given the increased bleeding risk compared to conventional forceps biopsies, a conservative approach would be to hold all medications potentially associated with increased bleeding risk. Thrombocytopenia (<50 × 10/L) is suggested to be a contraindication for biopsies during flexible bronchoscopy [51]. These values may be accepted also for TLCB until data on this topic will become available. Patients with clinical or radiological signs of pulmonary hypertension should have a pre-procedural evaluation of pulmonary artery pressure by echocardiography or right heart catheterization. An estimated systolic pulmonary artery pressure >50 mmHg on echocardiography indicates an increased likelihood of pulmonary hypertension and, in the absence of more definitive data, is considered a relative contraindication to TLCB [51].

Other complications are anecdotal and can comprise transient respiratory failure, neurological manifestations (e.g., seizures), pneumomediastinum, prolonged air leak, and pulmonary abscess [52].

Regarding mortality, current data are showing that TLCB appears to be safer than surgical lung biopsy; a recent meta-analysis has revealed an overall mortality rate with this procedure of about 0.1% among approximately 1000 patients [15]. A more recent analysis of data published in the literature on cryobiopsy documents seven deaths within a month after the procedure: one patient died from respiratory failure due to carcinomatous lymphangitis, one from acute myocardial infarction manifesting weeks later, one from pulmonary edema from newly diagnosed severe aortic stenosis, one with organizing pneumonia and who was on palliative care, one from pulmonary embolism, and two patients from acute exacerbation of idiopathic pulmonary fibrosis (IPF) [14, 20, 25, 49] (in both cases of death from acute exacerbation of IPF, diffuse alveolar damage was the histological background on autopsy and the death developed after significant procedural complications: tension pneumothorax with subsequent ventilation with high positive airway pressures and severe bleeding). A more recent case of acute exacerbation of interstitial lung disease (ILD) as a complication of TLCB has been reported in a patient with nonspecific interstitial pneumonitis (although this case report does not describe the TLCB technique specifically analysis of histology, description of HRCT features, and clinical information documenting the presence of a stable disease or rapid progressive deterioration before the TLCB) [53]. The risk of acute exacerbation needs to be assessed before the procedure, particularly in case of recent worsening [37, 54, 55]: recent onset of patchy ground-glass areas on HRCT scan, functional deterioration and/or increased dyspnea on exertion in the last month, and/or high levels of inflammatory or more specific markers (KL-6) could be predictors of high acute exacerbation risk [56, 57]. Acute deterioration in respiratory status should be considered a relative contraindication, although the decision needs to be individualized based on assessment of benefits and risks [1].

Anecdotal data suggest that complications are more frequent when pulmonary function is severely impaired. Forced expiratory volume in the first second (FEV1) <0.8 L or <50% predicted, forced vital capacity (FVC) <50% predicted, and diffusing capacity of the lungs for carbon monoxide (DLCO) <35% or <50% predicted have been used to exclude biopsy candidates in some series, though not in all [2, 15, 25]; however, these limitations are drawn from data reported in studies dealing with SLB; additionally, in the subset of patients with severe fibrosing ILD, the risk-benefit analysis is less advantageous, because in these patients it seems that the prognostic significance of an exact histological diagnosis is reduced [58] and data on the efficacy of a specific “anti-fibrotic” drug on patients with severe IPF are still scanty [58,59,60,61,62,63]. Our large cohort of 699 patients who underwent transbronchial lung cryobiopsy for suspected diffuse parenchymal lung diseases, pneumothorax incidence, was significantly higher when FVC was <50% (p 0.008), but it was not influenced by DLCO (p 0.7842), while bleeding appeared independent by the lung function tests (both FVC and DLCO). We suggest that FVC < 50% should be considered as a relative contraindication to transbronchial lung biopsy on safety grounds while baseline DLCO should be evaluated together with other clinical, radiological, and laboratory features [1]. Significant hypoxemia, defined as PaO2 < 55–60 mmHg on room air or while receiving 2 L/min of nasal oxygen, has also been considered a contraindication by some but not others [3, 6, 25]. A high body mass index (BMI > 35) can result in failure of the procedure [1, 25], mainly because of desaturation in intubated and spontaneously breathing patients. Additionally, a study evaluated TBCB in mechanically ventilated patients in the intensive care unit, though the experience remains anecdotal at this time [1, 64]. No age limit has been suggested at this time, as TBCB has been performed safely in a wide age range of patients, which need to be carefully evaluated in terms of comorbidities and fitness for anesthesia [1].