Transbronchial lung biopsy with a cryoprobe, or cryobiopsy, is a novel technique allowing bronchoscopic retrieval of lung biopsies an order of magnitude larger than conventional forceps-based techniques, without crush artifact.1–3 Initial reports have explored the diagnostic capability of cryobiopsy in diffuse parenchymal lung disease (DPLD). Conventional forceps transbronchial biopsies, diagnostic in only a third of cases,4 are generally viewed as insufficient in this setting; current guidelines suggest surgical lung biopsy (SLB) be pursued when clinical and radiographic features are insufficient to establish a confident diagnosis.5,6 However, the risks associated with SLB in this population7–10 are sometimes perceived as prohibitive by clinicians or DPLD patients. In most centers, the most common diagnosis in patients presenting with DPLD is idiopathic pulmonary fibrosis (IPF), which relies on clinical criteria and a specific histologic or radiologic pattern of usual interstitial pneumonia (UIP).5,11 As high-resolution computed tomography (HRCT) of the chest is frequently insufficient for confident diagnosis, and newly available therapeutics have changed management strategies for IPF, a safe alternative to SLB is urgently needed. Cryobiopsy appears ideally positioned to meet this need. However, data remain limited and the techniques used poorly standardized.12 Whether transbronchial cryobiopsy via outpatient flexible bronchoscopy is sufficient to establish histopathologic diagnoses in DPLD, including UIP, remains unclear.
PATIENTS AND METHODS
Patient Selection and Data Extraction
This study was approved by the Vanderbilt University Institutional Review Board (IRB# 151099) with a waiver of informed consent. Our institution transitioned from SLB to cryobiopsy as the predominant initial lung biopsy modality in indeterminate DPLD in January 2012. Records of patients who underwent cryobiopsy through the date of IRB approval (August 17, 2015) were reviewed; biopsies to characterize DPLD were included. Each underwent the procedure once. Pertinent data were retrospectively extracted from the medical record by 2 authors (R.J.L. and T.M.T.). HRCTs within 1 month of cryobiopsy were classified per current guidelines5 as UIP pattern, possible UIP, or inconsistent with UIP by a thoracic radiologist. Subjects included in an early report of cryobiopsy from our institution are included in this analysis, as management changes in these subjects were not previously assessed.1
Bronchoscopy and Cryobiopsy Procedure
Procedures were performed by an attending interventional pulmonologist or fellow under direct supervision in a dedicated bronchoscopy operating room with standard monitoring at least 7 days after last antiplatelet or anticoagulant medication use. A certified registered nurse anesthetist provided intravenous deep sedation and adequate spontaneous respiratory effort was maintained.
After topical anesthesia of the oropharynx, vocal cords, and upper trachea with lidocaine, each subject was intubated with an 8.5 mm wire-reinforced endotracheal tube (ETT) (Smiths Medical, St Paul, MN) advanced over flexible bronchoscope (Olympus XT180 or BF-1TH190; Olympus America Inc., Center Valley, PA). This style of ETT was chosen for its combination of flexibility and strength in this procedure with rapid exits from and entries into the airway, and the ease with which it can be advanced into either mainstem bronchus in the event of significant bleeding. Oxygen was delivered per tube but the cuff was not inflated.
Areas of significant abnormality on preprocedure computed tomography scan were targeted. The lower lobes were preferentially biopsied; each biopsy was taken from a different segment. Before each biopsy, a fluoroscopic reference image was recorded with the bronchoscope wedged in the desired segment to permit fluoroscopic navigation back to the segment in the event of significant bleeding. A 1.9 mm diameter carbon dioxide-cooled cryoprobe measuring 90 cm in length (ERBE Elektromedizin GmBH, Tübingen, Germany) was then advanced until resistance was felt or the probe was seen fluoroscopically to reach the pleura. After pulling back 1.5 to 2 cm, the cryoprobe was activated for 4 to 5 seconds before cryoprobe and bronchoscope were pulled out of the airway en bloc. The cryobiopsy was submerged in room temperature saline to rapidly thaw and release the tissue, allowing the cryoprobe to be removed as the bronchoscope was reintroduced into the airway. The bronchoscope was rewedged in the biopsied segment, iced saline was administered, and tamponade was maintained until hemostasis was achieved. Two to 3 cryobiopsies were sought per case at the discretion of the operator.
Bronchial blockers and a rigid bronchoscope were immediately available to manage complications. Starting January 2015, we began employing prophylactic bronchial blockade using a balloon occlusion catheter (Fogarty; Edwards Lifesciences, Irvine, CA) or endobronchial blocker (Arndt; Cook Medical, Bloomington, IN) as previously described13 while the bronchoscope was out of the airway after each biopsy, in response to a major hemorrhage.
Tissue Processing and Pathology
Cryobiopsies were immediately transferred from saline to 10% formalin, then embedded in paraffin and stained with hematoxylin and eosin, with additional stains at the discretion of the pathologist. All slides were reviewed by 1 of 2 expert lung pathologists. Diagnoses of UIP were made in accordance with current guidelines5; confident non-UIP diagnoses (eg, other idiopathic interstitial pneumonias, granulomatous diseases, bronchiolitides) were based on identification of characteristic features.14,15
Major hemorrhage was scored if additional airway management or bronchial blockade (other than prophylactic use) was required, the procedure was stopped early, transfusion was required, or hemodynamic instability or respiratory failure resulted. Respiratory failure was defined as hypoxemia requiring initiation or escalation of supplemental oxygen or increased level of monitoring for respiratory status (outpatient to inpatient, inpatient floor to ICU). Pneumothoraces, unplanned hospital admissions, and periprocedural arrhythmias were also recorded.
Diagnostic Yield and Management Changes
Two authors with experience in DPLD (R.J.L. and J.A.K.) independently scored the confidence with which diagnoses were established. A confident histopathologic diagnosis was scored if the pathologist identified a specific disease process or strongly favored one disease over, at most, one additional consideration. A confident consensus diagnosis was scored if both retrospective reviewers independently agreed that the multidisciplinary consensus diagnosis was reached with a high degree of certainty and no contradictory data were obtained during the clinical follow-up of that subject. The same authors independently reviewed medical records for management changes attributed to cryobiopsy results.
Continuous variables are reported as mean±SD or median with interquartile range. Categorical variables are reported as frequencies. Between-group comparisons were conducted using Wilcoxon rank-sum test or Pearson χ2 for continuous and categorical variables, respectively. A 2-sided P value of ≤0.05 was considered significant. All analyses were performed using SPSS Statistics v.23 (IBM Corp., Armonk, NY).
Subject, Procedure, and Cryobiopsy Characteristics
A total of 104 subjects underwent cryobiopsy for the diagnosis of a DPLD during the study interval. Baseline subject characteristics are summarized in Table 1.16 The majority (88%) were outpatients at the time of the procedure. Most were referrals from our institution’s Interstitial Lung Disease Clinic when a DPLD was insufficiently characterized by clinical and radiographic data according to the evaluating pulmonologist after review of the HRCT with thoracic radiology. Referrals from outside our institution were evaluated in Interventional Pulmonary Clinic and discussed by our multidisciplinary DPLD committee before the procedure. Preprocedure HRCTs demonstrated a UIP pattern in 7 cases, possible UIP in 11, and were inconsistent with UIP in 86. Subjects with UIP pattern on HRCT proceeded to cryobiopsy because of clinical concern for an alternate diagnosis (eg, hypersensitivity pneumonitis in the context of ongoing avian exposure). A total of 228 cryobiopsies were obtained with mean greatest dimension of 6.9 mm. Alveoli were absent from only 1 case and bronchioles from 3. Lower lobes were most frequently biopsied; in 21 cases (20%), the right upper lobe was biopsied in addition to the right lower lobe. Additional biopsy features are detailed in Table 2.
Diagnostic Yields and Management Changes
Cryobiopsy established a confident histopathologic diagnosis in 46 of 104 cases (44.2%) and a confident consensus diagnosis in 71 (68.3%) (Table 3). All cases with a confident histologic diagnosis received a confident consensus diagnosis upon review by multidisciplinary committee. In an additional 25 cases, less-than-definite histopathologic findings still allowed for a confident consensus diagnosis when considered alongside clinical and radiographic data. UIP, the most common histopathologic diagnosis (Fig. 1), comprised 18% of all cases and 41% of those with a histopathologic diagnosis. IPF was the most common consensus diagnosis, comprising 21% of the series and 31% of those in which a confident consensus diagnosis was established. Of the 18 subjects with UIP pattern or possible UIP pattern on HRCT, half received a histologic diagnosis of UIP and two thirds a consensus diagnosis of IPF (Table 4). There were 10 diagnoses of UIP among 86 subjects with HRCTs interpreted as inconsistent with UIP pattern. SLB, pursued in 5 subjects without a confident diagnosis after cryobiopsy, established 3 additional diagnoses: UIP/IPF, T-cell lymphoma, and organizing pneumonia. A flow chart of the diagnostic process is depicted by Figure 2.
Cryobiopsy changed management in 73 cases (70%), more often when cryobiopsy established a clinical diagnosis than when it did not (88% vs. 39%; 95% confidence intervals, 79%-96% vs. 21%-58%; P<0.01). Changes are detailed in Table 5; the relationship between histopathologic diagnosis, consensus diagnosis, and management change is illustrated in Figure 3.17
Major complications occurred in 8 cases (7.7%). There were 4 major hemorrhages. Two were well-controlled with bronchoscopic tamponade, iced saline, and lung-down positioning but stopped the procedure after 1 biopsy; estimated blood loss (EBL) was 100 mL in each case. The third, occurring after the third biopsy, resulted in brief hypoxemic PEA arrest requiring cuff-up positive pressure ventilation, bronchoscopic tamponade, and decubitus positioning with EBL 300 mL; this subject was extubated the next morning with full neurological recovery. In response, prophylactic bronchial blockade was used after each biopsy in the final 16 cases in this series. Reinflation of this blocker after prophylactic use in one of these latter subjects was scored as the fourth major hemorrhage, as our prespecified definition of major hemorrhage included bleeding necessitating bronchial blockade for control. In this case, deflation of the bronchial blocker revealed a brisk bleed after the first biopsy; reinflating the blocker established rapid control of the hemorrhage and hemostasis was achieved after 3 minutes with EBL 50 mL. Further biopsies were then pursued. Minimal or minor bleeding occurred in the remaining 100 cases; overall mean EBL was 22±19 mL (range, 0 to 75 mL).
Three pneumothoraces occurred, including 1 in the fourth major bleed subject, all requiring small-bore chest tube thoracostomy and 1 to 2 day hospitalizations. There were 2 episodes of respiratory failure: the postarrest subject and an inpatient who developed bronchospasm hours postprocedure prompting brief medical intensive care unit observation. The final serious complication was a subject admitted for overnight observation after returning several hours after discharge with malaise but who did not have pneumothorax, new hypoxemia, or hemoptysis.
In this retrospective analysis of 104 consecutive transbronchial cryobiopsy procedures evaluating indeterminate DPLD, we determined that: (1) cryobiopsy established confident histopathologic diagnoses in 46 cases (44%) and confident consensus diagnoses in 71 (68%); (2) 18% of subjects received a histopathologic diagnosis of UIP and 21% received a consensus diagnosis of IPF; (3) cryobiopsy changed management in 70% of cases; (4) the rate of major complications was 7.7%.
The prevalence of UIP/IPF in existing series of cryobiopsy for DPLD varies widely, likely reflecting variations in study populations and procedural aspects.18 We diagnosed 22 of 104 subjects with IPF for a diagnostic prevalence of 21%, higher than reported in most previous studies.2,18–24 Cryobiopsies in 19 of these 22 subjects were definitive for UIP. Histologic features in the other 3 cases were consistent with UIP but not sufficient for a confident diagnosis; all 3 were felt to have IPF by multidisciplinary committee considering clinical and radiographic data.
The only other large series with a high prevalence of UIP/IPF (31%)18 used a markedly different cryobiopsy technique: 2.4 mm diameter cryoprobe (vs. our 1.9 mm), biopsy <10 mm from the pleura (vs. 15 to 20 mm from the pleura), most biopsies in only 1 segment (vs. different segments, and in 21 cases, different lobes), and intubation with a rigid bronchoscope (vs. orotracheal intubation facilitated by flexible bronchoscope). Despite diagnosing UIP/IPF at similar rates, we note a significantly lower incidence of pneumothorax (2.9% vs. 20.2%) and hospitalization (5.8% vs. 94.6%) in our series. We did have 4 major hemorrhages (3.8%) and have adopted prophylactic bronchial blockade following each biopsy in all patients. Our overall complication rate of 7.7% compares favorably with the previously reported range of 0% to 25%.2,18–24
The relationship between the incidence of the 2 primary complications of cryobiopsy, pneumothorax and significant hemorrhage, has been hypothesized to be reciprocal; higher rates of pneumothorax have been reported when the cryoprobe is activated very close to the pleura,18,23,25 whereas more proximal biopsies near larger-caliber vessels might be at higher risk for major hemorrhage.12 Pneumothorax rates might also be influenced by ventilation strategy, with higher risk of pneumothorax hypothesized in the setting of positive pressure or jet ventilation. For this reason, we utilized deep sedation rather than general anesthesia to permit maintenance of adequate spontaneous ventilation around the cuff-down ETT used in our subjects. More study is necessary to determine an optimal technique which limits the risk of these competing complications while maximizing diagnostic yield.
Our overall consensus diagnostic yield of 68% (71 of 104) falls within the previously reported range of 50% to 98%.2,20–23 Our histopathologic diagnostic yield of 44% (46 of 104), however, falls short of the published range of 51% to 83%18–24 in prior studies of cryobiopsy. We offer several potential explanations. We performed cryobiopsy for any diffuse process if an indication existed for SLB and declined to biopsy only 1 referral (on a high level of mechanical ventilatory support) during the study period. We therefore cryobiopsied a wide range of diffuse lung diseases, not just fibrotic disease suspicious for IPF or other idiopathic interstitial pneumonia (entities with fairly robust histopathologic and clinical-radiologic-pathologic diagnostic criteria). As shown in Table 4, histopathologic diagnostic rate was higher in cases with HRCT consistent with UIP pattern (4 of 7; 57%) and possible UIP pattern (6 of 11; 55%) than those with HRCT inconsistent with UIP (36 of 86; 42%). Despite this, we were still able to establish 19 different specific diagnoses, compared with a range of 7 to 15 diagnoses in prior investigations.2,18–23 This broader scope of pathology might also explain why our diagnostic prevalence of IPF was not higher; UIP was not suspected in subjects with HRCTs concerning for primary small airway processes or diffuse nodules but these patients underwent cryobiopsies nonetheless and were included in this analysis, whereas they were excluded in many prior investigations. Finally, we employed a more stringent definition of histopathologic diagnosis compared with several prior studies,21,22 by including only specific idiopathic interstitial pneumonias and other well-defined entities (eg, malignancy, sarcoidosis, protein alveolar proteinosis14). Our diagnostic rates are well above that reported for traditional forceps transbronchial lung biopsy, with histopathologic and consensus diagnostic rate 20% to 32% in this setting.4,26
No prior studies report how often cryobiopsy affects clinical management. A management change attributed to cryobiopsy occurred in 70% of our cases. Expectedly, changes were more likely if cryobiopsy established a diagnosis, although 39% of nondiagnostic cryobiopsies still led to management changes. As a result, 4 of 5 subjects benefited from cryobiopsy in the form of a diagnosis or a management change (frequently, both), depicted by Figure 3.
Five subjects without a consensus diagnosis following cryobiopsy proceeded to SLB. Surgical pathology was consistent with cryobiopsy pathology (nonspecific small airway-centric inflammation) in 2 cases and no diagnosis was rendered. Two were clear examples of cryobiopsy sampling error: a case of UIP/IPF in which no alveolated parenchyma was recovered (only 1 biopsy obtained due to significant bleeding; likely due to inadvertently proximal biopsy location) and a case of organizing pneumonia with particularly patchy distribution on HRCT. In the fifth case, cryobiopsy demonstrated a necrotic malignancy subsequently diagnosed as T-cell lymphoma by SLB. SLB was offered but declined by the patient in several additional cases with still-indeterminate disease following cryobiopsy.
The primary strength of this study is a relatively high diagnostic prevalence of UIP/IPF using a more conservative outpatient-based flexible bronchoscopic approach than other centers reporting similar rates of UIP/IPF, where rigid bronchoscopy is typically performed and complication rates are higher. This was accomplished despite pursuing cryobiopsy in a broader range of subjects indicated for SLB, including those with suspected primary small airway disease or diffuse nodules. Strict diagnostic definitions in accordance with current guidelines were used to avoid inflating diagnostic yields. All cases were reviewed by a multidisciplinary team of pulmonologists, radiologists, and pathologists at this center with extensive interstitial lung disease experience. Our procedural technique has been meticulously detailed to permit comparison with other practices.
There are several important limitations in the current study. Although the confidence with which diagnoses were established was not always prospectively documented, we attempted to limit bias in the retrospective interpretation of these diagnoses this by having multiple authors independently assess the confidence in each and review clinical follow-up to ensure no contradicting data arose postcryobiopsy. The single-center nature of this series may limit generalizability of these results, although diagnoses were established based on current consensus statement guidelines. Although our sample size is relatively large compared with many other studies of cryobiopsy, this series is still of modest size. No direct comparison with gold standard SLB for histopathologic diagnosis is offered, so safety and yield comparisons to SLB rely on previously published data. Large studies of SLB suggest a range of operative risk depending on preoperative status and comorbidities10,27; our sample was not large enough to permit robust stratification of risk on this basis. Accordingly, comparison between cryobiopsies and SLB remain subject to significant bias at this time. Finally, procedures were performed or supervised by an interventional pulmonologist; complication rates could be higher with less experienced operators. Given the small but clearly present risk of life-threatening hemorrhage during this procedure, in its current state we feel strongly that its performance should be limited to experienced operators with advanced bronchoscopic training who are adept at managing serious hemorrhage.
Transbronchial cryobiopsy continues to show promise as a relatively safe alternative to SLB for the diagnosis of indeterminate DPLD. The outpatient-based flexible bronchoscopic approach we utilized appears capable of diagnosing UIP/IPF as well as a wide range of other diseases. Comparison with SLB remains challenging; directly comparative research is warranted and will help elucidate where cryobiopsy is best situated in the diagnostic algorithm for indeterminate DPLD. Standardization of a cryobiopsy technique that optimizes yield and minimizes complications should be proposed by an expert panel and supported by additional data.
1. Kropski JA, Pritchett JM, Mason WR, et al. Bronchoscopic cryobiopsy
for the diagnosis of diffuse parenchymal lung disease. PLoS One. 2013;8:e78674.
2. Babiak A, Hetzel J, Krishna G, et al. Transbronchial cryobiopsy
: a new tool for lung biopsies. Respiration. 2009;78:203–208.
3. Griff S, Ammenwerth W, Schönfeld N, et al. Morphometrical analysis of transbronchial cryobiopsies. Diagn Pathol. 2011;6:53.
4. Berbescu EA, Katzenstein A-LA, Snow JL, et al. Transbronchial biopsy
in usual interstitial pneumonia. Chest. 2006;129:1126–1131.
5. Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis
: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183:788–824.
6. Wells AU, Hirani N. On behalf of the BTS Interstitial Lung Disease
Guideline Group, a subgroup of the British Thoracic Society Standards of Care Committee, in collaboration with the Thoracic Society of Australia and New Zealand and the Irish Thoracic Society. Interstitial lung disease
guideline. Thorax. 2008;63(suppl 5):v1–v58.
7. Fibla JJ, Brunelli A, Cassivi SD, et al. Aggregate risk score for predicting mortality after surgical biopsy for interstitial lung disease
. Interact Cardiovasc Thorac Surg. 2012;15:276–279.
8. Blackhall V, Asif M, Renieri A, et al. The role of surgical lung biopsy in the management of interstitial lung disease
: experience from a single institution in the UK. Interact Cardiovasc Thorac Surg. 2013;17:253–257.
9. Lee Y-C, Wu C-T, Hsu H-H, et al. Surgical lung biopsy for diffuse pulmonary disease: experience of 196 patients. J Thorac Cardiovasc Surg. 2005;129:984–990.
10. Hutchinson JP, Fogarty AW, McKeever TM, et al. In-hospital mortality after surgical lung biopsy for interstitial lung disease
in the United States. 2000 to 2011. Am J Respir Crit Care Med. 2016;193:1161–1167.
11. Nalysnyk L, Cid-Ruzafa J, Rotella P, et al. Incidence and prevalence of idiopathic pulmonary fibrosis
: review of the literature. Eur Respir Rev. 2012;21:355–361.
12. Poletti V, Hetzel J. Transbronchial cryobiopsy
in diffuse parenchymal lung disease: need for procedural standardization. Respiration. 2015;90:275–278.
13. Hohberger LA, DePew ZS, Utz JP, et al. Utilizing an endobronchial blocker and a flexible bronchoscope for transbronchial cryobiopsies in diffuse parenchymal lung disease. Respiration. 2014;88:521–522.
14. Travis WD, Costabel U, Hansell DM, et al. An Official American Thoracic Society/European Respiratory Society Statement: update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2013;188:733–748.
15. Ryu JH. Classification and approach to bronchiolar diseases. Curr Opin Pulm Med. 2006;12:145–151.
16. Rice TW, Wheeler AP, Bernard GR, et al. Comparison of the SpO2/FIO2 ratio and the PaO2/FIO2 ratio in patients with acute lung injury or ARDS. Chest. 2007;132:410–417.
17. Micallef L, Rodgers P. euler APE: drawing area-proportional 3-Venn diagrams using ellipses. PLoS One. 2014;9:e101717.
18. Ravaglia C, Bonifazi M, Wells AU, et al. Safety and diagnostic yield of transbronchial lung cryobiopsy
in diffuse parenchymal lung diseases: a comparative study versus video-assisted thoracoscopic lung biopsy and a systematic review of the literature. Respiration. 2016;91:215–227.
19. Griff S, Schönfeld N, Ammenwerth W, et al. Diagnostic yield of transbronchial cryobiopsy
in non-neoplastic lung disease: a retrospective case series. BMC Pulm Med. 2014;14:171.
20. Pajares V, Puzo C, Castillo D, et al. Diagnostic yield of transbronchial cryobiopsy
in interstitial lung disease
: a randomized trial: transbronchial cryobiopsy
. Respirology. 2014;19:900–906.
21. Hernández-González F, Lucena CM, Ramírez J, et al. Cryobiopsy
in the diagnosis of diffuse interstitial lung disease
: yield and cost-effectiveness analysis. Arch Bronconeumol. 2015;51:261–267.
22. Fruchter O, Fridel L, El Raouf BA, et al. Histological diagnosis of interstitial lung diseases by cryo-transbronchial biopsy
23. Hagmeyer L, Theegarten D, Wohlschläger J, et al. The role of transbronchial cryobiopsy
and surgical lung biopsy in the diagnostic algorithm of interstitial lung disease
in interstitial lung disease
. Clin Respir J. 2015;10:589–595.
24. Ussavarungsi K, Kern RM, Roden AC, et al. Transbronchial cryobiopsy
in diffuse parenchymal lung disease: retrospective analysis of 74 cases. Chest. 2017;151:400–408.
25. Casoni GL, Tomassetti S, Cavazza A, et al. Transbronchial lung cryobiopsy
in the diagnosis of fibrotic interstitial lung diseases. PLoS One. 2014;9:e86716.
26. Sheth JS, Belperio JA, Fishbein MC, et al. Utility of transbronchial versus surgical lung biopsy in the diagnosis of suspected fibrotic interstitial lung disease
. Chest. 2017;151:389–399.
27. Raj R, Brown KK. Mortality related to surgical lung biopsy in patients with interstitial lung disease
. the devil is in the denominator. Am J Respir Crit Care Med. 2016;193:1082–1084.