Introduction: Cryobiopsy (CB) using a 1.1-mm cryoprobe under fluoroscopic guidance is feasible and safe for diagnosis of ground glass opacity (GGO) lesions. However, the efficacy of CB combined with cone-beam CT (CBCT) for GGO-predominant pulmonary nodules remains elusive. Methods: We retrospectively studied patients who underwent CB combined with conventional biopsy under CBCT guidance for GGO-predominant pulmonary nodules with a consolidation-to-tumour ratio <50.0%. Results: A total of 32 patients with GGO-predominant pulmonary nodules were enrolled: 17 pure GGOs and 15 mixed GGOs. The mean lesion diameter was 15.81 ± 5.52 mm and the overall diagnostic yield was 71.9%. Seven lesions were diagnosed by CB alone, which increased the diagnostic outcomes by 21.9%. Diagnostic yields for CB, forceps biopsy (FB), brushing, and guide sheath flushing were 65.6%, 46.9%, 15.6%, and 14.3%, respectively. Univariate analysis revealed that positive computed tomography (CT) bronchus sign (p = 0.035), positive CBCT sign (p < 0.01), and CB-first biopsy sequence (p = 0.036) were significant predictive factors for higher diagnostic yield. Specimens obtained by CB had larger mean sample size (p < 0.01), lower blood cell area (p < 0.01), and fewer crush artefacts (p < 0.01) than specimens from FB. No severe bleeding or other complications occurred. Conclusion: CB using a 1.1-mm cryoprobe under CBCT guidance increased diagnostic yield for GGO-predominant pulmonary nodules based on conventional biopsy. Further, it provided larger and nearly intact samples compared with forceps.

Ground glass nodules (GGNs) are defined as lesions <3 cm with an area of increased attenuation and preservation of the underlying vessels and bronchi on computed tomography (CT). GGNs can be categorised as pure (no solid component) or mixed (heterogeneous attenuation with some solid components) [1]. GGNs are radiologically nonspecific findings, which may present as benign lesions (e.g., infection or inflammation) or lung adenocarcinoma [2, 3]. Therefore, it is crucial to obtain specimens of GGNs for definitive diagnosis prior to treatment.

Various forms of guided bronchoscopy have been used to locate ground glass opacity (GGO) lesions [4]. The diagnostic yield of endobronchial ultrasound (EBUS) combined with a guide sheath (GS) to diagnose GGO lesions fluctuated between 57.0% and 65.0% [5, 6]. However, when the lesions were small or had a low consolidation-to-tumour ratio (CTR), the diagnostic yield was only 25.0–50.0% [5, 6]. There is still space for improvement in transbronchial lung biopsy (TBLB) for GGO-predominant pulmonary nodules. Cone-beam CT (CBCT) can provide three-dimensional information with standard fluoroscopy to precisely locate target lesions [7]. Several studies have shown the feasibility and potential of CBCT-guided TBLB for diagnosis of pulmonary nodules [8, 9]. Therefore, CBCT may have diagnostic potential for GGNs.

In addition to confirmation of tool-in-lesion, the use of appropriate biopsy tools is also key to acceptable results. Transbronchial lung cryobiopsy (CB) can acquire large and well-preserved samples that contribute to histopathological diagnosis [10, 11]. Previous studies reported that CB using a 1.1-mm cryoprobe for pulmonary nodules was feasible and safe based on conventional biopsy, which increased the diagnostic yield by 18.2–20.0% [12, 13]. We previously applied a 1.1-mm cryoprobe to diagnose GGO lesions and achieved a diagnostic yield of 82.61% [14]. However, there are few published reports of its use in conjunction with CBCT for GGNs.

This study evaluated the feasibility and safety of CBCT-guided CB combined with conventional TBLB for GGO-predominant pulmonary nodules. Additionally, we explored the factors that may influence diagnostic outcomes and assessed the quality of specimens.

Patients

We retrospectively included consecutive patients who received TBLB under CBCT guidance in Shanghai Chest Hospital between November 2020 and March 2023. Inclusion criteria were (1) age >18 years, (2) presence of GGO-predominant pulmonary nodules (CTR <50.0%) suspicious for malignancy in need of nonsurgical biopsy, and (3) lesions biopsied by both CB and conventional biopsy in the same bronchoscopy procedure. The exclusion criteria were (1) concomitant endobronchial lesions that were visualised using bronchoscopy, (2) diffuse lesions, and (3) CB and FB samples were mixed together. This study was approved by the Ethics Committee of Shanghai Chest Hospital (IS23038). The need for informed consent was waived by the Ethics Committee of Shanghai Chest Hospital.

Navigational Bronchoscopy

In the planning phase, each patient underwent chest CT (slice thickness: 0.5–1.0 mm, interval: 0.5–1.0 mm) before bronchoscopy. The data were transferred to a virtual bronchoscopic navigation system (VBN; DirectPath; Olympus, Tokyo, Japan) or electromagnetic navigation bronchoscopy system (ENB; LungCare navigation system, LungCare Medical Technologies Ltd., Inc., Suzhou, China, or SuperDimension system, Medtronic, Minneapolis, USA) to construct virtual images and determine the biopsy route before examination.

A standard bronchoscope (BF-1TQ290, outer diameter 5.9 mm, working channel 3.0 mm; Olympus) was paired with a K-203 GS (outer diameter 2.55 mm; Olympus) or an extended working channel (EWC) (outer diameter 2.6 mm; Medtronic). A thin bronchoscope (BF-P290, outer diameter 4.2 mm, working channel 2.0 mm; Olympus) was paired with a K-201 GS (outer diameter 1.95 mm; Olympus). Ultrathin bronchoscopy (BF-MP290F, outer diameter 3.0 mm, working channel 1.7 mm; Olympus) was carried out without a GS because of the restricted working channel. The choice of bronchoscopes and guidance techniques was based on the operator’s estimation of procedural feasibility and difficulty, taking into account lesion characteristics and safety. In principle, a thin bronchoscope combined with VBN was recommended for most procedures. When lesions were located in the more peripheral bronchus (e.g., 7th bronchus generation or more distal), we preferred utilising an ultrathin bronchoscope with VBN or a standard bronchoscope with ENB.

CBCT Scan and Adjustment

All bronchoscopy procedures were performed in an operating room with CBCT (Discovery IGS 730; GE HealthCare, Boston, USA, or Cios Spin; Siemens Healthcare, Forchheim, Germany). Patients undergoing general anaesthesia were intubated and positioned in the supine position on the angiographic table. An 8 s scan protocol for Discovery IGS 730 (0.36 μGy/frame, 400 frames and 200° gantry rotation) or a 30 s scan protocol for Cios Spin (0.14 μGy/frame, 400 frames, and 196° gantry rotation) was used for CBCT imaging. All CBCT scans were performed during inspiratory breath holding with an adjustable pressure-limiting valve of 10–20 cm H2O. CBCT images from three dimensions (axial, coronal, and sagittal) were initially obtained to identify the target lesions. When the bronchoscope was close enough to the lesion under fluoroscopy and navigation, an EBUS probe with a diameter of 1.4 mm (UM-S20-17; Olympus) was introduced into the working channel with/without a GS. After obtaining EBUS images of distinctive signals, a second CBCT was performed to confirm the relationship between the probe and the lesion.

CBCT images were categorised into three types based on the positional relationship between the EBUS probe and the target: positive (inside or next to the lesion), negative (not touching), and inconclusive (difficult to judge whether EBUS probe has reached the lesion). If the EBUS probe did not touch the target lesion, it was adjusted based on CBCT images and fluoroscopic guidance. A maximum of three adjustments under CBCT guidance in each patient were permitted to limit radiation exposure. Once the EBUS probe reached the target lesion, the fluoroscopic images of the EBUS probe were retained as markers, and the probe was withdrawn, maintaining the bronchoscope or GS in situ. Then, biopsy procedures were performed under live fluoroscopic guidance with reference to marked fluoroscopic images.

Sampling Procedure

According to the different guided bronchoscopy methods, samples harvested by a 1.1-mm cryoprobe (20402-401; Erbe, Tübingen, Germany) were retrieved as follows: (1) the cryoprobe was removed from the K-203 GS or EWC, keeping the standard bronchoscope and GS/EWC in situ to facilitate repeated biopsy and reduce bleeding; (2) the cryoprobe was removed en bloc with the K-201 GS, keeping the thin bronchoscope in situ to handle bleeding under direct vision; and (3) the cryoprobe and ultrathin bronchoscope were removed together. A CB-first biopsy sequence with a freezing time of 3–5 s was advised because GGO lesions are more prone to bleed than solid lesions during biopsy, which may impact CB effectiveness [14].

Conventional biopsy, including FB (FB-233D or FB-231D; Olympus) and brushing (BC-204D or BC-202D; Olympus), was also performed according to the fluoroscopic images. A 1.9-mm forceps was combined with a standard or thin bronchoscope when appropriate tissue was not obtained using a 1.5-mm forceps. All procedures were performed by two experienced experts (Junxiang Chen and Jiayuan Sun). Three CB samples and 5–10 FB samples were recommended for pathological examination. GS flushing was used in cases using GS during examination. Finally, bronchoscopy was performed again to investigate airway bleeding. Rapid on-site cytopathological evaluations were not conducted.

Histopathological Evaluation

The samples harvested by CB and FB were separately plunged into 10% formalin and fixed for 12–24 h at room temperature. Formalin-fixed samples were embedded in paraffin and stained with haematoxylin and eosin. The 3-μm sections were scanned by the NanoZoomer 2.0-HT Slide Scanner (Hamamatsu Photonics, Japan) and used for morphological assessment and quantitative morphometric analysis. An experienced pathologist blinded to the biopsy technique evaluated available sections using the NDP.view2 viewing software [15]. Online supplementary Figure S1 (for all online suppl. material, see https://doi.org/10.1159/000535236) shows an example of representative samples for histopathological evaluation.

Study Outcomes

The primary endpoint was the overall diagnostic yield and that of each biopsy method. Malignant or specific benign findings resulting in a definitive diagnosis were considered diagnostic (e.g., definitive malignancy or granuloma). Findings unspecific or unqualified were regarded as nondiagnostic (e.g., atypical cells, normal lung tissue, or inflammation). Follow-up results were not included in the calculation of the diagnostic yield [16]. The secondary endpoints were size and quality of specimens, safety, and factors associated with the diagnostic yield. Haemorrhage was assessed according to the standardised definitions of bleeding [17].

Statistical Analysis

Baseline characteristics were recorded as frequency, percentage, median interquartile range, or mean ± SD. Comparisons were performed by t test, Mann-Whitney test, Fisher’s exact test, or Pearson’s χ2 test as appropriate. Statistical analysis was performed using SPSS version 22.0 (IBM, Armonk, NY, USA). p < 0.05 was considered statistically significant.

Patients

Totally, 32 patients with GGO-predominant nodules were eligible and enrolled in this study. Specimens were not available for histopathological evaluation in 1 patient diagnosed with lung cancer by CB alone (shown in Fig. 1). Baseline characteristics and histopathological findings are shown in Table 1. There were 17 pure GGNs and 15 mixed GGNs. The mean lesion size was 15.81 ± 5.52 mm. All lesions were invisible by fluoroscopy but were visualised with CBCT all prior to bronchoscope insertion. Positive and negative bronchus signs on CT were 65.6% and 34.4%, respectively.

Fig. 1.

Flowchart of study inclusion. CB, cryobiopsy; CBCT, cone-beam computed tomography; PPNs, peripheral pulmonary nodules.

Fig. 1.

Flowchart of study inclusion. CB, cryobiopsy; CBCT, cone-beam computed tomography; PPNs, peripheral pulmonary nodules.

Close modal
Table 1.

Patient characteristics

Characteristics
Total patients, n 32 
Age, years 64 (50–70) 
Gender 
 Male 12 (37.5) 
 Female 20 (62.5) 
Lesion size, mm 15.81±5.52 
 8<n ≤ 20 23 (71.9) 
 20<n ≤ 30 9 (28.1) 
Lesion location 
 Right upper lobe 13 (40.6) 
 Right middle lobe 2 (6.2) 
 Right lower lobe 8 (25.0) 
 Left upper lobe 6 (18.8) 
 Left lower lobe 3 (9.4) 
Appearance of lesions on CT 
 Pure GGO 17 (53.1) 
 Mixed GGO 15 (46.9) 
  0<CTR ≤25% 10 (31.3) 
  25%<CTR <50% 5 (15.6) 
Distance to pleura, mm 15.07±9.21 
CT bronchus sign 
 Positive 21 (65.6) 
 Negative 11 (34.4) 
Histopathological findings 
 Specific findings 23 (71.9) 
  Adenocarcinoma 23 (71.9) 
 Non-specific findings 9 (28.1) 
Characteristics
Total patients, n 32 
Age, years 64 (50–70) 
Gender 
 Male 12 (37.5) 
 Female 20 (62.5) 
Lesion size, mm 15.81±5.52 
 8<n ≤ 20 23 (71.9) 
 20<n ≤ 30 9 (28.1) 
Lesion location 
 Right upper lobe 13 (40.6) 
 Right middle lobe 2 (6.2) 
 Right lower lobe 8 (25.0) 
 Left upper lobe 6 (18.8) 
 Left lower lobe 3 (9.4) 
Appearance of lesions on CT 
 Pure GGO 17 (53.1) 
 Mixed GGO 15 (46.9) 
  0<CTR ≤25% 10 (31.3) 
  25%<CTR <50% 5 (15.6) 
Distance to pleura, mm 15.07±9.21 
CT bronchus sign 
 Positive 21 (65.6) 
 Negative 11 (34.4) 
Histopathological findings 
 Specific findings 23 (71.9) 
  Adenocarcinoma 23 (71.9) 
 Non-specific findings 9 (28.1) 

Data are illustrated as number (%), mean ± SD or median (interquartile range).

CT, computed tomography; CTR, consolidation-to-tumour ratio.

Procedures

Procedural details are presented in Table 2. Thin bronchoscopy (68.8%) and GS (87.5%) were commonly used in bronchoscopy procedures. All patients underwent CB, FB, and brushing. Of them, 23 were in the CB-first biopsy sequence and 9 in the FB-first biopsy sequence. Patients in the FB-first biopsy sequence were biopsied using a thin bronchoscope with GS. GS flushing was additionally applied to 28 patients. Details for the CBCT procedure are presented in Table 3. The median number of intraoperative CBCT scans was 1 (interquartile range: 1–2). The primary CBCT scan showed 18 (56.3%) positive, 12 (37.5%) negative, and 2 (6.2%) inconclusive CBCT images, respectively. After CBCT-based adjustments, the final EBUS images included 28 blizzard and four mixed blizzard signs; the final CBCT findings revealed 27 (84.4%) positive cases and 5 (15.6%) inconclusive cases. In the latter, one was local atelectasis, and four were image interferences caused by bleeding or secretion.

Table 2.

Procedural details

VariableValue
Type of bronchoscope used 
 BF-P290 22 (68.8) 
 BF-MP290F 3 (9.4) 
 BF-1TQ290 7 (21.9) 
Biopsy method  
 CB 32 (100.0) 
 FB 32 (100.0) 
  1.5-mm forceps 22 (68.8) 
  1.9-mm forceps 10 (31.2) 
 Brushing 32 (100.0) 
 GS flushing 28 (87.5) 
Biopsy sequence 
 CB-first 23 (71.9) 
 FB-first 9 (28.1) 
Biopsy number 
 CB 3 (3–4) 
 FB 8 (7–11) 
  1.5-mm forceps 8 (7–11) 
  1.9-mm forceps 9 (7–12) 
GS 
 K-201 21 (65.6) 
 K-203 1 (3.1) 
 EWC 6 (18.8) 
 Without GS 4 (12.5) 
EBUS image 
 Blizzard sign 28 (87.5) 
 Mixed blizzard sign 4 (12.5) 
Navigation bronchoscope 
 VBN 19 (59.4) 
 ENB 13 (40.6) 
VariableValue
Type of bronchoscope used 
 BF-P290 22 (68.8) 
 BF-MP290F 3 (9.4) 
 BF-1TQ290 7 (21.9) 
Biopsy method  
 CB 32 (100.0) 
 FB 32 (100.0) 
  1.5-mm forceps 22 (68.8) 
  1.9-mm forceps 10 (31.2) 
 Brushing 32 (100.0) 
 GS flushing 28 (87.5) 
Biopsy sequence 
 CB-first 23 (71.9) 
 FB-first 9 (28.1) 
Biopsy number 
 CB 3 (3–4) 
 FB 8 (7–11) 
  1.5-mm forceps 8 (7–11) 
  1.9-mm forceps 9 (7–12) 
GS 
 K-201 21 (65.6) 
 K-203 1 (3.1) 
 EWC 6 (18.8) 
 Without GS 4 (12.5) 
EBUS image 
 Blizzard sign 28 (87.5) 
 Mixed blizzard sign 4 (12.5) 
Navigation bronchoscope 
 VBN 19 (59.4) 
 ENB 13 (40.6) 

Data are illustrated as number (%) or median (interquartile range).

EBUS, endobronchial ultrasound; ENB, electromagnetic navigation bronchoscopy; VBN, virtual bronchoscopic navigation.

Table 3.

Intraoperative CBCT scan and adjustment

Intraoperative CBCT scanCaseRelationship of EBUS probe and the target lesion
insidenext tonegativeinconclusive
First scan 32 (100.0) 12 (37.5) 6 (18.8) 12 (37.5) 2 (6.2) 
Second scan 12 (37.5) 2 (6.2) 4 (12.5) 6 (18.8) 
Third scan 6 (18.8) 2 (6.2) 2 (6.2) 2 (6.2) 
Fourth scan 2 (6.2) 1 (3.1) 1 (3.1) 
Final result 32 (100.0) 17 (53.1) 10 (31.3) 5 (15.6) 
Intraoperative CBCT scanCaseRelationship of EBUS probe and the target lesion
insidenext tonegativeinconclusive
First scan 32 (100.0) 12 (37.5) 6 (18.8) 12 (37.5) 2 (6.2) 
Second scan 12 (37.5) 2 (6.2) 4 (12.5) 6 (18.8) 
Third scan 6 (18.8) 2 (6.2) 2 (6.2) 2 (6.2) 
Fourth scan 2 (6.2) 1 (3.1) 1 (3.1) 
Final result 32 (100.0) 17 (53.1) 10 (31.3) 5 (15.6) 

Data are illustrated as number (%).

EBUS, endobronchial ultrasound; CBCT, cone-beam computed tomography.

Diagnostic Outcomes

The Venn diagram of diagnostic outcomes is presented in Figure 2. The overall diagnostic yield of the CBCT-guided CB combined with conventional biopsy for GGO-predominant pulmonary nodules was 71.9% (23/32), higher than that of FB alone (p = 0.042). CB had the highest diagnostic yield (65.6%, 21/32), followed by FB (46.9%, 15/32), brushing (15.6%, 5/32), and GS flushing (14.3%, 4/28). CB improved the diagnostic yield of conventional biopsy by additionally diagnosing 7 patients (21.9%, 7/32). In univariate analysis, a positive CT bronchus sign (leading to the lesion; 85.7 vs. 45.5%, p = 0.035), a positive CBCT sign (85.2% vs. 0, p < 0.01), and CB-first biopsy sequence (73.9 vs. 43.5%, p = 0.036) were significant factors predicting higher diagnostic yield (Table 4). The diagnostic yield of CB and FB corresponding to the CBCT sign is shown in online supplementary Table S1.

Fig. 2.

Venn diagram of diagnostic yield of the bronchoscopy procedures. CB, cryobiopsy; FB, forceps biopsy; GS, guide sheath.

Fig. 2.

Venn diagram of diagnostic yield of the bronchoscopy procedures. CB, cryobiopsy; FB, forceps biopsy; GS, guide sheath.

Close modal
Table 4.

Univariate analysis

VariableUnivariate analysis of biopsy
diagnostic yield (%)p
Lesion size (mm)  
 8 < n ≤ 20 16/23 (69.6)  
 20 < n ≤ 30 7/9 (77.8)  
Lesion location  0.249 
 RUL/LUL 12/19 (63.2)  
 Others 11/13 (84.6)  
Appearance of lesions on CT  
 Pure GGO 12/17 (70.6)  
 Mixed GGO 11/15 (73.3)  
EBUS image  0.303 
 Blizzard sign 19/28 (67.9)  
 Mixed blizzard sign 4/4 (100.0)  
CT bronchus sign  0.035 
 Positive 18/21 (85.7)  
 Negative 5/11 (45.5)  
CBCT sign  <0.01 
 Positive 23/27 (85.2)  
 Inconclusive 0/5 (0)  
CB-first biopsy sequence  0.036 
 CB 17/23 (73.9)  
 FB 10/23 (43.5)  
FB-first biopsy sequence  
 CB 4/9 (44.4)  
 FB 5/9 (55.6)  
Navigation bronchoscope  
 VBN 14/19 (73.7)  
 ENB 9/13 (69.2)  
VariableUnivariate analysis of biopsy
diagnostic yield (%)p
Lesion size (mm)  
 8 < n ≤ 20 16/23 (69.6)  
 20 < n ≤ 30 7/9 (77.8)  
Lesion location  0.249 
 RUL/LUL 12/19 (63.2)  
 Others 11/13 (84.6)  
Appearance of lesions on CT  
 Pure GGO 12/17 (70.6)  
 Mixed GGO 11/15 (73.3)  
EBUS image  0.303 
 Blizzard sign 19/28 (67.9)  
 Mixed blizzard sign 4/4 (100.0)  
CT bronchus sign  0.035 
 Positive 18/21 (85.7)  
 Negative 5/11 (45.5)  
CBCT sign  <0.01 
 Positive 23/27 (85.2)  
 Inconclusive 0/5 (0)  
CB-first biopsy sequence  0.036 
 CB 17/23 (73.9)  
 FB 10/23 (43.5)  
FB-first biopsy sequence  
 CB 4/9 (44.4)  
 FB 5/9 (55.6)  
Navigation bronchoscope  
 VBN 14/19 (73.7)  
 ENB 9/13 (69.2)  

RUL, right upper lobe; LUL, left lower lobe; CT, computed tomography; GGO, ground glass opacity; EBUS, endobronchial ultrasound; CBCT, cone-beam computed tomography; CB, cryobiopsy; FB, forceps biopsy; ENB, electromagnetic navigation bronchoscopy; VBN, virtual bronchoscopic navigation.

Specimen Size and Quality

The histopathological evaluation was performed on 31 patients (Table 5). The total sample size for CB was 6.54 (3.69–8.50) mm2, compared with 4.11 (2.06–5.45) mm2 for FB, 2.46 (1.54–4.25) mm2 for 1.5-mm forceps, and 5.45 (5.00–6.82) mm2 for 1.9-mm forceps. The total sample size of CB was higher than that of FB (p < 0.01) and 1.5-mm forceps (p < 0.01). The total area of blood cells in CB samples was lower than that of FB samples (0.03 mm2 vs. 1.60 mm2, p < 0.01), as was the proportion of crush artefacts in CB samples compared with that in FB samples (p < 0.01). The quality of samples related to the biopsy sequence is illustrated in online supplementary Table S2.

Table 5.

Histopathological evaluation

VariableCB (N = 31)FB (N = 31)1.5-mm forceps (N = 21)1.9-mm forceps (N = 10)
Number of obtained samples 3 (2–3) 6 (5–7)** 5 (5–7)** 7 (6–8)**,†† 
Total sample size, mm2 6.54 (3.69–8.50) 4.11 (2.06–5.45)** 2.46 (1.54–4.25)** 5.45 (5.00–6.82)†† 
Mean sample size, mm2 2.18 (1.37–3.10) 0.59 (0.44–0.85)** 0.49 (0.31–0.62)** 0.83 (0.71–0.89)**,†† 
Total area of blood cells in the samples, mm2 0.03 (0–1.04) 1.60 (0.50–2.64)** 0.96 (0.35–1.79)** 2.96 (1.78–3.28)**,†† 
Mean area of blood cells in the samples, mm2 0.01 (0–0.41) 0.24 (0.08–0.38)* 0.16 (0.07–0.34) 0.34 (0.25–0.53)*,† 
Crush artefact, % 0 (0) 15 (48.4)** 10 (47.6)** 5 (50.0)** 
VariableCB (N = 31)FB (N = 31)1.5-mm forceps (N = 21)1.9-mm forceps (N = 10)
Number of obtained samples 3 (2–3) 6 (5–7)** 5 (5–7)** 7 (6–8)**,†† 
Total sample size, mm2 6.54 (3.69–8.50) 4.11 (2.06–5.45)** 2.46 (1.54–4.25)** 5.45 (5.00–6.82)†† 
Mean sample size, mm2 2.18 (1.37–3.10) 0.59 (0.44–0.85)** 0.49 (0.31–0.62)** 0.83 (0.71–0.89)**,†† 
Total area of blood cells in the samples, mm2 0.03 (0–1.04) 1.60 (0.50–2.64)** 0.96 (0.35–1.79)** 2.96 (1.78–3.28)**,†† 
Mean area of blood cells in the samples, mm2 0.01 (0–0.41) 0.24 (0.08–0.38)* 0.16 (0.07–0.34) 0.34 (0.25–0.53)*,† 
Crush artefact, % 0 (0) 15 (48.4)** 10 (47.6)** 5 (50.0)** 

Data are illustrated as median (interquartile range) or number (%).

CB, cryobiopsy; FB, forceps biopsy.

*p < 0.05, **p < 0.01 when CB compared with other biopsy methods; p < 0.05, ††p < 0.01 when 1.5-mm forceps compared with 1.9-mm forceps.

Safety

Mild and moderate bleeding was identified in 26 and 6 patients, respectively. Mild bleeding was controlled by <1 min of simple bronchoscopic suction. Of the six patients with moderate bleeding, three required >1 min of bleeding suction and three required cold saline infusion. No fatal bleeding or other complications occurred.

This study, for the first time, demonstrated that the combination of CB and conventional biopsy under CBCT guidance are a feasible and safe method for diagnosis of GGO-predominant pulmonary nodules, with an overall diagnostic yield of 71.9%. The samples harvested by CB were larger and better preserved, which improved the diagnostic yield by 21.9% compared with conventional biopsy.

It is difficult to diagnose GGO lesions by EBUS-GS-guided transbronchial biopsy under fluoroscopy, given the challenge of pinpointing the lesions [5, 6, 18, 19]. Previous studies showed a diagnostic yield of 57–69% TBLB under fluoroscopy for GGO-predominant lesions [5, 18]. CBCT could provide real-time information of tool-in-lesion [20, 21]. Thus, our study utilised CBCT-guided CB combined with conventional biopsy to harvest 32 GGO-predominant pulmonary nodules (mean size 15.81 ± 5.52 mm and CTR <50.0%). For lesions <15 mm, our diagnostic yield was 62.5% (10/16) (data not shown), which was higher than previous studies (38–40%) [5, 18]. For lesions ≥15 mm, although our diagnostic yield of 81.3% (13/16) seemed slightly higher than those studies (63–74%) [5, 18], the lesions we included were <30 mm. Therefore, this study suggested that CBCT-guided CB combined with conventional biopsy is a promising biopsy method for GGN, especially for lesions <15 mm.

CBCT can recognise false-positive EBUS signs and improve the diagnostic yield for peripheral pulmonary nodules by about 20% [22, 23]. In univariate analysis, this study revealed that positive CBCT signs were a significant factor for the higher diagnostic yield (85.2%, 23/27), while inconclusive CBCT signs decreased the overall diagnostic yield to 71.9% (23/32). Although distinctive EBUS signs were obtained from inconclusive cases, the probe may not actually reach the GGO-predominant nodule. Initial bleeding or secretion may blur GGO lesions, resulting in false-positive EBUS signs that can mislead an operator [22]. In addition, a positive CT bronchus sign yielded a significantly higher diagnostic rate (85.7 vs. 45.5%, p = 0.035), something consistent with previous studies [19, 24]. With respect to the selection of biopsy sequence, FB followed by CB is usually deployed in solid lesions to facilitate biopsy efficiency [12, 13]. However, this study demonstrated that the CB-first biopsy sequence for GGO-predominant pulmonary nodules was related to a higher diagnostic yield and lower blood cell area in CB specimens (p < 0.05).

The presence of malignant GGO-predominant pulmonary nodules is indicative of pre-invasive adenocarcinoma, minimally invasive adenocarcinoma, and invasive adenocarcinoma with predominant lepidic pattern [25]. The histopathological diagnosis of GGO-predominant pulmonary nodules in small biopsies could be more challenging, given the difficulty of distinguishing normal and cancerous tissue with small tumour components and no obvious invasive features. Furthermore, specimens obtained by forceps were susceptible to compression and bleeding, which affected the pathological diagnosis. In contrast, CB can freeze the surrounding tissue, allowing for lateral biopsies and acquire large and artefact-free samples to provide a more accurate histopathological diagnosis [26, 27]. Oki et al. [12] found that the combination of FB and CB provided a specific diagnosis in 74.0% of cases, while the diagnostic yield of FB and CB was 54.0% and 62.0%, respectively. Kim et al. [13] demonstrated that CB increased the diagnostic yield of conventional biopsy for pulmonary nodules by 18.2%. Although CB can increase the diagnostic yield for solid pulmonary nodules, the improvement in GGO-predominant nodules was thus far unknown. Our study revealed that CB enhanced the diagnostic yield by 21.9% for GGO-predominant nodules compared to conventional biopsy. Especially, when the biopsy instruments were next to the lesions, the CB may have an advantage over FB in the diagnosis of GGO-predominant nodules due to the large sample and the lateral biopsy of the cryoprobe [26].

Several studies showed that specimens taken with a 1.1-mm cryoprobe were larger and of higher quality than other small biopsies [12‒14, 28]. For a freezing time of 6–8 s, the mean surface area of CB samples was 18.5 mm2, larger than that obtained with 1.5 (3.4 mm2) and 1.9-mm forceps (3.7 mm2) [13]. Our study showed a larger mean sample size of CB (2.18 mm2) than that of 1.5 (0.49 mm2) and 1.9-mm forceps (0.83 mm2). In addition, the quality of CB samples was better than that of FB samples in blood cell area and number of crush artefacts. However, there were some differences between ours and previously reported results. In our study, the sample size was measured on HE-stained sections, not in gross specimens. Although gross specimens are larger than sectioned ones, the latter are more frequently utilised for histopathological diagnosis. Our freezing time was short at 3–5 s. Freezing time has a significant effect on sample size [27, 29]; that is, the longer the freezing time, the larger the sample size acquired. However, given the risk of bleeding, the appropriate freezing time needs to be further investigated.

A previous study reported that the complication rate of CB plus conventional biopsy was 10.5%; most of them uncontrolled bleeding after CB (8.6%) that required using therapeutic bronchoscopy with/without a balloon catheter [13]. In our study, most mild bleeding (81.3%, 26/32) was controlled by simple bronchoscopic suction. The moderate bleeding observed in six cases (18.8%, 6/32) required prolonged suction or cold saline infusion without intubation of a balloon/bronchial blocker. This indicated the advantages of our bronchoscopy method (maintaining the bronchoscope or GS in place): (1) observing and handling bleeding in a timely manner; (2) avoiding repeated positioning targets; and (3) facilitating multiple sampling.

Our study has several limitations. Firstly, this retrospective study had a limited sample size and was performed at a single institution. A further prospective multicentre trial is required to compare the diagnostic performance of CB and FB. Secondly, our procedures were heterogeneous (e.g., bronchoscopy, navigational method, with/without a GS, biopsy instrument, and biopsy sequence). Because this study is in an early stage of exploration, the optimal guidance and biopsy method need to be investigated by randomised controlled trials. Last, considering the added dose and procedure time of CBCT, whether its overall benefit is greater than that of fluoroscopy alone needs to be further explored.

In conclusion, CB using a 1.1-mm cryoprobe combined with conventional biopsy under CBCT guidance can be a feasible and safe method for diagnosing GGO-predominant pulmonary nodules. CBCT has the potential to locate fluoroscopically invisible GGNs. The GGNs obtained by CB facilitate histopathological diagnosis.

The authors thank Dr. Haohua Teng (Department of Pathology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China) for valuable help and support for the pathological evaluation.

This study was reviewed and approved by the Ethics Committee of Shanghai Chest Hospital (IS23038). The need for informed consent was waived by the Ethics Committee of Shanghai Chest Hospital.

The authors have no conflicts of interest to report.

This work was supported by National Multi-disciplinary Treatment Project for Major Diseases [2020NMDTP]; Science and Technology Commission of Shanghai Municipality [21Y11901600].

The conception and design of the study: Jiayuan Sun and Meng Shi; the acquisition of data for the study: Junxiang Chen, Yongzheng Zhou, and Shuaiyang Liu; the analysis of data for the study: Zhihong Huang, Junxiang Chen, and Fangfang Xie; the interpretation of data for the study: Zhihong Huang, Junxiang Chen, Meng Shi, and Jiayuan Sun; drafting the work: Zhihong Huang and Junxiang Chen; revising the work: Jiayuan Sun and Meng Shi; and final approval of the version to be published: all authors.

Additional Information

Zhihong Huang and Junxiang Chen contributed equally to this work.

Data are not publicly available due to ethical reasons. Further enquiries can be directed to the corresponding author.

1.
Mazzone
PJ
,
Lam
L
.
Evaluating the patient with a pulmonary nodule: a review
.
JAMA
.
2022
;
327
(
3
):
264
73
.
2.
Hattori
A
,
Matsunaga
T
,
Takamochi
K
,
Oh
S
,
Suzuki
K
.
Oncological characteristics of radiological invasive adenocarcinoma with additional ground-glass nodules on initial thin-section computed tomography: comparison with solitary invasive adenocarcinoma
.
J Thorac Oncol
.
2016
;
11
(
5
):
729
36
.
3.
Howington
JA
,
Blum
MG
,
Chang
AC
,
Balekian
AA
,
Murthy
SC
.
Treatment of stage I and II non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines
.
Chest
.
2013
143
5 Suppl
e278S
313S
.
4.
Nadig
TR
,
Thomas
N
,
Nietert
PJ
,
Lozier
J
,
Tanner
NT
,
Wang Memoli
JS
.
Guided bronchoscopy for the evaluation of pulmonary lesions: an updated meta-analysis
.
Chest
.
2023
;
163
(
6
):
1589
98
.
5.
Ikezawa
Y
,
Sukoh
N
,
Shinagawa
N
,
Nakano
K
,
Oizumi
S
,
Nishimura
M
.
Endobronchial ultrasonography with a guide sheath for pure or mixed ground-glass opacity lesions
.
Respiration
.
2014
;
88
(
2
):
137
43
.
6.
Izumo
T
,
Sasada
S
,
Chavez
C
,
Tsuchida
T
.
The diagnostic utility of endobronchial ultrasonography with a guide sheath and tomosynthesis images for ground glass opacity pulmonary lesions
.
J Thorac Dis
.
2013
;
5
(
6
):
745
50
.
7.
Setser
R
,
Chintalapani
G
,
Bhadra
K
,
Casal
RF
.
Cone beam CT imaging for bronchoscopy: a technical review
.
J Thorac Dis
.
2020
;
12
(
12
):
7416
28
.
8.
Ali
EAA
,
Takizawa
H
,
Kawakita
N
,
Sawada
T
,
Tsuboi
M
,
Toba
H
.
Transbronchial biopsy using an ultrathin bronchoscope guided by cone-beam computed tomography and virtual bronchoscopic navigation in the diagnosis of pulmonary nodules
.
Respiration
.
2019
;
98
(
4
):
321
8
.
9.
Sobieszczyk
MJ
,
Yuan
Z
,
Li
W
,
Krimsky
W
.
Biopsy of peripheral lung nodules utilizing cone beam computer tomography with and without trans bronchial access tool: a retrospective analysis
.
J Thorac Dis
.
2018
;
10
(
10
):
5953
9
.
10.
Hvidtfeldt
M
,
Sverrild
A
,
Pulga
A
,
Frøssing
L
,
Silberbrandt
A
,
Sanden
C
.
Mucosal cryobiopsies: a new method for studying airway pathology in asthma
.
ERJ Open Res
.
2022
;
8
(
1
):
00666
2021
.
11.
Suzuki
M
,
Matsumoto
Y
,
Imabayashi
T
,
Teishikata
T
,
Tsuchida
T
,
Asamura
H
.
Cryobiopsy as a reliable technique for the preoperative identification of micropapillary/solid components in early-stage lung adenocarcinoma
.
Lung Cancer
.
2021
;
162
:
147
53
.
12.
Oki
M
,
Saka
H
,
Kogure
Y
,
Niwa
H
,
Yamada
A
,
Torii
A
.
Ultrathin bronchoscopic cryobiopsy of peripheral pulmonary lesions
.
Respirology
.
2023
;
28
(
2
):
143
51
.
13.
Kim
SH
,
Mok
J
,
Jo
EJ
,
Kim
MH
,
Lee
K
,
Kim
KU
.
The additive impact of transbronchial cryobiopsy using a 1.1-mm diameter cryoprobe on conventional biopsy for peripheral lung nodules
.
Cancer Res Treat
.
2023
;
55
(
2
):
506
12
.
14.
Jiang
S
,
Liu
X
,
Chen
J
,
Ma
H
,
Xie
F
,
Sun
J
.
A pilot study of the ultrathin cryoprobe in the diagnosis of peripheral pulmonary ground-glass opacity lesions
.
Transl Lung Cancer Res
.
2020
;
9
(
5
):
1963
73
.
15.
Zheng
PP
,
van der Weiden
M
,
Kros
JM
.
Fast tracking of co-localization of multiple markers by using the nanozoomer slide scanner and NDPViewer
.
J Cell Physiol
.
2014
;
229
(
8
):
967
73
.
16.
Vachani
A
,
Maldonado
F
,
Laxmanan
B
,
Kalsekar
I
,
Murgu
S
.
The impact of alternative approaches to diagnostic yield calculation in studies of bronchoscopy
.
Chest
.
2022
;
161
(
5
):
1426
8
.
17.
Folch
EE
,
Mahajan
AK
,
Oberg
CL
,
Maldonado
F
,
Toloza
E
,
Krimsky
WS
.
Standardized definitions of bleeding after transbronchial lung biopsy: a delphi consensus statement from the nashville Working Group
.
Chest
.
2020
;
158
(
1
):
393
400
.
18.
Ikezawa
Y
,
Shinagawa
N
,
Sukoh
N
,
Morimoto
M
,
Kikuchi
H
,
Watanabe
M
.
Usefulness of endobronchial ultrasonography with a guide sheath and virtual bronchoscopic navigation for ground-glass opacity lesions
.
Ann Thorac Surg
.
2017
;
103
(
2
):
470
5
.
19.
Nakai
T
,
Matsumoto
Y
,
Suzuk
F
,
Tsuchida
T
,
Izumo
T
.
Predictive factors for a successful diagnostic bronchoscopy of ground-glass nodules
.
Ann Thorac Med
.
2017
;
12
(
3
):
171
6
.
20.
Reisenauer
J
,
Duke
JD
,
Kern
R
,
Fernandez-Bussy
S
,
Edell
E
.
Combining shape-sensing robotic bronchoscopy with mobile three-dimensional imaging to verify tool-in-lesion and overcome divergence: a pilot study
.
Mayo Clin Proc Innov Qual Outcomes
.
2022
;
6
(
3
):
177
85
.
21.
Yu
KL
,
Yang
SM
,
Ko
HJ
,
Tsai
HY
,
Ko
JC
,
Lin
CK
.
Efficacy and safety of cone-beam computed tomography-derived augmented fluoroscopy combined with endobronchial ultrasound in peripheral pulmonary lesions
.
Respiration
.
2021
;
100
(
6
):
538
46
.
22.
Casal
RF
,
Sarkiss
M
,
Jones
AK
,
Stewart
J
,
Tam
A
,
Grosu
HB
.
Cone beam computed tomography-guided thin/ultrathin bronchoscopy for diagnosis of peripheral lung nodules: a prospective pilot study
.
J Thorac Dis
.
2018
;
10
(
12
):
6950
9
.
23.
Kheir
F
,
Thakore
SR
,
Uribe Becerra
JP
,
Tahboub
M
,
Kamat
R
,
Abdelghani
R
.
Cone-beam computed tomography-guided electromagnetic navigation for peripheral lung nodules
.
Respiration
.
2021
;
100
(
1
):
44
51
.
24.
Jiang
S
,
Xie
F
,
Mao
X
,
Ma
H
,
Sun
J
.
The value of navigation bronchoscopy in the diagnosis of peripheral pulmonary lesions: a meta-analysis
.
Thorac Cancer
.
2020
;
11
(
5
):
1191
201
.
25.
Travis
WD
,
Brambilla
E
,
Nicholson
AG
,
Yatabe
Y
,
Austin
JHM
,
Beasley
MB
.
The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification
.
J Thorac Oncol
.
2015
;
10
(
9
):
1243
60
.
26.
Kho
SS
,
Chan
SK
,
Yong
MC
,
Tie
ST
.
Performance of transbronchial cryobiopsy in eccentrically and adjacently orientated radial endobronchial ultrasound lesions
.
ERJ Open Res
.
2019
;
5
(
4
):
00135
2019
.
27.
Hetzel
J
,
Linzenbold
W
,
Boesmueller
H
,
Enderle
M
,
Poletti
V
.
Evaluation of efficacy of a new cryoprobe for transbronchial cryobiopsy: a randomized, controlled in vivo animal study
.
Respiration
.
2020
;
99
(
3
):
248
56
.
28.
Oberg
CL
,
Lau
RP
,
Folch
EE
,
He
T
,
Ronaghi
R
,
Susanto
I
.
Novel robotic-assisted cryobiopsy for peripheral pulmonary lesions
.
Lung
.
2022
;
200
(
6
):
737
45
.
29.
Yarmus
LB
,
Semaan
RW
,
Arias
SA
,
Feller-Kopman
D
,
Ortiz
R
,
Bösmüller
H
.
A randomized controlled trial of a novel sheath cryoprobe for bronchoscopic lung biopsy in a porcine model
.
Chest
.
2016
;
150
(
2
):
329
36
.