Integrative use of conventional mode and elastography in endobronchial ultrasound for the diagnosis of malignant intrathoracic lymphadenopathy

Introduction

Background

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a minimally invasive alternative to mediastinoscopy for diagnosing intrathoracic lymphadenopathy, offering up to 90% accuracy in the diagnosis and staging of lung cancer (1,2). Ultrasonographic features assist in differentiating malignant from benign lymph nodes, provide valuable guidance in selecting lymph nodes for sampling (3). Conversely, the presence of features suggestive of benignity may help reduce unnecessary sampling. Nonetheless, lymph node sampling should ultimately be guided by clinical context and preliminary diagnosis, as sonographic features alone cannot substitute for histopathological confirmation.

Rationale and knowledge gap

In conventional B-mode, several features suggestive of malignancy such as round shape, distinct margin, heterogeneous echogenicity, and the presence of a coagulation necrosis sign, have reported accuracy ranging from 63.8% to 86.0% (4). Combined features have also been incorporated into an aggregate scoring system, requiring at least two of the four predictive factors—round shape, absence of CHS, presence of matting, and nonhilar vascular pattern perfusion—and demonstrating a diagnostic accuracy of 82.7% (5). Power or color Doppler further contributes to lymph node assessment by evaluating vascular grading. Grades II (characterized by a few punctiform or rod-shaped flow signals) and III (indicating rich blood flow) demonstrated a diagnostic accuracy of 78%, while the presence of a bronchial artery (BA) inflow sign yields an accuracy of 80.3% (6). However, diagnosis using the conventional mode relies on multiple parameters, most of which are qualitative in nature and potentially subject to observer variability.

Elastography is a novel EBUS modality that visualizes tissue stiffness through color-coded mapping, typically using a three-color classification system—predominantly non-blue (benign), part-blue/part non-blue, and predominantly blue (malignant)—with reported diagnostic accuracy ranging from 85.2% to 96.7% (7-10). However, the part-blue/part non-blue pattern remains inconclusive, posing limitations in interpretation. Quantitative parameters, including strain ratio (SR), stiff area ratio, blue color proportion, blue pixel ratio, and mean strain histogram, have been utilized to enhance diagnostic reliability. However, several of these parameters require, specialized image-analysis software, potentially limiting their use in routine clinical practice. The SR is calculated by comparing the strain of a reference area surrounding the lymph node with that of the lymph node, with values directly obtained from the ultrasound processor. However, SR is subject to considerable variability and has shown lower accuracy than the color-classification method. Published studies report a wide range of cut-off values 2.47–32.07, indicating the need for further work to establish consistent thresholds for clinical application (11-14).

Objective

This study aims to evaluate the diagnostic accuracy of an integrative approach combining conventional EBUS features and elastography in discriminating malignant intrathoracic lymph nodes. We present this article in accordance with the STARD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1386/rc).

Methods

Participants

We conducted a retrospective cross-sectional study to evaluate the added diagnostic value of combining the conventional mode with elastography in patients who underwent EBUS-TBNA for the diagnosis of intrathoracic lymphadenopathy at Rajavithi Hospital, Thailand, between July 2015 and December 2018, using a consecutive series of eligible cases during the study period. The eligibility criteria included: (I) patients age ≥15 years; (II) intrathoracic lymphadenopathy with a short-axis diameter ≥0.5 cm on chest computed tomography; and (III) no contraindications to the EBUS-TBNA procedure. Patients were excluded if the cytopathological diagnosis from EBUS-TBNA was inadequate or non-diagnostic, or if video recordings of the EBUS procedure were unavailable. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethical Committees for Research Involving Human Subjects, Rajavithi Hospital, Thailand (No. 003/2563). Informed consent was waived as this was a retrospective study.

EBUS-TBNA procedure and EBUS features

All procedures were performed according to the standard protocol of our institution, which is largely consistent with the recommendation, under local anesthesia and conscious sedation, using a convex probe EBUS bronchoscope (BF-UC180F; Olympus, Tokyo, Japan), and an ultrasound processor (EU-ME2; Olympus, Tokyo, Japan), with a 10 MHz ultrasound setting. A 21- or 22-gauge TBNA needle (NA-201SX-4021/NA-201SX-4022; Olympus) was employed, as previous evidence demonstrated no significant difference in diagnostic yield between the two needle sizes (15). The procedures were conducted by a single interventional pulmonologist. Intrathoracic lymph nodes were evaluated sequentially using conventional B-mode, power/color Doppler mode, elastographic dominant color pattern and finally, SR. The entire procedure was continuously video recorded directly from the ultrasound processor for documentation.

Conventional mode features

Conventional B mode included the following features, as described by Fujiwara (4) (Figure 1): (I) short-axis diameter (in mm); (II) shape-round if the long-to-short axis ratio <1.5, and oval if ≥1.5; (III) margin-distinct if >50% of hyperechoic border was clearly identified, and indistinct if the margin was unclear; (IV) echogenicity-homogenous (uniform echogenicity) or inhomogeneous (non-uniform echogenicity); (V) central hilar structure (CHS; a linear hyperechoic area in the center of the lymph node)—presence or absence; (VI) coagulation necrosis (hypoechoic area within lymph node)—presence or absence.

Figure 1 Conventional B-mode features of intrathoracic lymph nodes. (A) Round shape. (B) Oval shape. (C) Distinct margin. (D) Indistinct margin. (E) Inhomogeneous echogenicity. (F) Homogeneous echogenicity. (G) Coagulation necrosis. (H) Central hilar structure.

Power/color Doppler mode included the following features (4,6) (Figure 2): (I) vascularity-absent or minimal vascularity (no blood flow or minimal signals) or abundant vascularity (multiple, rich, strip-like blood flow signals); (II) BA inflow sign—a presence or absence of a blue signal indicating flow from BA toward lymph node. The features of conventional mode favor malignancy include: size >10 mm, round shape, distinct margin, heterogeneous echogenicity, absence of CHS, presence of coagulation necrosis, vascular grade II or III, and presence of BA inflow sign.

Figure 2 Power/color Doppler mode of intrathoracic lymph nodes. (A) Vascular grading: no/minimal vascularity. (B) Vascular grading: abundant vascularity. (C) Bronchial artery inflow sign.

Elastographic features

Elastography included the following features, as described by Izumo (8) (Figure 3): (I) dominant color pattern of the entire lymph node-classified as type I (predominantly non-blue), type II (part-blue/part non-blue), and type III (predominantly blue); (II) SR, calculated by selecting the largest area within the lymph node as the region of interest (A), and the surrounding tissue as the reference area (B). The SR (B/A) was automatically calculated by the ultrasound processor. The features of elastography favor malignancy include: dominant color type III (predominantly blue) pattern, and the optimal cut-off SR.

Figure 3 Elastographic features of intrathoracic lymph nodes, demonstrating in three-dominant color pattern and strain ratio. (A) Type I predominantly non-blue. (B) Type II part-blue/part non-blue. (C) Type III predominantly blue. (D) Strain ratio (B/A): A is the ROI within the lymph node, and B is the reference area in the surrounding tissue. ROI, region of interest.

Transbronchial needle aspiration (TBNA) was performed on all accessible lymph nodes. For each targeted lymph node, a minimum of three passes were obtained to optimize diagnostic yield. All specimens were submitted to cytological, histopathological, and microbiological examinations after the procedure without rapid on-site evaluation (ROSE).

Diagnosis of intrathoracic lymphadenopathy

Each lymph node station was assessed and analyzed independently. The diagnosis was determined based on cytological or histopathological findings obtained from EBUS-TBNA specimens, which served as the reference standard. Microbiology studies were performed to diagnosed infectious etiologies. The cytopathologists were blinded to all EBUS imaging findings. The diagnosis based on EBUS features cannot replace tissue sampling. Lymph nodes were defined as malignant if malignancy was confirmed by cytology and/or histopathology. Benign nodes were defined by: (I) cytology and/or histopathology indicating infection, inflammation, reactive hyperplasia, or absence of malignancy; (II) microbiological confirmation of infection; or (III) no clinical or radiological progression was observed after ≥6 months of follow-up. Lymph nodes with inadequate or inconclusive results were excluded from the analysis.

Data collection

All procedures were continuously video recorded and subsequently reviewed by the performing operator under blinded conditions with respect to clinical information and final diagnoses, ensuring that all cases were assessed in an identical manner and minimizing recall and information bias.

Statistical analysis

All the data were analyzed using STATA version 17.0 (StataCorp LLC, College Station, TX, USA). Lymph node characteristics were reported as mean ± standard deviation (SD) or median with interquartile range (IQR) for continuous variables, and as frequency with percentage for categorical variables. Comparisons between malignant and benign nodes were performed using Student’s t-test or Mann-Whitney U test, as appropriate. Diagnostic performance was evaluated using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and the area under the receiver operating characteristic (ROC) curve (AUC). Multivariable logistic regression analysis was used to evaluate diagnostic odds ratio (DOR). Variables without multicollinearity, with either a P<0.2 or an AUC >0.60, were included in the multivariable model. The optimal SR cut-off value was determined by ROC curve analysis to differentiate malignant from benign lymph nodes. AUC values for individual and combined modalities were compared to assess incremental diagnostic value. Exploratory subgroup analyses were performed to evaluate the diagnostic performance of EBUS features according to lymph node size categories (<10, 10–20, and >20 mm). Partially missing EBUS feature data were handled using complete case analysis.

Results

Characteristic of intrathoracic lymph nodes

Of the 104 participants included, 68 (65.4%) were female, with a mean age of 58.8±15.7 years. A total of 226 lymph nodes were initially assessed; 16 nodes were excluded—8 due to unavailable video recordings and 8 due to non-diagnostic results—leaving 210 lymph nodes for analysis (Figure 4). The mean short-axis diameter was 17.4±9.5 mm. The distribution of lymph node stations is summarized in Table 1. Among these, 124 (59%) were malignant and 86 (41%) were benign. No serious complications were observed during or after the EBUS-TBNA procedure.

Figure 4 Diagnostic flow diagram of intrathoracic lymph nodes evaluated by EBUS-TBNA. EBUS-TBNA, endobronchial ultrasound-guided transbronchial needle aspiration.

EBUS features in the diagnosis of intrathoracic lymph nodes

In the conventional mode, EBUS features significantly associated with malignant lymph nodes included a short-axis >20 mm (48.4% in malignant vs. 11.6% in benign; P<0.001), round shape (88.4% vs. 59.5%; P<0.001), absence of CHS (96.0% vs. 74.4%; P<0.001), and the presence of coagulation necrosis sign (69.1% vs. 34.9%; P<0.001). No significant associations were observed for margin, echogenicity, vascular grading, or the presence of BA inflow sign. In elastography, a predominantly blue pattern was suggestive of malignancy (98.4% vs. 46.5%; P<0.001). The median SR for malignant nodes was 44.0 (IQR, 22.4–136.3) compared to 8.4 (IQR, 3.8–24.5) for benign nodes (Table 2).

Diagnostic performance of EBUS features

In the conventional mode, lymph node size was categorized into three groups: <10, 10–20, and >20 mm. Vascular grading was simplified into two categories: no or minimal vascularity (grades 0/I) and abundant vascularity (grades II/III). Elastography patterns were assessed based on the predominant color using a three-color classification system. The diagnostic performance of EBUS features was evaluated in terms of sensitivity, specificity, PPV, NPV, accuracy, and the AUC.

The absence of CHS, the presence of coagulation necrosis, and a round shape were associated with high sensitivity in detecting malignant lymph nodes. The absence of CHS demonstrated a sensitivity of 96.0%, specificity of 25.6%, PPV of 65.0%, NPV of 81.5%, accuracy of 67.1% and an AUC of 0.61 (95% CI: 0.56–0.66). The presence of a coagulation necrosis yielded a sensitivity of 94.4%, specificity of 9.3%, PPV of 60.0%, NPV of 53.3%, accuracy of 67.4%, and AUC of 0.67 (95% CI: 0.61–0.74). A round shape showed a sensitivity of 88.4%, specificity of 40.5%, PPV of 68.2%, NPV of 70.8%, accuracy of 61.4% and an AUC of 0.64 (95% CI: 0.58–0.7). The presence of the BA inflow sign, demonstrated the highest specificity (96.5%) among conventional mode features, though its sensitivity was limited. Similarly, a lymph node size >20 mm demonstrated high specificity (88.4%) but limited sensitivity, and accuracy. Margin, echogenicity, and vascular grading demonstrated low diagnostic accuracy for discriminating malignancy in this study.

In elastography, the part-blue/part non-blue group (type II), due to its inconclusive nature and the need for tissue sampling, was combined with the predominantly blue group (type III) and classified as “non-type I”. This category demonstrated a high sensitivity of 98.4%, specificity of 23.3%, PPV of 64.9%, NPV of 90.9%, accuracy of 67.6%, and an AUC of 0.61 (95% CI: 0.56–0.65). A SR >12.48 yielded a high sensitivity of 92.7%, accuracy of 80.0%, and an AUC of 0.77; 95% CI: 0.72–0.83. The diagnostic performance of both conventional mode and elastography is summarized in Table 3.

Multivariable analysis of EBUS features for malignant discrimination

Both EBUS features, conventional mode and elastography were included in the logistic regression analysis after confirming the absence of multicollinearity. Univariable analysis identified predictive parameters from both conventional mode and elastography. For the conventional mode, these included absence of CHS, lymph node size >20 mm, round shape, and presence of coagulation necrosis. From elastography, non-type I pattern (types II and III) and a SR >12.48 were significant predictors. In the multivariable analysis, a SR >12.48, absence of CHS, round shape, presence of coagulation necrosis, and lymph node size >20 mm remained significant predictors of malignancy, with DORs of 11.0 (P<0.001), 4.8 (P=0.01), 3.5 (P=0.01), 3.4 (P=0.003), and 3.3 (P=0.01), respectively (Table 4).

Integrative use of conventional mode and elastography in the diagnosis of malignant intrathoracic lymphadenopathy

The diagnostic performance of the combined model—incorporating an elastographic SR >12.48 along with conventional mode features including lymph node size >20 mm, presence of coagulation necrosis, round shape, and absence of CHS—was compared with individual models using either conventional mode or elastography alone. The integrative use of EBUS features significantly improved diagnostic performance for distinguishing malignant lymph nodes, with an AUC of 0.90 (95% CI: 0.85–0.94), compared to 0.83 (95% CI: 0.78–0.89; P=0.002) for conventional mode alone, and 0.79 (95% CI: 0.73–0.85; P<0.001) for elastography alone (Figure 5A). Moreover, the integration of SR >12.48 with any single conventional feature—lymph node size >20 mm, coagulation necrosis, round shape, or absence of CHS—yielded consistently high and comparable diagnostic accuracy (AUC 0.84, 0.84, 0.83, and 0.81), with no significant differences among combinations (P=0.40) (Figure 5B).

Figure 5 ROC curves of integrative EBUS features for differentiating malignant lymph nodes. (A) ROC curves demonstrating the diagnostic accuracy of the EBUS combination model compared with conventional mode alone and elastography alone. (B) ROC curves illustrating the diagnostic accuracy of strain ratio >12.48 in combination with individual conventional features (size >20 mm, round shape, absence of CHS, and coagulation necrosis). AUC, area under the receiver operating characteristic curve; CHS, central hilar structure; EBUS, endobronchial ultrasound; ROC, receiver operating characteristic; SR, strain ratio.

Discussion

The study evaluated the diagnostic accuracy of both conventional mode and elastography in differentiating malignant from benign lymph nodes, and examined the added value of combining both methods compared to using either one alone. Our findings support that combining both methods improves diagnostic accuracy.

Conventional mode assessment involves multiple qualitative parameters, some of which are difficult to distinguish or rely on subjective interpretation, such as vascular grading. As a result, interpretation remains operator-dependent, potentially affecting diagnostic consistency. In this study, diagnosis of intrathoracic lymphadenopathy using conventional EBUS features showed that absence of CHS, round shape, presence of coagulation necrosis, and short-axis size >20 mm were significantly associated with malignant lymph nodes. These findings are consistent with those of Fujiwara (4). However, distinct margin, inhomogeneous echogenicity, vascular grading, and the presence of the BA inflow sign were not significantly associated with malignancy in our study, which contrasts with the findings reported by Fujiwara and He (4,14). In this study, the absence of CHS, presence of coagulation necrosis, and round shape demonstrated high sensitivity for malignancy (96.0%, 94.4% and 88.4% respectively) but only moderate diagnostic accuracy (67.1%, 67.4% and 61.4% respectively), suggesting their potential utility as screening tools.

Vascular grades II and III, as described in the study by Nakajima et al. (6) were typically observed in malignant lymph nodes. In this study, abundant vascularity was more frequently observed in malignant nodes than in benign ones, but the difference was not statistically significant. Notably, benign lymph nodes particularly those associated with inflammation or infection can also exhibit marked vascularity. In our findings, among the 19 lymph nodes diagnosed with tuberculosis, 10 (52.6%) demonstrated abundant vascularity, while 9 (47.4%) showed no or minimal blood flow. The BA inflow sign is commonly seen in malignant lymph nodes and associated with abundant vascularity, indicating active neovascularization. In this study, the presence of BA inflow sign showed high specificity of 96.5% and PPV of 82.4% but limited sensitivity, indicating its usefulness in confirming malignant lymph nodes. However, BA inflow sign is observed in only 17 nodes, which may reflect underestimation. This may be explained, in part, by the relatively low proportion of lymph nodes with abundant vascularity (50.5%) and by the presence of coagulation necrosis in a considerable number of nodes (54.8%), which may have limited visualization on Doppler imaging. Further prospective studies with larger sample sizes are necessary to validate its diagnostic utility.

For elastography, a predominantly blue pattern (type III) was observed in 98.4% of malignant lymph nodes, consistent with our previous study (94.1%) (9). Classified as non-type I, the grouping of type II (part-blue/part non-blue, inconclusive) and type III (predominantly blue, highly suggestive of malignancy) was associated with malignant nodes in univariable analysis but not in multivariable analysis. Nevertheless, tissue confirmation remains essential in clinical practice, given the inherent limitations of the color classification system, to avoid misdiagnosing malignancy. False positives may occur in benign node with calcification and fibrotic component, whereas false negatives may arise in malignant nodes with necrosis or hypervascularity (10). Consequently, the optimal cut-off value of SR has also varied across studies.

In our previous study, a SR >2.5 provided a sensitivity of 100% for predicting malignant nodes, whereas the present study demonstrated a higher optimal cut-off value of 12.48 with a comparably high sensitivity of 92.7%. However, SR remains a parameter with substantial variability and a lack of standardized cut-off thresholds. Potential limitations include the influence of underlying etiologies, lymph node locations adjacent to vascular structures or affected by respiratory movement, inconsistency in region-of-interest selection, and operator-dependent interpretation. These challenges are particularly evident in very small lymph nodes (<10 mm), where both conventional mode and elastography may provide limited diagnostic information. By contrast, in lymph nodes of 10–20 mm, a size frequently encountered in clinical practice, predictive EBUS features for malignancy were consistent with those of larger nodes (>20 mm), suggesting their applicability across different size ranges (Table 5). Despite variability in cut-off thresholds, both qualitative and quantitative elastographic measurements demonstrated favorable diagnostic performance. Consistent with this, Wang et al. (16) reported superior diagnostic performance of elastography compared with conventional EBUS, although reliance on a single technique alone may be insufficient for optimal accuracy (17).

In the present study, we therefore aimed to integrate fewer parameters that are simple, reliable, and suitable for real-time evaluation during the EBUS-TBNA procedure. A combined model incorporating quantitative elastographic SR >12.8 with qualitative conventional EBUS features including absence of CHS, round shape, presence of coagulation necrosis, and lymph node size >20 mm enhanced the diagnostic accuracy for predicting malignant lymph nodes. Notably, comparable accuracy was achieved even when the SR was combined with only a single conventional feature (Figure 5). Representative images demonstrating lymph node evaluation using the integrative model that combines quantitative elastography and conventional EBUS features are shown in Figure 6.

Figure 6 Integrative use of conventional mode and elastography. (A-C) Small cell lung cancer; (D-F) reactive node. (A) Conventional mode: lymph node size 3 cm, round shape, presence of coagulation necrosis, and absence of central hilar structure. (B) Power/color Doppler mode: abundant vascularity and presence of bronchial artery inflow sign. (C) Elastography: type II color pattern (part-blue/part non-blue) with a strain ratio of 16.92. (D) Lymph node size 1 cm, oval shape, presence of central hilar structure. (E) Power/color Doppler mode: no or minimal vascularity. (F) Elastography: type I color pattern (predominantly non-blue) with a strain ratio of 1.89.

Taken together, our findings highlight the potential role of integrative EBUS features as complementary tools in clinical practice. In the diagnosis and staging of lung cancer, suspicious lymph nodes require sampling regardless of EBUS features, to avoid missed diagnoses. By contrast, in benign conditions such as tuberculosis in endemic regions, EBUS features suggestive of benignity may help reduce unnecessary punctures. Similarly, in suspected extrathoracic metastases with multiple lymph node stations, EBUS features may help optimize target selection, allowing sampling to be directed toward the most suspicious nodes instead of all stations.

There are several notable limitations. First, its retrospective design limited the completeness of video recording. Second, being a single-center, single-operator study may limit generalizability and introduce operator-dependent, recall and information bias. To minimize these potential biases, patient identifiers, clinical data, and final diagnoses were blinded. All video recordings were reviewed in a sequential manner, beginning with conventional mode and followed by elastography, to ensure consistency across cases and to minimize information bias. Third, for nodes classified as indefinite benign (absence of malignancy), pathological confirmation could not be obtained through other modalities; therefore, only clinical and radiologic follow-up was conducted. Fourth, this study has a temporal limitation, as the dataset was collected earlier and reanalyzed later. Nevertheless, the findings remain clinically relevant and may help optimize sampling strategies, minimize unnecessary procedures, and support the selective use of alternative modalities such as EBUS-guided transbronchial mediastinal cryobiopsy or intranodal forceps biopsy, particularly in cases where sarcoidosis or malignant lymphoma is suspected. Future studies with multiple operators, multicenter, prospective designs, a larger study population, and interobserver variability assessment are warranted to better reflect clinical practice and improve generalizability.

Conclusions

The integrative use of EBUS features, combining conventional mode and novel elastography, offers a non-invasive approach incorporating selected high-accuracy parameters that provides preliminary diagnostic information. This method may facilitate lymph node targeting, optimize sampling strategies, and support clinical decision-making, especially in regions where benign and malignant etiologies frequently overlap.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1386/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1386/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1386/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1386/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethical Committees for Research Involving Human Subjects, Rajavithi Hospital, Thailand (No. 003/2563). Informed consent was waived because this retrospective study used anonymized patient data.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Detterbeck FC, Jantz MA, Wallace M, et al. Invasive mediastinal staging of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132:202S-220S.
2. Silvestri GA, Gonzalez AV, Jantz MA, et al. Methods for staging 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:e211S-250S.
3. Wahidi MM, Herth F, Yasufuku K, et al. Technical Aspects of Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration: CHEST Guideline and Expert Panel Report. Chest 2016;149:816-35. [Crossref] [PubMed]
4. Fujiwara T, Yasufuku K, Nakajima T, et al. The utility of sonographic features during endobronchial ultrasound-guided transbronchial needle aspiration for lymph node staging in patients with lung cancer: a standard endobronchial ultrasound image classification system. Chest 2010;138:641-7. [Crossref] [PubMed]
5. Wang L, Wu W, Hu Y, et al. Sonographic Features of Endobronchial Ultrasonography Predict Intrathoracic Lymph Node Metastasis in Lung Cancer Patients. Ann Thorac Surg 2015;100:1203-9. [Crossref] [PubMed]
6. Nakajima T, Anayama T, Shingyoji M, et al. Vascular image patterns of lymph nodes for the prediction of metastatic disease during EBUS-TBNA for mediastinal staging of lung cancer. J Thorac Oncol 2012;7:1009-14. [Crossref] [PubMed]
7. Madan K, Madan M, Iyer H, et al. Utility of Elastography for Differentiating Malignant and Benign Lymph Nodes During EBUS-TBNA: A Systematic Review and Meta-analysis. J Bronchology Interv Pulmonol 2022;29:18-33. [Crossref] [PubMed]
8. Izumo T, Sasada S, Chavez C, et al. Endobronchial ultrasound elastography in the diagnosis of mediastinal and hilar lymph nodes. Jpn J Clin Oncol 2014;44:956-62. [Crossref] [PubMed]
9. Korrungruang P, Boonsarngsuk V. Diagnostic value of endobronchial ultrasound elastography for the differentiation of benign and malignant intrathoracic lymph nodes. Respirology 2017;22:972-7. [Crossref] [PubMed]
10. Lin CK, Yu KL, Chang LY, et al. Differentiating malignant and benign lymph nodes using endobronchial ultrasound elastography. J Formos Med Assoc 2019;118:436-43. [Crossref] [PubMed]
11. Çağlayan B, İliaz S, Bulutay P, et al. The role of endobronchial ultrasonography elastography for predicting malignancy. Turk Gogus Kalp Damar Cerrahisi Derg 2020;28:158-65. [Crossref] [PubMed]
12. He HY, Huang M, Zhu J, et al. Endobronchial Ultrasound Elastography for Diagnosing Mediastinal and Hilar Lymph Nodes. Chin Med J (Engl) 2015;128:2720-5. [Crossref] [PubMed]
13. Rozman A, Malovrh MM, Adamic K, et al. Endobronchial ultrasound elastography strain ratio for mediastinal lymph node diagnosis. Radiol Oncol 2015;49:334-40. [Crossref] [PubMed]
14. He HY, Chen JL, Ma H, et al. Value of Endobronchial Ultrasound Elastography in Diagnosis of Central Lung Lesions. Med Sci Monit 2017;23:3269-75. [Crossref] [PubMed]
15. Yarmus LB, Akulian J, Lechtzin N, et al. Comparison of 21-gauge and 22-gauge aspiration needle in endobronchial ultrasound-guided transbronchial needle aspiration: results of the American College of Chest Physicians Quality Improvement Registry, Education, and Evaluation Registry. Chest 2013;143:1036-43. [Crossref] [PubMed]
16. Wang Y, Zhao Z, Zhu M, et al. Diagnostic value of endobronchial ultrasound elastography in differentiating between benign and malignant hilar and mediastinal lymph nodes: a retrospective study. Quant Imaging Med Surg 2023;13:4648-62. [Crossref] [PubMed]
17. Zhou X, Li Y. Diagnostic utility of endobronchial ultrasound elastography for detecting benign and malignant lymph nodes: a retrospective study. J Thorac Dis 2025;17:614-22. [Crossref] [PubMed]