Sonographic target characterization in single-pass endobronchial ultrasound-guided transbronchial needle aspiration

We read with great interest the recent study by Qian et al. (1), which examined determinants of specimen adequacy at the individual-puncture level during endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA). By shifting the analytical focus from the lymph-node level to individual punctures, the authors addressed a clinically relevant procedural question: how to optimize specimen quality while minimizing unnecessary passes. Their finding that the needle pathway (central vs. peripheral) was not independently associated with the macroscopic on-site evaluation (MOSE) score is particularly thought-provoking.

We agree that a “central” puncture should not be regarded as intrinsically superior to a “peripheral” one. However, this negative finding should not be interpreted as evidence that intranodal target selection is unimportant. In our view, the central/peripheral classification primarily reflects the geometric location of the needle within the node rather than the endosonographic characteristics of the tissue actually sampled. For ultrasound-guided procedures, the clinically relevant question may concern not only where the needle tip is located within the node but also which sonographic tissue component is sampled. A conceptual comparison between geometric pathway classification and real-time endosonographic target characterization is shown in Figure 1.

Figure 1.

Conceptual comparison between geometric pathway classification and real-time endosonographic target characterization in EBUS-TBNA. (A) Central and peripheral needle pathways describe the geometric location of sampling within a lymph node. (B) Real-time endosonographic assessment may additionally characterize heterogeneous target regions according to echotexture, suspected necrosis, hyperechoic foci, vascular structures, and elastographic features, with attention to the accessible solid-appearing component. This framework is proposed for prospective evaluation of its association with single-pass specimen adequacy and is not a prescriptive sampling algorithm. EBUS-TBNA, endobronchial ultrasound-guided transbronchial needle aspiration.

Mediastinal and hilar lymph nodes frequently exhibit substantial intranodal heterogeneity on real-time EBUS. Conventional endosonographic features, including echotexture, suspected necrosis, hyperechoic foci, and vascular patterns, may help characterize the target region at the time of sampling (2-4). In addition, elastographic findings may provide supplementary information on intranodal heterogeneity and target selection (5). Although malignancy prediction is distinct from specimen adequacy, these studies suggest that biologically relevant heterogeneity can be recognized during real-time EBUS. Future studies should therefore complement geometric pathway classification with endosonographic target characterization and prospectively document the sonographic target before each pass. Standardized per-pass annotation could include echogenicity/homogeneity, necrotic change, hyperechoic foci, Doppler vascularity, elastographic features, and the estimated extent of viable solid tissue. In this context, a real-time strategy that considers accessible solid components in addition to geometric pathway classification may be better aligned with the nature of ultrasound-guided sampling. This proposition requires prospective validation; nevertheless, the present negative result suggests that simple geometric partitioning does not capture biologically meaningful intranodal heterogeneity.

The finding that final malignant pathology as an independent correlate of higher MOSE scores also merits caution. Although biologically plausible, final pathology is not available when the first puncture is planned and therefore cannot directly guide intra-procedural target selection. Moreover, if malignancy was established mainly from the same EBUS-TBNA specimens that were evaluated by MOSE, specimen quality itself may have influenced the probability of pathological confirmation. Clarification of the diagnostic reference standard and sensitivity analyses restricted to independently confirmed cases would strengthen interpretation of this association.

Finally, because 636 punctures were performed in 426 lymph nodes, some observations represented subsequent passes within the same node. These later passes may have been informed by the preceding MOSE result and accompanied by intentional changes in depth, trajectory, or target region. Although mixed-effects modeling accounts for clustering, it does not capture this sequential adaptive process. Analyses restricted to the first pass per node, or models incorporating pass order and prior MOSE findings, may better identify determinants of true first-pass adequacy.

We believe that the next step is to establish an integrated framework linking real-time endosonographic features with macroscopic, diagnostic, and molecular specimen adequacy. Standardized capture of sonographic features and per-pass target annotation would enable prospective evaluation of whether feature-informed target selection is associated with improved first-pass adequacy beyond geometric pathway classification. Such efforts may better define the contribution of dedicated ultrasound expertise to precision guidance in EBUS-TBNA.

Supplementary

The article’s supplementary files as

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.