Diagnostic yield and histopathological assessment of endobronchial ultrasound-guided transbronchial mediastinal cryobiopsy: a prospective comparison of two tunnelling approaches

Background Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is widely used for mediastinal and hilar lymph node evaluation, although its diagnostic performance may be limited in rare tumours, benign diseases, lymphoproliferative disorders and for detailed molecular analyses. Data on the role of EBUS-guided transbronchial mediastinal cryobiopsy (EBUS-TMC) remain limited. Methods We conducted a prospective cohort study in 90 patients undergoing EBUS-TBNA and EBUS-TMC for mediastinal or peribronchial lesions ≥1 cm. Two tunnelling approaches using 22-G (SonoTip TopGain) and 19-G (ViziShot 2 FLEX) TBNA needles were employed. TBNA and cryobiopsy with a 1.1-mm cryoprobe were performed sequentially in the same lymph node. Results 45 patients were assigned to each group. The diagnostic yield of EBUS-TMC was 91.1% (22-G) and 93.3% (19-G) (p=0.0005 and p=0.006, respectively, versus EBUS-TBNA). For EBUS-TBNA, yields were 68.8% (22-G) and 77.7% (19-G) (p=0.28). Diagnostic yield improved significantly between TBNA and TMC for both 22-G (p=0.0005) and 19-G (p=0.006). Mean± sem procedure time was 25.1±3.9 min (22-G) versus 26.7±5.3 min (19-G) (p=0.11) and sample size was 3.5±0.9 mm (22-G) versus 3.6±1.1 mm (19-G) (p=0.76). Needle switching from 19-G to 22-G was required in two patients due to anatomical constraints. Re-tunnelling was performed in two (22-G) and four (19-G) patients. No severe bleeding or complications were observed. Conclusions EBUS-TMC is a safe, feasible and minimally invasive method with superior diagnostic yield and histological quality compared to EBUS-TBNA. Both 22-G and 19-G approaches are effective; 22-G is more adaptable for distal targets, while 19-G is suited for centrally located or accessible lesions. Combining EBUS-TBNA and EBUS-TMC may further enhance diagnostic accuracy in complex cases.