Abstract
Mediastinal lymphadenopathy poses a significant challenge in the differential diagnosis of numerous diseases, including tuberculosis. Transbronchial biopsy under endosonographic guidance (EBUS-TBNA) is a minimally invasive diagnostic method, widely recognized as a preferred approach for invasive diagnostics. However, in the Russian Federation this method is taken with caution, and videothoracoscopic biopsy is often preferred. Aim: to investigate the diagnostic value of endosonographic patterns identified during transbronchial endosonography of mediastinal lymph nodes in the differential diagnosis of tuberculous involvement. Materials and methods: a retrospective analysis was conducted on a prospectively recruited cohort of 185 patients with mediastinal lymphadenopathy who underwent EBUS-TBNA followed by videothoracoscopic biopsy for diagnostic verification. Ultrasonographic characteristics of lymph nodes, including shape, margins, echogenicity, presence of central structures, signs of necrosis, calcification, and lymph node fusion, were analyzed. An economic evaluation was performed using a cost-minimization approach to compare EBUS-TBNA and videothoracoscopic biopsy. Results. Tuberculosis of intrathoracic lymph nodes was confirmed in 23 patients (12.4%), sarcoidosis stages 1–2 in 137 (74.1%), metastatic involvement in 11 (5.9%), and lymph node hyperplasia in 11 (5.9%). Statistically significant predictors of tuberculous involvement included oval/irregular shape (p=0.009), unclear margins (p=0.012), necrosis (p=0.038), and lymph node fusion (p=0.020). The combination of these features increased the likelihood of tuberculosis by 3.1–4.8 times (positive predictive value 30.2%), with a negative predictive value of 93%. The median cost of videothoracoscopic biopsy was 123.6 c.u., compared to 72.3 c.u. for EBUS-TBNA. Conclusion. Characteristic endosonographic patterns of tuberculous mediastinal lymph node involvement were identified, serving as additional differential diagnostic criteria. The absence of these patterns reliably excludes tuberculosis. However, definitive verification requires morphological and/or molecular-genetic confirmation, which is significantly more cost-effective than initial videothoracoscopic biopsy.