Abstract:

Objective: To describe the real-world treatment characteristics of inpatients with lymph node tuberculosis at a single center, with a focus on the distribution of total anti-tuberculosis treatment duration, local lesion complexity, indications for surgical intervention, and their relationship with prolonged treatment duration, thereby providing a reference for standardized clinical management. Methods: Medical records of 290 inpatients with lymph node tuberculosis treated at Beijing Chest Hospital, Capital Medical University, from 2019 to 2024 were retrospectively collected to analyze treatment modalities and total treatment duration. Among them, 288 patients had more detailed records of the diagnostic and treatment process and were further included in supplementary analyses of relapse, newly developed lesions during treatment, multiple lesions, reasons for surgery, and events occurring during treatment. A simplified stratification of lesion complexity was constructed based on three indicators—relapse, newly developed lesions, and multiple lesions—and treatment duration was descriptively compared among different subgroups. Results: Among the 290 patients, 245 underwent surgical treatment, corresponding to a surgical rate of 84.5%. Among 274 patients with evaluable total treatment duration, the median total treatment duration was 13.00 (8.00, 19.00) months. Treatment duration was ≥10 months in 186 patients (67.9%), ≥18 months in 97 patients (35.4%), and ≥24 months in 30 patients (10.9%). The median total treatment duration was 12.50 (7.00, 19.00) months in the surgical group and 17.50 (12.00, 19.25) months in the non-surgical group. Among the 288 patients included in the supplementary analyses, relapse occurred in 20 cases (6.9%), newly developed lesions during treatment in 24 cases (8.3%), and multiple lesions in 62 cases (21.5%); 81 patients (28.1%) had at least one complex manifestation. The median treatment duration was 15.00 (12.00, 24.00) months in patients with relapse and 15.00 (10.00, 20.25) months in those with multiple lesions, both of which were longer than those in patients without relapse (13.00 (7.50, 18.00) months) and those without multiple lesions (13.00 (7.00, 18.00) months), respectively. The median treatment duration was 15.00 (10.00, 21.00) months in patients with at least one complex manifestation, compared with 12.00 (7.00, 18.00) months in those without complex manifestations. Among 247 patients with documented reasons for surgery, rupture/liquefaction/wound-related complications, relapse/newly developed/multiple lesions, mass enlargement or lesion progression, and poor response to drug treatment/prolonged non-healing together accounted for 67.6% (167/247), suggesting that the high rate of surgical intervention was closely related to the concentration of patients with complex local lesions. Conclusion: The duration of anti-tuberculosis treatment among inpatients with lymph node tuberculosis at our center was overall longer than the routinely recommended regimen and showed a prominent long-tail distribution. Relapse, newly developed lesions, multiple lesions, and local lesion progression were closely associated with longer treatment duration and may serve as important clinical indicators of treatment difficulty. The high surgical rate observed in this study more likely reflects the concentration of patients with complex lesions among hospitalized and surgically managed cases and should not be interpreted as representing a routine treatment choice for patients with lymph node tuberculosis.

Key words: Tuberculosis, lymph node, Treatment outcome, Recurrence, Retrospective studies