Clinical application of rapid on-site evaluation of technology for the diagnosis of respiratory system tumors: a retrospective study.

Objectives: This study aimed to investigate the clinical application of rapid on-site evaluation (ROSE) technology in the diagnosis of respiratory system tumors.

Methods: A total of 175 consecutive hospitalized patients with a suspected respiratory system tumors between 2019 and 2024 were enrolled (124 from 2019 to 2022, and 20 from 2022 to 2024). Here, 144 cases underwent conventional pathological examination with ROSE assistance, while 31 patients received examination without ROSE. The concordance rate between ROSE and pathological results was compared, along with the timeliness of reports from 2019 to 2022. Using pathological examination results as the "gold standard," the difference in diagnostic positive rates between the groups with and without ROSE was compared.

Results: The concordance rate between ROSE and pathological diagnosis was 95.83%. The median time of DQ staining in 2019-2022 was 8.0 min, and the median time of pathological results was 5645.5 min, with statistical difference between the two groups (P < 0.05). The average time of pathological report in 2022-2024 was 1446.5 min, the longest time was 2562.0 min, and the shortest time was 1118.0 min. The concordance rate with pathological examination was 95.83% for the ROSE group and 83.87% for the non-ROSE group. The difference in positive rates between the ROSE group and the non-ROSE group was statistically significant (P < 0.05). Using pathological examination as the "gold standard," the concordance rate for dual-operator ROSE was 95.97%, and for single-operator ROSE was 95.00% (P = 0.845). The average blood loss of the searchable ROSE group (16 cases) and non-ROSE group (27 cases) in 2022-2024 was 4.87 mL, 2.37 mL, and there was no statistically significant difference between the two groups (P > 0.05).

Conclusions: ROSE technology demonstrated a higher positive rate, thereby improving the pathological diagnostic yield compared to conventional pathological examination alone. This enables clinicians to obtain preliminary diagnoses earlier, guides sampling site and quantity, possesses considerable clinical value, and warrants promotion.