Real-Life Comparison of Diagnostic Yield and Sample Adequacy of 22 G and 25 G EBUS-TBNB Needles: A Retrospective Study

<b>Background/Objectives</b>: EBUS-TBNA is a safe and minimally invasive procedure to evaluate hilar and mediastinal lymph nodes (LNs). The Franseen needle provides a transbronchial needle biopsy (TBNB). Various needle sizes are available. In the literature, diagnostic yield (DY) and sample adequacy (SA) between needle sizes are still being debated. <b>Methods</b>: In total, 88 patients with lymphadenopathy were consecutively enrolled from June to December 2021. Chest CT and PET/CT scans were performed. Dimension at imaging and EBUS and the standardized uptake value (SUV) were recorded. EBUS-TBNB was performed with 22 G or 25 G needle sizes. DY for cancer and SA for predictive markers were evaluated. Overall DY (ODY) was also evaluated. <b>Results</b>: A 22 G needle was used in 51 patients and a 25 G needle was used in 37 patients with no differences in sex and age. The 22 G population presented a larger median dimension of LN both at imaging and EBUS compared to the 25 G population. Median LN SUV was higher in the 22 G population. Notably, 70 out of 88 patients had LNs suspicious for malignancy, which was higher in the 22 G group compared to the 25 G group (n = 46, 90% vs. n = 24, 65%; <i>p</i> = 0.004). DY for cancer was similar in both groups (84% for 22 G; 91% for 25 G). Also, SA for predictive markers was similar. ODY values were 78% and 92%, respectively, for the 22 G and 25 G needles. <b>Conclusions</b>: The 25 G needle has a higher DY (even if not statistically significant) and SA for predictive markers similar to the 22 G needle; further studies are necessary to evaluate if 25 G is comparable to the 22 G needle.