Purpose <p>An earlier multicenter phase III study demonstrated the safety and efficacy of mitoxantrone hydrochloride injection for tracing (MHI) for detecting sentinel lymph nodes (SLNs) in breast cancer patients. This ancillary study was conducted to further analyze and discover differences between the two tracers that had not been reported previously.</p> Methods <p>This analysis used single-center data from a phase III trial. Patients with breast cancer were injected with both MHI and 99mTc as SLN tracers before surgery. To account for cluster effects of multiple SLNs from the same patient and further analyze the differences between the two methods and factors that could impact detection, Obuchowski’s modified McNemar test, the GLIMMIX model, and univariate and multivariate analyses were used as appropriate.</p> Results <p>In total, 463 SLNs from 109 patients were resected, 47 of which were pathologically positive. The SLN detection rate did not significantly differ between MHI and 99mTc. MHI blue staining was an independent predictor of pathological positivity (OR 3.58, 95% CI: 1.21–10.60, <i>p</i> = 0.02), while the 99mTc signal was not significant in the same model (<i>p</i> = 0.39).</p> Conclusion <p>MHI was comparable to 99mTc in safety and SLN detection rates but differed in sensitivity on positive SLNs.</p>