Objective <p>To explore the value of microflow patterns based on superb microvascular imaging (SMI) combined with greyscale ultrasound in thyroid nodule diagnosis and biopsy recommendation.</p> Materials and methods <p>Adult patients with thyroid nodules were recruited from May 2023 to February 2024. The greyscale features of nodules were evaluated according to the five ultrasound risk stratification systems (RSSs). The microflow patterns on SMI were used to adjust the category of nodules. The crab claw-like and the root hair-like patterns were malignant signs for upgrading, with the wheel-like and the arborescent patterns for downgrading. The diagnostic performance and biopsy recommendation of microflow patterns combined with greyscale ultrasound were analyzed.</p> Results <p>A total of 253 nodules (136 malignant, 117 benign) in 203 patients were included. Chinese Thyroid Imaging Reporting and Data System (C-TIRADS) owned the best diagnostic sensitivity and the largest AUC, with no significant improvement for the former (0.853 vs 0.904, <i>p</i> = 0.096) but an evident increase for the latter (0.834 vs 0.900, <i>p</i> &lt; 0.001) when combined with SMI. American College of Radiology (ACR) TI-RADS had the highest specificity, which was further enhanced by SMI (0.803 vs 0.855, <i>p</i> = 0.041). The unnecessary biopsy rates of RSSs were reduced by 1.27–20.30% according to microflow patterns on SMI. Among these, C-TIRADS had the lowest unnecessary biopsy rate and the largest reduction after the adjustment by SMI (38.04% vs 17.74%).</p> Conclusion <p>The microflow patterns on SMI could enhance the diagnostic validity of greyscale ultrasound for thyroid nodules. Besides, the unnecessary biopsy rate could also be decreased by combining with SMI.</p> Key Points <p><Emphasis Type="BoldItalic">Question</Emphasis> <i>The utility of combining superb microvascular imaging (SMI) with greyscale ultrasound for the diagnosis of thyroid nodules and the recommendation of biopsies remains unclear.</i></p> <p><Emphasis Type="BoldItalic">Findings</Emphasis> <i>Risk stratification systems adjusted by microflow patterns on SMI outperformed the original systems in distinguishing thyroid carcinoma and recommending biopsies.</i></p> <p><Emphasis Type="BoldItalic">Clinical relevance</Emphasis> <i>Microflow patterns combined with greyscale ultrasound can enhance the diagnostic validity and reduce the unnecessary biopsy rate for thyroid nodules.</i></p> Graphical Abstract <p></p>