HCC is a malignant tumor with a rising incidence globally, and MVI is one of the crucial factors affecting patient prognosis. Although traditional postoperative pathological assessment can identify the presence of MVI, its lagging nature limits its guiding role in preoperative treatment selection. This study investigates PDW and diabetes mellitus as indicators for predicting preoperative MVI in HCC, aiming to provide more accurate predictive tools for clinical guidance in treatment strategies, thereby improving patient survival rates and quality of life.

PDW is an indicator of platelet size and morphological heterogeneity, showing prognostic value in various malignant tumors. The results of this study indicate a significant correlation between PDW > 17 and MVI, suggesting a potential association between PDW and MVI in HCC. In previous studies, PDW > 16.8 has been identified as an indicator of poor prognosis in breast cancer (Yiru et al. 2018), and our previous research has also found PDW > 17.1 to be a risk factor for predicting recurrence within 3 years after HCC resection (Li et al. 2023), indicating PDW as a simple and reliable preoperative predictive indicator, offering new insights for clinical practice.

The potential mechanism by which PDW is associated with MVI in HCC may involve its reflection of platelet activation status. Activated platelets release pro-inflammatory cytokines and angiogenic factors, such as vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF), which are known to promote tumor progression, angiogenesis, and microvascular invasion (Minciuna et al. 2024). Higher PDW levels indicate increased platelet heterogeneity and activation, potentially leading to enhanced tumor cell adhesion to the vascular endothelium and facilitating invasion into microvessels (Xu and Guo 2023). Moreover, platelet activation is linked to epithelial-mesenchymal transition (EMT), a key process in HCC metastasis and MVI formation.

Similarly, diabetes mellitus may promote MVI in HCC through multiple mechanisms. Chronic hyperglycemia and insulin resistance contribute to a pro-inflammatory and pro-fibrotic liver microenvironment, increasing the likelihood of tumor invasion into microvessels (Ye et al. 2024). Diabetes mellitus is also associated with increased oxidative stress and systemic inflammation, which can accelerate tumor progression. Furthermore, diabetic patients often exhibit platelet hyperreactivity and increased coagulation activity, creating a microthrombotic state that may enhance tumor-endothelium interactions and promote MVI formation (Ranucci et al. 2019; Minno et al. 2012). This prothrombotic environment may explain why diabetes mellitus is an independent risk factor for MVI in HCC, as confirmed in our study.

Diabetes mellitus has been shown to be associated with an increased incidence of HCC (Teng et al. 2023; Yang et al. 2020). This study further confirms the close relationship between diabetes mellitus and MVI, with diabetes mellitus being one of the independent risk factors for MVI. This may be related to liver changes, steatosis, and inflammatory responses caused by diabetes mellitus, thereby affecting the development and invasion of HCC. Literature also reports that diabetes mellitus is characterized by platelet morphological and functional abnormalities and platelet hyperactivity, considered a prothrombotic state (Abd El-Ghany et al. 2021). Therefore, diabetic patients should receive more attention in HCC treatment, and clinicians should consider their unique physiological conditions when formulating treatment plans.

Through multivariate logistic regression analysis, we established a predictive model including PDW, diabetes mellitus, and other factors, visualized using a nomogram. This model demonstrated high predictive accuracy and stability in both the training and validation groups, providing a new method for accurate prediction of preoperative MVI. Deng et al. incorporated tumor size, AFP, and NLR indicators into the nomogram model for preoperative prediction of MVI risk (Deng et al. 2019). Chang et al.‘s nomogram model included preoperative FIB-4, AFU, AFP levels, liver cirrhosis, and tumor margin irregularity to predict MVI (Chang et al. 2023). Additionally, the broad clinical application prospects of this model can help clinicians identify high-risk patients, formulate personalized treatment plans, and improve surgical success rates and patient survival rates.

Although this study has achieved certain results, there are still some limitations. Firstly, it is a single-center retrospective study, which may have certain selection biases, and further multicenter, large-sample prospective studies are needed to validate the results. Secondly, this study did not consider all possible factors that may affect MVI, such as liver function and inflammatory indicators, and future research can further improve predictive models. Moreover, this study did not explore the mechanistic relationship between PDW and diabetes mellitus in depth, which needs to be elucidated through more in-depth experimental research, particularly regarding the role of platelet activation, inflammatory cytokines, and coagulation factors in MVI development.

In conclusion, this study confirms PDW and diabetes mellitus as reliable indicators for predicting preoperative MVI in HCC and establishes corresponding predictive models, providing a new predictive tool for clinical practice. Future research should further explore their potential mechanisms through in vitro and in vivo studies, particularly focusing on the interactions between platelets, endothelial cells, and tumor cells in the context of MVI. Strengthening validation in clinical applications will also help promote progress and development in the field of HCC treatment.