Background and Aims For patients with hepatocellular carcinoma (HCC), the effective prediction of presence or absence of microvascular invasion (MVI) is of great importance in clinical decision making, postoperative adjuvant therapy and systematic prognostic evaluation. Therefore, this study was conducted to investigate the risk factors for MVI in HCC and to establish a preoperative predictive nomogram, so as to provide a clinical reference.Methods The clinical data of 535 patients with HCC treated in Anhui Provincial Hospital from January 2017 to November 2020 were retrospectively analyzed. According to admission time, they were divided into model group (433 cases) and validation group (102 cases). Univariate and multivariate analyses were carried out to determine the independent risk factors for MVI. R software was used to establish a nomogram model to predict the preoperative MVI risk of hepatocellular carcinoma. Bootstrap analysis was used for internal validation of the model, and validation group was used for external validation of the model. C-index, calibration and receiver operating characteristic (ROC) curves were used to evaluate the predictive value of the nomogram.Results In the model group, multivariate analysis showed that NLR>2.282 (OR=1.864, 95% CI=1.184-2.933), GGT>60 IU/L (OR=2.554, 95% CI=1.631-4.001), lgAFP (OR=1.455, 95% CI=1.21-1.75), tumor size (OR=1.177, 95% CI=1.084-1.277) and absence of complete capsule (OR=2.019, 95% CI=1.286-3.171) were independent risk factors for preoperative prediction of MVI in patients with HCC. The C-index of the nomogram model established based no above factors were 0.785 (95% CI=0.742-0.828) and 0.824 (95% CI=0.737-0.91) in model group and validation group, respectively. The model fitted well with the calibration prediction curve. Based on the Youden index, the optimal critical value of the nomogram was 103. The sensitivity, specificity, positive predictive value and negative predictive value under the critical value were 86%, 61%, 67% and 82% in the model group, and 82%, 56%, 53% and 83% in the validation group, respectively.Conclusion NLR>2.282, GGT>60 IU/L, lgAFP, tumor size and absence of complete capsule were the independent risk factors for MVI in HCC. The established nomogram has a good preoperative performance in predicting MVI, which can directly analyze the preoperative risk of MVI and identify the high-risk population.