MRI-based preoperative markers combined with narrow-margin hepatectomy result in higher early recurrence

To investigate the applicability of MRI-based preoperative risk markers in assisting clinicians to define an appropriate surgical margin width for patients with solitary hepatocellular carcinoma.

Patients who underwent preoperative MRI with hepatectomy were randomly divided into development (65%) and internal validation (35%) datasets between January 2015 and January 2019. Multivariate logistic analysis was used to evaluate MRI-based markers of early recurrence (≤2 years) in the development dataset. Independent factors in the development dataset were investigated using a multivariable Cox analysis. The multivariable logistic and Cox models were verified using the risk score system in the validation dataset. Recurrence-free survival (RFS) was calculated according to MRI-based preoperative markers together with a narrow or wide margin in all datasets.

A narrow resection margin was identified as an independent risk factor for early postoperative recurrence (P < 0.001) according to multivariable Cox analysis. RFS was significantly shorter in patients with narrow resection margins than that of those with wide resection margins (P < 0.005). Patients with the three MRI-based preoperative markers (tumour size > 5 cm, substantial necrosis, and non-smooth margins) combined with a narrow resection margin had a shorter RFS than that of those with a wide margin (P < 0.005). Patients without these markers also benefitted from a wide margin (P < 0.05).

MRI-based preoperative risk markers in combination with narrow resection margins were associated with a higher recurrence rate compared with wide resection margins, and hence, such patients may benefit from a wide-margin hepatectomy.