The present observational study aims to assess the benefit of this quantitative multiparametric magnetic resonance imaging (MRI) in clinical practice, to quantify future liver remnant performance, and to accurately predict the risk of liver failure after major hepatectomy, among patients undergoing major liver resection.

The main questions to be answered are:

• Can multiparametric MRI predict the postoperative liver function?
• Can multiparametric MRI predict the postoperative liver-specific complications as well as mortality? With ethical approval and fully informed consent, patients being considered for major liver resection will undergo clinical assessment, blood sampling, and multiparametric MRI before surgery. For the primary outcome, 33 participants will be needed to detect a minimum correlation coefficient of 0.2 with 5% significance and 80% power.

In patients being considered for major liver resection, volumetric and functional evaluation of the future liver remnant should ideally be combined to reduce the risk of posthepatectomy liver failure (PHLF) and other adverse outcomes. Quantitative multiparametric magnetic resonance imaging (MRI) can measure liver health by assessing parenchyma fibrosis and inflammation. This has shown promising results in predicting postoperative liver performance. The current study aims to assess the benefit of this MRI-based technology in clinical practice, to quantify future liver remnant performance, and to accurately predict the risk of liver failure after major hepatectomy.

Prospective observational clinical study in a single hepatobiliary surgery center. With ethical approval and fully informed consent, patients being considered for major liver resection will undergo clinical assessment, blood sampling, and multiparametric MRI before surgery. Postoperative liver function and complications will be reported for 90 days after surgery. Preoperative MRI assessment scores and postoperative outcomes will be correlated to determine whether multiparametric MRI scans can accurately predict the risk of postoperative liver-specific complications (primary endpoint) as well as postoperative liver function, surgery-specific complications, the overall complication rate, quality of hospital care, and length of stay. For the primary outcome, 33 participants will be needed to detect a minimum correlation coefficient of 0.2 with 5% significance and 80% power. This study will include 33 patients and results are expected in 2024. If successful, this investigation will support the use of quantitative multiparametric MRI to guide surgical decision-making. This will represent a non-invasive, diagnostic, volumetric, and segmental functional test for the preoperative workup of patients being considered for major hepatectomy.

Inclusion Criteria:

• Adult (≥ 18 years old) patients with an indication for elective major liver resection.
• Any benign or malignant indication.
• Provision of informed consent.
• MRI scan performed ≤ 7 days before surgery

Exclusion Criteria:

• Previous liver resection.
• Previous volume-enhancing procedures, such as portal vein embolization.
• Previous microwave ablation of liver lesions.
• Contraindication and/or inability to undergo contrast-enhanced MRI scan (including implanted metallic devices or foreign bodies, claustrophobia and contraindication for Primovist® administration).

Participants will be recruited after they have been diagnosed and their treatment has been planned by a multidisciplinary tumor board. Patients indicated to undergo major liver resection (> 3 segments based on Brisbane classification) will be eligible for participation. No change in treatment or intervention is proposed in the present study. All evaluations and analyses will take place at Champalimaud Clinical Centre. Patients who are being considered for major liver resection and have given informed consent will undergo a preoperative assessment, during which baseline clinical data and blood samples will be collected and a multiparametric MRI scan of the liver will be performed.

Upon reasonable request, the data generated by the current research that supports our future article, would be made available as soon as possible, wherever legally and ethically possible.