1.1 Current evidence on the efficacy of chemosaturation via percutaneous hepatic artery perfusion and hepatic vein isolation for primary or metastatic liver cancer ('hepatic chemosaturation') is limited in quality and quantity. With regard to safety, there is a significant incidence of serious adverse effects. Therefore, this procedure should only be performed within the context of research, which may take the form of observational studies.
1.2 Patient selection should be done by an appropriate multidisciplinary team.
1.3 Hepatic chemosaturation should only be carried out by clinicians with specific training in its use and in techniques to minimise the risk of adverse effects from the procedure.
1.4 Research should document indications for treatment, details of patient selection and details of adjuvant and prior treatments. Outcome measures should include complications, survival and quality of life. Data from well-designed trials comparing the procedure against other forms of management would be particularly useful, but prospective observational studies may also be of value.