Retrograde in situ hypothermic perfusion of the liver during right hemihepatectomy

Currently, surgical resection is often the only curative treatment for primary
or secondary hepatic malignancies. Liver resections are nowadays performed with
low mortality and acceptable morbidity. As result of that, an increasing number
of patients is currently under consideration for resection of more complex or
large tumors, thus requiring extensive resection procedures. Application of
vascular exclusion (i.e., clamping of the portal vein and hepatic artery)
during such procedures reduces blood loss, which is one of the most important
factors affecting peri-operative outcomes. However, vascular exclusion causes
ischemia-reperfusion (I/R) injury of the liver parenchyma as an inevitable
side-effect, which adversely impacts postoperative liver function and
regeneration. Additional cooling of the liver by means of hypothermic perfusion
is expected to further reduce intraoperative blood loss, as well as to protect
the liver from I/R injury. Therefore, the aim of this pilot study is to cool
the future remnant liver (FRL) in situ during right hemihepatectomy under
vascular exclusion. Consequently, we expect an improvement in postoperative
outcomes, due to a decrease in intraoperative blood loss and less parenchymal
damage, as well as a better ability of the liver remnant to regenerate.

To reduce intraoperative blood loss and enhance tolerance of the FRL to I/R
injury during right hemihepatectomy under vascular exclusion by means of in
situ hypothermic perfusion with retrograde outflow (R-IHP) of the FRL.

The study is designed as a prospective pilot study in 18 patients (9
interventions and 9 controls) to assess the effects of the proposed
intervention. Additionally, 4 patients will be included separately for
assessment of the intervention*s feasibility prior to randomized inclusion.

During right hemihepatectomy, the ipsilateral branches of the portal vein and
hepatic artery are clamped and cut prior to parenchymal transection. A catheter
is introduced in the cut end of both the hepatic artery and portal vein branch,
which allows antegrade perfusion through the hepatic artery supplying the FRL,
and retrograde drainage via the corresponding portal vein branch. The main
portal vein and hepatic artery are clamped proximal to the cathethers and
subsequently the venous outflow of the FRL (i.e., the middle and left hepatic
vein) is clamped as well. A needle temperature probe is inserted into the FRL
parenchyma while cold perfusion with 4°C Ringer*s lactate solution is started.
As soon as the FRL has reached a temperature of 28°C, or after a period of 10
minutes, parenchymal transection is started. When the resected specimen is
removed, R-IHP is terminated and the vessels are decannulated. The cut ends are
ligated proximally to the cannulation site and following removal of the
vascular clamps the liver remnant is recirculated. The maximum time of vascular
exclusion is 60 minutes.

A more complex variety of IHP (i.e., total vascular exclusion with perfusate
drainage from a cavotomy combined with a venovenous shunt) of the liver is
feasible, as has been shown in large animal studies in our laboratory, as well
as in smaller patient series. Interestingly, although all clinical studies
utilized IHP in patients suffering from severe hepatic co-morbidities and/or
those in need of extensive surgery, remarkably few complications were seen,
especially when considering these patient characteristics.
Furthermore, isolated perfusion of the liver has been employed for local
administration of chemotherapeutic agents in patients with unresectable liver
metastases. Again, few complications were reported, all of which were most
likely attributable to the use of chemotherapeutics in these studies.

Our proposed R-IHP model encompasses a combination of the two methods described
above, applying intrahepatic cooling by means of a relatively simple operative
procedure. Since both the perfusate entry and drainage sites are readily
available, anatomical access is easily provided. Also, because a plain
electrolyte solution is used for perfusion (Ringer*s lactate solution),
possible leakage of perfusate into the systemic circulation is harmless. In
all, by circulating the FRL with chilled perfusion solution, bleeding from the
left side of the liver will be ameliorated while its parenchyma is
simultaneously protected against the detrimental effects of I/R injury.

IHP is however, associated with an increase in cumulative operating time and a
variety of technical difficulties can be encountered when cannulating the right
hepatic artery and portal vein branch. Additionally, inadequate cooling might
occur, in which the liver fails to reach the 28°C target temperature. However,
this is expected not to interfere with the course of the operation since
parenchymal transection will not be dependent on liver temperature.
Furthermore, there is a possibility of bleeding after taking the biopsies,
although this procedure is carried out under direct vision and any bleeding
site is immediately controlled. Clamping of the hepatic veins is associated
with a more serious risk of bleeding, although this procedure has been proven
low-risk in a recent large patient series. The additional investigational
treatments (i.e., biochemical/inflammatory parameters, HBS-SPECT, and
CT-volumetry) carry no risks for the patient, and the drawing of blood will be
combined with routine laboratory testing whenever possible. Since preoperative
HBS-SPECT and pre-/postoperative CT scans are performed as standard routine
these will carry no additional burden for the patient.