Portocaval shunt can optimize transhepatic flow following extended hepatectomy: a short-term study in a porcine model

The aim of this study was to evaluate whether the portocaval shunt (PCS) corrects these unwanted changes in transhepatic flow after extended hepatectomy (EH). Forty female Landrace pigs were divided into two main groups: (A) EH (75%) and (B) no EH. Group A was divided into 3 subgroups: (A1) EH without PCS; (A2) EH with side-to-side PCS; and (A3) EH with end-to-side PCS. Group B was divided into 2 subgroups: (B1) side-to-side PCS and (B2) end-to-side PCS. HAF, PVF, and PVP were measured in each animal before and after the surgical procedure. EH increased the PVF/100 g (173%, p < 0.001) and PVP (68%, p < 0.001) but reduced the HAF/100 g (22%, p = 0.819). Following EH, side-to-side PCS reduced the increased PVF (78%, p < 0.001) and PVP (38%, p = 0.001). Without EH, side-to-side PCS reduced the PVF/100 g (68%, p < 0.001) and PVP (12%, p = 0.237). PVP was reduced by end-to-side PCS following EH by 48% (p < 0.001) and without EH by 21% (p = 0.075). PCS can decrease and correct the elevated PVP and PVF/100 g after EH to close to the normal values prior to resection. The decreased HAF/100 g in the remnant liver following EH is increased and corrected through PCS.


Portal vein flow (PVF).
After EH, the PVF/100 g of remnant liver increased from 102.11 ± 9.09 to 279 ± 36.44 ml/min/100 g (173%, p < 0.001; Number 1, Fig. 1). This increased flow following EH was reduced to 60 ± 12.40 ml/min/100 g remnant liver (78%, p < 0.001) through side-to-side PCS (Number 2, Fig. 1). Without EH, side-to-side PCS reduced the PVF from 102.11 ± 9.09 to 32.04 ± 5.46 ml/min/100 g (68%, p < 0.001; Number 3, Fig. 1), which is less than the effect of PCS with EH (p = 0.142; Number 3 < Number 2). The side-to-side PCS corrected the post EH PVF changes so that close to a normal PVF value was reached (Number 4, Fig. 1). The end-to-side PCS reduced the PVF to zero both in the with and without EH groups, since there was no flow in the portal vein following the placement of this shunt (Fig. 1). Changes in PVF following each step are summarized in Table 1.
Portal vein pressure (PVP). EH increased the PVP from 9.5 ± 0.67 to 16 ± 1.29 mmHg (68%, p < 0.001; Fig. 3). The side-to-side PCS following EH reduced the PVP from 16 ± 1.29 to 9.9 ± 0.66 mmHg (38%, p = 0.001; Number 2, Fig. 3). The PVP following EH with side-to-side PCS was close to the normal value. Without EH, the PVP decreased from 9.5 ± 0.67 to 8.3 ± 0.71 mmHg (12%, p = 0.237) as a result of side-to-side PCS (Number 3, Fig. 3). The end-to-side PCS reduced PVP following EH from 16 ± 1.29 to 8.3 ± 1.04 mmHg (48%, p < 0.001) and without EH from 9.5 ± 0.67 to 7.5 ± 0.81 mmHg (21%, p = 0.075). PVP did not significantly (2) S-S PCS increases the reduced HAF/100 g following EH by 8%; (3) S-S PCS increases the HAF/100 g without EH (20%) more than with EH; (4) The HAF/100 g following EH with an E-S PCS is even higher than the normal HAF/100 g.  www.nature.com/scientificreports/ change as a result of PCS without EH. The side-to-side PCS after EH reduced PVP up to 80% the values after EH, whereas the end-to-side method reduced the pressure 10 percent more than the side-to-side method (48% vs. 38%; Number 4 vs. Number 2, Fig. 3). The changes in PVP following each step are summarized in Table 1.

Discussion
Despite the new developments in the field of liver surgery in recent decades, many patients are unable to undergo therapeutic liver resection due to the risk of SFSF syndrome 22 . Posthepatectomy liver failure remains one of the most serious complications of major liver resection and occurs in up to 10% of cases 22,23 . Several studies have reported posthepatectomy liver failure as a significant cause of morbidity and mortality 7 . Considering the important role of transhepatic flow in this regard 24 as well as its unwanted changes following EH 25 , it can be hypothesized that by a correction of unwanted flow changes following EH, SFSF syndrome may be prevented. It has been shown that reducing the PVF by clamping the splenic artery and performing splenectomy 26 in small for size grafts following partial liver transplantation can prevent the consequences of SFSF 19,[27][28][29] . PCS, which also reduces PVF, could show similar results in partial liver transplantation 30,31 . The effect of PCS on portal vein decompression has been deemed an important factor in preventing progressive necrosis and ultimately fatal liver failure following partial liver transplantation 32 . While some experimental studies have investigated the role of PCS in rat models following EH [33][34][35] , the feasibility and the effect of PCS following EH in correction of transhepatic flow, including its effect on hepatic artery flow, have never been systematically studied in human or large animal models. Due to the anatomical and physiological similarities between pigs and humans 36,37 , pigs are optimal models for evaluating the long-term effect of PCS following EH. However, evaluating the feasibility and effectiveness of PCS following EH in a short-term follow-up is one of the ethical and scientific prerequisites.
To systematically evaluate the immediate effect of PCS, we compared the transhepatic flow without PCS and with side-to-side and end-to-side PCS in livers without EH as well as following EH. Our study shows that following EH, the PVF and PVP increase, whereas HAF decreases significantly per 100 g of remnant liver. Moreover, we showed that the PCS, either side-to-side or end-to-side, can correct these variations by reducing the PVF and PVP and increasing the HAF. Following EH with side-to-side PCS, the PVF and PVP were even close to the normal values observed in an intact liver (without resection). This correction was also seen in HAF following EH through end-to-side PCS. As mentioned in our previous study, it is important to compare the results proportional to the remnant liver volume 25 . In the clinical setting and following EH, the remnant liver is exposed to increased PVF and PVP 24 . This leads to centrilobular arterial hypoperfusion and sinusoidal damage and finally to the occurrence of SFSF syndrome with a high rate of postoperative morbidity and mortality 32,38,39 . Following EH, the liver attempts to regenerate itself to be able to provide the needed function for the whole body mass 40 . It has been shown that following EH, the HAF remains constant 41 . It is also known that the liver will automatically compensate for a reduction in PVF by increasing the HAF to hold the transhepatic flow unchanged. This phenomenon is called the hepatic artery buffer response 42 . Therefore, performing a PCS not only correct the unwanted changes in PVF and PVP but can also improve the HAF. Our results could show and confirm this pattern following EH in a porcine model with a short-term evaluation.
PCS can be performed either side-to-side or end-to-side. Based on our results, the side-to-side PCS corrects the PVF and PVP changes following EH up to the normal values in an intact liver (without resection). The endto-side PCS could show better results in regard to the HAF. This is because the liver loses the whole PVF following an end-to-side PCS and attempts to compensate for this variation by increasing the HAF. In the clinical setting, side-to-side PCS is more practical than end-to-side PCS. Redirecting the whole PVF through an end-to-side PCS can result in high-stage hepatic encephalopathy as well as right heart failure 43,44 . However, end-to-side PCS can be an option in few cases with chronic portal vein obstruction. Although short-term follow-up results are necessary and form a basis for further long-term studies, the absence of a long-term follow-up is admittedly a limitation of this study.

Conclusion
Hepatic inflow modulation by PCS following EH is feasible. PCS following EH immediately reduces and corrects the increased PVF and PVP and increases the decreased HAF. Based on these findings and since the changes in HAF, PVF, and PVP are considered triggers of SFSF syndrome, further long-term experimental and clinical studies must be performed to evaluate the effectiveness of PCS in preventing SFSF syndrome following EH.

Study design. The study has been reported in accordance with the ARRIVE guidelines (Animals in Research:
Reporting In Vivo Experiments) 45 . This experimental study was conducted on forty female Landrace pigs. The animals were divided into two main groups: (A) EH (75%) and (B) no EH. Group A was further divided into 3 subgroups: (A1) EH without PCS; (A2) EH with side-to-side PCS; and (A3) EH with end-to-side PCS. Group B was divided into 2 subgroups: (B1) side-to-side PCS without EH and (B2) end-to-side PCS without EH (Fig. 4). Each group was made up of 8 animals. HAF, PVF, and PVP were measured in each animal before and after the surgical procedure. The baseline data of the animals before the surgery were considered representative of the group without EH and without PCS.
Anesthesia, hemodynamics, hepatic inflow and pressure monitoring. After a 12-h fasting time with free access to water, the pigs were anesthetized following our standardized narcotic protocol. Two catheters were placed in the carotid artery and internal jugular vein to monitor MAP and CVP, respectively. The heart rate, body temperature, and blood oxygen saturation were monitored during the entire procedure 37  Surgical procedure. The resection procedure was conducted in accordance with our standardized stapler hepatectomy method 46 . After EH, complete hemostasis was achieved by electrocoagulation and hand suturing.
To perform the PCS, the portal vein and vena cava were prepared and released from the surrounding tissues in the potential anastomosis place. Both veins were cross-clamped after heparinization with 30 IU/kg. For side-toside PCS placement, a portal vein-caliber size anastomosis was performed between the right lateral side of the portal vein and medial side of the vena cava using Prolene 5-0 (PROLENE® Ethicon, Norderstedt, Germany) in a continuous technique (Fig. 5). To place the end-to-side PCS, a portal vein-caliber size venotomy was performed on the vena cava after the portal vein in the liver hilum was cross-clamped and dissected. Anastomosis was performed continuously with the end-to-side technique (Fig. 6). At the end of the experiment, the remnant  All measured values are expressed as the mean ± standard deviation (SD), and the changes in the values are graphically represented. The differences in mean hemodynamic measurements between study groups were tested using independent samples t tests. A p value less than 0.05 was considered to be statistically significant in all tests.
Institutional protocol number. The study protocol was approved by the German Committee for Animal Care, Karlsruhe, Germany (AZ: 35-9185.81/G-45/12).