How to expand the safe limits in hepatic resections?

Introduction

Patients with extensive liver invasion by primary or secondary tumors have poor prognosis unless curative resection can be accomplished 1. Remarkably, the number of patients that benefit from curative hepatic resections have progressively increased as technical surgical refinements derived from Living Donor Liver Transplantation (LDLT) have been applied to patients with liver tumors 2. Neoadjuvant and adjuvant chemotherapy with new chemotherapeutic agents have also contributed to achieve improved survivals 3, 4. Therefore, the major limitation in the current oncosurgical strategies are no longer the number, size or location of the liver tumors 5-7. Instead, the main factor limiting the candidacy of patients to extended hepatectomies is the remnant liver mass 8.

A case of a patient developing alopecia, ascites and incomplete liver regeneration after an 85–90% hepatectomy was described in 1975 by Starlz et al. 9, but the capacity of the liver to regenerate after extended hepatectomies had been previously demonstrated in clinical and experimental studies 10. The progressive development of LDLT impelled the recognition of the “small-for-size” syndrome (SFSS) as a distinct clinical entity caused by a reduced mass of liver parenchyma that is insufficient to maintain proper liver function 11. The SFSS is clinically characterized by the development of cholestasis, coagulopathy and ascites, usually complicated by portal hypertension, sepsis and gastrointestinal bleeding 12. Many studies used the term “post-hepatectomy liver failure” to refer to the SFSS occurring in patients undergoing hepatic resections 13, 14, but the current trend is to include the “post-hepatectomy liver failure” as a type of SFSS 11, 12.

Predicting the risk of developing the SFSS is important in the evaluation of potential surgical strategies. Two major factors that have been identified to determine the development of the SFSS are the future remnant liver volume (FRLV) and some characteristics of the patients such as the Child–Pugh status, steatosis, diabetes mellitus, model for end stage liver disease (MELD) score and cholestasis 8. The estimation of the FRLV may be accomplished by preoperative volumetry using computed tomography (CT) scanning and specific software. The thresholds of FRLV that are usually considered to be safe are above 20% in healthy livers, above 30% when hepatocyte steatosis or history of chemotherapy cycles is present, and above 40% when the patients have cirrhosis 8. A number of strategies such as portal vein embolization (PVE), two-stage hepatectomy and others (described later in the text), may be used in eligible patients to increase the FRLV pre-operatively in order to prevent the development of the SFSS 8, 12, 15. Liver function tests such as the Indocyanine Green (ICG) clearance are also used to assess the risk of the SFSS, with an ICG retention rate at 15 min below 20% being considered safe in normal patients 12. In patients who have portal hypertension, a specific algorithm developed by Makuuchi et al. that includes the bilirubin, ICG clearance and the presence of ascites should be used 16. When a major resection is planned in this type of patient, a normal ICG-R15 value is required. Based on all these observations, helpful limits for a safe liver resection have been developed that, when surpassed, are associated with a high risk of developing the SFSS 8, 11, 13, 14. These limits, however, are not clear-cut, and some patients still develop the SFSS despite keeping within the safe “size limits”.

Should we change from “small-for-size” to “small-for-flow” syndrome?

A potential explanation for the SFSS developing in patients within the safe “size limits” could be an excessive increase of portal pressure and/or portal blood flow. Indeed, perisinusoidal and periportal hemorrhage are consistently observed in liver biopsies early after extended hepatic resections, while arterial vasoconstriction and ischemic cholangitis are observed at later stages 17. Portal blood flow was the main predictor of early graft dysfunction regardless of the size of the graft in a retrospective series of 134 patients undergoing LDLT 18. Furthermore, the relation between the presence of portal hyper-perfusion and the development of SFSS with impairment of liver regeneration has been well-established in patients with cirrhosis and hyperdynamic circulation who undergo liver transplantation 19, 20. The portal vein pressure after hepatectomy has also been recently shown to independently predict the development of liver failure and mortality in patients without cirrhosis undergoing major liver resections 21. Diverse perioperative strategies of portal blood flow modulation have successfully prevented the development of the SFSS in patients undergoing LDLT who received grafts below the lower limit (0.8%) of the graft-to-recipient body weight ratio (GRWR) 22, 23. Portal pressure and portal blood flow, however, are not generally measured in hepatic surgeries. Based on all these reports, we recently suggested using the term “small-for-flow” syndrome in order to emphasize the relevance of portal hyperperfusion as a major factor responsible for the development of the SFSS and to encourage the investigation of blood flow modulation strategies that could expand the limits of what today we consider a safe liver resection 24.

We show some of the implications, similarities and dissimilarities between the SFSS and the “small-for-flow” concepts in Table 1. In the “small-for-flow” concept, the critical factor determining the development of liver dysfunction would not be the size of the liver remnant but the proportion between the mass of the liver remnant and the blood flow that it receives, reflected by the values of portal blood flow and pressure. A disproportionate increase of portal blood flow and portal pressure would lead to sinusoidal injury, to ischemic injury due to a compensatory reduction of blood flow in the hepatic artery, and to impairment of liver regeneration 20. This could explain the requirement for bigger FRLVs to prevent the SFSS in patients with cirrhosis who have hyperdynamic circulation 25. Similarly, tolerance to portal hyperperfusion is diminished when the liver presents steatosis, cirrhosis, or exposure to long chemotherapeutic cycles, resulting also in requirement of bigger FRLVs. The notion of “small-for-flow” syndrome also implies that predicting the SFFS may be achieved by intraoperative monitoring of portal blood flow and pressure, and that strategies for perioperative modulation of these parameters may prevent the development of the SFSS 24.

Modulation of portal blood flow and pressure pushes the “size” limits in liver transplantation of small grafts

The scarcity of donor livers and the limitations for the use of brain-death donors in some countries has stimulated the use of reduced grafts in liver transplantation 26. As a result, split liver transplantation for two adult recipients and LDLT techniques were developed 27. The main limiting factor in both procedures was the development of the SFSS 11. The estimation of the minimum liver graft volume was based on the graft weight/recipient weight ratio (GRWR), and GRWRs smaller than 1% were found to deleteriously impact the prognosis of liver transplantation 28.

Portal hyperperfusion was identified as a major factor responsible for the development of the SFSS in experimental liver transplantation in pigs 20, pointing to the modulation of portal blood flow and pressure as a potential strategy to prevent the SFSS in liver transplantation. Diverse pharmacological 29 and surgical 30, 31 approaches to reduce portal hyperperfusion have recently shown to prevent the SFSS in animal models of liver transplantation with small grafts.

Portal pressure has also been shown to be important in the pathogenesis of the SFSS 20, 30, 32. Modulation of portal pressure by the embolization of the splenic artery, the creation of a portocaval shunt, or the perioperative administration of somatostatin have shown beneficial effects in clinical studies 22, 23. In a large series of liver transplantation patients receiving the left lobe, Ikegami et al. showed that portal pressure modulation with splenic artery embolization or portosystemic shunt creation was an effective strategy for the use of small grafts 23, and identified a portal pressure >20 mmHg, a MELD score >20, the inpatient status and the blood loss as risk factors for developing SFFS. Importantly, transplantation of grafts with a GRWR of 0.6% was as successful as those with GRWR >0.8%, demonstrating that the “size limits” of the liver grafts can be expanded by modulating portal blood flow and pressure. Potential mechanisms by which the modulation of portal blood flow and pressure could prevent the SFSS may involve the decrease of shear-stress and sinusoidal injury, and the improvement of hepatic microcirculation, oxygenation, and hepatocyte function.

Strategies for pushing the “size” limits in hepatic resections: focus on the liver volume

Approximately, 1–5% of patients undergoing a major hepatectomy will have a remnant liver mass that will not be sufficient to support the metabolic requirements, resulting in the development of the SFSS 11, 12. In 2011, an international consensus defined “post-hepatectomy liver failure” as a “postoperatively acquired deterioration in the ability of the liver to maintain its synthetic, excretory, and detoxifying functions, characterized by an increased INR and hyperbilirubinemia on or after postoperative day 5” 13. The “50-50” criteria is generally used to make the diagnosis, and comprise the presence of prothrombin rate inferior to 50% and serum bilirubin above 50 μmol/l in the third and fifth postoperative days 14. Being the FRLV the main limiting factor to indicate hepatic surgery in patients with liver tumors, efforts in the last 30 years have focused on the development of diverse strategies for increasing the FRLV and, therefore, the number of patients eligible for surgery (Table 2).

One of the most common approaches to increase the FRLV prior to surgery is the performance of a PVE 33, 34, which is reasonably safe and results in increases of 8–27% of FLRV 35. PVE is associated with improvement of the function of the liver remnant, with attenuation of the effects of portal hyperperfusion, and with preservation of the liver regenerative response 36. Performance of a PVE, however, is associated with several drawbacks. A major one is that the time to achieve the required increase of the FRLV may be as long as 4–6 weeks in more than 80% of the patients, and in some of them it is never achieved 12. Liver hypertrophy is known to be slower in diabetic patients 37. During this time, the risk of disease progression is high because patients cannot receive chemotherapy. Moreover, promotion of neoplastic cell growth in the non-embolized lobes may be observed during this time. As a result of these problems, a curative liver resection cannot be completed in as many as 20% of the patients undergoing a PVE.

A second potential approach is the so-called “two stage hepatectomy” 38. This strategy is used in patients with bilobar colorectal liver metastases, who have a poor prognosis unless a complete resection is achieved 39. Two stage hepatectomy consists of the resection of the tumor in the less affected lobe followed by portal embolization of the second affected lobe. If an estimated FRLV higher than 20% is achieved, a second surgery with extended hepatectomy is performed. This strategy aims to prevent the growth of tumor cells in the non-embolized lobe and to increase the number of patients who can benefit from a liver resection. Again, the delay from the first surgery to the final extended hepatectomy may be from 3 to 6 months. In a recent publication by Homayounfar et al., the second stage hepatectomy was accomplished in 79.2% of the patients and the mortality and morbidity rates were 5.3% and 57.9%, respectively 40.

Another new strategy for patients with limited FRLV has recently been described, consisting of associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) 41. This strategy increases up to 74% the FRLV in a mean of 9 days, and recent reports show increases of 69% of the FRLV as early as 3 days after ALPPS 42. Therefore, liver resection can be performed between 4 and 8 days after ALPPS. Despite inducing a rapid hepatic hypertrophy and reducing posthepatectomy liver failure, ALPPS is associated with high morbidity (68%) and mortality (12%) rates 43. ALPPS may also be used as a rescue therapy for patients with insufficient FRLV after PVE, as growth induction by ALPPS is not impaired 42.

Strategies for pushing the “size” limits in hepatic resections: focus on the flow

All the approaches described in the previous paragraphs focus on increasing the FRLV and they have shown to be useful in the management of patients with extensive liver disease 12. In the studies of patients treated with ALPPS, however, the rates of morbidity and mortality are unusually high even in patients that achieved FRLVs considered being safe 42. We believe that this observation reflects an incomplete knowledge of the mechanisms leading to the SFSS as well as of how these new surgical strategies influence such mechanisms.

The perioperative monitoring of portal blood flow and pressure is a potential way of increasing our understanding of the SFSS in extended hepatic resections as well as the effects of the new surgical strategies 24. The experience in LDLT shows that liver grafts that fulfill the “size” requirements may develop the SFSS if portal hyperperfusion is present 18 and, vice versa, the SFSS can be prevented in liver grafts that are considered “too small” if modulation of portal blood flow during LDLT is performed. These observations suggest that portal modulation maneuvers should also be implemented during the second procedure of a two-stage hepatectomy or ALPPS if portal hypertension is detected. Strategies such as somatostatin infusion or splenic artery ligation could be performed, with re-evaluation of the portal blood flow and pressure. In case of persistent portal hypertension refractory to the previous maneuvers, the creation of a porto-caval shunt surgically or with a transjugular intra-hepatic portal-systemic shunt (TIPS) could be performed. In this line, some studies have reported the attenuation of the SFSS in patients undergoing LDLT 44, 45.

Strategies for pushing the “size” limits in hepatic resections: focus on regeneration preconditioning

Based on the “regenerating liver” concept introduced by Nagano et al. 46, regeneration preconditioning is another potential hepatoprotective approach for improving the tolerance to extended hepatectomy (Fig. 1). In this approach, the aim would be to stimulate liver regeneration prior to surgery in order to increase the number of hepatocytes that are in a primed or proliferating state when the extended hepatectomy is performed, resulting in improved tolerance and faster liver regeneration. Based on theoretical modeling, it has been suggested that increasing the metabolic load prior to an extended hepatectomy or decreasing the metabolic load after an extended hepatectomy could also result in improved recoveries 47.

If confirmed by clinical and experimental studies, the concept of “regeneration preconditioning” could change the current assumption that surpassing a certain liver volume threshold after PVE is necessary for a successful subsequent extended hepatic resection. In addition, the most convenient period to perform the extended hepatectomy could be between 24 and 72 h after the PVE, when the peak of DNA synthesis takes place. The current practice of waiting 3–4 weeks after the PVE before performing the extended hepatectomy, therefore, probably results in the loss of the benefits of regeneration preconditioning. Regeneration preconditioning is a strategy for improving the safety of liver resections that should be further explored in experimental studies.

Conclusions

The increased use of small liver grafts and the increasing need for expanding the indications of curative liver surgery in patients with hepatic tumors has improved our understanding of the mechanisms responsible for the development of the SFSS. It is increasingly clear that size is not all that matters in the development of liver dysfunction after extended hepatectomies. Increasing the size of the future liver remnant, the introduction of the “small-for-flow” concept with the perioperative monitoring and modulation of portal blood flow and pressure, and the exploration of the potential effects of regeneration preconditioning are all promising strategies that could help to expand the indications and increase the safety of liver surgery.

Conflict of interest 

None declared.