Small-For-Size Syndrome and Graft Inflow Modulation Techniques in Liver Transplantation Free

Dr. Thomas Starzl, the leading pioneer in liver transplantation (LT), passed away on March 4, 2017, and it was because of his innovation and persistence that significant breakthroughs were made in the field of LT [1]. As a result, an increasing number of patients with end-stage liver disease have benefited from this life-saving procedure to date [2]. However, the apparent imbalance between the growing demand for LT and the shortage of suitable liver grafts remains a significant challenge for the LT community [3]. Due to traditional Asian culture and the continuous improvements of the legislation on deceased organ donation, access to deceased grafts is highly restricted in many Asian countries [4]. In this circumstance, partial LT, such as living donor LT (LDLT) and split LT, appears to be an effective solution to expand the donor pool [5, 6]. Based on previous experiences, liver grafts with a graft-to-recipient weight ratio (GRWR) <0.8% may not meet the metabolic demands of recipients, leading to posttransplant hepatic dysfunction (also known as small-for-size syndrome [SFSS]) [7‒10]. SFSS can be attributed to various factors in a broad sense, except for graft size. Other factors affecting the donor and recipient should also be taken into account. With targeted improvement measures, satisfactory clinical outcomes can be achieved in recipients with a higher risk of SFSS. This review article aimed to describe the etiology, pathological features, clinical manifestations, and diagnostic criteria of SFSS and strategies to improve graft function, with a particular focus on graft inflow modulation (GIM) techniques. Furthermore, we discuss the unmet needs and the future perspectives of SFSS in LT.

In undersized liver grafts, the extraordinary regenerative capacity of mature hepatocytes may be compromised [11], suggesting a significant association between small-for-size grafts (SFSGs) and the development of posttransplant SFSS. The term “SFSG” generally refers to a liver graft with a GRWR <0.8% – in other words, the actual volume of the graft is less than 40% of the standard liver volume [12‒14]. Damage to the hepatic sinusoidal and liver parenchyma caused by vascular shear stress relative to excessive portal vein flow (PVF) due to disproportionate graft volume was considered the primary pathogenesis of SFSS. Later, this hypothesis was proved by the Hemptinne group [15]. Aberrant gene expression was observed at the molecular level in recipients with SFSG. Endothelin-1 overexpression was involved in hemodynamic changes in the hepatic sinusoids, triggering portal hypertension that led to SFSS [16]. Hepatocytes in SFSG are typically more susceptible to oxidative damage from inflammation due to significantly downregulated expression levels of heme oxygenase-1 (HO-1) and heat shock protein-70 (HSP70) [17]. Furthermore, inflammation in SFSG may be exacerbated by the expression disorder of endogenous regulatory factors, such as zinc finger protein A20 (A20), early growth response 1 (EGR1), and inducible nitric oxide synthase (iNOS) [18].

Furthermore, SFSS after LT is associated with other factors except for graft size [19, 20]. According to the guideline developed by the International Liver Transplantation Society(ILTS) in 2017, graft injury and dysfunction in SFSS reflect graft size, graft quality, and the degree of recipient portal hypertension [21]. A wide range of factors, such as recipient preoperative status, surgical techniques, donor age, and graft quality, can also contribute to the development of SFSS in LT and require more attention and more in-depth studies (Fig. 1).

It is well known that enhancing patients’ preoperative conditions are essential for minimizing postoperative complications and improving outcomes [22, 23], and LT is no exception. In addition to the most commonly used nutritional indicators, such as serum albumin, body mass index, and subjective global assessment, careful assessment of sarcopenia in patients with end-stage liver disease has received considerable attention. Masuda et al. [24] reported that preoperative sarcopenia was significantly associated with mortality and sepsis after LDLT. Subsequent studies by the Uemoto group further strengthened the evidence linking sarcopenia with overall survival after LT [25] and recommended appropriate nutritional support during the perioperative period [26]. In addition, the degree of preoperative portal hypertension in the recipients is another crucial consideration, as excessive portal venous perfusion is considered to contribute to graft injury in LT [27].

Graft function and survival rates are also affected by hepatic venous outflow reconstruction, especially in recipients with SFSG [28]. Obstructed venous outflow, leading to graft congestion, will further result in graft failure. Therefore, it is crucial to maintain adequate hepatic venous drainage. Recent reports have shown that various outflow reconstruction strategies can be used when necessary [28‒32]. Similarly, posttransplant anastomotic stenosis of the hepatic artery and bile duct and hepatic artery thrombosis have been associated with inferior graft function. In conclusion, the exquisite surgical technique is vital in optimizing graft function.

Due to increasing age, aging of the liver impairs the ability of hepatocytes to regenerate [33]. Recent studies have shown a significant correlation between donor age and recipient prognosis. Independent studies in the Chen and Zheng groups have shown that the older donor is a significant risk factor for low overall survival in LT recipients [34, 35]. However, to date, the safety margin for donor age remains controversial. In a study investigating the impact of donor age on LDLT, Soejima et al. [36] found a statistically significant better prognosis in the group with a donor age <50 years compared to the group with a donor age ≥50 years, while the Uemoto group defined a safe margin of donor age in LDLT as 45 years [37]. Based on an extensive literature review, experts in the field strongly recommend using grafts from young donors as a preventive strategy for posttransplant dysfunction.

Due to the increasing incidence of nonalcoholic fatty-liver disease and [38] the persistent worldwide shortage of organs, the use of liver grafts with hepatic steatosis is now inevitable [39]. However, liver grafts with moderate-to-severe steatosis have been shown to be associated with increased incidences of primary nonfunction and decreased recipient survival [40‒42] due to impaired liver regeneration and ischemia-reperfusion injury [43]. Moreover, liver grafts from donors after cardiac death have long been considered marginal quality organs with a risk of early allograft dysfunction [44]. Given that liver fibrosis is a precursor to cirrhosis, grafts with chronic liver fibrosis characterized by collagen fiber deposition may be another potential risk factor for developing SFSS [45]. Unfortunately, there has not yet been a potent therapy that can successfully reverse liver fibrosis [46].

The pathological features of SFSS were first described in 1995 in a porcine orthotopic autotransplantation model designed by the Sugimachi group [47]. In this study, the histopathological examination of post-perfusion SFSG revealed severe ischemia, marked congestion within the sinusoids and portal veins, and extensive necrosis and exfoliation of endothelial cells. The Fung group described in detail the histopathologic features of SFSS, which can vary depending on the postoperative time point [48]. Endothelial damage in the portal vein, ductular reaction, hepatocyte ballooning, cholestasis, and centrilobular microvesicular steatosis were the primary pathologic changes detected on early postoperative liver biopsy. Ischemic bile duct necrosis and hepatic artery vasospasm with thrombosis developed during the second or third week after LT, and nodular regenerative hyperplasia of liver parenchymal injury was observed approximately 3 weeks later. Subsequently, they further demonstrated the unique role of impaired hepatic artery buffer response in the pathophysiology of SFSS [49].

The most frequent clinical manifestations of SFSS after LT include intractable ascites, cholestasis, coagulopathy, and grade III–IV encephalopathy. In addition, gastrointestinal dysfunction, renal insufficiency, and even septic shock can be observed in some severe cases [2]. However, the diagnostic criteria for SFSS in LT are still controversial. In 2003, the Maehara group first summarized the diagnostic criteria based on posttransplant total bilirubin levels and ascites volume [36]. Subsequently, they subdivided it into two distinct concepts: small-for-size dysfunction and small-for-size nonfunction [50]. In their study, the Everhart group proposed the concept of early allograft dysfunction to describe posttransplant hepatic dysfunction, defined as a serum total bilirubin level >10 mg/dL or an international normalized ratio >1.6 at postoperative day 7 [51]. To better understand SFSS in LT, more comprehensive diagnostic criteria have been recently proposed (Table 1).

As mentioned earlier, in a broad sense, SFSS can be attributed to a variety of factors. Over the past decades, significant efforts have been made to develop targeted improvements to minimize the risk of posttransplant SFSS. However, given the shortage of donors and the safety of living donors, neither cadaveric livers nor right hepatic lobes from living donors are always available in sufficient size for LT [55, 56]. Therefore, several attempts have been made to optimize the graft function. Short-term combination therapy involving diet, exercise, and medication has been introduced into clinical practice to improve hepatic steatosis in living liver grafts [57]. Perioperative nutritional support has received special attention, especially in SFSG recipients [24]. However, the most challenging but also the most potent prevention strategy is how to modulate postoperative PVF effectively [58‒60]. Previous studies have suggested that excessive PVF is positively correlated with the severity of portal hypertension and that recipients with portal vein pressure (PVP) ≥20 mm Hg have a significantly worse outcome compared to those with a PVP <20 mm Hg after reperfusion [7, 61, 62]. The Uemoto group, combined with portal pressure control, successfully reduced the lower limit of GRWR to 0.6% [63]. The most widely used GIM techniques are discussed in the following sections (Fig. 1).

The splenic vein is an essential branch of the portal vein system. Based on this fact, spleen inflow modulation techniques seem to hold promise for controlling excessive PVF.

Since the first successful use of splenic artery ligation (SAL) in LDLT, the procedure has been widely used as a first-line strategy to modulate PVF [64]. Excessive PVF, as a result of a disproportionate graft size, can be significantly reduced by simultaneous SAL, which in turn leads to a proportionate increase in hepatic artery flow [65‒67]. However, it should be noted that SAL may not be the best option in cases of severe portal hypertension [68, 69]. The most common complications of simultaneous SAL in LT are the development of splenic abscesses and pancreatic injury. Furthermore, a recent report that the development of collateral vessels after simultaneous SAL may lead to posttransplant gastrointestinal hemorrhage has raised concerns about the hemodynamic changes in the portal vein system after simultaneous SAL in LT surgery [70].

Hypersplenism is a common manifestation of portal hypertension in patients with cirrhosis. In turn, sinusoidal dilatation and hyperemia in the spleen increase PVF, thus establishing a vicious cycle that contributes to portal hypertension [71]. Furthermore, endothelin-1 from sinusoidal endothelial cells also contributes to the development of portal hypertension [72]. Based on this knowledge, simultaneous splenectomy during LT surgery has been reported to be beneficial in controlling excessive PVF to SFSG, especially in cases with large spleens [62, 73]. Simultaneous splenectomy is well documented to be more effective in modulating PVF than SAL [62, 74]. However, it should be noted that simultaneous splenectomy in LT surgery is not always trivial considering the potential adverse factors, including megalosplenia, extensive perisplenic collateral vessels, and severe adhesions.

Meanwhile, simultaneous splenectomy may increase the risk of postoperative infection, in particular overwhelming postsplenectomy infection. Overwhelming postsplenectomy infection has been reported in 20% of recipients who underwent simultaneous splenectomy during LT surgery, leading to high mortality [75]. Furthermore, it is now widely accepted that splenectomized individuals are at high risk of portal vein thrombosis. The Kodera group suggested that simultaneous splenectomy during LDLT should be avoided as much as possible due to an increased risk of postoperative portal vein thrombosis [76]. Other frequent complications associated with splenectomies, such as pancreatic leakage and intra-abdominal hemorrhage, can delay recovery and increase mortality.

Considering the high risk of fatal complications associated with splenectomy, Moon et al. [77] designed a modified procedure, splenic devascularization. They divided the gastrosplenic ligament and ligated both the splenic artery and the right gastroepiploic artery, leaving the intrapancreatic collateral of the superior mesenteric artery as the sole arterial supply to the spleen. With this strategy, they were able to reduce PVF and the risk of complications related to splenectomy, concluding that splenic devascularization could replace simultaneous splenectomy as an effective GIM technique during LDLT. However, it should be emphasized that this novel procedure is still in its infancy, and further studies are required to evaluate its availability and practicability before it can be widely used in clinical practice.

Control of posttransplant PVF is also crucial for recipients at risk of developing SFSS. Due to the rapid advances in interventional radiology, splenic artery embolization (SAE) has emerged as a powerful weapon in the regulation of PVF after LT. A previous study has shown that early SAE can significantly alleviate symptoms of SFSS and improve prognosis [78]. Moreover, Jiayin and colleagues [79] introduced selective SAE, which has been successfully applied to the treatment of SFSS following LDLT. In general, interventional therapies play an increasing role in the prevention and management of SFSS.

As a consequence of worsening portal hypertension, spontaneous portosystemic shunts (SPSS) are frequently seen in patients with advanced cirrhosis [80]. However, the necessity of occluding these collateral vessels during LT surgery remains controversial [81, 82]. Considering that collateral veins can alter PVF and even cause hepatofugal blood flow and portal steal syndrome, several studies have supported the routine ligation of large SPSS during LT surgery when feasible [83, 84]. In contrast, the Gocho group suggested that the SPSS, especially the communication between the portal vein system and the vena cava, should be preserved to prevent excessive PVF in SFSG transplantation and provide a potential approach for interventional therapy after LT [85]. Moreover, intraoperative reconstruction of the portosystemic shunt has been considered a promising and safe GIM technique to prevent excessive PVF when using SFSG [86]. In 2002, Pouyet and colleagues [87] successfully avoided graft overperfusion in a recipient with extra-SFSG (GRWR = 0.61%) by shunting the flow in the superior mesenteric venous flow through a mesocaval shunt with downstream ligation of the superior mesenteric vein. However, it should be noted that the uncontrollability of blood flow in the portal vein is a primary concern with these surgical reconstruction techniques [88, 89]. Hemi-portocaval shunt has been recommended as a feasible and effective option to prevent graft hyperperfusion and portal vein steal syndrome [90, 91].

Previous animal studies have indicated that low doses of somatostatin play an essential role in conferring organ protection in LT by attenuating acute phase shear stress associated with portal hypertension [92, 93]. Subsequently, the Troisi group systematically investigated the safety and efficacy of somatostatin in PVF modulation after LT [94]. In this study, 18 recipients received somatostatin, and the rest received placebo treatment. The results showed a significant decrease in hepatic venous portal gradient and portal venous blood flow and a corresponding increase in hepatic arterial blood flow in the somatostatin group. Furthermore, prostaglandin E1 and adenosine have been reported as potential liver inflow modulators, although more extensive studies are still needed to validate their efficacy and safety in LT recipients [49, 95].

Although SFSS has been controversially discussed in LT settings for many years, there is no universally accepted definition of the phenomenon. Most current versions are outlined by single centers with small sample sizes, leading to inherent bias [36, 50‒54]. Furthermore, it presents an important challenge for the prevention of SFSS in the near future, as the development of SFSS in LT, especially in LDLT, is a multifactorial event with risk factors that include the preoperative status of the recipients, surgical techniques, donor age, and graft quality except for graft size. A recent systematic review of SFSS and GIM techniques has unequivocally shown that in LT recipients with SFSG, modification of PVF contributed to reducing morbidity and mortality, thereby improving outcomes [96]. However, previous studies have not specified the metric PVF as a trigger for performing GIM techniques, and PVP has usually been used as a surrogate marker for PVF in most studies. What is the quantitative relationship between PVP and PVF? Further elucidation of the relationship between PVF and PVP will allow further evidence-based recommendations. Although it has been proposed that PVP of less than 15 mm Hg is the most favorable and less than 20 mm Hg is mandatory, the need to decompress portal inflow to minimize the PVP threshold for SFSS has been controversial to date [61, 62, 97].

Various pharmacological and surgical treatments have been implemented to reduce or partially divert excessive portal inflow, with varying levels of evidence to support their use, as previously mentioned. However, there is no consensus on the best option in the current guidelines. With this in mind, it is recommended that LT surgeons be prepared to apply these techniques on a case-by-case basis [85].

The challenge of SFSS will remain in the near future due to the increasing demand for organs, especially in Asian countries. Fortunately, in the last decade, there have been tremendous insights into portal hemodynamics and their impact on postoperative outcomes in LT [96]. The elucidation of specific mechanisms and microscopic changes in SFSS will allow the development of comprehensive preventive measures and individualized therapeutic strategies. In addition, novel GIM techniques will help reduce the risk of posttransplant SFSS in LT recipients. Finally, the increasing number of successful partial LT cases and the lower incidence of SFSS are particularly beneficial for expanding the donor pool and alleviating organ shortages.

In a broad sense, posttransplant SFSS can be attributed to various factors such as the preoperative status of the recipients, surgical techniques, donor age, and graft quality, except for graft size. Contemporary transplant surgeons should consider a combination of the various factors of donor and recipient and develop case-specific prevention strategies to improve the survival of the graft and recipient. Given the advantages and disadvantages of different GIM techniques, recipients will benefit from an individualized therapeutic plan after carefully weighing the benefits against potential risks. Furthermore, more extensive, prospective, multicenter studies are needed to better understand SFSS to improve graft and patient outcomes.

The authors declare no conflict of interest.

This study was supported by grants from the National Natural Science Foundation of China (81873591 and 81670591), the Natural Science Foundation of Guangdong Province (2016A030311028), the Science and Technology Planning Project of Guangdong Province (2018A050506030), the Science and Technology Program of Guangzhou (201704020073), the Guangdong Provincial Key Laboratory Construction Projection on Organ Donation and Transplant Immunology (2013A061401007, 2017B030314018, and 2020B1212060026), and the Guangdong Provincial International Cooperation Base of Science and Technology (Organ Transplantation) (2015B050501002 and 2020A0505020003).

Pengrui Cheng contributed to the manuscript’s composition, literature review, and drafting and finalization of the manuscript. Zhongqiu Li contributed to the literature review and search. Zongli Fu designed the figure. Qian Jian designed the table. Ronghai Deng contributed to the manuscript’s drafting and critical review. Yi Ma contributed to the approval of the final version of the manuscript.