Large-for-size syndrome prophylaxis in infant liver recipients with low body mass

Abstract

Transplantation of the left lateral section (LLS) of the liver is now an established practice for treating advanced diffuse and unresectable focal liver diseases in children, with variants of the LLS primarily used in infants. However, the surgical challenge of matching the size of an adult donor's graft to the volume of a child's abdomen remains significant. This review explores historical developments, various approaches to measuring the required functional liver mass, and techniques to prevent complications associated with large-for-size grafts in infants.

Key Words: Pediatric liver transplantation; Large-for-size syndrome; Preoperative evaluation of donor and recipient; Liver volumetry; Monosegmental transplantation; Left lateral sector graft; Reduced size liver graft; Abdominal wall reconstruction; Liver transplantation; Liver resection

Core Tip: Transplantation of the left lateral section is a well-established treatment for advanced liver diseases in children, particularly infants. This review highlights the ongoing challenge of matching adult donor grafts to the smaller abdominal capacity of pediatric recipients. It also discusses the evolution of surgical approaches, methods for accurately assessing required liver mass, and strategies for mitigating complications related to large-for-size grafts.

INTRODUCTION

The world's first attempt for performing pediatric liver transplantation (LT) was made by T. Starzl [1] in 1963 to a 3-year-old boy with biliary atresia. Still, the patient died of bleeding before the end of the surgical intervention. The subsequent series of liver transplants were unsuccessful until, in 1967, Starzl et al[1] reported the first successful pediatric LT from a deceased donor[1].

In 1984, Broelsch et al[2] in the United States and Bismuth et al[4] in France independently performed the first reduced-size LTs, thereby expanding the boundaries of transplant surgery in pediatric practice[2-6].

In 1988, Pichlmayr et al[5] reported the first split transplantation, which consisted of dividing the liver into 2 grafts and giving a chance for life to two patients at once. The growing number of patients on the waiting list and the shortage of organs from deceased donors, especially for children, prompted the search for new resources, one of which was a transplantation of part of the liver from a living donor. Broelsch et al[2] and co-authors developed the technique of related transplantation of the left lateral segment of the liver, which was implemented for the first time in the world in 1989[2,7,8]. In the USSR, the first attempt at transplanting the left lobe of the liver was made in 1977 by Galperin[9-12].

With the accumulation of experience, the surgical community was faced with the problem of matching the size of the graft with the volume of the child’s abdominal cavity. Low body weight in children suffering from chronic liver diseases leads to a discrepancy between the size of the recipient’s abdominal cavity and the donor’s liver fragment, which can cause the development of intra-abdominal hypertension when closing a postoperative wound. An increase in intra-abdominal pressure and, as a consequence, the development of abdominal compartment syndrome are formidable complications that require finding ways to solve them.

In anticipation of the story about the main problem, we should dwell on the most frequently encountered terms and concepts. Large-for-size syndrome usually occurs in pediatric LT due to the implantation of an excessively large liver graft into a small recipient cavity, resulting in increasing intra-abdominal pressure, and intra-abdominal hypertension which could lead to poor graft or recipient outcomes[13]. That is why the algorithm for preoperative preparation of patients must include accurate measurements: The volume of the liver graft, the size of the recipient’s abdominal cavity, and the liver mass coefficient necessary to meet the body’s needs.

According to the literature, today there is no clear consensus for measuring the volume of the intended graft, its mass, the required functional parenchyma, and methods for reconstructing the anterior abdominal wall during transplantation of the left lateral segment of the liver. We consider it necessary to dwell separately on the most common aspects of the above issues.

SURGERY PLANNING

Donor liver volumetry

LT is gradually advancing, including in developing countries[14]. Today, to accurately determine the size of the liver segments, two instrumental methods are used – ultrasound and computed tomography (CT) with 3D modeling. Since the density of the liver parenchyma is almost equal to the density of water, it is this property that is used in CT volumetry and allows one to measure its volume in cm³. Many authors give preference to CT in the preoperative assessment of left lateral section (LLS) volume. Others have an alternative point of view. Thus, Hatsuno et al[15] in his work compared the data on the average calculated volume of the LLS of the liver of 101 patients using ultrasound and CT. According to ultrasound 261 ± 118 cm³ and CT volumetry 274 ± 123 cm³, the relationship between the results of the two imaging methods was linear and statistically significant. Based on this, the author concluded that ultrasound has acceptable levels of sensitivity and specificity for routine measurement of the volume of the LLS[15].

Another important parameter in measuring the future LLS graft, which must be taken into account in pediatric practice, is its anteroposterior size. Kasahara et al[16] introduced the concept of “ratio of thickness” (RT)-the ratio of the maximum thickness of the graft to the dorso-ventral size of the recipient’s abdominal cavity[16].

Along with the calculation of the graft-to-recipient weight ratio (GRWR), the calculation of RT, according to Sakamoto et al[17], can be decisive in choosing surgical tactics. These indicators can be taken into account when deciding whether to reduce the LLS liver graft. He proposes the following algorithm for preoperative evaluation of graft type selection. As can be seen from the above diagram, when GRWR ≥ 4.0% or RT ≥ 1, there is a need to reduce the graft volume. Next, the RT indicator comes into force, which, if it exceeds one, dictates the need for transplantation of a single segment of the liver (segment II), or non-anatomical reduction of LLS when this coefficient is < 1. According to the algorithm proposed by the author, transplantation of the native LLS can only be performed with a combination of GRWR < 4% and RT < 1 (Figure 1)[17].

Figure 1 The preoperative algorithm for selecting the type of graft. LLS: Left lateral section; GRWR: Graft to recipient weight ratio.

A rather simple and interesting classification of liver LLS grafts by thickness is proposed by Shirouzu et al[18]. It divides grafts into 2 types: Blowfish (short and thick) and Flatfish (flat and long)[18]. Based on the characteristics of the graft obtained in the preoperative period, two surgical tactics are proposed. The Blowfish graft is hyperreducible. In turn, the Flatfish type LLS configuration does not require additional intervention and, according to surgeons, can be freely placed in the recipient’s abdominal cavity.

Based on data from CT volumetry or ultrasound of the donor’s liver, there is a need to accurately calculate the standard liver volume (SLV) for the recipient in order to avoid receiving a transplant that is too small or, on the contrary, too large. For this purpose, the formulas presented below in Table 1 were developed[19-27].

Table 1 Formulas for calculating standard liver volume.

Ref.	Formula	Used material (n)
DeLand et al[22], 1968	LV = 1020 × BSA -220	Autopsy (550)
Urata et al[19], 1995	LV = 706.2 × BSA +2.4	CT (96)
Noda et al[21], 1997	LV= 50.12 × BW 0.78	CT (54)
Heinemann et al[27], 1999	LV = 1072.8 × BSA - 345.7	Autopsy (1332)
Vauthey et al [26], 2002	LV = 18.51 × BW + 191.8	CT (292)
Yoshizumi et al[20], 2003	LV = 772 × BSA	Deceased donor grafts (1413)
Herden et al [23], 2013	Formula 1: For children from 0 month up to 1 year old: LV = -143.062973 + 4.274603051 × BH (cm) + 14.78817631 × BW (kg)	Autopsy (246)
	Formula 2: For children from year up to 16 years old: LV = -20.2472281 + 3.339056437: For BH (cm) + 13.11312561× BW (kg)	Autopsy (142)
Yang et al[24], 2018	LV = 331−4.1 × age + 41.6 × gender + 15.3 × BW; Gender: Male = 1, female = 0	CT (790)
Yang et al[25], 2021	BW ≤ 20 kg: LV = 707.12 × BSA 1.09; BW > 20 kg, males: LV = 691.90 × BSA 1.06; females: LV = 663.19 × BSA 1.04	CT (792)

LV: Liver Volume; BSA: Body Surface Area; BW: Body Weight; BH: Body High; CT: Computed tomography.

As can be seen from the many formulas given in the table, there is no consensus and formula for calculating the required graft mass. We will dwell in more detail on just some of them. The weight of the liver at birth is approximately 5% of the newborn’s body weight. This value is not constant throughout life and by the age of 16 it reaches a certain plateau, amounting to about 2%. According to Shirouzu et al[18] age, weight, or height are not accurate criteria for calculating the required liver mass, especially in sick children, who are often retarded in physical development. They consider it appropriate to take the body surface area as an individual value in each specific case as a reference point[19]. Based on this, using the method of simple regression of CT data to calculate (96 patients were studied, including 65 children), Urata et al[19] developed a formula for calculating the SLV (Table 1).

Noda et al[21] conducted another study to calculate a SLV appropriate for age. The study included 54 patients aged from 10 days to 22 years. Based on CT data, it was possible to calculate the liver volume ratio for individuals of different ages (Table 2), and thereby, like K. Urata, confirm the correlation between age and changes in liver volume[21].

Table 2 Average liver volume and liver volume concerning body weight in each age group, mean ± SD.

Groups	Quantity	Mean age	Liver volume (сm3/kg)	Liver volume / body weight (сm3/kg)
I	6	3.2 months	178.2 ± 81.9	34.09 ± 5.5
II	6	1 year 5.7 months	281.0 ± 51	28.56 ± 5.4
III	7	3 years 4.1 months	425.9 ± 94.9	31.83 ± 5.9
IV	10	7 years	596.5 ± 218.3	25.42 ± 4.5
V	8	13 years 9.4 months	1024.0 ± 210.3	23.77 ± 3.9
VI	17	18 years 10 months	1114.3 ± 192.9	20.17 ± 3.1

Based on the percentage of liver weight to recipient weight (GRWR), Kiuchi et al[28] propose to classify transplants into 5 groups: (1) Very small (extra-small-for-size)-XS; GRWR < 0.8%; (2) Small (small-for-size)-S; GRWR 0.8%-1.0%; (3) Medium (medium-for-size)-M; GRWR1-3.0%; (4) Large (large-for-size)-L; GRWR 3%-5.0%; and (5) Extra-large-for-size-XL; GRWR ≥ 5.0%.

After analyzing data from 276 related LTs, he concluded that the use of small grafts (less than 1% of the recipient’s body weight) leads to a decrease in his survival. This probably occurs due to greater trauma to hepatocytes after reperfusion (the so-called “portal attack”). At the same time, large transplants, although they have some anatomical and immunological disadvantages, demonstrate higher survival[28].

Based on two- and three-dimensional CT volumetry data, Yoshizumi et al[20] analyzed data from 70 healthy Japanese adults (46 men, 24 women, aged 20 to 65 years) and applied all the formulas that existed at that time (2008) for comparative analysis of liver volume. He concluded that the F. DeLand and A. Heinemann formulas overestimate liver volume compared to CT data, while the K. Urata formula, on the contrary, underestimates it[20].

Analysis of the formulas of Noda et al[21] and Yoshizumi et al[20] did not show a significant difference, however, they tend to underestimate the volume of the liver if its volume exceeds 1600 cm³. Such deviations are most likely associated with differences between the racial groups studied. This thesis is confirmed by the fact that DeLan et al[22] and Heinemann et al[27] calculated their formulas based on research data from a Western group of people, and Urata et al[19], Noda et al[21] and Yoshizumi et al[20] from an Eastern one. Concluding his analysis, Yoshizumi et al[20] indicates that it is impossible to recognize one or another formula as unsuitable, and each of them requires an individual approach[20,29]. The Herden et al[23] formula presented in Table 1 (called “Hamburg”) also deserves special attention, since it is more focused on young children, who are of particular interest in our work. The development of a formula for calculating SLV for children was based on autopsy data from 388 white patients under the age of 16 years. This formula was then applied to retrospectively recalculate all pediatric transplants performed between January 2000 and December 2010 at a single center. Depending on the graft weight and the calculated standard liver weight, the children were divided into 4 groups: (1) Small-for-size grafts ≤ 0.5; (2) Adequate graft (Size-matched grafts) > 0.5 to ≤ 1.5; (3) Large-for-size grafts > 1.5 to ≤ 2; and (4) Extra-large-for-size grafts > 2.

In addition, the GRWR was calculated for the patients included in the study, and the children were also divided into four groups: (1) Small-for-size grafts < 1%; (2) Size-matched grafts ≥ 1% to < 3%; (3) Large-for-size grafts ≥ 3% to < 4%; and (4) Extra-large-for-size grafts ≥ 4%.

When comparing the results of the Hamburg formula and GRWR, a significant difference was noted in groups “b” and “c”, while no significant differences were found in groups “a” and “d”. The next step was to compare graft survival in all four groups, where an increased risk of liver transplant rejection was found in children who received a small-sized liver fragment, while “large grafts” showed better survival rates[23].

It is important to note that there is no “ideal” formula for calculating SLV. Some formulas may overestimate or underestimate liver volume, which can be attributed to the fact that calculations may vary based on the results of CT volumetric analysis or GRWR indices depending on the index population. Consequently, data obtained in one region of the world may not always be directly applicable to another. For instance, it is well known that Asian adults tend to have smaller overall body sizes compared to European adults[20,29]. This variability is one of the reasons why there is no international consensus on the various published indices.

However, having knowledge of CT and/or ultrasound data-volumetry of the donor liver, as well as calculating the proper graft coefficient, the surgeon often intraoperatively has to deal with problems that a “large transplant” may entail. This mainly concerns the impossibility of primary closure of the wound due to the high tension of the tissues of the anterior abdominal wall. In order to prevent such complications and prevent the development of intra-abdominal hypertension, graft reduction methods, the use of monosegmental grafts and various options for plastic surgery of the anterior abdominal wall have been developed. It is advisable to dwell in more detail on each of the existing approaches

3D modeling

Currently, in the era of modern technology, methods for preventing large-for-size syndrome using 3-D printing have begun to be developed. Thus, a group of scientists from Korea, based on a CT scan of a patient, printed a 3D model of his abdominal cavity. Also, a graft printout was performed based on a CT scan of the donor’s liver. After this, the graft was “fitted in”. Thus, 16 patients, including children of the first year of life, were examined and prepared for LT. All the 16 cases with 3D printed abdominal cavity showed appropriate fitting of the donor’s liver graft to both the 3D printed model and actual recipient’s abdominal cavity with no large-for-size syndrome after LT[30].

SURGERY

Monosegmental transplantation

In the 1990s, Houssin et al[31] described a new graft reduction technique. It was proposed to use only one segment, resection of segment III of the LLS of the liver and use it as a graft. Since then, many different experiments have been carried out, but there is no single concept regarding monosegmental transplantation. According to Enne et al[32], with GRWR > 4, transplantation of the third segment is necessary, which seems technically simpler than transplantation of the second segment. The reduction method, from the author’s point of view, does not pose a danger to the afferent vascular pedicle or the left hepatic vein. In addition, the advantage of this technique is that the vascular anastomoses are performed in the same way as in LLS transplantation, and are not isolated intraparenchymatously as in SII transplantation. This study is based on 13 segment III transplantations. Patients operated on were aged from 27 to 454 days (average age was 211 days and with body weight from 2.45 kg to 7.4 kg (average body weight-5.2 kg). In the postoperative period, surgical complications occurred in 29, 4% of cases. This includes pleural effusion, diaphragm paralysis, stricture of biliary anastomoses, and biliary fistula. Of these, vascular complications accounted for 7.4% (1 case of portal vein thrombosis and 1 case of hepatic artery thrombosis were encountered). In this series of observations, there was no need for liver retransplantation. Patient survival was 85% over a follow-up period of up to 21 months[32].

Speaking about monosegmental transplantations, one cannot fail to mention the Japanese School of Transplant Surgery, which has achieved great success in this area. For the first time, Makuuchi et al[33] mentioned resections of individual liver segments using intraoperative ultrasound in Japan[33]. Subsequently, this technique was improved by Mizuta et al[34], who successfully performed 3 transplantations of the 2^nd segment of the liver in newborns weighing up to 3 kg. The second segment of the liver, in its anatomical configuration, is thinner than the third, which was the reason for the choice in its favor when planning transplantation for recipients weighing less than 5 kg. Mizuta et al[34] technique is as follows: For the most accurate anatomical resection, the portal vessels between segments II and III are ligated and intersected. Next, using the ultrasound contrast agent Sonazoid and ultrasound itself, the demarcation line between the second and third segments was determined intraoperatively, following which the parenchyma was divided[34]

In addition, in their study, Mizuta et al[34] substantiate the need to apply a temporary porto-caval shunt to children under 3 weeks of age. From his point of view, the imposition of an anastomosis between the recipient’s portal vein and his umbilical vein, which at this age is not yet obliterated, prevents the development of intestinal edema in very young children without portal hypertension. In cases of transplantation to slightly older children, when the umbilical vein is already obliterated, a direct portacaval shunt is applied. Thus, having performed 3 successful liver transplants in children weighing up to 3 kg and up to 27 days of age, Mizuta et al[34] consider it advisable to use segment II as a graft, especially in cases of fulminant liver failure in newborns.

Another Japanese surgeon, Sakamoto et al[17], shares a similar opinion, considering transplantation of SII for children weighing less than 5 kg to be more justified and safer than transplantation of SIII or the entire LLS of the liver. In addition to calculating the GRWR coefficient, which should not exceed 4%, it also calculates the above-mentioned TR coefficient, which in turn should not exceed one. From the point of view of surgical technique, his resection method also consists in ligating the portal branches to the third segment, which leads to the appearance of a demarcation line along which the division of the hepatic parenchyma is carried out. In preoperative assessment of the type of graft and selection of the optimal volume of resection, Sakamoto et al[17] suggests using 3D computer modeling[17].

Reduced and hyper-reduced grafts of the LLS

As an alternative to monosegmental transplantation, many authors[2,17,35-43], including Attia et al[44], suggest to use of a two-segment reduced liver graft. From his point of view, this technique has several advantages. Firstly, vascular anastomoses are applied as during transplantation of the native LLS of the liver, without disturbing the vascular architecture. Second, preserving the medial half of the second segment prevents torsion of the left hepatic vein, which sometimes occurs in third segment liver transplants. Note that Attia et al[44] were the first to perform a hyperreduced split transplant using this technique. The LLS was reduced by two additional parenchymal transections. The first plane of parenchyma dissection was determined by the portal branch of the third segment into which the probe was inserted. The main portal inflow in the third segment was necessarily preserved. A second resection plane passed through the second segment to further reduce the graft mass (Figure 2).

Figure 2  The scheme of hyper-reduction of the left lateral section graft, performed in both vertical and horizontal planes.

In 2008, Attia et al[44] published the results of his work. Of the 4 children, one died 5 days after surgery from massive hemorrhage in the brain with normal graft function. The second child died 10 months later, also with a normally functioning graft, due to primary pulmonary hypertension. The author cites only one case of arterial thrombosis, which was successfully resolved by re-anastomosing the hepatic artery. Data for thrombosis of the portal and hepatic veins, as well as biliary complications are not provided in the work[44].

Around the same time, in February 2008, Japanese surgeon Kasahara et al[45] published his experience of using hyper reduced LLS liver grafts in three children weighing less than 7 kg. Unlike Attia et al[44], who performed reductions of grafts obtained from deceased donors, he describes reductions during related transplantation. Moreover, all manipulations with the graft were carried out in the donor’s body under conditions of preserved blood circulation, which made it possible to avoid prolonged warm and cold ischemia. To select the optimal resection line, Kasahara et al[45] suggests using intraoperative ultrasound with Doppler ultrasound[45].

The technique for performing vascular anastomoses did not differ from that for transplanting the entire LLS. Biliary reconstruction was carried out with the jejunal loop switched off according to Roux. There were no complications in donors, nor vascular complications in recipients. In the above series of observations, 3-month survival after transplantation was 100%.

Another representative of the Japanese school of pediatric transplant surgeons, Shehata et al[46], when choosing the optimal transplant size, gives the leading importance to the calculation of GRWR and, in cases where this coefficient exceeds 4%, suggests hyperreduction of the LLS of the liver[46,47]. According to the author, from September 2000 to December 2009, 49 Liver transplants were performed on children whose average age was 7 months and weight 5.45 kg. Of these, 5 patients underwent partial resection of the liver, 26 patients underwent monosegmental transplantation, and 18 patients underwent hyperreduction of the liver monosegment. All graft reduction procedures were performed “in situ” during surgery at the donor. Postoperative complications were distributed as follows: Hepatic artery thrombosis occurred in two patients (4.1%), thrombosis and portal vein strictures in eight patients (16.3%). The average graft survival rate during the first year after surgery was 83.7%. Thus, M. Shehata believes that reducing the size of the LLS with GRWR > 4% should be a necessary procedure for LT in children weighing less than 5 kg[46].

Argentine colleagues Ardiles et al[48], in 2013, published their 14 years of experience in LT in pediatric practice. The surgical aspects of the liver graft reduction technique are similar to those presented above[48].

Balci et al[49] report about 4 infants who underwent living donor LT with heterotopically implanted reduced monosegmental or left lateral segment grafts. Reduced monosegment III grafts were used in 2 cases, and reduced left lateral segment grafts were used in the other 2 patients. All patients recovered from the LT operation and are doing well at a mean follow-up of 8 months[49].

Park et al[50] reported the Korean experience of the use of reduced and hyper-reduced LLS grafts. Thus, the average age and body weight of the patients were 4.0 ± 1.7 months and 5.3 ± 1.4 kg, respectively, for three cases of pediatric LT with HRLLS graft. The mean weight of the hyper-reduced LLS grafts was 191.7 ± 62.1 g and the GRWR was 3.75% ± 1.57%. All patients recovered from their respective LT operation and are actually in good health, more than 6 years after the LT. There was one case of pediatric deceased donor LT with in situ size reduction of recipient-graft size. Another case presented dextroplantation of a reduced LLS graft. A case of pediatric living donor LT using a monosegment graft procured by pure 3-dimensional laparoscopic LLS resection and in situ reduction was also reported. Authors concluded that making a hyper-reduced LLS or monosegment graft during living donor LT and split LT can be a useful option for treating pediatric patients[50].

It is important to mention that most authors who use liver graft reduction techniques perform it before implantation in situ at the donor stage of the operation, and in cases of using organs from a deceased donor–ex-situ at the “back table” stage. In situ, graft reduction is justified by the fact that the related donor is always a somatically healthy person who does not suffer from disorders of the hemostatic system, which significantly reduces the risk of blood loss when performing reduction compared to conditions of coagulopathy in the recipient. However, the literature also describes a case of partial reduction of the graft after implantation. Zenitani et al[51] described a clinical case that required reduction of the LLS liver graft after implantation. According to preoperative calculations, GRWR did not exceed 3.8%, but intraoperatively this coefficient was 4.3%. After the blood flow was started, zones of graft hypoperfusion appeared, and therefore a decision was made to perform partial resection. The author used a technique that was previously described for liver resections in adult patients, but this method was not used in children. A soft intestinal sponge was applied to the area of the graft that was to be reduced to flatten its surface, and then the liver parenchyma was cut off along the resulting line using a linear stapler. Emphasizing the positive aspects of this method of resection, the author notes the reduction in the duration of the operation and the minimization of surgical complications, such as bleeding or bile leakage[51].

It should be borne in mind that in patients with liver cirrhosis and ascites, the abdominal cavity is enlarged compared to that of healthy peers. Thanks to this circumstance, favorable conditions are created for the placement of a “large” transplant in the abdominal cavity of the recipient - a child. In the absence of portal hypertension, ascites, or in the development of fulminant liver failure, when cirrhosis and its complications do not have time to form, the size of the recipient's abdominal cavity may not correspond to the size of the graft, not only in the total volume but also in the anteroposterior plane. Based on this, in addition to the term “graft-to-recipient weight ratio”, the term “thickness ratio”, which was mentioned earlier, was introduced into clinical practice. That is why, along with classical methods of graft reduction, new non-anatomical types of reduction are being introduced. For example, to overcome the discrepancy between the anterior-posterior size of the graft and the abdominal cavity of the recipient, Kasahara et al[16] have resorted to removing the anterior surface of the graft since 2013 to reduce its thickness by more than 40%. In turn, this resection method increases the wound surface of the graft and the duration of the operation but provides a great advantage for the primary closure of the postoperative wound[16].

In 2021, Gavriilidis and Hidalgo[52] published a systematic review on the outcomes of using large grafts in children and surgical techniques for preventing large-for-size syndrome[52]. The review included case reports and retrospective studies on neonatal transplants where either mono-segmental transplantation or reduction of the LLS graft was performed. The authors concluded that a GRWR > 4% is a significant risk factor for the development of large-for-size syndrome. Mono-segmental transplantation and LLS graft reduction help decrease graft volume and thickness, thus providing the smallest possible graft. Medial reduction, in particular, deserves special mention, as it involves a non-anatomical reduction of segment 3, altering both the volume and thickness of the graft and enabling primary abdominal wall closure. One of the key points noted by the authors is that donor graft volumetry is not the only factor influencing the development of large-for-size syndrome or the choice of surgical strategy for its prevention. They recommend considering the volume of the child’s abdominal cavity, which can be affected by comorbid conditions such as liver size, ascites, sarcopenia, abdominal wall elasticity, and the thickness of the liver graft. The main causes of graft loss and patient mortality were infections (sepsis), thrombosis of the graft's afferent vessels, and graft rejection. Retransplantation was required in 10 patients, six of whom needed it in the early postoperative period following the initial surgery.

The experience of various centers with the use of reduced grafts and monosegmental transplantation is presented in Table 3. The largest cumulative experience among transplant centers involves the use of reduced grafts, with reductions performed both in situ and during the "back table" phase. Monosegmental transplantation is performed almost twice as rarely, likely due to the technical challenges associated with monosegmental resection. Nevertheless, the outcomes of these methods are comparable[52].

Table 3 Left lateral section graft reduction techniques, performed in various transplant centers.

Ref.	Number of patients	Recipient age, month, median	Recipient weight (kg), median (range)	Donor type	Graft type
Strong et al[67], 1982	1	5	4.7	DDLT	Mono-segment III
Mentha et al[41], 1986	1	11	6.9	DDLT	Mono-segment III
Srinivasan et al[68], 1999	6	7	3.45 (2.45–5.46)	DDLT	Reduced LLS graft (segment III reduction)
Noujaim et al[69], 2002	2	N/A	2.6	DDLT	Mono-segment III
Kasahara et al[70], 2003	14	7	5.95 (3.5–7.4)	LDLT	Hyper-reduced LLS graft (segment III, n = 13, Segment II, n = 1)
Attia et al[44], 2008	4	2.8	4.9 (2.9–7.8)	DDLT	Hyper-reduced LLS graft (segment III reduction)
Grabhorn et al[73], 2008	10	0.5	3.25 (2–4)	DDLT	Reduced LLS graft (n = 5); hyper-reduced LLS graft (n = 4)
Enne et al[32], 2004	12	12	6.2 (4.6–9.0)	LDLT	Mono-segment III
Thomas et al [71], 2010	9	10	7.5	DDLT (n = 9); LDLT (n = 1)	Hyper-reduced LLS graft
Shehata et al[46], 2012	44	7	5.45 (2.8–8.0)	LDLT	Mono-segment III (n = 26); reduced mono-segment III (n = 18)
Kanazawa et al[39], 2013	31	7	5.8 (2.8–8.5)	LDLT	Hyper-reduced LLS graft
Sakamoto et al[17], 2014	5	5	6.2 (4.0–9.4)	LDLT	Mono-segment II
Sanada et al[72], 2017	13	5	3.3 (2.6–7.1)	LDLT	Mono-segment II (n = 12); Mono-segment III (n = 1)
Raices et al[76], 2019	59	14	8.0 (7.0–8.9)	LDLT	Hyper-reduced LLS graft
Kitajima et al[74], 2018	89	7.3	5.9 (2.4–9.4)	LDLT	Hyper-reduced LLS graft (n = 47), Mono-segment II (n = 42)
Hirata et al[75], 2009	25	10	7.1 (6.4–8.0)	LDLT	Reduced LLS graft (Medial reduction)
Gautier et al [64], 2015	6	6	5.3 (3.5-6.0)	DDLT (n = 2); LDLT (n = 4)	Reduced LLS graft (n = 5), Hyper-reduced LLS graft (n = 1)
Balci et al[49], 2021	4	7.5	5.9 (4.6-7.8)	LDLT	Reduced LLS graft (n = 2), Reduced segment III graft (n = 2)
Park et al[50], 2022	6	4	5.3 (4.4-5.7)	LDLT (n = 5) DDLT (n = 1)	Reduced LLS graft (n = 3), Hyper-Reduced LLS graft (n = 3)

LLS: Left lateral section; DDLT: Deceased donor liver transplantation; LDLT: Living donor liver transplantation.

Anterior abdominal wall reconstruction

Some authors have a completely different point of view and in their works, they demonstrate that there is no need to use mono-segmental, reduced, or hyper-reduced liver grafts. Primary abdominal wall closure after pediatric LT is neither always possible nor advisable, given the graft-recipient size discrepancy and its potential large-for-size scenario. Some authors present delayed sequential abdominal wall closure. For example, Molino et al[53] demonstrate 27 cases of delayed sequential abdominal wall closure after pediatric LT due to large-for-size syndrome using a temporary silicone insert. All these patients had longer ICU and overall hospital stay but complete abdominal closure was achieved in 26 of 27 patients[53].

Zakaria et al[54] in their work present that temporary abdominal closure by silicon patch was done in 39 patients (52.7%) in the large-for-size group and 29 patients (11.3%) in the other group with a significant difference between both groups. Staged reduction in the size of the silicon patch until complete abdominal closure was done in 9 patients in the large-for-size group and the median time of abdominal re-closure was 5 days range 2–17 days[54].

In German surgeons Schulze et al[55] opinion, all mono- and reduced-size liver transplants increase surgical risks for the donor, increase the cold ischemia time, and for the recipient the incidence of biliary complications and impaired venous outflow from the graft increases. In addition, when transplanting the SIII monosegment, the main problem of pediatric transplantation is not solved-a reduction in the ventro-dorsal size of the liver, since the main thickness of the parenchyma falls on the third segment. To confirm his views, the author cites the experience of 41 successful liver transplants in children weighing less than 10 kg, where in all cases only LLS graft was used. When the GRWR was greater than 4% and the anterior abdominal wall could not be closed primarily, he used a temporary silicone insert. Overall patient and graft survival rates according to the authors data were 97% and 93%, respectively. The frequency of vascular complications was distributed as follows: Two arterial thromboses and one portal vein thrombosis in the early postoperative period. The author did not provide data on the presence of biliary complications. Thus, in approximately 50% of cases, the author was able to close the anterior abdominal wall primarily, subject to the following conditions: An intraoperative picture in which the edges of the surgical wound were brought together without tension; the second criterion was portal perfusion of the graft, measured by intraoperative duplex ultrasound. When the portal vein velocity decreased to 10 mL/min after wound closure, plastic surgery of the anterior abdominal wall was resorted to. The third condition was peak airway pressure. If it increased after suturing the surgical wound, a temporary silicone insert was used. Schulze et al[55] believe that ultimately, if these three criteria are taken into account, it is possible to avoid the need for reduction of the liver graft or plastic surgery of the anterior abdominal wall using artificial materials, and perform primary closure of the surgical wound[55].

There is no clear opinion regarding the use of various synthetic and biological materials for plastic surgery of postoperative wounds. Some authors believe that the lack of primary wound closure is a mandatory indication for the use of these materials. Others resort to all sorts of options for reducing the size of the graft, which were mentioned above. The latter's concerns are explained by the high risk of infectious complications when using inserts of a different nature, especially in children receiving immunosuppressive therapy.

In addition, fluid loss through the wound increases, the length of stay of children in the intensive care unit increases, and the need for repeated surgical interventions arises. However, despite all the positive and negative aspects of plastic materials, the frequency of their use and the pace of development are growing every year. All materials used for plastic surgery of abdominal wall defects can be divided into 2 types: Synthetic and biological (obtained from tissues of various animals).

According to the opinions of some authors, the use of biomaterials significantly reduces the risk of developing infectious complications compared to synthetic ones[56-59]. Technically, performing plastic surgery of the anterior abdominal wall does not present any great difficulties. As a rule, the material is fixed to the peritoneum. In this case, the skin is separated from the underlying layers for 3-5 cm in advance to create greater elasticity, then it is sutured with separate sutures[60].

Caso et al[56] from Spain in 2014 used a biomaterial obtained from the dermal matrix of a pig. A retrospective analysis of 6 cases of its use proved the validity of this approach. Materials created based on collagen stimulate neoangiogenesis, and increase the migration of fibroblasts to the site of their application, thereby improving engraftment rates and reducing the risks of infectious complications. The observation period averaged 22 months and, according to the data provided by the author, did not lead to wound infection in any case. In addition, the author is convinced that repeated intervention in the long-term postoperative period is not required and the abdominal wall is not subject to reconstruction[56].

The opposite opinion is shared by Australian surgeon Karpelowsky et al[59], who in their work cites the experience of using biomaterial made from the small intestine of a pig as a temporary patch for plastic surgery of the anterior abdominal wall. The study group included 10 children who underwent LT at the age of 5 to 15 months. The biomaterial was removed 3-6 days after transplantation. The author reported only one case of wound infection. In long-term follow-up, not a single case of wound infection or the appearance of a ventral hernia has been described[59].

Several published works are devoted to the plastic surgery of postoperative wounds using synthetic materials, often silicone or polytetrafluoroethylene. The previously mentioned some authors resort to the use of silicone inserts during LT and consider it necessary to remove them on the fifth postoperative day. By this time, the swelling of the graft and the anterior abdominal wall decreases, which makes final closure of the wound possible[46,53].

Another surgical option, described by Canadian researchers, is not to close the abdominal fascia. Thus, the surgical team determined the type of abdominal wall closure that would be performed based on the amount of tension required to reapproximate skin and/or fascia, and any evidence of respiratory compromise (increased airway pressures) by simulated abdominal wall closure. If the surgical team felt that fascial closure was not possible, closure of skin only was attempted. If skin closure was not safely possible, then a silastic mesh was cut to size and sutured circumferentially to the fascia with non-absorbable suture, and then vacuum system applied[61].

Belgian colleagues led by Lafosse et al[62] describe an interesting experience in the combined use of a patch of biological origin with an intradermal expander. A 2.5-year-old patient who underwent 3 Liver fragment transplantations required reconstructive surgery on the portal vein. The “old” postoperative scar formed after excision cast doubt on the possibility of primary closure of the wound after surgery. In this regard, expanders with 40 mL of 0.9% NaCl were introduced subcutaneously into the patient's upper abdominal cavity. Over 2 months, their volume was gradually increased by adding saline through the valves. After 10 weeks, when with the help of expanders, it was possible to increase the area of healthy skin, an operation was performed to apply a meso-Rex shunt. The aponeurotic layer was sutured using a biological mesh from porcine intestines, and thanks to preliminary preparation, skin sutures were applied without difficulty[62].

Goetz et al[63] reported about using biologic meshes for abdominal wall expansion in 6 cases of pediatric split LT[63]. Biologic mesh implantation for abdominal wall expansion was done in median 7 days (range: 3–11 days) after transplantation when signs of abdominal compartment syndrome with portal vein thrombosis in 3 and of the liver artery in 1 case occurred. In 2 cases, bovine acellular collagen matrix and 4 cases ovine reinforced tissue matrix was used. Median follow-up was 12.5 months (range: 4–28 months) and showed good liver perfusion by sonography and normal corporal development without signs of ventral hernia. One patient died because of fulminant graft rejection and emergency re-transplantation 11 months after the initial transplantation.

In the Russian transplant center named after academician V. I. Shumakov[64], sufficient experience in the use of “large” liver transplants in children with low body weight was accumulated. Despite numerous indications of the risk of developing compartment syndrome when using “large” LLS and descriptions of methods for preventing SIAH found in modern literature, the center has accumulated unique experience in the use of “large” LLS grafts in children with low body weight, which indicates the real possibility of closing the anterior abdominal wall when using “large” LLS, even with GRWR 4.0 or more. The report, that in all cases, native LLS graft was used[64,65]. They used a special technique for applying caval and portal anastomoses, as well as the original method of positioning the graft (in which the graft is placed in the position of the removed liver more to the right, with the resection surface of the graft towards the diaphragm), in the vast majority of cases they were able to close the surgical wound primarily and without tension. In the observed groups, there were no cases of the development of abdominal compartment syndrome, except in one case, during simultaneous transplantation of LLS of the liver and small intestine, where a xenopericardial flap was used for plastic surgery of the anterior abdominal wall. Also, they used silicone mesh sutured circumferentially to the fascia and applied vacuum-assisted closure dressing (Figure 3). The experience of various centers with the use of various surgical techniques for abdominal wall reconstruction is presented at Table 4.

Figure 3 Partial abdominal closure technique using vacuum aspiration abdominal wound dressing. A: Overall appearance after the implantation of a large left lateral section graft. Primary closure of the abdominal cavity is not feasible; B: Appearance after the application of a vacuum-assisted closure system.

Table 4 Abdominal wall plastic surgery, performed in various transplant centers.

Ref.	Number of patients	Surgery type	Material used for abdominal cavity protection
Molino et al[53], 2022	27	Partial abdominal wall closure	Silicone inserts
Zakaria et al[54], 2021	37	Partial abdominal wall closure	Silicone inserts
Schulze et al[55], 2011	11	Partial abdominal wall closure	Silicone inserts
Caso et al[56], 2014	6	Partial abdominal wall closure	Patches made from dermal matrix of a pig
Karpelowsky et al[59], 2009	10	Partial abdominal wall closure	Patches made from pig intestine
Goldaracena et al[61], 2020	18	Not to close abdominal fascia	N/A
Lafosse et al[62], 2012	1	Skin expander	N/A
Gautier et al[64], 2015	4	Partial abdominal wall closure	Xenopericardial flaps, vacuum aspiration abdominal wound dressings
Shehata et al[46], 2012	24	Partial abdominal wall closure, Skin-only closure	Polytetrafluoroethylene insert, silicone inserts
Goetz et al[63], 2023	6	Partial abdominal wall closure	Biologic meshes

Discussion of limitations of methods for preventing large-for-size syndrome

Currently, there is no consensus on the most preferable method for preventing large-for-size syndrome. The calculation of SLV is important, although, as mentioned earlier, there is no "ideal" formula for determining SLV. Some formulas may overestimate or underestimate liver volume, which can be attributed to the fact that calculations may differ based on the results of CT volumetric analysis or GRWR indices depending on the index population. Data obtained in one region of the world may not always be directly applicable to another. For example, it is known that Asian adults tend to have smaller overall body sizes compared to European adults. This variability is one of the reasons why no international consensus exists on these various published indices. This was partially confirmed in studies on SLV by various authors, where analysis of different population groups yielded varying SLVs[20,23,29]. However, there is limited data on this topic, and further research is needed. Nevertheless, CT volumetry of the donor liver plays a critical role in determining transplantation strategy, particularly when assessing the size compatibility of the potential graft with the recipient’s abdominal cavity. For this reason, it is essential not only to evaluate GRWR but also to consider the thickness of the LLS graft and the volume of the recipient's abdominal cavity.

Transplantation of a mono-segment carries certain surgical risks for the patient. Transplantation of the liver's third segment is rarely practiced globally nowadays, although attempts to use it have been documented[33,41,52,58,67,68]. This is due to the fact that the resection of the third liver segment is technically more challenging. Some authors note, that in aspect of implantation, transplantation of segment III is more challenging, particularly in terms of vascular reconstruction[64]. Additionally, some authors argue that transplanting the third segment does not address the issue of reducing the ventro-dorsal size of the graft, since the main thickness of the parenchyma falls on the third segment[55]. Consequently, transplantation of segment II is the preferred technique today. However, the resection of a mono-segment suitable for subsequent transplantation can be a technically complex procedure. In some clinics, the LLS of the liver is removed, followed by ex-situ resection of the third segment. This approach may increase the duration of cold ischemia and result in a more severe reperfusion syndrome[55,66], which could, in turn, affect the postoperative course for the recipient.

Reduction and hyper-reduction of the LLS graft are performed either ex-situ during the "back table" stage or after the graft has been implanted in the recipient's body[69]. One of the primary disadvantages of this method is the presence of a large number of wound surfaces on the graft, which could potentially lead to additional complications, such as bleeding or bile leakage[64,70,71]. Technically, reduction is simpler to perform than mono-segment resection, which can be considered one of the advantages of this procedure[72]. Also, reduced grafts were also utilized in emergency transplantations for children suffering from acute liver failure, where there is often insufficient time for detailed surgical planning[73]. Despite that, multivariate analysis revealed that fulminant liver failure, hepatofugal portal vein flow before LT, and non-anatomical reduction of LLS graft were associated with poor graft survival[74].

The use of silicone inserts, xenopericardial flaps or negative pressure wound therapy systems for temporary closure of the abdominal wall has shown promising results in cases where primary closure of the abdominal cavity is not possible[55,61,64]. However, the primary drawback of this approach is the necessity for one or more additional surgical procedures to achieve definitive closure of the abdominal cavity. Although in most cases the abdominal cavity can be successfully closed in several stages, there have been reports of cases where closure was not possible, albeit these reports are rare[53]. This increases the duration of hospitalization and may lead to associated complications, such as wound infections, anesthesia-related issues, and surgical complications. These potential complications are also relevant when using subcutaneous expanders[62].

Despite these limitations, the use of techniques aimed at preventing the development of large-for-size syndrome is justified[75]. Such interventions expand the pool of available donor organs and provide the opportunity to save children who might otherwise not survive without these preventive measures, especially in children with low body mass[76].

Undoubtedly, all approaches aimed at ensuring the safety of transplantation in low-weight pediatric patients are crucial, including initial donor liver assessment, volumetry, and the evaluation of the child's abdominal cavity capacity. The further development of innovative technologies, such as 3D modeling, has the potential to enhance these approaches[30]; however, these technologies are not available in all clinics due to the lack of appropriate equipment and their high cost.

It is also important to note that donor selection criteria, surgical approaches, and the experience in performing such procedures can vary significantly across different centers. As a result, institutional, national, underpowered sample, selection, detection, and learning curve biases may have influenced the outcomes[52].

CONCLUSION

Upon reviewing contemporary literature regarding the use of LLS grafts in pediatric patients, it becomes evident that despite the array of global practices, the optimal technical approach for employing "large" LLS grafts remains undetermined. Factors such as the potential for intra-abdominal hypertension, stemming from discrepancies between pediatric abdominal cavity volume and donor graft size, alongside the diversity of diseases and their varying severity, necessitate the application of diverse preventive measures. Common strategies include non-anatomical reduction, partial or hyperreduction of the graft, monosegmental transplantation, and abdominal wall reconstruction. Nonetheless, the definitive course of action typically emerges intraoperatively. Surgeons should utilize all available preoperative data, particularly CT volumetry of the donor liver and the native size of the child’s liver. This allows for a clear determination, even at the donor evaluation stage, of which surgical technique will be employed to prevent large-for-size syndrome. In cases where preoperative planning is not feasible (e.g., urgent split LT), graft reduction and/or anterior abdominal wall reconstruction (delayed abdominal wall closure) should be considered. Additionally, it is important to take into account the resources available at the clinic, as not all facilities may have access to the necessary tools for abdominal wall reconstruction.