During liver transplantation (LT) in pediatric patients, graft size matching to the recipient’s abdomen is important because implantation of a large-for-size graft might hinder the primary closure of the abdomen and thus induce various vascular complications [1,2]. To make a left lateral section (LLS) graft as small as possible, the LLS can be reduced to a hyper-reduced LLS (HRLLS) or monosegment or graft [1-4]. Anatomical resection of the segment II or III matches a segment III or II monosegment graft, but it requires thorough evaluation of the intrahepatic vascular and biliary anatomy [5-8]. In contrast, HRLLS grafts are produced by non-anatomical resection of the peripheral liver parenchyma and its feasibility is higher as compared to monosegment grafts. The demand for LT for pediatric patients has been persistently present, but the number of LT using HRLLS or monosegment grafts is very limited in Korea. Therefore, this study reviews the use of HRLLS or monosegment grafts in pediatric patients in Korea and delineates their characteristics.

The shape of the donor LLS is important for deciding whether non-anatomical size reduction will be performed. Non-anatomical resection is easy and effective, if the LLS looks like a flat fish. In contrast, if the LLS looks like a puffy fish, non-anatomical resection is not recommended because effective reduction of the graft thickness will be difficult [9]. Reducing the graft size is important in order to obtain an estimated graft-recipient weight ratio (GRWR) of less than 4%. Considering that size reduction rate following non-anatomical resection is usually less than 50%, donors with LLS volume comparable to 4% to 8% of GRWR on computed tomography (CT) volumetry are theoretically indicated for size reduction. The graft thickness-to-anteroposterior diameter in the recipient’s abdominal cavity ratio should be taken into account. If this ratio of thickness is less than 1.0, the donors are indicated for size reduction. If the GRWR is less than 4% and the thickness ratio is less than 1.0, the LLS grafts might not require size reduction [9].

The surface marking is made along the falciform ligament to split a usual LLS graft, and additional markings are made transversely at the ventral part of the segment III and vertically at the lateral part of the segment II. Firstly, the liver is transected to make a LLS graft. Thereafter, the parenchymal transection is extended along the additional marking lines. The transection plane should be perpendicular to the intrahepatic glissonian branches in order to prevent unnecessary ischemia. In living donor LT (LDLT), the HRLLS is recovered along the pace of recipient operation. In split LT, the HRLLS graft can be recovered before harvesting other organs. This method is usually adopted if the pediatric and adult recipients are treated in different hospitals. If the two recipients of the split liver grafts are treated in the same hospital, whole liver harvesting after in situ splitting is preferred, because complete separation of two split liver grafts at the back table is more convenient and safe [9].

This study was intended to describe the detailed techniques for harvesting and implanting HRLLS grafts developed in a high-volume liver transplantation center. The average age and body weight of the recipients were 4.0±1.7 months (range: 3–6 months) and 5.3±1.4 kg (range: 4.1–6.9 kg), respectively. Progressive familial intrahepatic cholestasis was diagnosed in two cases and biliary atresia was diagnosed in the other case. LDLT was performed in one case and split LT was performed in the two other cases. Non-anatomical size reduction was performed to the transected LLS grafts. The mean weight of the HRLLS grafts was 191.7±62.1 g (range: 120–230 g) and the GRWR was 3.75%±1.57% (range: 2.45%–5.49%). Widening venoplasty was applied to the graft left hepatic vein outflow orifice. Vein homograft interposition was used for portal vein hypoplasia. Regarding abdomen wound closure, primary repair, two-staged closure with a mesh, and three-staged repair with a silo and a mesh, were employed. The three patients recovered from the LT operation without any complications and are doing well, more than 6 years after LT. It can be concluded that making a HRLLS graft through non-anatomical resection during LDLT and split LT could be a useful option for treating pediatric patients [9].

We presented a case of pediatric deceased donor LT using a reduced whole liver graft in a 25-month-old boy weighing 12.7 kg. Following Kasai portoenterostomy for biliary atresia, the general condition of the patient deteriorated progressively. He was on the waiting list for LT with pediatric end-stage liver disease score of 15. The donor was a 51-month-old boy weighing 20 kg. The donor-to-recipient body weight ratio was 1.58%. The liver graft appeared to be larger than the recipient's abdominal cavity and thus, in situ size reduction was planned. The recipient surgery was performed following standard procedures. We performed a graft outflow vein reconstruction using a modified piggyback technique similar to the double inferior vena cava method. Since the portal vein was hypoplastic, side-to-side anastomosis was used. We also performed intraoperative portogram to embolize venous collaterals. After completing the graft implantation, it was found that the liver graft was too large to be accommodated within the recipient’s abdomen. After in situ resection of the LLS parenchyma, we successfully performed the primary closure of the abdominal wound. The patient experienced episodes of acute rejection and was in good health during four years after LT [10].

This case report describes a donor who underwent pure laparoscopic left lateral sectionectomy and in situ reduction using 3-dimensional laparoscopy and indocyanine green near-infrared fluorescence cholangiography to obtain a monosegment. A 43-year-old woman offered to donate part of her liver to her daughter, who required LT for acute liver failure after Kasai operation for biliary cirrhosis caused by biliary atresia. The donor’s height was 150.4 cm, her body weight was 56.8 kg, and her body mass index was 25.1 kg/m2. Liver dynamic CT showed a left lateral liver volume of 223 mL and an estimated GRWR of 4.4%. The entire procedure including the in situ reduction was performed under 3-dimensional laparoscopic view. The optimal bile duct division point was determined by real time indocyanine green fluorescence cholangiography. The total operation time was 320 minutes, with no transfusion required and no intraoperative complications. Intraoperative real time indocyanine green fluorescence cholangiography revealed the donor's bile duct anatomy and identified the optimal division point. The final graft weighed 167 g, 48 g was removed in situ and the GRWR was 3.3%. The donor was discharged 8 days following the operation with no complications. It was concluded that pure 3-dimensional laparoscopic left lateral sectionectomy and in situ reduction are feasible for obtaining a donor monosegment for pediatric LDLT [7].

Graft size matching is essential for successful LT in infant recipients. Herein, we presented our technique of graft dextroplantation used in an infant who underwent LDLT using a LLS graft. The patient was an 11-month-old female infant weighing 7.8 kg with hepatoblastoma. She was partially responsive to systemic chemotherapy and thus, LDLT was performed to treat the tumor. The living donor was the mother of the patient, a 34-year-old female. After non-anatomical size reduction, the weight of the reduced LLS graft was 235 g, with GRWR of 3.0%. Recipient hepatectomy was performed according to standard procedures for pediatric LDLT. At the beginning of graft implantation, the graft was temporarily placed at the abdomen to determine the implantation location. The graft portal vein was anastomosed with an interposed external iliac vein homograft. Since the liver graft was not too large and could be partially accommodated in the right subphrenic fossa, the abdominal wall wound was primarily closed. The patient recovered without complications. An imaging study revealed deep accommodation of the graft within the right subphrenic fossa. The patient has been doing well for six months without any vascular complications. This case suggests that dextroplantation of a reduced LLS graft could be a useful technique for LDLT in pediatric patients [11].

The position of the left side liver graft is important, since it could lead to complications of the hepatic vein and portal vein, especially in a small child using a variant LLS (vLLS) graft. The purpose of this study was to evaluate the outcome of a novel technique for the implantation of a vLLS graft to the right side (dextroplantation) in infants. For 3 years, 10 infants have undergone dextroplantation using vLLS grafts (group D). The graft was implanted to the right side of the recipient after 90° counterclockwise rotation; the left hepatic vein graft was anastomosed to the inferior vena cava using the extended right and middle hepatic vein stump, and the portal vein was reconstructed using oblique anastomosis without angulation. The surgical outcomes were compared with the historical control group (n=17, group C) who underwent conventional LT using vLLS during infancy. Group D recipients were smaller than group C (body weight <6 kg: 50.0% vs. 11.8%; p=0.03). The rate of GRWR >4% was higher in group D (60.0%) than in group C (11.8%; p=0.01). Surgical drains were removed earlier in group D than in group C (15 vs. 18 postoperative days; p=0.048). Each group showed 1 portal vein complication (10.0% vs. 5.9%); no hepatic vein complication occurred in group D, but 3 hepatic vein complications (17.6%) occurred in group C (p>0.05). Hospital stay was shorter in group D than in group C (20 vs. 31 postoperative days; p=0.02). Dextroplantation of a vLLS graft, even a large-for-size one, was successful in infants without compromising venous outcomes, compared with conventional vLLS transplantation. Since the surgical drains were removed earlier, hospital stays were reduced in the case of dextroplantation [12].

This report is the first systematic review of the existing evidence of reduced LLS and monosegment grafts in pediatric LT over the last 25 years [13]. The overall median weight of this cohort was 5.8 kg (2.6–8 kg) and it represented a complex group of recipients, where a significant donor-recipient weight ratio mismatch could be expected. This might result in a compartment syndrome which might compromise the inflow and outflow of the graft.

Regarding the results of surgical manipulation of LLS comparable to standard LLS, this cohort of 330 modified LLSs, albeit heterogeneous, provided an overall graft and patient survival of 84% and 89%, respectively, with a median follow-up of 39 months. The incidence of hepatic artery thrombosis and portal vein thrombosis were 1.5% and 4.2%, respectively. The comparison should be made against recipients with similar weight (5 kg) receiving a standard-whole LLS. A significant consequence in this approach would be the need for delayed/secondary abdominal wall closure, bridging the gap with a prosthetic or biological mesh [14,15].

At birth, the liver represents approximately 3.5% of our total body weight (range, 2.1–4.7) and 2%–1.5% (range, 1.8–2.8) in individuals >17-year-old [16]. An adult LLS might represent 16% (±4%) of the standard total liver volume (1,518±353 mL), averaging 242±79 mL [17]. Accordingly, a 5 kg-weighing child receiving a LLS graft, will likely face a GRWR close to 5%. It is well documented that GRWR is a strong predictor of graft survival and the vast majority of reports concur that grafts exceeding 4% will likely generate a conflict with the abdominal cavity’s capacity [1,15,16,18,19]. Hence, some groups advocate for reduction or conversion of LLS into a monosegment. However, a dogmatic view based only on the volume should be avoided, since other factors, such as, ascites, sarcopenia and specially graft thickness, might influence the LT (the myth of 4?) [20]. There is a clear preference for monosegment graft in Japan, whereas reduction seems to be more prevalent in western countries. In Japan, recipients of monosegment grafts were younger and lighter, with a bigger donor-recipient weight ratio mismatch and monosegment produced smaller grafts (median of 160 g), which finally achieved GRWR below 4%.

The large majority of reports would recommend aiming for a GRWR <4% to reduce the risks of a large-for-size scenario. Despite the technical complexities, reduced LLS and monosegment grafts are both valid options to facilitate pediatric LT with results comparable to standard LLS. Opting for monosegment graft diminishes the volume and thickness, thus providing the smallest possible graft, particularly in segment II.

LT in pediatric patients has increased worldwide, but it is still regarded as challenging because of large-for-size graft-related problems. The main problems related to large-for-size grafts include the risk of abdominal compartment syndrome due to the recipient’s small abdominal cavity, size discrepancies in vessel size, and insufficient portal circulation and tissue oxygenation [1,21-23]. To solve these critical problems, it is necessary to reduce the size of LLS grafts as much as possible, by making HRLLS or monosegment grafts.

The targeted graft size reduction for LT in pediatric patients is to make the estimated GRWR less than 4% [9]. Anatomically, a monosegment graft would be ideal for making a very small liver graft with effective reduction in graft thickness. However, it is more difficult to make a monosegment graft than a HRLLS graft. It is essential to assess the donor liver anatomy thoroughly to make a monosegment graft and hence this technique is not usually practical during split LT. In contrast, in situ non-anatomical resection used to make a HRLLS graft is technically easier and more intuitive than an anatomical monosegmentectomy.

The shape of a HRLLS graft is also important because non-anatomical resection does not effectively reduce the graft thickness [1,9]. The graft shape is evaluated using the graft thickness-to-anteroposterior diameter in the recipient’s abdominal cavity ratio [19]. If the ratio exceeds 1.0, the primary abdominal wall closure can induce graft compression and abdominal compartment syndrome, so temporary closure with a prosthetic mesh should be taken into account. However, it is more difficult to assess the graft thickness during split LT, since liver imaging studies are not always available. Some patients who underwent split LT failed to achieve primary abdominal wound closure, because the HRLLS grafts were still too large and too thick to fit within the abdominal cavity.

Implantation of a large-for-size graft can induce various vascular complications. The anastomosis site of the graft hepatic vein could be compressed or twisted because of the graft compression induced by the tight abdominal wall, which could lead to hepatic vein outflow obstruction. To prevent such complication, customized unification venoplasty could be conducted in order to facilitate outflow vein drainage [24,25]. Since the amount of portal blood flow is small in pediatric patients, there is a potential risk of portal hypoperfusion of the graft. Any anastomotic stenosis could interfere with the portal blood supply to the graft, so branch patch of the recipient portal vein is usually used for recipients with normal portal vein. If there is portal vein hypoplasia, which is often observed in biliary atresia, interposition of vein homograft is used to prevent anastomotic stenosis and portal hypoperfusion [26].

The surgical technique for in situ size reduction to make a HRLLS is the same for both living and deceased donors, primarily because it is performed on non-anatomy basis and surgeons’ experience-based intuition. However, HRLLS is not effective for reducing the graft thickness and thus, judicious selection of a suitable donor is important.

In conclusion, making a HRLLS or monosegment graft through non-anatomical and anatomical resection during LDLT and split LT can be a useful option for treating pediatric patients.

There was no funding related to this study.

All authors have no conflicts of interest to declare.

Conceptualization: CSP. Data curation: All. Formal analysis: All. Investigation: All. Methodology: All. Writing – original draft: All. Writing – review & editing: All.