Short limb in biliary tract reconstruction: A retrospective study of living donor liver transplantation in children with biliary atresia

Abstract

Roux-en-Y cholangiojejunostomy is a standard procedure for biliary reconstruction in pediatric living donor liver transplantation (LDLT). However, there is uncertainty on whether the adult standard of Roux branch limb is suitable for pediatric LDLT and its impact on postoperative biliary complications (BC). This study aimed to explore the effect of the short Roux limb and standard limb on pediatric LDLT biliary reconstruction. According to the length of the Roux limb, 168 LDLT children were divided into the routine limb group (n = 108) and the short limb group (n = 60). The incidences of postoperative biliary tract complications between the 2 groups were compared retrospectively. The mean Roux limb length in the short limb group was significantly shorter than that in the routine limb group group (P < .01). There were significant differences in age, height, and weight between the 2 groups (P < .01). However, there were no significant differences in graft-to-recipient weight ratio, intraoperative blood loss, cold ischemia time, and operation time between the 2 groups (P > .01). Moreover, postoperative BC, including refluxing cholangitis, were similar between the 2 groups (P = .876). Furthermore, the history of Kasai surgery, the history of postoperative RC of Kasai, and whether or not the Roux limb was reconstructed had no significant effect on the occurrence of postoperative RC. There was no significant difference in postoperative BC between the short limb and the routine limb in children with living donor liver transplantation.

1. Introduction

Liver transplantation is the only effective treatment for children with end-stage liver diseases. In recent years, the development of surgical technology has significantly improved the long-term survival of patients and grafts, but postoperative biliary complications (BC) still seriously affect patient survival and prolong hospital stays.^[1] Although different transplant centers have different clinical outcomes, most central studies confirm that the incidence of BC in pediatric liver transplantation ranges from 10% to 45%.^[2] Especially in infant living donor liver transplantation (LDLT), the occurrence of BC severely affects survival and quality of life. Therefore, the way of biliary reconstruction and the prevention of BC after liver transplantation become the focus of clinical research.

The bile duct reconstruction after LDLT mainly includes duct-to-duct choledochocholedochostomy and Roux-en-Y cholangiojejunostomy (RYHJ). In the vast majority of cases, RYHJ is the inevitable choice for bile duct reconstruction in children with LDLT. Intestinal loop reconstruction is a key component of RYHJ bile duct reconstruction. At present, in the process of pediatric live donor liver transplantation, most surgeons, except for a few, still adhere to the traditional adult standard of 40 cm long ROUX intestinal loop length in RYHJ intestinal loop reconstruction. Unfortunately, unnecessary length often leads to some postoperative complications, such as cholestasis,^[3] reflux Cholangitis,^[4] and malabsorption of fat and fat-soluble vitamins. Congenital biliary atresia (BA) is one of the common surgical indications for liver transplantation in children. However, a considerable portion of these children may have undergone Kasai surgery before liver transplantation. Due to the surgical habits of different surgeons and the impact of postnatal growth and development on children, the original ROUX intestinal loop may be very long or short. During the liver transplantation process in these children, if there is no tension in the ROUX intestinal loop, surgeons usually use the original Kasai ROUX intestinal loop for biliary reconstruction. Therefore, utilizing the characteristics of liver transplantation in children with BA, we compared the results of shorter Rouxl limb lengths with traditional limb lengths in pediatric LDLT for biliary reconstruction.

2. Patients and methods

2.1. Patients

The study was approved by the Medical Ethics Committee of the Beijing Friendship Hospital, Capital Medical University. All child guardians had signed informed consent forms. A total of 168 pediatric patients who underwent liver transplantation for BA in Beijing Friendship Hospital, Capital Medical University between January 2016 and January 2022 were included in the study. There were 86 males and 82 females. Inclusion criteria: Age < 18 years old; Liver transplantation for the first time because of BA; Not incorporating other organ transplants; Complete clinical data; The primary disease is BA; The follow-up period is ≥ 3 years. Exclusion criteria: Nonanastomotic stenosis; Perform ≥ 2 liver transplantation surgeries; Died before being discharged from the hospital after liver transplantation.

2.2. Grouping and surgery

The 168 children were divided into short limb group (SLG) groups and routine limb group (RLG). In SLG, children had undergone Kasai operation. If there is no tension during biliary reconstruction during liver transplantation, the original Kasai postoperative intestinal loop may still be used, but the residual ROUX intestinal loop may be <40 cm. These patients were included in the SLG group. For children with a history of Kasai surgery, if there is tension during the reconstruction of the original ROUX intestinal loop during liver transplantation, the ROUX intestinal loop should be rebuilt with a standard length of 40 cm. These patients were included in the RLG group. Finally, there were 60 children in RLG and 108 children in SLG.

2.3. Outcome measures

Bile leakage, the bilirubin level in the drainage fluid of the abdominal drainage tube exceeds the level of bilirubin in serum. Biliary intestinal anastomotic stenosis, after intrahepatic bile duct dilatation is found on ultrasound or computer tomography, the anastomotic stenosis is directly diagnosed by percutaneous transhepatic cholangiography or endoscopy.^[5] Refluxing cholangitis (RC), after liver transplantation in BA children, there is a high fever that cannot be explained by other reasons, including body temperature over 38°C, blood leukocytosis with or without neutrophil elevation, hyperbilirubinemia or lighter stool color, without evidence of other site infection or ultrasound imaging of biliary duct dilatation.^[6]

2.4. Statistic analysis

SPSS 26.0 software was used for analysis. The Kolmogorov–Smirnov test was used to test the normality of quantitative data. Data with normal distribution were expressed as mean ± standard deviation and compared by t test. Data with non-conforming distribution were expressed as median and quartile distance M (P25, P75) and compared by the Mann–Whitney U test. Qualitative data were represented by frequency and component ratio (n [%]), and χ² test was used for inter-group comparison, and Fisher exact test was used if the expected frequency of any cell was < 5. P < .05 was considered statistically significant.

3. Results

3.1. General characteristics between the 2 groups

Among the 168 BA children, there were 60 cases in the SLG and 108 in the RLG, including 86 males (51%) and 82 females (49%). The general characteristics of the 2 groups were shown in Table 1. The minimum age was 4 months and the maximum age was 119 months. The minimum limb length was 20 cm. Compared with RLG, the SLG group had obvious differences in age, height, weight, and length of intestinal loop (P < .05). However, there were no significant differences in graft-to-recipient weight ratio, cold ischemia time (CIT), intraoperative bleeding, operation time, and postoperative BC between the 2 groups (P < .05). For the occurrence of postoperative RC, most of them occurred within 1 month after surgery, and the incidence in SLG and RLG was 18.33% and 15.74%, respectively (P > .05). There were 2 cases of postoperative intestinal (incomplete bowel obstruction) complications in the RLG group.

Table 1.

Comparison of the basic conditions between SLG and RLG.

Items	SLG (n = 60)	RLG (n = 108)	P values
Male n (%)	34 (56.67)	52 (48.15)	.290
Age (mo)	12 (10, 15.75)	9 (7, 15.75)	.001
Height (cm)	73.00 ± 9.6	70.00 ± 6.2	.039
Weight (kg)	9.00 ± 4.4	8.50 ± 2.5	.041
GRWR	2.70 (2.10, 3.30)	3.00 (2.40, 3.40)	.230
Limb (cm)	22.00 (20.00, 25.00)	40.00 (40.00, 40.00)	.001
RC (n [%])	11 (18.33)	17 (15.74)	.876
CIT (min)	97.5 ± 37.3	90.2 ± 32.6	.350
Intraoperative bleeding (mL)	349.9 ± 180.0	275.2 ± 130.2	.312
Anastomotic leakage n (%)	2 (3.33)	5 (4.63)	.423
Anastomotic stricture n (%)	4 (6.67)	6 (5.56)	1.000
Intestinal complications n (%)	0 (0.00)	2 (1.85)	.538
Time of operation (min)	510.00 ± 81	502.00 ± 112	.663

CIT = cold ischemia time, GRWR = graft-to-recipient weight ratio, RC = refluxing cholangitis, RLG = routine limb group, SLG = short limb group.

3.2. The influence of Kasai surgical history on the occurrence of RC after liver transplantation

Of the enrolled children with BA, 72 had received Kasai surgery before liver transplantation and 56 had not received Kasai surgery. All children in the SLG group were patients after Kasai surgery, and some patients in the RLG group were patients after Kasai surgery. To explore whether Kasai surgery affected postoperative RC, the patients with or without Kasai surgery before transplantation were compared. As shown in Table 2, the history of Kasai surgery before transplantation had no significant effect on the occurrence of postoperative RC (P = .209).

Table 2.

Comparison of RC after transplantation with or without Kasai surgical history.

Items	RC after LT	Without RC after LT	P value
Kasai group (n = 96)	19 (19.80%)	77 (80.20%)	.209
No Kasai group (n = 72)	9 (12.50%)	63 (87.50%)

LT = liver transplant, RC = refluxing cholangitis.

3.3. Effect of postoperative RC on RC after LT

Due to intrahepatic bile duct dysplasia and intestinal content reflux, RC is a common and difficult-to-treat complication after Kasai surgery. Most RC occurred within 1 year after surgery, and its incidence was as high as 60% to 90%.^[7] To explore whether postoperative RC after Kasai surgery affected the occurrence of RC after liver transplantation, we compared the 2 groups with or without RC after Kasai surgery. As shown in Table 3, there was no significant difference in the incidence of RC after transplantation between the 2 groups with or without RC after Kasai surgery (P = .238).

Table 3.

Comparison of the effects of postoperative RC of Kasai on post-transplant RC.

Items	RC after LT	Without RC after LT	P values
RC after Kasai (n = 49)	12 (24.50%)	37 (75.50%)	.238
No RC after Kasai (n = 47)	7 (14.90%)	40 (85.10%)

LT = liver transplant, RC = refluxing cholangitis.

3.4. Effects of gestural limb retention and reconstruction on postoperative RC during BR in LT

Among 96 LT children after Kasai, 66 had gestural limb preservation and 30 had reconstructed intestinal limbs during biliary reconstruction. The incidence of post-transplantation RC in the intestinal loop preservation group and the intestinal limb reconstruction group was 16.7%, and there was no significant difference between the 2 groups (P = 1, Table 4).

Table 4.

Comparison of intestinal limb retention and reconstruction in RC.

Items	RC after LT	No RC after LT	P value
Roux limb retention (n = 66)	11 (16.70%)	55 (83.30%)	1.000
Roux limb reconstruction (n = 30)	5 (16.70%)	25 (83.30%)

LT = liver transplant, RC = refluxing cholangitis.

4. Discussion

In 1893, Cesar Roux first used RYHJ anastomosis in gastrointestinal surgery,^[8] and then this operation was gradually applied to biliary reconstruction. After long-term clinical practice, RYHJ anastomosis has been the most common and widely used surgical method in biliary reconstruction. In many cases, RYHJ anastomosis must be used for biliary reconstruction after liver transplantation in children. Congenital BA, sclerosing cholangitis, etc. The recipient has no normal biliary tract that can be effectively used or has received a Kasai operation; The biliary duct in children is small and seriously mismatched with the diameter of the biliary tract of the donor liver; There are multiple bile ducts in the donor liver; Split or reduced volume donor liver.^[9] However, due to the loss of the physiological function of the Oddi sphincter, there is a risk of BC, such as RC after RYHJ.

However, in the process of pediatric liver transplantation, many surgeons habitually adopt the adult standard and apply the 40-cm long Roux limb to children, which leads to unnecessary long length of Roux limb, especially for younger infants.^[10] Children have different growth characteristics from adults. The Roux intestinal loop grows and enlarges as children get older. The average small intestine length of infants is 275 cm, which can reach 380 cm at the age of 1 year, while the average small intestine length of adults can reach 600 cm.^[11] Excessive length of the Roux limb often leads to obstruction of bile drainage, stasis, and retention of food residues,^[3] which may cause repeated infection and gallstone formation. Because of the different growth rates between the intestine and the mesentery, the Roux limb can become twisted and obstructed.^[12] In addition, bacterial overgrowth increases the loss of bile salts, resulting in impaired gut circulation and impaired absorption of fat and fat-soluble vitamins A, D, E, and K.^[13,14]

The reconstruction of RYHJ is one of the keys to the success of pediatric liver transplantation. In clinical practice, the length of the residual Roux limb after the Kasai operation is usually <40 cm, but the choledochojejunostomy is tension-free and the original limb can be reserved for choledochojejunostomy. It has been reported that the incidence of BC of RYHJ with a short Roux limb is not significantly different from that of RYHJ with a standard Roux limb of 40 cm.^[15,16] When RYHJ is first used in gastrointestinal surgery, the length of the Roux Limb was only 12 cm.^[8] Later, some animal models suggest that the ideal Roux limb length should be 30 cm, which is because the regurgitation of gastrointestinal contents can be effectively prevented at this length.^[17] Some surgical works^[18] show that the Roux limb ranges from 40 to 75 cm. The refluxing of digestive juice and food residue into the biliary tract through the nonfunctional intestinal loop is the cause of RC. Therefore, it is predicted that the longer the intestinal loop, the more the regurgitation of food residue can be prevented, and then the postoperative RC can be prevented. It has been demonstrated that a prolonged Roux limb can be used to treat postoperative RC.^[19] However, common bile duct duodenum anastomosis and duodenum anastomosis are the most likely procedures to cause intestinal contents to enter the bile duct in the clinic. However, the incidence of RC in these procedures does not appear to be high.^[20,21] Therefore, it has been suggested that reflux of intestinal contents and food residues are not the causes of RC, and a comparative study has confirmed that prolonged Roux limb is not effective in preventing postoperative RC.^[22]

Based on the above reasons and considering the small size and growing characteristics of children, there is no need for a long length of Roux intestinal limb in biliary reconstruction after liver transplantation in children. On the contrary, with the growth of children, the length of the Roux intestinal limb will become longer and longer, which is easy to lead to bile stasis, RC, and even stone formation.^[10] Considering that there is no absolute relationship between Roux limb length and RC, the length of Roux limb should not be preset, but should be designed according to the individual situation of children. It has been reported that the length should be the distance from the umbilical cord to the hilar.^[16]

In addition, with the continuous progress of endoscopic technology, enteroscopy is more and more widely used in the treatment of BC after pediatric liver transplantation. RYHJ reconstruction is closely related to the success rate of endoscopic treatment. There is no doubt that the long Roux limb will make endoscopic treatment difficult.^[23]

The biggest feature of pediatric liver transplantation is the continuous growth of children after the operation. Therefore, the biggest deficiency of the study is that the follow-up time is still insufficient, and the long-term effect remains to be observed.

5. Conclusion

Our current study confirms that there are no significant differences in BC after pediatric liver transplantation between short Roux intestinal limb RYHJ and traditional RYHJ. In line with the characteristics of children continuous growth, individualized or short Roux limb seems to be more suitable for biliary reconstruction after pediatric liver transplantation.

Author contributions

Conceptualization: Yuhong Zhang, Zhijun Zhu.

Data curation: Yuhong Zhang, Zhijun Zhu.

Investigation: Yuhong Zhang, Zhijun Zhu.

Methodology: Yuhong Zhang, Zhijun Zhu.

Project administration: Yuhong Zhang, Zhijun Zhu.

Resources: Yuhong Zhang, Zhijun Zhu.

Software: Yuhong Zhang, Zhijun Zhu.

Supervision: Yuhong Zhang, Zhijun Zhu.

Validation: Yuhong Zhang, Zhijun Zhu.

Visualization: Yuhong Zhang, Zhijun Zhu.

Writing – original draft: Yuhong Zhang, Zhijun Zhu.

Writing – review & editing: Yuhong Zhang, Zhijun Zhu.