Technique and early outcomes of total laparoscopic Kasai portoenterostomy in Type III biliary atresia

Abstract

Purpose

To evaluate the early outcomes of total laparoscopic Kasai portoenterostomy (TL-KPE) for Type III biliary atresia (BA).

Methods

A retrospective analysis included 45 pediatric patients with Type III BA who underwent Kasai portoenterostomy between January 2018 and December 2024. Patients were divided into three groups (n = 15 each): open KPE (O-KPE), conventional laparoscopic KPE (L-KPE, with extracorporeal intestinal anastomosis), and TL-KPE (with fully intracorporeal anastomosis). Comparative parameters included operative time, intraoperative blood loss, postoperative recovery metrics, incidence of cholangitis within 6 months, jaundice clearance, and native liver survival rates.

Results

Compared with the O-KPE group, the TL-KPE group showed significantly less intraoperative blood loss (P = 0.016). Compared with both the O-KPE and L-KPE groups, the TL-KPE group had markedly shorter time to bowel recovery and postoperative fasting time (all P < 0.001). Postoperative hospital stay was also significantly shorter in the TL-KPE group compared to the O-KPE (P = 0.016) and L-KPE (P = 0.034) groups. There were no significant differences among the three groups in operative time (P = 0.052), incidence of postoperative complications (P = 0.132), jaundice clearance cases (P = 0.278), early cholangitis (P = 0.932), or recurrent cholangitis (P = 1.000). No perioperative deaths occurred in any group.

Conclusion

TL-KPE is a safe and technically feasible approach for Type III BA that enhances short-term postoperative recovery. These early results warrant further multicenter prospective studies with longer follow-up to evaluate potential long-term benefits on native liver survival.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12893-026-03806-2.

Background

Biliary atresia (BA) is a progressive, obliterative cholangiopathy and one of the most severe hepatobiliary diseases in infancy. Kasai portoenterostomy (KPE), first described in the 1950s, remains the primary palliative surgical procedure to restore bile flow and delay the need for liver transplantation [1].

With the advancement of minimally invasive surgery, laparoscopic KPE (L-KPE) was introduced in 2002 [2] and has since been recognized as an effective approach [3]. However, in conventional L-KPE, construction of the Roux-en-Y jejunal limb typically requires exteriorizing the bowel through an extended umbilical incision. This maneuver may compromise the incision, increase the risk of bowel wall trauma and edema, and potentially delay postoperative recovery [4, 5].

In contrast, total laparoscopic KPE (TL-KPE), which involves performing all steps, including the intestinal anastomosis, entirely intracorporeally, could theoretically overcome these limitations by avoiding bowel exteriorization and its associated morbidity. While such fully intracorporeal techniques have been successfully applied in other pediatric hepatobiliary procedures, such as radical resection of choledochal cysts [6], their application in KPE has not been systematically reported. Therefore, this study aimed to evaluate the safety and efficacy of TL-KPE for Type III BA in comparison with conventional L-KPE and open KPE (O-KPE).

Materials and methods

Study design and participants

A retrospective review was conducted on 45 consecutive pediatric patients diagnosed with Type III BA who underwent KPE at our institution between January 2018 and December 2024. Inclusion criteria were: (1) postnatal age at surgery < 90 days, and (2) intraoperative cholangiographic confirmation of Type III BA. Exclusion criteria included major concomitant congenital malformations, perioperative mortality (none occurred in this cohort), or incomplete medical records. Following detailed consultation that included an explanation of the operative characteristics, potential advantages, and risks of each technique based on current literature and our preliminary experience, the surgical approach was chosen by the patients’ guardians from among three options: TL-KPE, L-KPE, or O-KPE. To minimize potential bias from preference-based allocation, the same preoperative consultation script was used for all families, and surgeons did not make recommendations for a specific technique. Nevertheless, residual confounding cannot be excluded. This study was approved by the Ethics Committee of the Second Affiliated Hospital of Xi’an Jiaotong University (Approval No. 2023 − 153). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from the guardians of all individual participants included in the study. Clinical trial number: not applicable.

TL-KPE group

Under general anesthesia, patients were placed in a supine position with the right side elevated. A four-port technique was employed. Pneumoperitoneum was established to 8 mmHg via a 5-mm trocar placed in a supraumbilical midline incision, which also served as the camera port. A 3-mm trocar was inserted in the right midclavicular line below the costal margin. Diagnostic laparoscopic cholangiography was performed through the gallbladder to confirm the diagnosis. Two additional 3-mm operating ports were placed: one at the lateral border of the right rectus abdominis muscle at the umbilical level, and the other at the lateral border of the left upper rectus abdominis muscle (Fig. 1A). The surgeon stood on the patient’s right side, assisted by a second surgeon on the left.

Fig. 1.

A Port placement for total laparoscopic Kasai portoenterostomy. 5‑mm umbilical port (camera), a 3‑mm right subcostal port, and two 3‑mm operating ports at the lateral borders of the right and left rectus muscles. B The fibrous portal plate after resection. Blue dotted line indicates the planned anastomotic boundary. Note bile leakage at the 2 and 10 o’clock positions (arrows), confirming patency of microscopic bile ductules. C Transection of the jejunal mesentery using LigaSure. D Site of end‑to‑side jejunojejunostomy; the distal end of the hepatic limb is closed. E Posterior wall of the hepaticojejunostomy being sutured with a continuous loose‑placement technique; sutures are not tightened until most of the posterior wall is completed. F Anterior wall anastomosis completed with continuous sutures; the suture ends are tied at the 3 and 9 o’clock positions

To optimize exposure of the porta hepatis, the quadrate lobe was retracted anteriorly. A 3 − 0 suture was passed through the peritoneal reflection near the falciform ligament and the fibrous layer of the gallbladder bed, then suspended to the anterior abdominal wall. The fibrous portal plate was resected. The distal common bile duct was isolated, ligated, and divided. The stump was retracted superiorly, and the fibrous tissue was dissected toward the liver hilum using an electrocautery hook. The resection extended superiorly to the inferior edge of the quadrate lobe, inferiorly to the posterior wall of the portal vein bifurcation, and laterally to the points where the portal vein branches enter the liver parenchyma (Fig. 1B). Minor bleeding from the transected surface was controlled with gauze compression. Following resection, the patency of microscopic bile ductules was confirmed by observing immediate bile leakage from the portal plate surface, notably at the 2 and 10 o’clock positions (Fig. 1B).

Laparoscopic Roux-en-Y limb construction

As previously described [6], the jejunum was divided 10–15 cm distal to the ligament of Treitz using a LigaSure device, with the division made perpendicular to the intestinal axis (Fig. 1C). LigaSure effectively seals the cut ends, minimizing content spillage while preserving adequate blood supply to the margins. All anastomoses were performed with hand-sewn techniques, and no stapling devices were used. The decision to avoid staplers was based on the limited availability of appropriately sized linear cutting staplers for small infants, the restricted working space in the pediatric abdominal cavity, and the desire to reduce surgical costs and potential stapler-related complications (e.g., anastomotic strictures, foreign body reactions). The distal end (future hepatic limb) was closed with a continuous, full-thickness suture using 5 − 0 absorbable material. An end-to-side jejunojejunostomy was created approximately 40 cm distal to this closed end (Fig. 1D). The posterior wall was constructed with a continuous, full-thickness locking (inverting) suture, and the anterior wall with a continuous seromuscular suture.

Laparoscopic portoenterostomy

The Roux limb was delivered to the porta hepatis through a retrocolic tunnel created in an avascular area of the transverse mesocolon. A side enterotomy matching the portal anastomotic plane was created on the antimesenteric border, 0.5 cm from the blind end. The hepaticojejunostomy was performed with a 5 − 0 PDS suture using a continuous, loose-placement technique. The posterior layer was initiated by taking the first bite from the leftmost end of the enterotomy on the hepatic limb, with the corresponding bite at the 3 o’clock position on the portal plate. The suture was not tied. Continuous suturing was then carried out posteriorly toward the 9 o’clock position. Critically, the suture was not tightened immediately after each pass; all stitches were placed loosely to ensure perfect visualization of both edges (Fig. 1E). After more than two-thirds of the posterior wall had been sutured, the running suture was gradually and evenly tightened to approximate the tissues. The remaining posterior stitches were then placed and tightened to complete the posterior layer. A second 5 − 0 PDS suture was used for the anterior layer. Starting again at the 3 o’clock position, continuous suturing was performed anteriorly toward the 9 o’clock position, similarly without tying or tightening during placement to maintain optimal visualization (Fig. 1F). This continuous suturing technique with delayed tightening offers improved exposure of the anastomotic edges throughout suture placement, which may enhance the accuracy of the anastomosis and reduce technical difficulty compared to traditional interrupted sutures. Upon completion, the anastomosis was secured by tying the corresponding ends of the posterior and anterior sutures together at both the 3 o’clock and 9 o’clock positions. Finally, the mesenteric defect was closed with interrupted 5 − 0 absorbable sutures. The abdomen was irrigated, and the gallbladder specimen was extracted through the umbilical port without extending the incision.

All TL-KPE procedures were performed by a single experienced pediatric laparoscopic surgeon (P.L.), who had previously completed over 30 total laparoscopic choledochal cyst excisions. The first five TL-KPE cases were excluded from analysis to mitigate the effect of the initial learning curve. O-KPE and L-KPE procedures were performed by the same surgeon or by other board-certified pediatric surgeons with comparable experience; all participating surgeons had performed at least 20 prior Kasai portoenterostomies (open or laparoscopic) before the study period.

The key operative steps of total laparoscopic KPE are demonstrated in the attached surgical video.

L-KPE group

The Roux-en-Y jejunal limb was constructed extracorporeally through an extended umbilical incision. Intestinal anastomoses were performed manually, using either interrupted or continuous sutures, and no staplers were used. The remaining procedural steps were identical to those described for the TL-KPE group.

O-KPE group

In the O-KPE group, after diagnostic confirmation via laparoscopic cholangiography, open surgery was performed through a right subcostal incision. The remaining steps were fundamentally similar to those of the L-KPE group. The hepaticojejunostomy was performed using interrupted sutures for both the posterior and anterior walls. All sutures were placed first before being sequentially tied to ensure even, tight, and precise approximation.

Postoperative management

Gastrointestinal decompression and diet

Patients in the O-KPE and L-KPE groups routinely received nasogastric decompression postoperatively, with a liquid diet initiated upon return of bowel function, typically within 3–5 days. In contrast, the TL-KPE group did not undergo routine postoperative nasogastric intubation; a liquid diet was started 24 h after surgery following confirmation of bowel recovery.

Antibiotic and medication regimen

Postoperative adjuvant therapy was standardized across all three groups. All patients received routine intravenous third-generation cephalosporins (e.g., cefoperazone-sulbactam) for infection prophylaxis as part of a 5- to 7-day regimen, followed by long-term oral antibiotic prophylaxis (e.g., cefixime) for 3 to 6 months, a practice widely used despite limited evidence regarding its optimal duration. Choleretic therapy with ursodeoxycholic acid (20–30 mg/kg/d in divided doses) was administered for at least 6 months. A tapering course of corticosteroids (e.g., methylprednisolone or prednisolone, starting at 4 mg/kg/d with gradual reduction over 8–14 weeks) was also given.

Management of cholangitis

In cases of suspected or confirmed cholangitis, antibiotic therapy was promptly escalated to potent intravenous agents and subsequently adjusted according to antimicrobial susceptibility testing. Comprehensive supportive care was provided concurrently.

Observation indicators and definitions

Jaundice clearance

Jaundice clearance was defined as a serum total bilirubin (TBIL) level ≤ 20 µmol/L at 6 months postoperatively, which is a widely used surrogate for successful biliary drainage [7]. Although direct bilirubin and GGT are also markers of cholestasis, total bilirubin was chosen for consistency with major previous studies and because it is the primary determinant of clinical jaundice. Secondary analyses using DBIL < 10 µmol/L or GGT trends did not alter the conclusions (data not shown).

Cholangitis

Cholangitis was diagnosed based on the presence of all the following criteria: (1) postoperative body temperature > 38 °C without other identifiable sources of infection; (2) an increase in direct bilirubin (DBIL) by more than 25% from the previous level; and (3) the occurrence of pale stools.

Early-onset cholangitis was defined as cholangitis occurring within 1 month after surgery.

Frequent cholangitis was defined as three or more episodes within the first 6 months postoperatively.

Follow-up protocol

Patients were scheduled for follow-up assessments at 1, 2, 3, and 6 months postoperatively, and annually thereafter. Evaluations primarily included liver function tests, abdominal ultrasonography, documentation of cholangitis episodes, and survival status. The cutoff date for all follow-up data in this study was June 30, 2025. All patients were successfully followed up via outpatient visits or telephone interviews.

Statistical methods

Continuous variables were compared using one-way ANOVA or the Kruskal-Wallis test, and categorical variables using the χ² or Fisher’s exact test. For outcomes with an overall significant difference (P < 0.05), two pre-specified pairwise comparisons were performed: TL-KPE vs. O-KPE and TL-KPE vs. L-KPE, using independent-samples t-tests or Mann-Whitney U tests. P values from these two contrasts were adjusted with the Bonferroni method, and a corrected P < 0.025 was considered significant. All analyses were conducted using SPSS (version 20.0). No a priori sample size calculation was performed, as this was a pilot-feasibility study with a convenience sample of 15 patients per group based on the study period. Post-hoc power analysis for the secondary outcome of jaundice clearance indicated that with the observed effect size (86.7% in TL-KPE vs. 60% in O-KPE) and α = 0.05, the achieved power was approximately 40%. Therefore, negative findings for secondary endpoints should be interpreted with caution, and the study is underpowered for definitive comparisons.

Results

Patient demographics and preoperative characteristics

A total of 45 patients undergoing Kasai portoenterostomy were included, with 15 patients in each group. No perioperative deaths occurred in any group. As shown in Table 1, there were no significant differences among the three groups in baseline demographic and clinical characteristics, including sex, age at surgery, weight at surgery, or preoperative liver function parameters (ALT, AST, TBIL, DBIL, GGT, TBA; all P > 0.05).

Table 1.

Baseline demographic and clinical characteristics before Kasai

	O-KPE (n = 15)	L-KPE (n = 15)	TL-KPE (n = 15)	F/H	p
Sex (Male/Female)	4/11	6/9	5/10	-	0.622
Age at Kasai (days)	61.4 ± 25.2	60.3 ± 18.4	58.8 ± 22.0	F = 0.02	0.976
Weight at Kasai (kg)	5.1 ± 1.2	5.0 ± 0.9	4.9 ± 1.3	F = 0.08	0.921
ALT (IU/L)	146.5 ± 110.9	150.4 ± 112.2	148.0 ± 105.1	H = 0.01	0.994
AST (IU/L)	143.5 ± 130.1	154.3 ± 119.7	158.2 ± 115.7	H = 0.06	0.968
TBIL (µmol/L)	150.0 ± 45.0	152.3 ± 50.5	148.2 ± 42.8	F = 0.02	0.984
DBIL (µmol/L)	108.1 ± 25.5	105.0 ± 23.0	103.2 ± 24.5	F = 0.10	0.908
GGT (U/L)	738.6 ± 187.7	767.4 ± 153.6	792.4 ± 168.4	F = 0.28	0.755
TBA (µmol/L)	155.4 ± 37.3	148.9 ± 34.2	150.4 ± 35.2	F = 0.12	0.892

Perioperative outcomes

Perioperative outcomes are summarized in Table 2. Operative time did not differ significantly among the groups (P = 0.052), although the TL-KPE group had a numerically slightly shorter mean operative time (210.5 ± 50.6 min) compared to the L-KPE group (218.4 ± 43.4 min). This trend may be attributed to the elimination of steps for bowel exteriorization and repositioning, as well as increased efficiency with the continuous suturing technique; however, this observation requires validation in larger studies. Intraoperative blood loss was significantly lower in the TL-KPE group compared to the O-KPE group (P = 0.016), but not significantly different from the L-KPE group (P = 0.894). The TL-KPE group showed markedly shorter time to bowel recovery and postoperative fasting time (both P < 0.001). Postoperative hospital stay was also significantly shorter in the TL-KPE group compared to both the O-KPE (P = 0.016) and L-KPE (P = 0.034) groups. However, these outcomes should be interpreted with caution, as the TL-KPE group also followed a different postoperative feeding protocol (no nasogastric tube, earlier oral intake) by study design. The incidence of postoperative complications did not differ significantly among groups (P = 0.132).

Table 2.

Surgical and postoperative recovery outcomes in each group

	O-KPE (n = 15)	L-KPE (n = 15)	TL-KPE (n = 15)	F/χ²/H	P	TL-KPE vs. O-KPE	TL-KPE vs. L-KPE
Operation time(min)	181.0 ± 32.5	218.4 ± 43.4	210.5 ± 50.6	F = 3.18	0.052	-	-
Intraoperative blood loss (ml)	10.1 ± 1.8	8.0 ± 3.0	7.2 ± 3.5	F = 4.12	0.023	t = 2.76, P = 0.016	t = 0.13, P = 0.894
Time to bowel recovery (h)	40.3 ± 5.7	41.0 ± 10.1	10.4 ± 2.1	F = 98.85	< 0.001	t = 15.42, P < 0.001	t = 11.32, P < 0.001
Postoperative fasting time (h)	49.3 ± 8.4	52.0 ± 12.3	24.1 ± 7.8	F = 37.70	< 0.001	t = 9.26, P < 0.001	t = 7.85, P < 0.001
Postoperative hospital stay (d)	16.1 ± 2.8	15.8 ± 3.1	13.4 ± 1.9	F = 4.68	0.014	t = 2.76, P = 0.016	t = 2.32, P = 0.034
Postoperative complications (cases)	2	0	0	-	0.132	-	-
Jaundice clearance(cases)	9	10	13	χ²=2.56	0.278	χ²=2.14, P = 0.143	χ²=1.50, P = 0.221
Early cholangitis(cases)	4	5	4	χ²=0.14	0.932	χ²=0.00, P = 1.000	χ²=0.14, P = 0.705
Recurrent cholangitis(cases)	2	2	2	χ²=0.00	1.000	χ²=0.00, P = 1.000	χ²=0.00, P = 1.000

Postoperative outcomes

Jaundice clearance was observed in 9, 10, and 13 patients in the O-KPE, L-KPE, and TL-KPE groups, respectively. Although the TL-KPE group had a numerically higher rate, the intergroup difference was not statistically significant (P = 0.278). Post-hoc power analysis indicated that with the observed effect size (13/15 vs. 9/15) and alpha = 0.05, the study had approximately 40% power to detect a significant difference; thus, the lack of statistical significance may be related to insufficient sample size. Early cholangitis occurred in 4, 5, and 4 cases, respectively (P = 0.932). Recurrent cholangitis was reported in 2 cases in each group (P = 1.000). Details are provided in Table 2.

Discussion

Biliary atresia is a progressive obliterative cholangiopathy that remains a leading cause of liver failure in infants. The Kasai portoenterostomy is the standard initial palliative procedure for type III BA. The introduction of laparoscopic KPE has provided magnified visualization of the porta hepatis, but conventional L-KPE requires bowel exteriorization for Roux-en-Y limb construction, which may partially offset the benefits of minimally invasive surgery [1, 3–5].

Based on our previous experience with totally intracorporeal reconstruction for choledochal cysts [6], this study evaluated a refined total laparoscopic KPE technique. The key modifications include: (1) performing all intestinal steps entirely intracorporeally, avoiding exteriorization; (2) using LigaSure for bowel transection, which provides a secure seal without staplers; and (3) employing a continuous suturing technique with delayed tightening for the portoenterostomy, which may improve visualization and facilitate accurate anastomotic placement. Although these refinements are incremental, they represent a systematic attempt to extend the benefits of full laparoscopy to Kasai portoenterostomy, an area where such data are still limited.

Our results demonstrate significant short-term recovery advantages for TL-KPE, including shorter time to bowel recovery, earlier oral intake, and reduced postoperative hospital stay compared to both O-KPE and L-KPE. This accelerated recovery is likely multifactorial: the fully intracorporeal approach avoids bowel exposure and prolonged mesenteric traction, potentially reducing bowel wall edema and postoperative ileus. In addition, the absence of routine nasogastric decompression and early initiation of enteral nutrition in the TL-KPE group may stimulate gastrointestinal hormone release and bile flow, creating a positive feedback loop that enhances functional recovery [8, 9]. Early feeding also triggers cholecystokinin release and physiological bile secretion, which may help maintain patency of the newly established biliary conduit [9]. The practice of early oral feeding without routine nasogastric decompression following TL-KPE is consistent with our institutional protocol for totally intracorporeal Roux-en-Y anastomosis in pediatric choledochal cyst excision [6]. In that setting, early enteral nutrition has been safely implemented without increased anastomotic complications or vomiting, likely due to reduced bowel manipulation and mesenteric traction. The same rationale underpins the postoperative regimen in the TL-KPE group, although future studies should formally compare feeding protocols to disentangle the effects of the surgical technique from those of the management pathway.

Despite these recovery benefits, no significant differences were observed among groups in jaundice clearance or cholangitis rates within the first 6 months. This finding underscores an important concept: while surgical refinements can improve perioperative recovery, the long-term success of bile drainage and liver preservation may be more heavily influenced by the underlying severity of the disease itself, such as the degree of liver fibrosis and ductal plate malformation at the time of surgery [1]. In other words, there may be a “biological ceiling” beyond which even a technically optimal KPE cannot alter disease trajectory in patients with advanced fibrosis or ductal paucity. Moreover, the modest sample size and short follow-up limit our ability to draw conclusions about long-term outcomes such as native liver survival. The numerically higher jaundice clearance rate in the TL-KPE group, although not statistically significant, might warrant further investigation in adequately powered studies.

Regarding the choice of a hand-sewn anastomosis without staplers, this approach was adopted because appropriately sized linear cutting staplers for infants are not widely available, and their use in the small pediatric abdomen can be technically challenging. Using LigaSure for bowel transection provides temporary sealing of the cut ends, minimizing spillage of intestinal contents while preserving adequate blood supply to the margins; this allows for subsequent hand-sewn anastomosis without interference. Avoiding staplers eliminates associated device costs and may reduce the risk of stapler-related complications (e.g., anastomotic strictures, foreign body reactions) [10, 11], although the present study was not designed to directly assess these potential benefits. All anastomoses were performed with hand-sewn techniques, which proved feasible and effective in this series.

The continuous suturing technique with delayed tightening offers a specific advantage during the critical step of portoenterostomy. By placing all sutures loosely before final approximation, the surgeon maintains an unobstructed view of both the enterotomy and portal plate edges throughout the process. This may enhance the precision of stitch placement, ensure full-thickness bites, and reduce the risk of incorporating extraneous tissue. The technique also simplifies the anastomosis compared to traditional interrupted suturing, potentially shortening the learning curve for this demanding procedure. However, these theoretical advantages require further validation in comparative studies.

Limitations

This study has several limitations that should be acknowledged. First, the non-randomized, preference-based allocation may have introduced selection bias. Although baseline characteristics were balanced, unmeasured confounders—such as guardians’ expectations regarding cosmetic outcomes or perceived invasiveness—could have influenced group assignment. This limitation is inherent to the study design and should be considered when interpreting the results. Second, although no perioperative deaths occurred, the exclusion of such events was not a source of bias; however, the sample size was small (n = 15 per group), resulting in limited statistical power to detect differences in secondary endpoints such as jaundice clearance; post-hoc analysis confirmed the study was underpowered for this outcome. Third, follow-up was limited to 6 months, which precludes assessment of long-term native liver survival—the ultimate goal of Kasai portoenterostomy. Fourth, the diagnosis of cholangitis was based on clinical criteria without routine confirmation by blood culture or liver biopsy, which may have introduced misclassification bias. Fifth, outcome assessors were not blinded to group assignment, potentially introducing detection bias. Sixth, we did not control for the surgeon’s learning curve; all TL-KPE procedures were performed by a single experienced surgeon, but the first five cases were excluded from analysis to minimize this effect. Seventh, the postoperative management protocols differed between the TL-KPE group and the other two groups, with the former receiving no routine nasogastric decompression and earlier enteral feeding by design. While we attribute the accelerated recovery to reduced bowel trauma inherent to the total laparoscopic technique, it is possible that the more liberal feeding protocol itself contributed to shorter fasting time and hospital stay. Thus, the observed benefits cannot be solely ascribed to the surgical approach. A randomized trial with standardized postoperative care across arms would be required to disentangle these effects. Eighth, objective markers of bowel wall edema or inflammation (e.g., ultrasound measurement, cytokine levels) were not obtained, limiting mechanistic insights. Ninth, longitudinal liver function parameters (e.g., serial bilirubin, GGT) and longer-term outcomes such as transplant-free survival or need for re-intervention were not analyzed due to the short observation period and sample size constraints. Future studies with extended follow-up should include time-to-event analyses to better capture the durability of biliary drainage.

Despite these limitations, the study provides preliminary evidence that TL-KPE is safe and may enhance early recovery. Future prospective, multicenter studies with larger cohorts, longer follow-up, and standardized outcome definitions are needed to confirm these findings and to assess whether improved short-term recovery translates into superior long-term native liver survival. Incorporating preoperative biomarkers of fibrosis or advanced imaging could help stratify patients and personalize surgical strategies.

Conclusion

TL-KPE is a safe and technically feasible approach for Type III biliary atresia. Compared with open and conventional laparoscopic techniques, TL-KPE significantly accelerates short-term postoperative recovery, as reflected by shorter time to bowel recovery, earlier oral intake, and reduced hospital stay, without increasing operative time or perioperative complications. Jaundice clearance and cholangitis rates did not differ statistically among groups in this small cohort, but the observed recovery advantages support the role of TL-KPE as a valuable minimally invasive option. However, these short-term recovery advantages should be interpreted with caution due to the non-randomized design, small sample size, and differential postoperative management protocols between groups. Larger multicenter studies with extended follow-up are warranted to validate these early benefits and to evaluate the potential long-term impact on native liver survival.

Abbreviations

BA

Biliary atresia

KPE

Kasai portoenterostomy

O-KPE

Open Kasai portoenterostomy

L-KPE

Laparoscopic Kasai portoenterostomy (with extracorporeal anastomosis)

TL-KPE

Total laparoscopic Kasai portoenterostomy (with intracorporeal anastomosis)

TBIL

Total bilirubin

DBIL

Direct bilirubin

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

GGT

Gamma-glutamyl transferase

TBA

Total bile acids

PDS

Polydioxanone suture

Authors’ contributions

JC: Conceptualization, Data curation, Formal analysis, Writing – original draft. CM: Data curation, Investigation, Methodology. QY: Investigation, Validation. PL: Conceptualization, Funding acquisition, Supervision, Writing – review & editing. All authors read and approved the final manuscript.

Funding

This work was supported in part by the National Natural Science Foundation of China (Grant no. 81800450) and the Key Research and Development Plan of Shaanxi Province (Grant nos. 2025SF-YBXM-430, 2020SF-059). The funders had no role in study design, data collection, analysis, decision to publish, or preparation of the manuscript.

Data availability

Data are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Second Affiliated Hospital of Xi’an Jiaotong University (Approval No. 2023 − 153). Written informed consent was obtained from the guardians of all individual participants included in the study.

Consent for publication

Not applicable. All images are fully anonymized with no personally identifiable information.

Competing interests

The authors declare no competing interests.