Introduction

A choledochal cyst (CC) is a congenital hepatic bile duct anomaly related to a pancreaticobiliary maljunction [1,2,3]. The standard treatment for CC is total excision of the cyst and resection of the extrahepatic bile duct with bilioenteric anastomosis separating the mutual countercurrents of bile and pancreatic juice [4].

Most frequent late postoperative complications among patients with CC include cholangitis, hepatobiliary malignancies, and postoperative anastomotic strictures (PAS) [1]. PAS is a well-known complication after correcting CC. Although the exact cause of PAS is unknown, various risk factors such as Todani classification type IV-A [5], hepaticoduodenostomy (HD) [6], laparoscopic surgery [7], and narrow anastomosis [8] have been known to be associated with PAS. As far as we know, there is no report with a cumulative analysis of these risk factors of PAS. This systematic review and meta-analysis aimed to investigate the risk factors of PAS following surgical correction of CC in children.

Material and methods

Study selection

A systematic literature search was performed in the PubMed, CINAHL, Cochrane Library, and Ichushi-web electronic database using the combinations of the following terms “Congenital biliary dilatation”, “choledochal cyst”, “Stenosis”, “Stricture”, and “Complication” for studies published between 1973 and 2022. The relevant cohorts of PAS were systematically searched for clinical presentation and outcomes. Reference lists of relevant articles were manually searched for further cohorts. Duplicates were deleted. The resulting publications were reviewed in detail for clinical features and type of operation performed.

Data extraction

The relevant articles were reviewed by title, keywords, and abstracts by the authors (RT, HN, and TD) and a full-text assessment of selected articles was performed.

Inclusion criteria and exclusion criteria

The inclusion criteria were the study that reported pediatric patients with PAS after correcting CC and the study that was written in English or Japanese. The exclusion criteria were the study that reported adult patients with PAS or the study did not mention the age of the patients clearly.

Statistical analysis

A systematic review and meta-analysis were conducted based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [9]. Pooled odds ratios (OR) were calculated for dichotomous variables using the Mantel–Haenszel method, and pooled mean differences were measured for continuous variables using the inverse variance method in meta-analysis. The odds ratio represents the odds of an adverse event occurring in the treatments. The confidence interval (CI) was established at 95%, a p value of less than 0.05 was considered statistically significant, and an odds ratio of less than 1 favored the treatment group.

Statistical heterogeneity was assessed using I2. A fixed effects model was used if I2 < 50%, and a random effects model was used if I2 ≥ 50%. Statistical analysis was conducted by Review Manager 5.4 (Cochrane Collaboration).

Results

The initial search yielded a total of 795 publications, of which 772 were identified by electronic database searching and 23 from cross-referencing (Fig. 1). After confirmation of duplicate listed articles, 625 titles, keywords, and abstracts were screened. Of these, 403 non-relevant studies were excluded. The remaining 223 publications were assessed in full text for eligibility and 152 articles were excluded because they did not address any of the selection criteria. In total, data from 70 studies [1, 3, 6,7,8, 10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74] (published between 1976 and 2022) met the defined inclusion criteria and were included in the cumulative analysis (Table 1). All 70 included studies are retrospective studies.

Fig. 1
figure 1

This is an information flow diagram, demonstrating the process of selection and exclusion of articles from the literature search for the purposes of systematic review and meta-analysis

Full size image
Table 1 Study Characteristics
Full size table

Characteristics of patients

The total number of PAS was 206, with an incidence of 2.1%. The mean age at the radical surgery for the patient without PAS was 43.66 months old, and 84.13 months old for the patient with PAS. Statistical evaluation was not possible because of the small number of cases. Four studies reported the diameter of the anastomosis at the primary surgery. The mean diameter was 12.5 mm.

Todani classification

Fifteen studies reported the Todani classification with 51 of 206 PAS patients. There were 21 cases with type I and 30 cases with Type IV-A. In particular, six studies compared the patient with PAS and without PAS regarding the Todani classification. The incidences of PAS were 2.0% in Type I and 10.1% in Type IV-A, respectively. Meta-analysis did not show a significant difference between the 2 groups [OR 0.09, 95% CI (0.02 0.40), p = 0.001] (Fig. 2).

Fig. 2
figure 2

Forest plots on the incidence of PAS between type I and type IV-A

Full size image

Hepaticojejunostomy (HJ) versus HD

Fourteen studies reported a comparison between HJ and HD, including 1333 patients (824 underwent HJ and 509 underwent HD). The total number of PAS was 30 (22 underwent HJ and 8 underwent HD). Meta-analysis did not show a significant difference between the two groups [OR 1.02, 95% CI (0.62 2.60), p = 0.51] (Fig. 3).

Fig. 3
figure 3

Forest plots on the incidence of PAS between hepaticojejunostomy and hepaticoduodenostomy

Full size image

Laparoscopic versus open

Nine studies reported a comparison between laparoscopic surgery and open surgery, including 1534 patients (761 underwent laparoscopic surgery and 773 underwent open surgery). The total number of PAS was 33 (13 underwent laparoscopic surgery and 20 underwent open surgery). Meta-analysis did not show a significant difference between the two groups [OR 1.43, 95% CI (0.34 5.96), p = 0.63] (Fig. 4).

Fig. 4
figure 4

Forest plots on incidence of PAS between laparoscopic surgery and open surgery

Full size image

Treatments

Fifty-one studies reported the treatment of PAS. Thirteen patients were treated with endoscopic balloon dilatation, and 136 patients were treated with surgical revision of anastomosis.

Discussion

The present study was the first systematic review and meta-analysis to focus on the exact cause of PAS following the correction of CC. The cumulative incidence of PAS was 2.1%. Several reports suggested the cause of PAS such as Todani classification type IV-A, laparoscopic surgery, HD, and narrow anastomosis. In the present study, pooled data analysis has revealed that Todani classification Type IV-A was a significantly higher risk of PAS.

Todani classification type IV-A was the dilatation in the common bile duct and intrahepatic bile duct. Kim et al. [5] reported the risk factors of PAS in the adult patient and suggested that PAS could be influenced by type IV-A. Miyano et al. [75] reported that the pathological findings of the cystic type of CC showed atrophy of the lamina propria and prominent inflammatory cell infiltration. Kim et al. [5] also investigated the pathological findings of the bile duct and the cyst excised in the patient with PAS in the adult. They showed no significant differences in the thickness of cyst wall, fibrosis, loss of smooth muscle, and loss of mucosa between the patient with PAS and not [5]. On the other hand, they reported that there was a significant difference in the infiltration of inflammatory cells in the patient with PAS in adults [5]. In our study, although incidence of PAS in type IV-A was statistically higher than type I, it was unable to evaluate the inflammation of the cyst wall because only a few papers investigated the cyst wall in detail pathologically (e.g., pathological findings, inflammatory cytokines, and characterization of T-cell populations).

Yamataka et al. [25] reported that there were no PAS in the patients with CC whose ages were 5 years old or younger. It has been reported that inflammation of the cyst wall was milder in the younger patient than 10 years old and more severe in the older patient with CC, indicating that histological damage to the common hepatic duct was more severe in older pediatric patients and adults [25]. In the present study, there was a trend that the age at the radical surgery in the patients with PAS was higher than the patient without PAS. However, a statistical analysis was not appropriate due to the small number of samples.

It was thought that HD was more prone to ascending cholangitis and thus to PAS [37, 76], but Todani et al. [77] found no difference in the frequency of ascending cholangitis between HD and HJ, and some surgeons [2, 69] also found no difference in the frequency of PAS between HD and HJ. Similarly, in a systematic review by Narayanan et al. [78] comparing HD and HJ, there was no significant difference in the frequency of PAS. In the present meta-analysis, no significant difference between HJ and HD.

The laparoscopic procedure has been getting adopted as the initial treatment for CC world wide since Farello et al. [79] reported the laparoscopic procedure for CC in 1995. Some surgeons are concerned about the disadvantages of the laparoscopic procedures for pediatric patients with CC due to small operative space, risk of injury to vital structures, and low quality of the anastomosis. Xie et al. [7] reported that the rate of PAS in the laparoscopic procedures was higher than in open procedures, but there was no significant difference, while Diao et al. [27] concluded that laparoscopic procedures were safe and effective, with lower morbidities of not only PAS, but also bile lake, intrahepatic stone formation, intrahepatic reflux, cholangitis, pancreatic leak, pancreatitis, and intestinal obstructions. Our meta-analysis showed no significant difference between laparoscopic procedures and open procedures.

Todani et al. [6] reported the creation of a wide anastomotic stoma was essential to prevent ascending cholangitis resulting in PAS. Kubota et al. [73] reported a case of PAS with an anastomotic diameter of 12 mm, which seemed to be sufficient, and in their case, intraoperative cholangiography showed stenosis at the bifurcation of the right and left hepatic ducts, where cholangitis occurred, which may have spread to the anastomosis and resulted in PAS. In our study, the mean diameter of anastomoses was 12.5 mm, but there were few articles reporting the detail of cholangitis and the information on intrahepatic bile duct stenosis.

There were some limitations in this study. First, there were no randomized controlled trials that met our criteria. Second, there were variations in surgical techniques among facilities, and unified procedures could not be used to evaluate the results.

Based on this study, we plan to conduct further investigation focusing on (1) therapeutic strategies for PAS, (2) the pathological evaluation of the cyst wall in patients with PAS, and (3) looking for the adequate timing of the radical surgery for patients with CC.

In conclusion, this study suggests that close careful follow-up is essential in patients with type IV-A of CC who underwent excision surgery, considering the possibility of PAS. Since the age at the radical surgery would be associated with the inflammation of the cyst wall, further studies must be required in the future.