Characteristics, management and outcomes of choledochal malformations in Finnish adult patients

Abstract Aim: We aimed to assess management, comorbidities, biliary histopathology and outcomes of choledochal malformations (CM) in adults. Methods: All adult CM patients managed during 1987–2021 at the Helsinki University Hospital were retrospectively reviewed. Results: Of the 39 patients (females 79%) identified, 19 (49%) underwent surgery, 5 (13%) therapeutic ERCP with papillotomy for type 3 CM and 15 (38%) were managed conservatively. Twenty-five (66%) patients had ≥1 comorbidity, and five (13%) patients had associated inflammatory bowel disease (IBD); ulcerative colitis (n = 3) and Crohn’s disease (n = 2). Most surgically treated patients underwent hepaticojejunostomy (n = 13) or pancreaticoduodenectomy (n = 3) and they were younger at presentation than other patients [29 (21 – 48) vs. 59 (41 – 71), p = 0.001]. Three patients (23%) suffered from pancreatitis following diagnostic ERCP. Early surgical complications occurred in six (32%) patients, including one anastomotic leakage. Biliary epithelial dysplasia was found in three (15%) patients in resected specimens, while no neoplastic changes were observed. Nine patients (47%) had long-term postoperative complications, cholangitis (n = 5) and anastomotic stricture (n = 3) being the most common ones. No evidence of malignant change was observed after median follow-up of 2.3 (0.60 − 4.5) years included all patient cohorts. Conservatively managed patients remained symptomless with unprogressive biliary tract imaging findings and normal liver biochemistry values during follow-up. Conclusions Nearly half of operated patients developed long-term postoperative complications. A novel association between CMs and IBD was observed. Although no hepatobiliary malignancies regardless of treatment modality were encountered, the number of patients and length of follow-up remained limited.

Choledochal malformations (CM) are congenital disorders characterized as abnormal cystic or fusiform dilatation of the biliary tract occurring in the Caucasian population in 1 in every 38,000 -59,000 births [1][2][3][4][5] (Figure 1).Most CMs are diagnosed and thus treated, during childhood due to associated morbidity typically including abdominal pain, cholestasis and pancreatitis, and the possibility of developing hepatobiliary malignancies, especially later in life [1,6,7].
The gold standard treatment for type 1 and 4 CMs is a surgical resection of the extrahepatic bile ducts and gallbladder with hepaticojejunostomy (HJ) to prevent the adverse effects of untreated CMs [8][9][10].Biliary malignancy risk is low in type 3 CMs, which should preferably be treated by sphincterotomy in ERCP [11].However, management of CMs in an elderly patient population with frequent comorbidities may be more ambiguous, although surgical treatment is often advocated due to the overall malignancy risk of 11% and the increasing malignancy rate with each passing decade [10,[12][13][14].Relatively high shortand long-term surgical complication rates, affecting 13-50% and 25-35% of the adult patients respectively, call for careful treatment planning with consideration of the degree of comorbidity, life expectancy versus the risk of malignancy and patients own informed opinion [10,[12][13][14][15][16][17][18][19].In selected elderly patients or in the presence of significant comorbidities, an endoscopic or expectant approach may be the safest option.
So far, comprehensive information on management and comorbidities in adults with CM is scarce, as most reports have concentrated on surgical outcomes.Our aim was to assess the characteristics, comorbidities, management and outcomes of CMs in adults treated at our referral hospital to compare and improve different management options.

Ethics
The Helsinki University Hospital Ethics Committee and the Institutional Review Board approved this study.

Patients and study design
This was a single-center retrospective follow-up study consisting of all adult patients (age ≥ 18 years at the time of diagnosis) diagnosed with a CM at the Helsinki University Hospital during 1987 and 2021.In Finland, management of CMs, for both adults and children, is centralized in Helsinki University Hospital, which is national referral center for hepatobiliary surgery.Eligible patients were identified through institutional records and ICD codes.Hospital records were reviewed for baseline patient characteristics, comorbidities, surgeries, endoscopic retrograde cholangiopancreatography (ERCP) procedures, magnetic resonance cholangiopancreatography (MRCP), computed tomography (CT), ultrasound (US) imaging results, liver biochemistry, histopathological and cytological findings of all liver, bile duct and gallbladder specimens obtained at surgery or during ERCP, and follow-up data.Additionally, in patients with associated inflammatory bowel disease (IBD), endoscopy reports and biopsy findings were reviewed.
To compare proportions of adult and pediatric CM patients, the number of pediatric patients (age < 18 years at the time of diagnosis) diagnosed during the same period was also recorded.

CM Management
Patients with symptomatic CMs were proposed surgery.Conservative approach was proposed to elderly patients, who were asymptomatic with normal liver biochemistry and no suspicion of malignancy during follow-up.The decision was made after informing patients about the risks related to both operative and conservative approaches.Type 3 CMs were managed by endoscopic sphincterotomy.Aim of the surgery was the complete removal of CM.

Definitions
The CMs were categorized according to the King's College Hospital modification of the original Todani classification based on radiological imaging findings and operative notes [1] (Figure 1).The early postoperative complications were classified according to the Clavien-Dindo classification [20].Comorbidities were scored according to Charlson Comorbidity Index (CCI) [21,22].Post-ERCP complications were rated according to the Cotton Criteria [23].

Statistics
All analyses were performed using SPSS version 26.Data are given as median values with interquartile ranges (IQR) or frequencies and percentages.Mann Whitney U test and Fisher exact test were used for statistical analyses.The level of significance was set at p < 0.05.

Baseline characteristics and incidence
In total, 39 adult CM patients were identified (Table 1).Median age at presentation was 40 years, and the most frequent presenting symptom was abdominal pain.Patients with type 4 CM (33%) were diagnosed at a younger age [28 (21 -44) years vs. 51 (39 -70), p = 0.001], whereas patients with type 3 CM presented at an older age [73 (48 -76) years vs 41 (28 -61), p = 0.024].The proportion of type 1 fusiform CMs was higher in males than females [n = 4/8 (50%) vs n = 3/31 (9.7%), p = 0.022].Notably, 10 patients (26%) had previously undergone cholecystectomy due to gallstone disease.Overall, 37 (95%) patients had undergone diagnostic MRCP and/or ERCP to assess the type of CM including pancreaticobiliary junction.Of them, a diagnostic ERCP was performed in 13 (33%) patients and combined with a sphincterotomy in two of them to treat cholestasis at presentation prior to definitive CM treatment.Eight (62%) of these patients underwent operative treatment for CM subsequently.Three patients (23%) suffered from pancreatitis following diagnostic ERCP, other complications were not reported.The post-ERCP pancreatitis episodes were rated as moderate in one patient and severe in two patients according to the Cotton criteria [22].One of the latter patients died of prostate cancer a few months later and was excluded from further analysis.
When patients were categorized into three groups according to the year of presentation (1987-1998, 1999-2010, 2011-2021), an increase in the number of CMs was observed (n = 2, n = 8 and n = 29, respectively).During the same period, a total of 62 pediatric CM patients were diagnosed in Finland.Thus, 39% of all CM patients were adults at the time of diagnosis.

Therapeutic interventions and complications
In total, 19 (49%) patients underwent surgery, 5 (13%) therapeutic ERCP with papillotomy for type 3 CM, and 15 (38%) were treated conservatively (Table 1).Most operatively treated patients were symptomatic, however, three (13%) underwent surgery for incidentally detected CMs.Most operations (74%) were standard HJs.In four patients with extensive pancreatic or hepatic involvement of CM, complete cyst removal required extension of HJ to cholangiojejunostomy (n = 1, type 1 CM) or pancreaticoduodenectomy (n = 3, type 4 CM), which was combined with resection of liver segments 2-4 in one.One patient underwent resection of incidental type 2 CM during cholecystectomy.Surgically treated patients were younger than others (Table 2).
Six patients (32%) had early surgical complications (Table 3).One patient underwent re-operation due to volvulus, one patient with HJ leakage was treated with drainage and antibiotics, and complications in the remaining four patients were treated conservatively.Age, CM subtype, preoperative symptoms and operation type were unrelated to the occurrence of early clinical complications.

Conservative management
Altogether 14 patients (37%) were managed expectantly.Out of them, four asymptomatic patients refused surgery; nine asymptomatic patients were recommended a conservative approach due to normal liver biochemistry values and no progression of biliary tract imaging findings.One patient with a concomitant pancreatic neuroendocrine tumor is listed for HJ following pancreatic resection.During the median follow-up of 3.2 (0.58 − 5.9) years, all patients remained symptom-free, MRCP imaging showed no progression of biliary tract findings and liver biochemistry values were normal.

Bile duct histopathology
Overall, histological samples were obtained from 24 (64%) patients, including 20 (83%) patients who underwent surgical treatment or therapeutic ERCP and papillotomy.The remaining four patients had samples obtained during a diagnostic ERCP or previous cholecystectomy (Supplemental Table 1).Available histological samples included removed gallbladder in 18 patients, resected CM in 16 patients, liver biopsy in 4 patients and ERCP brush cytology in 9 patients.Abnormal histopathological findings were found in 14 (74%) surgically treated patients and all patients that underwent ERCP and papillotomy as a therapeutic treatment.Inflammation was the most common finding, while dysplasia of either the extrahepatic biliary tree (n = 2) or the gallbladder (n = 1) was found in three (15%) patients aged 30.2 (21.8 − 55.3) years at the time of surgery.Dysplasia of the gallbladder was found during a prior cholecystectomy for gallstones.The patient had a type 1f CM and did not undergo further surgical treatment.The two patients with dysplasia of the extrahepatic bile ducts underwent surgery for type 4 CM.No neoplastic changes were encountered.Abnormal histological and brush cytology samples were more frequent in females than males (87% n = 13/15 vs 25% n = 3/4, p = 0.014 and 100% n = 7/7 vs 0% n = 0/2, p = 0.028), and less frequent in patients with a type 1 fusiform CM compared to other patients (25% n = 1/4 vs 86% n = 12/14, p = 0.028 and 0% n = 0/2 vs 100% n = 7/7, p = 0.018).

Long-term complications
The median follow-up time for all patients was 2.3 (0.60 − 4.5) years (Table 2).No hepatobiliary malignancies were encountered during the follow-up.Five patients (13%) died.Apart from the patient who died of ERCP pancreatitis, other deaths were unrelated to CM. Postoperative long-term complications occurred in 9 (47%) patients, cholangitis (n = 5) and anastomotic stricture (n = 3) occurring most frequently (Table 3).Cholangitis episodes were treated with intravenous antibiotics, while apart from one patient, the patients with cholelithiasis (n = 1) and/or anastomotic stricture (n = 2) were treated successfully with double-balloon ERCP or PTC.One patient with both cholelithiasis and anastomotic stricture underwent endoscopic anastomotic stricture dilatations twice followed by surgical revision of HJ 15 years after the initial surgery.One patient underwent removal of necrotic Roux loop and re-HJ due to vascular complications occurring during surgery for a gastrointestinal stromal tumor 2.5 years after initial HJ.One patient developed portal vein thrombosis 2.2 years after HJ at the age of 35 years.No prothrombotic disorders or other risk factors were found.Portal flow was successfully restored with anticoagulant therapy.
The follow-up period was longer in patients with long-term complications [2.8 (2.1 − 6.8) years vs 1.0 (0.40 − 3.0) years, p = 0.027] than in others.Although all patients who underwent pancreaticoduodenectomy had long-term complications, including cholangitis, anastomotic stricture and exocrine pancreatic insufficiency, no statistically significant association between the type of surgery and complications were observed.The occurrence of long-term complications was also unrelated to CM subtype, comorbidities, and presentation or operation age.
Liver biochemistry results were assessed after a median follow-up of 1.5 (0.40 − 4.5) years at the age of 43 (27 -68) years.Alanine aminotransferase, alkaline phosphatase and/ or glutamyl transferase were modestly elevated in 5 (21%) patients, while bilirubin levels were within the normal range.Tumor markers CEA, Ca19-9 and AFP were assessed overall from 24 patients (62%) during follow-up and were all within normal range.

Discussion
In the present study, the comprehensively collected incidence of long-term complications was relatively high at 47%.Accordingly, in previous reports, the long-term surgical complication rate of adult CMs ranges between 25% and 35% [12,15,16,18,24].Complication rates in pediatric patients are typically somewhat lower as seen in a previous report from the Nordic countries, where 21% of children were affected with long-term complications [25].The higher complication rate in adults associated with more advanced biliary  pathology necessitates more complex surgical procedures including resection of the liver and pancreas for complete removal of CM instead of standard HJ in several patients [18].In this regard, diagnosis and treatment of CMs early in life seem beneficial.In Finland, over 60% of CMs were detected and managed during childhood in the same time period.However, the rate of cholangitis, the most common long-term complication in both children and adults, was comparable (17% of children versus 26% adults) [25].
Although no hepatobiliary malignancies were found either prior to CM diagnosis or perioperatively, histological samples revealed dysplasia of either the gallbladder or extrahepatic ducts in three (14%) patients at a relatively young age.During the follow-up none of the patients developed hepatobiliary malignancies regardless of whether they underwent surgery or conservative monitoring.However, the risk of developing cancer even after HJ is well recorded, calling for a standardized postoperative follow-up, especially in patients with a type 1 or 4 CM [7,13,[26][27][28].
The proportion of patients who underwent surgical treatment in this study was relatively low, when considering the established malignancy rate affecting 10-38% of unoperated CM patients [13,14,26,27,29].The only statistically significant difference found between the operated and unoperated patients was that conservatively managed patients were diagnosed at an older age.Other factors supporting the decision to choose conservative monitoring included patients being asymptomatic with normal liver biochemistry and repeated MRCP and/or ERCP imaging studies showing no progression of the hepatobiliary changes.Nevertheless, monitoring of conservatively managed patients should be also considered as the malignancy risk increases with advancing age [14,27,30,31].
We also studied comorbidities to better understand if they would influence surgical outcomes or management options, as little data on the subject has been reported previously.We did not find comorbidities to significantly associate with surgical outcomes or the chosen management modality, although comorbidities occurred more frequently among conservatively treated patients.However, this difference remained statistically insignificant, possibly due to a small number of patients.
Interestingly, our findings also suggest IBD to be unusually prevalent in patients with CMs, especially type 1fusiform CM.Prevalence of IBD among CM patients was 13% as opposed to 1% in normal Finnish adult population [32].Whilst the association between CM and IBD has not been reported in previous studies, the significance of this finding remains unclear but clearly requires further confirmative research and may bear pathophysiological significance.Type 1 fusiform CMs also differed from the usual presentation in that most patients with type 1 fusiform CM in our study were men instead of the typically seen higher female-to-male ratio.In our previous study in pediatric CM patients type 1fusiform CM also related to the presence of associated disorders [25].
The current follow-up practice included outpatient visits roughly every 6 months for 2 years postoperatively, after which the frequency is determined individually, usually between every 1-5 years.Six patients were referred to their local central hospital for follow-up.However, in one case, routine follow-up was stopped after 6 months following HJ as the patient was asymptomatic with normal liver biochemistry and normal postoperative MRCP findings, highlighting the importance of a standardized approach to follow-up.The conservatively managed patients underwent infrequent follow-ups every 1-5 years with MRCP and/or ERCP with accompanying liver biochemistry measurements.While none of the patients that were managed conservatively became symptomatic or showed disease progression during the follow-up, no generally accepted follow-up guidelines for these patients are available.
Generally accepted surgical indications for asymptomatic adult patients with CM are lacking.Although most authors advocate operative treatment, individualized assessment is crucial when planning management options, taking into consideration the relatively high complication rate of 47% against the malignancy rate of 0%, as seen in our study.Limitations of our study included the retrospective design with a relatively limited number of patients and follow-up.Conclusive conclusions on the optimal treatment strategy, balancing the complication rate with malignancy risk, requires repeated assessments in years to come.

Table 2 .
comparison of different management modalities.