Presentation and treatment of biliary atresia Witt, Mauri
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Witt, M. (2018). Presentation and treatment of biliary atresia. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Prognosis of biliary atresia after initial
presentation with vitamin K deficiency bleeding
Mauri Witt, Jan. B.F. Hulscher, Janneke L.M. Bruggink, Ruben H.J. de Kleine, Henkjan J. Verkade, Also, on behalf of NeSBAR Other NeSBAR members: J.H. Escher, L.W.E. van Heurn, R.H.J. Houwen, A. Kindermann, B. Koot, C. Sloots, I. de Blaauw, G. Damen and D.C. van der ee
ABSTRACT
Objective
Vitamin K deficiency bleeding (VKDB) in infants can be the first symptom of cholestasis, including biliary atresia. VKDB is associated with high morbidity and mortality. We analysed whether the presence of VKDB at initial presentation of patients with biliary atresia affects their long-term prognosis.
Design
We retrieved data from the Netherlands Study group on Biliary Atresia Registry (NeSBAR), containing all patients with biliary atresia in the Netherlands between 1987 and 2017. Clearance of jaundice, native liver survival, liver transplantation (LTx) and pre-transplant mortality were compared between patients with and without VKDB at initial presentation.
Results
VKDB at initial presentation occurred in 57/281 (20%) patients. Clearance of jaundice was achieved in 36% of patients with VKDB and in 35% of patients without VKDB. Median native liver survival of patients with VKDB was 17 months versus 15 months for patients without VKDB, with a median follow-up of 70 and 95 months respectively. Relatively more patients with VKDB underwent LTx (70%, 40/57), compared with patients without VKDB (54%, 121/224). There was no significant difference in pre-transplant mortality rate between the two groups.
Conclusion
VKDB at initial presentation of BA does not grossly affect survival but is associated with a significantly higher rate of liver transplantation during follow-up.
4
INTRODUCTION
Vitamin K deficiency can cause severe bleeding in infants. Vitamin K deficiency bleeding (VKDB) in infants has been associated with high morbidity and mortality, especially when the bleeding is located intracranially.1-3 Especially breastfed
infants are at risk for VKDB, due to insufficient amounts of vitamin K in breast milk. Vitamin K is a fat-soluble vitamin and its absorption is dependent on intestinal solubilisation by bile acids. Breastfed infants with unrecognised cholestasis are therefore at highest risk for VKDB, because of the combination of low amounts of vitamin K present in breast milk and decreased intestinal availability of bile acids for its solubilisation and absorption.4, 5 Whereas children after the diagnosis of a
cholestatic liver disease depend on proper treatment with high dosages vitamin K, children with still unrecognised cholestasis (i.e. before the diagnosis) merely depend on the local or national vitamin K prophylactic regimen.6
To prevent VKDB in breastfed infants in general, different prophylactic regimens are used worldwide.7 The Netherlands has had several oral regimens of vitamin K
prophylaxis, consisting of a single oral dose of 1 mg vitamin K at birth followed by a daily oral supplementation of vitamin K in a dose of 25 μg (until 2011) or 150 μg (since 2011) between week 2 and 13. We previously reported, however, that each of these two regimens failed to prevent VKDB in breastfed children with unrecognised cholestasis.8, 9 Under the Dutch regimen of 25 μg of Vitamin K, VKDB occurred in
82% of breastfed BA patients and in 40% of patients this consisted of intracranial haemorrhages. Under the current regimen of 150 μg, VKDB has still been present at the initial presentation in 82% of all breastfed patients with BA, and in 27% this concerned intracranial haemorrhages.8 Based on these data the national vitamin K
prophylaxis regimen is expected to change towards intramuscular administration at birth.
As stated above, we have demonstrated that the high incidence of VKDB as presenting symptom of biliary atresia is due to a failing prophylactic regime in the Netherlands.8, 9 It is unknown whether BA patients who initially presented with
any form of VKDB or, more specifically, with VKDB-intracranial haemorrhage have a different prognosis compared to other BA patients. Although unfortunate, the Dutch situation of vitamin K prophylactic regimens offer the possibility to address this question, through analysis of our national registry. In the present study we analysed the unique cohort in which VKDB was an initial presenting symptom of BA and investigated whether this VKDB affected their long term prognosis with respect to response to Kasai portoenterostomy (KPE). We compared the postoperative clearance of jaundice, survival with native liver, age and percentage of liver transplantation (LTx) and mortality. Our data indicate that VKDB at initial presentation of BA does not grossly affect overall mortality but is associated with a significantly higher rate of liver transplantation during follow-up.
METHODS
This study was performed according to the guidelines of the Medical Ethical Committee of the University Medical Center Groningen. Upon specific request, the Medical Ethical Committee confirmed that the study concerned an anonymised, retrospective analysis of patient data, for which ethical approval was not required according to the Dutch laws.
All data were obtained from the Netherlands Study group on Biliary Atresia Registry (NeSBAR). As described previously,8, 10-12 NeSBAR is based on a joint effort
of the Dutch Society for Paediatrics - Section Gastroenterology, Hepatology and Nutrition, and the Dutch Society for Paediatric Surgeons. Patient data of all BA patients treated in the Netherlands after January 1987 have been registered in this database. The diagnosis of BA was confirmed by operative findings and/or pathology results. The collected data include gestational age, birth weight, sex, associated anomalies, date of first presentation, bleeding at first presentation, results of diagnostic procedures, laboratory values (e.g. liver synthesis, coagulation), date of KPE, post-operative treatment such as corticosteroids and antibiotics, clearance of jaundice, date of LTX, follow-up and outcome. We included all patients registered in the NeSBAR between March 1987 and January 2017. Patients who did not undergo KPE were analysed separately. Conform previous studies VKDB was defined in the present study as bruising, bleeding or intracranial haemorrhage in infants younger than six months, not due to other coagulopathies, in combination with normalisation of the coagulopathy (partial thromboplastin time or activated partial thromboplastin time) after administration of vitamin K.5, 8, 9
In order to study the predictive value of VKDB at initial presentation for the severity of the liver disease and the development of end-stage liver failure we analysed the following outcome parameters in patients with and without VKDB; the presence of cirrhosis in liver biopsy before or during KPE, clearance of jaundice after KPE (defined as a total serum bilirubin <20 µmol/L within six months after surgery), LTx rate, native liver survival and pre-transplant mortality. Indications for LTx were based on the previously described classification by Sundaram et al.,13 e.g.
irreversible jaundice, failure to thrive, bacterial cholangitis, complications of portal hypertension (variceal bleeding, ascites or spontaneous bacterial peritonitis), severe pruritus, hepatopulmonary syndrome, hepatorenal syndrome and hepatic malignancy.
4
To determine statistical differences in patient characteristics between patients with and without VKDB we used a Fisher’s exact test or Chi-square test for dichotomous variables, a Student T test for continuous variables with a normal distribution and a Mann-Whitney U test for continuous variables with a non-normal distribution. Kaplan Meier survival analysis was used for native liver survival. A
p-value <.05 was considered statistically significant. All statistical analyses were
performed with SPSS (version 23.0; IBM Corp, Armonk, NY). RESULTS
Between March 1, 1987 and January 1, 2017, 301 patients with BA have been treated in the Netherlands and included in the NeSBAR. Five patients (2%) were lost to follow-up and excluded from analysis. Of the remaining 296 patients, 281 patients (95%) underwent KPE en were included for further analyses. Fifteen patients did not undergo KPE, 10 patients underwent primary LTx, 5 patients died before KPE or LTx. These patients were analysed separately.
In total 57 (20%) patients presented with VKDB at initial presentation. Twenty-three (8%) patients presented with intracranial haemorrhage, 17 (6%) with gastrointestinal bleeding, 8 (3%) with umbilical bleeding, 31 (11%) with skin bleeding, 15 (5%) with prolonged bleeding after vein puncture. In 27 (10%) patients there were multiple bleeding sites.
Patient characteristics are described in Table 1. One (2%) of the patients who presented with VKDB was born before a gestational age of 37 weeks versus 27 (13%) of the patients without VKDB (p=0.02). Mean gestational age was 39.6 (+ 1.9) versus 38.5 (+2.6) weeks for the patients with or without VKDB, respectively (p<0.001). Of the patients with VKDB 95% were breastfed, versus 14% of patients without VKDB (p<0.001), confirming the previously reported increased risk on VKDB in breastfed infants.4,9,15 Laboratory values before KPE surgery did not
significantly differ between the two groups. Liver biopsies obtained before or at KPE showed cirrhosis in 6/37 (16%) patients with VKDB and in 11/130 (8%) patients without VKDB (p=0.17).
Table 1: Patients characteristics. The group is divided into patients who presented with VKDB versus patients without VKDB at initial presentation (p value for comparison between these two groups). Values are expressed as n (%), mean ± SD, or median [range].
VKDB
N= 57 No VKDBN= 224 P value Male gender 20/57 (35%) 110/224 (49%) 0.06 Birth weight, grams,
N available 3417 (+ 458)50 3196 (+666)205 0.006 Preterm (GA<37 weeks) 1/55 (2%) 27/204 (13%) 0.02 GA, (weeks)
N available 39.6 (+ 1.9)45 38.5 (+2.6)192 0.001 Breastfeeding 54/57 (95%) 31/224 (14%) <0.001 Congenital anomalies 9/54 (17%) 50/210 (24%) 0.51 Total bilirubin pre KPE, mmol/L
N available 170 [74-418]55 161 [60-454]216 0.05 Direct bilirubin pre KPE, mmol/L
N available 131 [57-287]51 126 [40-403]205 0.09 Aspartate transaminase pre KPE, U/L
N available 200 [28-635]55 173 [26-2250]210 0.11 Alanine transaminase pre KPE, U/L
N available 129 [15-458]55 123 [18-666]206 0.84 GGT pre KPE, U/L
N available 558 [89-2816]53 482 [43-2630]201 0.81 AP pre KPE, U/L
N available 488 [33-1065]53 462 [88-5564]201 0.34 Albumin pre KPE, g/L
N available 35 +5 36 +6 0.08 PT Pre KPE(after vitamin K), sec
N available 12.3 [10.1 – 59.6]38 12.7 [9.9-131]114 0.59 Cirrhosis in liver biopsy 6/37 (16%) 11/130 (8%) 0.17 Age at KPE, days
N available 52 [31-101]57 62 [20-210]224 <0.001 KPE < 60 days 38/57 (67%) 105/224 (47%) 0.01 Steroids post KPE 16/55 (29%) 36/213 (17%) 0.04 AB post KPE 38/57 (67%) 144/224 (64%) 0.74
Patients with VKDB underwent KPE at younger age compared to patients without VKDB, median ages were 52 [range 31-101] days versus 62 [range 20-210] days, respectively (p<0.001). In the VKDB group 38/57 (67%) patients underwent KPE before the age of 60 days, versus 105/224 (47%) in the group without VKDB
4
The median follow-up of patients with VKDB was 70 [6-310] months versus 95 months [1-361] in patients without VKDB (p=0.23). Outcome parameters are described in Table 2. Clearance of jaundice was achieved in 36% of patients with VKDB and in 35% of patients without VKDB (p=0.99). There were no statistically significant differences in biochemical and clinical parameters during assessment for liver transplantation between patients with or without VKDB. During follow-up, patients with VKDB underwent LTx in 40/57 (70%) of cases versus 121/224 (54%) patients without VKDB (p=0.03). Main indications for LTx were progressive liver failure demonstrated by irreversible jaundice and complications of portal hypertension. This did not significantly differ between the two groups. There was no difference in native liver survival (p= 0.40, Figure 1). Median native liver survival of patients with VKDB was 17 [5-264] months versus 15 [1-350] months for patients without VKDB (p=0.89).
Table 2: Outcome parameters of patients who presented with VKDB versus patients without VKDB at initial presentation. Values are expressed as n (%), mean ± SD, or median [range].
VKDB
N= 57 No VKDBN= 224 P value Total bilirubin 6 months post KPE, mmol/L
n available 81 [1-664]49 73 [2-764]142 0.49 Direct bilirubin 6 months post KPE, mmol/L
n available 117 [1-591]42 68 [1-526]117 0.76 ASAT 6 months post KPE, U/L
n available 173 [53-1505]49 135 [16-5402]139 0.04 ALAT 6 months post KPE, U/L
n available 108 [20-775]46 94 [17-1414]140 0.10 GGT, 6 months post KPE, U/L
n available 233 [29-1788]48 225 [17-1154]136 0.68 AP, 6 months post KPE, U/L, median [range]
n available 600 [183- 1417]42 564 [145-2508]128 0.52 Platelet 6 months post KPE, 10^9
n available 201 [23-473]42 191 [31-603]26 0.84 Clearance of jaundice 20/56 (36%) 78/220 (35%) 0.99 Liver transplantation 40/57 (70%) 121/224 (54%) 0.03 Age at LTx, months,
n available 16 [5-196]40 12 [4-346]120 0.45 Native liver survival at end of FU 10/57 (18%) 56/224 (25%) 0.24 Native liver survival, months
N available 17 [5-264]57 15 [1-350]220 0.89 Pre-transplant mortality 7/57 (12%) 47/224 (21%) 0.14 Age at (pre-transplant) death, months
n available 9 [6-22]7 9 [1-219]46 0.99 Follow-up, months 70 [6-310] 95 [1-361] 0.23
Pre-transplant mortality was 12% in patients with VKDB and 21% in patients without VKDB (p=0.14). Median age at pre-transplant death in patients with and without VKDB was 9 months in both groups. Six (11%) of the patients with VKDB died while on waiting list for LTx. In patients without VKDB, 28 (13%) died while on the waiting list for LTx.
Figure 1: Native liver survival in patients who presented with Vitamin K deficiency bleeding compared to patients without Vitamin K deficiency bleeding at initial presentation.
(Log rank 0.40)
Outcome data of patients who did not undergo KPE are described separately in Table 3. Outcome data did not significantly differ between both groups.
Table 3: Outcome data of patients who did not undergo KPE. The group is divided into patients who presented with VKDB versus patients without VKDB at initial presentation (p value for comparison between these two groups). Values are expressed as n (%), mean ± SD, or median [range].
VKDB
N= 3 No VKDBN= 12 P value Liver transplantation 1/3 (33%) 9/12 (75%) 0.17 Native liver survival at end of FU 0/3 0/12 -Native liver survival, months
4
DISCUSSION
In this study we analysed whether vitamin K deficiency bleeding at initial presentation of patients with BA is related to the long-term prognosis of these patients. Our data show that the clearance of jaundice after KPE was similar in these patients. Yet, our data do also show that during follow-up, a higher fraction of BA patients with VKDB undergo LTx than patients without VKDB, 70% versus 54% respectively (p=0.03). There were no statistically significant differences in age at liver transplantation or pre-transplant mortality between the two groups. The higher percentage of LTx in the VKDB group did not coincide with a significant difference in native liver survival between BA patients with or without VKDB. For this study we used the Dutch nationwide database, in which all patients with BA treated in The Netherlands are registered. In this way we minimised the risk of selection bias. In our study 20% of all BA patients presented with VKDB, which is similar to previous reported incidences.9 Our most recent analysis of the Dutch
Vitamin K prophylaxis (2016) reported an incidence of 82% of VKDB at initial presentation of breastfed patients with BA over the last few years.8 In the present
study on data since 1987, 64% of all breastfed infants presented with VKDB, which might suggest underreporting of VKDB in the early years. The presence of VKDB does not seem related to severity of liver disease, given the similar clearance rates of jaundice within 6 months after KPE. In the present series, starting in 1987, patients with VKDB achieved clearance of jaundice in 36% of cases, versus 35% of patients without VKDB (p=0.99). However, at least two differences in patients’ characteristics might have influenced these results.
Firstly, the fraction of preterm infants was higher in the group without VKDB than in the group with VKDB. This might be due to the fact that preterm infants receive additional, often intramuscular, vitamin K prophylaxis in The Netherlands, which could have exerted a preventive effect on VKDB. Secondly, the groups with and without VKDB differ in age at KPE surgery. Patients with VKDB underwent KPE at a younger age than patients without VKDB, 52 versus 62 days respectively (p<0.01). Our previous study showed that parents and primary healthcare doctors do not always recognise alarm symptoms of neonatal cholestasis.14 Insufficient
recognition of other symptoms than VKDB (e.g. discoloured stools) in patients with undiagnosed BA might underlie this delay in diagnosis and treatment. VKDB in combination with neonatal cholestasis seems to be an alarm symptom in which BA is recognised earlier. Previous studies have shown that KPE at younger age results in higher rates of clearance of jaundice and longer native liver survival.15-21
Thus,patients with VKDB underwent KPE surgery at younger age, which might have masked a worse outcome and resulted in a higher clearance of jaundice rate in this group.
A minority of patients was treated with steroids after KPE, with a slight but significant difference between the two groups (17% of patients without VKDB vs. 29% of patients with VKDB; P=0.04). Although the value of post-operative steroids has been disputed and does not seem to affect long-term outcome, we cannot exclude the possibility of a small effect on the clearance of jaundice.
Neither the effects of younger age at KPE surgery nor the use of steroids did prevent against the need for liver transplantation, which was higher in the group with VKDB: 70% versus 55% in patients without VKDB, respectively (p=0.03). The exact cause remains to be elucidated. Although not statistically significant, liver biopsy at diagnosis showed cirrhosis in 16% of patients with VKDB versus in 8% of patients without VKDB. More severe liver damage before KPE might be an underlying cause for the higher number of liver transplantations in the VKDB group. We cannot exclude that larger groups would have resulted in statistical significance. Alternatively, we cannot exclude either that, the increased number of transplants was due to chance.
Not only irreversible jaundice is an indication for LTx, also complications as recurrent cholangitis, portal hypertension and hepatopulmonary syndrome are an indication for transplantation. Yet, in our study, we did not find differences in indications between the both groups. Native liver survival was similar between the two groups, as was pre-transplant mortality. At the end of follow-up 18% of patients with VKDB still lived with their native liver, versus 25% of patients without VKDB (not significant). Interpretation of the Kaplan Meier curve may suggest that there could be a (presently non-significant) difference in native liver survival at the disadvantage of children with VKDB. Although present group sizes are considerable for this rare disease, we cannot exclude that even larger groups would have resulted in statistical significance.
In the present analysis clearance of jaundice rates were inferior to other countries.22, 23 Furthermore, pretransplant mortality was still very high. In total 34
patients (12%) died while on the waiting list for LTx. In the past decade several improvements in the care of BA in our country have been made. Centralisation of care and the introduction of a multidisciplinary team has been realised (only) since 2012, which has resulted in higher clearance of jaundice rates. In this study, a large amount of patients was treated before centralisation. The major burden of pretransplant mortality was from the first part of the study period, in which the BA care was not yet centralised and no living-related liver transplantation program was available. Present pretransplant mortality rate for BA in the Netherlands is
4
In conclusion, the present, unique study shows the results of a large cohort of BA patients with or without VKDB as presenting symptom with an extensive follow-up. Our data indicate that BA patients with VKDB at initial presentation have similar short-term outcome as patients without VKDB. Notwithstanding the similar rates of clearance of jaundice, however, a higher fraction of patients with VKDB underwent liver transplantation than patients without VKDB, during the follow-up of our study. Finally, the presence of VKDB at initial presentation of BA did not affect the pre-transplant mortality or native liver survival. We feel that our present results will be helpful to provide relevant prognostic information to parents of patients with BA.
Acknowledgements
We gratefully thank the other members of the Netherlands Study Group for Biliary Atresia Registry; J.H. Escher, L.W.E. van Heurn, R.H.J. Houwen, A. Kindermann, B. Koot, C. Sloots, I. de Blaauw, G. Damen and D.C. van der Zee, as well as W. de Vries, M. den Dulk and D.B.E. van Wessel for their invaluable help with collection of the data.
REFERENCES
1. von Kries R. Oral versus intramuscular phytomenadione: safety and efficacy compared. Drug Saf 1999;21:1-6.
2. Sutor AH, Dagres N, Niederhoff H. Late form of vitamin K deficiency bleeding in Germany. Klin Padiatr 1995;207:89-97 doi:10.1055/s-2008-1046519.
3. Alatas FS, Hayashida M, Matsuura T, et al. Intracranial hemorrhage associated with vitamin K-deficiency bleeding in patients with biliary atresia: focus on long-term outcomes. J Pediatr Gastroenterol Nutr 2012;54:552-7 doi:10.1097/ MPG.0b013e3182421878 [doi].
4. Shneider BL, Magee JC, Bezerra JA, et al. Efficacy of fat-soluble vitamin supplementation in infants with biliary atresia. Pediatrics 2012;130:e607-14 doi:10.1542/peds.2011-1423. 5. Van Winckel M, De Bruyne R, Van De Velde S, et al. Vitamin K, an update for the
paediatrician. Eur J Pediatr 2009;168:127-34 doi:10.1007/s00431-008-0856-1; 10.1007/ s00431-008-0856-1.
6. Mahadevan SB, Beath SV, McKiernan PJ, et al. The vitamin K debacle and infants with cholestatic liver disease. Arch Dis Child 1999;81:281.
7. Health Council of the Netherlands. Advisory letter Vitamin K supplementation in infants. The Hague: Health Council of the Netherlands; 2010.;Publication no. 2010/11E. 8. Witt M, Kvist N, Jorgensen MH, et al. Prophylactic Dosing of Vitamin K to Prevent
Bleeding. Pediatrics 2016;137:10.1542/peds.2015-4222 doi:10.1542/peds.2015-4222 [doi].
9. van Hasselt PM, de Koning TJ, Kvist N, et al. Prevention of vitamin K deficiency bleeding in breastfed infants: lessons from the Dutch and Danish biliary atresia registries. Pediatrics 2008;121:e857-63 doi:10.1542/peds.2007-1788; 10.1542/peds.2007-1788. 10. de Vries W, de Langen ZJ, Aronson DC, et al. Mortality of biliary atresia in children not
undergoing liver transplantation in the Netherlands. Pediatr Transplant 2011;15:176-83 doi:10.1111/j.1399-3046.2010.01450.x [doi].
11. de Vries W, de Langen ZJ, Groen H, et al. Biliary atresia in the Netherlands: outcome of patients diagnosed between 1987 and 2008. J Pediatr 2012;160:638,644.e2 doi:10.1016/j.jpeds.2011.09.061 [doi].
12. van Wessel DBE, Boere T, Hulzebos CV, et al. Preterm Infants With Biliary Atresia: A Nationwide Cohort Analysis From The Netherlands. J Pediatr Gastroenterol Nutr 2017;65:370-4 doi:10.1097/MPG.0000000000001692 [doi].
13. Sundaram SS, Mack CL, Feldman AG, et al. Biliary atresia: Indications and timing of liver transplantation and optimization of pretransplant care. Liver Transpl 2017;23:96-109 doi:10.1002/lt.24640 [doi].
14. Witt M, Lindeboom J, Wijnja C, et al. Early Detection of Neonatal Cholestasis: Inadequate Assessment of Stool Color by Parents and Primary Healthcare Doctors.
4
16. Hsiao CH, Chang MH, Chen HL, et al. Universal screening for biliary atresia using aninfant stool color card in Taiwan. Hepatology 2008;47:1233-40 doi:10.1002/hep.22182. 17. Lin YC, Chang MH, Liao SF, et al. Decreasing Rate of Biliary Atresia in Taiwan: A Survey,
2004-2009. Pediatrics 2011 doi:10.1542/peds.2011-0742.
18. Schreiber RA, Barker CC, Roberts EA, et al. Biliary atresia: the Canadian experience. J Pediatr 2007;151:659,65, 665.e1 doi:10.1016/j.jpeds.2007.05.051.
19. Serinet MO, Wildhaber BE, Broue P, et al. Impact of age at Kasai operation on its results in late childhood and adolescence: a rational basis for biliary atresia screening. Pediatrics 2009;123:1280-6 doi:10.1542/peds.2008-1949.
20. Lien TH, Chang MH, Wu JF, et al. Effects of the infant stool color card screening program on 5-year outcome of biliary atresia in Taiwan. Hepatology 2011;53:202-8 doi:10.1002/hep.24023; 10.1002/hep.24023.
21. Chardot C, Buet C, Serinet MO, et al. Improving outcomes of biliary atresia: French national series 1986-2009. J Hepatol 2013;58:1209-17 doi:10.1016/j.jhep.2013.01.040; 10.1016/j.jhep.2013.01.040.
22. Davenport M, De Ville de Goyet J, Stringer MD, et al. Seamless management of biliary atresia in England and Wales (1999-2002). Lancet 2004;363:1354-7 doi:10.1016/S0140-6736(04)16045-5.
23. Lampela H, Ritvanen A, Kosola S, et al. National centralization of biliary atresia care to an assigned multidisciplinary team provides high-quality outcomes. Scand J Gastroenterol 2012;47:99-107 doi:10.3109/00365521.2011.627446; 10.3109/00365521.2011.627446.
