Development and validation of a model for early diagnosis of biliary atresia.
Biliary atresia
Diagnostic model
Neonatal cholestasis
Journal
BMC pediatrics
ISSN: 1471-2431
Titre abrégé: BMC Pediatr
Pays: England
ID NLM: 100967804
Informations de publication
Date de publication:
31 10 2023
31 10 2023
Historique:
received:
11
10
2022
accepted:
06
10
2023
medline:
2
11
2023
pubmed:
1
11
2023
entrez:
1
11
2023
Statut:
epublish
Résumé
Early diagnosis of biliary atresia (BA), particularly distinguishing it from other causes of neonatal cholestasis (NC), is challenging. This study aimed to design and validate a predictive model for BA by using the data available at the initial presentation. Infants presenting with NC were retrospectively identified from tertiary referral hospitals and constituted the model design cohort (n = 148); others were enrolled in a prospective observational study and constituted the validation cohort (n = 21). Clinical, laboratory, and abdominal ultrasonographic features associated with BA were assessed. A prediction model was developed using logistic regression and decision tree (DT) analyses. Three predictors, namely, gamma glutamyl transpeptidase (γGT) level, triangular cord sign (TC sign), and gallbladder abnormalities, were identified as factors for diagnosing BA in multivariate logistic regression, which was used to develop the DT model. The area under the receiver operating characteristic (ROC) curve (AUC) value for the model was 0.905, which was greater than those for γGT level, TC sign, or gallbladder abnormalities alone in the prediction of BA. A simple prediction model combining liver function and abdominal ultrasonography findings can provide a moderate and early estimate of the risk of BA in patients with NC.
Sections du résumé
BACKGROUND AND AIMS
Early diagnosis of biliary atresia (BA), particularly distinguishing it from other causes of neonatal cholestasis (NC), is challenging. This study aimed to design and validate a predictive model for BA by using the data available at the initial presentation.
METHODS
Infants presenting with NC were retrospectively identified from tertiary referral hospitals and constituted the model design cohort (n = 148); others were enrolled in a prospective observational study and constituted the validation cohort (n = 21). Clinical, laboratory, and abdominal ultrasonographic features associated with BA were assessed. A prediction model was developed using logistic regression and decision tree (DT) analyses.
RESULTS
Three predictors, namely, gamma glutamyl transpeptidase (γGT) level, triangular cord sign (TC sign), and gallbladder abnormalities, were identified as factors for diagnosing BA in multivariate logistic regression, which was used to develop the DT model. The area under the receiver operating characteristic (ROC) curve (AUC) value for the model was 0.905, which was greater than those for γGT level, TC sign, or gallbladder abnormalities alone in the prediction of BA.
CONCLUSION
A simple prediction model combining liver function and abdominal ultrasonography findings can provide a moderate and early estimate of the risk of BA in patients with NC.
Identifiants
pubmed: 37907911
doi: 10.1186/s12887-023-04370-x
pii: 10.1186/s12887-023-04370-x
pmc: PMC10617173
doi:
Types de publication
Observational Study
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
549Informations de copyright
© 2023. The Author(s).
Références
Dan Med J. 2016 Aug;63(8):
pubmed: 27477793
Acta Paediatr Taiwan. 2007 Jul-Aug;48(4):196-200
pubmed: 18265540
J Pediatr Surg. 2010 Dec;45(12):2361-3
pubmed: 21129545
J Pediatr Gastroenterol Nutr. 2004 Aug;39(2):115-28
pubmed: 15269615
Hepatology. 2018 Dec;68(6):2069-2077
pubmed: 30153340
Pediatr Surg Int. 2018 Apr;34(4):363-380
pubmed: 29397405
Abdom Radiol (NY). 2016 Sep;41(9):1867-8
pubmed: 27040406
J Ultrasound Med. 2017 Oct;36(10):2027-2038
pubmed: 28556285
BMC Pediatr. 2015 Nov 20;15:192
pubmed: 26589959
Pediatrics. 2011 Dec;128(6):e1428-33
pubmed: 22106076
J Pediatr Gastroenterol Nutr. 2013 Apr;56(4):344-54
pubmed: 23263590
AJR Am J Roentgenol. 2016 May;206(5):W73-82
pubmed: 27010179
Eur J Pediatr. 2011 Mar;170(3):279-84
pubmed: 21249394
Radiology. 1982 Sep;144(4):873-5
pubmed: 7111740
J Hepatol. 2014 Jul;61(1):116-23
pubmed: 24657403
Radiology. 2018 Jul;288(1):262-269
pubmed: 29634437
Indian J Pediatr. 2017 Sep;84(9):685-690
pubmed: 28687948
J Pediatr. 2016 Mar;170:211-7.e1-2
pubmed: 26725209
AJR Am J Roentgenol. 1998 Dec;171(6):1693-8
pubmed: 9843315
J Pediatr Gastroenterol Nutr. 2016 Sep;63(3):370-3
pubmed: 26963938
EBioMedicine. 2018 Aug;34:223-230
pubmed: 30077722
Turk J Pediatr. 2008 May-Jun;50(3):253-9
pubmed: 18773671
Pediatrics. 2015 Dec;136(6):e1663-9
pubmed: 26620065
Korean J Radiol. 2015 Nov-Dec;16(6):1364-72
pubmed: 26576128
Pak J Biol Sci. 2009 Jan 1;12(1):95-7
pubmed: 19579927
Pediatr Surg Int. 2011 Jul;27(7):675-9
pubmed: 21207229
J Pediatr (Rio J). 2010 Nov-Dec;86(6):473-9
pubmed: 21140036
Eur Radiol. 2017 May;27(5):1812-1821
pubmed: 27553925