Initial assessment of the infant with neonatal cholestasis-Is this biliary atresia?

Childhood Liver Disease Research Network

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Introduction Optimizing outcome in biliary atresia (BA) requires timely diagnosis. Cholestasis is a presenting feature of BA, as well as other diagnoses (Non-BA). Identification of clinical features of neonatal cholestasis that would expedite decisions to pursue subsequent invasive testing to correctly diagnose or exclude BA would enhance outcomes. The analytical goal was to develop a predictive model for BA using data available at initial presentation. Methods Infants at presentation with neonatal cholestasis (direct/conjugated bilirubin >2 mg/dl [34.2 μ M]) were enrolled prior to surgical exploration in a prospective observational multi-centered study (PROBENCT00061828). Clinical features (physical findings, laboratory results, gallbladder sonography) at enrollment were analyzed. Initially, 19 features were selected as candidate predictors. Two approaches were used to build models for diagnosis prediction: a hierarchical classification and regression decision tree (CART) and a logistic regression model using a stepwise selection strategy. Results In PROBE April 2004-February 2014, 401 infants met criteria for BA and 259 for Non-BA. Univariate analysis identified 13 features that were significantly different between BA and Non-BA. Using a CART predictive model of BA versus Non-BA (significant factors: gammaglutamyl transpeptidase, acholic stools, weight), the receiver operating characteristic area under the curve (ROC AUC) was 0.83. Twelve percent of BA infants were misclassified as Non-BA; 17% of Non-BA infants were misclassified as BA. Stepwise logistic regression identified seven factors in a predictive model (ROC AUC 0.89). Using this model, a predicted probability of >0.8 (n = 357) yielded an 81% true positive rate for BA; <0.2 (n = 120) yielded an 11% false negative rate. Conclusion Despite the relatively good accuracy of our optimized prediction models, the high precision required for differentiating BA from Non-BA was not achieved. Accurate identification of BA in infants with neonatal cholestasis requires further evaluation, and BA should not be excluded based only on presenting clinical features.

Original languageEnglish (US)
Article numbere0176275
JournalPloS one
Volume12
Issue number5
DOIs
StatePublished - May 1 2017
Externally publishedYes

Fingerprint

Biliary Atresia
cholestasis
Cholestasis
abnormal development
Decision trees
Logistics
Peptidyl Transferases
Decision Trees
Ultrasonography
Logistic Models
ROC Curve
Bilirubin
Area Under Curve
Gallbladder
Testing

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Initial assessment of the infant with neonatal cholestasis-Is this biliary atresia? / Childhood Liver Disease Research Network.

In: PloS one, Vol. 12, No. 5, e0176275, 01.05.2017.

Research output: Contribution to journalArticle

Childhood Liver Disease Research Network. / Initial assessment of the infant with neonatal cholestasis-Is this biliary atresia?. In: PloS one. 2017 ; Vol. 12, No. 5.
@article{dbd4996d15644e73908dc311aa8f64ee,
title = "Initial assessment of the infant with neonatal cholestasis-Is this biliary atresia?",
abstract = "Introduction Optimizing outcome in biliary atresia (BA) requires timely diagnosis. Cholestasis is a presenting feature of BA, as well as other diagnoses (Non-BA). Identification of clinical features of neonatal cholestasis that would expedite decisions to pursue subsequent invasive testing to correctly diagnose or exclude BA would enhance outcomes. The analytical goal was to develop a predictive model for BA using data available at initial presentation. Methods Infants at presentation with neonatal cholestasis (direct/conjugated bilirubin >2 mg/dl [34.2 μ M]) were enrolled prior to surgical exploration in a prospective observational multi-centered study (PROBENCT00061828). Clinical features (physical findings, laboratory results, gallbladder sonography) at enrollment were analyzed. Initially, 19 features were selected as candidate predictors. Two approaches were used to build models for diagnosis prediction: a hierarchical classification and regression decision tree (CART) and a logistic regression model using a stepwise selection strategy. Results In PROBE April 2004-February 2014, 401 infants met criteria for BA and 259 for Non-BA. Univariate analysis identified 13 features that were significantly different between BA and Non-BA. Using a CART predictive model of BA versus Non-BA (significant factors: gammaglutamyl transpeptidase, acholic stools, weight), the receiver operating characteristic area under the curve (ROC AUC) was 0.83. Twelve percent of BA infants were misclassified as Non-BA; 17{\%} of Non-BA infants were misclassified as BA. Stepwise logistic regression identified seven factors in a predictive model (ROC AUC 0.89). Using this model, a predicted probability of >0.8 (n = 357) yielded an 81{\%} true positive rate for BA; <0.2 (n = 120) yielded an 11{\%} false negative rate. Conclusion Despite the relatively good accuracy of our optimized prediction models, the high precision required for differentiating BA from Non-BA was not achieved. Accurate identification of BA in infants with neonatal cholestasis requires further evaluation, and BA should not be excluded based only on presenting clinical features.",
author = "{Childhood Liver Disease Research Network} and Shneider, {Benjamin L.} and Jeff Moore and Nanda Kerkar and Magee, {John C.} and Wen Ye and Karpen, {Saul J.} and Kamath, {Binita M.} and Molleston, {Jean P.} and Bezerra, {Jorge A.} and Murray, {Karen F.} and Loomes, {Kathleen M.} and Whitington, {Peter F.} and Philip Rosenthal and Squires, {Robert H.} and Guthery, {Stephen L.} and Ronen Arnon and Schwarz, {Kathleen B.} and Turmelle, {Yumirle P.} and Sherker, {Averell H.} and Sokol, {Ronald J.} and Hertel, {Paula M.} and Alonso, {Estella M.} and Fredericks, {Emily M.} and Haber, {Barbara H.} and Wang, {Kasper S.} and Sorensen, {Lisa G.} and Ng, {Vicky Lee} and Lee Bass and Henry Lin and Goodrich, {Nathan P.} and Kieran Hawthorne and Heubi, {James E.} and Rachel Sheridan and Lin Fei and Jeffrey Teckman and Spino, {Catherine A.}",
year = "2017",
month = "5",
day = "1",
doi = "10.1371/journal.pone.0176275",
language = "English (US)",
volume = "12",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "5",

}

TY - JOUR

T1 - Initial assessment of the infant with neonatal cholestasis-Is this biliary atresia?

AU - Childhood Liver Disease Research Network

AU - Shneider, Benjamin L.

AU - Moore, Jeff

AU - Kerkar, Nanda

AU - Magee, John C.

AU - Ye, Wen

AU - Karpen, Saul J.

AU - Kamath, Binita M.

AU - Molleston, Jean P.

AU - Bezerra, Jorge A.

AU - Murray, Karen F.

AU - Loomes, Kathleen M.

AU - Whitington, Peter F.

AU - Rosenthal, Philip

AU - Squires, Robert H.

AU - Guthery, Stephen L.

AU - Arnon, Ronen

AU - Schwarz, Kathleen B.

AU - Turmelle, Yumirle P.

AU - Sherker, Averell H.

AU - Sokol, Ronald J.

AU - Hertel, Paula M.

AU - Alonso, Estella M.

AU - Fredericks, Emily M.

AU - Haber, Barbara H.

AU - Wang, Kasper S.

AU - Sorensen, Lisa G.

AU - Ng, Vicky Lee

AU - Bass, Lee

AU - Lin, Henry

AU - Goodrich, Nathan P.

AU - Hawthorne, Kieran

AU - Heubi, James E.

AU - Sheridan, Rachel

AU - Fei, Lin

AU - Teckman, Jeffrey

AU - Spino, Catherine A.

PY - 2017/5/1

Y1 - 2017/5/1

N2 - Introduction Optimizing outcome in biliary atresia (BA) requires timely diagnosis. Cholestasis is a presenting feature of BA, as well as other diagnoses (Non-BA). Identification of clinical features of neonatal cholestasis that would expedite decisions to pursue subsequent invasive testing to correctly diagnose or exclude BA would enhance outcomes. The analytical goal was to develop a predictive model for BA using data available at initial presentation. Methods Infants at presentation with neonatal cholestasis (direct/conjugated bilirubin >2 mg/dl [34.2 μ M]) were enrolled prior to surgical exploration in a prospective observational multi-centered study (PROBENCT00061828). Clinical features (physical findings, laboratory results, gallbladder sonography) at enrollment were analyzed. Initially, 19 features were selected as candidate predictors. Two approaches were used to build models for diagnosis prediction: a hierarchical classification and regression decision tree (CART) and a logistic regression model using a stepwise selection strategy. Results In PROBE April 2004-February 2014, 401 infants met criteria for BA and 259 for Non-BA. Univariate analysis identified 13 features that were significantly different between BA and Non-BA. Using a CART predictive model of BA versus Non-BA (significant factors: gammaglutamyl transpeptidase, acholic stools, weight), the receiver operating characteristic area under the curve (ROC AUC) was 0.83. Twelve percent of BA infants were misclassified as Non-BA; 17% of Non-BA infants were misclassified as BA. Stepwise logistic regression identified seven factors in a predictive model (ROC AUC 0.89). Using this model, a predicted probability of >0.8 (n = 357) yielded an 81% true positive rate for BA; <0.2 (n = 120) yielded an 11% false negative rate. Conclusion Despite the relatively good accuracy of our optimized prediction models, the high precision required for differentiating BA from Non-BA was not achieved. Accurate identification of BA in infants with neonatal cholestasis requires further evaluation, and BA should not be excluded based only on presenting clinical features.

AB - Introduction Optimizing outcome in biliary atresia (BA) requires timely diagnosis. Cholestasis is a presenting feature of BA, as well as other diagnoses (Non-BA). Identification of clinical features of neonatal cholestasis that would expedite decisions to pursue subsequent invasive testing to correctly diagnose or exclude BA would enhance outcomes. The analytical goal was to develop a predictive model for BA using data available at initial presentation. Methods Infants at presentation with neonatal cholestasis (direct/conjugated bilirubin >2 mg/dl [34.2 μ M]) were enrolled prior to surgical exploration in a prospective observational multi-centered study (PROBENCT00061828). Clinical features (physical findings, laboratory results, gallbladder sonography) at enrollment were analyzed. Initially, 19 features were selected as candidate predictors. Two approaches were used to build models for diagnosis prediction: a hierarchical classification and regression decision tree (CART) and a logistic regression model using a stepwise selection strategy. Results In PROBE April 2004-February 2014, 401 infants met criteria for BA and 259 for Non-BA. Univariate analysis identified 13 features that were significantly different between BA and Non-BA. Using a CART predictive model of BA versus Non-BA (significant factors: gammaglutamyl transpeptidase, acholic stools, weight), the receiver operating characteristic area under the curve (ROC AUC) was 0.83. Twelve percent of BA infants were misclassified as Non-BA; 17% of Non-BA infants were misclassified as BA. Stepwise logistic regression identified seven factors in a predictive model (ROC AUC 0.89). Using this model, a predicted probability of >0.8 (n = 357) yielded an 81% true positive rate for BA; <0.2 (n = 120) yielded an 11% false negative rate. Conclusion Despite the relatively good accuracy of our optimized prediction models, the high precision required for differentiating BA from Non-BA was not achieved. Accurate identification of BA in infants with neonatal cholestasis requires further evaluation, and BA should not be excluded based only on presenting clinical features.

UR - http://www.scopus.com/inward/record.url?scp=85018873789&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018873789&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0176275

DO - 10.1371/journal.pone.0176275

M3 - Article

VL - 12

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 5

M1 - e0176275

ER -