scholarly journals Predictive Models for Neonatal Follow-Up Serum Bilirubin: Model Development and Validation (Preprint)

2020 ◽  
Author(s):  
Joseph H Chou
Keyword(s):  
Neural Network ◽  
Predictive Models ◽  
Serum Bilirubin ◽  
Predictive Accuracy ◽  
Characteristic Curve ◽  
Receiver Operator Characteristic Curve ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Test Set

BACKGROUND Hyperbilirubinemia affects many newborn infants and, if not treated appropriately, can lead to irreversible brain injury. OBJECTIVE This study aims to develop predictive models of follow-up total serum bilirubin measurement and to compare their accuracy with that of clinician predictions. METHODS Subjects were patients born between June 2015 and June 2019 at 4 hospitals in Massachusetts. The prediction target was a follow-up total serum bilirubin measurement obtained &lt;72 hours after a previous measurement. Birth before versus after February 2019 was used to generate a training set (27,428 target measurements) and a held-out test set (3320 measurements), respectively. Multiple supervised learning models were trained. To further assess model performance, predictions on the held-out test set were also compared with corresponding predictions from clinicians. RESULTS The best predictive accuracy on the held-out test set was obtained with the multilayer perceptron (ie, neural network, mean absolute error [MAE] 1.05 mg/dL) and Xgboost (MAE 1.04 mg/dL) models. A limited number of predictors were sufficient for constructing models with the best performance and avoiding overfitting: current bilirubin measurement, last rate of rise, proportion of time under phototherapy, time to next measurement, gestational age at birth, current age, and fractional weight change from birth. Clinicians made a total of 210 prospective predictions. The neural network model accuracy on this subset of predictions had an MAE of 1.06 mg/dL compared with clinician predictions with an MAE of 1.38 mg/dL (<i>P</i>&lt;.0001). In babies born at 35 weeks of gestation or later, this approach was also applied to predict the binary outcome of subsequently exceeding consensus guidelines for phototherapy initiation and achieved an area under the receiver operator characteristic curve of 0.94 (95% CI 0.91 to 0.97). CONCLUSIONS This study developed predictive models for neonatal follow-up total serum bilirubin measurements that outperform clinicians. This may be the first report of models that predict specific bilirubin values, are not limited to near-term patients without risk factors, and take into account the effect of phototherapy. CLINICALTRIAL

scholarly journals Predictive Models for Neonatal Follow-Up Serum Bilirubin: Model Development and Validation

10.2196/21222 ◽  
2020 ◽  
Vol 8 (10) ◽  
pp. e21222
Author(s):  
Joseph H Chou
Keyword(s):  
Neural Network ◽  
Predictive Models ◽  
Serum Bilirubin ◽  
Predictive Accuracy ◽  
Characteristic Curve ◽  
Receiver Operator Characteristic Curve ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Test Set

Background Hyperbilirubinemia affects many newborn infants and, if not treated appropriately, can lead to irreversible brain injury. Objective This study aims to develop predictive models of follow-up total serum bilirubin measurement and to compare their accuracy with that of clinician predictions. Methods Subjects were patients born between June 2015 and June 2019 at 4 hospitals in Massachusetts. The prediction target was a follow-up total serum bilirubin measurement obtained <72 hours after a previous measurement. Birth before versus after February 2019 was used to generate a training set (27,428 target measurements) and a held-out test set (3320 measurements), respectively. Multiple supervised learning models were trained. To further assess model performance, predictions on the held-out test set were also compared with corresponding predictions from clinicians. Results The best predictive accuracy on the held-out test set was obtained with the multilayer perceptron (ie, neural network, mean absolute error [MAE] 1.05 mg/dL) and Xgboost (MAE 1.04 mg/dL) models. A limited number of predictors were sufficient for constructing models with the best performance and avoiding overfitting: current bilirubin measurement, last rate of rise, proportion of time under phototherapy, time to next measurement, gestational age at birth, current age, and fractional weight change from birth. Clinicians made a total of 210 prospective predictions. The neural network model accuracy on this subset of predictions had an MAE of 1.06 mg/dL compared with clinician predictions with an MAE of 1.38 mg/dL (P<.0001). In babies born at 35 weeks of gestation or later, this approach was also applied to predict the binary outcome of subsequently exceeding consensus guidelines for phototherapy initiation and achieved an area under the receiver operator characteristic curve of 0.94 (95% CI 0.91 to 0.97). Conclusions This study developed predictive models for neonatal follow-up total serum bilirubin measurements that outperform clinicians. This may be the first report of models that predict specific bilirubin values, are not limited to near-term patients without risk factors, and take into account the effect of phototherapy.

Efficacy and Safety of Fenofibrate in Uncomplicated Hyperbilirubinemia in Newborn: A Randomized Trial, with a 6-month Follow-up

2020 ◽  
pp. 37-42
Author(s):  
Jayendra R. Gohil ◽  
Vishal S. Rathod ◽  
Bhoomika D. Rathod
Keyword(s):  
Single Dose ◽  
Serum Bilirubin ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Control Group ◽  
Term Neonates ◽  
Significant Difference ◽  
Group A ◽  
Group B

Objective: To study the effect and safety of Fenofibrate in uncomplicated hyperbilirubinemia in newborn with 6-month follow-up. Materials and Methods: This is a randomized controlled clinical trial conducted in 60 normal term neonates admitted for uncomplicated hyperbilirubinemia in NICU at Sir T G Hospital, Bhavnagar from January 2012 to December 2012. The data included: age, sex, total serum bilirubin (TSB), weight and duration of phototherapy. All neonates enrolled in the study received phototherapy. They were divided in two groups of 30 each: control group A and group B receiving Fenofibrate (100 mg/kg single dose). There was statistically insignificant difference between the parameters of age, sex, weight and TSB between the two groups at hospitalization. Data was analyzed by using appropriate statistical methods. Results: Mean values for total serum bilirubin in Fenofibrate group B at 24 and 48 hours after admission were significantly lower than those for control group A (p<0.0001,  p=0.0001). There was no significant difference in fall of TSB between 24 and 48 hours. The mean duration of phototherapy in Fenofibrate group (44.8h: 24-72h) was significantly shorter than that in control group (55.2 h: 24‐96 h) (P=0.02). There were no side effects of the drug observed during the study and during 6 months follow up period. Conclusion: Fenofibrate as a single 100 mg/kg dose in healthy full term neonates, is effective and a safe drug (till six-month follow-up) for neonatal hyperbilirubinemia, that can decrease the time needed for phototherapy and hence hospitalization. Effect of a single dose seems to wane after 24 hours.

33 272 Infants, 7-Year Follow-up: Total Serum Bilirubin, Transfusions Reexamined

PEDIATRICS ◽  
2002 ◽  
Vol 110 (5) ◽  
pp. 1032-1032
Author(s):  
Thomas B. Newman ◽  
Mark Klebanoff
Keyword(s):  
Serum Bilirubin ◽  
Total Serum Bilirubin ◽  
Total Serum

scholarly journals Can bilirubin/albumin ratio predict neurodevelopmental outcome in severe neonatal hyperbilirubinemia? A 3-month follow up study

2021 ◽  
Vol 69 (1) ◽  
Author(s):  
Reem M. Soliman ◽  
Iman F. Iskander ◽  
Esraa A. Elmazzahy ◽  
May A. K. Abdellatif
Keyword(s):  
Serum Bilirubin ◽  
Tertiary Care ◽  
Neurodevelopmental Outcome ◽  
Neonatal Hyperbilirubinemia ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Early Management ◽  
Albumin Ratio ◽  
Auditory Evoked Response

Abstract Background The risk of kernicterus and BIND may be in part determined by total serum bilirubin (TSB) and by the level of non-albumin bound free bilirubin, which can easily pass the blood–brain barrier. Free bilirubin (Bf) seems a more reliable predictor for bilirubin neurotoxicity. Bilirubin/albumin ratio (B/A) is considered a surrogate parameter for Bf and has been more useful than TSB. The aim of the study is to determine whether B/A ratio correlates with BIND in newborns with severe hyperbilirubinemia and if it can predict poor neurologic outcome at 3 months follow up. Results This prospective study included one hundred seventeen outborn neonates ≥ 35 weeks admitted in a tertiary care neonatal intensive care unit, between May and December 2012, with TSB ≥ 20 mg/dl or necessitating exchange transfusion. Total serum bilirubin and serum albumin were done on admission and bilirubin/albumin ratio was calculated. BIND score was calculated. At the age of 3 months, 112 neonates were followed up with a detailed neurological assessment. Babies who depicted any abnormal motor examination were subjected to brain stem auditory evoked response and MRI examination. Seven infants (6.2%) presented with kernicterus on follow up. BIND scores on admission, mean TSB, and bilirubin/albumin ratio was significantly higher in kernicteric infants compared with those having normal neurological outcome at 3 months of age (P 0.001). The lowest TSB level at which kernicterus occurred in our study was 31 mg/dl. Receiver operation characteristics analysis identified B/A ratio cut off value for predicting kernicterus of 9.6 with sensitivity of 100% and specificity of 91.4%, whereas TSB cut off value of 30 mg/dl showed sensitivity of 100% and specificity of 83%. Conclusion B/A ratio is a strong indicator for the risk of kernicterus. B/A is more specific than TSB and should be used in the early management of neonatal hyperbilirubinemia.

Follow-up of Neonates With Total Serum Bilirubin Levels >=25 mg/dL: A Danish Population-Based Study

PEDIATRICS ◽  
2012 ◽  
Vol 130 (1) ◽  
pp. 61-66 ◽  
Author(s):  
P. K. Vandborg ◽  
B. M. Hansen ◽  
G. Greisen ◽  
M. Jepsen ◽  
F. Ebbesen
Keyword(s):  
Serum Bilirubin ◽  
Population Based ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Danish Population ◽  
Population Based Study

scholarly journals Poor odor identification predicts mortality risk in older adults without neurodegenerative diseases: the Shanghai Aging Study

2020 ◽  
Author(s):  
Zhenxu Xiao ◽  
Qianhua Zhao ◽  
Xiaoniu Liang ◽  
Wanqing Wu ◽  
Yang Cao ◽  
...  
Keyword(s):  
Older Adults ◽  
Predictive Models ◽  
Predictive Accuracy ◽  
Characteristic Curve ◽  
Cox Proportional Hazards ◽  
Predictive Values ◽  
All Cause Mortality ◽  
Long Term Mortality

AbstractBackgroundIrrespective of neurodegeneration, the decline of olfactory function might also reflect a wider range of pathological conditions contributing to mortality. However, the potential explanations and their predictive values have rarely been reported.MethodsA total of 1,433 older adults aged ≥ 60 years without neurodegenerative disease were administered a follow-up of 8.6 years on average. Sniffin’ Sticks Screening Test was used to assess the olfactory identification at baseline. Survival status during the follow-up was obtained from the local Center for Disease Control and Prevention. Bidirectional stepwise Cox proportional hazards regression was used to identify variables associated with mortality. Two predictive models were constructed by statistical learning methods.ResultsAll-cause mortality rate was 1.1/100 person-years during the follow-up. Sex (HR = 0.65, 95%CI 0.46 - 0.93), age (HR = 1.12, 95%CI 1.10 - 1.15), chronic kidney disease (HR = 1.88, 95%CI 1.09 - 3.25), low density lipoprotein (HR = 0.81, 95%CI 0.67 - 0.99), anemia (HR = 2.74, 95%CI 1.19 - 6.30), and fail to identify coffee odor (HR = 1.96, 95%CI 1.19 - 3.23) were significantly associated with all-cause mortality. The logistic regression and the random forest predictive models showed similar predictive accuracy (0.91 and 0.90) and area under the receiver operating characteristic curve (0.77 and 0.75).ConclusionsThe association between poor olfactory and long-term mortality was verified among Chinese older population. Certain odors identification ability may contribute to the prediction of long-term mortality along with other important risk factors.

33 272 Infants, 7-Year Follow-up: Total Serum Bilirubin, Transfusions Reexamined

PEDIATRICS ◽  
2002 ◽  
Vol 110 (5) ◽  
pp. 1032-1032 ◽  
Author(s):  
T. B. Newman ◽  
M. Klebanoff
Keyword(s):  
Serum Bilirubin ◽  
Total Serum Bilirubin ◽  
Total Serum

Two Different UGT1A1 Mutations causing Crigler–Najjar Syndrome types I and II in an Iranian Family

2015 ◽  
Vol 24 (4) ◽  
pp. 523-526 ◽  
Author(s):  
Yoshihiro Maruo ◽  
Mahdiyeh Behnam ◽  
Shinichi Ikushiro ◽  
Sayuri Nakahara ◽  
Narges Nouri ◽  
...  
Keyword(s):  
Serum Bilirubin ◽  
Total Serum Bilirubin ◽  
Total Serum ◽  
Compound Heterozygous ◽  
Type I ◽  
Syndrome Type ◽  
Type Ii ◽  
Wild Type ◽  
Iranian Family ◽  
Udp Glucuronosyltransferase

Background: Crigler–Najjar syndrome type I (CN-1) and type II (CN-2) are rare hereditary unconjugated hyperbilirubinemia disorders. However, there have been no reports regarding the co-existence of CN-1 and CN-2 in one family. We experienced a case of an Iranian family that included members with either CN-1 or CN-2. Genetic analysis revealed a mutation in the bilirubin UDP-glucuronosyltransferase (UGT1A1) gene that resulted in residual enzymatic activity.Case report: The female proband developed severe hyperbilirubinemia [total serum bilirubin concentration (TB) = 34.8 mg/dL] with bilirubin encephalopathy (kernicterus) and died after liver transplantation. Her family history included a cousin with kernicterus (TB = 30.0 mg/dL) diagnosed as CN-1. Her great grandfather (TB unknown) and uncle (TB = 23.0 mg/dL) developed jaundice, but without any treatment, they remained healthy as CN-2. Results: The affected cousin was homozygous for a novel frameshift mutation (c.381insGG, p.C127WfsX23). The affected uncle was compound heterozygous for p.C127WfsX23 and p.V225G linked with A(TA)7TAA. p.V225G-UGT1A1 reduced glucuronidation activity to 60% of wild-type. Thus, linkage of A(TA)7TAA and p.V225G might reduce UGT1A1 activity to 18%–36 % of the wild-type. Conclusion: Genetic and in vitro expression analyses are useful for accurate genetic counseling for a family with a history of both CN-1 and CN-2. Abbreviations: CN-1: Crigler–Najjar syndrome type I; CN-2: Crigler–Najjar syndrome type II; GS: Gilbert syndrome; UGT1A1: bilirubin UDP-glucuronosyltransferase; WT: Wild type; TB: total serum bilirubin.

Sign in / Sign up

Export Citation Format

Share Document