Proteomics Improves the Prediction of Burns Mortality: Results from Regression Spline Modeling

Celeste Finnerty, Hyunsu Ju, Heidi Spratt, Sundar Victor, Marc G. Jeschke, Sachin Hegde, Suresh Bhavnani, Bruce A. Luxon, Allan R. Brasier, David Herndon

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Prediction of mortality in severely burned patients remains unreliable. Although clinical covariates and plasma protein abundance have been used with varying degrees of success, the triad of burn size, inhalation injury, and age remains the most reliable predictor. We investigated the effect of combining proteomics variables with these three clinical covariates on prediction of mortality in burned children. Serum samples were collected from 330 burned children (burns covering >25% of the total body surface area) between admission and the time of the first operation for clinical chemistry analyses and proteomic assays of cytokines. Principal component analysis revealed that serum protein abundance and the clinical covariates each provided independent information regarding patient survival. To determine whether combining proteomics with clinical variables improves prediction of patient mortality, we used multivariate adaptive regression splines, because the relationships between analytes and mortality were not linear. Combining these factors increased overall outcome prediction accuracy from 52% to 81% and area under the receiver operating characteristic curve from 0.82 to 0.95. Thus, the predictive accuracy of burns mortality is substantially improved by combining protein abundance information with clinical covariates in a multivariate adaptive regression splines classifier, a model currently being validated in a prospective study.

Original languageEnglish (US)
Pages (from-to)243-249
Number of pages7
JournalClinical and Translational Science
Volume5
Issue number3
DOIs
StatePublished - Jun 2012

Fingerprint

Burns
Splines
Proteomics
Mortality
Blood Proteins
Inhalation Burns
Clinical Chemistry
Principal component analysis
Body Surface Area
Assays
Principal Component Analysis
Classifiers
ROC Curve
Cytokines
Prospective Studies
Survival
Wounds and Injuries
Proteins
Serum

Keywords

  • Cytokines
  • Mortality
  • Pediatrics
  • Stress

ASJC Scopus subject areas

  • Medicine(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Neuroscience(all)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

Proteomics Improves the Prediction of Burns Mortality : Results from Regression Spline Modeling. / Finnerty, Celeste; Ju, Hyunsu; Spratt, Heidi; Victor, Sundar; Jeschke, Marc G.; Hegde, Sachin; Bhavnani, Suresh; Luxon, Bruce A.; Brasier, Allan R.; Herndon, David.

In: Clinical and Translational Science, Vol. 5, No. 3, 06.2012, p. 243-249.

Research output: Contribution to journalArticle

Finnerty, Celeste ; Ju, Hyunsu ; Spratt, Heidi ; Victor, Sundar ; Jeschke, Marc G. ; Hegde, Sachin ; Bhavnani, Suresh ; Luxon, Bruce A. ; Brasier, Allan R. ; Herndon, David. / Proteomics Improves the Prediction of Burns Mortality : Results from Regression Spline Modeling. In: Clinical and Translational Science. 2012 ; Vol. 5, No. 3. pp. 243-249.
@article{881edc974e614df88c9d776b4d862546,
title = "Proteomics Improves the Prediction of Burns Mortality: Results from Regression Spline Modeling",
abstract = "Prediction of mortality in severely burned patients remains unreliable. Although clinical covariates and plasma protein abundance have been used with varying degrees of success, the triad of burn size, inhalation injury, and age remains the most reliable predictor. We investigated the effect of combining proteomics variables with these three clinical covariates on prediction of mortality in burned children. Serum samples were collected from 330 burned children (burns covering >25{\%} of the total body surface area) between admission and the time of the first operation for clinical chemistry analyses and proteomic assays of cytokines. Principal component analysis revealed that serum protein abundance and the clinical covariates each provided independent information regarding patient survival. To determine whether combining proteomics with clinical variables improves prediction of patient mortality, we used multivariate adaptive regression splines, because the relationships between analytes and mortality were not linear. Combining these factors increased overall outcome prediction accuracy from 52{\%} to 81{\%} and area under the receiver operating characteristic curve from 0.82 to 0.95. Thus, the predictive accuracy of burns mortality is substantially improved by combining protein abundance information with clinical covariates in a multivariate adaptive regression splines classifier, a model currently being validated in a prospective study.",
keywords = "Cytokines, Mortality, Pediatrics, Stress",
author = "Celeste Finnerty and Hyunsu Ju and Heidi Spratt and Sundar Victor and Jeschke, {Marc G.} and Sachin Hegde and Suresh Bhavnani and Luxon, {Bruce A.} and Brasier, {Allan R.} and David Herndon",
year = "2012",
month = "6",
doi = "10.1111/j.1752-8062.2012.00412.x",
language = "English (US)",
volume = "5",
pages = "243--249",
journal = "Clinical and Translational Science",
issn = "1752-8054",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Proteomics Improves the Prediction of Burns Mortality

T2 - Results from Regression Spline Modeling

AU - Finnerty, Celeste

AU - Ju, Hyunsu

AU - Spratt, Heidi

AU - Victor, Sundar

AU - Jeschke, Marc G.

AU - Hegde, Sachin

AU - Bhavnani, Suresh

AU - Luxon, Bruce A.

AU - Brasier, Allan R.

AU - Herndon, David

PY - 2012/6

Y1 - 2012/6

N2 - Prediction of mortality in severely burned patients remains unreliable. Although clinical covariates and plasma protein abundance have been used with varying degrees of success, the triad of burn size, inhalation injury, and age remains the most reliable predictor. We investigated the effect of combining proteomics variables with these three clinical covariates on prediction of mortality in burned children. Serum samples were collected from 330 burned children (burns covering >25% of the total body surface area) between admission and the time of the first operation for clinical chemistry analyses and proteomic assays of cytokines. Principal component analysis revealed that serum protein abundance and the clinical covariates each provided independent information regarding patient survival. To determine whether combining proteomics with clinical variables improves prediction of patient mortality, we used multivariate adaptive regression splines, because the relationships between analytes and mortality were not linear. Combining these factors increased overall outcome prediction accuracy from 52% to 81% and area under the receiver operating characteristic curve from 0.82 to 0.95. Thus, the predictive accuracy of burns mortality is substantially improved by combining protein abundance information with clinical covariates in a multivariate adaptive regression splines classifier, a model currently being validated in a prospective study.

AB - Prediction of mortality in severely burned patients remains unreliable. Although clinical covariates and plasma protein abundance have been used with varying degrees of success, the triad of burn size, inhalation injury, and age remains the most reliable predictor. We investigated the effect of combining proteomics variables with these three clinical covariates on prediction of mortality in burned children. Serum samples were collected from 330 burned children (burns covering >25% of the total body surface area) between admission and the time of the first operation for clinical chemistry analyses and proteomic assays of cytokines. Principal component analysis revealed that serum protein abundance and the clinical covariates each provided independent information regarding patient survival. To determine whether combining proteomics with clinical variables improves prediction of patient mortality, we used multivariate adaptive regression splines, because the relationships between analytes and mortality were not linear. Combining these factors increased overall outcome prediction accuracy from 52% to 81% and area under the receiver operating characteristic curve from 0.82 to 0.95. Thus, the predictive accuracy of burns mortality is substantially improved by combining protein abundance information with clinical covariates in a multivariate adaptive regression splines classifier, a model currently being validated in a prospective study.

KW - Cytokines

KW - Mortality

KW - Pediatrics

KW - Stress

UR - http://www.scopus.com/inward/record.url?scp=84862158518&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862158518&partnerID=8YFLogxK

U2 - 10.1111/j.1752-8062.2012.00412.x

DO - 10.1111/j.1752-8062.2012.00412.x

M3 - Article

C2 - 22686201

AN - SCOPUS:84862158518

VL - 5

SP - 243

EP - 249

JO - Clinical and Translational Science

JF - Clinical and Translational Science

SN - 1752-8054

IS - 3

ER -