A multivariable model of clinical variables predicts advanced fibrosis in chronic hepatitis C

Mazen Alsatie, Paul Y. Kwo, Joel R. Gingerich, Rong Qi, George Eckert, Oscar Cummings, Thomas Imperiale

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

BACKGROUND: A noninvasive method to identify advanced hepatic fibrosis (AHF) in chronic hepatitis C (CHC) could preclude the need for routine liver biopsy. Recent evidence suggests that obesity may contribute to hepatic fibrosis in hepatitis C virus infection. GOALS: To determine whether clinical variables, including body mass index (BMI), can predict risk of AHF. STUDY: Retrospective review of untreated CHC patients evaluated between 1993 and 2002 without clinical or physical evidence of end-stage liver disease. Liver biopsies were scored for fibrosis, steatosis, and inflammation. Multivariable analysis was used to derive and internally validate a prediction equation. A clinical index was created from the equation by assigning points for each variable. The risk of AHF was measured for each risk category. RESULTS: Two hundred eighty-six satisfied inclusion criteria, of which 86 (30%) had AHF. In the derivation subgroup (N=190), 5 factors were independently associated with AHF: diabetes mellitus, platelets count <150,000, aspartate aminotransferase ≥65 IU/mL, international normalized ratio ≥1.1, and bilirubin ≥0.85 mg/dL. The corresponding risk index contained 3 categories: low-risk (score of 0), intermediate risk (scores of 1 to 3), and high risk (scores of ≥4), in which the respective risks of AHF were 9%, 34%, and 92%. Inclusion of BMI did not improve model performance. CONCLUSIONS: A model for estimating AHF risk in CHC performed well in this population. BMI had no effect on the risk of AHF. If this model can be validated in other patient cohorts, it could preclude the need for liver biopsy in patients with scores of 0 or ≥4.

Original languageEnglish
Pages (from-to)416-421
Number of pages6
JournalJournal of Clinical Gastroenterology
Volume41
Issue number4
DOIs
StatePublished - Apr 2007

Fingerprint

Chronic Hepatitis C
Fibrosis
Liver
Body Mass Index
Biopsy
End Stage Liver Disease
International Normalized Ratio
Inclusion Bodies
Virus Diseases
Aspartate Aminotransferases
Platelet Count
Bilirubin
Hepacivirus
Diabetes Mellitus
Obesity
Inflammation

Keywords

  • Cirrhosis
  • Hepatic fibrosis
  • Hepatitis C
  • Obesity
  • Steatosis

ASJC Scopus subject areas

  • Gastroenterology

Cite this

A multivariable model of clinical variables predicts advanced fibrosis in chronic hepatitis C. / Alsatie, Mazen; Kwo, Paul Y.; Gingerich, Joel R.; Qi, Rong; Eckert, George; Cummings, Oscar; Imperiale, Thomas.

In: Journal of Clinical Gastroenterology, Vol. 41, No. 4, 04.2007, p. 416-421.

Research output: Contribution to journalArticle

Alsatie, Mazen ; Kwo, Paul Y. ; Gingerich, Joel R. ; Qi, Rong ; Eckert, George ; Cummings, Oscar ; Imperiale, Thomas. / A multivariable model of clinical variables predicts advanced fibrosis in chronic hepatitis C. In: Journal of Clinical Gastroenterology. 2007 ; Vol. 41, No. 4. pp. 416-421.
@article{6b963d8f1e464008b18e3011427d5185,
title = "A multivariable model of clinical variables predicts advanced fibrosis in chronic hepatitis C",
abstract = "BACKGROUND: A noninvasive method to identify advanced hepatic fibrosis (AHF) in chronic hepatitis C (CHC) could preclude the need for routine liver biopsy. Recent evidence suggests that obesity may contribute to hepatic fibrosis in hepatitis C virus infection. GOALS: To determine whether clinical variables, including body mass index (BMI), can predict risk of AHF. STUDY: Retrospective review of untreated CHC patients evaluated between 1993 and 2002 without clinical or physical evidence of end-stage liver disease. Liver biopsies were scored for fibrosis, steatosis, and inflammation. Multivariable analysis was used to derive and internally validate a prediction equation. A clinical index was created from the equation by assigning points for each variable. The risk of AHF was measured for each risk category. RESULTS: Two hundred eighty-six satisfied inclusion criteria, of which 86 (30{\%}) had AHF. In the derivation subgroup (N=190), 5 factors were independently associated with AHF: diabetes mellitus, platelets count <150,000, aspartate aminotransferase ≥65 IU/mL, international normalized ratio ≥1.1, and bilirubin ≥0.85 mg/dL. The corresponding risk index contained 3 categories: low-risk (score of 0), intermediate risk (scores of 1 to 3), and high risk (scores of ≥4), in which the respective risks of AHF were 9{\%}, 34{\%}, and 92{\%}. Inclusion of BMI did not improve model performance. CONCLUSIONS: A model for estimating AHF risk in CHC performed well in this population. BMI had no effect on the risk of AHF. If this model can be validated in other patient cohorts, it could preclude the need for liver biopsy in patients with scores of 0 or ≥4.",
keywords = "Cirrhosis, Hepatic fibrosis, Hepatitis C, Obesity, Steatosis",
author = "Mazen Alsatie and Kwo, {Paul Y.} and Gingerich, {Joel R.} and Rong Qi and George Eckert and Oscar Cummings and Thomas Imperiale",
year = "2007",
month = "4",
doi = "10.1097/01.mcg.0000225593.93577.64",
language = "English",
volume = "41",
pages = "416--421",
journal = "Journal of Clinical Gastroenterology",
issn = "0192-0790",
publisher = "Lippincott Williams and Wilkins",
number = "4",

}

TY - JOUR

T1 - A multivariable model of clinical variables predicts advanced fibrosis in chronic hepatitis C

AU - Alsatie, Mazen

AU - Kwo, Paul Y.

AU - Gingerich, Joel R.

AU - Qi, Rong

AU - Eckert, George

AU - Cummings, Oscar

AU - Imperiale, Thomas

PY - 2007/4

Y1 - 2007/4

N2 - BACKGROUND: A noninvasive method to identify advanced hepatic fibrosis (AHF) in chronic hepatitis C (CHC) could preclude the need for routine liver biopsy. Recent evidence suggests that obesity may contribute to hepatic fibrosis in hepatitis C virus infection. GOALS: To determine whether clinical variables, including body mass index (BMI), can predict risk of AHF. STUDY: Retrospective review of untreated CHC patients evaluated between 1993 and 2002 without clinical or physical evidence of end-stage liver disease. Liver biopsies were scored for fibrosis, steatosis, and inflammation. Multivariable analysis was used to derive and internally validate a prediction equation. A clinical index was created from the equation by assigning points for each variable. The risk of AHF was measured for each risk category. RESULTS: Two hundred eighty-six satisfied inclusion criteria, of which 86 (30%) had AHF. In the derivation subgroup (N=190), 5 factors were independently associated with AHF: diabetes mellitus, platelets count <150,000, aspartate aminotransferase ≥65 IU/mL, international normalized ratio ≥1.1, and bilirubin ≥0.85 mg/dL. The corresponding risk index contained 3 categories: low-risk (score of 0), intermediate risk (scores of 1 to 3), and high risk (scores of ≥4), in which the respective risks of AHF were 9%, 34%, and 92%. Inclusion of BMI did not improve model performance. CONCLUSIONS: A model for estimating AHF risk in CHC performed well in this population. BMI had no effect on the risk of AHF. If this model can be validated in other patient cohorts, it could preclude the need for liver biopsy in patients with scores of 0 or ≥4.

AB - BACKGROUND: A noninvasive method to identify advanced hepatic fibrosis (AHF) in chronic hepatitis C (CHC) could preclude the need for routine liver biopsy. Recent evidence suggests that obesity may contribute to hepatic fibrosis in hepatitis C virus infection. GOALS: To determine whether clinical variables, including body mass index (BMI), can predict risk of AHF. STUDY: Retrospective review of untreated CHC patients evaluated between 1993 and 2002 without clinical or physical evidence of end-stage liver disease. Liver biopsies were scored for fibrosis, steatosis, and inflammation. Multivariable analysis was used to derive and internally validate a prediction equation. A clinical index was created from the equation by assigning points for each variable. The risk of AHF was measured for each risk category. RESULTS: Two hundred eighty-six satisfied inclusion criteria, of which 86 (30%) had AHF. In the derivation subgroup (N=190), 5 factors were independently associated with AHF: diabetes mellitus, platelets count <150,000, aspartate aminotransferase ≥65 IU/mL, international normalized ratio ≥1.1, and bilirubin ≥0.85 mg/dL. The corresponding risk index contained 3 categories: low-risk (score of 0), intermediate risk (scores of 1 to 3), and high risk (scores of ≥4), in which the respective risks of AHF were 9%, 34%, and 92%. Inclusion of BMI did not improve model performance. CONCLUSIONS: A model for estimating AHF risk in CHC performed well in this population. BMI had no effect on the risk of AHF. If this model can be validated in other patient cohorts, it could preclude the need for liver biopsy in patients with scores of 0 or ≥4.

KW - Cirrhosis

KW - Hepatic fibrosis

KW - Hepatitis C

KW - Obesity

KW - Steatosis

UR - http://www.scopus.com/inward/record.url?scp=34147098113&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34147098113&partnerID=8YFLogxK

U2 - 10.1097/01.mcg.0000225593.93577.64

DO - 10.1097/01.mcg.0000225593.93577.64

M3 - Article

C2 - 17413613

AN - SCOPUS:34147098113

VL - 41

SP - 416

EP - 421

JO - Journal of Clinical Gastroenterology

JF - Journal of Clinical Gastroenterology

SN - 0192-0790

IS - 4

ER -