Non-invasive response prediction in prophylactic carvedilol therapy for cirrhotic patients with esophageal varices

Hwi Young Kim, Young Ho So, Won Kim, Dong Won Ahn, Yong Jin Jung, Hyunsik Woo, Donghee Kim, Moon Young Kim, Soon Koo Baik

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background & Aims: Non-selective beta-blockers (NSBBs) are the mainstay of primary prophylaxis of esophageal variceal bleeding in patients with liver cirrhosis. We investigated whether non-invasive markers of portal hypertension correlate with hemodynamic responses to NSBBs in cirrhotic patients with esophageal varices. Methods: In this prospective cohort study, 106 cirrhotic patients with high-risk esophageal varices in the derivation cohort received carvedilol prophylaxis, and completed paired measurements of hepatic venous pressure gradient, liver stiffness (LS), and spleen stiffness (SS) at the beginning and end of dose titration. LS and SS were measured using acoustic radiation force impulse imaging. A prediction model for hemodynamic response was derived, and subject to an external validation in the validation cohort (63 patients). Results: Hemodynamic response occurred in 59 patients (55.7%) in the derivation cohort, and in 33 patients (52.4%) in the validation cohort, respectively. Multivariate logistic regression analysis identified that ΔSS was the only significant predictor of hemodynamic response (odds ratio 0.039; 95% confidence interval 0.008–0.135; p <0.0001). The response prediction model (Model ΔSS = 0.0490–2.8345 × ΔSS; score = (exp[Model ΔSS ])/(1 + exp[Model ΔSS ]) showed good predictive performance (area under the receiver-operating characteristic curve [AUC] = 0.803) using 0.530 as the threshold value. The predictive performance of the Model ΔSS in the validation set improved using the same threshold value (AUC = 0.848). Conclusion: A new model based on dynamic changes in SS exhibited good performance in predicting hemodynamic response to NSBB prophylaxis in patients with high-risk esophageal varices. Lay summary: Non-selective beta-blockers are the mainstay of primary prophylaxis to prevent variceal bleeding in patients with cirrhosis and high-risk esophageal varices. This prospective study showed that a prediction model based on changes in spleen stiffness before vs. after dose titration might be a non-invasive marker for response to prophylactic non-selective beta-blocker (carvedilol) therapy in patients with cirrhosis and high-risk esophageal varices. ClinicalTrials.gov Identifier: NCT01943318.

Original languageEnglish
Pages (from-to)412-422
Number of pages11
JournalJournal of Hepatology
Volume70
Issue number3
DOIs
Publication statusPublished - 2019 Mar

Fingerprint

Esophageal and Gastric Varices
Spleen
Hemodynamics
Therapeutics
Area Under Curve
Liver
Fibrosis
carvedilol
Prospective Studies
Hemorrhage
Elasticity Imaging Techniques
Venous Pressure
Portal Hypertension
ROC Curve
Liver Cirrhosis
Cohort Studies
Logistic Models
Odds Ratio
Regression Analysis
Confidence Intervals

All Science Journal Classification (ASJC) codes

  • Hepatology

Cite this

Kim, Hwi Young ; So, Young Ho ; Kim, Won ; Ahn, Dong Won ; Jung, Yong Jin ; Woo, Hyunsik ; Kim, Donghee ; Kim, Moon Young ; Baik, Soon Koo. / Non-invasive response prediction in prophylactic carvedilol therapy for cirrhotic patients with esophageal varices. In: Journal of Hepatology. 2019 ; Vol. 70, No. 3. pp. 412-422.
@article{156eede8751b4ff7a8a869fa6ede2725,
title = "Non-invasive response prediction in prophylactic carvedilol therapy for cirrhotic patients with esophageal varices",
abstract = "Background & Aims: Non-selective beta-blockers (NSBBs) are the mainstay of primary prophylaxis of esophageal variceal bleeding in patients with liver cirrhosis. We investigated whether non-invasive markers of portal hypertension correlate with hemodynamic responses to NSBBs in cirrhotic patients with esophageal varices. Methods: In this prospective cohort study, 106 cirrhotic patients with high-risk esophageal varices in the derivation cohort received carvedilol prophylaxis, and completed paired measurements of hepatic venous pressure gradient, liver stiffness (LS), and spleen stiffness (SS) at the beginning and end of dose titration. LS and SS were measured using acoustic radiation force impulse imaging. A prediction model for hemodynamic response was derived, and subject to an external validation in the validation cohort (63 patients). Results: Hemodynamic response occurred in 59 patients (55.7{\%}) in the derivation cohort, and in 33 patients (52.4{\%}) in the validation cohort, respectively. Multivariate logistic regression analysis identified that ΔSS was the only significant predictor of hemodynamic response (odds ratio 0.039; 95{\%} confidence interval 0.008–0.135; p <0.0001). The response prediction model (Model ΔSS = 0.0490–2.8345 × ΔSS; score = (exp[Model ΔSS ])/(1 + exp[Model ΔSS ]) showed good predictive performance (area under the receiver-operating characteristic curve [AUC] = 0.803) using 0.530 as the threshold value. The predictive performance of the Model ΔSS in the validation set improved using the same threshold value (AUC = 0.848). Conclusion: A new model based on dynamic changes in SS exhibited good performance in predicting hemodynamic response to NSBB prophylaxis in patients with high-risk esophageal varices. Lay summary: Non-selective beta-blockers are the mainstay of primary prophylaxis to prevent variceal bleeding in patients with cirrhosis and high-risk esophageal varices. This prospective study showed that a prediction model based on changes in spleen stiffness before vs. after dose titration might be a non-invasive marker for response to prophylactic non-selective beta-blocker (carvedilol) therapy in patients with cirrhosis and high-risk esophageal varices. ClinicalTrials.gov Identifier: NCT01943318.",
author = "Kim, {Hwi Young} and So, {Young Ho} and Won Kim and Ahn, {Dong Won} and Jung, {Yong Jin} and Hyunsik Woo and Donghee Kim and Kim, {Moon Young} and Baik, {Soon Koo}",
year = "2019",
month = "3",
doi = "10.1016/j.jhep.2018.10.018",
language = "English",
volume = "70",
pages = "412--422",
journal = "Journal of Hepatology",
issn = "0168-8278",
publisher = "Elsevier",
number = "3",

}

Non-invasive response prediction in prophylactic carvedilol therapy for cirrhotic patients with esophageal varices. / Kim, Hwi Young; So, Young Ho; Kim, Won; Ahn, Dong Won; Jung, Yong Jin; Woo, Hyunsik; Kim, Donghee; Kim, Moon Young; Baik, Soon Koo.

In: Journal of Hepatology, Vol. 70, No. 3, 03.2019, p. 412-422.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Non-invasive response prediction in prophylactic carvedilol therapy for cirrhotic patients with esophageal varices

AU - Kim, Hwi Young

AU - So, Young Ho

AU - Kim, Won

AU - Ahn, Dong Won

AU - Jung, Yong Jin

AU - Woo, Hyunsik

AU - Kim, Donghee

AU - Kim, Moon Young

AU - Baik, Soon Koo

PY - 2019/3

Y1 - 2019/3

N2 - Background & Aims: Non-selective beta-blockers (NSBBs) are the mainstay of primary prophylaxis of esophageal variceal bleeding in patients with liver cirrhosis. We investigated whether non-invasive markers of portal hypertension correlate with hemodynamic responses to NSBBs in cirrhotic patients with esophageal varices. Methods: In this prospective cohort study, 106 cirrhotic patients with high-risk esophageal varices in the derivation cohort received carvedilol prophylaxis, and completed paired measurements of hepatic venous pressure gradient, liver stiffness (LS), and spleen stiffness (SS) at the beginning and end of dose titration. LS and SS were measured using acoustic radiation force impulse imaging. A prediction model for hemodynamic response was derived, and subject to an external validation in the validation cohort (63 patients). Results: Hemodynamic response occurred in 59 patients (55.7%) in the derivation cohort, and in 33 patients (52.4%) in the validation cohort, respectively. Multivariate logistic regression analysis identified that ΔSS was the only significant predictor of hemodynamic response (odds ratio 0.039; 95% confidence interval 0.008–0.135; p <0.0001). The response prediction model (Model ΔSS = 0.0490–2.8345 × ΔSS; score = (exp[Model ΔSS ])/(1 + exp[Model ΔSS ]) showed good predictive performance (area under the receiver-operating characteristic curve [AUC] = 0.803) using 0.530 as the threshold value. The predictive performance of the Model ΔSS in the validation set improved using the same threshold value (AUC = 0.848). Conclusion: A new model based on dynamic changes in SS exhibited good performance in predicting hemodynamic response to NSBB prophylaxis in patients with high-risk esophageal varices. Lay summary: Non-selective beta-blockers are the mainstay of primary prophylaxis to prevent variceal bleeding in patients with cirrhosis and high-risk esophageal varices. This prospective study showed that a prediction model based on changes in spleen stiffness before vs. after dose titration might be a non-invasive marker for response to prophylactic non-selective beta-blocker (carvedilol) therapy in patients with cirrhosis and high-risk esophageal varices. ClinicalTrials.gov Identifier: NCT01943318.

AB - Background & Aims: Non-selective beta-blockers (NSBBs) are the mainstay of primary prophylaxis of esophageal variceal bleeding in patients with liver cirrhosis. We investigated whether non-invasive markers of portal hypertension correlate with hemodynamic responses to NSBBs in cirrhotic patients with esophageal varices. Methods: In this prospective cohort study, 106 cirrhotic patients with high-risk esophageal varices in the derivation cohort received carvedilol prophylaxis, and completed paired measurements of hepatic venous pressure gradient, liver stiffness (LS), and spleen stiffness (SS) at the beginning and end of dose titration. LS and SS were measured using acoustic radiation force impulse imaging. A prediction model for hemodynamic response was derived, and subject to an external validation in the validation cohort (63 patients). Results: Hemodynamic response occurred in 59 patients (55.7%) in the derivation cohort, and in 33 patients (52.4%) in the validation cohort, respectively. Multivariate logistic regression analysis identified that ΔSS was the only significant predictor of hemodynamic response (odds ratio 0.039; 95% confidence interval 0.008–0.135; p <0.0001). The response prediction model (Model ΔSS = 0.0490–2.8345 × ΔSS; score = (exp[Model ΔSS ])/(1 + exp[Model ΔSS ]) showed good predictive performance (area under the receiver-operating characteristic curve [AUC] = 0.803) using 0.530 as the threshold value. The predictive performance of the Model ΔSS in the validation set improved using the same threshold value (AUC = 0.848). Conclusion: A new model based on dynamic changes in SS exhibited good performance in predicting hemodynamic response to NSBB prophylaxis in patients with high-risk esophageal varices. Lay summary: Non-selective beta-blockers are the mainstay of primary prophylaxis to prevent variceal bleeding in patients with cirrhosis and high-risk esophageal varices. This prospective study showed that a prediction model based on changes in spleen stiffness before vs. after dose titration might be a non-invasive marker for response to prophylactic non-selective beta-blocker (carvedilol) therapy in patients with cirrhosis and high-risk esophageal varices. ClinicalTrials.gov Identifier: NCT01943318.

UR - http://www.scopus.com/inward/record.url?scp=85057501740&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057501740&partnerID=8YFLogxK

U2 - 10.1016/j.jhep.2018.10.018

DO - 10.1016/j.jhep.2018.10.018

M3 - Article

C2 - 30389550

AN - SCOPUS:85057501740

VL - 70

SP - 412

EP - 422

JO - Journal of Hepatology

JF - Journal of Hepatology

SN - 0168-8278

IS - 3

ER -