磁共振弹性成像
磁共振成像
肝病
医学
弹性成像
瞬态弹性成像
疾病
放射科
核磁共振
病理
内科学
物理
超声波
肝活检
活检
作者
Kathleen E. Corey,Nabih Nakrour,Emily D. Bethea,Jessica E.S. Shay,Karin L. Andersson,Irun Bhan,Lawrence S. Friedman,Avinash Kambadakone,Laura E. Dichtel,Raymond T. Chung,Mukesh G. Harisinghani
摘要
ABSTRACT Background Clinical guidelines emphasise identifying patients at risk of chronic liver disease progression. To avoid biopsy drawbacks, noninvasive imaging tests (NITs) have become part of standard‐of‐care. We assessed the real‐world clinical profile, referral trends, and use of magnetic resonance imaging (MRI)‐based tests, multiparametric MRI (mpMRI) and magnetic resonance elastography (MRE), as part of chronic liver disease management. Methods Patients referred for abdominal imaging as part of standard‐of‐care were eligible for inclusion irrespective of liver aetiology or referral pathway. Liver fibrosis was assessed using MRE and disease severity using mpMRI (disease activity [iron‐corrected T1, cT1], liver fat content [LFC] and iron). T ‐tests were used for group comparisons; Kaplan–Meier analyses for disease progression and area under the receiver operating characteristic (AUC) for diagnostic accuracy. Results Over 18 months, 256 patients (53 years, 51% female, 48% with BMI > 30 kg/m 2 ) were referred for liver imaging. The majority (66%) had steatotic liver disease (SLD). Of those with low MRE (73%) and low FIB‐4 (42%), 36% had elevated cT1 (> 875 ms). Those with MRE > 5 kPa had cT1 > 875 ms. During follow‐up, those with low MRE (< 3.14 kPa) but elevated cT1 (> 800 ms) had significant disease worsening (HR: 3.1, p = 0.0035) compared to all others. In the SLD group, cT1 (AUC: 0.71) outperformed LFC (AUC: 0.64) and MRE (AUC: 0.53) in predicting disease progression. Conclusion Regardless of aetiology, patients with low fibrosis risk (MRE) but high disease activity (cT1) face a three‐times higher risk of progression. Integrating both biomarkers into standard care, especially for SLD, can guide management adjustments.
科研通智能强力驱动
Strongly Powered by AbleSci AI