Evaluation of subclinical ventricular systolic dysfunction assessed using global longitudinal strain in liver cirrhosis: A systematic review, meta-analysis, and meta-regression

Abstract

Global longitudinal strain (GLS) can identify subclinical myocardial dysfunction in patients with cirrhosis. This systematic review aims to provide evidence of a possible difference in GLS values between patients with cirrhosis and patients without cirrhosis. Studies from inception to August 11, 2021, were screened and included based on the inclusion criteria. The Newcastle Ottawa Scale was used to assess the quality of nonrandomized studies. Meta-analyses were conducted with subsequent sensitivity and subgroup analyses according to age, sex, cirrhosis etiology, and severity. Publication bias was evaluated using Begg’s funnel plot, Egger’s test, and rank correlation test with subsequent trim-and-fill analysis. The systematic database search yielded 20 eligible studies. Random effect showed a significant reduction of left ventricular (LV) GLS (MD:-1.43;95%; 95%CI,-2.79 to -0.07; p = 0.04; I2 = 95% p<0.00001) and right ventricular (RV) GLS (MD:-1.95; 95%CI,-3.86 to -0.05, p = 0.04; I2 = 90%, p<0.00001) in the group with cirrhosis. A sensitivity test on subgroup analysis based on the study design showed a -1.78% lower LV-GLS in the group with cirrhosis (I2 = 70%, p = 0.0003). Meta-regression analysis showed that the severity of cirrhosis was significantly related to GLS reduction. This research received no specific grants from any funding agency in the public, commercial, or not-for-profit sectors. The study protocol was registered at PROSPERO (CRD42020201630). We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement guidelines.

The Contemporary Role of Speckle Tracking Echocardiography in Cirrhotic Cardiomyopathy

“…In 19 of these studies reporting GLS values, a total of 1017 patients and 609 controls were included. The investigators found lower absolute GLS values in patients with liver cirrhosis than in controls [Mean Difference −1.43% (−2.79%-0.07%), p = 0.04] [46]. However, this analysis reported increased heterogeneity between studies (I 2 = 0.95, p < 0.00001), did not include all relevant studies [36,47], including a study in non-cirrhotic patients and namely those with chronic hepatitis B [48], and incorporated a study that calculated longitudinal strain with cardiac magnetic resonance [49].…”