Agreement of endoscopic ultrasound–guided portal pressure gradient with hepatic venous pressure gradient for portal hypertension severity classification: a systematic review and meta-analysis

Background The severity of portal hypertension (PH) is central to risk stratification in chronic liver disease, yet transjugular HVPG testing is resource-intensive. EUS-guided portal pressure gradient (EUS-PPG) has emerged as an endoscopic alternative, but comparative agreement evidence has not been synthesized for severity classification. Methods We performed a systematic review and meta-analysis registered in PROSPERO [CRD420261325781]. MEDLINE (PubMed), Embase, Web of Science, Scopus, and Cochrane CENTRAL were searched from inception with citation searching. We identified 274 records, removed 104 duplicates, screened 170 titles/abstracts, assessed 20 full texts, and included 4 prospective, non-randomized paired comparative studies. Random-effects models were used. The primary outcome was quantitative agreement, assessed using correlation coefficients, intraclass correlation, and Bland–Altman metrics. Secondary outcomes included technical success and procedure-related adverse events. Risk of bias was assessed using ROBINS-I, and certainty of evidence was rated using GRADE. Results Four prospective, non-randomized comparative studies (73 enrolled participants; 69 paired measurements available for analysis) met the inclusion criteria. EUS-PPG showed a high pooled correlation with HVPG ( r = 0.82, 95% CI: 0.72–0.89; p < 0.00001; I² = 0%), with intraclass correlation coefficients ranging from 0.86 to 0.97 and minimal bias on Bland–Altman analysis. The classification accuracy for clinically significant portal hypertension (CSPH) was 100% in two studies. Technical success rates were equivalent (RR 1.00, 95% CI 0.93–1.07), and no significant safety differences were observed (RR 1.23, 95% CI 0.34–4.45). Subgroup analysis by etiology (cirrhosis vs. mixed/non-cirrhotic) showed no difference in agreement. The certainty of evidence was rated very low due to the study design and imprecision. Conclusions EUS-PPG strongly correlates with transjugular HVPG/TJ-PPG and appears to be as feasible and safe as the latter, according to limited comparative data. Until larger outcome-driven studies validate the threshold-based classification, EUS-PPG should be viewed as a complementary hemodynamic tool to inform selected clinical decisions rather than a universal replacement. Trial registration [ID: CRD420261325781].