Insufficient accuracy of computed tomography-based portal pressure assessment in hepatitis B virus-related cirrhosis: An analysis of data from CHESS-1601 trial

Abstract

We read with great interest the paper ''Accurate computed tomography-based portal pressure assessment in patients with hepatocellular carcinoma", in which Iranmanesh et al. proposed a computed tomography (CT)-based model to non-invasively predict hepatic venous pressure gradient (HVPG). 1 The noninvasive CT-based model combining the liver/spleen volume ratio with the peri-hepatic ascites demonstrated a high accuracy for predicting HVPG >10 mmHg (area under the receiver operating characteristic curve [AUROC]: 0.911 [0.847-0.975]), with sensitivity, specificity, and positive and negative predictive values of 92%, 79%, 91%, and 81%, respectively. 1 However, the diagnostic performance of the CT-based model (HVPG score) remains to be further externally validated in portal hypertension, with certain etiologies.As one of the global public health issues, hepatitis B virus (HBV) infection affects 20 million people in China, which accounts for more than 70% of patients with cirrhosis nationwide. 2 HVPG is currently the only validated approach to measure portal pressure and to risk stratify patients with HBV-related cirrhosis. However, the measurement is invasive and requires extensive experience. Thus, it fails as a routine examination. 3,4 Non-invasive models that correlate with HVPG are necessary. Thus, we aim to evaluate the performance of the HVPG score established by Iranmanesh et al. 1 in patients with HBV-related cirrhosis. In a pre-planned exploratory analysis of a multicenter prospective trial (CHESS-1601, ClinicalTrials.gov identifier: NCT02842697), a cohort of 131 adult patients with HBV-related cirrhosis was recruited. The male:female ratio was 103:28, with a mean age of 45.5 ± 10.6 years, a mean BMI of 22.68 ± 3.08 kg/m 2 and a mean HVPG of 16.58 ± 5.84 mmHg. In overall analysis, the HVPG score did not correlate well with HVPG (R = 0.048, p = 0.584) ( Fig. 1A), which was also confirmed by Bland-Altman plot (Fig. 1B). For diagnosing clinically significant portal hypertension (CSPH, HVPG >10 mmHg), the CT-based model also showed

Virtual Hepatic Venous Pressure Gradient with CT Angiography (CHESS 1601): A Prospective Multicenter Study for the Noninvasive Diagnosis of Portal Hypertension

“…According to our results, the performance of liver stiffness measured with transient elastography was just fair in our study (AUC for CSPH, 0.71 [95% CI: 0.48, 0.95]), likely because of the limited sample size and wider-ranging causes of cirrhosis compared with previous studies (16). Due to the increasing relevance of extrahepatic factors in portal hypertension progression, HVPG cannot be reliably estimated by means of liver stiffness in severe portal hypertension (6,7,17). Therefore, we suggest that virtual HVPG could serve as an auxiliary parameter for liver stiffness, especially in patients with obesity, liver necrotic inflammation, or severe ascites.…”

“…Another CT-based model, namely the CT-based portal pressure score, was proposed by Iranmanesh and colleagues (13,18) and showed promising results for diagnosing CSPH in patients with hepatocellular carcinoma, with an AUC of 0.911 (95% CI: 0.847, 0.975). However, this model showed insufficient performance in hepatitis B virus-related cirrhosis, with an AUC of 0.57 (95% CI: 42%, 79%) (17). Thus, we must be aware that the cause of cirrhosis may have an influence on the performance of a diagnosis method.…”

Development and validation of a radiomics signature for clinically significant portal hypertension in cirrhosis (CHESS1701): a prospective multicenter study

“…Our HBV-dominant (72·2%, 278/385) cirrhotic cohort might explain this slight fluctuation. Acute exacerbation occurs more frequently in HBV patients, in which the overestimated liver stiffness by FibroScan tends to parallel with the degree of necroinflammation, corroborated by the reports showing significant correlation between the falsely high liver stiffness with increased alanine aminotransferase levels and the decrease of liver stiffness during the recovery phase [ 29 , 30 ]. It is also noted that patients with obesity are associated with unreliable liver stiffness measurement by FibroScan [ 5 , 6 , 27 ], which is not the limitation of r HVPG signature.…”

“…It is also noted that patients with obesity are associated with unreliable liver stiffness measurement by FibroScan [ 5 , 6 , 27 ], which is not the limitation of r HVPG signature. Also, the accuracy of liver stiffness by FibroScan might be hindered by severe portal hypertension because the increasingly relevance of extrahepatic factors such as collateral formation in disease progression [ 30 , 31 ]. Moreover, a recent study has emphasized the disconnection of portal pressure in further progression stage of cirrhosis [ 10 ].…”

Spleen in hepatocellular carcinoma: More complexity and importance than we knew

“…For example, a computed tomographybased model had a high accuracy for prediction of hepatic venous pressure gradient in individuals with alcoholic and HCV dominant cirrhosis, but not for HBV-related cirrhosis in China. 3 Thus, we are concerned that the applicability of SSM in predicting the late recurrence of HCC should not be general-ized to populations with different races/ethnicities and etiologies of HCC without further validation.…”