Abstract
Objectives
To investigate the value of spectral CT-derived extracellular volume fraction (fECV) in assessing liver functional reserve and predicting cirrhosis-related complications in clinically stable cirrhotic patients.
Materials and methods
This retrospective study enrolled clinically stable patients with cirrhosis who underwent contrast-enhanced spectral CT examinations, as well as a control group without major diseases. Iodine concentrations in the liver and aorta were measured on equilibrium phase images to calculate the fECV. The diagnostic performance of fECV for Child-Pugh classification was evaluated using receiver operating characteristic (ROC) curve analysis. The predictive ability of fECV for cirrhosis-related complications was assessed using the Kaplan-Meier time-dependent ROC curves, and Cox proportional hazards regression models.
Results
A total of 116 patients with cirrhosis (mean age, 60.20 ± 12.11 years) and 38 control subjects (mean age, 60.16 ± 8.93 years) were enrolled in this study. The area under the ROC curve (AUC) of fECV for differentiating Child-Pugh classes was 0.852-0.881 (all p < 0.001). The high fECV group (≥38.48%) demonstrated a significantly higher cumulative risk of complications (p < 0.001). fECV showed superior predictive performance for long-term (3-year) complications (AUC = 0.813, p < 0.001). Furthermore, fECV was identified as an independent risk factor for cirrhosis-related complications (Model 1: HR = 1.073, p = 0.004; Model 2: HR = 1.078, p < 0.001).
Conclusion
fECV is a promising imaging biomarker for liver functional reserve assessment and complication prediction in cirrhosis.
Keywords: Spectral CT, Extracellular volume fraction, Liver cirrhosis
Introduction
Cirrhosis represents the advanced pathological stage of chronic liver disease, characterized by the replacement of normal hepatic architecture with regenerative nodules, ultimately leading to liver failure [1]. It is estimated that over 120 million individuals worldwide are affected by cirrhosis [2], and mortality rates among these patients are projected to continue rising over the next decade [3]. The clinical course of cirrhosis includes compensated and decompensated stages, with the latter significantly impairing patients’ quality of life and survival due to frequent severe complications [4]. Therefore, accurate assessment of hepatic functional reserve is crucial for prognosis prediction and therapeutic decision-making in cirrhotic patients.
Medical imaging plays a vital role in the diagnosis and monitoring of cirrhosis. In recent years, techniques such as ultrasound elastography and magnetic resonance elastography (MRE) have emerged for staging liver fibrosis [5–7]. However, ultrasound elastography is operator-dependent and susceptible to interference from factors like obesity and ascites. MRE is costly, has limited availability, and is contraindicated in patients with certain metallic implants [8]. CT is a more accessible imaging modality used for cirrhosis screening and complication assessment. Nevertheless, conventional CT primarily provides morphological information, creating a need for complementary functional parameters.
The excessive deposition of extracellular matrix (ECM) components, such as collagen, is a central pathological feature of cirrhosis [9]. The extracellular volume fraction (fECV) is a quantitative imaging parameter that reflects the proportion of extracellular space within tissue. Previous studies have confirmed a correlation between the collagen proportional area obtained from liver biopsy and CT-derived fECV. Furthermore, fECV can effectively reveal dynamic changes in the ECM and the degree of hepatic fibrosis [10, 11]. However, few studies have investigated the relationship between fECV and the prognosis of cirrhosis.
Herein, we investigated the diagnostic value of fECV derived from spectral CT iodine maps for the assessment of liver functional reserve as defined by Child-Pugh classification, and further explored the value of fECV in predicting complications of cirrhosis.
Materials and methods
Study population
Imaging and clinical data were retrospectively collected from patients with clinically stable cirrhosis who underwent contrast-enhanced spectral CT examinations at our hospital between May 2021 and September 2023. Clinically stable cirrhosis was defined as patients with cirrhosis who had no hospitalization due to acute decompensation events within 4 weeks before CT examination [12]. Acute decompensation events were defined as follows: (1) new-onset or recurrent grade 2 or grade 3 ascites developing within 2 weeks; (2) initial episode or recurrence of acute hepatic encephalopathy in patients with intact baseline consciousness; (3) acute gastrointestinal bleeding; (4) any type of acute bacterial infection [13]. During the same period, control subjects were randomly selected from patients initially suspected of having a digestive system disease and undergoing enhanced CT, who were subsequently diagnosed without major diseases such as malignancy or chronic liver disease. The exclusion criteria are illustrated in Fig. 1. Ultimately, a total of 116 patients with cirrhosis and 38 control subjects were enrolled in this study.
Fig. 1.
Flowchart of patient inclusion and exclusion criteria
The diagnosis of cirrhosis was confirmed based on imaging findings, histological biopsy, or standard laboratory criteria. The following laboratory test results were collected: platelet count, albumin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin, prothrombin time, estimated glomerular filtration rate (eGFR), creatinine, serum sodium, and hematocrit (Hct). Patients with cirrhosis were stratified into three groups according to the Child-Pugh classification: Child-Pugh A group, Child-Pugh B group, and Child-Pugh C group.
CT scan acquisition
All patients underwent triple-phase contrast-enhanced CT scans of the upper abdomen using a GE Revolution 256-slice spectral CT scanner (GE Healthcare). The scanning range extended from 1 cm above the dome of the right hemidiaphragm to the inferior border of the liver. An iodine-based contrast agent (320 mg I/mL) was administered intravenously via the antecubital vein using a high-pressure injector at a dose of 1.2 mL/kg body weight and a flow rate of 2.0 mL/s. The scanning parameters were set as follows: tube voltage 80/140 kVp (instantaneous switching mode), tube current 405 mA, pitch 0.992:1, rotation speed 0.5 s/rotation, and reconstructed slice thickness 1.25 mm. The scanning delays for the arterial, portal venous, and equilibrium phases were set at 30, 60, and 180 s after contrast injection, respectively.
Data measurement
Equilibrium phase images were transferred to a GE AW 4.7 workstation (GE Healthcare), and iodine maps were generated using the GSI Viewer software. Regions of interest (ROIs) were manually drawn in the left lateral lobe, left medial lobe, right anterior lobe, and right posterior lobe of the liver. Each ROI was a circular area of 100 mm² and was carefully placed to avoid intrahepatic vessels, bile ducts, and focal lesions. An ROI of identical size was placed in the aorta at the corresponding slice level, avoiding aortic thrombi and calcifications (Fig. 2).
Fig. 2.
ROIs were drawn on the iodine maps generated from equilibrium phase images in the (a) left lateral lobe, (b) left medial lobe, (c) right anterior lobe, and (d) right posterior lobe of the liver, as well as in the aorta at their corresponding levels
All contouring and measurements were independently performed by two radiologists with more than 5 years of experience in abdominal CT interpretation. Importantly, both radiologists were strictly kept blinded to the clinical information, Child-Pugh classification, and patient outcomes of all subjects during the delineation and measurement of ROIs, so as to eliminate subjective bias. For the assessment of inter-observer reproducibility, the intraclass correlation coefficient (ICC) was calculated. After confirming acceptable inter-observer agreement, the average of the measurements from the two radiologists was adopted as the final iodine concentration (IC) values for the liver and aorta. For the evaluation of intra-observer reproducibility, each radiologist randomly selected 40 patients and conducted repeated ROI delineation and measurement two months after the initial assessment, with the ICC calculated to quantify the intra-observer agreement. The fECV and normalized iodine concentration (NIC) were calculated using the following formulas:
 |
 |
Follow-up and definition of complications
The follow-up date for patients was set until December 31, 2025. The primary endpoint of this study was the occurrence of cirrhosis-related complications, including ascites, variceal bleeding, overt hepatic encephalopathy, and infection. Although these complications have distinct pathophysiological mechanisms, they collectively indicate the onset of hepatic decompensation. Therefore, this study adopted a composite endpoint to comprehensively assess the overall risk of clinical deterioration in patients with cirrhosis. For patients who already had these complications at baseline, a new complication was defined as either the worsening of an existing condition or the development of a new one. Follow-up for a patient was terminated if they experienced all-cause mortality or were diagnosed with hepatocellular carcinoma during the follow-up period.
The severity of ascites was classified into three grades based on the volume of ascitic fluid [14]. Progression of ascites was defined as an increase in ascites grade or progression from uncomplicated ascites to complicated ascites. Complicated ascites refers to ascites that is diuretic-resistant, ascites that is diuretic-intractable due to diuretic-induced complications that preclude effective diuretic therapy, or ascites requiring more than three therapeutic paracenteses despite optimal medical treatment [15]. Variceal hemorrhage was defined as the presence of active bleeding from esophageal, gastric, or ectopic varices observed during endoscopy, or the identification of stigmata suggesting recent hemorrhage [16]. Overt hepatic encephalopathy was defined as West-Haven criteria grades 2–4 [1, 17]. The diagnostic criterion for spontaneous bacterial peritonitis (SBP) was an ascitic fluid neutrophil count ≥ 250 cells/µl [18]. Other infections apart from SBP (such as urinary tract infections, pneumonia, and soft tissue infections) were diagnosed using the same criteria applied to the general population [19].
Statistical analysis
Continuous variables are expressed as mean ± standard deviation, and categorical variables are presented as numbers and percentages. The inter-observer agreement between the two radiologists and intra-observer reproducibility of each radiologist were assessed using ICC. Normality of data distribution was tested with the Shapiro-Wilk test, and homogeneity of variances was examined using Levene’s test. If data met the assumptions of normal distribution and homogeneity of variances, one-way analysis of variance was employed to compare differences in means among groups; otherwise, the Kruskal-Wallis H test was used. Post-hoc pairwise comparisons were performed using Dunn’s test with Bonferroni adjustment for multiple comparisons. Differences in categorical variables across groups were evaluated using the chi-square test. Spearman’s correlation coefficient was used to assess the relationships between fECV/NIC and Child-Pugh score. Analysis of covariance was used to evaluate the independent effects of ascites and sex on fECV. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the diagnostic efficacy of CT parameters for Child-Pugh classification, and the optimal cutoff value was determined by the Youden index.
The cumulative probability of cirrhosis-related complications was estimated using the Kaplan-Meier method, and differences between groups were compared using the log-rank test. Time-dependent ROC (tROC) curves were applied to evaluate the predictive ability of fECV and Child-Pugh score for complications in patients with cirrhosis. Cox proportional hazards regression models were used to analyze factors associated with the occurrence of cirrhosis complications. Variables found to be significant (p < 0.20) in univariate analysis were included in the multivariate analysis to identify independent predictors.
Statistical analyses were performed using SPSS 29.0 (IBM) and R software (R Foundation for Statistical Computing). Statistical significance was set at p < 0.05.
Results
Patient characteristics
A total of 154 patients were enrolled in this study, including 38 in the control group, 52 in the Child-Pugh A group, 48 in the Child-Pugh B group, and 16 in the Child-Pugh C group. Statistically significant differences (all p < 0.05) were observed among the groups in laboratory parameters, including AST, ALT, albumin, bilirubin, INR, platelet count, and serum sodium. The detailed clinical characteristics and laboratory profiles of the cirrhosis patients and the control group are presented in Table 1.
Table 1.
Characteristics of patients with cirrhosis stratified by Child-Pugh class
| Variable | Controls (n = 38) |
Child-Pugh A (n = 52) | Child-Pugh B (n = 48) | Child-Pugh C (n = 16) | p value |
|---|---|---|---|---|---|
| Demographic and clinical variables | |||||
| Age (years) | 60.16 (8.93) | 60.31 (12.53) | 61.38 (12.01) | 56.31 (10.84) | 0.500 |
| Sex | 0.027 | ||||
| Male | 17 (44.74) | 34 (65.38) | 34 (70.83) | 13 (81.25) | |
| Female | 21 (55.26) | 18 (34.62) | 14 (29.17) | 3 (18.75) | |
| BMI (kg/m²) | 24.48 (2.10) | 24.21 (3.34) | 23.84 (3.34) | 23.91 (3.14) | 0.411 |
| Hypertension | 9 (23.68) | 20 (38.46) | 12 (25.00) | 2 (12.50) | 0.151 |
| Diabetes Mellitus | 7 (18.42) | 8 (15.38) | 12 (25.00) | 3 (18.75) | 0.677 |
| Alcohol-related | 6 (15.79) | 13 (25.00) | 21 (43.75) | 7 (43.75) | 0.019 |
| Hepatitis B | 0 (0.00) | 23 (44.23) | 16 (33.33) | 6 (37.50) | < 0.001 |
| Ascites | 0 (0.00) | 2 (3.85) | 35 (72.92) | 15 (93.75) | < 0.001 |
| Laboratory parameters | |||||
| AST (U/L) | 20.61 (6.05) | 38.78 (59.11) | 62.26 (107.23) | 81.85 (64.07) | < 0.001 |
| ALT (U/L) | 19.74 (9.67) | 44.78 (127.02) | 46.68 (107.03) | 79.47 (123.56) | 0.004 |
| Albumin (g/L) | 41.38 (4.74) | 39.00 (5.25) | 30.51 (5.80) | 25.69 (3.15) | < 0.001 |
| Bilirubin (µmol/L) | 11.56 (3.23) | 16.51 (8.02) | 37.95 (49.86) | 76.63 (60.95) | < 0.001 |
| INR | 0.87 (0.08) | 0.99 (0.15) | 1.14 (0.16) | 1.49 (0.33) | < 0.001 |
| Platelets (×10⁹/L) | 227.21 (57.85) | 157.56 (101.20) | 120.48 (99.45) | 76.62 (52.92) | < 0.001 |
| eGFR (mL/min/1.73 m2) | 97.99 (6.68) | 95.96 (20.27) | 95.97 (23.89) | 93.75 (18.74) | 0.750 |
| Creatinine (µmol/L) | 64.60 (14.25) | 67.91 (23.13) | 67.74 (37.01) | 68.63 (28.85) | 0.462 |
| Sodium (mmol/L) | 140.55 (1.96) | 139.96 (2.42) | 139.28 (3.77) | 137.43 (6.11) | 0.007 |
| CT imaging parameters | |||||
| fECV (%) | 28.95 (3.94) | 33.04 (5.39)* | 39.73 (5.51)* | 46.91 (5.89)* | < 0.001 |
| NIC | 48.67 (5.68) | 53.89 (8.81)* | 59.10 (7.96)* | 70.39 (9.88)* | < 0.001 |
Continuous variables were compared using one‑way analysis of variance when assumptions of normality and homogeneity of variance were satisfied, or the Kruskal‑Wallis H test otherwise, followed by Dunn’s post‑hoc test with Bonferroni correction for multiple comparisons. Categorical variables were assessed using the chi‑square test
* indicates a statistically significant difference compared to the preceding group in post-hoc pairwise comparisons
BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio; eGFR, estimated glomerular filtration rate; fECV, extracellular volume fraction; NIC, normalized iodine concentration
Group differences in fECV and NIC among Child-Pugh classes and their correlations with child-pugh score
The inter-observer ICCs for CT measurement parameters ranged from 0.885 (0.845–0.915) to 0.924 (0.897–0.944), and the intra-observer ICCs ranged from 0.931 (0.873–0.963) to 0.965 (0.922–0.983). These results demonstrated excellent inter- and intra-observer agreement for the CT measurement parameters.
The fECV values were 28.95% ± 3.94%, 33.04% ± 5.39%, 39.73% ± 5.51%, and 46.91% ± 5.89% for the control, Child-Pugh A, Child-Pugh B, and Child-Pugh C groups, respectively. The Kruskal-Wallis H test indicated a statistically significant difference in fECV among the groups (p < 0.001). Subsequent post-hoc pairwise comparisons using the Dunn’s test with Bonferroni adjustment revealed statistically significant differences between Child-Pugh A, B, and C groups compared to their preceding groups (p = 0.012, p < 0.001, and p = 0.045, respectively). These results suggest an increasing trend of fECV with the progression of Child-Pugh classification. Similar results were obtained from the statistical analysis of NIC, as detailed in Table 1.
Furthermore, Spearman’s correlation analysis showed a significant positive correlation between fECV and Child-Pugh score (r = 0.669, p < 0.001), while NIC exhibited a relatively weaker positive correlation with the Child-Pugh score (r = 0.488, p < 0.001) (Fig. 3).
Fig. 3.
Spearman correlation coefficients of (a) fECV and (b) NIC with Child-Pugh score. fECV, extracellular volume fraction; NIC, normalized iodine concentration
To exclude potential confounding factors, we performed analysis of covariance to determine whether ascites or sex independently affects fECV. After adjusting for Child-Pugh classification, neither ascites (p = 0.347) nor sex (p = 0.655) showed a significant independent association with fECV.
Diagnostic performance of fECV and NIC for child-pugh classification in cirrhosis
The diagnostic performance of fECV and NIC for Child-Pugh classification in cirrhosis was assessed using ROC curve analysis (Fig. 4). In distinguishing the control group from patients with cirrhosis, fECV demonstrated strong diagnostic efficacy (AUC = 0.868, p < 0.001), which was superior to that of NIC (AUC = 0.807, p < 0.001). The diagnostic performance of fECV and NIC was further compared in detail for differentiating between each pair of adjacent cirrhosis groups. For differentiating Child-Pugh A from B, the AUCs for fECV and NIC were 0.852 (p < 0.001) and 0.723 (p < 0.001), respectively. For differentiating Child-Pugh B from C, the AUCs for fECV and NIC were 0.871 (p < 0.001) and 0.803 (p < 0.001), respectively. Furthermore, in distinguishing Child-Pugh A from B + C, the AUCs for fECV and NIC were 0.881 (p < 0.001) and 0.771 (p < 0.001), respectively. Overall, fECV exhibited better diagnostic performance than NIC for Child-Pugh classification in cirrhosis. The detailed diagnostic performance parameters are presented in Table 2.
Fig. 4.
ROC curve analysis for fECV and NIC in differentiating (a) control vs. cirrhosis, (b) Child-Pugh A vs. B, (c) Child-Pugh B vs. C, and (d) Child-Pugh A vs. B + C. ROC, receiver operating characteristic; fECV, extracellular volume fraction; NIC, normalized iodine concentration
Table 2.
Diagnostic performance of fECV and NIC for Child-Pugh classification in cirrhosis
| AUC | p value | Cutoff value | Sensitivity (%) | Specificity (%) | ||
|---|---|---|---|---|---|---|
| Control vs. Cirrhosis | fECV | 0.868 (0.811–0.926) | < 0.001 | 35.04 | 62.93 (53.86–71.17) | 94.74 (82.71–99.06) |
| NIC | 0.807 (0.737–0.877) | < 0.001 | 52.49 | 75.86 (67.33–82.74) | 76.32 (60.79–87.01) | |
| Child-Pugh A vs. B | fECV | 0.852 (0.774–0.930) | < 0.001 | 34.94 | 91.67 (80.45–96.71) | 73.08 (59.75–83.23) |
| NIC | 0.723 (0.620–0.826) | < 0.001 | 56.56 | 70.83 (56.82–81.76) | 73.08 (59.75–83.23) | |
| Child-Pugh B vs. C | fECV | 0.871 (0.785–0.957) | < 0.001 | 42.15 | 93.75 (71.67–99.68) | 81.25 (68.06–89.81) |
| NIC | 0.803 (0.674–0.933) | < 0.001 | 63.14 | 75.00 (50.50−89.82) | 79.17 (65.74–88.27) | |
| Child-Pugh A vs. B + C | fECV | 0.881 (0.815–0.947) | < 0.001 | 36.47 | 84.38 (73.57–91.29) | 82.69 (70.27–90.62) |
| NIC | 0.771 (0.683–0.859) | < 0.001 | 56.83 | 75.00 (63.18–83.99) | 75.00 (61.79–84.77) |
fECV, extracellular volume fraction; NIC, normalized iodine concentration
Ability of fECV to predict complications of cirrhosis
The median follow-up time for the 116 patients with cirrhosis was 636 days. During follow-up, 39 patients (33.62%) experienced new or worsening cirrhosis-related complications, including ascites (n = 17, 14.66%), variceal bleeding (n = 7, 6.03%), overt hepatic encephalopathy (n = 4, 3.45%), and infection (n = 11, 9.48%).
First, the cumulative probability of developing cirrhosis-related complications was evaluated using Kaplan-Meier curves (Fig. 5). The high fECV group (≥ 38.48%) had a significantly higher cumulative complication probability than the low fECV group (< 38.48%) (p < 0.001), indicating that elevated fECV was significantly associated with an increased risk of complications in cirrhotic patients.
Fig. 5.
Kaplan-Meier curves for cirrhosis-related complications. fECV, extracellular volume fraction
Subsequently, tROC curve analysis was used to assess the predictive ability of fECV and Child-Pugh score for complications at 6 months, 1 year, and 3 years (Fig. 6). fECV (AUC = 0.755, p < 0.001) and Child-Pugh score (AUC = 0.747, p = 0.001) showed comparable predictive performance for 6-month complications. Similar results were observed for the prediction of 1-year complications (AUC = 0.748, 0.738, respectively; p < 0.001 for both). Notably, fECV demonstrated superior predictive performance for 3-year complications compared to Child-Pugh score (AUC = 0.813, 0.689, respectively; p < 0.001, p = 0.016, respectively). These results indicate that fECV and Child-Pugh score have comparable efficacy in predicting short-term (≤ 1 year) complications, while fECV shows greater advantage in long-term (3-year) risk assessment for cirrhosis.
Fig. 6.
tROC curve analysis evaluating the predictive performance of fECV and Child-Pugh classification for complications in patients with cirrhosis at (a) 6 months, (b) 1 year, and (c) 3 years. tROC, time-dependent receiver operating characteristic; fECV, extracellular volume fraction
Finally, univariate Cox regression analysis identified age, diabetes, albumin, eGFR, ascites, fECV, and Child-Pugh class as significant prognostic factors for cirrhosis complications (all p < 0.20). In multivariate Cox regression analysis, Model 1 (including Child-Pugh class) identified diabetes (HR = 2.461, p = 0.021) and fECV (HR = 1.073, p = 0.004) as independent risk factors. In Model 2 (including components of the Child-Pugh class), diabetes (HR = 2.371, p = 0.031) and fECV (HR = 1.078, p < 0.001) remained significant (Table 3). These results indicate that fECV is an independent prognostic factor for cirrhosis complications. For every 1% increase in fECV, the complication risk increased by 7.3% to 7.8%. The predictive value of fECV remained stable in multivariate models and was not confounded by the Child-Pugh class or its components.
Table 3.
Univariate and multivariable Cox regression analyses of cirrhosis-related complications
| Variable | Univariable analysis | Multivariable analysis for Model 1 | Multivariable analysis for Model 2 | |||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | |
| Age | 1.022 (0.994–1.051) | 0.124 | 1.018 (0.984–1.053) | 0.301 | 1.018 (0.985–1.053) | 0.289 |
| Sex | 1.255 (0.651–2.417) | 0.498 | ||||
| BMI | 1.010 (0.924–1.105) | 0.823 | ||||
| Hypertension | 1.214 (0.612–2.406) | 0.579 | ||||
| Diabetes Mellitus | 1.857 (0.922–3.739) | 0.083 | 2.461 (1.142–5.301) | 0.021 | 2.371 (1.081–5.202) | 0.031 |
| Alcohol-related | 0.976 (0.507–1.879) | 0.943 | ||||
| Hepatitis B | 1.034 (0.537–1.992) | 0.919 | ||||
| AST | 1.000 (0.996–1.004) | 0.967 | ||||
| ALT | 0.999 (0.995–1.004) | 0.808 | ||||
| Albumin | 0.915 (0.866–0.967) | 0.002 | 0.964 (0.912–1.019) | 0.191 | ||
| Bilirubin | 1.001 (0.993–1.008) | 0.872 | ||||
| INR | 1.619 (0.549–4.777) | 0.383 | ||||
| Platelets | 1.001 (0.998–1.004) | 0.512 | ||||
| eGFR | 0.989 (0.976–1.002) | 0.108 | 1.002 (0.986–1.018) | 0.815 | 1.001 (0.985–1.018) | 0.906 |
| Creatinine | 1.006 (0.997–1.015) | 0.226 | ||||
| Sodium | 0.948 (0.864–1.040) | 0.257 | ||||
| Ascites | 3.370 (1.641–6.920) | < 0.001 | 1.509 (0.653–3.487) | 0.336 | ||
| fECV | 1.093 (1.056–1.132) | < 0.001 | 1.073 (1.022–1.126) | 0.004 | 1.078 (1.033–1.125) | < 0.001 |
| Child-Pugh class | ||||||
| A | Ref. | Ref. | ||||
| B | 4.081 (1.653–10.076) | 0.002 | 2.303 (0.876–6.056) | 0.091 | ||
| C | 7.236 (2.668–19.622) | < 0.001 | 3.031 (0.878–10.460) | 0.079 | ||
For multivariate Cox regression analyses, Model 1 incorporated Child-Pugh class but excluded its components; Model 2 incorporated the components of Child-Pugh class
BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; INR, international normalized ratio; eGFR, estimated glomerular filtration rate; fECV, extracellular volume fraction
Discussion
This study confirmed the correlation between spectral CT-derived fECV and Child-Pugh classification. Prognostic analysis revealed that an elevated fECV is significantly associated with an increased risk of complications in cirrhotic patients and serves as an independent predictor of adverse outcomes in cirrhosis.
ECM constitutes an intricate and dynamic network that provides physical scaffolding for cells. Its primary components include collagen, glycoproteins, and proteoglycans [20]. When the liver suffers prolonged injury, hepatic stellate cells become activated, transforming into myofibroblasts and promoting the synthesis of ECM components such as type I and III collagen [21, 22]. Excessive deposition of the ECM distorts the normal liver architecture, leading to fibrotic scarring and ultimately resulting in cirrhosis [23]. Furthermore, relevant studies have confirmed that abnormal ECM deposition is closely associated with the prognosis of patients with cirrhosis [24].
Spectral CT performs scans using X-rays at two different energy levels, separating signals from specific materials (such as iodine) by leveraging their differential X-ray attenuation properties [25]. The fECV value reflects the relative volume of the extracellular space by precisely quantifying the iodine concentration retained within this interstitial compartment [26]. When pathological proliferation of the ECM causes expansion of this space, the fECV value increases accordingly. Therefore, fECV obtained from spectral CT serves as a promising imaging biomarker for quantitatively assessing the hepatic ECM.
Early studies have already established the value of fECV in grading liver fibrosis. For instance, Morita et al. [27] found that ECV values measured by spectral CT significantly increased with the progression of liver fibrosis stages. Cirrhosis, as the advanced pathological stage of liver fibrosis, is characterized by a further exacerbation of excessive ECM deposition based on fibrosis, accompanied by more complex hemodynamic and functional remodeling. In this work, we demonstrated a positive correlation between fECV and Child-Pugh score in cirrhosis (r = 0.669). Furthermore, fECV effectively distinguished between Child-Pugh A and B, as well as between Child-Pugh B and C, with AUC of 0.852 and 0.871, respectively, indicating superior diagnostic performance compared to NIC. Similarly, Mesropyan et al. [28] confirmed that ECV values measured by MRI correlated with Child-Pugh score in cirrhosis (r = 0.64), and its diagnostic performance in differentiating various Child-Pugh classes (AUC = 0.785–0.944) was higher than that of native T1. It can be inferred that fECV, by measuring the proportion of extracellular space relative to the total tissue volume, reflects the extent of liver structural remodeling, thereby quantifying the severity of cirrhosis.
Nevertheless, ascites may alter systemic hemodynamics and contrast distribution, thereby interfering with the quantitative measurement of fECV. However, the standardized fECV calculation itself effectively avoids such interference: equilibrium-phase imaging ensures stable extracellular contrast distribution, while normalization to aortic iodine concentration corrects systemic hemodynamic fluctuations [29, 30]. These measures effectively isolate the true signal of hepatic extracellular matrix deposition. In this study, we confirmed via analysis of covariance that ascites had no independent effect on fECV after adjusting for cirrhosis severity. This validates fECV as a robust biomarker unaffected by ascites-related interference.
This study further confirms the value of fECV in predicting cirrhosis-related complications. Survival analysis demonstrated a higher cumulative probability of complications in the high fECV group (≥ 38.48%). Cox regression analysis identified fECV as an independent predictor for the occurrence of cirrhosis-related complications, with every 1% increase in fECV corresponding to a 7.3%–7.8% increase in complication risk. Previous studies have also proposed CT-based prognostic markers for cirrhosis. For instance, CT-measured liver volume, spleen volume, and their ratios have been significantly associated with the prognosis of cirrhotic patients [31]. However, these parameters primarily reflect macroscopic morphological changes in organs and are limited in their ability to precisely capture alterations in the hepatic microenvironment. In contrast, the abnormal deposition of ECM quantified by fECV forms the direct structural basis for increased intrahepatic vascular resistance and portal hypertension, thereby elevating the risk of complications such as ascites and variceal hemorrhage [32]. This pathological foundation extends to the clinical manifestations of hepatic decompensation. On one hand, a reduction in functional hepatocyte mass and portosystemic shunting impairs ammonia clearance. Ammonia crossing the blood-brain barrier exerts various neurotoxic effects, which can subsequently trigger hepatic encephalopathy [4]. On the other hand, portal hypertension leads to intestinal congestion, compromised mucosal barrier integrity, ascites, and reduced immunity, collectively promoting the occurrence of infections [4, 33]. Furthermore, this study found that fECV exhibited superior predictive performance for long-term (3-year) complications compared to the Child-Pugh score. This is likely because the Child-Pugh score primarily reflects the current state of hepatic functional reserve, whereas fECV captures the cumulative effect of pathological changes within the liver parenchyma, rendering it more suitable for predicting long-term complications in cirrhosis.
This study has several limitations. First, its single-center, retrospective design may introduce selection bias. Further multicenter prospective studies are warranted in the future to validate the generalizability of the present findings across different scanners and populations. Second, the enrollment of patients with clinically stable cirrhosis resulted in a relatively small sample size for the Child-Pugh C group, which might affect statistical power. Third, the prognostic performance of fECV was only compared with the Child-Pugh score, and not with established scoring systems such as MELD and ALBI. Finally, in routine clinical practice, many patients with clinically diagnosed cirrhosis who present typical imaging features and consistent laboratory indicators are not suitable for invasive puncture biopsy. Accordingly, only a subset of patients in this study received a definitive diagnosis confirmed by histopathological examination.
In conclusion, this study demonstrates that the fECV derived from spectral CT serves as a promising imaging biomarker for assessing hepatic functional reserve in patients with clinically stable cirrhosis. fECV exhibits considerable value in predicting the risk of cirrhosis-related complications. Incorporating this quantitative parameter, which reflects microscopic structural alterations within the liver, into risk assessment frameworks holds promise for achieving earlier and more precise risk stratification in cirrhosis.
Acknowledgements
Not applicable.
Abbreviations
- fECV
Extracellular volume fraction
- ROC
Receiver operating characteristic
- AUC
Area under the curve
- MRE
Magnetic resonance elastography
- ECM
Extracellular matrix
- AST
Aspartate aminotransferase
- ALT
Alanine aminotransferase
- eGFR
Estimated glomerular filtration rate
- Hct
Hematocrit
- ROIs
Regions of interest
- IC
Iodine concentration
- NIC
Normalized iodine concentration
- SBP
Spontaneous bacterial peritonitis
- ICC
Intraclass correlation coefficient
- tROC
Time-dependent receiver operating characteristic
- BMI
Body mass index
- INR
International normalized ratio
Author contributions
LZ: Data curation, Formal analysis, Methodology, Writing – original draft; YS: Software, Visualization; XZ: Visualization, Resources; ZZ: Investigation, Validation, Writing – review and editing; YJ: Conceptualization, Supervision, Writing – review and editing. All authors read and approved the final manuscript.
Funding
The authors state that this work has not received any funding.
Data availability
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and was approved by the Ethics Committee of Qingdao Municipal Hospital (2025-KY-180). Informed consent was waived by the committee due to the use of anonymized historical data with full protection of patient privacy.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Zhen Zhang, Email: zhangzhen202688@163.com.
Yang Ji, Email: jiyang202602@163.com.
References
- 1.Ginès P, Krag A, Abraldes JG, Solà E, Fabrellas N, Kamath PS. Liver cirrhosis. Lancet. 2021;398:1359–76. 10.1016/S0140-6736(21)01374-X. [DOI] [PubMed] [Google Scholar]
- 2.Devarbhavi H, Asrani SK, Arab JP, Nartey YA, Pose E, Kamath PS. Global burden of liver disease: 2023 update. J Hepatol. 2023;79:516–37. 10.1016/j.jhep.2023.03.017. [DOI] [PubMed] [Google Scholar]
- 3.Huang DQ, Terrault NA, Tacke F, Gluud LL, Arrese M, Bugianesi E, Loomba R. Global epidemiology of cirrhosis - aetiology, trends and predictions. Nat Rev Gastroenterol Hepatol. 2023;20:388–98. 10.1038/s41575-023-00759-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fadlallah H, El Masri D, Bahmad HF, Abou-Kheir W, El Masri J. Update on the Complications and Management of Liver Cirrhosis. Med Sci (Basel). 2025;13:13. 10.3390/medsci13010013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ozturk A, Olson MC, Samir AE, Venkatesh SK. Liver fibrosis assessment: MR and US elastography. Abdom Radiol (NY). 2022;47:3037–50. 10.1007/s00261-021-03269-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shi Y, Qi YF, Lan GY, Wu Q, Ma B, Zhang XY, Ji RY, Ma YJ, Hong Y. Three-dimensional MR Elastography Depicts Liver Inflammation, Fibrosis, and Portal Hypertension in Chronic Hepatitis B or C. Radiology. 2021;301:154–62. 10.1148/radiol.2021202804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Idilman IS, Li J, Yin M, Venkatesh SK. MR elastography of liver: current status and future perspectives. Abdom Radiol (NY). 2020;45:3444–62. 10.1007/s00261-020-02656-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dajti E, Huber AT, Ferraioli G, Berzigotti A. Advances in imaging-Elastography. Hepatology. 2025. 10.1097/HEP.0000000000001342. ahead of print. [DOI] [PubMed] [Google Scholar]
- 9.Parola M, Pinzani M. Liver fibrosis in NAFLD/NASH: from pathophysiology towards diagnostic and therapeutic strategies. Mol Aspects Med. 2024;95:101231. 10.1016/j.mam.2023.101231. [DOI] [PubMed] [Google Scholar]
- 10.Bandula S, Punwani S, Rosenberg WM, Jalan R, Hall AR, Dhillon A, Moon JC, Taylor SA. Equilibrium contrast-enhanced CT imaging to evaluate hepatic fibrosis: initial validation by comparison with histopathologic sampling. Radiology. 2015;275:136–43. 10.1148/radiol.14141435. [DOI] [PubMed] [Google Scholar]
- 11.Salahshour F, Abkhoo A, Sadeghian S, Safaei M. Reliability assessment of CT enhancement rate and extracellular volume in liver fibrosis prediction. BMC Gastroenterol. 2025;25:101. 10.1186/s12876-025-03678-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tranah TH, Ballester MP, Carbonell-Asins JA, Ampuero J, Alexandrino G, Caracostea A, Sánchez-Torrijos Y, Thomsen KL, Kerbert AJC, Capilla-Lozano M, Romero-Gómez M, Escudero-García D, Montoliu C, Jalan R, Shawcross DL. Plasma ammonia levels predict hospitalisation with liver-related complications and mortality in clinically stable outpatients with cirrhosis. J Hepatol. 2022;77:1554–63. 10.1016/j.jhep.2022.07.014. [DOI] [PubMed] [Google Scholar]
- 13.D’Amico G, Bernardi M, Angeli P. Towards a new definition of decompensated cirrhosis. J Hepatol. 2022;76:202–7. 10.1016/j.jhep.2021.06.018. [DOI] [PubMed] [Google Scholar]
- 14.Moore KP, Wong F, Gines P, Bernardi M, Ochs A, Salerno F, Angeli P, Porayko M, Moreau R, Garcia-Tsao G, Jimenez W, Planas R, Arroyo V. The management of ascites in cirrhosis: report on the consensus conference of the International Ascites Club. Hepatology. 2003;38:258–66. 10.1053/jhep.2003.50315. [DOI] [PubMed] [Google Scholar]
- 15.Arroyo V, Ginès P, Gerbes AL, Dudley FJ, Gentilini P, Laffi G, Reynolds TB, Ring-Larsen H, Schölmerich J. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. Int Ascites Club Hepatol. 1996;23:164–76. 10.1002/hep.510230122. [DOI] [PubMed] [Google Scholar]
- 16.Lesmana CRA, Shukla A, Kumar A, Shalimar, Qi X, Gani RA, Zhuang ZH, et al. Management of acute variceal bleeding: updated APASL guidelines. Hepatol Int. 2025;19:1003–31. 10.1007/s12072-025-10894-4. [DOI] [PubMed] [Google Scholar]
- 17.American Association for the Study of Liver Diseases; European Association for the Study of the Liver. Hepatic encephalopathy in chronic liver disease: 2014 practice guideline by the European Association for the Study of the Liver and the American Association for the Study of Liver Diseases. J Hepatol. 2014;61:642–59. 10.1016/j.jhep.2014.05.042. [DOI] [PubMed] [Google Scholar]
- 18.Rimola A, García-Tsao G, Navasa M, Piddock LJ, Planas R, Bernard B, Inadomi JM. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. Int Ascites Club J Hepatol. 2000;32:142–53. 10.1016/s0168-8278(00)80201-9. [DOI] [PubMed] [Google Scholar]
- 19.Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control. 2008;36:309–32. 10.1016/j.ajic.2008.03.002. [DOI] [PubMed] [Google Scholar]
- 20.Arteel GE. Hepatic Extracellular Matrix and Its Role in the Regulation of Liver Phenotype. Semin Liver Dis. 2024;44:343–55. 10.1055/a-2404-7973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Friedman SL. Hepatic stellate cells: protean, multifunctional, and enigmatic cells of the liver. Physiol Rev. 2008;88:125–72. 10.1152/physrev.00013.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Iwaisako K, Brenner DA, Kisseleva T. What’s new in liver fibrosis? The origin of myofibroblasts in liver fibrosis. J Gastroenterol Hepatol. 2012;27(Suppl 2):65–8. 10.1111/j.1440-1746.2011.07002.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Arriazu E, Ruiz de Galarreta M, Cubero FJ, Varela-Rey M, Pérez de Obanos MP, Leung TM, Lopategi A, Benedicto A, Abraham-Enachescu I, Nieto N. Extracellular matrix and liver disease. Antioxid Redox Signal. 2014;21:1078–97. 10.1089/ars.2013.5697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wu Y, Cao Y, Xu K, Zhu Y, Qiao Y, Wu Y, Chen J, Li C, Zeng R, Ge G. Dynamically remodeled hepatic extracellular matrix predicts prognosis of early-stage cirrhosis. Cell Death Dis. 2021;12:163. 10.1038/s41419-021-03443-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Greffier J, Villani N, Defez D, Dabli D, Si-Mohamed S. Spectral CT imaging: Technical principles of dual-energy CT and multi-energy photon-counting CT. Diagn Interv Imaging. 2023;104:167–77. 10.1016/j.diii.2022.11.003. [DOI] [PubMed] [Google Scholar]
- 26.Zheng T, Qu Y, Chen J, Yang J, Yan H, Jiang H, Song B. Noninvasive diagnosis of liver cirrhosis: qualitative and quantitative imaging biomarkers. Abdom Radiol (NY). 2024;49:2098–115. 10.1007/s00261-024-04225-8. [DOI] [PubMed] [Google Scholar]
- 27.Morita K, Nishie A, Ushijima Y, Takayama Y, Fujita N, Kubo Y, Ishimatsu K, Yoshizumi T, Maehara J, Ishigami K. Noninvasive assessment of liver fibrosis by dual-layer spectral detector CT. Eur J Radiol. 2021;136:109575. 10.1016/j.ejrad.2021.109575. [DOI] [PubMed] [Google Scholar]
- 28.Mesropyan N, Kupczyk PA, Dold L, Praktiknjo M, Chang J, Isaak A, Endler C, Kravchenko D, Bischoff LM, Sprinkart AM, Pieper CC, Kuetting D, Jansen C, Attenberger UI, Luetkens JA. Assessment of liver cirrhosis severity with extracellular volume fraction MRI. Sci Rep. 2022;12:9422. 10.1038/s41598-022-13340-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Zopfs D, Reimer RP, Sonnabend K, Rinneburger M, Hentschke CM, Persigehl T, Lennartz S. Große Hokamp N. Intraindividual Consistency of Iodine Concentration in Dual-Energy Computed Tomography of the Chest and Abdomen. Invest Radiol. 2021;56:181–7. 10.1097/RLI.0000000000000724. [DOI] [PubMed] [Google Scholar]
- 30.Cicero G, Mazziotti S, Silipigni S, Blandino A, Cantisani V, Pergolizzi S, D’Angelo T, Stagno A, Maimone S, Squadrito G, Ascenti G. Dual-energy CT quantification of fractional extracellular space in cirrhotic patients: comparison between early and delayed equilibrium phases and correlation with oesophageal varices. Radiol Med. 2021;126:761–7. 10.1007/s11547-021-01341-z. [DOI] [PubMed] [Google Scholar]
- 31.Patel M, Tann M, Liangpunsakul S. CT-scan Based Liver and Spleen Volume Measurement as a Prognostic Indicator for Patients with Cirrhosis. Am J Med Sci. 2021;362:252–9. 10.1016/j.amjms.2020.10.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Iwakiri Y, Trebicka J. Portal hypertension in cirrhosis: Pathophysiological mechanisms and therapy. JHEP Rep. 2021;3:100316. 10.1016/j.jhepr.2021.100316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Piotrowski D, Boroń-Kaczmarska A. Bacterial infections and hepatic encephalopathy in liver cirrhosis-prophylaxis and treatment. Adv Med Sci. 2017;62:345–56. 10.1016/j.advms.2016.11.009. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.