Published online Aug 7, 2026. doi: 10.3748/wjg.118090
Revised: February 24, 2026
Accepted: April 15, 2026
Published online: August 7, 2026
Processing time: 206 Days and 5.7 Hours
Severe hepatic encephalopathy (SHE) following transjugular intrahepatic por
To develop and validate a predictive model combining the FIPS score with nutri
This retrospective study included 590 cirrhotic patients who underwent TIPS at four Chinese hospitals between April 2015 and March 2025. Patients from Center I were divided into training and internal validation cohorts at a ratio of 7:3, while those from the remaining centers formed the external validation cohort. Inde
Independent predictors of post-TIPS SHE included FIPS [hazard ratio (HR) = 1.753; 95%CI: 1.266-2.428; P = 0.001], myosteatosis (HR = 1.921; 95%CI: 1.114-3.312; P = 0.019), and sarcopenia (HR = 2.722; 95%CI: 1.596-4.641; P < 0.001). The modified model demonstrated area under the receiver operating characteristic curves of 0.766, 0.745, and 0.751 at 3 months, 6 months, and 12 months in the training cohort; 0.733, 0.718, and 0.793 in the internal validation cohort; and 0.811, 0.797, and 0.854 in the external validation cohort, respectively, outperforming the original FIPS score. Predictive accuracy improved by 4.4% (P = 0.001), with calibration and decision curve analyses confirming its clinical utility.
The modified model demonstrated superior predictive performance for post-TIPS SHE and provides a reliable tool for risk stratification and clinical decision-mak
Core Tip: In recent years, transjugular intrahepatic portosystemic shunt (TIPS) has been widely applied in patients with portal hypertension. TIPS is characterized by its minimally invasive nature, proven efficacy, and rapid postoperative recovery, making it a primary treatment option for portal hypertension. However, the development of postoperative hepatic encephalopathy remains a major challenge associated with this technique. Existing predictive scoring models for this complication are limited. This study proposes a novel risk prediction model incorporating nutritional status-related indicators and the Freiburg post-TIPS survival score, which identifies patients at high risk of severe hepatic encephalopathy following TIPS and provides guidance for clinical decision-making.
- Citation: Zhang JQ, Yin L, Zhu YJ, Dong L, Hou CL, Tian SJ, Chen P, Huang XZ, Xu H, Chen ZY, Xu XJ, Zhou CZ, Cheng DL. Modified model incorporating sarcopenia and myosteatosis for predicting severe hepatic encephalopathy after transjugular intrahepatic portosystemic shunt: Multicenter study. World J Gastroenterol 2026; 32(29): 118090
- URL: https://www.wjgnet.com/1007-9327/full/v32/i29/118090.htm
- DOI: https://dx.doi.org/10.3748/wjg.118090
The transjugular intrahepatic portosystemic shunt (TIPS) procedure effectively reduces portal hypertension-related rebleeding and refractory ascites; however, its clinical benefit is often limited by the risk of post-procedural hepatic encephalopathy (HE)[1-3]. HE is a neuropsychiatric syndrome resulting from liver failure and/or portosystemic shunting, manifesting as cognitive impairments ranging from subclinical changes to coma. The incidence of post-TIPS HE among patients with liver cirrhosis ranges from 10% to 50%[4]. The most severe form, ranging from drowsiness to coma, classified as grades III-IV according to the West-Haven criteria, is referred to as severe HE (SHE)[5]. Patients with SHE frequently require hospitalization and face a significantly increased mortality risk[6-8].
Accurate and actionable preoperative risk prediction is crucial for maximizing the clinical benefits of TIPS and mitigating complications[3,9]. However, specific predictive tools for clinically assessing the risk of post-TIPS HE are lacking. In our previous work, we systematically compared various existing prognostic scores and demonstrated that the Freiburg post-TIPS survival (FIPS) score and the Chronic Liver Failure Consortium-Acute Decompensation (CLIFC-AD) score have superior discriminatory ability for predicting post-TIPS SHE[10]. A stratified analysis revealed that the models (FIPS and CLIFC-AD) performed significantly better in predicting SHE among patients with sarcopenia but were less effective in those without. Specifically, in non-sarcopenic patients, the concordance index (C-index) values for predicting SHE at 1 month, 3 months, 6 months, and 12 months post-TIPS using the FIPS model were 0.661, 0.594, 0.633, and 0.659, respectively, all below the threshold of 0.7, indicating poor predictive performance[10]. These findings suggest that combining the FIPS score with nutritional indicators, such as sarcopenia, may afford enhanced predictive accuracy.
However, the parameters constituting the currently available prognostic scores – such as bilirubin, renal function, albumin, international normalized ratio, serum sodium, and age – primarily reflect the liver-kidney axis function and biochemical homeostasis. Although a previous study[11] did explore the use of sarcopenia-related nutritional indices in models predicting survival and disease progression in patients with cirrhosis undergoing TIPS, no predictive models specifically targeting the outcome of SHE have yet been developed. At present, there is a lack of standardized methods to objectively assess muscle quantity, quality, or overall nutritional reserves, especially in patients with cirrhosis. Sarcopenia is a common complication in patients with cirrhosis and is associated with adverse clinical outcomes[12]. Furthermore, sarcopenia and myosteatosis (fatty infiltration of muscle) have been identified as key risk factors for post-TIPS HE[13]. The omission of these variables leads to a structural limitation. The FIPS and CLIFC-AD scores may indirectly identify patients with sarcopenia who often exhibit more severe organ dysfunction, which partly accounts for their better performance in this subgroup. Conversely, the lack of nutritional and muscular metrics in these models results in suboptimal performance in non-sarcopenic patients, thereby limiting their precision and cross-cohort applicability.
We chose the FIPS score as the framework for developing our model because of its simplicity, standardization, and ease of calculation. Additional indicators reflecting nutritional and muscular dimensions were incorporated to create a streamlined and comprehensive tool for predicting SHE. This study aims to provide clinicians with a precise and practical tool for identifying high-risk SHE patients post-TIPS, ultimately enabling personalized treatment and better clinical outcomes.
To address the limitations of existing prognostic models, we developed a modified FIPS model that integrates nutritional and muscular indicators, specifically sarcopenia and myosteatosis, into the framework of the FIPS score. Unlike traditional models, this novel approach provides a more comprehensive assessment of risk factors for SHE post-TIPS, particularly by incorporating previously underutilized yet clinically significant parameters. By enhancing predictive accuracy and enabling individualized risk stratification, the modified FIPS model serves as a practical and innovative tool for clinicians to identify high-risk patients and optimize treatment strategies, ultimately improving clinical outcomes.
This retrospective analysis was conducted on patients with cirrhosis who underwent TIPS between April 2015 and March 2025 at four independent hospitals in China. A total of 590 patients were included (Supplementary Figure 1). Of these, 419 patients from Center I (The First Affiliated Hospital of the University of Science and Technology of China) were randomly divided into a training cohort (TC) (n = 295) and an internal validation cohort (IVC) (n = 124) in a ratio of 7:3. The remaining 171 patients from Centers II, III, and IV (Affiliated Hospital of Xuzhou Medical University, Jiangyin Hospital Affiliated to Nantong University, and Lu’an Hospital Affiliated to Anhui Medical University, respectively) formed the external validation cohort (EVC). Data were extracted from the retrospective databases from each center, including baseline clinical and laboratory characteristics, detailed TIPS procedural records, and comprehensive follow-up assessments. All prognostic scores were calculated based on laboratory results obtained within three days before TIPS. The study was approved by the Institutional Review Board of the First Affiliated Hospital of the University of Science and Technology of China (No. 2023-RE-283), the Institutional Review Board of Affiliated Hospital of Xuzhou Medical University (No. HX-2025-063-01), the Institutional Review Board of Jiangyin Hospital Affiliated to Nantong University (No. 2025-KY052-01), and the Institutional Review Board of Lu’an Hospital Affiliated to Anhui Medical University (No. 2025-KY056-01) and was conducted in accordance with the ethical principles delineated in the Declaration of Helsinki (1975). The ethics committee waived the need for additional informed consent for this retrospective analysis, as all patient data were anonymized and the initial written informed consent had been obtained at the time of database entry. This study was conducted as per the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis guidelines[14].
Inclusion criteria: (1) Clinical diagnosis of cirrhosis, confirmed by clinical presentation, laboratory results, and imaging findings; and (2) Patients who underwent TIPS due to portal hypertension-related complications (e.g., variceal bleeding or refractory ascites).
Exclusion criteria: (1) Age < 18 years; (2) Prior TIPS placement; (3) Advanced hepatocellular carcinoma exceeding the Milan Criteria for liver transplantation (single tumor diameter £ 5 cm, or fewer than three nodules with the maximum diameter of each nodule £ 3 cm); (4) History of liver transplantation; (5) Comorbid extrahepatic malignancies or severe systemic diseases (e.g., significant renal insufficiency, severe cardiopulmonary dysfunction, or major cerebrovascular events) limiting 12-month survival; (6) Absence or poor quality of computed tomography (CT) imaging within one week prior to TIPS, precluding paraspinal muscle measurement; (7) Loss to follow-up within 12 months; and (8) Death or liver transplantation within 12 months without a prior documented episode of SHE.
Randomization was performed using R software (version 4.3.3) with the createDataPartition function from the caret package. To ensure reproducibility, a fixed random seed was used during the process. Patients were stratified by SHE occurrence to ensure event distribution balance between cohorts.
All TIPS procedures were performed by 2-4 interventional radiologists in accordance with the clinical practice guidelines established by relevant professional societies in China[15]. The procedure involved creating a shunt between the hepatic and portal veins using a puncture system, followed by the insertion of either an 8-mm VIATORR self-expanding polytetrafluoroethylene (PTFE)-covered stent (W.L. Gore, Flagstaff, AZ, United States), a PTFE-covered “controlled expansion” TIPS stent (VCX, W.L. Gore, Flagstaff, AZ, United States), or a combination of an 8-mm PTFE-covered stent (Fluency/Fluency Plus, Bard, Tempe, AZ, United States) with an 8-mm bare metal stent (Luminexx, Bard, Tempe, AZ, United States). A coil and/or medical glue were used for variceal embolization when necessary. The portal pressure gradient (PPG) was measured before and after shunt creation to evaluate procedural success. The target was to reduce the PPG to < 12 mmHg or achieve a ≥ 50% reduction from baseline. If the target PPG was not achieved after initial dilation, further staged dilation or additional embolization was performed based on hemodynamic and bleeding risk assessments.
All patients underwent protocolized follow-ups, which included outpatient clinic visits at 1 month, 3 months, and 6 months post-TIPS, followed by visits every 6 months thereafter. Quarterly telephone interviews were conducted between clinic visits to supplement the follow-up data. Evaluations at each visit comprised laboratory tests, abdominal ultrasound, contrast-enhanced CT scans, and clinical assessment for complications, including variceal bleeding, ascites, HE, and survival status. In cases of suspected shunt dysfunction, interventions such as balloon angioplasty or parallel stent placement were performed. The follow-up period continued until September 2025 or until the occurrence of liver transplantation or death.
CT is considered the gold standard for assessing sarcopenia in patients with cirrhosis[16]. In this study, sarcopenia was defined based on the transversal psoas muscle thickness-to-height ratio (TPMT/H). Abdominal CT images for all patients were retrieved from institutional picture archiving and communication system and analyzed independently by two experienced radiologists blinded to clinical data and outcomes. The measurements were performed by two radiologists with ≥ 3 years of experience in abdominal imaging. Discrepancies > 10% between the two measurements were resolved by a senior radiologist through consensus. The intra-class correlation coefficient (ICC) for TPMT measurements was 0.94 (95%CI: 0.91-0.96), and the inter-center ICC was 0.89 (95%CI: 0.84-0.93). The coefficient of variation for TPMT measurements was < 8%. Muscle quality was assessed by measuring the mean radiation attenuation [in Hounsfield Units (HU)] of the right psoas muscle at the third lumbar vertebra (L3) level (designated as the L3-CT value). The mean radiation attenuation (L3-CT value) of the right psoas muscle at the L3 vertebral level was also measured independently by the same two radiologists. The ICC for L3-CT value measurements was 0.92 (95%CI: 0.89-0.95), and the coefficient of variation for L3-CT measurements was < 5%. All measurements were performed on GE AW 4.6 or 4.7 workstations. Sarcopenia was defined as TPMT/H < 10.7 mm/m for men and < 7.8 mm/m for women[13]. Myosteatosis was defined as an L3-CT value of < 43.6 HU, based on the optimal cutoff identified by receiver operating characteristic (ROC) curve analysis applied to the overall study cohort.
HE was graded using the West-Haven classification system[5]. The West-Haven classification categorizes HE severity into five grades, ranging from grade 0 to IV, with grades III-IV indicating SHE. Specifically, grade III HE is characterized by somnolence to semi-stupor, with responses to verbal stimuli, disorientation, and marked confusion. Grade IV HE is marked by coma, with no response to verbal or painful stimuli. In this study, SHE was used as the endpoint event.
Statistical analyses were performed using SPSS (version 27.0) and R software (version 4.3.3). Normally distributed continuous variables were expressed as mean ± SD and compared using the t-test. Non-normally distributed data were expressed as median and interquartile range and compared using the rank-sum test. Categorical variables were presented as n (%) and analyzed using the χ² test or Fisher’s exact test when appropriate.
Patients at Center I were randomly assigned to the TC and IVC at a ratio of approximately 7:3, while those at Centers II, III, and IV were used as the EVC. The TC was used for variable selection and model development, whereas the IVC and EVC were employed for validating the TC-derived results. Risk factors were identified, and a predictive nomogram was constructed using univariable and multivariable Cox regression analyses. Various metrics were used to evaluate the performance of the model. The time-dependent C-index and the area under the ROC curve (AUC), which are related measures of the discriminatory ability of a model, were calculated at the same milestone time points of 3 months, 6 months, and 12 months across all cohorts. The 95%CI for AUC were estimated using the bootstrap method with 2000 resamples to ensure the robustness and interpretability of the estimates. Integrated discrimination improvement (IDI) analysis was performed to assess improvements in the predictive accuracy of the model, while its calibration performance was evaluated using calibration curves. Patients were stratified into lower-risk and higher-risk groups based on the cutoff value derived from the total nomogram score. Kaplan-Meier analysis and the log-rank test were used to compare differences between the survival curves. All tests were two-tailed. A P value < 0.05 was considered statistically significant.
Missing data, ranging from 5% to 20%, were imputed using the “mice” package. Specifically, the missing rates for the key variables were as follows: (1) Creatinine: 8.2%; (2) FIPS score: 9.9%; (3) Albumin: 5.7%; (4) Bilirubin: 6.9%; (5) L3-CT value: 6.3%; and (6) L3-Area: 8.5%. To address the issue of missing data, we applied multiple imputation using five imputed datasets to estimate and replace the missing values with plausible ones. Creatinine and the FIPS score, with missing rates of approximately 8%-10%, were among the imputed variables. To account for competing risks such as death or liver transplantation, sensitivity analyses were conducted using the Fine-Gray subdistribution hazard model to estimate the subdistribution hazard ratio (SHR) (“cmprsk” package). The Fine-Gray sensitivity analysis was also performed based on these imputed datasets. The “riskRegression” package was used for time-dependent ROC curve analysis, while “ggplot2” and “ggprism” were used for visualization. Clinical decision curve analysis (DCA) was performed using the “rmda” package. Survival analysis was carried out using the “survival” package. Plots were generated using the “survminer” package. Model calibration was performed using the “rms” package, and IDI analysis was conducted using the “dplyr” package.
The baseline demographic and clinicopathological characteristics of 590 patients are listed in Table 1. During the study period, 776 patients underwent TIPS. Of these, 186 patients did not meet the study criteria and were excluded, resulting in a final cohort of 590 patients (Supplementary Figure 1). The final cohort included 414 men and 176 women, with a median age of 54 years (range: 47-62 years). Sarcopenia was present in 30.5% patients (n = 180); the remaining patients (69.5%; n = 410) were non-sarcopenic. Myosteatosis was identified in 41.0% patients (n = 242), while 59.0% patients (n = 348) did not exhibit myosteatosis. Viral hepatitis was the most common cause of cirrhosis, affecting 57.3% (n = 338) of the study population. Among the entire study cohort, 12.9% patients (n = 76) with hepatocellular carcinoma met the Milan criteria. Ascites was present in 68.3% patients (n = 403), while diabetes was observed in 16.9% patients (n = 100). The median total bilirubin level was 21.10 μmol/L (range: 14.10-31.05 μmol/L), the median albumin level was 32.60 g/L (range: 29.28-36.20 g/L), the median prothrombin time was 15.00 seconds (range: 13.50-16.80 seconds), and the median creatinine level was 63.00 μmol/L (range: 52.00-78.00 μmol/L). TIPS was performed to prevent rebleeding from esophagogastric varices in 66.4% patients (n = 392), refractory ascites in 23.6% patients (n = 139), and both conditions in 10% patients (n = 59). The median FIPS score for the cohort was -1.04, ranging from -1.65 to -0.47, and the median TPMT/H ratio was 11.30 mm/m (range: 9.41-13.95 mm/m). The mean ± SD of the L3 psoas muscle CT attenuation value was 45.60 ± 10.17 HU. The median follow-up duration was 27.50 months (range: 14.00-47.40 months). SHE occurred in 108 patients (18.3%) within 12 months after TIPS, including 57 patients (19.3%) in the TC, 21 patients (16.9%) in the IVC, and 30 patients (17.5%) in the EVC.
| Characteristics | All patients | Center I | Centers II, III, IV | P value | |
| Training cohort | Internal validation cohort | External validation cohort | |||
| Severe hepatic encephalopathy | 108 (18.3) | 57 (19.3) | 21 (16.9) | 30 (17.5) | 0.808 |
| Age (years) | 54.00 (47.00, 62.00) | 53.00 (47.00, 60.50) | 55.00 (49.00, 65.00) | 57.00 (51.00, 65.00) | 0.230 |
| Gender | |||||
| Male | 414 (70.2) | 202 (68.5) | 100 (80.6) | 112 (65.5) | 0.013 |
| Female | 176 (29.8) | 93 (31.5) | 24 (19.4) | 59 (34.5) | |
| Sarcopenia | |||||
| No | 410 (69.5) | 208 (70.5) | 89 (71.8) | 113 (66.1) | 0.500 |
| Yes | 180 (30.5) | 87 (29.5) | 35 (28.2) | 58 (33.9) | |
| Myosteatosis | |||||
| No | 348 (59.0) | 162 (54.9) | 65 (52.4) | 121 (70.8) | < 0.001 |
| Yes | 242 (41.0) | 133 (45.1) | 59 (47.6) | 50 (29.2) | |
| Etiology | |||||
| Viral hepatitis | 338 (57.3) | 171 (58.0) | 71 (57.3) | 96 (56.1) | 0.929 |
| Other | 252 (42.7) | 124 (42.0) | 53 (42.7) | 75 (43.9) | |
| Ascites | |||||
| No | 187 (31.7) | 88 (29.8) | 41 (33.1) | 58 (33.9) | 0.615 |
| Yes | 403 (68.3) | 207 (70.2) | 83 (66.9) | 113 (66.1) | |
| Diabetes | |||||
| No | 490 (83.1) | 242 (82.0) | 109 (87.9) | 139 (81.3) | 0.263 |
| Yes | 100 (16.9) | 53 (18.0) | 15 (12.1) | 32 (18.7) | |
| Child-Pugh score | 7.00 (6.00, 9.00) | 7.00 (6.00, 9.00) | 7.00 (6.00, 8.00) | 7.00 (6.00, 8.00) | 0.033 |
| Meld score | 10.00 (7.00, 12.00) | 10.00 (7.00, 13.00) | 10.00 (7.00, 12.00) | 9.19 (7.26, 11.47) | 0.796 |
| Meld-Na score | 10.20 (8.00, 14.00) | 10.60 (8.00, 14.00) | 10.00 (7.00, 13.00) | 7.98 (4.00, 11.39) | < 0.001 |
| Freiburg index of post-transjugular intrahepatic portosystemic shunt survival score | -1.04 (-1.65 to -0.47) | -1.07 (-1.65 to -0.48) | -0.96 (-1.63 to -0.44) | -0.90 (-1.50 to -0.01) | |
| Prothrombin time (seconds) | 15.00 (13.50, 16.80) | 15.10 (13.55, 16.90) | 14.90 (13.40, 16.42) | 14.40 (12.93, 15.97) | 0.058 |
| Blood urea nitrogen (mmol/L) | 5.85 (4.33, 8.10) | 5.85 (4.33, 8.20) | 5.85 (4.24, 7.84) | 6.74 (4.73, 8.66) | 0.040 |
| Platelets (× 109) | 67.00 (46.00, 106.00) | 68.00 (48.00, 106.00) | 65.50 (43.28, 103.25) | 72.00 (51.00, 106.00) | 0.195 |
| Lymphocytes (× 109) | 0.70 (0.45, 1.10) | 0.70 (0.46, 1.18) | 0.70 (0.44, 0.99) | 0.70 (0.46, 1.00) | 0.742 |
| Hemoglobin (g/L) | 80.00 (65.00, 99.50) | 80.00 (65.00, 99.50) | 80.50 (65.75, 98.50) | 78.00 (65.00, 89.00) | 0.062 |
| Albumin (g/L) | 32.60 (29.28, 36.20) | 32.50 (29.00, 36.40) | 32.65 (29.52, 35.52) | 32.60 (29.50, 36.30) | 0.876 |
| Alanine aminotransferase (U/L) | 22.00 (15.00, 33.00) | 22.00 (15.00, 32.45) | 23.00 (15.75, 35.70) | 19.70 (14.00, 32.00) | 0.431 |
| Aspartate aminotransferase (U/L) | 30.20 (22.50, 42.45) | 30.00 (21.60, 42.45) | 31.50 (24.00, 42.25) | 27.60 (20.88, 42.91) | 0.100 |
| Creatinine (μmol/L) | 63.00 (52.00, 78.00) | 62.00 (52.00, 80.00) | 66.00 (53.00, 75.00) | 62.00 (51.50, 73.30) | 0.388 |
| Total bilirubin (μmol/L) | 21.10 (14.10, 31.05) | 21.10 (14.30, 31.95) | 21.10 (13.88, 26.75) | 18.04 (13.60, 25.79) | 0.113 |
| Maximal transverse diameter of the right psoas muscle at the L3 level (cm) | 1.90 (1.52, 2.37) | 1.89 (1.52, 2.33) | 1.91 (1.54, 2.39) | 1.87 (1.49, 2.47) | 0.099 |
| Cross-sectional area of the right psoas muscle at the L3 level (cm2) | 5.49 (3.83, 7.02) | 5.57 (3.78, 7.02) | 5.39 (3.92, 7.00) | 5.06 (3.50, 6.64) | 0.012 |
| Mean muscle attenuation in HU of the right psoas muscle at the L3 level (HU) | 45.60 ± 10.17 | 44.14 ± 9.35 | 43.65 ± 10.47 | 49.54 ± 10.29 | < 0.001 |
| L3_transversal psoas muscle thickness-to-height (mm/m) | 11.30 (9.41, 13.95) | 11.30 (9.36, 13.85) | 11.30 (9.56, 14.60) | 11.33 (9.00, 14.11) | 0.217 |
| Time (months) | 27.50 (14.00, 47.40) | 26.53 (14.00, 47.70) | 29.75 (14.50, 47.00) | 21.50 (12.10, 24.30) | < 0.001 |
To develop a modified predictive model and validate its performance, patients from Center I were randomly divided into TC (n = 295) and IVC (n = 124) cohorts at a ratio of approximately 7:3, while patients from the other three centers were assigned to the EVC (n = 171). Baseline demographic, etiological, hepatic function, nutritional, muscle mass, and complication data for each cohort are summarized in Table 1. Univariable and multivariable Cox regression analyses were performed in the TC cohort to identify independent predictors of post-TIPS SHE in patients with cirrhosis (Table 2).
| Characteristics | Univariate | Multivariate | ||
| HR (95%CI) | P value | HR (95%CI) | P value | |
| Age (years) | 1.042 (1.018-1.067) | < 0.001 | - | - |
| Gender (female/male) | 0.609 (0.328-1.131) | 0.116 | - | - |
| Sarcopenia (yes/no) | 2.598 (1.545-4.370) | < 0.001 | 2.722 (1.596-4.641) | < 0.001a |
| Freiburg index of post-transjugular intrahepatic portosystemic shunt survival score | 1.801 (1.349-2.405) | < 0.001 | 1.753 (1.266-2.428) | 0.001a |
| Prothrombin time (seconds) | 1.023 (0.930-1.124) | 0.644 | - | - |
| Blood urea nitrogen (mmol/L) | 1.050 (1.013-1.089) | 0.008 | 1.002 (0.957-1.049) | 0.932 |
| White blood cell (109/L) | 0.941 (0.851-1.041) | 0.237 | - | - |
| Platelets (109/L) | 0.997 (0.992-1.001) | 0.156 | - | - |
| Lymphocyte (109/L) | 0.687 (0.424-1.112) | 0.127 | - | - |
| Hemoglobin (g/L) | 1.003 (0.993-1.013) | 0.542 | - | - |
| Albumin (g/L) | 0.968 (0.921-1.017) | 0.2 | - | - |
| Alanine aminotransferase (U/L) | 1.004 (1.001-1.008) | 0.022 | 1.004 (1.000-1.008) | 0.085 |
| Creatinine (μmol/L) | 1.008 (1.003-1.013) | 0.002 | - | - |
| Total bilirubin (μmol/L) | 1.015 (1.003-1.026) | 0.013 | - | - |
| Maximal transverse diameter of the right psoas muscle at the L3 level (cm) | 0.594 (0.363-0.973) | 0.038 | - | - |
| Cross-sectional area of the right psoas muscle at the L3 level (cm2) | 0.909 (0.802-1.031) | 0.137 | - | - |
| Mean muscle attenuation in HU of the right psoas muscle at the L3 level (HU) | 0.964 (0.938-0.990) | 0.007 | - | - |
| Myosteatosis | 2.245 (1.316-3.830) | 0.003 | 1.921 (1.114-3.312) | 0.019a |
| L3_transversal psoas muscle thickness-to-height (mm/m) | 0.907 (0.832-0.988) | 0.025 | - | - |
Univariable Cox regression analysis identified significant associations between post-TIPS SHE incidence and age, sarcopenia, FIPS score, blood urea nitrogen, alanine aminotransferase, creatinine, bilirubin, maximum transverse diameter of the right psoas muscle at the L3 level (L3-TD), mean CT attenuation value of the psoas muscle at the L3 level (L3-CT), myosteatosis, and L3_TPMT.H (Table 2). Given the superior discriminatory ability of the FIPS scoring model demonstrated in previous studies[10], FIPS score, sarcopenia, blood urea nitrogen, alanine aminotransferase, and myosteatosis were included in a multivariable Cox proportional hazards model using a stepwise backward selection approach, after eliminating multicollinearity. The final model identified sarcopenia [hazard ratio (HR) = 2.722; 95%CI: 1.596-4.641; P < 0.001], FIPS score (HR = 1.753; 95%CI: 1.266-2.428; P = 0.001), and myosteatosis (HR = 1.921; 95%CI: 1.114-3.312; P = 0.019) as independent risk factors for post-TIPS SHE in patients with cirrhosis (Table 2). A nomogram was constructed based on these three independent predictors (Figure 1) to quantify the risk of SHE at 3 months, 6 months, and 12 months post-TIPS.
Sensitivity analysis using the Fine-Gray subdistribution hazard model confirmed that the relative effects of the independent predictors (FIPS score, sarcopenia, and myosteatosis) on the risk of post-TIPS SHE were consistent with those observed in the Cox model (Supplementary Table 1). The maximum relative difference between the HR from the Cox model and the subdistribution HR (SHR) from the Fine-Gray model was less than 5%, indicating that the models yielded comparable results and that the impact of competing risks (death or liver transplantation) on the primary endpoint (SHE-free survival) was minimal.
Performance evaluation and subsequent validation demonstrated that the modified predictive model consistently exhibited strong discriminatory ability across all cohorts. The modified model consistently outperformed the FIPS scoring system in terms of time-dependent C-index across all cohorts (Figure 2). Furthermore, it demonstrated approximately a 7% increase in the C-index compared to that achieved using the FIPS score at multiple time points across all cohorts. ROC curves comparing the modified model with the original FIPS score are illustrated in Figure 3. Notably, at 12 months post-TIPS, the AUC values for the modified predictive model and the FIPS scoring model in the TC cohort were 0.751 (95%CI: 0.673-0.829) and 0.681 (95%CI: 0.601-0.761), respectively (Figure 3C). In the IVC cohort, the AUC values were 0.793 (95%CI: 0.682-0.905) for the modified model and 0.726 (95%CI: 0.592-0.859) for the FIPS scoring model (Figure 3F). In the EVC cohort, the AUC values were 0.854 (95%CI: 0.722-0.936) for the modified model and 0.793 (95%CI: 0.678-0.907) for the FIPS scoring model (Figure 3I), demonstrating superior predictive performance for the risk of SHE. Compared to the original model, the final model demonstrated an IDI of 4.4% (95%CI: 1.9%-7.0%, P = 0.001) in the entire cohort, 7.1% (95%CI: 3.7%-10.5%, P < 0.001) in the TC cohort, 4.7% (95%CI: -0.5% to 9.9%, P = 0.074) in the IVC cohort, and 9.3% (95%CI: 3.4%-15.2%, P = 0.002) in the EVC cohort (Table 3). The calibration curves showed good consistency between the model nomogram predictions and actual observations in all cohorts (Figure 4). The DCA (Figure 5) demonstrated that the net benefit curves of the nomogram in all three cohorts were consistently higher than those of the FIPS across the major clinical decision threshold range.
| Dataset | Integrated discrimination improvement | 95%CI | P value |
| All cohorts | 0.044 | 0.019-0.070 | 0.001 |
| Training cohort | 0.071 | 0.037-0.105 | < 0.001 |
| Internal validation cohort | 0.047 | -0.005 to 0.099 | 0.074 |
| External validation cohort | 0.093 | 0.034-0.152 | 0.002 |
The optimal cutoff values determined by the surv_cutpoint function (TC: 75.353; IVC: 91.565; EVC: 103.741) were used to divide the study cohorts into low-risk and high-risk groups for post-TIPS SHE. These values represent continuous cutoffs specific to each cohort. Kaplan-Meier curves for these two groups were plotted based on the modified predictive model nomogram, followed by a log-rank test for comparison. The cumulative incidence of SHE was significantly higher in the high-risk group than it was in the low-risk group. Specifically, the HRs of the model in the TC, IVC, and EVC cohorts were 4.485 (95%CI: 2.514-8.000), 4.133 (95%CI: 1.709-9.997), and 5.773 (95%CI: 2.745-12.143), respectively, with all P values < 0.05 (Figure 6).
This study focused on a high-impact endpoint – SHE within 12 months post-TIPS – and proposed as well as validated a simplified nomogram model derived from the FIPS score. The model integrates only three independent variables: The FIPS score, which represents the core axis of biochemical and functional reserve; sarcopenia, which reflects muscle mass; and myosteatosis, which reflects muscle quality and tissue composition.
Compared to the original FIPS score alone, the modified model demonstrated superior discrimination, stable calibration, and higher clinical net benefit in both the training set and various validation cohorts, with DCA advantages covering the primary intervention threshold range. Additionally, reclassification improvement was confirmed by the IDI. Risk stratification based on the total nomogram score revealed that the incidence of SHE in the high-risk group increased by more than fourfold, providing a clear operational value for “triggering interventions”.
This study performed targeted functional optimization based on the FIPS score rather than simply adding variables. By incorporating key indicators related to muscle-nutrition-metabolic buffering, the model addressed the limitations of the traditional liver/kidney biochemical framework. This approach maximized information gain with fewer variables, offering a more streamlined and practical strategy for model optimization.
Bettinger et al. incorporated age, bilirubin, albumin, and creatinine into a novel risk score called the FIPS[17]. Although the FIPS model was initially developed to predict survival time after TIPS, recent studies have demonstrated a strong correlation between the FIPS score and the risk of post-TIPS HE[18,19]. The present study also confirmed that the FIPS score is an independent risk factor for predicting post-TIPS SHE through univariate Cox regression analysis (HR = 1.801; 95%CI: 1.349-2.405; P < 0.001) and multivariate Cox regression analysis (HR = 1.753; 95%CI: 1.266-2.428; P = 0.001).
The FIPS score effectively predicts the risk of post-TIPS SHE by addressing three critical components: (1) Input; (2) Processing; and (3) Tolerance. Following TIPS, nitrogenous toxins bypass the liver and directly enter the systemic circulation. Impaired liver function (elevated bilirubin and decreased albumin) and reduced renal function (increased creatinine) weaken the body’s ability to eliminate toxins[20,21]. Additionally, advanced age and hypoalbuminemia further lower the tolerance threshold of the central nervous system[22,23]. The aforementioned four variables (i.e., age, bilirubin, albumin, and creatinine) in the FIPS score reflect hepatic secretion and metabolism, renal clearance function, and brain susceptibility. These dimensions complement each other and create a synergistic effect, enabling more accurate prediction of SHE risk.
However, the FIPS score considers only organ function-related parameters and lacks an assessment of nutritional status. Malnutrition is a common complication in patients with liver cirrhosis[24]. Previous systematic reviews and meta-analyses have shown that sarcopenia and myosteatosis are strongly associated with complications and poor overall survival in patients with various diseases[25,26]. Moreover, a previous study[27] confirmed that an abnormal nutritional status is associated with a higher incidence of post-TIPS HE compared to a normal nutritional status.
In the present study, both univariate and multivariate Cox regression analyses confirmed that sarcopenia and myosteatosis are independent risk factors for predicting post-TIPS SHE. Sarcopenia and myosteatosis were found to be complementary: The former indicates a loss of muscle volume, while the latter reflects a decline in muscle function and metabolic quality. The assessment of sarcopenia and myosteatosis can be easily performed using skeletal muscle CT imaging, which is straightforward, technically feasible, and highly applicable in clinical practice. Following TIPS, increased portosystemic shunting allows nitrogenous toxins to enter the systemic circulation more easily, making ammonia clearance increasingly dependent on muscle function. Sarcopenia may result in reduced muscle mass and strength, along with a potential decrease in glutamine synthetase substrates and enzyme activity reserves. Myosteatosis, characterized by fat replacing functional muscle fibers, may induce mitochondrial and metabolic dysfunction, which could impair the ability of glutamine synthetase to convert ammonia into non-toxic glutamine, thereby potentially reducing the uptake of ammonia and detoxification and exacerbating ammonia toxicity[28,29]. Additionally, fat infiltration-associated inflammation and insulin resistance may suppress protein synthesis and promote proteolysis, driving a vicious cycle of “muscle loss-limited detoxification-elevated ammonia load-increased brain susceptibility”[30,31]. Thus, in the context of portosystemic shunting, sarcopenia and myosteatosis increase systemic ammonia levels and brain exposure through independent and synergistic pathways, thereby potentially elevating the risk of HE.
These mechanisms and cumulative risk effects may further drive the progression of HE to its most severe form – SHE. The present study developed a modified model for predicting post-TIPS SHE by integrating the FIPS score, sarcopenia, and myosteatosis. In this study, 18.3% of patients developed SHE within one year of undergoing TIPS, which is consistent with the previous observation[27]. A comparison revealed that the modified model achieved higher AUCs at all time points for predictive discrimination compared to that obtained using the FIPS score. Further analysis demonstrated that the modified model provides superior net benefit within a reasonable range, making it a valuable alternative risk prediction tool for clinicians. By using the optimal cutoff values determined by the surv_cutpoint function (TC: 75.353; IVC: 91.565; EVC: 103.741), Kaplan-Meier analysis effectively distinguished the high-risk and low-risk groups with different incidences of SHE. The risk of SHE occurrence in the high-risk group was more than four times that of the low-risk group. Additionally, the differences between the two groups were statistically significant (P < 0.05 for all comparisons). Furthermore, to enhance clinical applicability, we propose a three-tier risk stratification system based on the tertiles of the total nomogram score. This risk stratification system demonstrated good performance in identifying the risk of SHE across different populations (Supplementary Figure 2 and Supplementary Table 2). This classification provides a straightforward framework for clinicians to identify patients at varying levels of risk for post-TIPS SHE and to tailor individualized interventions accordingly.
The decision to “augment” rather than “replace” the FIPS score was based on three key considerations. First, the FIPS score consists of a small number of complementary and stable variables with minimal intercorrelation and cross-domain overlap, reducing the risk of “redundant scoring” and leaving clear structural space for the inclusion of muscle and imaging phenotypes. Second, while the FIPS score primarily reflects central organ function and hemodynamics (e.g., liver and kidney), it lacks coverage of the “muscle-nutrition-inflammation” axis, which is critical for peripheral ammonia buffering and systemic susceptibility. The addition of sarcopenia and myosteatosis helps capture key signals of peripheral compensatory capacity and inflammatory susceptibility. Third, the existing clinical workflow for FIPS collection and calculation minimizes the learning curve and system modifications when choosing “augmentation” over “replacement”. The modified nomogram retains FIPS’s transparency, interpretability, and ease of implementation while providing continuous risk probabilities, enabling precise stratification, threshold adaptation, and dynamic monitoring over time.
By integrating muscle quality and imaging indicators, the modified model significantly improves the predictive accuracy and clinical utility for post-TIPS SHE, providing a reliable tool for identifying high-risk patients and optimizing treatment decisions. The nomogram could be implemented in routine practice using thin-slice abdominal CT images with a slice thickness of < 5 mm obtained within one week prior to the TIPS procedure, combined with relevant laboratory results collected within three days before the procedure. These practical steps would enable standardized and individualized risk stratification, helping clinicians to better identify high-risk patients and guide personalized intervention strategies.
This study has several limitations as well. First, as a retrospective study, it is subject to selection and information biases. Second, the proposed model represents a functional modification of the existing FIPS score rather than an entirely new model, and its superiority over independent models requires further validation. Third, the modified predictive model demonstrated excellent performance in Chinese centers with predominantly viral-etiology cirrhosis. However, its applicability to non-viral or non-Asian populations remains uncertain. External validation in other geographic regions or etiologic cohorts should be performed to assess the generalizability of the model. Additionally, the lack of absolute risk estimates or predicted probabilities across clinically meaningful strata limits the practical utility of the model in tailoring early interventions for high-risk patients. Future research should address these gaps to improve the interpretability and real-world applicability of the model. Furthermore, the model focuses on predicting SHE as the endpoint and does not provide risk probabilities for the full spectrum or overt HE, which may limit its applicability in certain clinical scenarios, such as early intervention strategies. Finally, it is important to emphasize that the model proposed in this study represents a functional enhancement of the existing FIPS score rather than an entirely novel predictive tool. Therefore, conclusions regarding its clinical application, particularly those related to dynamic monitoring and individualized interventions, should be regarded as preliminary findings. These conclusions require further confirmation through prospective, multicenter cohort studies for external validation.
We have demonstrated that the FIPS score, sarcopenia, and myosteatosis are independent risk factors for predicting post-TIPS SHE. The modified predictive model holds promise for future applications in the refined risk stratification and dynamic monitoring of SHE, with the potential to improve personalized management of patients with cirrhosis after TIPS in clinical practice. By providing individualized risk assessments, this model can assist clinicians in optimizing TIPS timing and tailoring post-procedural strategies to better mitigate SHE risk.
| 1. | Boike JR, Thornburg BG, Asrani SK, Fallon MB, Fortune BE, Izzy MJ, Verna EC, Abraldes JG, Allegretti AS, Bajaj JS, Biggins SW, Darcy MD, Farr MA, Farsad K, Garcia-Tsao G, Hall SA, Jadlowiec CC, Krowka MJ, Laberge J, Lee EW, Mulligan DC, Nadim MK, Northup PG, Salem R, Shatzel JJ, Shaw CJ, Simonetto DA, Susman J, Kolli KP, VanWagner LB; Advancing Liver Therapeutic Approaches (ALTA) Consortium. North American Practice-Based Recommendations for Transjugular Intrahepatic Portosystemic Shunts in Portal Hypertension. Clin Gastroenterol Hepatol. 2022;20:1636-1662.e36. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 196] [Cited by in RCA: 180] [Article Influence: 45.0] [Reference Citation Analysis (3)] |
| 2. | Deltenre P, Zanetto A, Saltini D, Moreno C, Schepis F. The role of transjugular intrahepatic portosystemic shunt in patients with cirrhosis and ascites: Recent evolution and open questions. Hepatology. 2023;77:640-658. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 58] [Cited by in RCA: 56] [Article Influence: 18.7] [Reference Citation Analysis (0)] |
| 3. | de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C; Baveno VII Faculty. Baveno VII - Renewing consensus in portal hypertension. J Hepatol. 2022;76:959-974. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 2244] [Cited by in RCA: 2113] [Article Influence: 528.3] [Reference Citation Analysis (17)] |
| 4. | Friis KH, Thomsen KL, Laleman W, Montagnese S, Vilstrup H, Lauridsen MM. Post-Transjugular Intrahepatic Portosystemic Shunt (TIPS) Hepatic Encephalopathy-A Review of the Past Decade's Literature Focusing on Incidence, Risk Factors, and Prophylaxis. J Clin Med. 2023;13:14. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 29] [Reference Citation Analysis (1)] |
| 5. | Weissenborn K. Hepatic Encephalopathy: Definition, Clinical Grading and Diagnostic Principles. Drugs. 2019;79:5-9. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 262] [Cited by in RCA: 211] [Article Influence: 30.1] [Reference Citation Analysis (2)] |
| 6. | Cheng DL, Zhou CZ, Zhu YJ, Yin L, Zhang YH, Lv WF. [Impact of sarcopenia on patients with portal hypertension after transjugular intrahepatic portosystemic shunt]. Zhonghua Fangshexue Zazhi. 2021;55:425-430. [DOI] [Full Text] |
| 7. | Jain A, Sharma BC, Mahajan B, Srivastava S, Kumar A, Sachdeva S, Sonika U, Dalal A. L-ornithine L-aspartate in acute treatment of severe hepatic encephalopathy: A double-blind randomized controlled trial. Hepatology. 2022;75:1194-1203. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 70] [Cited by in RCA: 65] [Article Influence: 16.3] [Reference Citation Analysis (0)] |
| 8. | Stewart CA, Malinchoc M, Kim WR, Kamath PS. Hepatic encephalopathy as a predictor of survival in patients with end-stage liver disease. Liver Transpl. 2007;13:1366-1371. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 227] [Cited by in RCA: 191] [Article Influence: 10.1] [Reference Citation Analysis (1)] |
| 9. | European Association for the Study of the Liver. EASL Clinical Practice Guidelines on nutrition in chronic liver disease. J Hepatol. 2019;70:172-193. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 900] [Cited by in RCA: 799] [Article Influence: 114.1] [Reference Citation Analysis (5)] |
| 10. | Xu XJ, Yin L, Zhu YJ, Lu D, Huang XZ, Lv WF, Zhou CZ, Cheng DL. Comparing efficacy of different scoring models to predict hepatic encephalopathy after TIPS in cirrhotic patients. Ann Med. 2025;57:2514082. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 3] [Cited by in RCA: 3] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 11. | Zhang Q, Long L, Zhu HL, Peng H, Luo XH, Zhu KS, Wang RP. Predicting disease progression in cirrhotic patients after transjugular intrahepatic portosystemic shunt implantation: A sex-stratified analysis. World J Gastroenterol. 2023;29:5768-5780. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 12. | Xiao HJ, Han T. [Assessment and intervention strategies for sarcopenia in patients with liver cirrhosis]. Zhonghua Gan Zang Bing Za Zhi. 2021;29:199-203. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 1] [Reference Citation Analysis (0)] |
| 13. | Yin L, Chu SL, Lv WF, Zhou CZ, Liu KC, Zhu YJ, Zhang WY, Wang CX, Zhang YH, Lu D, Cheng DL. Contributory roles of sarcopenia and myosteatosis in development of overt hepatic encephalopathy and mortality after transjugular intrahepatic portosystemic shunt. World J Gastroenterol. 2023;29:2875-2887. [PubMed] [DOI] [Full Text] |
| 14. | Collins GS, Reitsma JB, Altman DG, Moons KG. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3053] [Cited by in RCA: 2837] [Article Influence: 257.9] [Reference Citation Analysis (4)] |
| 15. | Chinese College of Interventionalists. [CCI clinical practice guidelines: Management of TIPS for portal hypertension (2019 edition)]. Linchuang Gandanbing Zazhi. 2019;35:2694-2699. [DOI] [Full Text] |
| 16. | Ebadi M, Bhanji RA, Dunichand-Hoedl AR, Mazurak VC, Baracos VE, Montano-Loza AJ. Sarcopenia Severity Based on Computed Tomography Image Analysis in Patients with Cirrhosis. Nutrients. 2020;12:3463. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 37] [Cited by in RCA: 33] [Article Influence: 5.5] [Reference Citation Analysis (0)] |
| 17. | Bettinger D, Sturm L, Pfaff L, Hahn F, Kloeckner R, Volkwein L, Praktiknjo M, Lv Y, Han G, Huber JP, Boettler T, Reincke M, Klinger C, Caca K, Heinzow H, Seifert LL, Weiss KH, Rupp C, Piecha F, Kluwe J, Zipprich A, Luxenburger H, Neumann-Haefelin C, Schmidt A, Jansen C, Meyer C, Uschner FE, Brol MJ, Trebicka J, Rössle M, Thimme R, Schultheiss M. Refining prediction of survival after TIPS with the novel Freiburg index of post-TIPS survival. J Hepatol. 2021;74:1362-1372. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 44] [Cited by in RCA: 134] [Article Influence: 26.8] [Reference Citation Analysis (0)] |
| 18. | Cai W, Zheng B, Lin X, Wu W, Chen C. Prediction of Patient Hepatic Encephalopathy Risk with Freiburg Index of Post-TIPS Survival Score Following Transjugular Intrahepatic Portosystemic Shunts: A Retrospective Study. Int J Gen Med. 2022;15:4007-4016. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 9] [Reference Citation Analysis (0)] |
| 19. | Sturm L, Schultheiss M, Stöhr F, Labenz C, Maasoumy B, Tiede A, Praktiknjo M, Seifert LL, Auer TA, Fehrenbach U, Piecha F, Harberts A, Kluwe J, Bruns T, Pollmanns MR, Chang J, Grobelski J, Jansen C, Meyer C, Reincke M, Rohrer C, Philipp Arbabi SR, Kimmann M, Ripoll C, Zipprich A, Hinrichs J, Koehler M, Trebicka J, Kloeckner R, Engelmann C, Thimme R, Bettinger D; German Cirrhosis Study Group of the DGVS. Freiburg index of post-TIPS survival (FIPS) identifies patients at risk of further decompensation and ACLF after TIPS. J Hepatol. 2025;83:348-357. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3] [Cited by in RCA: 12] [Article Influence: 12.0] [Reference Citation Analysis (0)] |
| 20. | Issac AG, Yu MA, Rogers DM, Subramanian RM. Case Report: Efficacy of albumin dialysis for the reversal of bile cast nephropathy-induced acute kidney injury. Front Nephrol. 2023;3:1256672. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 2] [Article Influence: 0.7] [Reference Citation Analysis (0)] |
| 21. | Ginès P, Solà E, Angeli P, Wong F, Nadim MK, Kamath PS. Hepatorenal syndrome. Nat Rev Dis Primers. 2018;4:23. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 221] [Cited by in RCA: 185] [Article Influence: 23.1] [Reference Citation Analysis (2)] |
| 22. | Ngcobo NN. Correction: Influence of Ageing on the Pharmacodynamics and Pharmacokinetics of Chronically Administered Medicines in Geriatric Patients: A Review. Clin Pharmacokinet. 2025;64:463. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 1] [Cited by in RCA: 36] [Article Influence: 36.0] [Reference Citation Analysis (0)] |
| 23. | Castellazzi M, Morotti A, Tamborino C, Alessi F, Pilotto S, Baldi E, Caniatti LM, Trentini A, Casetta I, Granieri E, Pugliatti M, Fainardi E, Bellini T. Increased age and male sex are independently associated with higher frequency of blood-cerebrospinal fluid barrier dysfunction using the albumin quotient. Fluids Barriers CNS. 2020;17:14. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 22] [Cited by in RCA: 58] [Article Influence: 9.7] [Reference Citation Analysis (0)] |
| 24. | Traub J, Reiss L, Aliwa B, Stadlbauer V. Malnutrition in Patients with Liver Cirrhosis. Nutrients. 2021;13:540. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 139] [Cited by in RCA: 119] [Article Influence: 23.8] [Reference Citation Analysis (3)] |
| 25. | Wijarnpreecha K, Werlang M, Panjawatanan P, Kroner PT, Cheungpasitporn W, Lukens FJ, Pungpapong S, Ungprasert P. Association between sarcopenia and hepatic encephalopathy: A systematic review and meta-analysis. Ann Hepatol. 2020;19:245-250. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 16] [Cited by in RCA: 50] [Article Influence: 10.0] [Reference Citation Analysis (1)] |
| 26. | Ahn H, Kim DW, Ko Y, Ha J, Shin YB, Lee J, Sung YS, Kim KW. Updated systematic review and meta-analysis on diagnostic issues and the prognostic impact of myosteatosis: A new paradigm beyond sarcopenia. Ageing Res Rev. 2021;70:101398. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 24] [Cited by in RCA: 169] [Article Influence: 33.8] [Reference Citation Analysis (1)] |
| 27. | Huang WT, Cheng DL, Yin L, Dong L, Zhu YJ, Zhang L, Lu D, Zhou CZ, Lv WF. A GLIM-defined malnutrition-based model for predicting prognosis of cirrhotic patients with portal hypertension following TIPS procedure. BMC Gastroenterol. 2025;25:586. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 1] [Reference Citation Analysis (0)] |
| 28. | Cruzat V, Macedo Rogero M, Noel Keane K, Curi R, Newsholme P. Glutamine: Metabolism and Immune Function, Supplementation and Clinical Translation. Nutrients. 2018;10:1564. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 877] [Cited by in RCA: 823] [Article Influence: 102.9] [Reference Citation Analysis (5)] |
| 29. | Holecek M. Evidence of a vicious cycle in glutamine synthesis and breakdown in pathogenesis of hepatic encephalopathy-therapeutic perspectives. Metab Brain Dis. 2014;29:9-17. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 48] [Cited by in RCA: 60] [Article Influence: 5.0] [Reference Citation Analysis (0)] |
| 30. | Kumar R, Prakash SS, Priyadarshi RN, Anand U. Sarcopenia in Chronic Liver Disease: A Metabolic Perspective. J Clin Transl Hepatol. 2022;10:1213-1222. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 14] [Cited by in RCA: 20] [Article Influence: 5.0] [Reference Citation Analysis (1)] |
| 31. | Bhanji RA, Moctezuma-Velazquez C, Duarte-Rojo A, Ebadi M, Ghosh S, Rose C, Montano-Loza AJ. Myosteatosis and sarcopenia are associated with hepatic encephalopathy in patients with cirrhosis. Hepatol Int. 2018;12:377-386. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 189] [Cited by in RCA: 166] [Article Influence: 20.8] [Reference Citation Analysis (3)] |