Screening and management for portal hypertension in hepatocellular carcinoma: A nationwide survey in China

1

Introduction

Hepatocellular carcinoma (HCC) and portal hypertension (PHT) are two common complications of cirrhosis that often co-exist. The incidence of PHT ranges from approximately 35% to 52% in early-stage HCC [1], while it is higher in advanced stages due to more severe underlying liver dysfunction and potential development of portal vein tumor thrombosis. Complications of PHT, including ascites, esophagogastric varices (EGVs) and hepatic encephalopathy, pose common barriers to implementing curative, bridging, and palliative therapies for HCC, thereby portending a poor prognosis [2].

Recent advancements in systemic therapy have dramatically improved overall survival (OS) of HCC patients. However, emerging systemic therapies, especially those targeting the vascular endothelial growth factor receptor, may influence portal vein pressure and consequently increase the risk of variceal bleeding [1]. In the IMbrave150 study, Atezolizumab plus Bevacizumab (Atezo/Bev) was associated with a greater incidence of bleeding events than sorafenib (25.2% vs. 17.3%), including gastrointestinal bleeding (7% vs. 4.5%) and PHT-related acute variceal bleeding (AVB) (2.4% vs. 0.6%) [3]. The intricate interplay between HCC and PHT, along with the potential increased risk of variceal bleeding from systemic treatment, underscores the necessity of early identification and management in these patients [8–10].

A prior study profiling French physicians' management of PHT in patients with advanced HCC revealed substantial heterogeneity in screening and treatment practices. These findings highlight the need to improve clinician knowledge and conduct dedicated studies on the prevention of AVB in patients with HCC. Nevertheless, this survey has several limitations such as a small sample size and the presence of specific ethnic and disease-spectrum differences, limiting the generalizability of those findings [7]. Currently, no data are available on the attitude of Chinese physicians toward the screening and management of PHT in HCC. To address this knowledge gap, we conducted a nationwide, cross-sectional survey with the following objectives: (i) to characterize real-world practices for PHT in HCC across China; (ii) to identify factors influencing physicians' clinical decision-making, and (iii) to inform strategies for optimizing PHT management and improving patient outcomes.

2

Materials and methods

2.1

Study design and participants

This cross-sectional survey was conducted consecutively from March 1 to June 30, 2024, among hepatologists, gastroenterologists, digestive oncologists, and hepatobiliary surgeons from 132 secondary and tertiary hospitals across China.

2.2

Questionnaire design and development

The questionnaire was developed by two experts in digestive oncology and hepatology, aligned with the Baveno VII consensus [12] and the guidelines for the diagnosis and management of HCC [13]. It comprised four parts with a total of 57 items: physician and hospital demographics, PHT screening strategies in HCC, PHT management strategies in HCC, and the impact of PHT on systemic treatment decisions for HCC. Following its initial development, the questionnaire was revised based on feedback from a multidisciplinary panel of four experts. Subsequently, a pilot study involving 92 participants was conducted. The results prompted further revisions to several items, such as imposing an upper limit on the number of selections for multiple-choice questions to improve discriminative power and data quality; and categorizing patients screened for PHT as either newly diagnosed or follow-up cases to better reflect real-world practice.

The questionnaire was administered online using “Wen Juan Xing,” a predominant survey platform in China, and distributed through the ubiquitous social media application WeChat. To ensure data integrity under the platforms' inherent safeguards—including WeChat's real-name verification and Wen Juan Xing's single submission per IP address—each participant was required to use a unique WeChat account and was permitted only one submission.

2.3

Definitions

The Baveno VI criteria include liver stiffness measurement (LSM) and platelet count (Plt); patients with favorable criteria display LSM <20 kPa and Plt >150,000/mm³. The Baveno VII criteria include LSM and Plt; a patient with favorable criteria has LSM ≤15 kPa and Plt ≥150,000/mm³. A large EGV was defined as a diameter greater than or equal to 5 mm with or without a red sign, or a diameter less than 5 mm with a red sign, and a small EGV was described as a diameter less than 5 mm without a red sign.

2.4

Sample size and statistical analysis

According to our previous survey, the rate of endoscopic screening for PHT associated with HCC was approximately 30%. The sample size was calculated using the formula for cross-sectional studies: n=(Z^2 × P × (1-P))/(e^2), where Z=1.96 when α=.05, the assumed degree of variability of p=0.3, and e is an admissible error (which was 3% here). This calculation yielded a minimum of 897 participants. Accounting for a 20% potential non-response rate, the target sample size was 1075. Ultimately, 1,584 completed questionnaires from participating physicians were included in the analysis.

Descriptive statistics were performed for all the study variables. Categorical variables are presented as the number of cases and percentages, and were compared using the χ2 or Fisher’s exact test as appropriate. Differences were considered statistically significant at two-tailed p-values < 0.05. Data were analyzed and managed using SPSS (version 26.0; IBM SPSS, Armonk, NY, USA) and GraphPad Prism (version 10.0; Boston, MA, USA).

2.5

Ethical statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Review Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (No. 2024-0365). In addition, the survey was registered in the Chinese Clinical Trial Registry (ChiCTR2400084630).

3

Results

3.1

Characteristics of the study population

The study cohort consisted of 1,584 physicians. The distribution of specialties was as follows: gastroenterologists (763, 48.2%), hepatobiliary surgeons (339, 21.4%), digestive oncologists (299, 18.9%), and hepatologists (183, 11.6%). The participants were predominantly male (917, 57.9%), with a mean age of 37 years [32-43]. Most practiced in tertiary hospitals (1,131, 71.4%). 656/1584 (41.4%) of respondents were attending doctors, and over half reported more than 10 years of clinical experience. Regarding self-rated knowledge of PHT, 46.7% of participants reported a basic understanding, while only 15.8% claimed comprehensive knowledge (Table 1).

Table 1.

Characteristics of the survey population.

Variable	Overalln=1584	Hepatologistsn=183	Gastroenterologistsn=763	Digestive oncologistsn=299	Hepatobiliary surgeonsn=339	P
Male (%)	917 (57.9)	100 (54.6)	384 (50.3)	131 (43.8)	302 (89.1)	<0.001
Age (years) IQR	37 (32,43)	39 (33,45)	36 (32,42)	36 (30,42)	38 (32,44)
Years in practice (%)						0.001
<5	270 (17.0)	27 (14.8) ^a^,^b	139 (18.2) ^b	64 (21.4) ^b	40 (11.8) ^a
5-10	407 (25.7)	46 (25.1) ^a^,^b	209 (27.4) ^b	79 (26.4) ^a^,^b	73 (21.5) ^a
10-20	572 (36.1)	61 (33.3) ^a	274 (35.9) ^a	104 (34.8) ^a	133 (39.2) ^a
>20	335 (21.1)	49 (26.8) ^a	141 (18.5) ^b	52 (17.4) ^b	93 (27.4) ^a
Professional title (%)						0.003
Residency	347 (21.9)	38 (20.8) ^a^,^b^,^c	168 (22.0) ^c	83 (27.8) ^b	58 (17.1) ^a^,^c
Attending	656 (41.4)	76 (41.5) ^a	309 (40.5) ^a	132 (44.1) ^a	139 (41.0) ^a
Associate Senior	413 (26.1)	41 (22.4) ^a	202 (26.5) ^a^,^b	66 (22.1) ^a	104 (30.7) ^b
Senior	168 (10.6)	28 (15.3) ^a	84 (11.0) ^a	18 (6.0) ^b	38 (11.2) ^a
Hospital level (%)						0.165
Secondary	453 (28.6)	55 (30.1) ^a^,^b	212 (27.8) ^a^,^b	75 (25.1) ^b	111 (32.7) ^a
Tertiary	1131 (71.4)	128 (69.9) ^a^,^b	551 (72.2) ^a^,^b	224 (74.9) ^b	228 (67.3) ^a
Self-rated PHT knowledge (%)						0.016
Comprehensive	250 (15.8)	38 (20.8) ^a	123 (16.1) ^a	33 (11.0) ^b	56 (16.5) ^a
Basic	739 (46.7)	83 (45.4) ^a^,^b	381 (49.9) ^b	131 (43.8) ^a^,^b	144 (42.5) ^a
Simple	447 (28.2)	49 (26.8) ^a	201 (26.3) ^a	95 (31.8) ^a	102 (30.1) ^a
Some	136 (8.6)	11 (6.0) ^a	55 (7.2) ^a	36 (12.0) ^b	34 (10.0) ^a^,^b
Little	12 (0.8)	2 (1.1) ^a	3 (0.4) ^a	4 (1.3) ^a	3 (0.9) ^a

PHT, portal hypertension; IQR, Interquartile range.

a

indicate that the same sequence has a significant difference at the 0.05 level, and there is no difference if the letters are the same.

b

indicate that the same sequence has a significant difference at the 0.05 level, and there is no difference if the letters are the same.

c

indicate that the same sequence has a significant difference at the 0.05 level, and there is no difference if the letters are the same.

3.2

Screening strategies for PHT in patients with HCC

Screening for PHT in HCC patients was reported by 82.6% (1309/1584) of physicians. Complications of PHT, including gastrointestinal bleeding and ascites, were the most significant factors influencing this decision (Table 2). Computed tomography (CT) was the predominant screening modality in China, employed by 50.5% of respondents, whereas endoscopy was preferred by only 30.8% (Fig. 1A). Among non-invasive screening methods, the Baveno criteria was used by 28.0% of respondents; however, 20.3% of these users reported difficulty in applying them correctly, with no significant difference in proficiency between the Baveno VI and VII criteria (37.5% vs. 42.2%, P = 0.24). Regarding monitoring intervals, 48.3% of physicians opted for every 3-6 months, while 35.2% did not perform regular monitoring (Fig. 1B).

Table 2.

Screening and management of portal hypertension in hepatocellular carcinoma.

Variable	Overalln=1584	Hepatologistsn=183	Gastroenterologistsn=763	Digestive oncologistsn=299	Hepatobiliary surgeonsn=339	P
Screened for PHT (%)	82.6	87.4	81.7	85.6	79.6	0.060
Factors influencing screening^†(%)						0.018
History of cirrhosis	980 (41.2)	123 (43.8) ^a	486 (40.7) ^a	140 (36.3) ^a	231 (44.7) ^a
Clinical manifestations of PHT	1021 (42.9)	122 (43.4) ^a	518 (43.4) ^a	164 (42.5) ^a	217 (42)^a
Care about therapy	378 (15.9)	36 (12.8) ^a	189 (15.8) ^a	82 (21.2) ^b	69 (13.3) ^a
Screening modality (%)						<0.001
Clinical/laboratory tests	64 (4.9)	6 (3.8) ^a^,^b	19 (3.0) ^b	17 (6.8) ^a	22 (8.3) ^a
Endoscopy	400 (30.8)	64 (40.0) ^a	208 (33.4) ^a^,^b	72 (28.8) ^b	56 (21.1) ^c
Image (CT/MRI)	656 (50.5)	75 (46.9) ^a^,^b	325 (52.2) ^b^,c	99 (39.6) ^a	157 (59.0) ^c
HVPG measurement	109 (8.4)	11 (6.9) ^a	58 (9.3) ^a	19 (7.6) ^a	21 (7.9) ^a
LSM (VCTE)	70 (5.4)	4 (2.5) ^a	13 (2.1) ^a	43 (17.2) ^b	10 (3.8) ^a
Baveno criteria utilization (%)	28.0	22.5	22.5	39.8	33.0	<0.001
Baveno VI/VII (%)						0.706
Baveno VI	113 (37.5)	16 (50.0) ^a	50 (37.3) ^a	18 (38.3) ^a	29 (33.0) ^a
Baveno VII	127 (42.2)	12 (37.5) ^a	58 (43.3) ^a	18 (38.3) ^a	39 (44.3) ^a
Reported incorrect application	61 (20.3)	4 (12.5) ^a	26 (19.4) ^a	11 (23.4) ^a	20 (22.7) ^a
Provided primary prophylaxis (%)	1502 (94.8)	177 (96.7)	730 (95.7)	281 (94)	314 (92.6)	0.1
Preferred NSBB for primary prophylaxis (%)						<0.001
propranolol	1103 (73.3)	137 (77) ^a	537 (73.6) ^a	205 (73) ^a	224 (71.1) ^a
carvedilol	343 (22.8)	38 (21.3) ^a	185 (25.3) ^a	54 (19.2) ^a	66 (21) ^a
nadolol	56 (3.7)	3 (1.7) ^a	8 (1.1) ^b	20 (7.1) ^b	25 (7.9) ^b
Preferred NSBB for secondary prophylaxis (%)						<0.001
propranolol	971 (61.3)	119 (65)^a	489 (64.1) ^a	165 (55.2) ^a	198 (58.4) ^a
carvedilol	503 (31.8)	61 (33.3) ^a	251 (32.9) ^a	89 (29.8) ^a	102 (30.1) ^a
nadolol	106 (6.7)	2 (1.1) ^b	23 (3) ^b	43 (14.4) ^b	38 (11.2) ^b

CT, computed tomography; HVPG, hepatic venous pressure gradient; LSM, liver stiffness measurement; MRI, magnetic resonance imaging; NSBBs, non-selective beta-blockers; VCTE, vibration-controlled transient elastography.

a

indicate that the same sequence has a significant difference at the 0.05 level, and there is no difference if the letters are the same.

b

indicate that the same sequence has a significant difference at the 0.05 level, and there is no difference if the letters are the same.

†

This item allowed the selection of multiple responses.

Fig. 1.

Practices for portal hypertension screening and monitoring in hepatocellular carcinoma patients. Preferred modalities for PHT screening. (B) Intervals for routine PHT monitoring. Atezo/Bev, atezolizumab plus bevacizumab; HVPG, hepatic venous pressure gradient; LSM, liver stiffness measurement. The “Atezo/Bev subgroup” refers to patients receiving atezolizumab plus bevacizumab therapy. “Clinical tests” denotes the use of non-invasive serum biomarkers (e.g., platelet count, Aspartate aminotransferase-to-platelet ratio index). “Imaging” encompasses computed tomography, ultrasound, or magnetic resonance.

3.3

Management of PHT in patients with HCC

The vast majority of physicians (94.8%) report primary prophylaxis for PHT in HCC patients. Endoscopic findings, including varix size and red signs, were the most critical factors influencing prophylaxis strategies (Fig. 2A, Table 2). For small EGVs, close monitoring and non-selective beta-blockers (NSBBs) were the most common primary prophylaxis, while secondary prophylaxis primarily combined NSBB with endoscopic therapy (Table 3). Propranolol was the predominant NSBBs choice, with only 22.8% (343/1584) selecting carvedilol. (Table 2). For the management of AVB, 37.0% of physicians opted for endoscopic treatment, a proportion that decreased to 33.9% in patients with advanced HCC (Fig. 3A). Interventional treatment was selected by 24.7% for AVB management. Regarding transjugular intrahepatic portosystemic shunt (TIPS), 29% of the physicians identified HCC stage as the primary decision factor followed by liver function, TIPS-related complications, and patient-specific considerations (Fig. 2C, Fig. 3A). Factors associated with physicians' engagement in secondary prevention included the HCC stage, a history of AVB, and patient-specific considerations (Fig. 2B). The combination of NSBB with endoscopic therapy remained the cornerstone of secondary prevention, irrespective of varix size. While propranolol was the most frequently selected NSBB, the utilization of carvedilol was significantly higher in secondary than in primary prevention (31.8% vs. 22.8%, P < 0.01) (Table 2).

Fig. 2.

Physician-reported factors influencing clinical decision-making in hepatocellular carcinoma patients with portal hypertension. (A) Factors affecting the decision to initiate primary prophylaxis for EGVs. (B) Factors affecting the decision to initiate secondary prophylaxis. (C) Factors affecting the decision to place a TIPS. AVB, acute variceal bleeding; EGVs, esophagogastric varices; HCC, hepatocellular carcinoma; PHT, portal hypertension; TIPS, transjugular intrahepatic portosystemic shunt. “Patient factors” encompass physical status and socioeconomic considerations (e.g., financial capacity). “TIPS side effects” include concerns regarding tumor dissemination and liver failure.

Table 3.

Primary and secondary prevention means of portal hypertension in hepatocellular carcinoma.

Management Strategy	PrimaryPrevention means		SecondaryPrevention means		P
	Small EGVs (%)	Large EGVs (%)	Small EGVs (%)	Large EGVs (%)	<0.001
NSBB alone	620(41.2)	238(15.8)	420(26.5)	137(8.6)
Endoscopy alone	175(11.6)	358(23.8)	387(24.4)	298(18.8)
Combination therapy	146(9.7)	794(52.8)	527(33.3)	888(56.1)
TIPS	15(1.0)	69(4.6)	58(3.7)	224(14.1)
Close monitoring	548(36.4)	45(3)	192(12.1)	37(2.3)

EGVs, esophagogastric varices; NSBBs, non-selective beta-blockers; TIPS, transjugular intrahepatic portosystemic shunt.

Fig. 3.

Treatment preferences for acute variceal bleeding and the impact of acute variceal bleeding history on systemic therapy selection in hepatocellular carcinoma. (A) Physician preferences for AVB management modalities across three patient cohorts: the overall HCC population, patients with advanced HCC, and patients treated with Atezo/Bev. (B) Physician preferences for first-line systemic therapy regimens for HCC, stratified by a history of AVB. Atezo/Bev, atezolizumab plus bevacizumab; AVB, acute variceal bleeding; ICI, immune checkpoint inhibitor; TKI, tyrosine kinase inhibitor.

To investigate factors influencing the correct application of the Baveno VI/VII criteria, we performed a subgroup analysis. The results revealed significant variations by specialty, with hepatologists, gastroenterologists, and surgeons demonstrating higher proficiency than oncologists. Furthermore, correct application was positively associated with educational attainment, as physicians with bachelor's, master's, or doctoral degrees outperformed those with associate degrees (Supplementary Table 1).

3.4

Effect of PHT on HCC systemic treatment

The impact of PHT on subsequent treatment was considered by 92.6% (1,467/1,584) of physicians prior to initiating systemic therapy. For general HCC, Atezo/Bev was the preferred regimen for 32.9% of respondents. However, a history of variceal bleeding within six months resulted in a pronounced reduction in this preference (19.6% vs. 32.9%, P<0.001). This preference was observed in populations with a history of variceal bleeding over 6 months. A total of 22.7%, 22.3%, and 18.8% of physicians reported uncertainty regarding the optimal systemic therapy for general HCC, HCC with a history of AVB <6 months, and AVB ≥6 months, respectively (Fig. 3B). These respondents were excluded from subsequent analyses to minimize bias. Among the remaining physicians (n=912), those who preferred Atezo/Bev demonstrated significantly higher rates of PHT screening and shorter monitoring intervals. Notably, while 35.2% of all physicians did not perform regular PHT monitoring, this proportion was significantly lower among those using the Atezo/Bev regimen (3.4% vs. 35.2%, P<0.01) (Fig. 1B).

In the event of AVB during Atezo/Bev therapy, 79.5% of physicians recommended permanent discontinuation of bevacizumab, whereas 20.5% advocated for its continuation after hemostasis was achieved. Preferences for the interval between endoscopy therapy and bevacizumab initiation varied: 34.5% suggested a 4-week, 27.0% proposed 2 weeks, 22.6% advocated for 3 months, and 15.1% believed no interval was necessary (Table 4).

Table 4.

Management strategies and timing intervals for acute variceal bleeding during Atezolizumab-Bevacizumab therapy (n=912).

Variable	Respondents, n (%)
Management strategy for AVB during Atezo/Bev therapy
Permanent discontinuation of Atezo/Bev	368 (40.4)
Continue Atezo/Bev after variceal eradication	187 (20.5)
Continue atezolizumab alone (discontinue bevacizumab)	263 (28.8)
Switch to an alternative systemic regimen	80 (8.8)
Other	14 (1.5)
Recommended interval between EVL and Atezo/Bev re-initiation
No interval required	138 (15.1)
2 weeks	246 (27.0)
1 month	315 (34.5)
3 months	206 (22.6)
Other	7 (0.8)

Atezo/Bev, atezolizumab plus bevacizumab; AVB, acute variceal bleeding; EVL, endoscopic variceal ligation.

3.5

Disparities between different provinces, hospitals and departments

To assess the impact of regional medical education levels on clinical decision-making, we stratified provinces into tertiles based on their educational attainment and compared the top and bottom tertiles. Physicians from top-tier provinces demonstrated superior performance in the comprehensive management of PHT in HCC, exhibiting significantly higher screening rates, shorter monitoring intervals, and greater proficiency in systemic treatment selection (Supplementary Table 2).

No significant differences were observed between secondary and tertiary hospitals in PHT screening rates or methodologies. Both types of institutions predominantly monitored PHT in patients with HCC at intervals of less than six months. However, adherence to the protocol for regular PHT monitoring was significantly lower among physicians in secondary hospitals. (39.3% vs. 33.5%, P<0.01). Primary prophylaxis rates and their influencing factors were comparable across hospital grades. In the selection of NSBBs, tertiary hospital physicians showed a greater preference for carvedilol, while their counterparts in secondary hospitals more frequently chose propranolol (Supplementary Table 3).

Practice patterns varied considerably by specialty. Hepatologists demonstrated the highest rates of endoscopic screening (40% vs. 30.8% overall) and primary prophylaxis (96.7%). They most frequently prescribed NSBBs for small varices (50.8%), whereas digestive oncologists and surgeons preferred monitoring. Gastroenterologists performed the most endoscopic treatments for AVB (41.6%). Despite digestive oncologists reporting the highest Baveno criteria utilization (39.8%), proficiency did not differ significantly by specialties. Hepatobiliary surgeons relied most on imaging (59%) and had the highest rate of irregular monitoring (42.2%) (Fig. 4).

Fig. 4.

Variation in portal hypertension management practices across different departments. Proportion of endoscopy utilization in PHT screening. (B) Baveno VI/VII utilization rate. (C) Primary prevention rate. (D) Proportion of NSBB alone in primary prevention of small EGVs. (E) Proportion considering HCC systemic treatment for primary prevention. (F) Proportion of HCC patients not regularly monitored for PHT. EGVs, esophagogastric varices; NSBB, non-selective beta-blockers.

4

Discussion

The effective screening and management of PHT are critical components of comprehensive care of patients with HCC. Yet, international management strategies exhibit substantial heterogeneity, influenced by diverging clinical experiences and socioeconomic contexts. Data specifically reflecting the attitudes and practices of Chinese physicians have been lacking. Our study therefore represents the first large-scale investigation to systematically examine these practices across China.

Our study found that 82.6% of physicians in China screen for PHT in HCC patients, a rate substantial yet lower than the universal screening reported (100%) in another study [7]. Complications of PHT are the primary motivators for physicians to conduct screening. Imaging is the main modality for PHT screening in patients with HCC across China because of its non-invasive and reproducible characteristics. While cost-effective for varix risk stratification, CT is limited in detecting small varices and the high-risk “red signs” [10]. Consequently, endoscopy remains the diagnostic gold standard. Its utilization in China, however, was notably low (30.8%), in sharp contrast to the rate of approximately 80% reported in a French survey [7]. We posit that this disparity may be attributed to the poorer performance status of patients with advanced HCC, constrained endoscopic resources in certain regions, and the fact that a significant proportion of HCC patients in China are managed outside hepatology departments. Based on these findings, we recommend a tiered approach: routine PHT assessment at initial HCC diagnosis, utilizing CT as an initial tool, with mandatory and expedited endoscopic evaluation for high-risk patients (e.g., those with a history of variceal hemorrhage, portal vein tumor thrombosis, or indicative imaging features).

Primary prophylaxis for PHT was universally acknowledged by over 90% of Chinese physicians. Nevertheless, a significant guideline-practice gap emerged in its execution. While NSBBs were consistently employed as the cornerstone of prophylaxis, aligning with the Baveno VII consensus, a critical divergence was observed in the specific agent selected. Specifically, propranolol remained the predominant choice (73.3%), whereas the guideline-endorsed first-line agent, carvedilol—preferred for its dual α1- and β-adrenergic blockade and superior hemodynamic efficacy [11], was utilized by only 22.8% of physicians. Interestingly, this preference shifted significantly in secondary prophylaxis, with carvedilol usage rising to 31.8% (P < 0.01). This may reflect heightened therapeutic vigilance and a greater adherence to guidelines when managing patients with a history of variceal bleeding. The overall underutilization of carvedilol in primary prevention likely stems from a lack of awareness and familiarity with this newer agent among clinicians. Enhancing physician education on the pharmacological advantages of carvedilol is thus crucial to optimize PHT management in China. Furthermore, while secondary prophylaxis with combined NSBB and endoscopic variceal ligation (EVL) is guideline-concordant, its timing relative to bevacizumab initiation requires careful coordination. The ulceration induced by EVL must be balanced against the bleeding risk from bevacizumab. A 4-week interval is generally recommended; however, this may be shortened to 7–14 days depending on procedural invasiveness and ulcer healing status [12,13].

Our investigation revealed a notably low rate of endoscopic treatment for variceal bleeding (37% overall), which was further reduced in patients with advanced HCC (33.9%). This clinical reality is likely multifactorial. First, the predominance of advanced-stage HCC at diagnosis means a substantial proportion of patients have a fragile physical condition, rendering them suboptimal candidates for invasive procedures. Second, this challenge is compounded by the uneven distribution of endoscopic expertise and resources, particularly in economically underdeveloped regions of China, which can preclude timely intervention. The low rate of endoscopic treatment was consistent with the low rate of endoscopic screening for EGVs within the HCC patient population in China.

TIPS is an established intervention for uncontrolled variceal bleeding and ascites [8]. However, its application in HCC patients is often constrained by theoretical risks of post-procedural liver failure and tumor dissemination [14], Moreover, our survey revealed that Chinese physicians share similar concerns. Contrary to these apprehensions, a growing body of evidence suggests that TIPS can be safely performed in selected HCC patients. A systematic review indicated that TIPS resolved bleeding or ascites in over 60% of patients, with procedural bleeding occurring in less than 5% of patients, and no fatal complications reported [15]. Furthermore, a large retrospective study by Qiu et al. achieved a procedural success rate of 97% without serious procedure-related complications [16]. While the association between TIPS and tumor dissemination remains incompletely defined, the procedure can effectively alleviate ascites and may even serve as a bridge to subsequent locoregional therapies or liver transplantation under close monitoring [16,17]. Therefore, we propose that TIPS should be reconsidered as a viable option for HCC patients, particularly those with failed endoscopic hemostasis or refractory ascites, when performed in specialized centers with careful patient selection.

Recent advancements in systemic treatment for HCC have markedly improved OS. Compared with sorafenib, Atezo/Bev have been recognized as first-line treatment for unresectable HCC, with OS times superior to those of sorafenib in the IMbrave 150 trial (median OS of 19.2 months and 10.7 months, respectively) [6]. However, bevacizumab administration is associated with an increased risk of impaired wound healing and bleeding events, particularly gastrointestinal bleeding [18–20]. A striking international variation was evident in the management of AVB during Atezo/Bev therapy, with rates of permanent discontinuation being substantially higher in China (79.5%) than those reported in France (14.7%). Our data reflect this clinical challenge, demonstrating significant physician reluctance to prescribe Atezo/Bev following recent AVB. This finding underscores the critical need for proactive risk mitigation. Robust evidence indicates that preemptive endoscopic intervention prior to bevacizumab initiation can substantially reduce bleeding risk [19]. Therefore, systematic screening and management of esophageal varices should be integrated as a mandatory component of the treatment pathway for any HCC patient being considered for anti-angiogenic therapy.

Our survey identified prominent specialty-specific variations in PHT management, underscoring the need for a more unified approach. Hepatologists demonstrated greater utilization of endoscopy, digestive oncologists focused on systemic therapy implications, while hepatobiliary surgeons exhibited less consistent monitoring. These divergent perspectives highlight the critical value of a standardized, multidisciplinary team (MDT) model for managing HCC with PHT. The multidisciplinary team should develop tailored protocols for complex cases through structured interdepartmental discussion. Furthermore, to address the disparities linked to regional education levels, we advocate establishing supportive partnerships between advanced and regional medical centers. Such initiatives can enhance early patient identification and facilitate timely referrals, ultimately improving care standardization across diverse healthcare settings.

This study has several limitations. First, the survey was restricted to closed-ended questions and did not include open-ended responses, which potentially limits the diversity of clinical screening and management practices. Nonetheless, the questionnaire was meticulously designed in alignment with the established guidelines for PHT and HCC, ensuring that it encompassed the core screening and management strategies for PHT. Second, the questionnaire did not incorporate the staging of HCC, as clinical data indicated that majority of HCC cases are diagnosed at an advanced stage. Nevertheless, our analysis specifically examined management differences in variceal bleeding between the general and advanced HCC populations. Third, we only conducted the survey in China, reflecting its unique healthcare infrastructure and socioeconomic context, which limits direct extrapolation to other regions. Future multinational studies are warranted to validate and compare these findings across diverse healthcare systems.

5

Conclusions

This national study underscores a critical gap in the screening and management of PHT among Chinese physicians compared to Western counterparts, particularly in the underutilization of endoscopy. While primary and secondary prophylaxis strategies generally aligned with the Baveno VII consensus, we identified significant heterogeneity in practice patterns across medical specialties and hospital settings. Nevertheless, although over 90% of physicians recognized the effect of systemic treatment on PHT, a significant proportion remains unclear about the specific systemic treatment modalities and its exact impacts, particularly among those from regions with less medical education and secondary hospitals. Strategies for screening and management remain very heterogeneous among different departments. These findings highlight an imperative for targeted educational initiatives and standardized multidisciplinary protocols to optimize care and improve outcomes for HCC patients in China.

Funding

This work was supported by the National Key Research and Development Program of China [2023YFC2507405] and the Interdisciplinary Innovative Talents Foundation from Renmin Hospital of Wuhan University [JCRCFZ-2022-01].

Authors’ contributions

Consensus statement concept and design (XYL, HYC), investigation and data collection (QQZ, STW, YPW, YXC), data analysis (XYL, HYC), technical support (QQZ, STW, YPW), drafting the manuscript (XYL, HYC), study supervision (JS, MKC). All authors have read and approved the final version of the manuscript and agreed to its publication.

Availability of data and materials

The data that support the findings of this study are available on request from the corresponding author.

Uncited References

[4,5]

Acknowledgements

The authors gratefully acknowledge all participating physicians and institutions for their contributions.