Validation of A-VENA Criteria in Distinguishing Tumor from Bland Portal Vein Thrombus in Liver Transplant Candidates with Hepatocellular Carcinoma

doi:10.21203/rs.3.rs-2737342/v1

Download PDF

Research Article

Validation of A-VENA Criteria in Distinguishing Tumor from Bland Portal Vein Thrombus in Liver Transplant Candidates with Hepatocellular Carcinoma

https://doi.org/10.21203/rs.3.rs-2737342/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: The A-VENA criteria for distinguishing tumor portal vein thrombus(PVT) from bland PVT has not been validated.

Methods: Clinical data from January 2018 to December 2021 of Hepatocellular Carcinoma (HCC) with PVT patients who underwent liver transplantation (LT) were retrospectively collected. The diagnostic performance of A-VENA criteria was analyzed by receiver operating characteristic (ROC) curve. Diagnostic factors for tumor PVT were evaluated by multivariable analysis. Modified criteria were proposed based on the results of multivariable analysis.

Results: Of 49 HCC with PVT patients, 26 were pathologically confirmed as tumor PVT and 23 were confirmed as bland PVT. Tumor PVT group has a larger proportion of ‘AFP level >1000 ng/mL’ (61.5% vs 17.4%, P=0.002), ‘enhancement of PVT’ (92.3% vs 13.0%, P<0.001), ‘neovascularity’ (65.4% vs 8.7%, P<0.001), and ‘PVT adjacent to HCC’ (88.5% vs 21.7%, P<0.001). Portal vein expansion was not statically different in two groups (50.0% vs 34.8%, P=0.285). The area under the ROC curve (AUC) was 0.918 for A-VENA criteria (≥3) in distinguishing tumor PVT from bland PVT. Neovascularity (P=0.986) were not independently significant diagnostic factors. The AFP level > 400 ng/mL, enhancement of PVT, and adjacent to HCC were significant in multivariable analysis. The presence of ≥2 criteria of the three (AEA criteria) best characterized tumor PVT with an AUC of 0.978, 100% sensitivity, 95.7% specificity, 96.3% positive predictive value, and 100% negative predictive value.

Conclusions: A-VENA criteria could accurately distinguish tumor PVT from bland PVT in LT candidates. Proposed AEA criteria (the presence of at least 2 of the following: AFP >400 ng/dL, PVT enhancement, and adjacent to HCC) is an alternative tool for the diagnosis of tumor PVT.

A-VENA Criteria

Hepatocellular carcinoma

Liver transplantation

Portal vein thrombus

Portal vein thrombus (PVT) occurs in approximately 5–26%(1) of patients with cirrhosis at evaluation or at the time of liver transplantation (LT). In patients with hepatocellular carcinoma (HCC), distinguishing tumor PVT from bland PVT is significant. Bland PVT is mainly caused by sluggish portal blood flow and a hypercoagulable state, which does not affect the long-term survival of HCC patients or the recurrence rate after LT(2). In contrast, tumor PVT is associated with a poor survival time of 2.7 months without treatment(3) and 5.6 to 8.1 months after sorafenib treatment(4). Due to the high rate of HCC recurrence after transplantation, tumor PVT is considered a contraindication for LT(5, 6). Thus, it is critical to identify tumor PVT in LT candidates to exclude these patients from LT. It is also important to avoid misdiagnosing bland PVT as tumor PVT, which would result in the loss of transplantation opportunities for patients with potentially curable disease.

Traditionally, the gold standard for diagnosing tumor PVT and bland PVT is the pathological examination of a fine-needle biopsy(7–9). However, biopsy is an invasive procedure that is often not feasible in routine clinical practice, especially for patients with coagulation dysfunction(10). Several researchers have proposed noninvasive diagnostic criteria based on preoperative imaging, including contrast-enhanced computed tomography (CT)(11–13), magnetic resonance (MR) imaging(14–16) and contrast-enhanced ultrasound (CEUS)(17–19). However, these studies have not been validated, and the criteria are not widely used. Generally, tumor PVT is characterized as a hyperenhancing soft tissue in the artery phase adjacent to an HCC lesion with neovascularity and portal vein (PV) expansion in some cases; bland PVT always manifests as a hypodense filling defect without contrast enhancement at the artery phase and is not adjacent to the primary tumor(10–13, 20). In 2019, Sherman et al.(21) proposed the CT/MR-based A-VENA criteria to distinguish tumor PVT from bland PVT in LT candidates based on the presence of ≥ 3 of the following: an alpha-fetoprotein (AFP) level > 1000 ng/dL; venous expansion; thrombus enhancement; neovascularity; and PVT adjacent to HCC. In their study, the A-VENA criteria accurately characterized tumor PVT with 100% sensitivity and 93.6% specificity. Subsequently, the A-VENA criteria has been widely recognized and employed for diagnosing tumor PVT(22–24). However, the A-VENA criteria has a major limitation: tumor PVT in Sherman’s study was mainly diagnosed radiologically, and pathological evidence was absent(21), potentially leading to circular reasoning. In addition, to the best of our knowledge, the A-VENA criteria has not yet been validated externally. Both of these factors limit the wide use of the A-VENA criteria in clinical practice.

As several studies have revealed that patients with HCC and tumor PVT can benefit from LT after successful downstaging therapy(23–27), our center conducted a preliminary study to investigate the outcome of selective LT after downstaging therapy for HCC with tumor PVT from January 2018, and the research data have not yet been published. Therefore, we enrolled a patient cohort with pathologically confirmed HCC and PVT who underwent LT after downstaging. The aim of the present study was to use this cohort to validate the A-VENA criteria.

Patients

This was a single-center retrospective cohort study. Clinical data were obtained from the clinical database for 174 patients with HCC who underwent LT at the Hepato-Biliary-Pancreatic Center of Beijing Thsinghua Changgung Hospital from January 2018 to December 2021 and were retrospectively reviewed. Forty-nine of these patients were diagnosed with PVT before LT. In the explant pathology examination, 26 patients were confirmed to have tumor PVT (all down-staged before LT), and 23 had bland PVT (Fig. 1). The current study was approved by the the Ethics Committee of Beijing Tsinghua Changgung Hospital Ethics Committee. Because of the retrospective nature of the study and because no intervention was employed, the requirement for informed consent was waived.

Data Collection

The preoperative clinical variables were collected, including age, sex, the etiology of the patient’s liver disease, the Model for End-Stage Liver Disease (MELD) score, the Child-Turcotte-Pugh (CTP) score and the AFP level at the time of PVT diagnosis (before downstaging). The pre-LT three-phase enhanced CT at the time of PVT diagnosis was reviewed by an experienced radiologist. The parameters regarding the tumor number, the diameter of the largest tumor, the sum of the tumor diameter, the PVT location and the PV diameter were measured and recorded. The four imaging characteristics of the A-VENA criteria(21) were determined as follows: venous expansion (defined as a portal vein diameter ≥20 mm), enhancement of PVT (defined as a difference of Hounsfield units >20 at artery phase), neovascularity (inside the PVT) and whether the PVT was adjacent to HCC (distance <2 cm). The A-VENA score was then calculated combined with AFP level, and patients were subsequently grouped into the A-VENA-positive (score ≥3) and A-VENA-negative (score <3) groups. The explant pathology of the PVT was reviewed by an experienced pathologist as the gold-standard for diagnosing tumor PVT and bland PVT.

Statistical Analyses

Statistical analyses were performed using SPSS version 23.0 (IBM Corp., Armonk, NY, USA). Continuous variables are presented as the medians with interquartile ranges (IQRs). Categorical variables are expressed as numbers and proportions. Mann‒Whitney U tests were used to evaluate the significance of the differences between 2 groups for continuous variables. The differences between 2 groups of categorical variables were assessed using the chi-square or Fisher’s exact tests. The cutoff value for the diagnosis of tumor PVT was determined using a receiver operating characteristic (ROC) curve. The optimal cutoff value was defined as the point on the ROC curve that maximized the Youden index (sensitivity + specificity − 1). Univariable and multivariable regression analyses were performed to investigate the independent diagnostic factors of tumor PVT. The diagnostic performance of the A-VENA criteria was evaluated by the area under the receiver operating curve (AUC). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Two-sided P values <0.05 were considered statistically significant.

Baseline Characteristics

The baseline characteristics of the 49 patients with PVT are shown in Table 1. The median age at the diagnosis of PVT was 54 years, and 98% of the patients were male. The most common etiology of liver disease was hepatitis B virus infection (85.7%). The median MELD and CTP scores were 7.11 and 7, respectively. The median diameter of the largest tumor was 4.8 cm, and the median sum of the tumor diameters was 7.00 cm. Most patients (55.1%) had 1 tumor at the time of PVT diagnosis, 26.5% had 2 tumors, and 18.4% had 3 or more tumors. The median AFP level when PVT was diagnosed was 89.33 ng/mL. Regarding the PVT location, 59.2% were located in the main PV, 26.5% in the right or left PV and 14.3% in the PV segmental branches.

Comparing the Parameters of Tumor PVT and Bland PVT

The comparison of tumor PVT (n=26) and bland PVT (n=23) is shown in Table 1. The age, sex, etiology of the patient’s liver disease, MELD score, CTP score and PVT location did not differ significantly between the two groups (all P>0.05). The diameter of the largest tumor in the PVT group was greater than that in the bland PVT group (median: 7.70 cm versus 2.50 cm; P<0.001); the sum of the tumor diameters was also greater in the tumor PVT group than in the bland PVT group (median: 9.25 cm versus 3.10 cm; P=0.001), as was the AFP level at PVT diagnosis (median: 1270.73 ng/mL versus 14.86 ng/mL, P=0.001). Of note, a larger PV diameter was observed in the tumor PVT group (median: 20.0 mm versus 17.0 mm), but this difference was not statistically significant (P=0.097).

External Validation of the A-VENA Criteria for Distinguishing Tumor PVT from Bland PVT

The proportions of the five characteristics were investigated in the tumor PVT and bland PVT groups to validate the A-VENA criteria (Table 1). In the tumor PVT group, 16 patients (61.5%) had an AFP level >1000 ng/mL, while four patients (17.4%) met this criterion in the bland PVT group (P=0.002). Neovascularity was observed in 17 patients (65.4%) in the tumor PVT group and two patients (8.7%) in the bland PVT group (P<0.001). The proportion of patients whose PVT was adjacent to HCC was also significantly larger in the tumor PVT group (n=23, 88.5%) than in the bland PVT group (n=5, 21.7%; P<0.001). Almost all patients in the tumor PVT group showed an enhancement of PVT (n=24, 92.3%), while only three patients (13.0%) demonstrated PVT enhancement in the arterial phase in the bland PVT group (P<0.001). The proportion of patients with venous expansion was not significantly different between the tumor (n=13, 50.0%) and bland PVT groups (n=8, 34.8%).

In total, 24 patients (92.3%) in the tumor PVT group were classified into the A-VENA-positive group (score ≥3), and 21 patients (91.3%) in the bland PVT group were classified into the A-VENA-negative group (score <3); thus, the A-VENA criteria correctly diagnosed over 90% of PVTs. However, two patients (7.7%) with tumor PVT were classified as A-VENA negative (score <3) and misdiagnosed with bland PVT. Two patients (8.7%) with bland PVT were classified as A-VENA positive (score ≥3) and misdiagnosed with tumor PVT. ROC curve analyses were performed to evaluate the diagnostic performance of the A-VENA score (Fig. 2a) and A-VENA positivity (score ≥3) (Fig. 2b). The AUC of the A-VENA score for the differentiation of tumor PVT from bland PVT was 0.957 (95% CI: 0.893-1.000; P<0.001). The optimal cutoff value was 3, with a sensitivity of 92.7% and a specificity of 91.3%. The AUC of A-VENA positivity (score ≥3) was 0.918 (95% CI: 0.828-1.000; P<0.001).

Univariable and Multivariable Analyses to Investigate the Independent Factors to Distinguish Tumor PVT from Bland PVT

For further univariable and multivariable analyses, ROC curve analyses were performed to find the optimal cutoff value of the AFP level (Fig. 3a) and the diameter of the largest tumor at PVT diagnosis (Fig. 3b). The AUC of the AFP level for the diagnosis of tumor PVT was 0.773 (95% CI: 0.633-0.912; P<0.001), and the optimal cutoff value of 322 ng/mL had a sensitivity of 69.2% and a specificity of 82.6%. The AUC of the diameter of the largest tumor was 0.819 (95% CI: 0.702-0.939; P<0.001), and the optimal cutoff value of 4.0 cm had a sensitivity of 84.6% and a specificity of 69.6%. For the convenience of further analyses, cutoff values of 400 ng/mL (AFP level) and 5.0 cm (diameter of the largest tumor) were determined. Both univariable and multivariable analyses were performed to identify the significant diagnostic factors associated with tumor PVT (Table 3). The multivariable analysis showed that an AFP level >400 ng/mL (odds ratio [OR]: 26.108; 95% confidence interval [CI]: 1.114-611.717; P=0.043), enhancement of PVT (OR: 98.712; 95% CI: 2.324-4192.171; P=0.016) and PVT adjacent to HCC (OR: 59.264; 95% CI: 1.950-1801.320; P=0.019) were independent diagnostic factors of tumor PVT. However, the diameter of the largest tumor >5 cm and neovascularity were significant in the univariable analysis (P=0.004 and P<0.001, respectively) but not in the multivariable analysis (P=0.721 and P=0.986, respectively). Venous expansion was not significant in the univariable analysis (P=0.285).

Establishment of Modified Diagnostic Criteria to Distinguish Tumor PVT from Bland PVT

A modified diagnostic score (AEA score) to distinguish tumor from bland PVT was established by three independent diagnostic factors: an AFP level >400 ng/mL, enhancement of PVT and PVT adjacent to HCC (1 count for each presence). ROC curve analysis (Fig. 4a) was used to evaluate the diagnostic performance of the AEA score (AUC: 0.967; 95% CI: 0.903-1.000; P<0.001). The optimal cutoff value was determined to be 2 (sensitivity: 100%, specificity: 95.7%). The AUC of an AEA score ≥2 in distinguishing tumor from bland PVT was 0.978 (P<0.001). Specific information on the diagnostic power of AEA scores ≥1, ≥2 and ≥3 for the differentiation of tumor from bland PVT is shown in Table 3. An AEA score ≥2 had the highest sensitivity of 100%, specificity of 95.7%, PPV of 96.3% and NPV of 100%; thus, an AEA score ≥2 was determined to be the AEA criteria. All tumor PVT cases (26/26, 100%) and almost all bland PVT cases (22/23, 95.7%) in the current cohort were correctly diagnosed by the AEA criteria.

In the current study, we externally validated the A-VENA criteria in a pathologically confirmed PVT cohort who underwent LT. Our results revealed that the A-VENA criteria show excellent performance capabilities for distinguishing tumor PVT from bland PVT in HCC patients. Furthermore, based on the results of the multivariable analysis, alternative criteria (the AEA criteria) was established and showed better simplicity and equivalent efficacy for differentiating tumor PVT from bland PVT.

Although several studies indicated that LT after downstaging therapy could significantly improve the outcome of HCC patients with tumor PVT(23–27), this condition was still considered a contraindication of LT in the current guidelines due to the high recurrence rate of HCC after LT in such patients(5, 6). Therefore, distinguishing tumor PVT from bland PVT is critical since the staging and treatment choices for the two diseases are different. The gold standard for diagnosing tumor PVT depends on pathological evidence obtained from fine-needle biopsy analysis(7–9); however, biopsies are often not feasible in routine clinical practice due to the invasive nature of the procedure(10), especially in patients with coagulation dysfunction.

Several researchers have attempted to establish noninvasive diagnostic criteria for tumor PVT based on CT or MR imaging. Tublin et al.(12) retrospectively analyzed 58 patients with cirrhosis and PVT and found that the coexistence of ‘the main PVT diameter ≥ 23 mm’ or ‘PVT neovascularity’ had a sensitivity of 86% and a specificity of 100% for diagnosing tumor PVT. Sandrasegaran et al.(16) proposed an MR-based criterion in which the presence of at least two of the three following had a sensitivity of 100% and specificity of 90% for the diagnosis of tumor PVT: distance from tumor to PVT less than 2 cm, HCC size greater than 5 cm, and arterial enhancement of PVT. According to the Liver Imaging Reporting And Data System (LI-RADS)(20), the only criterion of tumor PVT was unequivocal arterial phase hyperenhancing soft tissue in the PV. In 2019, Sherman et al.(21) retrospectively reviewed a cohort of LT candidates with HCC and PVT (n = 59) and analyzed the diagnostic power of previously suggested radiographic features of tumor PVT and AFP > 1000 ng/dL. As a result, they established the A-VENA criteria, including the presence of at least 3 of the following: AFP > 1000 ng/dL, venous expansion, thrombus enhancement, neovascularity, and PVT adjacent to HCC. The A-VENA criteria had a sensitivity of 100% and a specificity of 93.6% in distinguishing tumor PVT from bland PVT. The A-VENA criteria combined four imaging characteristics proposed by former studies and a novel biological feature (AFP level) and therefore was considered as the best criteria for diagnosing tumor PVT and has been widely used in subsequent studies(22–24). However, pathological evidence for patients with tumor PVT was unavailable in Sherman’s study since these patients did not receive LT, which was a major limitation. In addition, the A-VENA criteria has not been validated externally. Both of these factors limit the wide use of the A-VENA criteria in clinical practice.

Hence, we conducted the current study to validate the A-VENA criteria by retrospectively analyzing a cohort of 49 patients with HCC and pathologically confirmed PVT who underwent LT. The results showed that the A-VENA criteria has promising diagnostic power. An A-VENA score ≥ 3 has an AUC of 0.918, a sensitivity of 92.7% and a specificity of 91.3% in distinguishing tumor PVT from bland PVT. We also examined the five specific features included in the A-VENA criteria. Compared with bland PVT, the tumor PVT cohort had a higher rate of ‘AFP level > 1000 ng/mL’ (61.5% vs. 17.4%, P = 0.002), ‘enhancement of PVT’ (92.3% vs. 13.0%, P < 0.001), ‘neovascularity’ (65.4% vs. 8.7%, P < 0.001), and ‘PVT adjacent to HCC’ (88.5% vs. 21.7%, P < 0.001). However, a significantly higher rate of ‘venous expansion’ in the tumor PVT cohort was not observed in the current study (50.0% vs. 34.8%). Additionally, the diameter of the main PV trunk was not significantly different in the tumor PVT and bland PVT groups, which was inconsistent with former studies(12, 21, 28). The probable cause of these differences was the different baselines of the cohorts among different studies, and bland PVT may appear as a luminal defect with PV expansion in the acute phase and for this reason PV expansion may not a proper diagnostic factor for tumor PVT(29, 30). Further large-scale studies are required to clarify the diagnostic power of PV expansion and PV diameter.

Including ‘venous expansion’ might not improve the diagnostic power of the A-VENA criteria. We investigated the independent diagnostic factors of tumor PVT via multivariable analysis and subsequently modified the A-VENA criteria. The results showed that AFP level > 400 ng/mL, enhancement of PVT and PVT adjacent to HCC were independently related to tumor PVT. Neovascularity was significant in the univariable analysis but not in the multivariable analysis. Hence, we propose modified criteria (the AEA criteria) as follows: AFP > 400 ng/mL, enhancement of PVT and PVT adjacent to HCC. The presence ≥ 2 of AEA criteria has an AUC of 0.978, a sensitivity of 100%, a specificity of 95.7%, a PPV of 96.3% and an NPV of 100%. The AEA criteria is a CT-based noninvasive diagnostic standard with high accuracy for diagnosing tumor PVT.

In Sherman’s study, the A-VENA criteria was mostly based on CT imaging (88.1% CT versus 11.9% MR). For the best consistency with the original study, the current study was entirely based on the three-phase enhanced CT scan. As a routine examination to diagnose and stage HCC, contrast-enhanced CT has the advantages of being noninvasive, objective and repeatable with stable parameters and therefore is considered a satisfactory modality in diagnosing PVT. In recent years, emerging imaging modalities, such as CEUS(17–19) and gadoxetic acid–enhanced MR imaging(14), have been used to diagnose tumor PVT and bland PVT. Further studies should be conducted to compare the diagnostic power of this modality with that of contrast-enhanced CT.

There are several obvious limitations in this study. First, there was likely selection bias because of the retrospective nature of the study, and the sample size was small and the statistical power of multivariate analysis was limited. Second, in the current study, tumor PVT was present at initial diagnosis in several patients, which might not be consistent with the design of Sherman et al.’s study, in which PVT occurred in patients on the LT waitlist. Finally, the conclusion of this single-center study needs to be further validated externally and confirmed in a large-scale prospective multicenter study. However, our study supported the A-VENA criteria by providing pathological evidence of PVT and therefore resolved the most serious limitation of the original research.

In conclusion, the A-VENA criteria can accurately distinguish tumor PVT from bland PVT in LT candidates. The proposed AEA criteria (the presence of at least 2 of the following: AFP > 400 ng/dL, PVT enhancement, and PVT adjacent to HCC) is an alternative tool for diagnosing tumor PVT.

AFP, alpha fetoprotein; AUC, area under the curve; CEUS, contrast-enhanced ultrasound; CI, confidence interval; CT, computed tomography; CTP, Child-Turcotte-Pugh; HCC, hepatocellular carcinoma; IQR, interquartile range; LT, liver transplantation; MELD, model for end-stage liver disease; MR, magnetic resonance; NPV, negative predictive value; OR, odds ratio; PPV, positive predictive value; PV, portal vein; PVT, portal venous thrombus; ROC, receiver operating characteristic

Ethics approval and consent to participate

The current study was approved by the the Ethics Committee of Beijing Tsinghua Changgung Hospital Ethics Committee. Because of the retrospective nature of the study and because no intervention was employed, the requirement for informed consent was waived.

Consent for publication

Written informed consent for publication was obtained.

Availability of data and materials

The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the grants from the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences [grant number: 2019-I2M-5-056] and National Natural Science Foundation of China [grant number: 81930119, 82090050, 82090053].

Authors' contributions

TBJ: Conception, analysis, writing - original draft. LSJ: Study design, analysis, interpretation of data, writing - original draft. TX: Investigation, interpretation of data, writing – review & editing. TR: Investigation, interpretation of data, writing – review & editing. AA, Investigation, interpretation of data. XY: Investigation, interpretation of data. ZC: Investigation, interpretation of data. LQ: Investigation, writing – review & editing, supervision. DJH: Investigation, writing – review & editing, supervision. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

Francoz C, Valla D, Durand F. Portal vein thrombosis, cirrhosis, and liver transplantation. J Hepatol. 2012;57(1):203-12.
Hibi T, Nishida S, Levi DM, Selvaggi G, Tekin A, Fan J, et al. When and why portal vein thrombosis matters in liver transplantation: a critical audit of 174 cases. Ann Surg. 2014;259(4):760-6.
Josep M. Llovet JB, Antoni Castells,Ramon Vilana,Maria Del Carmen Ayuso,Margarita Sala, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: Rationale for the design and evaluation of therapeutic trials. Hepatology. 1999;29(1).
Bruix J, Raoul JL, Sherman M, Mazzaferro V, Bolondi L, Craxi A, et al. Efficacy and safety of sorafenib in patients with advanced hepatocellular carcinoma: subanalyses of a phase III trial. J Hepatol. 2012;57(4):821-9.
Reig M, Forner A, Rimola J, Ferrer-Fàbrega J, Burrel M, Garcia-Criado Á, et al. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. J Hepatol. 2022;76(3):681-93.
Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, et al. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2018;67(1):358-80.
Cedrone A, Rapaccini GL, Pompili M, Aliotta A, Trombino C, De Luca F, et al. Portal vein thrombosis complicating hepatocellular carcinoma. Value of ultrasound-guided fine-needle biopsy of the thrombus in the therapeutic management. Liver. 1996;16(2):94-8.
Vilana R, Bru C, Bruix J, Castells A, Sole M, Rodes J. Fine-needle aspiration biopsy of portal vein thrombus: value in detecting malignant thrombosis. AJR Am J Roentgenol. 1993;160(6):1285-7.
Rammohan A, Jeswanth S, Sukumar R, Anand L, Kumar PS, Srinivasan UP, et al. Percutaneous ultrasound-guided fine-needle aspiration of portal vein thrombi as a diagnostic and staging technique for hepatocellular carcinoma. Abdom Imaging. 2013;38(5):1057-60.
Cannella R, Taibbi A, Porrello G, Dioguardi Burgio M, Cabibbo G, Bartolotta TV. Hepatocellular carcinoma with macrovascular invasion: multimodality imaging features for the diagnosis. Diagn Interv Radiol. 2020;26(6):531-40.
Qian LJ, Zhu J, Zhuang ZG, Xia Q, Cheng YF, Li JY, et al. Differentiation of neoplastic from bland macroscopic portal vein thrombi using dual-energy spectral CT imaging: a pilot study. Eur Radiol. 2012;22(10):2178-85.
Tublin ME, Dodd GD, 3rd, Baron RL. Benign and malignant portal vein thrombosis: differentiation by CT characteristics. AJR Am J Roentgenol. 1997;168(3):719-23.
Rossi S, Ghittoni G, Ravetta V, Torello Viera F, Rosa L, Serassi M, et al. Contrast-enhanced ultrasonography and spiral computed tomography in the detection and characterization of portal vein thrombosis complicating hepatocellular carcinoma. Eur Radiol. 2008;18(8):1749-56.
Kim JH, Lee JM, Yoon JH, Lee DH, Lee KB, Han JK, et al. Portal Vein Thrombosis in Patients with Hepatocellular Carcinoma: Diagnostic Accuracy of Gadoxetic Acid-enhanced MR Imaging. Radiology. 2016;279(3):773-83.
Li C, Hu J, Zhou D, Zhao J, Ma K, Yin X, et al. Differentiation of bland from neoplastic thrombus of the portal vein in patients with hepatocellular carcinoma: application of susceptibility-weighted MR imaging. BMC Cancer. 2014;14:590.
Sandrasegaran K, Tahir B, Nutakki K, Akisik FM, Bodanapally U, Tann M, et al. Usefulness of conventional MRI sequences and diffusion-weighted imaging in differentiating malignant from benign portal vein thrombus in cirrhotic patients. AJR Am J Roentgenol. 2013;201(6):1211-9.
Raza SA, Jang HJ, Kim TK. Differentiating malignant from benign thrombosis in hepatocellular carcinoma: contrast-enhanced ultrasound. Abdom Imaging. 2014;39(1):153-61.
Kwon JH, Yoo SH, Nam SW, Lee YJ, Shin YR. Diagnostic Role of Contrast-enhanced Ultrasound in the Discrimination of Malignant Portal Vein Thrombosis in Patients With Hepatocellular Carcinoma. Anticancer Res. 2020;40(8):4351-63.
Piscaglia F, Gianstefani A, Ravaioli M, Golfieri R, Cappelli A, Giampalma E, et al. Criteria for diagnosing benign portal vein thrombosis in the assessment of patients with cirrhosis and hepatocellular carcinoma for liver transplantation. Liver Transpl. 2010;16(5):658-67.
American College of Radiology. CT/MRI Liv- er imaging reporting and data system v2018 Core. https://www.acr.org/Clinical-Resources/ Reporting-and-Data-Systems/LI-RADS/CT-MRI- LI-RADS-v2018. Accessed on October 2019.
Sherman CB, Behr S, Dodge JL, Roberts JP, Yao FY, Mehta N. Distinguishing Tumor From Bland Portal Vein Thrombus in Liver Transplant Candidates With Hepatocellular Carcinoma: the A-VENA Criteria. Liver Transpl. 2019;25(2):207-16.
Tsai HM, Han MZ, Lin YJ, Chang TT, Chen CY, Cheng PN, et al. Real-world outcome of immune checkpoint inhibitors for advanced hepatocellular carcinoma with macrovascular tumor thrombosis. Cancer Immunol Immunother. 2021;70(7):1929-37.
Soin AS, Bhangui P, Kataria T, Baijal SS, Piplani T, Gautam D, et al. Experience With LDLT in Patients With Hepatocellular Carcinoma and Portal Vein Tumor Thrombosis Postdownstaging. Transplantation. 2020;104(11):2334-45.
Assalino M, Terraz S, Grat M, Lai Q, Vachharajani N, Gringeri E, et al. Liver transplantation for hepatocellular carcinoma after successful treatment of macrovascular invasion - a multi-center retrospective cohort study. Transpl Int. 2020;33(5):567-75.
Han DH, Joo DJ, Kim MS, Choi GH, Choi JS, Park YN, et al. Living Donor Liver Transplantation for Advanced Hepatocellular Carcinoma with Portal Vein Tumor Thrombosis after Concurrent Chemoradiation Therapy. Yonsei Med J. 2016;57(5):1276-81.
Jeong Y, Shin MH, Yoon SM, Song GW, Kim KH, Ahn CS, et al. Liver Transplantation After Transarterial Chemoembolization and Radiotherapy for Hepatocellular Carcinoma with Vascular Invasion. J Gastrointest Surg. 2017;21(2):275-83.
Lee KW, Suh SW, Choi Y, Jeong J, Yi NJ, Kim H, et al. Macrovascular invasion is not an absolute contraindication for living donor liver transplantation. Liver Transpl. 2017;23(1):19-27.
Shah ZK, McKernan MG, Hahn PF, Sahani DV. Enhancing and expansile portal vein thrombosis: value in the diagnosis of hepatocellular carcinoma in patients with multiple hepatic lesions. AJR Am J Roentgenol. 2007;188(5):1320-3.
Carneiro C, Brito J, Bilreiro C, Barros M, Bahia C, Santiago I, et al. All about portal vein: a pictorial display to anatomy, variants and physiopathology. Insights Imaging. 2019;10(1):38.
Alzubaidi S, Patel I, Saini A, Knuttinen G, Naidu S, Kriegshuaser S, et al. Current concepts in portal vein thrombosis: etiology, clinical presentation and management. Abdom Radiol (NY). 2019;44(10):3453-62.

TABEL 1. Characteristics of patients with tumor PVT and bland PVT.
Variables	Overall (n=49)	Tumor PVT (n=26)	Bland PVT (n=23)	P value
Age, y, median (IQR)	54.0 (48.5-61.0)	52.0 (45.5-58.5)	57.0 (50.0-63.0)	0.079
Sex, male, n (%)	48 (98.0)	25 (96.2)	23 (100.0)	1.000
Etiology of liver disease, n (%)				0.531
Hepatitis B	42 (85.7)	23 (88.5)	19 (82.6)
Hepatitis C	2 (4.1)	1 (3.8)	1 (4.3)
Alcoholic liver disease	1 (2.0)	1 (3.8)	0 (0)
Others	4 (4.1)	1 (3.8)	3 (13.0)
MELD score, median (IQR)	7.11 (4.20-9.22)	6.94 (4.02-8.33)	7.33 (4.81-9.82)	0.674
CTP score, median (IQR)	7 (6-8)	7 (6-8)	6 (6-9)	0.775
Sum of tumor diameter, cm, median (IQR)	7.00 (3.00-11.70)	9.25 (5.00-12.20)	3.10 (2.10-7.10)	0.001
Diameter of largest tumor, cm, median (IQR)	4.80 (2.50-8.20)	7.70 (4.50-12.00)	2.50 (1.95-4.90)	<0.001
Tumor number, n (%)				0.159
1	27 (55.1)	16 (61.5)	11 (47.8)
2	13 (26.5)	4 (15.4)	9 (39.1)
≥3	9 (18.4)	6 (23.1)	3 (13.0)
AFP, ng/mL, median (IQR)	89.33 (7.76-1978.35)	1270.73 (45.67-2000.00)	14.86 (2.45-89.33)	0.001
PVT location, n (%)				0.150
Main PV	29 (59.2)	13 (50.0)	16 (69.6%)
Right or left PV	13 (26.5)	7 (26.9)	6 (26.1%)
PV segmental branches	7 (14.3)	6 (23.1)	1 (4.3%)
PV diameter, mm, median (IQR)	18.0 (15.0-23.0)	20.0 (17.0-23.5)	17.0 (15.0-21.0)	0.097
A-VENA criteria
AFP >1000 ng/mL, n (%)	20 (40.8)	16 (61.5)	4 (17.4)	0.002
Venous expansion*, n (%)	21 (42.9)	13 (50.0)	8 (34.8)	0.283
Enhancement of PVT, n (%)	27 (55.1)	24 (92.3)	3 (13.0)	<0.001
Neovascularity, n (%)	19 (38.8)	17 (65.4)	2 (8.7)	<0.001
Adjacent to HCC, n (%)	28 (57.1)	23 (88.5)	5 (21.7)	<0.001
A-VENA score, n (%)				<0.001
≥3	26 (53.1)	24 (92.3)	2 (8.7)
<3	23 (46.9)	2 (7.7)	21 (91.3)
PVT, portal venous thrombus; IQR, interquartile range; MELD, model for end-stage liver disease; CTP, Child-Turcotte-Pugh; AFP, alpha fetoprotein; HCC, hepatocellular carcinoma. *Venous expansion was defined as the diameter of main trunk of portal vein ≥20mm.

TABLE 2. Univariable and multivariable analysis for the differentiation of tumor PVT from bland PVT.
	Univariable		Multivariable
Factors	OR (95%CI)	P Value	OR (95%CI)	P Value
Diameter of largest tumor >5cm versus not	6.375 (1.828-22.227)	0.004	0.594 (0.034-10.350)	0.721
AFP >400ng/mL versus not	10.687 (2.736-41.741)	0.001	26.108 (1.114-611.717)	0.043
Venous expansion* versus not	1.875 (0.592-5.934)	0.285
Enhancement of PVT versus not	80.000 (12.146-526.926)	<0.001	98.712 (2.324-4192.171)	0.016
Neovascularity versus not	19.833 (3.770-104.366)	<0.001	0.971 (0.036-26.218)	0.986
Adjacent to HCC versus not	27.600 (5.808-131.159)	<0.001	59.264 (1.950-1801.320)	0.019
PVT, portal venous thrombus; AFP, alpha fetoprotein; HCC, hepatocellular carcinoma; OR, odds ratio; CI, confidence interval. *Venous expansion was defined as the diameter of portal vein ≥20mm.

TABLE 3. Performance of AEA score for the differentiation of tumor PVT from bland PVT.
	N	Tumor PVT (n=26)	Bland PVT (n=23)	P Value	Sensitivity	Specificity	PPV	NPV
AEA ≥1				<0.001	100%	56.5%	72.2%	100%
Yes	36	26 (100%)	10 (43.5%)
No	13	0 (0%)	13 (56.5%)
AEA ≥2				<0.001	100%	95.7%	96.3%	100%
Yes	27	26 (100%)	1 (4.3%)
No	22	0 (0%)	22 (95.7%)
AEA =3				<0.001	50%	95.7%	92.90%	62.9%
Yes	14	13 (50.0%)	1 (0%)
No	35	13 (50.0%)	22 (100%)
PVT, portal venous thrombus; PPV, positive predictive value; NPV, negative predictive value.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Validation of A-VENA Criteria in Distinguishing Tumor from Bland Portal Vein Thrombus in Liver Transplant Candidates with Hepatocellular Carcinoma

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Version 1