Correlation of pretreatment imaging features with lung shunt fraction in patients with large hepatocellular carcinoma (maximum diameter >8 cm) planned for ⁹⁰Y-SIRT

Introduction

Hepatocellular carcinoma (HCC) constitutes approximately 75–85% of primary liver cancers and represents a major global health burden, ranking as the sixth most common malignant tumor and the third leading cause of cancer-related mortality worldwide (1). Each year, HCC accounts for approximately 800,000 deaths (1,2). Due to the frequent absence of clinically apparent symptoms in the early stages of HCC, approximately 70% of patients present with intermediate- or advanced-stage disease at diagnosis (3), precluding curative-intent therapies such as radical resection, ablation, or transplantation as first-line options (4).

Yttrium-90 selective internal radiation therapy (⁹⁰Y-SIRT) provides a safe and effective therapeutic modality for patients with unresectable, intermediate-to-advanced HCC (5). The lung shunt fraction (LSF) is the proportion of radioactivity entering the lungs through the hepatic arteriovenous shunt, which is measured by imaging with technetium-99m macroaggregated albumin (^99mTc-MAA) before ⁹⁰Y-SIRT. An elevated LSF significantly increases the risk of radiation pneumonitis and severe pulmonary toxicity following ⁹⁰Y radioembolization (6). Consequently, the LSF is crucial for treatment decision-making. For resin microspheres, an LSF >10% indicates dose reduction, while an LSF >20% typically contraindicates treatment. Similarly, for glass microsphere treatment, the risk of radiation pneumonitis increases when the pulmonary absorbed dose per treatment exceeds 30 Gy or the cumulative dose surpasses 50 Gy (7-9). Furthermore, the LSF serves as a prognostic indicator for disease progression, metastasis, and the response of liver tumors to ⁹⁰Y-SIRT (10).

The assessment of the LSF relies on angiography combined with ^99mTc-MAA single-photon emission computed tomography/computed tomography (SPECT/CT) imaging. This approach not only evaluates LSF but also predicts the anticipated ⁹⁰Y microsphere biodistribution, providing vital dosimetric information for treatment planning (11). Recently, several studies (12-16) have investigated noninvasive imaging predictors of elevated LSF, aiming to optimize treatment stratification and reduce reliance on invasive procedures. However, conflicting evidence has emerged regarding the specific tumor characteristics associated with a high LSF. For instance, Gaba et al. (12) systematically verified a significant correlation between certain liver tumor characteristics and high LSF (>20%) and found that infiltrative morphology, tumor burden >50%, main portal vein invasion, and arterioportal shunting were independent predictors. Conversely, multivariate analyses by Olorunsola et al. (13) and Choi et al. (15) identified hepatic vein tumor invasion or occlusion, but not portal vein tumor thrombus, as a robust independent risk factor for a high LSF. These discrepancies suggest that the mechanisms underlying an elevated LSF may be influenced by differences in tumor biology and the specific anatomical sites of vascular invasion.

Other studies (12-16) have demonstrated that larger HCC tumors have a higher risk of a high LSF, but some large HCC tumors do not have a high LSF and are still suitable for ⁹⁰Y-SIRT. Currently, the data on the impact of large HCC tumors on LSF remain insufficient. Moreover, patients with HCC are more prone to arteriovenous shunting, which leads to elevated LSF, necessitating a reduction in the treatment dose or even postponing or terminating ⁹⁰Y-SIRT (12). Therefore, this study aimed to investigate the association between preprocedural imaging features and the LSF in patients with HCC scheduled for ⁹⁰Y-SIRT, particularly those with large tumors (maximum diameter >8 cm). We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2206/rc).

Methods

Research participants

A retrospective analysis was conducted on 218 patients with HCC who underwent procedural hepatic arteriography combined with ^99mTc-MAA imaging for LSF assessment prior to planned ⁹⁰Y-SIRT at The First Affiliated Hospital of Jinan University from August 2022 to April 2025. Clinical data of all patients were collected, including baseline data, laboratory tests, and imaging examinations. The inclusion criteria were as follows: (I) age ≥18 years; (II) HCC diagnosis confirmed as per the relevant guidelines (17,18) via histopathology or combination with imaging and serological tests, with at least one lesion being measurable; (III) Child-Pugh class A/B; and (IV) availability of complete preprocedural baseline data. Meanwhile, the exclusion criteria were as follows: (I) severe contrast agent allergy; (II) severe liver and kidney dysfunction; (III) severe coagulation dysfunction; and (IV) Child-Pugh class C. A total of 218 patients (197 males and 21 females) were ultimately included, with an average age of 56.6±11.5 years. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and was approved by the Ethics Committee of The First Affiliated Hospital of Jinan University (approval No. KY-2025-134). Informed consent was waived due to the retrospective nature of the study and the anonymization of data.

Pretreatment imaging examination

Patients underwent multiphasic contrast-enhanced computed tomography (CECT) with a multidetector CT scanner (NeuViz Epoch, Neusoft Medical Systems, Liaoning, China). The CT protocol included noncontrast, hepatic arterial, portal venous, and delayed phases. The imaging parameters were as follows: voltage, 100–120 kVp; tube current, 150–250 mAs; slice thickness, 2.5–3.0 mm; and reconstruction interval, 2.0–3.0 mm. Additionally, magnetic resonance imaging (MRI) was conducted according to standard liver protocols, which often included dynamic contrast-enhanced sequences (such as arterial, portal venous, and delayed phases) and T2-weighted and diffusion-weighted imaging.

Angiography and ^99mTc-MAA injection

All patients underwent diagnostic angiography with a digital subtraction angiography (DSA) system (Artis Zeego Q, Siemens Healthineers, Erlangen, Germany) in the interventional department. Cone beam CT was routinely performed to clarify the hepatic vascular anatomy and evaluate the blood supply of liver tumors, which served as the pretreatment preparation for SIRT. After angiography, 75–222 MBq of ^99mTc-MAA was injected via the artery to simulate the distribution of ⁹⁰Y resin microspheres in the liver, lungs, and potential extrahepatic abdominal organs. Within 1 hour after the injection of ^99mTc-MAA, whole-body planar imaging and SPECT/CT imaging of the chest and abdomen were performed via a SPECT/CT device (Discovery NM/CT640, GE HealthCare, Chicago, IL, USA) (19).

SPECT/CT image acquisition

The LSF was quantified via planar scintigraphy (anterior-posterior projection; matrix: 256×1,024; speed: 20 cm/min) to assess pulmonary complication risk. Subsequently, SPECT/CT acquisition was conducted with a low-dose CT protocol (tube voltage: 120 kV; tube current: 20 mA; slice thickness: 2.5 mm; matrix: 512×512) co-registered with emission data acquired via a low-energy high-resolution collimator on a dual-detector system (energy peak: 140 keV ±10%; matrix: 128×128; 6° per projection; 20–35 seconds per projection). Image reconstruction was performed via the ordered subset expectation maximization algorithm with corrections for attenuation and scatter. The acquired three-dimensional SPECT/CT data were used for quantitative analysis, which included the calculation of the tumor-to-normal ratio and identification of any extrahepatic shunting. These parameters were applied to the ⁹⁰Y-resin microsphere partition model for personalized dosimetry planning.

Image processing and data analysis

All images were independently reviewed by two abdominal radiologists with 5 and 8 years of experience, respectively, using a dedicated picture archiving and communication system workstation, with multiplanar reconstruction images employed as needed. Both reviewers, blinded to all clinical outcomes, assessed the imaging key features using the intraclass correlation coefficient (ICC) for continuous variables and the Cohen kappa coefficient for categorical variables. Any discrepancies in image interpretation were resolved through consensus review, with unresolved cases adjudicated by a senior abdominal radiologist.

Tumor burden was quantified via software-based volumetry with Affinity 5.0 on the Hermia Workstation v. 2.17 (Hermes Medical Solutions, Stockholm, Sweden), and the cumulative volume of all detectable hepatic tumors was manually segmented. This total tumor volume was expressed as a percentage of the automatically derived whole-liver volume, and tumor burden was categorized as <25%, 25–50%, or >50%. Additional tumor characteristics were systematically evaluated according to predefined criteria: maximum tumor diameter (<3, 3–8, or >8 cm) (20), morphology (circumscribed vs. infiltrative), distribution (unilobar vs. bilobar), and tumor number (solitary, oligonodular, or multinodular). The presence or absence of the following imaging features was documented: tumor capsule, hepatic vein invasion or occlusion, portal vein tumor thrombus, and early hepatic vein enhancement (the premature opacification of tumor-draining hepatic veins during the arterial phase relative to uninvolved veins). Abnormal intratumoral vessels were identified as enhancing linear structures within the tumor on arterial phase imaging that did not correspond to a hepatic artery branch and that measured >3 mm in diameter (15). Standard Liver Imaging Reporting and Data System (LI-RADS) imaging features, including arterial phase hyperenhancement and nonperipheral washout, were assessed (21). Liver cirrhosis was determined based on established imaging signs, such as nodular liver contour, lobar disproportion, and features of portal hypertension. Lipiodol deposition from previous therapies was recorded based on its presence on noncontrast scans.

According to the relevant consensus (22), the appropriate color scale on the Xeleris 3.1 SPECT/CT workstation (GE HealthCare) was used to display the contours of the corresponding left and right lungs and the liver. Six regions of interest were drawn for the anterior and posterior views of the liver and both lungs, respectively, and the corresponding radioactive counts were recorded. The LSF was calculated by dividing the geometric mean of the net counts from the lungs by the sum of the geometric mean of the net counts from the lungs and the liver:

LSF=anteriorlungcount×posteriorlungcountanteriorlivercount×posteriorlivercount+anteriorlungcount×posteriorlungcount[1]

Statistical analysis

Statistical analysis was performed with IBM SPSS software version 27.0 (IBM Corp., Armonk, NY, USA). Continuous variables conforming to a normal distribution are presented as the mean ± standard deviation (SD). Categorical variables are summarized as frequencies and percentages. Univariate analysis of categorical variables was performed via the Chi-squared test. To avoid overlooking potential imaging-related factors, independent variables with P<0.15 in the univariate analysis were included in the ordered logistic regression for the overall multivariate analysis. Multivariate analysis of the HCC subgroup with a maximum tumor diameter greater than 8 cm was conducted with binary logistic regression to identify factors that differentiate LSF risk within this specific high-risk population, with variance inflation factors (VIFs) being computed for all included variables to assess multicollinearity. To evaluate the predictive performance of the independent risk factors identified by the multivariate analysis, receiver operating characteristic (ROC) curves were generated with the area under the curve (AUC) computed for each factor. A P value <0.05 was considered statistically significant.

Results

Baseline characteristics of patients

A total of 218 patients with HCC, comprising 197 men and 21 women, were included in this study, with a median age of 53 years (range, 27–87 years). Among the 218 patients, the median LSF was 10.1% (range, 1.5–64.8%). There were 108 patients (49.5%) with a low LSF (LSF <10%), 57 patients (26.1%) with a moderate LSF (10%≤ LSF ≤20%), and 53 patients (24.3%) with a high LSF (LSF >20%). The disease-specific nonimaging features included Barcelona Clinic Liver Cancer stage, Chinese Liver Cancer stage, Child-Pugh grade, Eastern Cooperative Oncology Group performance status, and underlying diseases (hepatitis B/C virus), along with the levels of albumin, alpha-fetoprotein, and total bilirubin (Table 1).

Interobserver agreement

The ICC for maximum tumor diameter was 0.78 [95% confidence interval (CI): 0.72–0.83]. The kappa values for hepatic vein invasion or occlusion, portal vein tumor thrombus, and abnormal intratumoral vessels were 0.89 (95% CI: 0.85–0.92), 0.85 (95% CI: 0.81–0.89), and 0.76 (95% CI: 0.70–0.81), respectively, which suggested good interobserver reproducibility and consistency for the key imaging features.

Univariate analysis

LSF levels were correlated with the maximum tumor diameter (χ²=22.176; P<0.001), tumor burden (χ²=12.416; P=0.015), tumor morphology (χ²=3.985; P=0.136), tumor distribution (χ²=4.387; P=0.112), portal vein tumor thrombus (χ²=6.501; P=0.039), hepatic vein invasion or occlusion (χ²=42.740; P<0.001), and abnormal intratumoral vessels (χ²=10.546; P=0.005). Meanwhile, LSF levels were not correlated with tumor number (χ²=3.376; P=0.497), early hepatic vein enhancement (χ²=2.620; P=0.243), tumor capsule (χ²=0.353; P=0.838), arterial hyperenhancement (χ²=0.016; P=0.992), nonperipheral washout (χ²=0.118; P=0.943), liver cirrhosis (χ²=0.936; P=0.626), or lipiodol deposition (χ²=2.618; P=0.270) (Table 2).

Overall multivariate analysis

Before ordered logistic regression was conducted, a parallel lines test (χ²=5.973; P=0.543) was conducted to verify the proportional odds assumption. The test results indicated that there was no coefficient heterogeneity among the various levels of LSF. The ordered logistic regression analysis, which was statistically significant (likelihood ratio test χ²=54.263; P<0.001), identified the independent imaging factors associated with elevated LSF to be hepatic vein invasion or occlusion [odds ratio (OR) =5.654; 95% CI: 2.859–11.182; P<0.001] and maximum tumor diameter (OR =2.340; 95% CI: 1.240–4.416; P=0.009), with an overall prediction accuracy rate of 57.34%. Tumor morphology, distribution, burden, portal vein tumor thrombus, and abnormal intratumoral vessels did not exhibit independent statistical significance (all P values >0.05). The VIFs for all variables were below 5, indicating no substantial multicollinearity. The goodness-of-fit statistic for the multivariate analysis yielded a McFadden R² of 0.119, indicating that the LSF is likely influenced by other unmeasured factors (Table 3 and Figure 1).

Figure 1 Overall multivariate regression forest plot of risk factors for LSF level. LSF, lung shunt fraction; OR, odds ratio.

ROC curves were plotted to assess the performance of two independent factors identified by the ordinal logistic regression, namely maximum tumor diameter (AUC =0.653; 95% CI: 0.575–0.732) and hepatic vein invasion or occlusion (AUC =0.701; 95% CI: 0.620–0.782), in predicting LSF >20%. The combination of these two factors yielded an AUC of 0.736 (95% CI: 0.663–0.808) in predicting LSF >20%. At the optimal threshold determined by the Youden index, the sensitivity and specificity were 71.7% and 68.5%, respectively. These results indicate that the combined use of these two factors provides improved discriminatory ability compared to either factor alone, supporting its potential utility in preoperative risk stratification (Figure 2).

Figure 2 ROC curves of maximum tumor diameter and hepatic vein invasion or occlusion in predicting an overall LSF >20%. AUC, area under the curve; CI, confidence interval; LSF, lung shunt fraction; ROC, receiver operating characteristic curve.

Subgroup multivariate analysis

The binary logistic regression analysis, which produced statistically significant results (likelihood ratio test χ²=17.342; P<0.001), identified the independent imaging factors associated with elevated LSF to be hepatic vein invasion or occlusion (OR =6.452; 95% CI: 2.413–17.253; P<0.001), with a prediction accuracy rate of 74.24%. Tumor morphology, distribution, burden, portal vein tumor thrombus, and abnormal intratumoral vessels did not demonstrate independent statistical significance (all P values >0.05). The VIFs for all variables were below 5, indicating no substantial multicollinearity. The goodness-of-fit statistic for the multivariate analysis yielded a McFadden R² of 0.103, indicating that the LSF is likely influenced by other unmeasured factors (Table 4). The overall treatment decisions for the 218 cases of HCC were made after multidisciplinary team consultation and were based on the patient’s LSF level, liver function status, and preferences, among other factors (Table 5).

Discussion

The safety profile of ⁹⁰Y-SIRT for intermediate-to-advanced HCC is dependent on an accurate preprocedural assessment of the LSF. This retrospective study investigated the association between a comprehensive set of pretreatment imaging features and LSF. Our analysis identified hepatic vein invasion or occlusion and a larger maximum tumor diameter as independent imaging risk factors correlated with elevated LSF levels (Figure 3). Among cases of HCC with a largest tumor diameter greater than 8 cm (n=132), even in cases with liver vein invasion or occlusion (n=78), a large proportion of patients (n=44) still had an LSF ≤20%, and they were still suitable for ⁹⁰Y-SIRT. These findings establish a noninvasive, quantifiable framework for clinicians to anticipate abnormal shunting, thereby optimizing the pretreatment risk assessment workflow for ⁹⁰Y-SIRT. Although the prediction accuracy rates for the overall group and subgroup (>8 cm) in the multivariate analysis were 57.34% and 74.24% respectively, their explanatory power was relatively low, which indicates that other unmeasured patient-specific or tumor biological factors also influence LSF. However, this study focused mainly on conventional and easily assessable imaging features in clinical practice.

Figure 3 A 47-year-old male with a large (maximum diameter 18.1 cm) HCC. ^99mTc-MAA planar imaging showed a high LSF of 38.1% (A). Contrast-enhanced CT in the portal phase showed that the middle hepatic vein (yellow arrows) was invaded by the tumor (red arrows) and appeared blurred. (B1-B3: coronal views; C1-C3: axial views). ^99mTc-MAA, technetium-99m macroaggregated albumin; CT, computed tomography; HCC, hepatocellular carcinoma; LSF, lung shunt fraction.

In our cohort, maximum tumor diameter was independent risk factor for an elevated LSF, a finding consistent with previous reports linking larger tumors to increased shunting (13-16). Notably, although tumor burden demonstrated significance in the univariate analysis, it was not independently predictive in the multivariate analysis. This suggests that maximum tumor diameter may be a more robust indicator of vascular distortion than volumetric ratios in this Asian cohort, potentially reflecting its closer relationship with vascular architectural disruption. Hepatic vein invasion or occlusion was found to be an independent risk factor for an elevated LSF both in the overall HCC group and large-HCC subgroup, which is consistent with previous studies (13,15). Another recent study (20), despite its limited sample size, also reported that large HCC tumors (>8 cm) without evidence of vascular invasion had comparable LSF characteristics to other, smaller HCC tumors (3–8 cm). The results from these studies and our own suggest that the historical association of a high LSF with large HCC is more related to the increased risk of vascular invasion rather than tumor size alone. The strength of this association underscores the need for heightened vigilance in LSF estimation and dosimetry planning in these patients.

Furthermore, neither portal vein tumor thrombus nor abnormal intratumoral vessels were found to be independent predictors of LSF despite initial univariate associations. Since multicollinearity was negligible (VIF <5), this suggests that hepatic vein invasion is a more critical determinant of shunting than is portal vein involvement or specific intratumoral vessel morphology in this patient population. This finding, coupled with the nonsignificance of abnormal vessels—which contrasts with reports in other cohorts (15)—suggests that the pre-existing vascular alterations in our patients may fundamentally modulate shunting mechanisms, making generalized predictors less applicable.

Our study has several advantages, including a relatively large cohort of patients (n=53) with a high LSF and a subgroup cohort (n=132) with a maximum tumor diameter greater than 8 cm. This enhances the universality of our finding that a large tumor does not necessarily suggest a high LSF. In clinical practice, individualized assessment should be conducted for patients with large tumors. To our knowledge, this is also the first study to systematically evaluate the association of lipiodol deposition and hepatitis B virus-related cirrhosis with LSF, and no significant associations were found. These findings contribute to a more nuanced understanding of the shunt mechanisms in Asian HCC populations and suggest that the pathological basis of cirrhosis in this context may involve vascular regulatory disturbances rather than anatomical shunts (23).

Certain limitations to this study should be acknowledged. First, we employed a single-center, retrospective design, and internal and multicenter validation or calibration was lacking. Future studies should verify these associations in a larger independent cohort. Second, due to the limited study scale, the multivariate analysis may underestimate the influence of other factors on the LSF. Third, the planar LSF used in the analysis yielded inaccurate values and was higher compared to the LSF based on SPECT/CT. Therefore, some planar LSF values were likely significantly higher than those that would have been computed with SPECT/CT.

Conclusions

Hepatic vein invasion or occlusion and maximum tumor diameter was found to be independent risk factors for elevated LSF in patients with HCC scheduled to undergo ⁹⁰Y-SIRT. These imaging features can aid in the preoperative risk stratification but cannot replace ^99mTc-MAA-based shunt assessment. Notably, even among patients with large tumors (>8 cm) and hepatic vein invasion or occlusion, over half (56.4%) remained eligible for treatment, emphasizing the importance of individualized evaluation. Further research is needed to better identify which high-risk patients are suitable candidates for treatment.

Acknowledgments

We would like to thank the staff members of the Department of Nuclear Medicine and PET/CT-MRI Centre, The First Affiliated Hospital of Jinan University, for their excellent technical support.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2206/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2206/dss

Funding: This study was supported by Yttrium Little Red Flower Health Fund Project of Sunshine Medical and Health Development Foundation in Henan Province, China (No. HKP2024002), and the Medical Scientific Research Foundation of Guangdong Province (No. A2024035).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2206/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and was approved by the Ethics Committee of The First Affiliated Hospital of Jinan University (approval No. KY-2025-134). Informed consent was waived due to the retrospective nature of the study and the anonymized data collection.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Rumgay H, Ferlay J, de Martel C, Georges D, Ibrahim AS, Zheng R, Wei W, Lemmens VEPP, Soerjomataram I. Global, regional and national burden of primary liver cancer by subtype. Eur J Cancer 2022;161:108-18. [Crossref] [PubMed]
3. Zhou J, Sun H, Wang Z, Cong W, Wang J, Zeng M, et al. Guidelines for the Diagnosis and Treatment of Hepatocellular Carcinoma (2019 Edition). Liver Cancer 2020;9:682-720. [Crossref] [PubMed]
4. Bruix J, Sherman MPractice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. Hepatology 2005;42:1208-36. [Crossref] [PubMed]
5. Kishore SA, Bajwa R, Madoff DC. Embolotherapeutic Strategies for Hepatocellular Carcinoma: 2020 Update. Cancers (Basel) 2020;12:791. [Crossref] [PubMed]
6. Salem R, Parikh P, Atassi B, Lewandowski RJ, Ryu RK, Sato KT, Gates VL, Ibrahim S, Mulcahy MF, Kulik L, Liu DM, Riaz A, Omary RA, Kennedy AS. Incidence of radiation pneumonitis after hepatic intra-arterial radiotherapy with yttrium-90 microspheres assuming uniform lung distribution. Am J Clin Oncol 2008;31:431-8. [Crossref] [PubMed]
7. Murthy R, Nunez R, Szklaruk J, Erwin W, Madoff DC, Gupta S, Ahrar K, Wallace MJ, Cohen A, Coldwell DM, Kennedy AS, Hicks ME. Yttrium-90 microsphere therapy for hepatic malignancy: devices, indications, technical considerations, and potential complications. Radiographics 2005;25:S41-55. [Crossref] [PubMed]
8. Loke KS, Padhy AK, Ng DC, Goh AS, Divgi C. Dosimetric considerations in radioimmunotherapy and systemic radionuclide therapies: a review. World J Nucl Med 2011;10:122-38. [Crossref] [PubMed]
9. Lewandowski RJ, Salem R. Yttrium-90 radioembolization of hepatocellular carcinoma and metastatic disease to the liver. Semin Intervent Radiol 2006;23:64-72. [Crossref] [PubMed]
10. Xing M, Lahti S, Kokabi N, Schuster DM, Camacho JC, Kim HS. 90Y Radioembolization Lung Shunt Fraction in Primary and Metastatic Liver Cancer as a Biomarker for Survival. Clin Nucl Med 2016;41:21-7. [Crossref] [PubMed]
11. Garin E, Lenoir L, Rolland Y, Edeline J, Mesbah H, Laffont S, Porée P, Clément B, Raoul JL, Boucher E. Dosimetry based on 99mTc-macroaggregated albumin SPECT/CT accurately predicts tumor response and survival in hepatocellular carcinoma patients treated with 90Y-loaded glass microspheres: preliminary results. J Nucl Med 2012;53:255-63. [Crossref] [PubMed]
12. Gaba RC, Zivin SP, Dikopf MS, Parvinian A, Casadaban LC, Lu Y, Bui JT. Characteristics of primary and secondary hepatic malignancies associated with hepatopulmonary shunting. Radiology 2014;271:602-12. [Crossref] [PubMed]
13. Olorunsola OG, Kohi MP, Behr SC, Kolli PK, Taylor AG, Tong RT, LaBerge JM, Kerlan RK, Fidelman N. Imaging Predictors of Elevated Lung Shunt Fraction in Patients Being Considered for Yttrium-90 Radioembolization. J Vasc Interv Radiol 2015;26:1472-8. [Crossref] [PubMed]
14. Kallini JR, Gabr A, Hickey R, Kulik L, Desai K, Yang Y, Gates VL, Riaz A, Salem R, Lewandowski RJ. Indicators of Lung Shunt Fraction Determined by Technetium-99 m Macroaggregated Albumin in Patients with Hepatocellular Carcinoma. Cardiovasc Intervent Radiol 2017;40:1213-22. [Crossref] [PubMed]
15. Choi TW, Joo I, Kim HC. Association of dysmorphic intratumoral vessel with high lung shunt fraction in patients with hepatocellular carcinoma. Sci Rep 2022;12:14248. [Crossref] [PubMed]
16. Hamm CA, Busch F, Pöhlmann A, Shewarega A, He Y, Schmidt R, Xu H, Wieners G, Gebauer B, Savic LJ. Non-Invasive Imaging Biomarkers to Predict the Hepatopulmonary Shunt Fraction Before Transarterial Radioembolization in Patients with Hepatocellular Carcinoma. J Hepatocell Carcinoma 2023;10:27-42. [Crossref] [PubMed]
17. Omata M, Cheng AL, Kokudo N, Kudo M, Lee JM, Jia J, et al. Asia-Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int 2017;11:317-70. [Crossref] [PubMed]
18. Li XH, Wei L. The comparison among the guidelines for the diagnosis and treatment of hepatocellular carcinoma in China, AASLD and EASL. Zhonghua Gan Zang Bing Za Zhi 2019;27:236-40. [PubMed]
19. Young S, Flanagan S, D’Souza D, Todatry S, Ragulojan R, Sanghvi T, Golzarian J. Lung shunt fraction calculations before Y-90 transarterial radioembolization: Comparison of accuracy and clinical significance of planar scintigraphy and SPECT/CT. Diagn Interv Imaging 2023;104:185-91. [Crossref] [PubMed]
20. Thomas MA, Alamuri T, Lee RC, Giardina D, Karageorgiou J, Mani N, Braga DA, Malone CD. Modeling Clinical Relevance and Risk in Yttrium-90 Radioembolization of the Liver: Lung Shunt Fraction Variability According to Imaging Modality, Cancer Type, and Tumor Size. J Vasc Interv Radiol 2026;37:107834. [Crossref] [PubMed]
21. Chernyak V, Fowler KJ, Kamaya A, Kielar AZ, Elsayes KM, Bashir MR, Kono Y, Do RK, Mitchell DG, Singal AG, Tang A, Sirlin CB. Liver Imaging Reporting and Data System (LI-RADS) Version 2018: Imaging of Hepatocellular Carcinoma in At-Risk Patients. Radiology 2018;289:816-30. [Crossref] [PubMed]
22. Levillain H, Bagni O, Deroose CM, Dieudonné A, Gnesin S, Grosser OS, et al. International recommendations for personalised selective internal radiation therapy of primary and metastatic liver diseases with yttrium-90 resin microspheres. Eur J Nucl Med Mol Imaging 2021;48:1570-84. [Crossref] [PubMed]
23. Kisseleva T, Brenner D. Molecular and cellular mechanisms of liver fibrosis and its regression. Nat Rev Gastroenterol Hepatol 2021;18:151-66. [Crossref] [PubMed]