Predictive factors of severe abdominal pain during and after transarterial chemoembolization for hepatocellular carcinoma.

OBJECTIVES: To prospectively assess the frequency of severe abdominal pain during and after transarterial chemoembolization (TACE) for hepatocellular carcinoma (HCC) using the visual analog scale (VAS), and to identify predictive factors. METHODS: Ninety-eight TACE performed in 80 patients (mean 65 ± 12 years old, 60 men) were consecutively and prospectively included. Abdominal pain was considered severe if the VAS ≥ 30/100 after treatment administration, or if opioid analgesic (grades 2-3) intake was required during hospitalization. Patient and tumor characteristics as well as technical factors associated with severe pain were identified by binary logistic regression. RESULTS: The criterion for severe pain was met in 41/98 (42%) of procedures (peri-procedural pain 30/98 [31%] and opioid consumption during hospitalization 24/98 [25%]). Multivariate analysis identified age (odds ratio [OR] = 0.943 (95% confidence interval 0.895-0.994), p = 0.029), cirrhosis (OR = 0.284 (0.083-0.971), p = 0.045), and alcoholic liver disease (OR = 0.081 (0.010-0.659), p = 0.019) as negative predictive factors of severe abdominal pain. Severe abdominal pain occurred in or after 1/13 (8%), 8/34 (24%), 22/41 (54%), and 10/10 (100%) TACE sessions when none, one, two, and three of the protective factors were absent, respectively (p < 0.001). The area under the ROC curve of the combination of factors for the prediction of severe abdominal pain was 0.779 (CI 0.687-0.871). CONCLUSION: Severe abdominal pain was frequent during and after TACE revealing a clinically relevant and underestimated problem. A predictive model based on three readily available clinical variables suggests that young patients without alcoholic liver disease or cirrhosis could benefit from reinforced analgesia. KEY POINTS: • Severe abdominal pain occurs in 43% of TACE for HCC. • Younger age, absence of cirrhosis, and absence of alcoholic liver disease were identified as independent predictive factors of severe abdominal pain. • A simple combination of the three abovementioned features helped predict the occurrence of severe abdominal pain.

Factors Associated with Tumor Progression After Percutaneous Ablation of Hepatocellular Carcinoma: Comparison Between Monopolar Radiofrequency and Microwaves. Results of a Propensity Score Matching Analysis.

PURPOSE: To identify risk factors for local and distant intrahepatic tumor progression after percutaneous ablation of HCC and to compare MWA with monopolar RFA. MATERIALS AND METHODS: Consecutive patients with early or very early HCC who underwent percutaneous monopolar RFA or MWA were included. Factors associated with local and distant tumor progression were identified. Propensity score matching (PSM) was used to limit bias. Statistical analyses were performed with the Kaplan-Meier method using the log-rank test and Cox regression models. RESULTS: One hundred ninety HCC (mean diameter 23 ± 8.6 mm) were treated by RFA (n = 90, 47%) or MWA (n = 100, 53%) in 152 patients (mean age 63 ± 11, 79% men) between 2009 and 2016. The technical success rate was 97.4% (n = 185 HCC). After a median follow-up of 24.6 months (IQR: 9.7-37.2), 43 (23%), HCC showed local tumor progression [after a median of 13.4 months (IQR: 5.8-24.3)] and 91 (63%) patients had distant intrahepatic tumor progression (after a median of 10.4 months (IQR: 5.7-22). The cox model after PSM identified treatment by RFA (HR, 2.89; P = 0.005), HCC size ≥ 30 mm (HR, 3.12; P = 0.007) and vascular contact (HR, 3.43; P = 0.005) as risk factors for local progression. Factors associated with distant intrahepatic progression were HCC ≥ 30 mm (HR, 1.94; P = 0.013), serum AFP > 100 ng/mL (HR, 2.56; p = 0.002), and hepatitis B carrier (HR, 0.51; p = 0.047). CONCLUSION: The rate of local HCC progression was lower after MWA than monopolar RFA, regardless of tumor size and vascular contact. The ablation technique did not influence the risk of distant intrahepatic tumor progression.

Clinical impact of a new cone beam CT angiography respiratory motion artifact reduction algorithm during hepatic intra-arterial interventions.

OBJECTIVES: To assess the impact of recently developed respiratory motion correction software on contrast-enhanced cone beam CT angiography (CBCT-a) for intraprocedural image guidance during intra-arterial liver-directed therapy. METHODS: From 2015 to 2017, two groups of patients who underwent intra-arterial liver-directed therapy with (breathing, n = 30) or without (still, n = 30) significant respiratory motion artifacts were retrospectively included. All CBCT-a were processed with and without dedicated respiratory motion correction software. Four readers independently assessed the following in both reconstructions (motion correction ON and OFF): (1) overall image quality on a 0-to-5 point scale, and (2) presence of relevant peri-procedural information on tumor and vasculature (overall vessel geometry, visibility of extrahepatic vessels, target tumor conspicuity, visibility of tumor feeders). RESULTS: Motion correction increased the average image quality in the breathing group from 2.0 ± 0.9 to 2.9 ± 1.0 (p < 0.01). The visibility of vessel geometry, extrahepatic vessels, and tumor feeders was significantly improved for all readers, and tumor conspicuity was improved for three readers. The average image quality was not significantly different between reconstructions in the still group (motion correction ON and OFF), for any of the readers (4.0 ± 0.6 vs 4.2 ± 0.6; p = 0.12). There was no change in the visibility of vessel geometry, extrahepatic vessels, tumor feeders, or tumor conspicuity for the four readers using the respiratory motion correction software in this group. CONCLUSIONS: Using the dedicated respiratory motion correction software during intra-arterial liver-directed procedures increases the visualization of relevant peri-procedural information and image quality in CBCT-a corrupted by respiratory motion artifacts without affecting these elements in still CBCT-a. KEY POINTS: • The use of respiratory motion correction software could reduce the need for cone beam CT angiography acquisition retake. • Motion correction software significantly increases the visibility of vessel geometry, extrahepatic vessels, and tumor feeders, as well as tumor conspicuity in cone beam CT angiography corrupted by respiratory motion artifacts. • The use of respiratory motion correction software on cone beam CT angiography uncorrupted by respiratory motion artifact does not result in decreased image quality.

Quantitative correlation between uptake of Gd-BOPTA on hepatobiliary phase and tumor molecular features in patients with benign hepatocellular lesions.

PURPOSE: The purpose of our study was to correlate the quantitative analysis of benign hepatocellular tumor uptake on delayed hepatobiliary phase (HBP) imaging with the quantitative level of OATP expression. METHODS: This single-center retrospective study, which took place between September 2009 and March 2015, included 20 consecutive patients with a proven pathologic and immunohistochemical (IHC) diagnosis of FNH or HCA, including quantification of the OATP expression. The patients underwent Gd-BOPTA-enhancement MRI, including an HBP. The analysis of HBP uptake was performed using the liver-to-lesion contrast enhancement ratio (LLCER). Mean LLCER and OATP expressions were compared between FNH and HCA, and the expression of OATP was correlated with the LLCER value. RESULTS: Of the 23 benign hepatocellular tumors, 9 (39%) were FNH and 14 (61%) were HCA, including 6 inflammatory, 2 HNF1a inactivated, 3 ß-catenin-mutated and 3 unclassified HCAs. On HBP, 100% of the FNH appeared hyper- or isointense, and 79% of the adenomas appeared hypointense. The mean OATP expression of FNH (46.67 ± 26.58%) was significantly higher than that of HCA (22.14 ± 30.74%) (p = 0.0273), and the mean LLCER of FNH (10.66 ± 7.403%) was significantly higher than that of HCA (-13.5 ± 12.25%) (p < 0.0001). The mean LLCER of ß-catenin-mutated HCA was significantly higher than that of other HCAs (p = 0.011). Significant correlation was found between the OATP expression and LLCER values (r = 0.661; p = 0.001). CONCLUSION: In benign hepatocellular tumors, the quantitative analysis of hepatobiliary contrast agent uptake on HBP is correlated with the level of OATP expression and could be used as an imaging biomarker of the molecular background of HCA and FNH. KEY POINTS: • Gd-BOPTA uptake on HBP correlates with the OATP level in benign hepatocellular tumors • FNH and ß-catenin-mutated HCA showed an increased lesion-to-liver contrast enhancement ratio (LLCER) • Increased LLCER may be explained by activation of the Wnt ß-catenin pathway.

Impact of hepatobiliary phase liver MRI versus Contrast-Enhanced Ultrasound after an inconclusive extracellular gadolinium-based contrast-enhanced MRI for the diagnosis of benign hepatocellular tumors.

PURPOSE: To compare the added values of hepatobiliary phase (HBP) MRI and contrast-enhanced ultrasound (CEUS) in addition to inconclusive extracellular gadolinium-based contrast-enhanced MRI (CE-MRI) to characterize benign hepatocellular tumors (BHT). METHODS: Eighty-three BHT-46 focal nodular hyperplasia (FNH) and 37 hepatocellular adenomas (HCA)-with inconclusive CE-MRI in 54 patients (43 women and 11 men, mean age 42 years old ± 14.8) were retrospectively analyzed. All patients underwent both HBP-MRI and CEUS. Two radiologists independently reviewed 2 sets of images, SET-1: CE-MRI and HBP-MRI; SET-2: CE-MRI and CEUS, and classified lesions as "definite FNH," "possible FNH," or "definitely not FNH." Sensitivity (Se) and specificity (Spe) were compared between the two sets; subgroup analyses according to the lesion's size were performed. RESULTS: Regardless of lesion size, the respective Se and Spe of both datasets were not statistically different (95.7 and 100% vs. 76.1 and 94.6% for set-1 and -2 respectively; p = 0.18). For lesions larger than 35 mm, although both sets had similar specificity (100%), sensitivity was higher for SET-1 (100% vs. 40%); p = 0.04. Tumor classifications using SET-1 and SET-2 could have changed patient management in 35/54 (64.8%) and 33/54 (61.1%) of all patients, respectively. CONCLUSIONS: HBP-MRI or CEUS should be performed after an inconclusive CE-MRI. Both can change patient management by avoiding unnecessary biopsy or surveillance. The use of HBP-MRI should be advocated over CEUS in larger (>35 mm) lesions.

Differentiation of focal nodular hyperplasia from hepatocellular adenomas with low-mechanical-index contrast-enhanced sonography (CEUS): effect of size on diagnostic confidence.

PURPOSE: The purpose of this study was to assess the diagnostic performance of contrast-enhanced sonography (CEUS) for the differentiation of focal nodular hyperplasia (FNH) from hepatocellular adenoma (HCA) according to lesion size. MATERIALS AND METHODS: Forty patients with a definite diagnosis of FNH or HCA who underwent CEUS were included in this institutional review board (IRB)-approved study. A total of 43 FNHs and 20 HCAs, including 15 inflammatory HCAs and five unclassified HCAs, were analysed. Two radiologists reviewed the diagnostic CEUS parameters separately and in consensus, including the presence or absence of centrifugal filling and central vessels. The sensitivity (Se), specificity (Sp), and inter-observer confidence (Kappa) of CEUS diagnostic parameters were assessed. RESULTS: Inter-observer agreement of CEUS for FNH diagnosis was high (kappa = 0.81) with an overall Se of 67.4% [29/43 (CI 95%: 51.4-80.1 %)] and an Sp of 100% [20/20 (CI 95%: 81-100%)]. Significantly higher Se figures were found for lesions ≤ 35 mm than for lesions > 35 mm [respectively, 93 % (28/30) (CI 95%: 77.6-99.2) vs. 7.7% (1/13) (CI 95%: 0.2-36%), p = 0.002] with unchanged specificity. CONCLUSION: CEUS is highly specific for the diagnosis of FNH, with very good inter-observer agreement, whatever the size, but its sensitivity is significantly reduced in diagnosing lesions larger than 35 mm. KEY POINTS: • CEUS is highly specific for the diagnosis of FNH, regardless of lesion size • CEUS shows reduced sensitivity in diagnosing FNH lesions larger than 35 mm • The filling patterns of hepatocellular adenomas are not affected by lesion size.