Abstract
PURPOSE:
To determine the influence of pre-interventionally existing portal vein thrombosis on the ablation success of percutaneous tumor ablation of HCC.
MATERIALS AND METHODS:
15 patients with HCC and pre-existing portal vein thrombosis underwent thermal tumor ablation. We retrospectively analyzed the pre- and post-interventionally performed CT and MRI scans in terms of technical success as well as the complication rate. The portal vein thrombosis was classified into segmental, lobar and central thrombus.
RESULTS:
In 13/15 cases (87%) complete ablation with no evidence of residual tumor tissue was seen 6 weeks after the procedure in contrast-enhanced MRI scans and contrast-enhanced ultrasound (CEUS). No major and 2 minor complications were observed after the ablation procedure.
CONCLUSION:
Reduced perfusion due to pre-interventionally existing portal vein thrombosis has no significant impact on the ablation success or the complication rate.
Introduction
Hepatocellular carcinoma (HCC) is one of the most common malignant diseases worldwide. Because of its poor prognosis, it leads to estimated 746 000 cases of death each year [1]. In over 90% hepatocellular carcinomas, except the fibrolamellar carcinoma, emerge in a cirrhotic liver [2].
Bland portal vein thrombosis occurs in patients with and without malignant disease. Bland thrombus occurs in 4.5% –26% of patients with chronic liver disease [3] and in 42% of patients with HCC [4, 5]. In addition, bland and tumor thrombi can be coexistent. Diagnostic imaging can be a valuable instrument to differentiate between tumor thrombus and bland portal vein thrombosis. Considering that especially the arterial hypervascularization in contrast enhanced ultrasound (CEUS) [6] and the diffusion weighted imaging in MRI [7] are well established.
The management of HCC with portal vein tumor thrombus is complicated and controversial [8]. There are a few small sample size studies about the management of patients with HCC and portal vein tumor thrombus. On the scale of things tumor infiltration of the portal vein leads to a significantly worse patient outcome, especially in patients with central portal vein tumor thrombus [9]. In patients with segmental or lobar tumor thrombi, some authors recommend the thermal ablation both of the HCC lesion as well as the concomitant tumor thrombus although this technique is still controversially discussed.
In current literature, there is no published data concerning the impact of pre-interventionally existing bland portal vein thrombosis (PVT) on the technical success and complication rate after percutaneous tumor ablation as well as the potential side effects, such as varied microcirculation due to reduced heat-sink-effect. In this retrospective study, we analyzed the technical efficacy of thermal ablation in patients with HCC and preexisting bland PVT and evaluated the effects of the location and extent of the thrombus on the complication rate.
Materials and methods
Participant selection and study design
All cases of HCC tumor ablation between 08/2011 and 01/2017 were retrospectively reviewed by two readers (L.P.B. and B.P.). Pre-interventional (day 1–3 before ablation) CT and MRI scans for bland PVT were recorded. 15 patients with PVT were identified and included into this retrospective analysis. The location of PVT was divided into type I (segment-level), type II (lobe-level), type III (central) based on a classification proposed by Cheng et al. [10] (Fig. 1).
Classification of the portal vein thrombosis: type I (segment-level, right-handed image), type II (lobe-level, middle image) and type III (central, left-handed image).
This study conforms to the ethical guidelines of this journal [11].
All ablation procedures were performed under general anesthesia by an experienced interventionalist. In 12 out of 15 patients microwave ablation was conducted percutaneously under CT guidance using the Acculis microwave ablation system (AngioDynamics, Latham, NY, USA), which uses electromagnetic waves to induce tissue-heating effects causing tumor necrosis [12]. The other 3 patients underwent radiofrequency ablation using the StarBurst® radiofrequency ablation system (AngioDynamics, Latham, NY, USA), which delivers electric energy through the electrode to heat and destroy the surrounding cells [13].
Evaluation of technical success
CEUS using sulfur hexafluoride microbubbles (SonoVue®, Bracco, Milan, Italy) and MRI with liver-specific contrast agent Primovist® (Gd-EOB-DTPA disodium salt) [14] were performed in all patients after 6 weeks to assess technical success, i.e. complete ablation of the tumor. Complete ablation was defined as complete devascularisation of the former tumor and evidence of a surrounding 1 cm safety margin. All images were analyzed by two experienced radiologists (B.P. and I.W.) in consensus reading.
Complications
Complications were recorded for each ablation session by evaluating the radiologic images archived in the local Picture Archiving and Communication System (PACS) as well as the medical records. Complications were classified as minor and major according to the standardized grading system of the Society of Interventional Radiology [15]. Complications were considered as major if, untreated, threatened the life of the patient or resulted in substantial morbidity or in a lengthened hospital stay. Complications other than that were defined as minor. Frequent and expectable procedural side effects such as pain or transient increased liver enzyme levels were not regarded as complications.
Statistics
The analysis was performed using R (version 3.4.0, R Foundation for Statistical Computing, Vienna, Austria). To determine differences in technical success and complication rates between the three groups of tumor thrombi, we used Fisher’s exact test. A p value of p < 0.05 was considered the cut-off point for statistical significance.
Results
Baseline features of the study population are summarized in Table 1. All portal vein thrombi were classified according to their location in Table 2.
Patient characteristics. SD = standard deviation
Patient characteristics. SD = standard deviation
Classification of the portal vein thrombosis based on their location. Type I = segment-level, type II = lobe-level, type III = central
In the follow-up examinations after 6 weeks, residual tumor tissue was seen in 2 out of 15 cases, yielding a technical success rate of 87%. The technical success rate was 100% (8 out of 8) for Type I PVT, 75% (3 out of 4) for Type II PVT and 67% (2 out of 3) for Type III PVT. The differences were not statistically significant with a p value of 0.99 as calculated by the Fisher’s exact test.
No major complications according to the SIR grading system occurred, particularly no patient died due to percutaneous ablation. Minor complications occurred in 2 out of 15 cases (13%). One patient with Type II PVT presented with minor hemorrhage without requirement of any further therapy. Progression of pre-interventionally existing PVT, considered as a minor complication, was seen in one patient with Type III PVT. Again, the differences between the PVT groups were not statistically significant with a p value of 0.99.
Percutaneous tumor ablation in patients with HCC is of decisive importance in the therapeutic strategies by the current state of scientific knowledge. Although portal vein thrombosis is a concomitant feature in patients with HCC, there is only few published data with small sample size studies about the technical success or outcome in patients with portal vein tumor thrombus [16–19]. Our own literature research has shown no scientific publications about thermal ablation in patients with not tumor-related (bland) PVT.
Theoretically PVT should reduce the heat-sink-effect due to lowered macro- and microcirculation leading to a hardly predictable increase in size of the ablation zone (Fig. 2). The heat-sink-effect, which is the dissipation of thermal effect by blood flow, plays a crucial role in percutaneous thermal tumor ablation.
Patient with a pre-existing segment-level (type I) portal vein thrombosis who underwent percutaneous ablation of a HCC. The ablation zone was larger than expected, probably due to reduced macro- and microcirculation.
Although as shown above the size and shape of the ablation volume might change in patients with PVT due to reduced blood flow, in our study we were able to show that a very high technical success rate (87%) can be maintained. This corresponds with the previously published data about success rates of percutaneous thermal ablation of hepatic malignancies. A retrospective analysis of 1000 treatments of 2140 HCC lesions for example revealed technical success in 93% [20]. According to our results the localization of the PVT does not seem to have an impact on the technical success rate.
It can be assumed that the complication rate in patients with pre-interventionally existing PVT increases due to the difficult controllability of the size of the ablation zone. This fact could implicate a higher rate of complications, like laceration of the liver capsule with acute hemorrhage or occlusion of adjacent vessels [21].
In contrast to the described assumption our results of 0% major and only 13% minor complications after percutaneous ablation of patients with HCC and pre-interventionally existing PVT were in line with the data published in the current literature. Big studies about complications after percutaneous radiofrequency ablation showed similar complication rates with 2–7% for major and 2–5% for minor complications [20, 23].
One pre-existing PVT showed progression in the 6-week follow-up which was regarded as a minor complication. One explanatory model might be that therapeutic anticoagulation was discontinued in the peri-interventional setting to reduce the risk of bleeding. Another explanation is a reduced blood flow in the surrounding tissue of the ablation area which might lead to hemostasis and could therefore have a thrombogenic effect.
Limitations of our study are the low number of patients, the missing long-term follow-up and the confirmation of the benignancy of the PVT by imaging only without histopathologic correlation.
In conclusion, it can be stated that reduced perfusion of the liver due to pre-interventionally existing PVT does not seem to have an impact on the ablation success or the complication rate but further studies with higher number of patients and oncologic follow-up should be performed to validate our thesis.