Subclassification and Treatment Results of Ethanol Embolotherapy of Type II Arteriovenous Malformations of the Extremity and Body

Clinical question
How safe and efficacious is ethanol and coil embolization of type II arteriovenous malformations (AVM) using a new subtype classification?

Take-away point
Embolization of type II AVMs using ethanol and coils within the framework of the new subtype classification described is safe and efficacious. Additionally, there is a higher cure rate for type IIa AVMs compared to type IIc AVMs.

Reference
Ko, Seong Eun et al. Subclassification and Treatment Results of Ethanol Embolotherapy of Type II Arteriovenous Malformations of the Extremity and Body. Journal of Vascular and Interventional Radiology, Volume 30, Issue 9, 1443 - 1451

Click here for abstract

Study design
Retrospective review

Funding source
Self-funded or unfunded

Setting
Single institution 
 

 

Summary 

Recognizing that the treatment strategy of type II AVMs differs according to variations in the draining veins, the authors of this retrospective review propose a new subtype classification of type II AVMs according to the morphology of the draining vein.    

A total of 316 patients with congenital body and extremity AVMs were included in the study. After performing arteriography to define morphology and flow dynamics, AVMs were treated with a combination of coils and ethanol. Coils were utilized to reduce the flow velocity within the venous segment, followed by ethanol to obliterate the residual fistulae. For type IIa AVMs (shunts between multiple arterioles and a focal segment of the draining vein) the venous segment was accessed via direct puncture or transvenous cannulation followed by embolization with coils. Once slow flow was seen within the AVM post coil embolization, high-concentration ethanol was injected through a transarterial microcatheter or through the percutaneous needle or transvenous catheter to complete embolization. For type IIb AVMs (multiple shunts between multiple arterioles and a venous sac), the venous sac was punctured, embolized with coils, and ethanol then injected through the needle or transarterial microcatheter. For type IIc AVMs (shunts between multiple arterioles and a long segment of draining vein), direct puncture or transvenous cannulation was performed for coil embolization with ethanol injection performed between deployment of multiple coils. The median clinical follow-up period was 12 months among 75 patients. 

The overall treatment outcomes of ethanol embolotherapy with or without coils were cure in 70 lesions (83%), marked improvement in 6 lesions (7%), improvement in 5 lesions (6%), no change in 1 lesion, and treatment failure or aggravation in 2 lesions (2%). Among AVM subtypes, the cure rate was highest in type IIa AVMs (37 of 39; 95%), followed by types IIb (19 of 25; 76%) and IIc (13 of 20; 65%). There were 19 (10%) minor and 6 (3%) major complications (total of 189 treatment sessions). The complication rate was not significantly different among the subtypes of type II AVMs (P >.05). Major complications included compartment syndrome, acute pancreatitis, arm amputation due to infection, massive hematuria, and lymphedema. There was no mortality related to embolotherapy.


Commentary
Various sclerotic agents and methods for embolization of type II body and limb AVMs have been described in the literature. The authors have demonstrated that ethanol, which may be the preferred embolic agent for such AVMs due its strong devascularization effect, can be used with similar efficacy and less risk of complication compared to methods already described by reducing shunt volume by means of coil embolization of the draining vein and allowing for use of a smaller volume of ethanol. Therefore, understanding the drainage pattern of these AVMs is important and the alternative subtype classification described here for body and limb AVMs will provide an additional framework for how we think about-and approach-these complex vascular lesions.

Post Author
Zagum Bhatti, MD
Assistant Professor
Department of Radiology, Interventional Radiology Division
University of Texas Health Science Center at Houston, Houston, TX
@ZagumBhatti

Locoregional Therapy of Hepatocellular-Cholangiocarcinoma versus Hepatocellular Carcinoma: A Propensity Score-Matched Study 

Clinical question
Do patients with unresectable hepatocellular-cholangiocarcinoma have different outcomes following locoregional therapy compared to hepatocellular carcinoma patients?

Take-away point
Patients with hepatocellular-cholangiocarcinoma have reduced performance free survival and increased distant progression following locoregional therapy compared to hepatocellular carcinoma patients.

Reference
Yu-Hui Huang, et al. Locoregional Therapy of Hepatocellular-Cholangiocarcinoma versus Hepatocellular Carcinoma: A Propensity Score-Matched Study. Journal of Vascular and Interventional Radiology. Sept, 2019: 30; 9.

Click here for abstract

Study design

Retrospective

Funding source

Self-funded or unfunded

Setting

Single-center

Figure 1. Tumor response by modified Response Evaluation Criteria in Solid Tumors following LRT of (a) the entire cohort and (b) the propensity score–matched cohort. (c) On univariate analysis, objective response rate was 30% for HCC-CC and 71% for HCC (OR 0.2, 95% CI 0.05–0.8, P = .02). (d) After propensity score matching, objective response rate was 30% for HCC-CC and 50% for HCC (OR 0.5, 95% CI 0.12–2.2, P = .3). PD = progressive disease; SD = stable disease.

Summary

Mixed hepatocellular-cholangiocarcinoma (HCC-CC) tumors are rare malignancy, comprising less than 5% of primary liver malignancies. Previously published median survival data demonstrated poor outcomes following curative resection with additional high rates of recurrence. Additionally, there has been a reported high level of recurrence following liver transplantation in this population. Currently, there is no accepted treatment paradigm for unresectable disease. Small studies have demonstrated benefit of chemoembolization in this population. The authors intended to evaluate the efficacy of local regional therapy of unresectable HCC-CC patients with propensity-scored matched patients with hepatocellular carcinoma (HCC).

10 patients with HCC-CC treated were identified. Treatments included conventional chemoembolization, radiofrequency ablation, or radioembolization with both glass and resin microspheres. Imaging was performed 1-month following treatment and at 3 month intervals and were evaluated using Response Evaluation Criteria in Solid Tumors (RECIST criteria). Within the propensity-score matched cohorts, objective response was 30% for HCC-CC and 50% for HCC. Additionally, progression free survival was shorter for the HCC-CC population, 2.4 months, compared to 6 months for HCC. HCC-CC patients also had a higher rate of distant progression (60% versus 30%).

Commentary

While small in size, this study indicates that patients with HCC-CC have worse outcomes compared to similar patients with HCC after local regional therapy. This justifies a need for dedicated treatment algorithms for HCC-CC and that adjuvant therapy may be necessary. Additionally, further investigation is needed to evaluate the efficacy of the first and second line systemic agents currently approved for HCC in the HCC-CC population.

This study has the standard limitations that come with the small sample size and retrospective nature, however given the rarity of HCC-CC, much of this is expected and difficult to avoid. The use of propensity score-matched analysis attempted to minimize confounding variables. Additionally, the study is limited due to the heterogeneity of treatment provided to the HCC-CC cohort, including a mix of resin and glass microspheres for radioembolization.

Post Author:
David M Mauro, MD
Assistant Professor
Department of Radiology, Vascular and Interventional Radiology
University of North Carolina
@DavidMauroMD

Type IIIb Endoleak Is Not Extremely Rare and May Be Underdiagnosed after Endovascular Aneurysm Repair

Clinical Question
To investigate the incidence of type IIIb endoleak associated with the Zenith stentgraft.

Take-away Point
A 1.6% incidence of type IIIb endoleak in the Zenith stentgraft is more common than most practitioners may believe. Underdiagnosis and misdiagnosis of type IIIb endoleaks should always be considered in patients with sac expansion.

Reference
Fujimura N, Ichihashi S, Matsubara K, Shibutani S, Harada H, Obara H, Kichikawa K, Kitagawa Y. (2019). Type IIIb Endoleak Is Not Extremely Rare and May Be Underdiagnosed after Endovascular Aneurysm Repair. J Vasc Interv Rad, 30:1393-1399. doi:10.1016/j.jvs.2019.03.006

Click here for abstract

Study Design
Retrospective multicenter analysis

Funding Source
Self-funded

Setting
Multiple academic centers in Japan

Summary

Type IIIa endoleak represents a disconnection of the graft components while type IIIb endoleak is a deterioration of endoprosthesis material leading to aneurysm pressurization. A retrospective review of EVAR using the Zenith stent graft (Cook, Bloomington, Indiana) from 2007-2016 was performed across 11 centers in Japan, identifying 433 patients. A type IIIa endoleak was identified in 0.2% of patients (1/433) and type IIIb endoleak was identified in 1.6% (7/433). Of these seven patients, four had an increase in sac size and five underwent repeat intervention; four of the five repeat interventions patients were diagnosed as type IIIb at the time of reintervention. Three type IIIb endoleaks were from the limb component, three from the main body and one from both.

While type III endoleak was reportedly around 3% for older generation stent grafts, newer third-generation stent grafts were thought to have incidence much lower; the Zenith endograft has previously reported an incidence of 0.4-4.3% however this does not differentiate type IIIa and IIIb. The authors postulate that contrast-enhanced US may aid in the detection of type III endoleaks as it is a dynamic exam whereas CT may result in misdiagnosis.

 

Commentary

Type IIIb is an infrequently thought of cause of endoleak post EVAR. This Japanese multicenter study identified a 1.6% incidence of IIIb endoleaks for the Zenith stentgraft; a 1.8% incidence of all type III endoleaks which is within the reported literature range for Zenith stentgrafts. A recent review of the literature by Kwon et al. in J Vasc Surg on type IIIb endoleaks in contemporary EVARs yielded 46 case reports with 35% identified in the main body and 33% at the flow divider; 61% were treated endovascularly. They further conclude that identification of type IIIb endoleak is challenging and the true incidence is not well known.

It is important to keep in mind all types of endoleak during evaluation of sac enlargement. Perhaps dynamic studies such as contrast enhanced ultrasound may play an important role in the future for diagnosis of endoleak type when aneurysm sac expansion is noted on CTA but type of endoleak is not well identified.

Post Author
Nicole A. Keefe, MD
Fellow, Interventional Radiology
Department of Radiology and Medical Imaging
University of Virginia
@NikkiKeefe