Thursday, September 22, 2016

From the SIR Residents and Fellows Section (RFS)

Teaching Topic: Position Statement on Noninvasive Imaging of Peripheral Arterial Disease by the Society of Interventional Radiology and the Canadian Interventional Radiology Association

Dhanoa D, Baerlocher MO, Benko AJ, Benenati JF, Kuo MD, Dariushnia SR, Faintuch S, Midia M, Nikolic B. Position statement on noninvasive imaging of peripheral arterial disease by the Society of Interventional Radiology and Canadian Interventional Radiology Association. J Vasc Interv Radiol. 27: 947-51.

This manuscript reviews and provides recommendations for noninvasive lower-extremity imaging of PAD. This includes both functional and anatomic tests. The functional or physiologic tests include the ankle-brachial index (ABI), segmental limb pressures, pulse volume recordings (PVRs), segmental Doppler waveforms, and oxygen testing. The anatomic tests include duplex ultrasound (US), computed tomography (CT), and magnetic resonance (MR) imaging. Because of the complexities and degree of discussion needed for each study, CT and MR imaging will be discussed in a future manuscript. Given changes to our delivery of healthcare in the United States, the manuscript serves an important role in defining the appropriate use of noninvasive imaging to improve patient selection and documenting post-procedure outcomes.

Clinical Pearls

What imaging modalities constitute a complete noninvasive examination of peripheral arterial disease?

A typical noninvasive examination should always include an ABI with concomitant pulse volume recordings (PVRs), continuous-wave Doppler analysis, segmental pressures, and exercise testing.

Why is a position statement on noninvasive imaging of PAD needed?

The implementation of the Affordable Care Act has brought about drastic changes to the reimbursement models for medicine. Alternative methods have emerged to enact the intended paradigm shift towards value-based and outcome-oriented delivery of healthcare instead of the standard, traditional merit-based fee-for-service model. Thus, the appropriate use of noninvasive imaging to improve pre-procedural patient selection, as well as to objectively document post-procedure outcomes, is of critical significance.

Questions to Consider

How are ABIs calculated and how are results utilized to grade PAD?

ABIs are calculated by dividing the ankle systolic blood pressure by the brachial artery systolic blood pressure. Both upper extremity BP are obtained and the higher of the two are utilized. For the ankle systolic BP, the greater of the dorsalis pedis or posterior tibial artery should be used.

If there is a >15mmHg discrepancy between the upper extremity systolic blood pressures, hemodynamically significant disease should be considered to be present proximal to the brachial artery with the lower systolic blood pressure.

Stielger et al. proposed an ABI-Based Grading Scale as follows:
  • >1.3 : Falsely high value (suspicious for medial sclerosis) 
  • 0.9-1.3 : Normal 
  • 0.75-0.9 : Mild PAD 
  • 0.4-0.75 : Moderate PAD 
  • <0.4 : Severe PAD 
What are the important limitations to ABIs?

ABIs can be falsely elevated in patients with heavily calcified arteries. In these circumstances, it is recommended that toe brachial index be utilized as calcifications are rarely found at the great toe. A TBI of >0.65 is considered normal and TBI < 0.4 is considered severe PAD.

What are the four phases of a normal typical waveform in segmental PVR?

The four phases are:

1. A rapid systolic upstroke
2. Rapid diastolic downstroke
3. Prominent dicrotic notch (*The dicrotic notch denotes the closing of the Aortic Valve)
4. Normalization to baseline before the next cycle

How do you interpret arterial Doppler waveforms?

Normal arterial Doppler waveforms are triphasic consisting of a sharp systolic upstroke, reversal of flow below baseline and then a short forward component in late diastole. Mild PAD leads to a bisaphic waveform in which the short forward component in late diastole is lost. As PAD progresses to severe disease, the waveform continues to flatten and becomes monophasic with loss of the flow reversal. The waveform becomes rounded with a slow upstroke and slow downstroke creating the classic postobstructive tardus parvus signal.

Post Author:
Andrew Niekamp, MD
Diagnostic Radiology Resident, PGY-3
UT Houston

Thursday, September 15, 2016

From the SIR Residents and Fellows Section (RFS)

Teaching Topic: Endovascular Repair of Celiac Artery Aneurysm with the use of Stent Grafts

Zhang W, Fu YF, Wei PL, E B, Li DC, Xu J. Endovascular repair of celiac artery aneurysm with the use of stent grafts. J Vasc Interv Radiol. 2016. 27 (4): 514-8.

A recent article in JVIR evaluated the feasibility, safety, and long-term outcome of stent-graft insertion for endovascular repair of celiac artery aneurysms (CAAs). 10 patients with CAAs underwent endovascular repair via stent-graft insertion in a single center. Follow-up CTAs were performed at 1, 3, 6, and 12 months. There was no evidence of endoleak, stent obstruction, or splenic infarction during the follow-up period and all 10 patients had CAA sac shrinkage or increased CAA sac thrombus on follow-up imaging.

Clinical Pearls

What do we know about aneurysms involving visceral arteries?

Visceral arterial aneurysm (VAA) are rare, with an incidence of 0.1%–2%. Celiac artery aneurysms (CAA) constitute 4.8%–6.3% of all VAA cases. In this study, the treated aneurysms varied from 2.1 x 1.6 cm to 8.8 x 7.1 cm. They frequently present as a life-threatening emergency and are often fatal if associated with rupture. An aneurysm ≥ 20 mm in size is considered sufficient to warrant treatment if the patient’s overall condition permits it. Recent studies have reported treatment options for CAAs consisting of open surgery or embolization. Few publications have reported stent-graft insertion for endovascular repair in patients with CAAs.

What does the data show on endovascular treatment vs. open surgery?

In a recent article by Shukla et al. comparing outcomes between endovascular treatment (n = 122) and open surgery (n = 59) for VAAs, results show that endovascular treatment and open surgery are equally durable for patients with intact VAA, but endovascular treatment for ruptured VAAs was associated with a lower 30- day mortality rate (7.4% vs 28.6%; p<0.05) and better 2-year overall survival (69.4% vs 46.4%; p<0.05).

Questions to Consider

What are possible complications of CAA aneurysms and Stent Grafts for the CAA?

As mentioned above, the most dangerous complication of untreated CAAs is life-threatening hemorrhage. However, endovascular treatment of CAAs with stent grafts carries its own set of risks including dissection, stent thrombosis with end-organ ischemia, splenic infarction, and endoleaks.

While performing an endovascular Stent Grafting of a CAA, what do you have to watch for?

The anatomy of celiac artery aneurysms is complex, given that multiple branches can exit the aneuryms. Without identifying and potentially embolizing relevant branches, endoleaks may develop after stent graft placement. Equally, the interventionalist must define the distal landing zone for the endograft to assure a good seal and likelihood of long term patency. Pre-procedure planning with CTA and detailed catheter angiography are of utmost importance. As mentioned previously, endoleaks are of a concern as well, however, none resulted in the cases presented in the manuscript above. Many CAAs are associated with median arcuate ligament compression and post-stenotic dilation. Ligamentous compression may permanently deform balloon expandable stent grafts,

What are the limitations of the above study and why?

While the study sample was small (n=10), given low prevalence of the disease, this is nevertheless a meaningful number to show safety and efficacy. In addition, the retrospective nature introduces the possibility of selection bias. Can all CAAs be treated with stent graft? What can preclude a CAA from stent graft repair? If a CAA may be treated by embolization or stent graft, how should we decide the optimal treatment approach? 

What are the different types of Endoleaks?

While the classification system for endoleaks was originally intended for and applied to abdominal aortic aneurysms (AAAs), it can be used for discussion regarding stent graft treatment for aneurysm exclusion in other vascular territories.

Type I: Persistent filling of the aneurysm sac due to incomplete seal at the proximal or distal end of the stent graft.
Type II: Persistent filling of the aneurysm sack due to retrograde branch flow from collateral vessels.
Type III: Blood flow into the aneurysm sac due to ineffective sealing of overlapping graft joints.
Type IV: Blood flow into the aneurysm sac due to the porosity of the graft fabric, causing blood to pass through the graft joints or rupture of graft fabric.
Type V: Aneurysm sac expansion without clear evidence of endoleak origin.

Additional citations:
Shukla AJ, Eid R, Fish L, et al. Contemporary outcomes of intact and ruptured visceral artery aneurysms. J Vasc Surg 2015; 61:1442–1448.

Post author:
Ali Alikhani, MD
Diagnostic Radiology Resident, PGY-4
University of Tennessee Methodist Healthcare

Tuesday, August 16, 2016

Modifications to IRE: Are we there yet?

In an upcoming issue of JVIR, researchers from Stanford University present their experience in producing spherical ablation using irreversible electroporation (IRE) using a single electrode and a grounding pad. The drawbacks to IRE are evident to any interventionalist that regularly uses this treatment technique in clinical practice and any research that addresses these shortcomings should come as welcome news. In short, IRE has some limitations as an ablative tool including cardiac arrhythmias, requirement for chemical paralysis to prevent muscle contractions, and need for placing probes close together. This study investigated two modifications to attempt to address these issues. One is the use of high-frequency IRE (H-FIRE). Besides sounding awesome, H-FIRE uses a rapid burst of 0.25-5-┬Ás alternating-polarity pulses instead of the long-duration monopolar IRE pulse. This produces less susceptibility to distortions in the magnetic field distribution created by tissue heterogeneities and produces more predictable treatment zones. Further, H-FIRE inhibits stimulation of action potentials and treats without inducing muscles contractions. However, H-FIRE has been limited by the small size and “peanut” shape of the ablation zone. This study included a modified probe design to include a single electrode at the target site and a distal grounding pad. This differs from the traditional design with two separate electrical conductors co-located in the same tumor volume. Computed simulations were performed by using varying voltages, electrode exposure lengths, and tissue types. A vegetable (potato) tissue model was then used to compare ablations created by conventional and high-frequency IRE protocols by using the two collinear electrodes or a single electrode configured with a grounding pad. In addition, the new electrode configuration was evaluated in ex vivo liver tissue. Researchers found that the new probe + grounding pad configuration produced larger and more spherical ablation volumes than the traditional configuration in computed simulations and tissue models. Potato model studies verify that the probe + grounding pad configuration produces larger and more spherical ablations than those produced by the traditional method. The researchers made no comment regarding the flavor of the ablated potato.


This paper does a very good job at identifying an existing treatment modification (H-FIRE) and modifying that further with creating a single probe system with the use of a grounding pad. As many interventional oncologists would attest, it would be very advantageous in clinical practice to have an IRE system that used a single probe and was able to create spherical ablation zones. There are some additional hurdles for this technical modification. As mentioned in the paper, the creation of large ablations is theoretically possible with the new configuration if voltages are between 3,000 and 5,000 V. However, 3,000 V is the highest voltage reported for clinical IRE treatments, and the thermal implications of these higher voltages are not fully understood. In addition, in the new configuration, the electrical current will travel between the single electrode and the grounding pad. There is the potential for nontarget thermal and electrical effects. Despite these obstacles, many interventionalists will be looking forward to seeing this technology progress through development to the market.

Click here for abstract

Sano MB, Fan RE, Hwang GL, Sonn GA, Xing L. Production of Spherical Ablations Using Nonthermal Irreversible Electroporation: A Laboratory Investigation Using a Single Electrode and Grounding Pad. J Vasc Interv Radiol. doi: 10.1016/j.jvir.2016.05.032

Post Author:
Luke R. Wilkins, MD
University of Virginia

Monday, August 1, 2016

Results from the RENOVA study show superiority of stent graft to balloon angioplasty in the treatment of arteriovenous graft stenosis

A national team of researchers led by Dr. Ziv Haskal of the University of Virginia have recently published their 2-yr results from the Prospective, Randomized, Concurrently-Controlled Study of an ENdOVAscular Stent Graft. The RENOVA trial, was conducted at 28 clinical sites in the U.S. and was designed to compare the use of an ePTFE stent graft to PTA alone for the revision of AV access graft-venous anastomotic stenosis through 2 years. A total of 270 patients were enrolled with malfunctioning AVG anastomotic stenoses of ≥50%. 138 patients underwent placement of a stent graft (SG) and 132 received angioplasty (PTA) alone. At 12 months, treatment area primary patency (TAPP) was: SG 47.6% vs. PTA 24.8% (p< 0.001) and access circuit primary patency (ACPP) was SG 24% vs. PTA 11% (p=0.007). At 24 months, TAPP was: SG 26.9% vs. PTA 13.5% (p<0.001) and ACPP was SG 9.5% vs. PTA 5.5% (p=0.01). The estimated number of re-interventions before graft abandonment was 3.4 for the SG group vs. 4.3 for PTA group. There were no significant differences in the adverse events (p>0.05) excepting restenosis requiring reintervention rates of 82.6% in PTA patients vs. 63.0% in SG (p<0.001). The authors concluded that at 2 years, Stent graft use provided a “sustained, greater than two-fold advantage over angioplasty in treatment area and overall access patency.”


The FLAIR trial from 2010 showed 6-month superiority of the SG for both prolonging uninterrupted access function and patency of the target stenosis and led to FDA approval of the device. The RENOVA study has replicated and affirmed the pivotal trial results, extending the follow up to 2 years, and establishing the long-term superiority of this stent-graft for revision of failing anastomoses in prosthetic grafts as compared to repeated PTA. As detailed in the paper, the healthcare costs of maintaining a failed AV access are substantial. Further, given the burden associated repeat interventions upon patients, the use of SGs in this clinical scenario may offer significant advantages to the dialysis patient population and to the healthcare system as a whole.

Click here for full text

Haskal, Ziv J., et al. "Prospective, Randomized, Concurrently-Controlled Study of a Stent Graft versus Balloon Angioplasty for Treatment of Arteriovenous Access Graft Stenosis: 2-Year Results of the RENOVA Study." Journal of Vascular and Interventional Radiology 27.8 (2016): 1105-1114.

Post Author:
Luke R. Wilkins, MD
University of Virginia

Tuesday, July 12, 2016

Can IVUS-Guided TIPS be more than just “Fancy TIPS”?

Intravascular ultrasound guided access to a portal vein during TIPS creation is an alternate technique to the conventional method of fluoroscopically guided access following hepatoportal venography. This retrospective study was performed to compare three parameters: mean time to PV access, radiation dose, and capsular perforation, between the intravascular US guided access (n=55) and conventional fluoroscopically guided access (n=54) groups. Further, the results were stratified by operator experience (10 total operators; 5 with <20, and 5 with > 20 conventional TIPS procedures before this study, classified as inexperienced, and experienced, respectively). There was no significant difference in the sex, age, BMI, etiology of cirrhosis, or indication for TIPS, in the two study populations. The study found that parameter mean time to PV access was significantly lower for the intravascular US guided access group overall, but showed no difference when comparing experienced operators only. Further, parameter mean time to PV access continued to decrease over the course of the study in the US guided access group. In measuring radiation dose and capsular perforation, both were statistically lower in the intravascular US guided group than the conventional group.


Alternate techniques to gaining portal vein access during TIPS include three dimensional navigation using fusion, prior percutaneous PV marking, utilization of cone beam CT, transabdominal ultrasound, and intravascular ultrasound. While there have been studies comparing the conventional vs. intravascular US guided PV access techniques in the past, this is the first one to take operator experience into account, and has a slightly higher number of patients (109 vs. 100). This study also does not compare the number of needle passes performed to access the PV, as this was felt to be a subjective measurement. Further, time for femoral vein access and IVUS set-up was not included in this study nor was the added cost associated with the device. As technology progresses, cost, operator preference, and quality control will dictate whether such ancillary techniques will contribute in regularly to TIPS creation.

Click here for abstract

Pillai, Anil K., et al. Utility of Intravascular US–Guided Portal Vein Access during Transjugular Intrahepatic Portosystemic Shunt Creation: Retrospective Comparison with Conventional Technique in 109 Patients. Article in press. doi: 10.1016/j.jvir.2016.05.010

Post Author:
Nabeel Mecci, MD
University of Virginia
PGY-2 Radiology Resident

Thursday, June 2, 2016

New Study Investigates Retrieval and Complication Rates of Denali and Celect IVC Filter

In an article in press, Researchers from the University of Chicago compared the indwelling and retrieval complications of Denali and Celect filters placed in the infrarenal IVC. This retrospective review was conducted over 2 years and included 171 Denali and 162 Celect filters. 91% (303) of the filters were placed from a jugular approach. Follow-up CT images were evaluated in 58 filters from each group. The Denali group showed significantly lower incidence of caval strut penetration (1 vs 12) and tilt (1 vs 15). Retrieval attempts were made on 43 Denali filters and 53 Celect filters with similar indwelling times. There was significantly less fluoroscopy time in the Denali group (3.1 vs 6.0 min). In addition, there was a trend toward fewer “complex” retrievals in the Denali group (2 vs 10, p = 0.06). ​Complex retrieval was defined as “successful retrieval attempts in which conventional retrieval techniques, failed necessitating one of the adjunctive retrieval methods described earlier.” It is noted that only 5 of the 15 tilted Celect filters presented for retrieval and 3 of those 5 required snare-over-guide wire loop technique or forceps for retrieval. Complex retrievals were associated with increased fluoro time (14.3 vs 3.2 min) and increased dose (333.4 vs 149.5 mGy).


While the current study demonstrates that “complex” techniques are needed more frequently in retrieval of Celect filters when compared with Denali filters, it remains to be seen if this will significantly impact practice pattern. “Complex” retrievals are becoming much less, well.… complex. Given increased familiarity with advanced techniques, many centers are seeing a marked decrease in fluoro times for even the most challenging retrievals. When combined with overall decreasing rates of strut penetration and tilt (even in the Celect group), the result is low average rates of fluoroscopy time when compared to the group as a whole (Celect 6.0 min in the current study). Further, depending on local pricing, there is likely an increased cost for Denali when compared to some of the older generations of IVC filters. Before a practice considers changing filters, a comprehensive view of potential costs (clinical and financial) of retrieval should be thoroughly considered.

Click here for abstract

Bos AS, Tulius T, Patel M, Leef JA, Navuluri R, Lorenz JM, Van Ha TG. Indwelling and retrieval complication of Denali and Celect infrarenal vena cava filters. J Vasc Interv Radiol. June 2016. doi: 10.1016/j.jvir.2016.03.034

Post Author:
Luke R. Wilkins, MD
University of Virginia

Monday, May 23, 2016

Endologix Unibody Endograft performance in Aortoiliac occlusive disease

The field of endograft manufacture and design has been an active one with improved performance and lower profile devices under continuous development. A multicenter retrospective review was conducted to evaluate safety, efficacy and early patency rates of the 17Fr ipsilateral, 9Fr contralateral Endologix AFX unibody endograft in the treatment of aortoiliac occlusive disease and recently published in the European Journal of Vascular and Endovascular Surgery.

91 patients were treated for aortoiliac occlusive disease. Patients with concomitant aortic aneurysm were excluded. Mean follow up was 22.2 months. Technical success was 100%. Complication rate was 22% with 1 mortality in the 30 day follow up period from extensive pelvic thromboembolism. At 1 year 73% of patients experienced improvement in Rutherford stage of -3 or greater compared with baseline. At all points primary patency were >90%, assisted patency rates were >98% and secondary patency rates were 100%.


Use of the AFX unibody stent-graft had a high rate of technical success and a low morality rate. Patency rates were comparable to with open aortobifemoral bypass, kissing stent and CERAB techniques. This study is the largest so far in the evaluation of the AFX unibody graft performance. Notably 82% of patients treated in this study had baseline TASC D disease making the already favorable patency data arguably even more so. 41% of these cases were done percutaneously with hemostasis achieved using the Proglide (Abbott Vascular, Abbott Park, IL, USA). Although the authors indicate that the use of re-entry devices was not necessary, they say these may be helpful to obtain femoral-femoral wire access in some cases. It is not clear what proportion of aortoiliac lesions were successfully crossed in an intraluminal position and which if any required subintimal angioplasty (SIA). Furthermore if there were lesions which required SIA it is unclear if these were included.

Although the complication rate was high (22%) none of the complications appear to be device related. The majority of the complications appear to be related to the open femoral access utilized in 59% of cases. The incidence of groin related complications in this study is comparable with that found in open surgical repair and may reflect that this subset of patients is exposed to the same risks found in open surgery. Overall, this manuscript serves as continued support in using the Endologix device in cases of aortoiliac occlusive disease in which the cranial extent of occlusion makes kissing stents unfavorable.

Click here for abstract

Maldonado TS, Westin GG, Jazaeri O, et al. Treatment of aortoiliac occlusive disease with the Endologix AFX unibody endograft. Eur J Vasc Endovasc Surg. May 2016. doi: 10.1016/j.ejvs.2016.04.003

Post Author:
Marco Ugas, MD
University of Virginia
PGY-5 Radiology Resident