Evaluation of a Device Combining an Inferior Vena Cava Filter and a Central Venous Catheter for Preventing Pulmonary Embolism Among Critically Ill Trauma Patients

Summary

Critically ill trauma patients have elevated risk for pulmonary embolism (PE) due to a combination of immobility, depletion of anti-coagulation factors, and contraindications to anticoagulation (AC) in the setting of hemorrhage. In this population, PE is one of the leading causes of death in patients surviving >1 day. Tapson et al. conducted a prospective multicenter single-arm clinical trial to evaluate the efficacy and safety of a new device, the Angel Catheter (Bio2 Medical) which combines an inferior vena cava filter (IVC) with a central venous catheter. This device received 510(k) clearance from the Food and Drug Administration for use as a short term IVC filter and central venous catheter on 7/28/2016. The filter component is made of nitinol, closed cell in geometry, contains no wall anchoring hooks, and is permanently attached to the 9-F triple-lumen central venous catheter. The enrolled study patients had contraindications to pharmacologic venothromboembolism prophylaxis without or with confirmed lower extremity deep venous thrombosis (DVT) or PE. Between February 2015 and December 2015, 163 intensive care unit patients underwent placement of the Angel catheter, of whom 18/163 (11%) had confirmed proximal lower extremity deep venous thrombosis (DVT) at baseline. For the large majority of patients, device insertion was performed bedside in the ICU, with abdominal radiography to confirm post-deployment filter position between L1-L2 and L4-L5 interspaces. The primary endpoint was freedom from clinically significant pulmonary embolism (PE) 72 hours after device removal or discharge and was achieved in 100% of patients. The Angel catheters were removed in 143 (88%) patients, with the large majority of removals occurring once prophylactic anticoagulation was safely initiated, at a mean of 5.5 hospital days. Notably, 12 (8%) catheters which were removed inadvertently by the patient. Cavograms were performed in 129 patients prior to device removal, of which 31 (24%) showed thrombus in the filter component, including 14 (8.6%) with large clot burden (>25% volume of filter basket) and 17 (10%) with low clot burden (<25% volume of filter basket). These patients with thrombus subsequently underwent anticoagulation therapy and/or placement of a conventional IVC filter. Major bleeding occurred in 5 (3%) of all patients receiving the Angel catheter, and catheter related thrombosis occurred in 20 (12%) cases.



Commentary

Venothromboembolism (VTE) is known to be a major contributor of morbidity and mortality in critically ill trauma patients. Multicenter prospective randomized controlled trials for IVC filters have not included this patient population, but nonrandomized studies have shown decreased rates of PE and PE-related fatalities in patients who receive filters. Although conventional IVC filters may be placed in these patients, studies have demonstrated that the majority of filters are placed late in the ICU course, and therefore the patients may not be adequately protected from PE in their early hospitalization, when they may have contraindications to pharmacologic prophylaxis (reflected in the mean time to initiation of AC of 5.5 days in this study). IVC filters placed in the trauma setting have also been associated with low retrieval rates. The Angel catheter theoretically circumvents both of these pitfalls as a device that may be easily deployed and removed at bedside in the ICU. The results of this study and the European Angel Catheter Registry suggest that this device is both effective for PE prevention and acceptable safety profile in critically ill patients, and may be an exciting addition to the VTE prevention armamentarium in this patient population. As a single arm clinical trial, this study does not offer any comparison to placebo or conventional IVC filter placement, and provides relatively short duration of clinical follow-up, so additional evidence will be needed to elucidate optimal utilization of this device. Because no pre-placement cavography is performed and cross-sectional imaging may not be available in all patients, one could anticipate scenarios in which the Angel catheter could be malpositioned from variant caval or renal vein anatomy and provide suboptimal caval filtration. The relatively high rate of inadvertent catheter dislodgement reported in this study also raises concerns for potential associated morbidity if a filter is pulled out of a vessel without being adequately collapsed in patients not on pharmacology thromboprophylaxis. Despite these limitations, this study provides an important preliminary evaluation of the Angel catheter, which may one day assume a pivotal role in early PE prevention in critically ill patients.

Click here for abstract

Tapson VF, Hazelton JP, Myers J, et al. Evaluation of a Device Combining an Inferior Vena Cava Filter and a Central Venous Catheter for Preventing Pulmonary Embolism Among Critically Ill Trauma Patients. J Vasc Interv Radiol. 2017;28(9):1248-54.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
Venous Health Program Faculty
University of Michigan Health System
Michigan Medicine

James X. Chen, MD
Fellow in Interventional Radiology
Hospital of the University of Pennsylvania

From the SIR Residents and Fellows Section (SIRRFS)

Teaching Topic: Outcomes of Ultrasound–Guided Thrombin Injection of Nongroin Arterial Pseudoaneurysms

Valesano JC, Schmitz JJ, Kurup AN, Schmit GD, Moynagh MR, Atwell TD, Lewis BD, Lee RA, Callstrom MR. J Vasc Interv Radiol. 2017 Jun 1. pii: S1051-0443(17)30477-3. doi: 10.1016/j.jvir.2017.05.003. [Epub ahead of print]

Link to Abstract

Arterial pseudoaneurysms are among of the most frequently encountered complications of interventional vascular procedures. Pseudoaneurysms are contained pulsatile hematomas that arise from arterial puncture sites that fail to fully seal. Historically, pseudoaneurysms required surgical vessel closure, though now ultrasound-guided thrombin injection has now become the mainstay first-line treatment. The safety and efficacy of ultrasound-guided thrombin injection for femoral pseudoaneurysms has been well established in large trials and systematic reviews. [1] However, despite the increasing popularity of upper extremity endovascular access, reports of non-groin pseudoaneurysms treatment are limited.

In this study, Valesano et al. at the Mayo Clinic retrospectively review 39 cases of thrombin injection treatment for non-groin pseudoaneurysms. Of these 29 were in the upper extremity and 20 of those were of the brachial artery. Thirty-six were iatrogenic of nature and 22 of those were from arterial access. They report a 100% success rate in achieving zero Doppler flow immediately after thrombin injection and an 84.8% success rate in sustained thrombosis at 1-3 days after treatment

Clinical Pearls

Though there are many techniques, what are the general steps in ultrasound guided thrombin injection?

The authors describe their technique as follows:

1. Obtain baseline arterial duplex US imaging and Dopper waveforms of both the pseudoaneurysms and the origin artery proximal and distal to the site.

2. Evaluate pulses with ultrasound or palpation. Additionally, assess the relevant baseline sensorimotor function.

3. Using sterile technique and 1% lidocaine for anesthesia, advance a 20-25 gauge needle under ultrasound visualization until the tip of the needle is in the pseudoaneurysm sac.

4. Incrementally inject small aliquots of about 100 IU of 1000 IU/1 mL recombinant topical thrombin (The Medicines Company, Parsippany, New Jersey).

5. With intermittent Doppler imaging, monitor the pseudoaneurysm during injection until complete elimination of Doppler signal.

6. Reassess proximal and distal artery with US, pulses, and sensorimotor symptoms.

7. Follow-up imaging with duplex and Doppler US is recommended one day after the procedure.

What are the characteristics of pseudoaneurysms at higher risk of treatment failure?

It is difficult to draw conclusions from the small number of treatment failures from this series, however it appeared that the neck size was more predictive than the total size. The average size of pseudoaneurysms that remained successfully thrombosed at follow up were larger at 2.3cm +/- 1.6, while the average size for treatment failures were 2.0cm +/- 0.7. The neck diameter for treatment failures averaged 2.6 mm +/- 0.5 and 2.1 mm +/- 0.8 for treatment successes. Because of the small sample size, these results trended towards, but did not show statistical significance. Of the treatment failures, three were successfully treated with one retreatment, and the other two were successfully treated after two retreatments.

Questions to Consider

Though the authors reported no complications, what are the potential complications of this procedure?

In the small previous reports, there have been two incidents of brachial artery thrombosis. In 2014, Garvin et al reported 14 cases of treatment with thrombin in the upper extremities. [2] In one of these cases, brachial artery thrombosis was identified by decreased Doppler amplitude distal to the pseudoaneurysm fifteen minutes after treatment. Emergent surgical revascularization was performed. Kang et al reported 5 upper extremity treatments in 2000. [3] In one case, the neck of the pseudoaneurysm persisted after initial thrombin injection, and after a small amount was injected into the residual neck, the patient began to experience symptoms of hand ischemia. Pulses returned and symptoms resided in less than ten minutes after treatment with 5000 IU of heparin.

What special considerations should be taken in patients on anticoagulation therapy?

Successful pseudoaneurysm treatment with thrombin in the setting of concurrent anticoagulation therapy has been well documented for femoral pseudoaneurysms in patients on heparin, warfarin, clopidogrel, and aspirin. [4] While the authors of this study did not report the anticoagulation status of the treated patients, in the previously mentioned study of 14 cases, 9 patients were on anticoagulants for atrial fibrillation. The efficacy in anticoagulated patients compared to ultrasound-guided compression remains one of the leading justifications for thrombin therapy.

Additional Sources

1. Tisi PV, Callam MJ. Treatment for femoral pseudoaneurysms. Cochrane Database Syst Rev. 2013 Nov 29;(11):CD004981. doi: 10.1002/14651858.CD004981.pub4. Review. PMID: 24293322.

2. Garvin RP, Ryer EJ, Yoon HR, Kendrick JB, Neidrick TJ, Elmore JR, Franklin DP. Ultrasound-guided percutaneous thrombin injection of iatrogenic upper extremity pseudoaneurysms. J Vasc Surg. 2014 Jun;59(6):1664-9. doi: 10.1016/j.jvs.2014.01.009. Epub 2014 Feb 20. PMID: 24560862.

3. Kang SS, Labropoulos N, Mansour MA, Michelini M, Filliung D, Baubly MP, Baker WH. Expanded indications for ultrasound-guided thrombin injection of pseudoaneurysms. J Vasc Surg. 2000 Feb;31(2):289-98. PMID: 10664498.

4. Krueger K, Zaehringer M, Strohe D, Stuetzer H, Boecker J, Lackner K. Postcatheterization pseudoaneurysm: results of US-guided percutaneous thrombin injection in 240 patients. Radiology. 2005 Sep;236(3):1104-10. Epub 2005 Jul 29. PubMed PMID: 16055694.

Post Author
Charles Hyman, MS4
Warren Alpert Medical School of Brown University

Midterm Clinical Outcomes and MR Imaging Changes after Transcatheter Arterial Embolization as a Treatment for Mild to Moderate Radiographic Knee Osteoarthritis Resistant to Conservative Treatment

Summary

Osteoarthritis (OA) is a common and debilitating condition that leads to decreased functional status and quality of life in the affected population. Low grade inflammation with associated angiogenesis is now recognized as part of the etiopathophysiology of OA symptoms. Okuno et al. present a prospective, single-center observational study on outcomes following transcatheter arterial embolization for mild-to-moderate knee osteoarthritis. 95 knees in 72 patients with symptoms resistant to conservative therapy, of median duration 30 months, were included. All patients with osteonecrosis, local infection, malignancy, rheumatoid arthritis, and previous knee surgery were excluded. The embolization procedures were performed using antegrade femoral artery access, with a 3-French catheter to select the popliteal artery. Additional selective arteriography was performed with a 1.7-French catheter if needed. Abnormal vessels were identified from a mean of 3.2 arteries per knee. Embolization was performed with imipenem and cilastatin sodium in the majority of patients and 75 um Embozene in the minority of patients with hypersensitivity to the antibiotics. Embolization endpoint was suppression or reduction of filling of the abnormal popliteal vasculature and reduction in local tenderness by palpation. The authors state that patients described pain, itching, or heat sensation at the site of symptoms when the culprit vessel was embolized. All cases were technically successful. The most common adverse events were puncture site bleeding in 12 patients and transient cutaneous color changes in 4/7 patients treated with Embozene, with no major adverse events. At all follow-up time points, there were significant reductions in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores compared to baseline (mean 12.1 at baseline versus 6.2, 4.4, 3.7, 3.0, and 2.6 at 1, 4, 6, 12, and 24 months, respectively; all P< .001). There was no difference in clinical outcomes between embolization with imipenem and cilastatin sodium and Embozene. Magnetic resonance imaging follow-up in 35 knees of 29 patients at 2 years showed decrease in synovitis, with no osteonecrosis or other evidence of aggressive progression in degenerative OA changes.



Figure. Angiographic findings before and after transcatheter arterial microembolization in a 69-year-old patient treated with imipenem and cilastatin. (a) Selective arteriography from the descending genicular artery before embolization showing abnormal neovessels (arrow) adjacent to the medial condyle (MC). (LC ¼ lateral condyle.) (b) Post-embolization arteriography demonstrating elimination of hypervascularity.

Commentary

Although joint replacement is an effective management of severe OA, conservative treatment is used for most patients with mild-to-moderate OA. Many of these patients struggle with symptom control and new therapies could have an impactful role in improving quality of life and functional mobility. The recognition that synovitis and associated neovascularity underlie OA symptoms has raised interest of transcatheter embolization for symptomatic OA. At this nascent stage in investigation of this therapy, this study from Okuno et al. provides an important groundwork by delineating the technical approach, confirming a favorable safety profile, and demonstrating promising midterm clinical and imaging follow-up results. No patients in this study had any ischemic complications, including osteonecrosis, and did not have significant progression of OA from the decreased pain and increased mobility and joint loading. The authors also used a small volume of embolic material to reduce the risk of embolization-associated inflammation which could theoretically contribute to short-term worsening of joint inflammation. This study was limited by the lack of a control group and blinding and additional studies are certainly needed to validate efficacy and safety. Nonetheless, this work provides a promising foundation for subsequent studies into transcatheter embolization as a novel treatment approach for symptomatic OA.

Click here for abstract

Okuno Y, Korchi AM, Shinjo T, Kato S, Kaneko T. Midterm Clinical Outcomes and MR Imaging Changes after Transcatheter Arterial Embolization as a Treatment for Mild to Moderate Radiographic Knee Osteoarthritis Resistant to Conservative Treatment. J Vasc Interv Radiol. 2017;28(7):995-1002.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
Venous Health Program Faculty
University of Michigan Health System
Michigan Medicine

James X. Chen, MD
Resident in Radiology
Hospital of the University of Pennsylvania

From the SIR Residents and Fellows Sections (SIRRFS)

Teaching Topic: Randomized Controlled Trial of Octyl Cyanoacrylate Skin Adhesive versus Subcuticular Suture for Skin Closure after Implantable Venous Port Placement

Martin JG, Hollenbeck ST, Janas G, Makar RA, Pabon-Ramos WM, Suhocki PV, Miller MJ, Sopko DR, Smith TP, Kim CY. J Vasc Interv Radiol. 2017 Jan;28(1):111-116.

Click here for abstract

Implanting central venous port catheters has become an everyday occurrence for most interventional radiologists. In 1992, interventional radiologists implanted slightly less than 13 thousand central venous port catheters per Medicare claims data. By 2011, this number had exploded to over 83 thousand. [1] Likewise, the proportion placed by radiologists compared to other specialists increased from 17% to 27% in the same period. While port placement has become an essential tool in every interventional radiologist’s armamentarium, it remains one of the few procedures that require large incisions and full thickness skin closure. Traditionally, skin closure is performed with a deep dermal suture layer and a superficial subcuticular suture layer. The use of tissue adhesives made from octyl cyanoacrylate, such as Sure+Close II (Chemence Medical) and Dermabond (Ethicon), have been studied extensively in other specialties’ literature as an alternative for subcuticular suture closure. In this study, Martin et al. at Duke University conducted a randomized controlled trial comparing tissue adhesives and subcuticular sutures in port placement.

The authors randomly assigned 109 subjects after initial port placement and deep dermal suture closure to either receive subcuticular suture closure or adhesive closure. Subjects were followed up for infection, dehiscence, and photographs were obtained at 3-months post-procedure. Photographs were then examined by a blinded plastic surgeon and rated for cosmetic appearance on a validated 10-point scale. Additionally, each closure was timed by handheld stopwatch. There was no significant difference between groups in infection and dehiscence rates or cosmetic scores. However, using tissue adhesive was much faster taking only 1.4 minutes compared to the 8.6 minutes suturing required

Clinical Pearls

What is the purpose of superficial skin closure?

It is important to keep in mind that the ideal superficial skin closure is intended to appose epidermal edges with a slight eversion to prevent unsightly scar depression. The deep dermal stitches are used to minimize dead-space and should bear the nearly all the tension required for closure. The superficial tissues lack the strength to hold the incision closed and tension on the superficial layer causes scar widening. This is a particularly important consideration in port placement since the bulk of the port increases the tension across the incision.

How do these results compare to the use of skin adhesives in other procedures?

Skin adhesives have not shown statistically significant differences in cosmetic outcome in traumatic wound and surgical incision closure in many different procedures on various body parts. Perhaps the most relevant of these studies comes from the Plastic Surgery literature, where Nahas et al. compared the two methods side to side in mammoplasty and along the same incision in abdominoplasty. They found no significant cosmetic difference in 3, 6, and 12 month follow up. [2] This study design helps control for patient factors by using side to side or same wound comparison. Abdominoplasty incisions are usually closed under large amounts of tension and usually require various tension reducing techniques like quilting sutures and patient positioning to reduce the risk of dehiscence. Mammoplasty closure bears some similarities to port placement closure in that they are both on the chest and over and implanted foreign bodies. However, port placement is unique in that the implant is very superficial and the patients are theoretically at higher risk for poor wound healing secondary to the diseases which the port is intended to be used to treat.

Questions to Consider

How does radiation therapy affect your decision for skin closure?

Radiation can cause skin atrophy, fibrosis, ulceration, and vessel rupture. In radiated skin, all phases of the wound healing process are disrupted to some degree. [3] Fibroblasts and keratinocytes have have reduced production of crucial growth factors as well as disorganized collagen production. This study excluded patients with prior radiation therapy to the chest or plans for radiation. There is no RCT specifically comparing the different superficial closure techniques in radiated skin, but the safe use of tissue glues has been described in breast reconstruction after radiation. [4] It is likely that radiation increases the complication risks in both tissue glue and subcuticular suture closure.

What is the role of antibiotics in port placement?

The authors of this study prophylactically administered intravenous antibiotics (cefazolin 1g or clindamycin 600 mg) and irrigated the port pocket with a saline solution containing cefazolin or clindamycin. The use of antibiotic prophylaxis in implanted central venous access ports is somewhat controversial. While many retrospective studies have found decreased infection rates, the most recent Cochrane systematic review found that prophylactic antibiotics did not reduce Gram positive infection rates in RCT’s of long-term central venous catheters in oncology patients. However, flushing and locking the catheter with a combined antibiotic and heparin solution reduced the risk of Gram positive catheter-related sepsis. [5] The 2011 Guidelines for the Prevention of Intravascular Catheter-related Infections from the CDC recommend against prophylactic antibiotics. [6]

Additional Sources

1. Duszak R Jr, Bilal N, Picus D, Hughes DR, Xu BJ. Central venous access: evolving roles of radiology and other specialties nationally over two decades. J Am Coll Radiol. 2013 Aug;10(8):603-12. doi: 10.1016/j.jacr.2013.02.002. Epub 2013 Jun 14. PMID: 23770064.

2. Nahas FX, Solia D, Ferreira LM, Novo NF. The use of tissue adhesive for skin closure in body contouring surgery. Aesthetic Plast Surg. 2004 May-Jun;28(3):165-9. Epub 2004 Jul 30. PMID: 15383885.

3. Haubner F, Ohmann E, Pohl F, Strutz J, Gassner HG. Wound healing after radiation therapy: review of the literature. Radiat Oncol. 2012 Sep 24;7:162. doi: 10.1186/1748-717X-7-162.

4. Colwell AS. Current strategies with 1-stage prosthetic breast reconstruction. Gland Surgery. 2015;4(2):111-115. doi:10.3978/j.issn.2227-684X.2015.02.05.

5. van de Wetering MD, van Woensel JB, Lawrie TA. Prophylactic antibiotics for preventing Gram positive infections associated with long-term central venous catheters in oncology patients. Cochrane Database Syst Rev. 2013 Nov 25;(11):CD003295. doi: 10.1002/14651858.CD003295.pub3. PubMed PMID:24277633.

6. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for the Prevention of Intravascular Catheter-related Infections. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America. 2011;52(9):e162-e193. doi:10.1093/cid/cir257.

Post Author:
Charles Hyman, MS4
Warren Alpert Medical School of Brown University

Segmental Y90 vs Segmental TACE for Localized HCC

Summary

Embolotherapies play an important role in the management of hepatocellular carcinoma (HCC), but the optimal modality remains controversial. Transarterial chemoembolization (TACE) remains the most widely accepted option for HCC, but some studies have suggested that transarterial radioembolization (TARE) may have superior outcomes. Padia et al present a single center retrospective series comparing segmental TARE versus TACE for HCC. Due to inherent biases in the use of these modalities over time (i.e. preferential use of TACE to bridge for liver transplantation), there were baseline differences in the patient population including higher ECOG performance scores in the TACE group (p=0.003), larger tumor size in the TARE group (mean diameter 3.2 versus 2.6 cm, p<0.001), higher volume of tumor infiltration in the TARE group (23% versus 9%, p=0.01), and greater incidence vascular invasion in the TARE group (18% versus 1%, p<0.001). Recognizing the potential for bias from these baseline differences, the authors used a propensity score adjustment approach for the analyses. TARE was performed with glass microspheres (Therasphere) in all cases, with target doses of >200 Gy for ablative intent (median administered dose: 1.59 GBq). TACE was performed using drug eluting beads (100-300 um LC beads loaded with doxorubicin) or conventional ethiodized oil, doxorubicin emulsion following by 300-500 um microspheres. Tumor responses were assessed in the index lesion as well as whole liver per RECIST 1.1. TARE demonstrated significantly higher complete response rate, both by index lesion (TARE 92% versus TACE 74%, p=0.001) and overall liver (TARE 84% versus TACE 58%, p<0.001), with statistical significance in both unadjusted and propensity-score adjusted analyses. Survival analysis from the time of first embolotherapy demonstrated significantly longer median progression-free survival (PFS) in the TARE group (564 days) compared to TACE (271 days), p=0.02, with significance maintained on inverse probability of treatment weighting (IPTW) propensity score adjustment. Overall survival (OS) was; however, not significantly different between the two groups. At 1 year, the local tumor recurrent rate was significantly lower following TARE (8%) compared to TACE (30%). A significantly higher proportion of patients receiving TARE achieved optimal response after 1 treatment compared to TACE (93% versus 79%, p=0.012). Both modalities had low rates of clinical and biochemical toxicities (Common Terminology Criteria for Adverse Events grade 3 or higher), with no significant difference in the overall rate of complications, although there was a higher rate of pain in the TARE group (7.6% versus 1%), and higher rate of post-embolization syndrome in the TACE group (8.8% versus 2.3%).



Commentary

In HCC, extensive experience with TACE has cemented the modality as the mainstay choice for many providers and institutions, but there is a growing body of evidence supporting TARE. Comparisons between embolotherapy modalities are inherently challenging due to variations in patient selection and technical approach, particularly in retrospective studies, which contribute bias and limit generalizability of results. To circumvent some of these challenges, Padia et al focused on only cases of segmental TARE and TACE and applied propensity score adjustment to their outcomes analysis. The findings of improved index lesion and whole liver complete response rates and longer median PFS following segmental TARE using an “ablative” dose suggest that this approach may be superior to TACE. These results were echoed by findings of the PREMIERE trial (a prospectively randomized controlled trial comparing TARE [lobar and selective] with conventional TACE, Salem et al. Gastroenterology 2016) that also demonstrated significantly longer median time to progression in the TARE group, although no difference in imaging response rate was shown. In this study, additional benefits of TARE included lower 1-year local recurrence rates, which were similar to rates reported following thermal ablation or surgery, and ability to achieve optimal response after 1 treatment in a higher percentage of patients compared to TACE. There are several limitations to consider when evaluating these results, in addition to the inherent biases of retrospective approach. Although the authors used a sound statistical approach to reduce the impact of baseline differences in treatment populations, there may still be residual bias in the results. Furthermore, it is never stated whether the patients in the study were embolotherapy naïve, or the degree of treatment crossover that occurred following the first treatment, which could confound the survival analysis. Nonetheless, this study provides convincing support for the use of segmental TARE over TACE for HCC to achieve better local tumor control. Many institutions preferentially use TACE as a bridge to liver transplantation, and results like these suggest that TARE could potentially be the superior modality for reducing the rate of waitlist drop-out.

Click here for abstract

Padia SA, Johnson GE, Horton KJ, et al. Segmental Yttrium-90 Radioembolization versus Segmental Chemoembolization for Localized Hepatocellular Carcinoma: Results of a Single-Center, Retrospective, Propensity Score-Matched Study. J Vasc Interv Radiol 2017; 28:777-85 e1.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
University of Michigan Health Systems

James X. Chen, MD
Resident in Radiology
Hospital of the University of Pennsylvania

From the SIR Residents and Fellows Section

Teaching Topic: Influence of different flush methods on transfemoral cerebral angiography 

Lee HJ, Yang PS, Lee SB, et al. The influence of flush methods on transfemoral catheter cerebral angiography: continuous flush versus intermittent flush. J Vasc Interv Radiol. 2016; 27:651-657.

Click here for abstract

This manuscript compares different flush methods on transfemoral cerebral angiography (TFCA) in a single-blind randomized trial. Fifty patients were allocated to intermittent-flush (n = 25) and continuous-flush (n = 25). Researchers evaluated differences in procedure duration, amounts of contrast and heparinized saline used, , heparin dose, blood loss, fluoro time, radiation dose, and development of new embolic signal (NES) on diffusion-weighted imaging (DWI). There were noted differences in procedure duration, amount of contrast used, wasted heparinized saline, and aspirated blood. However, there were no differences in the occurrence of NES on DWI between the treatment groups.

Clinical Pearls

What is a new embolic signal (NES)?

NES identifies an area of brain hyperacutely affected by microembolism during a neurovascular, angiographic procedure. Diffusion-weighted imaging (DWI) is the gold standard to confirm these lesions. A new, diffusion-prolonged, foci can be considered a procedure-related embolic signal. Alternatively, transcranial Doppler (TCD) can be performed at the bedside and is easily repeated. Flushing with heparinized saline during TFCA is mandatory for protecting against thromboembolic complications.

What are the most common complications of cerebral angiography?

Access-site hematoma is the most common complication overall (4.2%), neurologic complications are seen in 2.63% with 0.14% being strokes with permanent disability. Factors associated with increased risk of neurologic complication include the indication of atherosclerotic cerebrovascular disease, indication of subarachnoid hemorrhage, and the comorbidity of frequent TIAs. Conversely, involvement of a trainee in the cerebral angiogram decreased the risk of complication1.



Figure

Preparation of the diagnostic catheter in each group. (a) The diagnostic catheter was connected to a 10-mL syringe filled with heparinized saline via a one-way connector in the intermittent-flush group. (b) The diagnostic catheter was connected to a Y connector in the continuous-flush group. The side arm of the Y connector was connected to the pressurized flushing line (A) and a connecting line (B) via a three-way connector. Another three-way connector was connected to the connecting line of the mechanical power injector (C) and a syringe for manual injection (D).

Questions to Consider

What types of flushing methods can be used?

Heparinized saline (5,000 U/L) was used for the flushes during TFCA in this study. A conventional continuous flushing system through a vascular sheath is formed by connecting the sheath to a plastic bag of heparinized saline surrounded by a pressure cuff inflated to 300 mmHg. A reducer permits a rate of 1 drop/sec into the sheath and catheter system. Some operators believe that the use of a continuous flushing method reduces the possibility of air embolism compared to intermittent flushing during the procedure, which requires blood aspiration into the flush syringe to ensure air bubbles are removed from the catheter prior to injection. It is generally accepted that intermittent flushing be performed whenever wires and catheters are removed and exchanged during the procedure.

How may the flush method affect the procedure?

Depending on the type of flush method used, procedure duration, amounts of contrast medium and heparinized saline used, heparin dose, blood loss, fluoroscopy time, radiation dose, and occurrence of new embolic signal (NES) on diffusion-weighted imaging (DWI) may differ and were monitored in this study. The authors found that although it is time-consuming to set up the more complex continuous-flush system, total procedure time in the continuous-flush group was significantly shorter than the total procedure time in the intermittent-flush group. Three NESs on DWI occurred in three of the 27 patients who underwent DWI in this study (11%). All lesions were asymptomatic and occurrence was lower than the previously reported prevalence (15%–26%). The amount of heparinized saline wasted, contrast used, contrast wasted, and blood aspirated were also significantly lower in the continuous-flush group.

Additional Citations:

Kaufmann TJ, Huston J, Mandrekar JN, Scleck CD, Thielen KR, Kallmes DF. Complications of diagnostic cerebral angiography: evaluation of 19,826 consecutive patients. Radiology. 2007; 243: 812-9.

Post Author:
Rajat Chand, MD
Diagnostic Radiology Resident, PGY-2
John H. Stroger Hospital of Cook County

Predictors of Recanalization for Incompetent Great Saphenous Veins Treated with Cyanoacrylate Glue

Summary

In recent years, several promising non-thermal endovenous ablation techniques have been introduced which do not require tumescent anesthesia. These include mechanochemical ablation (ClariVein; Vascular Insights, Quincy, Massachusetts), VariClose (Biolas, Ankara, Turkey), polidocanol endovenous microfoam (Varithena; BTG International, London, United Kingdom), and VenaSeal Closure System (Medtronic, Gorway, United Kingdom). Long-term outcomes following ablation with these non-thermal techniques remains limited compared to the mature evidence base underlying endovenous thermal ablation. This study from Chan et al is a retrospective series of 108 legs in 55 consecutive patients with saphenofemoral junction (SFJ) or great saphenous vein (GSV) incompetence which aimed to identify predictors of great saphenous vein (GSV) recanalization following treatment with VenaSeal. The included patients had Clinical-Etiology-Anatomy-Pathophysiology (CEAP) 3 (60%) or higher disease. Patients underwent VenaSeal with concurrent microphlebectomy and were followed with clinical evaluation and duplex ultrasound at 1-week, 1-month, 6-months, and 24-months post-procedure. VenaSeal was deposited 4 cm distal to the SFJ and at 3-cm intervals along the entire course of the targeted GSV (median treatment length: 28 cm). On follow-up, GSV closure rates were 97% (1 week), 94% (1 month), 89% (6 months), and 76% (12 months). The majority of recanalization occurred at one year and developed from the SFJ rather than segmentally within ablated areas. Cox regression for predictors of GSV recanalization found only GSV diameter ³6.6 mm as a significant predictor (p=0.016). Other evaluated variables, which were not found to be significant, included GSV length, presence of incompetent thigh perforators, clinical severity at presentation, and operator experience. The study also observed low pain and mild ecchymosis following the procedure as well as significant improvements in validated vein symptom scores post-procedure.



Figure 2B. Kaplan-Meier curve showing significantly lower closure rates in GSV ³ 6.6 mm (log-rank test, p=0.002)

Commentary

The treatment of saphenous vein insufficiency has been revolutionized in the last two decades with the development of endovenous ablation techniques using radiofrequency ablation (RFA) and endovenous laser ablation (EVLA) which offer high rates of technical success at far less morbidity than traditional surgical vein stripping. These techniques; however, do have some disadvantages including the requirement of tumescent anesthesia and ecchymosis and pain during the recovery period. Several new non-thermal endovenous ablation techniques have been developed to address these (and other) limitations of thermal ablation techniques and include cyanoacrylate based compounds (VenaSeal and Variclose), polidocanol foam (Varithena), and mechanochemical ablation (ClariVein). Similar to prior studies, this series demonstrated that VenaSeal has very low associated post-procedural pain which is a benefit over thermal ablation techniques. The prior VeCLose trial for Venaseal demonstrated 99% GSV closure rate at 3-months, but longer term follow-up with VenaSeal is limited. This study demonstrated an increasing rate of GSV recanalization at 12-months follow-up (24% patients from 11% patients at 6 months), highlighting the importance of long-term outcomes data. Notably, the study had substantial drop-off in number of patients who completed duplex follow-up at the later time points, with only 58% patients at 6 months, and 34% patients at 12 months. The authors suggested that the American patient population in the VeClose trial had smaller mean GSV diameters than those seen in the Asian population, but the median GSV diameter in the current study was 6.6 mm, similar to the 6.3 mm mean proximal GSV diameter in VeClose. The Cox regression finding of GSV diameter ³6.6 mm as a significant predictor for GSV recanalization provides useful data for counseling patients who are considering VenaSeal. Prior studies on endovenous thermal ablation techniques have also found larger GSV diameter as significant predictor for recanalization. Insurance reimbursement for VenaSeal remains challenging in many geographic areas and patients may be paying out-of-pocket so management of expectations and advising alternative modalities may be appropriate in patients with larger GSV diameters. The authors mention another series describing the use of larger VenaSeal volumes to achieve durable results in larger caliber GSVs, but definitive long-term outcomes data are lacking. Lastly, all patients in this study wore full-length compression stockings for at least one month post-procedure which is not a universal protocol (one of the marketed benefits of VenaSeal is the lack of need for post-procedure compression stockings) and may therefore affect the generalizability of the results. Non-thermal GSV closure techniques remain an exciting and growing area in the treatment of GSV reflux and more studies like this one are needed to elucidate the optimal technique and algorithm with which to apply these novel modalities.

Click here for abstract

Chan YC, Law Y, Cheung GC, Cheng SW. Predictors of Recanalization for Incompetent Great Saphenous Veins Treated with Cyanoacrylate Glue. J Vasc Interv Radiol 2017; 28:665-71.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
Venous Health Program Faculty
University of Michigan Health System
Michigan Medicine

James X. Chen, MD
Resident in Radiology
Hospital of the University of Pennsylvania

From the SIR Residents and Fellows Section

Teaching Topic: Percutaneous Image-Guided Irreversible Electroporation for the Treatment of Unresectable, Locally Advanced Pancreatic Adenocarcinoma

Narayanan G, Hosein PJ, Beulaygue IC, et al. Percutaneous image-guided irreversible electroporation for the treatment of unresectable, locally advanced pancreatic adenocarcinoma. J Vasc Interv Radiol. 2017; 28: 342-48.

Click here for abstract

Pancreatic adenocarcinoma is associated with five-year survival less than 5%. It is expected to surpass breast cancer to become the third leading cause of cancer-related deaths in the US. Narayanan et al. conducted a retrospective study of 50 patients with unresectable locally advanced pancreatic cancer (LAPC) to identify the treatment safety and efficacy of percutaneous irreversible electroporation (IRE). All patients received chemotherapy before IRE. Post-IRE, patients received follow-up contrast-enhanced CT at 1- and 3-month intervals. Repeat IRE was performed in 9 (18%) patients with unequivocal residual disease on follow-up CT. Three (6%) patients received surgical resection after IRE due to tumor downstaging. There were no treatment-related deaths or deaths within 30 days of treatment. Complications included abdominal pain, pancreatitis, sepsis, gastric leak, and non-fatal portal and splenic vein thrombosis. In univariate and multivariate analyses, tumor size <3 cm was the only factor associated with prolonged survival. The separation of survival curves between the small (<3 cm) and large (>3 cm) tumor groups near 12 months after diagnosis (correlating with the median time from diagnosis to IRE of 11.6 months) supported small tumor response to treatment. Study limitations include difficulty identifying residual disease due to the similar hypoattenuating appearances of tumor and ablation zone.

Clinical Pearls

What is the mechanism of action of IRE?

IRE is a nonthermal ablative technique that utilizes targeted high-voltage electrical pulses to create holes in the cell membrane, thereby irreversibly damaging cell homeostasis and inducing apoptosis. Pulses can be delivered through a bipolar electrode or a pair of unipolar electrodes. The ablation zone size is influenced by the length of the active tip, pulse number, duration of pulses, distance between probes, and voltage.

What are the advantages and disadvantages of IRE?

Given the technique does not depend on heating or cooling tissue, IRE is well-suited for treating tumors close to critical organs and vascular structures with less risk of thermal injury. It is also helpful for preserving sensitive structures such as nerves and bile duct. The IRE ablation zone can be difficult to predict, because it varies by tissue composition and electrical characteristics of tumor and surrounding tissue. Zones are also altered by conductivity of the local environment (e.g. the presence of metal biliary stent).

Questions to Consider

What are common complications following pancreatectomy? What is the role of the interventional radiologist in the management of these complications?

· Delayed gastric emptying (19-23%)

· Anastomotic leak or leak of bile or pancreatic enzymes (29-34%): Percutaneous catheter drainage placement can be guided by CT, or a combination of ultrasound and fluoroscopy. Catheter drainage with or without endoscopic intervention can avoid re-exploration in 94.7% of patients with pancreatic leaks. Bile leakage is defined as fluid from catheter drainage or abdominal collection with elevated bilirubin level three times greater than the serum bilirubin value.

· Intra-abdominal abscesses (9-13%): While most leaks resolve without intervention, abscesses can form if fluid collections are colonized by bowel contents or superinfected. Large-bore (up to 24F) catheters may be needed to drain purulent or viscous contents.

· Post-pancreatectomy hemorrhage (PPH) (1-8%): PPH can occur when pancreatic enzymes and bile erode into vascular structures. This complication is associated with high mortality, causing up to 38% of all post-pancreatectomy deaths. PPH is often managed with arterial embolization with rates of definitive therapy of 77-88%. Bleeds can be treated by placing a stent over the origin of the GDA for GDA stump leaks or positioning a covered stent over the extravasating site.

How do the results by Narayanan et al. compare to previous reports of open pancreatic IRE?

The largest series was reported by Martin et al. and was composed of 150 patients in the unresectable group. This group showed a median OS of 23.2 months from time of diagnosis and 19 months from time of procedure. The present study shows a median OS of 27 months from time of diagnosis and 14.2 months from time of IRE. While they show seemingly different results, in the series by Martin et al. the median time from diagnosis to IRE was 6.2 months (compared to 11.6 months from Narayanan et al.). In addition, the group from Martin et al. has diagnostic laparoscopy to exclude occult peritoneal mets before proceeding. While there are limitations in the present study and it is difficult to establish if IRE improves survival, the results are promising and warrant a prospective, randomized clinical trial.

Additional references:

Silk M, Tahour D, Srimathveeravalli G. The state of irreversible electroporation in interventional oncology. Sem Interv Radiol. 2014; 31: 111-17.

Martin RC, Kwon D, Chalikonda S, et al. Treatment of 200 locally advanced (stage III) pancreatic adenocarcinoma patients with irreversible electroporation: safety and efficacy. Ann Surg 2015; 262:486–494.

Post Author:
Maggie Chung, BA
SIR RFS Communications Co-Chair
Warren Alpert Medical School

Radiation Exposure of Patients and Interventional Radiologists during Prostatic Artery Embolization: A Prospective Single-Operator Study

Summary

Prostatic artery embolization (PAE) is a technically challenging, often lengthy procedure due to the complex, variable anatomy of the prostatic arteries, which may require multiple views, magnification, and cone-beam computed tomography (CT), all of which contribute to elevated radiation doses to the patient and interventional radiologist. Existing studies on PAE have reported indirect measures of radiation exposure including fluoroscopy times and dose area product (DAP), but none have evaluated direct radiation measures, which are a more accurate assessment of radiation dose. Andrade et al conducted a prospective evaluation of radiation exposure to the patient and practitioner during 25 PAE procedures. Patient peak skin dose (PSD) was measured with radiochromic film, placed under the patient’s hip. Operator radiation dose were attained from 9 pairs of thermoluminescent dosimeters, positioned at various body positions on both sides. All procedures were performed by a single interventional radiologist, whose experience with PAE included completion of a PAE course, assisting with 5 cases, and independent performance of 5 cases. Procedures were performed using an Artis Zee ceiling-mounted angiography system equipped with a flat-panel detector. Ceiling-suspended screen and table curtain were always employed. Fluoroscopy was performed at 15 images/second, digital subtraction angiography performed at 2 images per second. Cone-beam CT was performed only if deemed necessary. Mean patient weight was 71.4 kg (range: 54-88 kg), and prostate volume was 79 cm3 (range: 36-157 cm3). Embolization was performed with 100-200 um polyvinyl alcohol particles or 400-um microspheres to complete prostate artery occlusion. Average fluoroscopy time was 30.9 minutes (range: 15.5-48.3 minutes), with mean total DAP of 451 Gy-cm2 (range: 248-792 Gy-cm2), 75% of which was from fluoroscopy. Mean patient PSD was 2420 mGy (range: 1390-3616 mGy), which is in the range of other complex interventional radiology procedures like transjugular intrahepatic portosystemic shunt, transarterial chemoembolization, and neural embolization. Average effective dose to the practitioner was 17 uSv (range : 4-47 uSv), with higher doses to the left side of the body, including average 0.378 mSv to the left eye. Post-PAE clinical evaluation of patients, including skin check to the lower back and hip, at 15 days, 1 month, and 3 months demonstrated no sequelae of radiation exposure. 

Commentary
PAE is known to be a challenging procedure that may require lengthy fluoroscopy times and high radiation exposure, even in the hands of experienced practitioners. As more interventional radiologists undertake this complex procedure, it is essential to elucidate the degree of associated radiation exposure to patients and operators. In prior PAE series, the degree of radiation exposure has been reported through indirect measures, which may be inaccurate surrogates for direct measures of radiation dose. The direct measures of radiation dose assessed by Andrade et al therefore provide a unique and important addition to the PAE evidence base. For operators, the PAE exposure levels of average 17 uSv were similar to other complex interventional radiology procedures, highlighting the importance of optimizing radiation protection techniques and equipment when performing PAE. Although no patients developed skin radiation injury in this series, the mean patient PSD of >2 Gy with range up to 3.6 Gy, in combination with prior case reports of radiation dermatitis following PAE, suggest that vigilant clinical follow-up is merited. These high radiation doses are particularly relevant considerations for patients who undergo repeat PAE for whom attempts should be made to distribute the dose to a different skin region. Although this study was subject to biases of being a small, single center series with all cases performed by a single interventional radiologist, these results provide an important foundation to our understanding of radiation exposure during PAE. Reduction of these radiation exposure levels will be important aims for further refinement and maturation of PAE for the treatment of benign prostatic hyperplasia.

Click here for abstract

Andrade G, Khoury HJ, Garzón WJ, Dubourcq F, Bredow MF, Monsignore LM, Abud DG. Radiation Exposure of Patients and Interventional Radiologists during Prostatic Artery Embolization: A Prospective Single-Operator Study. J Vasc Interv Radiol 2017; 28:517-21.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
University of Michigan Health Systems
Michigan Medicine

James X. Chen, MD
Resident in Radiology
Hospital of the University of Pennsylvania

From the SIR Residents and Fellows Section

Teaching Topic: Determining efficacy of RF ablation for primary or metastatic lung cancer based on tumor characteristics and environment.

Qiuxia Yang, MD, Han Qi, MD, Rong Zhang, MD, Chao Wan, MD, Ze Song, MD, Liang Zhang, MD, and Weijun Fan, MD. Risk Factors for Local Progression after Percutaneous Radiofrequency Ablation of Lung Tumors: Evaluation Based on a Review of 147 Tumors. 2017. 28 (4): 481-9.

Click here for abstract

Tumor-related factors that influence local tumor progression (LTP) include, size, site, orientation, organ, histology, and biology. The decision to treat primary or metastatic lung cancer with surgery or minimally invasive RF ablation is generally based on size. Larger tumors > 3 cm are considered to be more amenable to surgery. For smaller tumors or those close to 3 cm, the authors retrospectively reviewed factors influencing the complete ablative rate (CAR) and LTP. 147 tumors with an average maximum diameter of 1.8 cm +/- 1.2 receiving a single ablation and followed > 6 months were evaluated in 93 patients. 26% of patients had primary lung cancer and 74% had metastatic disease.

Clinical Pearls

What is the mechanism of action of radiofrequency ablation?

The Bovie knife is the first well-known medical device to incorporate RF ablation, which functions as cautery (pulsed current) and cutting (continuous current). Radiofrequency refers to the 3 Hz – 300 GHz range of the electromagnetic spectrum, which can cause thermal ablation of tissue. As with the Bovie knife, RF ablation devices operate in a closed electrical circuit, with the cathode (RF electrode or probe) transmitting energy through the patient’s tissue toward an anode (dispersing pads). Coagulative necrosis is achieved as dipole molecules (mostly water) next to the electrode remain aligned to the direction of the current and are forced to vibrate as rapidly as the applied current. Frictional energy released by adjacent molecules is deposited in surrounding tissue, resulting in local temperature increase. As the energy is dispersed by the pads, tissue damage is limited to the area surrounding the electrode tip. Application of too high of a generator power too quickly can cause tissue to desiccate or “char,” which acts as insulation and limits further extension of ablation.

What is meant by the “heat sink” effect?

Induced coagulation necrosis = (energy deposited x local tissue interactions) – heat loss. “Heat sink” is a cooling effect, which limits the effect of all thermal ablation methods. Flowing blood in vessels 3 mm or larger adjacent to a target lesion limit temperature variation in that area. This is potential cause for residual, unablated tissue and local tumor progression. Other studies have also postulated that tumors with proximal vascular or bronchial extension might lead to larger microscopic extension beyond the edge of the tumor. In this study, contact between tumor and blood vessels was determined as an independent risk factor for incomplete ablation in this study, and the CAR was significantly reduced.

What is considered effective tumor ablation?

A slow method of energy deposition, as to avoid desiccation with quick temperature rises, is used to heat tissue to 50-100 degrees Celsius for 4-6 minutes. Data extrapolation from surgery has established a general guideline to achieve an ablation margin 0.5 – 1.0 cm of ablated normal tissue, which is thought to account for microscopic tumor extension beyond the visualized confines of the lesion. The authors determined a complete ablative margin in this study to correspond to ground-glass opacities completely encircling the ablated lesion. In practice, it is difficult to distinguish inner necrosis areas from outer peripheral hemorrhage and inflammatory reaction on CT images and as a result, the shortest distance of overall ablative margin was measured and recorded in this study. Tumors can be monitored for local progression by determining contrast enhancement at the ablation margins on follow-up CT studies, as was done in this retrospective review. In the first three months following ablation, an enhancement zone that is peripheral, concentric, symmetric, and uniform and has smooth inner margins can be considered to correspond to reactive hyperemia, inflammation, or granulation at the marginal parenchyma. For tumors that are PET-avid prior to ablation, follow up PET studies may also be used. 

Questions to Consider

Which factors related to the efficacy of complete tumor ablation in this study?

The authors retrospectively investigated variables of tumor related factors before ablation including, tumor type, tumor size, morphology (smooth margins, lobulated and/or spiculated), and contact with blood vessels. For tumors < 3 cm, the CAR was 68.55%. For tumors > 3 cm, the CAR was 17.39%. CAR of tumors with a smooth margin was significantly greater than CAR of tumors with lobulated and/or spiculated edges. CAR of tumors with no surrounding blood vessels was 75%, which was remarkably higher than CAR of tumors with blood vessel contact. In tumors with complete ablative margin, the CAR was 74.77%, whereas it was 16.67% for tumors with incomplete ablative margin. Further, in tumors with complete ablative margins, completely ablated tumors had significantly larger ablative margin than incompletely ablated tumors (5.04 mm +/- 2.29 vs 3.71 mm +/- 2.51). Multivariate analysis showed that incomplete ablative margin and an ablative margin of 1–4 mm were independent risk factors for incomplete RF ablation of lung tumor.

Which factors related to local tumor progression (LTP)?

LTP after RF ablation is not rare, and progression at the RF ablation site is associated with poor overall survival in published reports. In this study, the LTP rate was 52% for primary lung cancers and 36.9% for lung metastases. This was higher than rates reported in the previous studies mentioned in the article, which were calculated for a period of several years and included the condition that patients received repeated lung tumor RF ablations. Of the 58 tumors with incomplete ablation, 56 tumors developed local progression at the edge of ablated lesions. 52 tumors locally recurred at the site of incomplete or shortest ablative margin. The site of shortest ablative margin was usually located at the tumor edge of contact with the blood vessels, in front of the ablation electrode, with the maximum radius perpendicular to the ablation electrode, and with protrusion of lobulated and/ or spiculated area of tumors.

Additional Citations:
Hong K, Georgiades C. Radiofrequency Ablation: Mechanism of Action and Devices. J Vasc Interv Radiol. 2010. 21(8): S179-S186

Post Author:
Rajat Chand, MD
Diagnostic Radiology Resident, PGY-2
John H. Stroger Hospital of Cook County

Outcome Results on Ablation versus Surgery for HCC: A Report from the SEER Registry

Summary

The comparative efficacy of percutaneous ablation versus surgical resection for hepatocellular carcinoma (HCC) remains a controversial topic, with conflicting literature reporting equivalent outcomes between modalities or superior outcomes with surgery. This study by Mironov et al used the Surveillance, Epidemiology, and End Results (SEER) database to compare survival outcomes for small solitary HCCs treated with thermal ablation versus surgical resection. Only cases with available Ishak fibrosis score were included in the analysis to account for the effect of cirrhosis. To reduce confounding, patients with metastatic disease, treatment other than ablation or surgery, both surgery and ablation, and liver transplantation were excluded. There were baseline differences in the ablation and surgery patients including a higher prevalence of fibrosis in the ablation group (higher Ishak scores) and smaller tumors in the ablation group (mean 2.6 cm versus 3.0 cm, p<0.001). For tumors ≤2 cm (ablation = 264; resection = 79) and tumors between 2.1 and 4 cm (ablation = 335; resection = 209), there was no significant difference in observed or disease-specific survival between ablation and surgical resection. For tumors between 4.1 and 5 cm (ablation = 46; resection = 66), there was a significantly longer observed and disease-specific survival for surgical resection when stratified by presence of fibrosis (observed survival p=0.009, disease specific survival p=0.046). The 5-year observed survival was 72% (surgery) versus 29% (ablation) and disease-specific survival was 80% (surgery) versus 40% (ablation). Notably, the difference in disease-specific survival was not clinically significant by Cox regression with fibrosis covariate (p=0.145). When all tumors ≤4 cm were pooled, there was again no difference in survival outcomes between ablation and surgical resection. Significant predictive factors for observed and disease-specific survival by Cox model included tumor size and degree of fibrosis.

Commentary

Percutaneous ablation is maturing as an important part of the treatment armamentarium for HCC. The relative efficacy of ablation techniques compared to surgery remains controversial and current guidelines from the Barcelona Clinic Liver Cancer (BCLC) recommend ablation only for patients who are not surgical candidates. These recommendations are derived from a very limited evidence base, with only one prospective study demonstrating superior outcomes after surgery. The SEER registry offers a powerful resource to answer these questions, affording a large patient sample from diverse medical institutions and detailed survival outcome data. The authors of this paper effectively identified potential confounders including patients who had received both treatments or went on to receive a liver transplant. The results of their study demonstrate equivalent survival outcomes in tumors <4 cm, which suggests that it would be reasonable to consider ablation as an alternative to surgery in this patient population. The results in the 4-5 cm tumor group demonstrated superiority of surgery, reflecting limitations of ablation in larger tumor sizes. The SEER population included both patients who had been treated with radiofrequency and microwave ablation (and does not differentiate the two modalities), so it could not be determined in this study whether outcomes in these larger tumors may be superior with microwave. It is important to recognize limitations to the SEER data including lack of BCLC or Child-Pugh scores, performance status, or comorbidities, which may all be important contributors to survival. Nonetheless, this study serves as additional evidence that percutaneous ablation affords equivalent survival outcomes to surgery in HCC ≤4 cm and may help to further define the evolving role of ablation in the treatment of HCC patients.

Click here for abstract

Mironov O, Jaberi A, Kachura JR. Thermal Ablation versus Surgical Resection for the Treatment of Stage T1 Hepatocellular Carcinoma in the Surveillance, Epidemiology, and End Results Database Population. J Vasc Interv Radiol 2017; 28:325-33.

Post Authors:
Jeffrey Forris Beecham Chick, MD, MPH, DABR
Assistant Professor of Vascular and Interventional Radiology
Vice Quality Assurance and Safety Officer
Venous Health Program Faculty
University of Michigan Health System
Michigan Medicine

James X. Chen, MD
Resident in Radiology
Hospital of the University of Pennsylvania

From the SIR residents and fellows section

Topic: Gastric Artery Embolization Trial for the Lessening of Appetite Nonsurgically (GET LEAN): Six Month Preliminary Data 

Syed, M I, Morar K, Shaikh A, Craig P, Khan O, Patel S, Khabiri, H. Gastric Artery Embolization Trial for the Lessening of Appetite Nonsurgically (GET LEAN): Six Month Preliminary Data. J Vasc Interv Radiol. 2016. 27 (10): 1502-8.

Click here for abstract

In the October 2016 edition of JVIR, a report on the 6 month safety and efficacy results of a pilot study of left gastric artery (LGA) embolization for the treatment of morbid obesity was discussed. Four patients, three women, one man, with an average age of 41 y (range 30-54), with a mean weight of 259.3lbs, and mean BMI of 42.4 kg/m2 had their LGA embolized with 300-500-um Bead Block particles for the treatment of morbid obesity.



Weight loss was calculated as a percentage versus baseline, as well as b percentage excess body weight loss as follows, where IBW represents “Ideal Body Weight”, using the Devine formula.



Treatment with a PPI was started one week prior to embolization, and continued one month after the procedure. The procedure was performed via a femoral artery access or a left radial artery access.



No immediate complication other than nausea and mild vomiting was reported. Average body weight loss among the four patients at 6 months was 20.3 lbs. Average body weight loss as a percentage was 8.5%. Average excess body weight loss at 6 months was 17.2%. Patient 4, a diabetic patient taking only oral medications, showed improvement in hemoglobin A1c levels (7.4% to 6.3%) at 3 months, which remained at 6 months. QOL measures showed that the average physical component score improved by 9.5 (on an absolute scale of 0–100), and the average mental component score improved by 9.6 (on an absolute scale of 0–100), at 6 months.



In conclusion, preliminary data supports LGA embolization as a potentially safe procedure that warrants further investigation for weight loss in morbidly obese patients.

Clinical Pearls

What were some patient selection and clinical management steps considered for this procedure?

Patients with morbid obesity, (BMI ≥ 40 kg/m2) whose previous attempts at weight loss through diet, exercise, and behavior modification had failed were recruited for this study. These patients also declined to participate in bariatric surgery, as part of the consent process. A complete history and physical exam was performed prior to the procedure.

All patients had dietary consultations for preoperative evaluation and were followed the procedure by a dietician. Any patient with type II diabetes (n = 1) was evaluated by an endocrinologist before participation in the study. Blood glucose levels were monitored with adjustments of diabetic drugs as needed by the endocrinologist throughout the study. Any female patient of childbearing potential (n = 2) was required to use two forms of contraception during the study (oral and barrier), which was monitored by their primary care provider or gynecologist. An upper endoscopy study was performed at baseline and 3 days after the procedure in all patients.

If any patient had any abnormality on 3-day endoscopy (n = 3), upper endoscopy was repeated at 30 days. Fasting morning, plasma ghrelin, leptin, and CCK measurements, in addition to BMI and other baseline and follow-up tests or procedures, were performed at regular intervals.

What were the confounding factors and limitations of this study?

The presence of superficial symptomatic gastric ulcerations (n = 3) was a confounding variable regarding mechanism of weight loss. One known mechanism of weight loss caused by gastric ulcerations is the postprandial pain that creates a fear of food. This postprandial pain was absent in the patients who had superficial gastric ulcerations after the initial few days. 

One other mechanism of weight loss caused by gastric ulceration is appetite suppression. In the present study, all patients except for patient 2 subjectively reported appetite reduction that persisted beyond 30 days (after documented ulcer healing by endoscopy).

Another potential confounding variable was the medications used (PPI and sucralfate). There is no evidence that the short-term use of a PPI or sucralfate has any effect on weight loss. In fact, according to literature, long-term PPI use may result in weight gain. Notably, no dietary restrictions (to promote healing) were given to patients who had superficial gastric ulcerations.

Another confounding variable is that diet modification and nutritional supervision may have occurred that could have accounted for weight loss.

Limitations of this study included the small sample size and relatively short follow up period. Another limitation of the study was that diet and caloric intake records were not obtained.

Questions to Consider

What physiologic pathway is altered when LGA is embolized?

Left gastric artery (LGA) embolization may fulfill a role as a minimally invasive alternative to the current surgical treatment of gastric bypass or reduction surgery for morbidly obese patients. The LGA supplies the fundus of the stomach, where it is known that the hormone ghrelin (one of the hormones responsible for appetite) is produced. Ghrelin is a 28-amino acid hunger-stimulating peptide and hormone that is produced mainly by P/D1 cells lining the fundus of the stomach and epsilon cells of the pancreas. Ghrelin is the only known circulating orexigenic, or appetite-enhancing, hormone

What are the risks / complications of Bariatric Surgery to manage obesity?

Anastomotic leaks, bowel obstruction, deep vein thrombosis, pulmonary embolism, GI bleed, dumping syndrome, and anesthesia risks. Also reported, the 30 day mortality rate associated with BS is 0.31%, as of 2014, lower than previously reported in 2004. However, it is reported that repeat operation rate is 7%, and the overall complication rate is 17%. It is estimated that only 1% of eligible patients elect to undergo bariatric surgery.

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