Jeffrey W. McCann

G2 Inferior Vena Cava Filter: Retrievability and Safety

PURPOSE To assess the retrievability of the G2 inferior vena cava (IVC) filter and factors influencing the safety and technical success of retrieval. MATERIALS AND METHODS From October 2006 through June 2008, G2 IVC filters were placed in 140 consecutive patients who needed prophylaxis against pulmonary embolism (PE). General indications for filter placement included history of thromboembolic disease (n = 98) and high risk for PE (n = 42); specific indications included contraindication to anticoagulation (n = 120), prophylaxis in addition to anticoagulation (n = 16), and failure of anticoagulation (n = 4). Filter dwell time, technical success of filter retrieval, and complications related to placement or retrieval were retrospectively evaluated in patients who underwent filter removal. RESULTS Twenty-seven attempts at G2 filter removal were made in 26 patients (12 men; age range, 24-88 years; mean age, 55.4 y) after a mean period of 122 days (range, 11-260 d). Data were collected retrospectively with institutional review board approval. Filter removal was successful in all 27 attempts (100%). Tilting of the filter (> or =15 degrees ) occurred in five cases (18.5%), with probable filter incorporation into the right lateral wall of the IVC in one. Other complications of retrieval such as filter thrombosis, significant filter migration, filter fracture, and caval occlusion were not observed. CONCLUSIONS G2 IVC filter retrieval has a high technical success rate and a low complication rate. Technical success appears to be unaffected by the dwell time within the reported range.

Toxicities after Radioembolization with Yttrium-90 SIR-Spheres: Incidence and Contributing Risk Factors at a Single Center

PURPOSE To report the incidence of liver function test (LFT) toxicities after radioembolization with yttrium-90 ((90)Y) SIR-Spheres and review potential risk factors. MATERIALS AND METHODS Patients receiving (90)Y for radioembolization of primary or metastatic liver tumors had follow-up LFTs 29-571 days after treatment. The incidence and duration of bilirubin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) toxicities were documented using common terminology criteria. Factors that were assessed included previous intra-arterial (IA) therapy, systemic chemotherapy, low tumor-to-normal liver tissue ratio at mapping angiography, vascular stasis, and higher prescribed (90)Y doses. RESULTS There were 81 patients who underwent 122 infusions and had follow-up LFTs. Of 122 infusions, 71 (58%) were associated with toxicity. One patient died with radiation-induced liver disease. Grade 3 or greater toxicities occurred in seven (7%) patients after nine procedures. The median durations of laboratory elevations for bilirubin, AST, and ALT were 29 days, 29 days, and 20 days. Toxicity developed after 51 (71%) of 72 infusions with previous IA therapy versus 20 (40%) of 50 infusions in treatment-naïve areas (P = .0006). Absence of previous systemic therapy was associated with greater risk of toxicity versus previous chemotherapy (47% vs 66%, P = .03). Other factors were not associated with increased toxicity. CONCLUSIONS Mild hepatotoxicity developed frequently after infusion of SIR-Spheres using the body surface area method, with normalization of LFTs in most patients. Grade 3 or greater toxicities were seen in < 10% of infusions. Toxicity was strongly associated with previous IA therapy.

Moxifloxacin Prophylaxis for Chemoembolization or Embolization in Patients With Previous Biliary Interventions: A Pilot Study

OBJECTIVE Abscess formation is a common serious adverse event after intraarterial therapy for hepatic malignancy in patients with colonized bile ducts. The combination of antibiotic prophylaxis and bowel preparation has been used to prevent hepatic abscess. We describe our outcomes with moxifloxacin prophylaxis alone without bowel preparation. CONCLUSION Ten patients underwent 25 procedures and were followed for a median of 250 days. No abscesses developed. Our results suggest moxifloxacin alone may suffice for prophylaxis.

Early arterial stasis during resin-based yttrium-90 radioembolization: incidence and preliminary outcomes

OBJECTIVES This study was conducted to determine the incidence of early stasis in radioembolization using resin yttrium-90 (Y-90) microspheres, to evaluate potential contributing factors, and to review initial imaging outcomes. METHODS Patients in whom early stasis occurred were compared with those in whom complete delivery was achieved for tumour type and vascularity, tumour : normal liver ratio (T : N ratio) at technetium-99m-macroaggregated albumin (Tc-99m-MAA) angiography, previous intra-arterial therapy, and infusion site (left, right or whole liver). Tumour response was evaluated at 3 months and defined according to whether a partial response and stable disease versus progressive disease were demonstrated. RESULTS A total of 71 patients underwent 128 Y-90 infusions in which 26 (20.3%) stasis events occurred. Hypervascular and hypovascular tumours had similar rates of stasis (17.4% versus 27.8%; P = NS). The mean ± standard deviation T : N ratio was 3.03 ± 1.54 and 3.66 ± 2.79 in patients with and without stasis, respectively (P = NS). Stasis occurred in 14 of 81 (17.3%) and 12 of 47 (25.5%) infusions following previous intra-arterial therapy and in therapy-naïve territories, respectively (P = NS). Early stasis occurred in 15 of 41 (36.6%) left, 10 of 65 (15.4%) right and one of 22 (4.5%) whole liver infusions (P < 0.001). Rates of partial response and stable disease were similar in the stasis (88.3%) and non-stasis (76.0%) groups (P = NS). CONCLUSIONS Early stasis occurred in approximately 20% of infusions with similar incidences in hyper- and hypovascular tumours. Whole-liver therapy reduced the incidence of stasis. Stasis did not appear to affect initial imaging outcomes.

Side-Branch Embolization Before 90Y Radioembolization: Rate of Recanalization and New Collateral Development

OBJECTIVE The purpose of this study was to assess the rate of recanalization and collateral vessel formation after side-branch embolization during mapping angiography for planned (90)Y radioembolization. MATERIALS AND METHODS Patients who underwent side-branch embolization at mapping angiography before (90)Y administration were included. Embolized vessels included the gastroduodenal artery, right gastric artery, and accessory arteries. Four interventional radiologists reviewed follow-up angiograms to assess recanalization and new collateral formation of embolized vessels. The time to recanalization or new collateral formation was tracked within 60 days and after the final arteriographic study. Differences in outcome among patients who had and those who had not undergone previous arterial directed therapy were reviewed. RESULTS Fifty-six patients underwent side-branch embolization and follow-up arteriography; 124 treatments were performed after side-branch embolization (median, 2; range, 1-7), and the median follow-up period was 134 days (range, 7-684 days). Recanalization or new collateral vessel formation was found in 6 of 56 patients (10.7%) and in 8 of 56 patients (14.3%) 60 days after treatment or at final angiography, respectively. Embolization of 110 arteries was accomplished (42 gastroduodenal arteries, 46 right gastric arteries, and 22 accessory arteries). Two of 110 arteries (1.8%) recanalized, and four of 110 (3.6%) had new collateral vessels within 60 days. At final evaluation, 2 of 110 arteries (1.8%) had recanalized and 7 of 110 (6.4%) had new collaterals. Previous liver-directed therapy did not affect outcome (p > 0.05). No patient had symptomatic gastrointestinal ulceration. CONCLUSION In more than 89% of patients, side-branch embolization provides durable occlusion for (90)Y radioembolization without collateral development or recanalization for a bilobar cycle of therapy. Further recanalization and collateral development at longer-term follow-up are minimal.

Radiation Emission from Patients Treated with Selective Hepatic Radioembolization Using Yttrium-90 Microspheres: Are Contact Restrictions Necessary?

PURPOSE To estimate the possible radiation dose to other individuals from patients treated with yttrium-90 ((90)Y). MATERIALS AND METHODS Dosimetry data were analyzed after 143 consecutive administrations of (90)Y (124 resin, 19 glass) in 86 patients. External radiation exposure levels from patients were measured immediately after infusion. Total effective dose equivalent (TEDE) to maximally exposed individuals was calculated based on total body residence time and measured dose rate. These values were compared to Nuclear Regulatory Commission (NRC) regulations (maximum, 1 mSv) and other potential guidelines for caregivers, extensive caregivers, or pregnant contacts. RESULTS Mean administered activity for resin microspheres was 0.71 GBq ± 0.35 (range, 0.07-1.6GBq). Mean TEDE dose to the maximally exposed contact was 0.03 mSv (range, 0.0005-0.16 mSv). For glass microspheres, mean administered activity was 2.8 GBq ± 1.5 (range, 0.37-5.14 GBq). Mean TEDE dose to the maximally exposed contact was 0.06 mSv (range, 0.0023-0.23 mSv). All (90)Y treatments were within current NRC regulations for release without instructions. One, three, and one infusion were beyond potential thresholds for caregivers, extensive caregivers, or pregnant contacts, respectively. For any contact scenario, release without instruction was appropriate when administered activity was less than 3 GBq. CONCLUSIONS All patients treated with (90)Y hepatic radioembolization to a maximum administered activity of 5.14 GBq and maximum dose rate of 10 uSv/h were releasable without contact restrictions according to the NRC contact scenario. Patients who receive more than 3 GBq during infusion may require dose rate measurement if more restrictive contact scenarios are considered.

Chest Port Placement with Use of the Single-incision Insertion Technique

PURPOSE To evaluate the single-incision technique for the placement of subcutaneous chest ports. Advantages, technical success, and complications were assessed. MATERIALS AND METHODS From March 2007 through May 2008, 161 consecutive chest ports were placed with a modified single-incision technique and sonographic and fluoroscopic guidance via the right internal jugular vein (IJV; n = 130), right external jugular vein (n = 1), right subclavian vein (n = 1), or left IJV (n = 28). The primary indication was for long-term chemotherapy; all patients had malignancy. RESULTS All single-incision chest port insertions were technically successful. Ports were placed in patients 19 months to 93 years of age (mean, 56.3 y), with a mean follow-up of 203.6 device-days per patient and a total of 32,779 catheter access days. No procedure-related complications, pocket hematomas, venous thromboses, or pneumothoraces were observed. Minor delayed complications occurred in three patients. Premature catheter removal was required for two patients (1.2%; 0.006 per 100 catheter-days). One port was removed less than 30 days after implantation for infection of the pocket (0.61%; 0.003 per 100 catheter-days). Another catheter was removed because of patient dissatisfaction and unconfirmed concerns with arrhythmia (0.61%; 0.003 per 100 catheter-days). One minor superficial wound infection was successfully treated with oral antibiotics, with the port kept in place. CONCLUSIONS Use of a single-incision technique for chest port implantation in adult and pediatric oncology patients is feasible. This may be the preferred method of subcutaneous port placement, as it has a very low complication rate and a high success rate. Prospective evaluation is needed to compare it versus the conventional two-incision technique.

Pancreaticoduodenectomy: expected post-operative anatomy and complications

Pancreaticoduodenectomy is a complex, high-risk surgical procedure performed for tumours of the pancreatic head and other periampullary structures. The rate of perioperative mortality has decreased in the past number of years but perioperative morbidity remains high. This pictorial review illustrates expected findings in early and late post-operative periods, including mimickers of pathology. It aims to familiarize radiologists with the imaging appearances of common and unusual post-operative complications. These are classified into early non-vascular complications such as delayed gastric emptying, post-operative collections, pancreatic fistulae and bilomas; late non-vascular complications, for example, biliary strictures and hepatic abscesses; and vascular complications including haemorrhage and ischaemia. Options for minimally invasive image-guided management of vascular and non-vascular complications are discussed. Familiarity with normal anatomic findings is essential in order to distinguish expected post-operative change from surgical complications or recurrent disease. This review summarizes the normal and abnormal radiological findings following pancreaticoduodenectomy.

Pressure Wire Assessment of Hemodynamic Alterations after Chemoembolization of Hepatocellular Carcinoma

RATIONALE AND OBJECTIVES The end point of chemoembolization for hepatocellular carcinoma is qualitative. We intended to determine the feasibility of measuring intra-arterial pressure changes after chemoembolization and hypothesized that pressures would increase in the distal hepatic artery after the procedure. MATERIALS AND METHODS Before and after chemoembolization, systemic (S) systolic and mean pressures were measured along with celiac (C), lobar (L), and distal (D) hepatic artery pressures with a pressure wire. Corrected pressures were defined as a ratio with S as the denominator to account for intraprocedural S changes. Changes in the systolic and mean corrected pressures at each location (C/S, L/S, and D/S) were evaluated using paired t tests. Pressure changes in patients with and without tumor response using the Modified Response Evaluation Criteria in Solid Tumors were also compared. RESULTS Sixteen tumors were treated in 15 patients. One patient had bilobar tumors with separate supplying arteries. The only significant pressure change was systolic D/S (P = .02), while mean D/S approached significance (P = .08). C/S and L/S did not change significantly after chemoembolization. Eleven of 16 patients had a complete response, whereas the other five had a partial response after chemoembolization. When comparing complete to partial responders, no changes in systolic or mean C/S, L/S, or D/S reached statistical significance (all P > .05). CONCLUSIONS Measuring change in hepatic artery pressures is feasible. Distal intra-arterial corrected pressures increase significantly after chemoembolization. Further study to determine the ability to predict tumor necrosis at follow-up imaging is warranted.

Safety of Uterine Artery Embolization in Patients with Preexisting Hydrosalpinx

PURPOSE To determine (i) if preexisting hydrosalpinx poses a risk for development of pyosalpinx following uterine artery embolization (UAE) and (ii) the effect of UAE on fallopian tube diameter. MATERIALS AND METHODS Between 2005 and 2011, 429 women underwent UAE for treatment of symptomatic leiomyomas. Magnetic resonance (MR) imaging before UAE revealed 16 (3.7%) premenopausal women (median age, 47 y; range, 40-52 y) with preexisting hydrosalpinx. Bilateral UAE was performed by using 500-700-μm tris-acryl microspheres with the administration of routine prophylactic antibiotics. Pre- and post-UAE MR images were used to measure fallopian tube diameter. Fallopian tube diameters were classified as normal (1-4 mm) or mildly (< 10 mm), moderately (10-20 mm), or severely enlarged (> 20 mm). A diameter change ≥ 3 mm was considered significant. Resolution of hydrosalpinx was defined by the inability to identify fallopian tubes on post-UAE MR imaging or a normal fallopian tube diameter. Radiology and hospital records were reviewed to determine clinical outcomes. RESULTS All 16 patients underwent successful bilateral UAE. Clinical follow-up ranged from 14 to 1,531 days (median, 106 d). There was no clinical evidence of pyosalpinx after UAE. Two patients experienced minor complications unrelated to fallopian tube dilation. Post-UAE MR images were available in 13 of 16 patients (81.3%). There were no MR findings to suggest pyosalpinx after UAE. Resolution or improvement was noted in four of 15 hydrosalpinges (26.7%) for patients with follow-up imaging. CONCLUSIONS The results of the present study suggest that UAE is safe for patients with preexisting hydrosalpinx.