Fracture and distant migration of the Bard Recovery filter: a retrospective review of 363 implantations for potentially life-threatening complications.

PURPOSE: To report the occurrence of fracture of the Recovery filter and incidence of potentially life-threatening complications associated with fractured fragment migration. MATERIALS AND METHODS: A retrospective study of images obtained after placement of Recovery inferior vena cava (IVC) filters from 2003 to 2006 was conducted at a single tertiary-care center. Images were reevaluated for fracture and migration; complications related to filter fracture were investigated. Kaplan-Meier survival analysis was performed to investigate the relationship between time in situ and fracture. RESULTS: A total of 363 Recovery filters were placed; 97 were retrieved, leaving 266 filters in situ (135 patients subsequently died of other causes). The following images were evaluated: 130 chest computed tomography (CT) scans, 153 abdominal CT scans, 254 chest radiographs, 148 radiographs of the abdomen/pelvis, and 106 cavagrams. Mean imaging follow-up interval was 18.4 months (maximum, 81.3 mo). No en bloc migration occurred outside the IVC. Twenty-six limb fractures (all short limbs) were identified in 20 patients; the earliest occurred at 4.1 months. Eight fragment migrations occurred into pulmonary arteries, seven into iliac/femoral veins, one into the right ventricle, and one into the renal vein. Seven fragments were intracaval near the filter, one was extracaval, and one could not be located. Kaplan-Meier survival estimates predicted a fracture rate of 40% at 5.5 years. Of the 20 patients with filter fractures, three died of unrelated causes and 17 remain asymptomatic. CONCLUSIONS: Recovery filter fractures occurred at the short limb only, with a suggested 5.5-year fracture risk of 40%. No life-threatening events occurred in patients with filter fracture.

Effectiveness of coil embolization in angiographically detectable versus non-detectable sources of upper gastrointestinal hemorrhage.

PURPOSE: To determine whether the effectiveness of arterial embolization in patients with acute upper gastrointestinal hemorrhage is related to the visualization of contrast medium extravasation at angiography. MATERIALS AND METHODS: Transcatheter embolization was performed in 108 patients who experienced acute upper gastrointestinal hemorrhage during a 5-year period. Patient charts were retrospectively reviewed. Thirty-six patients who underwent embolization after angiography demonstrated active contrast medium extravasation from an involved artery. Seventy-two patients underwent embolization in the absence of contrast medium extravasation into a bowel lumen. Embolization technique, requirement for further blood products, need for further surgery, and 30-day mortality were recorded. RESULTS: The gastroduodenal artery (GDA) was embolized in 26 of the 36 patients (72%) with extravasation, and the left gastric artery was embolized in 10 (28%). The GDA was embolized in 64 of the 72 patients (89%) without extravasation, and the left gastric artery was embolized in 13 (18%). After embolization, 23 of the 36 patients (64%) with extravasation and 44 of the 72 (61%) without extravasation required additional blood product transfusions. Seven of the 36 patients (19%) with extravasation and 16 of the 72 (22%) without extravasation required subsequent surgery secondary to bleeding. Thirty-day hemorrhage-related mortality was 17% (six of 36 patients) in the positive extravasation group and 22% (16 of 72 patients) in the negative extravasation group. The treatment success rate was 44% (16 of 36 patients) in the positive extravasation group and 44% (32 of 72 patients) in the negative extravasation group. CONCLUSIONS: In patients with acute upper gastrointestinal hemorrhage, arterial embolization is equally effective in patients who demonstrate active contrast medium extravasation at angiography as in those who do not show contrast extravasation.

A critical evaluation of a percutaneous diagnostic and treatment strategy for chylothorax after thoracic surgery.

OBJECTIVE: Because chylothorax complicating thoracic surgery is difficult to diagnose and failure of nonoperative management necessitates further surgery, we critically evaluated an evolving percutaneous strategy for diagnosing and treating chylothorax. METHODS: After thoracic surgery, 37 patients with a clinical diagnosis of chylothorax were referred for lymphangiography for definitive diagnosis and percutaneous treatment. Successful localization of the cisterna chyli by lymphangiogram facilitated percutaneous cannulation of the thoracic duct and its embolization. In patients in whom cannulation was not possible, the thoracic duct was percutaneously disrupted. DIAGNOSIS: Lymphangiography was successful in 36 of the 37 patients (97%). Contrast extravasation, confirming clinical diagnosis, was present in 21 of the 36 (58%). MANAGEMENT: Twenty-one of 36 patients underwent 22 lymphangiographically directed percutaneous interventions: 12 embolizations and 10 disruptions. Mortality was zero, with two manageable complications. Patients without percutaneous intervention were discharged a median of 7 days (range 4-58) after first lymphangiography, 8 days (range 2-19) after percutaneous embolization, and 19 days (range 6-48) after first disruption. Eight patients had nine subsequent reoperations for chylothorax, two with negative lymphangiograms; no embolization patient required reoperation. CONCLUSIONS: There is a discrepancy between the clinical diagnosis of chylothorax after thoracic surgery and the presumed gold standard of diagnosis, contrast extravasation at lymphangiogram. Percutaneous treatment by thoracic duct embolization or disruption is safe and may obviate reoperation, but embolization of the thoracic duct is preferable to its disruption.

Radiation doses in interventional radiology procedures: the RAD-IR Study. Part III: Dosimetric performance of the interventional fluoroscopy units.

PURPOSE: To present the physics data supporting the validity of the clinical dose data from the RAD-IR study and to document the performance of dosimetry-components of these systems over time. MATERIALS AND METHODS: Sites at seven academic medical centers in the United States prospectively contributed data for each of 12 fluoroscopic units. All units were compatible with International Electrotechnical Commission (IEC) standard 60601-2-43. Comprehensive evaluations and periodic consistency checks were performed to verify the performance of each unit's dosimeter. Comprehensive evaluations compared system performance against calibrated ionization chambers under nine combinations of operating conditions. Consistency checks provided more frequent dosimetry data, with use of each unit's built-in dosimetry equipment and a standard water phantom. RESULTS: During the 3-year study, data were collected for 48 comprehensive evaluations and 581 consistency checks. For the comprehensive evaluations, the mean (95% confidence interval range) ratio of system to external measurements was 1.03 (1.00-1.05) for fluoroscopy and 0.93 (0.90-0.96) for acquisition. The expected ratio was 0.93 for both. For consistency checks, the values were 1.00 (0.98-1.02) for fluoroscopy and 1.00 (0.98-1.02) for acquisition. Each system was compared across time to its own mean value. Overall uncertainty was estimated by adding the standard deviations of the comprehensive and consistency measurements in quadrature. The authors estimate that the overall error in clinical cumulative dose measurements reported in RAD-IR is 24%. CONCLUSION: Dosimetric accuracy was well within the tolerances established by IEC standard 60601-2-43. The clinical dose data reported in the RAD-IR study are valid.

Radiation doses in interventional radiology procedures: the RAD-IR study: part II: skin dose.

PURPOSE: To determine peak skin dose (PSD), a measure of the likelihood of radiation-induced skin effects, for a variety of common interventional radiology and interventional neuroradiology procedures, and to identify procedures associated with a PSD greater than 2 Gy. MATERIALS AND METHODS: An observational study was conducted at seven academic medical centers in the United States. Sites prospectively contributed demographic and radiation dose data for subjects undergoing 21 specific procedures in a fluoroscopic suite equipped with built-in dosimetry capability. Comprehensive physics evaluations and periodic consistency checks were performed on each unit to verify the stability and consistency of the dosimeter. Seven of 12 fluoroscopic suites in the study were equipped with skin dose mapping software. RESULTS: Over a 3-year period, skin dose data were recorded for 800 instances of 21 interventional radiology procedures. Wide variation in PSD was observed for different instances of the same procedure. Some instances of each procedure we studied resulted in a PSD greater than 2 Gy, except for nephrostomy, pulmonary angiography, and inferior vena cava filter placement. Some instances of transjugular intrahepatic portosystemic shunt (TIPS) creation, renal/visceral angioplasty, and angiographic diagnosis and therapy of gastrointestinal hemorrhage produced PSDs greater than 3 Gy. Some instances of hepatic chemoembolization, other tumor embolization, and neuroembolization procedures in the head and spine produced PSDs greater than 5 Gy. In a subset of 709 instances of higher-dose procedures, there was good overall correlation between PSD and cumulative dose (r = 0.86; P <.000001) and between PSD and dose-area-product (r = 0.85, P <.000001), but there was wide variation in these relationships for individual instances. CONCLUSIONS: There are substantial variations in PSD among instances of the same procedure and among different procedure types. Most of the procedures observed may produce a PSD sufficient to cause deterministic effects in skin. It is suggested that dose data be recorded routinely for TIPS creation, angioplasty in the abdomen or pelvis, all embolization procedures, and especially for head and spine embolization procedures. Measurement or estimation of PSD is the best method for determining the likelihood of radiation-induced skin effects. Skin dose mapping is preferable to a single-point measurement of PSD.

Radiation doses in interventional radiology procedures: the RAD-IR study: part I: overall measures of dose.

PURPOSE: To determine patient radiation doses for interventional radiology and neuroradiology procedures, to identify procedures associated with higher radiation doses, and to determine the effects of various parameters on patient doses. MATERIALS AND METHODS: A prospective observational study was performed at seven academic medical centers. Each site contributed demographic and radiation dose data for subjects undergoing specific procedures in fluoroscopic suites equipped with built-in cumulative dose (CD) and dose-area-product (DAP) measurement capability compliant with International Electrotechnical Commission standard 60601-2-43. The accuracy of the dosimetry was confirmed by comprehensive measurements and by frequent consistency checks performed over the course of the study. RESULTS: Data were collected on 2,142 instances of interventional radiology procedures, 48 comprehensive physics evaluations, and 581 periodic consistency checks from the 12 fluoroscopic units in the study. There were wide variations in dose and statistically significant differences in fluoroscopy time, number of images, DAP, and CD for different instances of the same procedure, depending on the nature of the lesion, its anatomic location, and the complexity of the procedure. For the 2,142 instances, observed CD and DAP correlate well overall (r = 0.83, P <.000001), but correlation in individual instances is poor. The same is true for the correlation between fluoroscopy time and CD (r = 0.79, P <.000001). The correlation between fluoroscopy time and DAP (r = 0.60, P <.000001) is not as good. In 6% of instances (128 of 2,142), which were principally embolization procedures, transjugular intrahepatic portosystemic shunt (TIPS) procedures, and renal/visceral artery stent placements, CD was greater than 5 Gy. CONCLUSIONS: Most procedures studied can result in clinically significant radiation dose to the patient, even when performed by trained operators with use of dose-reducing technology and modern fluoroscopic equipment. Embolization procedures, TIPS creation, and renal/visceral artery stent placement are associated with a substantial likelihood of clinically significant patient dose. At minimum, patient dose data should be recorded in the medical record for these three types of procedures. These data should include indicators of the risk of deterministic effects as well as the risk of stochastic effects.