Jochen Trübenbach

In vivo efficiency of four commercial monopolar radiofrequency ablation systems: a comparative experimental study in pig liver.

Rationale and Objectives:To evaluate the efficiency of 4 radiofrequency (RF) systems by assessing the amount of delivered energy for each thermal induced lesion after perfusion mediated RF ablation and to compare the influence of perfusion mediation types on the energy efficiency. Methods:A total of 43 ablations in 16 male landrace pigs with 4 RF devices were performed strictly according to the manufacturers’ instructions. Total absorbed energy was computed and then related to 3D volumetry obtained after histopathological evaluation. Sixteen ablations were performed under physiological liver perfusion and 27 ablations with occlusion of portal vein, hepatic artery, or both vessels. Energy efficiency values of the RF systems for different vascular occlusion techniques were compared and analyzed by a nonparametrical rank sum test. Results:Under physiological perfusion, the average energy delivered to produce 1-cm3 lesion size was calculated to 1650 ± 929, 3097 ± 389, 8312 ± 2068, and 5493 ± 2306 Watt · s/cm3 for the Berchtold®, RadionicsTM, RadiotherapeuticsTM, and RITA® system, respectively. After perfusion-mediated RF ablation, artery occlusion was not as effective as portal vein occlusion, which reduced the energy to 587 ± 148, 869 ± 276, and 903 ± 394 Watt · s/cm3 for the Berchtold®, RadionicsTM, and RadiotherapeuticsTM system, respectively. The occlusion of vessels, portal vein, and artery or portal vein alone increased the energy efficiency compared with physiological liver perfusion or occlusion of the artery (P = 0,003). Conclusions:Under physiological liver perfusion the open perfused system and the internally cooled system provided the best efficiency values with lowest standard deviations. The energy efficiency was increased markedly for all systems after occlusion of the portal vein either alone or in combination with arterial occlusion. Occlusion of the hepatic artery did not improve the efficiency.

Biphasic spiral CT of the liver: automatic bolus tracking or time delay?

Abstract The aim of this study was to evaluate the value of automatic bolus tracking for biphasic spiral CT of the liver in comparison with time delay examinations. Forty patients scheduled for a biphasic spiral CT of the liver randomly were examined either with time delay of 25 s for the arterial phase and 55 s for the portal-venous phase (n = 20), or with an automatic scan start triggered by contrast enhancement in the aorta (n = 20). Examinations were performed with 120 ml of contrast material and a flow rate of 4.0 ml/s. Density measurements of the aorta, of the liver parenchyma, and of the spleen were obtained by means of regions of interest (ROI). The end of the arterial phase was considered when hepatic parenchymal enhancement was greater than 20 HU. In all patients of the group with automatic bolus tracking arterial scanning was completed in the arterial phase of the liver. In 25 % of patients with fixed time delay, however, an enhancement of liver parenchyma during arterial phase greater than 20 HU was observed. During the portal-venous phase there was no significant difference in parenchymal enhancement between both groups. Automatic bolus tracking allows an individualized timing of the arterial phase in biphasic spiral CT of the liver. The timing is more accurate than in time delay scanning.

Contrast enhanced MR-guided biopsy of hepatocellular carcinoma.

Magnetic resonance (MR)–guided liver biopsy was performed in three patients with hepatocellular carcinoma. The tumor was considered (n = 2) or proven (n = 1) inaccessible with ultrasound or computed tomographic guidance. Because all lesions were poorly delineated on nonenhanced MR imaging, contrast agents (Gd-BOPTA, n = 1; ferucarbotran, n = 2) were applied to facilitate biopsy in an open low-field scanner. Postcontrast tumor conspicuity was fair in the patient receiving Gd-BOPTA and excellent in both patients receiving ferucarbotran, and biopsy was successful in all cases.

Automatic bolus tracking in monophasic spiral CT of the liver: liver-to-lesion conspicuity

Abstract The aim of this study was to evaluate the value of automatic bolus tracking for monophasic spiral CT of the liver and to assess the liver-to-lesion conspicuity in comparison with time-delay examinations. In 40 patients scheduled for therapy control of known hypovascular hepatic metastases a monophasic spiral CT was completed either with time delay of 65 s (n = 20) or with automatic bolus tracking in the liver parenchyma (n = 20). Examinations were performed with 120 ml of contrast material and a flow rate of 3.0 ml/s. For automatic bolus tracking a parenchymal enhancement threshold of 40 HU was used. Contrast enhancement in the liver parenchyma and in liver lesions was obtained by means of regions of interest (ROI). Mean parenchymal enhancement was not significantly different between time delay and bolus-tracking group. In 4 of 20 patients in the bolus-tracking group the threshold level of 40 HU was not reached. With automatic bolus tracking a significantly higher liver-to-lesion density difference was observed (P < 0.0001). Automatic bolus tracking allows a better liver-to-lesion conspicuity in monophasic spiral CT. Contrary to recent studies, a significantly higher parenchymal enhancement was not found using automatic bolus tracking.

Magnetic resonance-guided transcortical biopsy of bone marrow lesions using a magnetic resonance imaging-compatible piezoelectric power drill: preliminary experience.

König CW, Trübenbach J, Böhm P, et al. Magnetic resonance-guided transcortical biopsy of bone marrow lesions using a magnetic resonance imaging-compatible piezoelectric power drill: Preliminary experience. Invest Radiol 2003;38:159–163. Rationale and Objectives. To test utility and specific properties of a commercially available MRI compatible power drill for MR guided transcortical bone biopsy. Methods. In 17 patients MR-guided bone biopsy was performed in an open low-field scanner (0.2 T), using a piezoelectrically powered drilling machine. Target lesions were osteoblastic in four and nonsclerosed intramedullary in 13 cases. Titanium drills sized 3 to 4 mm and an outer cannula were coaxially used for power assisted cortical trephination. For intramedullary lesion sampling, spring loaded biopsy guns and sharpened cannulas were additionally applied in seven and fluid aspiration in two patients, respectively. Results. The piezoelectric device proved to be fully MR compatible. Trephination and subsequent biopsy was successful in all patients without major complications. The drilling procedure could entirely be performed inside the magnet in case of a lateral approach (n = 11). The net drilling time averaged 7.8 minutes for trephination of nonimpaired diaphyseal bone, but was up to 50 minutes in case of thickened femoral bone. Procedures were complicated by frequent drill loosening, drill obstruction by cortical bone (n = 4) and impaired periosteal grip (n = 5) with damage to the outer cannula (n = 3). Conclusions. The piezoelectric power drill can be applied safely in a low-field MRI environment and is a valuable tool to facilitate transcortical bone biopsy.

Growth characteristics and imaging properties of the morris hepatoma 3924A in ACI rats: a suitable model for transarterial chemoembolization.

Purpose: For experimental studies investigating modalities and efficacy of transarterial chemoembolization (TACE) in hepatocellular carcinoma (HCC) an animal model resembling the human situation as closely as possible would be appropriate. Specifically, reproducible tumor growth characteristics with the capability for appropriate in vivo imaging to monitor treatment efficacy are required.Methods: Morris hepatoma 3924A was implanted into the liver of 30 ACI rats. Tumor growth was followed by angiography (n=10), ultrasound (US, n=30), native computed tomography (CT. n=16), and native magnetic resonance imaging (MRU n=30) between day 8 and day 36 after implantation. The radiological morphological characteristics were compared with the macroscopic and microscopic histological findings of the explanted tumors.Results: In all 30 animals a solitary liver tumor was found and macroscopically no signs of metastases, ascites, or peritoneal tumor were visible. On histopathological examination tumor sizes ranged between 27 ± 3 mm3 (day 8) and 3468 ± 79 mm3 (day 36). The first signs of tumor necrosis occurred at day 16. US allowed tumor visualization from day 8, MRI from day 8, angiography from day 10, and CT from day 14.Conclusions: The tumor model has the potential to be used for the visualization of tumor growth by MRI and US. The potential for monitoring therapeutic effects of TACE needs to be investigated.

Radiopacity of Current Endovascular Stents: Evaluation in a Multiple Reader Phantom Study

PURPOSE To evaluate the radiopacity of endovascular stents based on the fluoroscopy mode in a phantom of the human pelvis. MATERIALS AND METHODS The following stents were included in this study: Medtronic AVE Bridge, Medtronic AVE Bridge X, Cordis Covered Nitinol (Covent), Guidant Dynalink, Luminexx, Guidant Megalink, Memotherm Flexx, Palmaz Medium, Palmaz-Schatz Long-Medium, Palmaz Corinthian PQ394Q and PQ294Q, SelfX, SMART without markers, SMART with radiopaque markers, Easy Wallstent. To evaluate radiopacity, images of the stents placed in four different positions (lumbosacral junction left and right, iliosacral joint left and right) of a pelvic phantom were taken at the following modes: spotfilm, continuous fluoroscopy, 15 pulses per second, 7.5 pulses per second, and 3 pulses per second. Images were presented at random to four independent readers and radiopacity scores were assessed: 0 = not visible, 1 = poor visibility, 2 = average visibility, 3 = good visibility, and 4 = very good visibility. RESULTS The Covent stent had the highest overall radiopacity score (3.25), followed by the Luminexx (3.04) and the Medtronic AVE Bridge X (2.74) stents. At the spotfilm mode, the best visible stents were the Medtronic AVE Bridge X, the Covent and the Easy Wallstent stents and at the continuous fluoroscopy mode, the Covent, the Luminexx, and the Medtronic AVE Bridge X stents. Decreasing the fluoroscopy mode went hand in hand with a reduction of the radiopacity scores of all stents. At the standard fluoroscopy mode of 7.5 pulses per second, the Covent stent was seen well or very well in 96.9%, followed by the Luminexx (76.9%), and the Medtronic AVE Bridge X (41.25%) stents. CONCLUSIONS Stent radiopacity directly depends on the fluoroscopy mode; if the pulse frequency decreased, detecting the stents became more difficult. Stent mass correlates with stent radiopacity (e.g., Cordis Covered Nitinol, Bridge X). Radiopaque markers may improve stent radiopacity dramatically (e.g., Luminexx vs Memotherm Flexx).

Stent struts and articulations: their impact on balloon-expandable stents' hoop strength, pushability, and radiopacity in an experimental setting.

Wiskirchen J, Pusich B, Kramer U, et al. Stent struts and articulations: their impact on balloon-expandable stents’ hoop strength, pushability, and radiopacity in an experimental setting. Invest Radiol 2002;37:356–362. objectives and rationale. To evaluate the hoop strength, the pushability, and the radiopacity of endovascular stents depending on their design. material and methods. The Palmaz Medium (rectangular stent struts, no articulation = RE-NA), the Palmaz-Schatz Long-Medium (rectangular stent struts, central articulations = RE-CA), and the 39 mm and 29 mm version of the Palmaz Corinthian Stent (arch-like stent struts, omega hinges = AR-OH 1 and AR-OH 2) were included in the study. Hoop strength, pushability, and radiopacity were assessed in-vitro. A radiopacity score (RS)—ranging from 0 = invisible, 1 = poor, 2 = average, 3 = good, and 4 = very good—was used. results. Mass adjusted hoop strength was nearly the same for the AR-OH 1 and the AR-OH 2 (AR-OH 1: 159.2 N/g*cm; AR-OH 2: 156.5 N/g*cm). Their hoop strengths per unit mass significantly exceeded that of a RE-NA (125 N/g*cm) and that of a RE-CA (75.3 N/g*cm). The most pushable stent was the AR-OH 2 (0.54 1/N) followed by the AR-OH 1 (0.47 1/N) and the RE-CA (0.169 1/N). Because of multiple stent dislocations, the RE-NA was considered nonflexible. At spotfilm mode the RE-NA was the most visible stent (RS 3.5) followed by the AR-OH 1 (RS 3.25), the RE-CA (RS 3.12), and the AR-OH 2 (RS 1.9). At continuous fluoroscopy the AR-OH 1 (RS 1.8) was better than the RE-CA (RS 1.125), the RE-NA (RS 1.0), and the AR-OH 2 (RS 0.6). At 7.5 pulses per second the RE-NA (RS 1.3) performed slightly better than the AR-OH 1 (RS 1.25), and clearly better than the RE-CA (RS 0.8), and the AR-OH 2 (RS 0.3). conclusion. Overall, the ideal stent—having high hoop strength, a low profile, a good pushability, and a good radiopacity—still does not exist. However, by changing strut design (from rectangular to arch-like struts) and by inserting articulations, hoop strength and pushability can be improved without reducing radiopacity.

Experimental thermoradiotherapy in malignant hepatocellular carcinoma

PURPOSE The human liver is known to be a relatively radiosensitive organ that develops clinically relevant late radiation hepatitis subsequent to whole liver treatment with total doses above 30 Gy in conventional fractionation. Experimental data, as well as clinical series, have demonstrated that hyperthermia of solid tumors in addition to radiotherapy enhances tumor growth inhibition and tumor control probability. We therefore developed an experimental model for combined radiotherapy and hyperthermia of the liver in transplantable rat Morris hepatoma 3924A. METHODS AND MATERIALS A cube of approximately 8 mm(3) was implanted subcapsularly into the middle liver lobe of 59 male syngenic ACI rats weighing approximately 180-200 g. On Day 16 after tumor implantation, irradiation of the tumor-bearing liver with either 0 Gy/25 Gy/35 Gy/45 Gy total dose in 10 fractions +/- hyperthermia (target temperature 40-42 degrees C) twice a week was initiated. Energy deposition was monitored by temperature probes in the liver and esophagus of the rats. Determination of tumor volume with magnetic resonance imaging was performed 2 to 5 weeks after the end of therapy. The tumor growth rates could be estimated for 44 rats. If the growth rate was positive (37 rats), the inverse of the growth rate was interpreted as the time to 10-fold tumor volume. Otherwise the maximum observation time was considered as a censored value in a parametric survival analysis. RESULTS Intrahepatic temperature probes showed a temperature plateau of greater than 40 degrees C after 5 to 8 min subsequent to initiation of hyperthermia. The target temperatures could be maintained for at least 22 min > or =40 degrees C and 10 min > or =41 degrees C, respectively. Median plateau temperature in liver, esophagus, and epicutaneously was 41.2 degrees C (standard deviation [SD] 0.7 degrees C; range 38.2 to 43.3 degrees C), 40.4 degrees C (SD 1.08 degrees C; range 38.9 to 41.8 degrees C), and 40.8 degrees C (SD 0.8 degrees C; range 38.2 to 42.7 degrees C), respectively. Elevation of the temperature in the esophagus correlated with intrahepatic temperatures in the range of 39-42 degrees C, r = 0.957. The increase in time to 10-fold tumor volume for each step of irradiation dosage was by 34% (95% confidence interval [CI] 20% to 49%) without hyperthermia and by 60% (95% CI 47% to 80%) with hyperthermia (p < 0.0001). CONCLUSION Treatment outcome after experimental percutaneous thermoradiotherapy in intrahepatically implanted Morris hepatoma 3924A was related to total dose of irradiation and concurrently administered regional hyperthermia. An increased radiosensitivity due to hyperthermia (<42 degrees C) has to be assumed.