Asbjørn Ødegård

Three-dimensional ultrasound-based navigation combined with preoperative CT during abdominal interventions: a feasibility study.

AbstractPurpose: Three-dimensional (3D) intraoperative ultrasound may be easier to interpret when used in combination with less noisy preoperative image data such as CT. The purpose of this study was to evaluate the use of preoperative image data in a 3D ultrasound-based navigation system specially designed for minimally invasive abdominal surgery. A prototype system has been tested in patients with aortic aneurysms undergoing clinical assessment before and after abdominal aortic stent-graft implantation. Methods: All patients were first imaged by spiral CT followed by 3D ultrasound scanning. The CT volume was registered to the patient using fiducial markers. This enabled us to compare corresponding slices from 3D ultrasound and CT volumes. The accuracy of the patient registration was evaluated both using the external fiducial markers (artificial landmarks glued on the patient’s skin) and using intraoperative 3D ultrasound as a measure of the true positioning of anatomic landmarks inside the body. Results: The mean registration accuracy on the surface was found to be 7.1 mm, but increased to 13.0 mm for specific landmarks inside the body. CT and ultrasound gave supplementary information of surrounding structures and position of the patient’s anatomy. Fine-tuning the initial patient registration of the CT data with a multimodal CT to intraoperative 3D ultrasound registration (e.g., mutual information), as well as ensuring no movements between this registration and image guidance, may improve the registration accuracy. Conclusion: Preoperative CT in combination with 3D ultrasound might be helpful for guiding minimal invasive abdominal interventions.

Initial observations of endovascular aneurysm repair using Dyna-CT.

Purpose: To examine the feasibility of a new technology that provides images resembling computed tomographic (CT) slices on the operating table during elective endovascular abdominal aortic aneurysm (AAA) repair (EVAR). Technique: Commercially available Dyna-CT equipment was used in conjunction with Voxar 3D software for image reconstruction. During the preliminary evaluation of 9 patients (7 men; median age 73 years, range 67–84) with non-ruptured AAA undergoing elective EVAR with the Zenith Trifab stent-graft under regional anesthesia, the equipment functioned sufficiently well to produce the planned image dataset in 8 of 9 patients. Rotation failed in 1 overweight individual due to a short distance between the patients abdomen and the C-arm. In 7 of 8 patients, the radiologist was able to measure the diameter of the aneurysm, and it was possible to visualize the entire length of the treated aorta and the iliac arteries in all cases. The proximal aortic neck, including the stent-graft, was visualized in 7 cases. No endoleak or other complications was detected, which was confirmed on the pre-discharge CT scans. The time interval from the rotation of the C-arm until the appearance on the monitor of the first automatically generated 3D images was 7 minutes, which has subsequently been reduced to ∼2 minutes. Conclusion: Our preliminary experience with Dyna-CT is promising, but further research is necessary to define the place of this imaging modality in EVAR.

DynaCT during EVAR – A Comparison with Multidetector CT

OBJECTIVES We have explored the usefulness of an on-table, cross-sectional radiological imaging (DynaCT) in endovascular aortic repair (EVAR). DynaCT images were compared to images from a regular multidetector (16 slice) CT. In the comparison, we tested the accordance of firstly 5 relevant clinical measurements and secondly the visibility of 9 anatomical areas in the two different types of images. This imaging was carried out in addition to the usual angiographic imaging. DESIGN, MATERIAL AND METHOD: 20 patients with infrarenal abdominal aortic aneurysm (AAA) were prospectively enrolled in the study. We compared Images from DynaCT with two different doses of contrast medium to MDCT-images in two different ways. Firstly relevant arterial diameters and lengths and secondly, 9 anatomical areas were evaluated regarding visibility which was scored on a 4-point scale. RESULTS There were no significant differences in the measured arterial diameters and lengths. MDCT had a significantly higher visibility score than both DynaCT investigations. However, with the highest contrast medium dose we found acceptable diagnostic quality in 78-94% of the cases for 8 of the 9 investigated anatomical areas. CONCLUSION Our findings indicate that on-table DynaCT are of sufficient quality to give relevant information of arterial measurements, needed in endovascular repair of infrarenal aortic aneurysms.

A closed-chest pulmonary artery occlusion/reperfusion model in the pig: detection of experimental pulmonary embolism with MR angiography and perfusion MR imaging.

RATIONALE AND OBJECTIVES To establish a pig model suitable for imitating pulmonary emboli to facilitate research in the diagnosis of pulmonary embolism. METHODS Thirteen animals were anesthetized, mechanically ventilated, and subjected to pulmonary artery catheterization initiated from the right external jugular vein. With the use of a Swan-Ganz catheter, repetitive occlusion/reperfusion maneuvers were done at different locations of the pulmonary arterial tree. Conventional pulmonary angiography, MR angiography, and perfusion MR imaging were performed. RESULTS The model remained hemodynamically stable throughout the 13 experiments, without any significant difference between the blood pressure measurements at the start and at the end of the right-heart and pulmonary artery catheterizations. In each of the nine animal experiments that investigated MR imaging, four of four using perfusion MR imaging (proximal and distal occlusions) and five of five using MR angiography (larger pulmonary artery occlusions), all repeated pulmonary artery occlusions were successfully performed (reproducibility of 100%). CONCLUSIONS The closed-chest pulmonary artery occlusion/reperfusion model in the pig allowed repetitive, controlled imitations of pulmonary emboli at different levels of the pulmonary artery in the same experiment. MR angiography and perfusion MR imaging were adequate to detect the pulmonary artery occlusions and the nonperfused lung regions, respectively. The model may be a helpful tool for future research in this field.

Iohexol-Induced Neutrophil Myeloperoxidase Release and Activation upon Contact with Vascular Stent-Graft Material: A Mechanism Contributing to the Postimplantation Syndrome?:

Purpose: To investigate whether the contrast medium iohexol alone or in combination with vascular stent-graft material induces neutrophil degranulation. Methods: Human whole blood or isolated human neutrophils were incubated with or without iohexol and vascular stent-graft material. Samples were also drawn from 5 patients undergoing diagnostic angiography using iohexol. Myeloperoxidase and lactoferrin concentrations were determined by enzyme immunoassays. Results: Both in vitro and in the patients, iohexol induced neutrophil degranulation with considerable individual variation in dose sensitivity and timing. The in vitro effect was independent of the type of anticoagulant used (ethylenediamine tetra-acetic acid, heparin, lepirudin). Experiments using isolated neutrophils showed that degranulation was independent of complement activation or platelet-derived mediators. The dose for maximal response varied from 5 to 50 mg I/mL (10.7–107.6 mg/mL iohexol). In vitro, vascular stent-graft material alone did not induce neutrophil degranulation. As compared to iohexol alone, incubation with iohexol and vascular stent-graft material in combination substantially increased the release of myeloperoxidase. Conclusions: Iohexol induces neutrophil degranulation, which is greatly enhanced when combined with vascular stent-graft material. Thus, iohexol-induced neutrophil activation may contribute to an inflammatory response following stent-graft implantation. We speculate that neutrophil activation during other procedures combining catheters and iohexol (e.g., angiography) may induce inflammation, which might have detrimental effects.

DynaCT in Pre-treatment Evaluation of Aortic Aneurysm before EVAR

OBJECTIVE DynaCT(®) is a method for obtaining computed tomography (CT)-like images using a C-arm system. Our aim was to compare the accuracy of these images to multidetector CT (MDCT) images prior to endovascular aortic repair (EVAR). METHODS A non-consecutive group of 20 elective patients were prospectively exposed to MDCT and one additional DynaCT before EVAR. Six arterial measurements and nine anatomical areas were chosen to: (1) visualise the peri-aortic soft tissue and assess the possibility to diagnose a potential haemorrhage from a ruptured aneurysm and (2) make the pre-treatment measurements before insertion of stent graft. Differences between modalities and readers were statistically compared using a linear mixed model. RESULTS For maximum aortic diameter, a significant difference of 1.3 mm was found between techniques (p = 0.043). Visibility scores were significantly better for all areas in MDCT data. Pre-treatment evaluation with DynaCT before EVAR was possible for all areas; evaluation of the iliac arteries were suboptimal due to a limited imaging volume size. Significant inter-reader differences were found for all anatomical areas. CONCLUSION The result indicates that DynaCT gives sufficient information to determine the correct treatment and for selecting the proper stent graft before EVAR. A limited volume size reduces the evaluation of the iliac arteries.

Type III Endoleak from a Thoracic Aortic Stent-Graft

Purpose: To present an as yet unreported late complication of an Excluder thoracic endograft. Case Report: A 78-year-old man underwent surgery for a ruptured type V thoracoabdominal aortic aneurysm in 1996. Four years later, an aneurysm was detected in the proximal thoracic aorta and repaired with 2 Excluder endoprostheses. At 12 months, computed tomography showed an increase in the aneurysm sac diameter and a type III endoleak, which was traced to a hole in the stent-graft fabric on arteriography. No fracture of the metal components was detected in the stent-grafts. Another Excluder device was implanted within the distal endograft. Satisfactory exclusion of the leak has been maintained for 6 months. Conclusions: The risk of type III leaks must be minimized before stent-grafting can be regarded as a routine procedure in the treatment of thoracic aortic aneurysms.