Colour Doppler imaging of shunts from the left portal branch in portal hypertension. Description of a typical pattern

PURPOSE

To describe the typical colour Doppler appearance of a shunt through the parenchyma of the left lobe of the liver in portal hypertension.

MATERIAL AND METHODS

Ultrasound images of 141 patients with biopsy-verified cirrhosis were reviewed. Special note was taken of the appearance of shunts from the left portal branch.

RESULTS

In 28 patients, shunts from the left portal branch were detected ultrasonographically, 10 of which ran through the liver parenchyma on a course separated from the ligamentum teres. Seven of these 10 followed a tortuous course just below the surface of the liver creating a ball or corkscrew-like pattern.

CONCLUSION

Shunts from the left portal branch are not uncommon and may represent the only ultrasonographically detectable pathology in these patients. Recognition of the typical pattern will facilitate their detection.

Factors affecting color Doppler energy ultrasound recordings in an in-vitro model

Compared to conventional color Doppler ultrasound imaging, the new color Doppler modality "color Doppler energy" (CDE) has improved the possibility of visualizing blood vessels having low blood-flow velocities, but appears to be influenced by the settings of the ultrasound instrument and motion artefacts. The aim of this methodological study was to evaluate the effects of the different factors on the CDE signal. The CDE mode of a commercially available ultrasound system (Acuson 128 XP) was tested in an in vitro study. The effect of depth, angle of insonation, flow velocity, instrument power output, gain and other instrument settings were evaluated. The CDE signals obtained were stored on videotape and subsequently subjected to off-line computer analysis. The CDE signal intensity was found to be influenced mainly by fluid flow velocity, but was also affected by depth and instrument settings. Gain and power had, however, limited influence in this setting. Thus, the intensity of the CDE signal is influenced by several factors. Our results emphasize the need for optimum fixed preinstalled instrument settings when attempting to quantify organ perfusion by use of this new technique.

Color Doppler energy: computer analysis for detection of volume flow variations

The aims of this study were to (1) determine whether analysis of a color Doppler energy image can show small differences in volume flow and (2) determine the performance characteristics of color Doppler energy to aid future analyses. Flow through a silicone tube in a waterbath was insonated, and the images were analyzed by computer producing arbitrary values for the different flow volumes at different machine settings. A significant difference in the derived values was found at volume low changes of 10 ml/min. The differences are most pronounced when small amounts of color are shown in the vessel. This region of high sensitivity can be altered by changing the machine settings of filter, scale, or level. The angle of insonation must be taken into account. In conclusion, computer analysis of color Doppler energy can monitor changes in blood flow volume with a high sensitivity.

Color Doppler energy: computer analysis of color to assess angle dependency and detection of volume flow differences

The aims of this study were to assess (1) any angle dependency of color Doppler energy and (2) if computer analysis of the color can distinguish variations in volume flow. Fluid flowing through a silicon tube in a waterbath was insonated and the color in the images analyzed by a computer. The color presentation on the screen decreases rapidly as the angle approaches 90 degrees. At angles that did not produce a saturation of the color in the tube, the computer analysis detected significant differences between different flow volumes. We have concluded that monitoring of volume flow changes by analysis of color Doppler energy is possible, but the angle of insonation must be taken into account.

Incorporation versus infection of retroperitoneal aortic grafts: MR imaging features

The magnetic resonance (MR) imaging characteristics of normal aortic graft healing were compared with those of perigraft infection in 57 patients after aortic graft implantation. Thirty-three patients without postoperative complications underwent MR imaging in a 0.35-T unit 1 week after graft implantation, and 13 of those patients were reexamined 2-3 months after graft implantation. Twenty-four patients with clinically suspected perigraft infection underwent MR imaging 6 weeks to 18 years after graft implantation. Early normal postoperative changes were characterized by a perigraft collar of low to medium signal intensity on T1-weighted images and of high intensity on T2-weighted images in all 33 cases, consistent with perigraft fluid collection. In 10 of 13 patients reexamined 2-3 months postoperatively, the MR images demonstrated a collar of tissue consistent with perigraft fibrosis. In cases of clinical suspicion of retroperitoneal graft infection, MR imaging showed eccentric fluid collections of low to medium signal intensity on T1-weighted images and high intensity on T2-weighted images at more than 3 months after surgery. The MR findings were diagnostic of retroperitoneal perigraft infection in 17 of 20 patients shown to be infected at surgery. Retroperitoneal infection was correctly excluded on the basis of MR findings in four patients. Thus, MR imaging is an accurate imaging method for the diagnosis of aortic graft infection. In the early postoperative phase, resolving perigraft fluid cannot be differentiated from perigraft infection.

Diagnosis of prosthetic aortic graft infection by magnetic resonance imaging

The accuracy of magnetic resonance imaging (MRI) in the diagnosis of prosthetic aortic graft infection was evaluated in 18 patients with history and findings suggestive of this complication. The prospective interpretation of MRI was compared with surgical findings. Sixteen patients had a graft infection verified at operation. Fourteen patients had infection of the retroperitoneal portion of the graft; two patients had an infection limited to one of the groins; no graft infection was found at surgical exploration in the remaining two patients. Perigraft infection was correctly diagnosed on the basis of MRI findings in 14 of 16 cases; findings were false negative in one case, questionable in another case, and correctly excluded graft infection in two of two cases. MRI also defined the extent of infection in 14 of 16 cases. MRI findings that supported the clinical suspicion of graft infection were perigraft fluid collections remaining more than 3 months after surgery. Furthermore, local inflammation was suggested by an increased signal intensity of adjacent muscle on T2-weighted images in some cases. CT scans were performed in 12 patients; these enabled a correct diagnosis in five and provided indeterminate or false information in seven. These results indicate that MRI is helpful in the diagnosis of aortic graft infection. Furthermore, MRI provides information about the extent of infection crucial for planning therapy.

Aortorenal arterial autografts: the last two decades

Arterial autografts were introduced nearly a quarter century ago at the University of California, San Francisco and have proven their value for replacement in many demanding arterial problems. Renal artery fibrodysplasia is one of the more common lesions treated with arterial autograft. Arterial autografts that ideally match the renal artery and its branches are procured from the patient's own internal iliac artery. Either straight or branched configuration are available depending on the replacement requirements. In-situ aortorenal autografts are employed for lesions of the main renal artery or primary branches. Ex vivo repair involves temporary nephrectomy, pulsatile hypothermic perfusion and precise micro-vascular repair with unrestricted exposure, illumination, and an unhurried pace with no threat of renal ischemic insult. Autografts are attached proximally to the side of the aorta and distally to the disease-free end of the renal artery or a branch. The technique of arterial substitution for ex vivo repairs are identical except for the additional reanastomosis or reattachment of the renal vein. The arterial autograft exhibits the compliance characteristics which resemble a normal artery, maturation when used in the growing child, and durability essential for the long life span of this treated population. When the objective of a renal artery reconstruction is a normal renal arterial system, then the internal iliac artery autograft is the only choice.