Angiographic Analysis of Stenotic Renal Arteries

Improvement and Standardization of the Acquisition and Analysis of Angiographic Radiographs of Stenotic Renal Arteries

Dutch Kidney Foundation

December 1996 - December 1999
Hans van Assen, Henri Vrooman, Berend Stoel

Background

Figure4About 10 to 20 % of the Dutch population suffers from high blood pressure (hypertension). Renal artery stenosis is one of the few causes of hypertension that can be treated directly. If not treated, however,  the disease can progress rapidly in a period of 2 years. In about 10 percent of the cases, total occlusion of the renal artery occurs, causing a rapid loss of renal function. Therefore, the assessment of the progress of the disease is of critical importance to the clinical outcome. The common techniques to determine the location and significance of a stenosis is by visual assessment of digital angiograms. There are two major drawbacks, however: 1) often the optimal projection angle is hard to determine, which leads to an underestimation of the significance of the stenosis; and 2) the visual assessment is accompanied by high inter- and intra observer variability.
The project was particularly aimed at the development of an analytical workstation for an accurate quantification of renal artery stenoses using automatic contour detection algorithms. The advantages of such a radiological workstation are:1) a higher accuracy and reproducibility of the quantification of stenoses; 2) the application of quantitative measurement techniques, in particular during interventions ("on-line"); and 3) the opportunity to obtain additional anatomical (morphological) and functional data. 

Goals

The general aim of the project was to develop a software package that can accurately measure the diameter function of renal arteries in a reproducible manner from images acquired from conventional angiography and Digital Subtraction Angiography (DSA) and to perform a clinical evaluation of the developed software and the resulting measurement results.

Technical objectives:

  • The development of a software package for the quantification of renal artery stenosis.
  • The development of a calibration technique that can be used for catheters smaller than 6F.
  • To accurately measure the angle between both branches in a vessel bifurcation.

Clinical objectives:

  • The possibility to derive the functional significance of a stenosis based on morphologic parameters, as obtained from the software package.
  • To compare the morphologic data with the results from quantitative analysis of MR-angiography.
  • To compare the quantitative analysis results with visual assessment by radiologists
  • To evaluate the analysis procedure in terms of inter- and intra-observer variability

Approach

The quantitative method to analyze renal arteries consists of 5 consecutive steps:

1. Image calibration based on the distances between radiopaque markers mounted equidistantly on the catheter. Figure1
2. Detection of path line(s) roughly through the center of the vessel.

Figure2

3. Vessel wall detection, based on the detected path line(s). Figure3

4. Vessel reconstruction, resulting in a reference diameter.

 

Figure4

5. Parameter calculation, i.e. minimal lumen diameter, reference diameter, obstruction length and percentage diameter stenosis.

 

Figure5

 

The technique used for the detection of the path line(s) and the vessel wall is dependent on whether the vessel segment is a mid segment or a so-called ostial segment. In a mid segment, one path line is detected, similar to quantitative coronary analysis techniques, whereas the analysis of an ostial segment requires two path lines that coincide at some point. These path lines then form the basis for the vessel wall detection. Figure6

Status

The project finished in 2000.

Gallery

Figure7

Figure8

Contact

For further information, please contact:
B.C. Stoel, PhD.
Division of Image Processing
Department of Radiology, 1-C2S
Leiden University Medical Center
P.O. Box 9600
2300 RC Leiden
The Netherlands
Tel. +31 (0)71 526 1911
Fax. +31 (0)71 526 6801
e-mail: B.C.Stoel@lumc.nl