A MRI study of relative organ position in the upper abdomen for changes in patient orientation: application to image-registered endoscopic ultrasound

R. San Jose Estepar, Bruno Madore, Mark Yarmarkovich, Kirby G. Vosburgh
International Journal of Computer Assisted Radiology and Surgery
Volume 4, Number Supplement 1, Pages S257-S259
2009

Download full paper

Abstract

Purpose: This paper presents a study of abdominal organ shift due to changes in body posture. A previous study [1] has reported a noticeable qualitative shift for abdominal structures, comparing prone and supine CT Studies. Such measurements are relevant for the use of preoperative imaging in guiding endoscopic interventions. CT and MRI images of the torso are conventionally made with the patient supine, and this protocol is unlikely to change in conventional practice. While abdominal surgeons generally perform procedures with their patients supine, gastroenterologists use a variety of patient poses, depending on the examination. Thus the CT images made of a surgical patient will "fit" the relative organ positions observed by the surgeon in the OR, but these images will be less accurate in the endoscopy suite. Two specific examples are the conventional setting for gastroscopic studies, whether conventional or ultrasound, for which the patient is typically placed in the left lateral decubitus position, and endoscopic retrograde cholangio-pancreatography (ERCP), in which the patient is prone. Compared with their positions when the patient is supine, soft tissue structures such as the liver, the kidney (and most likely, the pancreas) will deform due to the changes in gravity and external body forces. We have carried out a MRI study with two subjects to assess the amount of motion with respect to the supine position for the liver and the left and right kidney. Since the pancreas is tied to relatively rigid structures in the retroperitoneum, we expect this effect to be smaller than with the liver, with the movement of the kidneys and the gall bladder being in-between.

Conclusions: The shifts in abdominal organs with patient pose are significant, relative to instrument targeting requirements for image-guided procedures. While our initial hypothesis was that the liver was going to show a larger effect due to patient position, our results suggests that the kidneys are more prone to shift. This motion can be explain by the little constrains offered by the facial spaces surrounding the kidney. In summary, global information from a supine CT should likely not be used for guiding interventions when the patient is lying on his side, unless corrections are made for these gravity-induced effects. It appears from this very limited study that the degrees of organ motion will be highly dependent on the individual patient; likely due to factors such as patient age, sex, and degree of obesity. The scale of organ motion effects and the limitations of current body models (and the practical inability to perform multi-position scanning in the clinical setting) imply that successful image guided interventions will require accurate local real time images to support navigation to clinical targets.

CARS 2009 presentation poster
Poster

Reference

San Jose Estepar R, Madore B, Yarmarkovich M, Vosburgh KG. A MRI study of relative organ position in the upper abdomen for changes in patient orientation: application to image-registered endoscopic ultrasound. International Journal of Computer Assisted Radiology and Surgery 2009;4(Supplement 1):S257-S259.

Bibtex entry

@Article{sanjoseCARS09,
  author         = {Raul {San Jose Estepar} and Bruno Madore and Mark          
                   Yarmarkovich and Kirby G. Vosburgh},                        
  title          = {A {MRI} study of relative organ position in the upper      
                   abdomen for changes in patient orientation: application to  
                   image-registered endoscopic ultrasound},                    
  journal        = {International Journal of Computer Assisted Radiology and   
                   Surgery},                                                   
  year           = {2009},                                                     
  volume         = {4},                                                        
  number         = {Supplement 1},                                             
  pages          = {S257-S259}
}