3-ccd (charge Coupled Device) Image Enhancement for Bowel Ischemia

Introduction: Determination of bowel perfusion in cases of strangulating hernias, SMA occlusion, and other ischemic states is currently a subjective decision of surgeons. 3-CCD image enhancement allows objective and quantitative criteria to be included in a surgeon’s decision as to whether bowel resection is warranted secondary to ischemia. 3-CCD technology separates visible light into its three primary wavelength regions (red, green, and blue); by using an easily implemented mathematical algorithm, the amount of light detected by the red, green, and blue CCDs can be directly correlated with tissue oxygenation. We have previously demonstrated the utility of this approach in vessel identification, partial nephrectomy, donor nephrectomy, and now extend these studies to bowel ischemia.

Materials and Methods: In a pilot study, two cases of ischemic bowel were analyzed using 3-CCD laparoscopic images and 3-CCD image enhancement. Ischemic bowel presented with lower values than healthy bowel. Watershed areas were then designated as either ischemic or well-perfused. To adjust for variations in brightness and clarity inherent to laparoscopic imaging, images were standardized such that fat always received a value of “1”. Fat is ubiquitous and well-perfused due to its numerous capillaries.

Results: In this scenario, healthy bowel received a value of “0.727” while ischemic bowel received a value of “0.486,” with a p-value <0.01. In transition areas, i.e. borderline areas between pink, healthy-appearing bowel and blue-black, swollen, ischemic appearing bowel, values greater than 0.653 suggested viable bowel.

Conclusions: Use of 3-CCD technology in laparoscopic surgery could help surgeons better identify areas of ischemic bowel and preserve healthy tissue. Further studies are planned to use 3-CCD technology to predict bowel viability in primates with ischemic injury.


Session: Podium Presentation

Program Number: S011

« Return to SAGES 2010 abstract archive