Comparison of hyperspectral imaging and fluorescence angiography for the determination of the transection margin in colorectal resections-a comparative study

Abstract

Purpose: One relevant aspect for anastomotic leakage in colorectal surgery is blood perfusion of both ends of the anastomosis. The clinical evaluation of this issue is limited, but new methods like fluorescence angiography with indocyanine green or non-invasive and contactless hyperspectral imaging have evolved as objective parameters for perfusion evaluation.

Methods: In this prospective, non-randomized, open-label and two-arm study, fluorescence angiography and hyperspectral imaging were compared in 32 consecutive patients with each other and with the clinical assessment by the surgeon. After preparation of the bowel and determination of the surgical resection line, the tissue was evaluated with hyperspectral imaging for 5 min before and after cutting the marginal artery and assessed by 6 hyperspectral pictures followed by fluorescence angiography with indocyanine green.

Results: In 30 of 32 patients, the image data could be evaluated and compared. Both methods provided a comparable borderline between well-perfused and poorly perfused tissue (p = 0.704). In 15 cases, the surgical resection line was shifted to the central position due to the imaging. The border zone was sharper in fluorescence angiography and best assessed 31 s after injection. With hyperspectral imaging, the border zone was visualized wider and with more differences between proximal and distal border.

Conclusion: Hyperspectral imaging and fluorescence angiography provide similar results in determining the perfusion border. Both methods allow a good and safe visualization of the blood perfusion at the central resection margin to create a well-perfused anastomosis.

Trial registration: This study was registered at Clinicaltrials.gov ( NCT04226781 ) on January 13, 2020.

Keywords: Anastomotic leak; Colorectal resection; Fluorescence angiography (FA); Hyperspectral imaging (HSI); Indocyanine green (ICG).

Conflict of interest statement

Hannes Köhler is an Employee of Diaspective GmbH. The authors declare that they have no conflict of interest.

Figures

Fig. 1

a Bland Altman plot of the measured distances in cm to the forceps in HSI and ICG. Scattering of the points around the zero line indicated agreement of both measured distances. b Bland Altman plot of the measured width in centimeters. Scattering of the points above the zero line suggested disagreement of the width of the border zone in HSI-and ICG. (SD = standard deviation)

Fig. 2

a–c An instrument marked the planned transection line. The ruler is best seen in the RGB picture (c). a, b Based on the respective minimum and maximum intensity of each image, border limit values were calculated. In this case, the HSI StO2 limit value 3 min after devascularisation was 70 and the FA/ICG value at the maximum initial flooding of the dye amounted to 131. The most central limit point in FA/ICG (a) and HSI (b) was marked and deviations to the instrument were measured. While FA pictures hardly any differences between the most proximal and distal point of the borderline (a), HSI shows a large distance between both points (b)

Fig. 3

Distribution of the distance from the planned transection line to the visualized borderline in HSI and ICG data in cemtimeters

Fig. 4

Most central and distal limit value in FA with ICG (a) and HSI StO2 (b). A similar course and width of both border zones are noticeable. b HSI StO2 image of a border zone with many color graduations between good proximal and poor distal perfusion. c Another patient’s HSI StO2 image with a clear and sharp borderline is showing the variety of border zones revealed with HSI