ID: 837 (Conflict of Interest: K)

A new tool in battling anastomotic leak: objective intraoperative perfusion assessment in real-time during complex abdominal surgery – a pilot study of colon interpositions

L.Lobbes1, R.Hoveling2, K.Beyer1, M. E.Kreis1, B.Weixler1
1Charité Universitätsmedizin Berlin (CBF), Berlin
2Quest Medical Imaging, Middenmeer


Anastomotic leak is a common complication after esophageal reconstruction with colon interposition in critically ill esophageal cancer patients. Ischemia as well as decreased microperfusion contribute significantly to this complication. Up to date, surgeons have to rely solely on clinical signs of ischemia when assessing colon graft perfusion intraoperatively. Near-infrared (NIR) fluorescent imaging with indocyanine green (ICG) could help in assessing intestinal perfusion objectively as during the procedure there is only limited information on impaired perfusion available to the surgeon. We investigated a novel software for NIR tissue perfusion imaging in esophageal cancer patients undergoing colon interposition surgery.

Material und Methoden

Three patients with esophageal cancer who underwent colon interposition for esophageal reconstruction in 2019 were included. Patients 1 and 2 each received an isoperistaltic left colonic interposition, patient 3 received an anisoperistaltic left colonic interposition.  

Tissue perfusion of the colon graft and the respective anastomoses was measured intraoperatively at three different time points with the Quest Spectrum® (Quest Medical Imaging, Middenmeer), a device for fluorescence guided surgery. At time point one after isolation of the colon graft) a dose of 5 mg ICG was administered intravenously. At time point two (after construction of the cervical anastomosis) again a dose of 5 mg ICG was administered intravenously. At time point three (after contruction of the abdominal anastomosis) a dose of 20mg methylene blue was administered intravenously.

During the administration of ICG and methylene blue, images were acquired for 90 seconds at the mentioned locations with the Quest Spectrum® to register the changes of the NIR fluorescence signal over time. Postoperative analysis of the signal changes was performed with the Quest Research Software for different regions of interest (ROI). ROIs where selected in the areas that were planned for anastomosis and for the oral and aboral side of the anastomosis in the cervical and abdominal region. Changes in the NIR-fluorescence signal were plotted in a graph to get an insight of the ingress and egress phase of the tissue perfusion.


Three patients (two male) were included in this pilot study. Patient 1 developed anastomotic micro-leak of the cervical site at postoperative day 7. In patient 2 the colon graft failed at postoperative day 5 with anastomotic leak at both sides. In patient 3 (female) no anastomotic leak occured. The tissue perfusion graphs for patient 1 show a deviation in ingress and egress in the region selected for the failed cervical anastomosis compared to the successful distal anastomosis in all 3 measurements. For patient 2 the perfusion graphs for the colon graft as well as for the colonic side of the cervical and abdominal anastomoses show low ingress rates and positive egress rates, which are comparable to the graphs observed for the failed anastomosis in patient 1. In the graphs of all measured phases of patient 3 perfusion graphs were observed with similar ingress and egress patterns as shown in the abdominal anastomosis region of patient 1.


Postoperative evaluation of our intraoperative imaging data shows that a significant decrease in perfusion was already detectable in the area of ​​future anastomotic leak intraoperatively with the aid of the software used here for the first time worldwide. Further studies on more patients are needed to elucidate the clinical value of this novel technique.