Near-infrared fluorescence cholangiography at a very low dose of indocyanine green: quantification of fluorescence intensity using a colour analysis software based on the RGB color model

Abstract

Background: Indocyanine green (ICG) near-infrared fluorescence cholangiography (NIRF-C) is widely used to visualize the biliary tract during laparoscopic cholecystectomy (LC). However, the ICG dose and its dosing time vary in the literature so there is not a standard ICG protocol. The objectives of this descriptive prospective study were to demonstrate that NIRF-C at a very low dose of ICG provides good visualization of the extrahepatic biliary tree while avoiding hepatic hyperluminescence and to assess the surgeon-perceived benefit. Furthermore, another additional aim was quantifying the amount of ICG dye in the liver tissue and biliary tract through a green colour intensity (GCI) analysis according to red green blue (RGB) color model and correlating it to surgeon-perceived benefit.

Method: Forty-four patients were scheduled for LC. We recorded demographics, surgical indication, intraoperative details, adverse reactions to ICG, hepatic hyperluminescence, visualization of the cystic duct (CD), the common bile duct (CBD) and the cystic duct-bile duct junction (CDBDJ) before and after dissection of Calot's triangle, operation time, surgical complications and subjective surgeon data. For all procedures, a unique dose of 0.25 mg of ICG was administered intravenously during the anaesthetic induction. ICG NIRF-C was performed using the overlay mode of the VISERA ELITE II Surgical Endoscope in all surgeries. Video recordings of all 44 LC were reviewed. Using a color analysis software, the GCI of CBD versus adjacent liver tissue was calculated using RGB color model.

Results: ICG NIRF-C was performed in all 44 cases. The mean operation time was 45 ± 15 min. There were no bile duct injuries (BDIs) or allergic reactions to ICG. The postoperative course was uneventful in all of cases. The mean postoperative hospital stay was 28 ± 4 h. ICG NIRF-C identified the CBD in 100% of the patients, the CD in 71% and the CDBDJ in 84%, with a surgeon satisfaction of 4/5 or 5/5 in almost 90% of surgeries based on a visual analogue scale (VAS). No statistically significant differences were found in the visualization of the biliary structures after the dissection of Calot's triangle in obese patients or with gallbladder inflammation. Furthermore, 25% of patients with a BMI ≥ 30, 27% of patients with a Nassar grade ≥ 3 and 21% of patients with gallbladder inflammation had a VAS score 5/5 compared to 6% of patients with a BMI < 30 (p = 0.215), 6% of patients with a Nassar grade < 3 (p = 0.083) and none of the patients without gallbladder inflammation (p = 0.037). Measured pixel GCI of CBD was higher than adjacent hepatic tissue for all cases regardless of the degree of gallbladder inflammation, the Nassar scale grades or the patient's BMI (p < 0.0001). In addition, a significant correlation was observed between surgeon-perceived benefit and the amount of ICG dye into the CBD according the RGB color model (p < 0.0001).

Conclusion: ICG NIRF-C at a very low dose of ICG (0.25 mg of ICG 20 min before surgery) enables the real-time identification of biliary ducts, thereby avoiding the hepatic hyperluminescence even in cases of obese patients or those with gallbladder inflammation.

Keywords: Hepatic hyperluminescence; Indocyanine green; Laparoscopic cholecystectomy; Near-infrared fluorescence cholangiography.