Animals
The animals were obtained commercially from KAC Company Limited (Kyoto, Japan). A total of three pigs (domestic, female, crossbred with Large Yorkshire and Landrace, mean weight 47.2 ± 0.6 kg, 3 to 4 months of age) were used in the current non-survival study. The study protocol was approved by the Committee for Ethics of Animal Experimentation of the National Cancer Center (Japan) (K18-022). Experiments were performed according to the Guidelines for Animal Experiments (Science Council of Japan: Guidelines for Proper Conduct of Animal Experiments, 2006) and Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines [13].
Novel oxygen saturation imaging endoscope system
A novel endoscope system EP-0002 for gastroenterological flexible endoscopy (FUJIFILM Corporation, Kanagawa, Japan) was used to capture the StO2 images (Fig. 1) [11, 12]. The EP-0002 provides the conventional observation modes of white light imaging (WLI). Additionally, the system provides a mode of oxygen saturation imaging (OSI), where the images of StO2 distribution in the tissue being observed can be obtained in real time. The WLI and OSI modes can be displayed simultaneously on separate monitors (Fig. 1). The OSI technology can derive the StO2 images from the differences in the absorption coefficient in the visible light region between oxy- and deoxy-hemoglobin using a small number of wavelengths (Fig. 2a). The light source is equipped with three types of laser diodes (405, 445, and 473 nm) for illumination. The laser light is guided to the tip of the endoscope, which excites the phosphors mounted at the tip to create white fluorescent light. The 405 and 445 nm laser light is used in the WLI mode. In the OSI mode, the unit alternately emits 445 and 473 nm light synchronously with the video frames at a speed of 30 frames per second (fps). Images are acquired with a red-green-blue (RGB) color charge-coupled device mounted at the tip of the endoscope. The dominant wavelength ranges of the sensitivity of the RGB image sensor are overlaid on both graphs (Fig. 2b). The processor unit controls the whole system and performs image processing for the conventional modes of WLI. In the OSI mode, the processor unit acquires two types of images sequentially and alternately that correspond to the illumination induced by the 445 and 473 nm lasers, respectively (Fig. 2c). The processor unit alternately extracts frames of 445 nm light from the input video signals and uses them to display the WLI images on the processor’s monitor at a speed of 15 fps (Fig. 2c). At the same time, the processor unit transfers the video signals to the image processing (IP) unit, and the IP unit creates an StO2 image by using two consecutive frames corresponding to 445 and 473 nm illuminations (Fig. 2c). The StO2 images are displayed by assigning the StO2 level to the RGB display color of the monitor in 1% increments, with the StO2 level of 100% as red, 50% as yellowish green, and 0% as dark blue. The StO2 images are displayed on the IP’s monitor at a speed of 15 fps almost in real time (Fig. 1).
Study protocol
After premedication with intramuscular injections of ketamine (10 mg/kg) and xylazine (2 mg/kg), general anesthesia was induced using 5% isoflurane. The airway was secured via endotracheal intubation, and general anesthesia was maintained with 1 to 3% isoflurane. The pigs were mechanically ventilated with the following settings: tidal volume, 10 to 15 ml/kg; respiratory rate, 12 breaths per minute; inspiratory to expiratory ratio, 1:2; positive end-expiratory pressure, 5 cmH2O; and fraction of inspired oxygen, 40%. Ringer’s lactate solution was administered intravenously through the auricular vein at 5 to 10 ml/kg/h. Heart rate, blood pressure, body temperature, and saturation of percutaneous oxygen (SpO2) were monitored throughout the procedure. An electrically heated mat was placed below each pig to prevent hypothermia.
Under general anesthesia, a midline laparotomy was performed. The marginal vessels of the rectum were dissected at the level of the planned anastomotic site, and the rectum was transected using a linear stapler (ECHELON FLEX ENDOPATH Stapler with a gold cartridge; Ethicon Endo-Surgery Inc, Cincinnati, OH, USA). The anvil was placed in the proximal side of the rectum, and end-to-end anastomosis was performed with the double stapling technique using a 25 mm diameter circular stapler (PROXIMATE ILS Curved Intraluminal Stapler; Ethicon Endo-Surgery Inc, Cincinnati, OH, USA). After the anastomosis, an air leak test was performed to check the completeness of the anastomosis via the endoscope. The root of the cranial rectal artery (CRA: equivalent to the superior rectal artery in humans) was dissected, and the porcine ischemia model was created in three pigs (Fig. 3). Prior to the assessment, the intestinal lumen was washed with saline to remove the stool. The contralateral side of the mesentery of the proximal rectum was marked with suture in each of the pigs. Bowel perfusion at the proximal side of the rectum was assessed intraluminally using a novel OSI endoscopy before and immediately after the anastomosis, as well as one minute and thirty minutes after CRA ligation.
At the end of the procedure, the pigs were sacrificed by an intravenous injection of a lethal dose of potassium chloride while under deep general anesthesia.
Quantitative evaluation of oxygen saturation imaging
Endoscopic procedures were recorded on videos and the stored still images were used to quantitatively evaluate the OSI with custom-made analysis software developed by FUJIFILM Corporation. A square region of interest was placed at three sites on the contralateral side of the mesentery of the proximal rectum in each of the pigs and each average value was calculated as the StO2 level. The StO2 levels were calculated at the following four time points: before and immediately after the anastomosis, and one minute and thirty minutes after CRA ligation. The average StO2 levels of all three pigs were compared among the different time points.
Statistical analysis
Statistical analyses were performed using JMP version 15.1.0 (SAS Institute, Cary, NC, USA). Values are presented as the mean ± standard deviation. For comparing the StO2 levels, a paired t-test was utilized. P-values < 0.05 were considered statistically significant.