Intravenous application of fluorescein for confocal laser scanning microscopy: evaluation of contrast dynamics and image quality with increasing injection-to-imaging time

• Published in: Gastrointestinal endoscopy 2008-08, Vol.68 (2), p.319-323
• Main Authors: Becker, Valentin, MD, von Delius, Stefan, MD, Bajbouj, Monther, MD, Karagianni, Artemisia, MD, Schmid, Roland M., MD, Meining, Alexander, MD
• Format: Article
• Language: English
• Subjects: Animals; Biological and medical sciences; Contrast Media - pharmacokinetics; Digestive system. Abdomen; Dose-Response Relationship, Drug; Endoscopy; Evaluation Studies as Topic; Female; Fluorescein - pharmacokinetics; Fluorescein - pharmacology; Gastric Mucosa - pathology; Gastric Mucosa - ultrastructure; Gastroenterology and Hepatology; Gastroenterology. Liver. Pancreas. Abdomen; Gastroscopy - methods; Image Enhancement - methods; Infusions, Intravenous; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Microscopy, Confocal - methods; Models, Animal; Probability; Sensitivity and Specificity; Statistics, Nonparametric; Stomach - pathology; Stomach - ultrastructure; Swine; Time Factors
• ISSN: 0016-5107; 1097-6779
• DOI: 10.1016/j.gie.2008.01.033

Background Confocal laser scanning microscopy (CLSM) is reported to be a promising tool for in vivo histopathology during an endoscopy. The most commonly used fluorophore is fluorescein sodium given intravenously. However, so far, there are no objective data on contrast dynamics and image quality over time after injection of the substance. Objective We aimed to investigate the ideal time period for the best CLSM imaging when using fluorescein sodium. Design The proximal stomachs of 4 anesthetized pigs were examined via standard endoscopes with a portable confocal miniprobe. After injection of 1% fluorescein intravenously (0.1 mL/kg body weight), consecutive sequences were recorded after 1, 2, 3, 5, 8, 20, 30, 40, 50, and 60 minutes. After the procedure, histograms, which represented signal intensity of 20 images (at respective time points), were analyzed from representative regions of interest (ROI). The mean (standard deviation [SD]) pixel value of the pixel values in the ROI were noted. A signal-to-noise ratio (SNR), defined as the ratio between the mean and the SD, was calculated. In addition, 5 images per time period were put into a random order and then blindly evaluated by 2 endoscopists. Results A total of 560 histograms from selected ROIs were analyzed. The SNR decreased significantly after 8 minutes, with the lowest values being reached after 30 minutes. Subjective evaluation of still images also demonstrated a continuous decrease, with the lowest image quality after 30 minutes. A slight increase of image quality after 40 minutes occurred because of extramucosal leakage of fluorescein, with fluorophore deposits in the crypt orifices. Conclusions In the pig model, fluorescein-based CLSM of the GI tract delivered the best contrast and image quality within the first 8 minutes after injection of the fluorophore. These data might facilitate standardization of the procedure for research purposes but may also help when applying this new technology in a daily routine manner.