Indocyanine green fluorescent imaging for the detection of parathyroid gland: a preclinical study

Yong Joon Suh, MD, Jung Woo Woo, MD, Young Joon Chai, MD, Hyungju Kwon, MD, Se Hyun Paek, MD, Eunyoung Kim, MD, Il Yong Chung, MD, Kyu Hyung Kim, MD, Su-jin Kim, MD, June Young Choi, MD, Kyu Eun Lee, MD, PhD, Yong Taik Lim, PhD, Yeo-Kyu Youn, MD, PhD. Department of Surgery, Seoul National University Hospital and College of Medicine, Seoul, Korea.

   The ability to visualize parathyroid glands has been a long-standing objective of endocrine surgeons. Near-infrared fluorescence (NIF) takes center stage recently and is applied to a wide range of medical areas. At the same time, indocyanine green (ICG) promotes this non-invasive and nonradiological technique as practically cost-effective and safe fluorophore. The present study investigated the fluorescence of parathyroid glands using ICG NIF in preclinical animal model in accordance with the regulations of the Institutional Animal Care and Use Committee (approval number 13-0167).

   The experiment was performed in three dogs under general anesthesia after the pilot study of five rats. Anatomic surgical dissection was performed first. And then the three dogs were injected with intravenous ICG of 12.5 µg/kg to 100 µg/kg. Tissue fluorescence was assessed serially under the illumination of a near-infrared light (785 nm, 500 mW diode laser). Real-time fluorescent images were obtained with the custom-built equipment fitted with an 835/45 nm band-pass filter and a 1394 ORCA-ERA charge-coupled device camera (Hamamatsu Photonics, Hamamatsu City, Japan). NIF imaging system was optimized for animal surgery, consisting of wide zoom lens (Nikon; Nikon Vision Co., Ltd., Tokyo, Japan). The fluorescent imaging was assessed sequentially after injection. And all images were used for data analysis with HCimage software (Hamamatsu Co. NJ, USA). Tissues were obtained and sent to a pathologist.

   The parathyroid glands of all three dogs were detected by a charge-couple device camera equipped with a 835/45 nm band-pass filter. Following ICG injection, the parathyroid glands did not show a green stain in an ambient light but could be visualized on fluorescent imaging. Increasing concentrations of ICG revealed a similar pattern but with increased intensity. Optimal visualization of parathyroid glands was obtained with a dose of 18.75 µg/kg of ICG. The excellent fluorescence of the parathyroid gland was recorded approximately 30 seconds after intravenous ICG injection. Peak fluorescence intensity was then noted approximately 20 seconds later and the fluorescence became notably diminished approximately 50 seconds later again. The duration of the fluorescence at adequate intensity was suitable to visualize the parathyroid gland. There was a positive correlation between NIF intensity and ICG dose up to 25 µg/kg. The pattern of this sequential graph was similar even at the thyroid gland. However, the fluorescence of the thyroid gland demonstrated visually different intensity and possessed different peak time of about fifty-five seconds. The fluorescence of thyroid glands did not hinder imaging of parathyroid glands. All tissues were confirmed by a pathologist.

   ICG NIF assists in the identification of parathyroid glands. This preclinical study suggests the possibility of an intraoperative anatomic navigation to save parathyroid glands during thyroidectomy or identify parathyroid glands during parathyroidectomy. Surgeons can also assess the viability of remained parathyroid glands. The results of this study support further investigation in clinical trials.

Fig. 1 Field image of anatomic dissection (a)

Fig. 1 Near-infrared image 50 seconds after indocyanine green injection (b)







Fig. 1 Histologic image of parathyroid gland (H&E, X100) (c)

Fig. 1 Histologic image of recurrent laryngeal nerve (H&E, X100) (d)






 Fig. 1 The different sequential net mean intensity graphs of parathyroid gland (e)

Fig. 1 The different sequential net mean intensity graphs of thyroid gland (f)

View Poster

« Return to SAGES 2014 abstract archive