Introduction: The intraoperative use of fluorescence imaging agents has started a new era in the field of image-guided surgery. For fluorescence guidance both a fluorescence imaging agent and a dedicated fluorescence camera system are required. Unfortunately current clinically available fluorescence camera systems require the lights in the operation room to be dimmed. This temporarily stalls the surgical procedure as the anatomical context is lost. As a first step to full embedment of the fluorescence guidance technique into clinical routine, we evaluated a prototype fluorescence camera that works under ambient light conditions. Findings with this camera (modified-PhotoDynamicEye; m-PDE) were compared to the findings with its predecessor (c-PDE).     

Materials and Methods: The evaluation was performed in a group of seven patients (melanoma: n=4; oral cavity carcinoma: n=3) that were scheduled to undergo sentinel node biopsy (SNB) using the hybrid tracer indocyanine green (ICG)-99mTc-nanocolloid. Written informed consent was obtained from all included patients. The feasibility of the hybrid tracer, which contains a radioactive (99mTc-nanocolloid) and a fluorescence signature (ICG), for SNB was previously shown.

                  An injection with hybrid tracer injection was given (4 deposits of 0.1 mL) peritumorally or surrounding the melanoma scar and was followed by preoperative lymphoscintigraphy and SPECT/CT imaging to determine the number and location of the SN(s). Via gamma tracing and fluorescence guidance the SN(s) were localized and excised. A portable gamma camera was used pre- and post-SN excision to confirm complete SN removal.  

                  For intraoperative fluorescence guidance, the prototype fluorescence camera (m-PDE) was compared to the conventional-PDE (c-PDE.) fluorescence camera for 1) Intraoperative fluorescence-based SN identification rate; 2) Experience of the surgeon(s).

Results:In the direct camera comparison the m-PDE was found superior over the c-PDE. Intraoperatively, 27 SNs were identified and excised. All (100%) were identified via fluorescence imaging using the m-PDE vs. 85.2% identified with the c-PDE. With the m-PDE 40.7% of the SNs could already be visualized transcutaneously vs. 22.2% using the c-PDE. Using the m-PDE, pseudo-coloring of the fluorescence signal to green (vs. black-and-white of the c-PDE) provided better signal-to-background contrast increasing SN detectability.

Conclusion:SN identification improved using the prototype m-PDE fluorescence camera in comparison to its predecessor. Because of its ability to provide real-time surgical guidance under ambient light conditions, the m-PDE enables the surgeon to link fluorescence-based findings directly to the anatomical background without stalling of the surgical procedure.