To prospectively assess the ideal dosing and the value of fluorescent sentinel lymph node (LN) detection with indocyanine green (ICG) for the detection of LN metastases in intermediate- and high-risk patients undergoing robot-assisted prostatectomy and extended pelvic LN dissection (ePLND).Patients and Methods
Twenty patients received transperineal prostatic injections of ICG. Patients were cycled through 5 doses (1.25, 2.5, 3.75, 5, and 7.5 mg) so optimal ICG dosing could be discovered early.Results
ICG injection was able to identify fluorescent LN (FLN) packets in all 20 patients. Compared to the higher ICG doses, the 1.25 and 2.5 mg doses had fewer FLN packets and were abandoned after 1 dose each. The median number of FLN packets was 4.0, 6.0, and 4.5 for the respective doses of 3.75, 5.0, and 7.5 mg. The external iliac group was the most common site of fluorescence in 27.2% of patients, followed by the common iliac (21.3%), obturator (20.3%), internal iliac (18.5%), and node of Cloquet (7.7%). Seven (35%) of 20 patients had node-positive disease. Of the 5 patients that had fluorescent tissue outside of our ePLND template, 1 had a positive node present in the anterior bladder neck fat. Across all patients, ICG had 62% sensitivity, 50% specificity, 8% positive predictive value, and 95% negative predictive value in detecting LN metastases.Conclusion
The low sensitivity of ICG for the detection of LN metastases highlights why FLN dissection with ICG does not represent an alternative to ePLND.Micro-abstract
We prospectively assessed fluorescent sentinel lymph node (LN) detection with indocyanine green (ICG) for the detection of LN metastases and the ideal ICG dosing in patients undergoing robot-assisted radical prostatectomy. ICG had low sensitivity for the detection of LN metastases. Fluorescent sentinel LN dissection with ICG does not represent an alternative to a well-performed extended pelvic LN dissection.