Limitation of indocyanine green fluorescence in identifying sentinel lymph node prior to skin incision in cutaneous melanoma

Background Although several reports have discussed the potential of the indocyanine green (ICG) fluorescence technique, its effectiveness and limitations have not been fully evaluated. For the accurate detection of sentinel lymph node (SLN) in cutaneous melanoma, which can arise in a wide variety of sites throughout the body, it is important to detect the SLN sites prior to skin incision. The purpose of this study was to evaluate the ICG fluorescence technique in the preoperative setting. Patients and methods We classified the detection process of SLN biopsy into three surgical stages: 'before skin incision,' 'after skin incision,' and 'ex vivo.' We retrospectively reviewed 93 node fields in 86 patients who underwent SLN biopsies using ICG and conventional techniques. Results The detection rate using ICG at the 'before skin incision,' 'after skin incision,' and 'ex vivo' surgical stages were 63.4, 98.9, and 100 %, respectively. At the 'before skin incision' surgical stage, while lymphoscintigraphy (but not ICG) detected 33 node fields, ICG (but not lymphoscintigraphy) detected only four node fields. The detection rates varied from 100 % in the groin region drained from trunk to 15.8 % in the axillary region from upper limb. The factors that influenced the failure of ICG to identify hot SLNs were an axillary node field (p < 0.001) and a high body mass index (BMI) (p = 0.046). Conclusion For reliable identification of SLNs in cutaneous melanoma, the ICG fluorescence technique may offer little benefit for patients with axillary node field drained from upper limb, or high BMI.