Indocyanine green (ICG) is a fluorescent molecule [1]. Incident infrared light of wavelength 780 nm provokes detectable photon emission at 820–830 nm [1]. Alongside a high definition camera and software imposed pseudo-colour, intravenously delivered ICG may be used to identify vessel perfusion and differentiate tissue density [2–4]. Indocyanine green was initially developed in the 1955 by Kodak photography, and received FDA approval in 1959 [5, 6]. Indocyanine green has a favourable safety profile, with an adverse event rate of 0.34 % including nausea, vomiting, and rarely shock [1 in 300,000]) [7]. The use of ICG is established in ophthalmology, dermatology, and cardiology for vascular identification [8, 9]. This article will summarise the current and future applications of ICG in robotic urology, with the da Vinci® robot (dVSS), (Intuitive Surgical Systems, Sunnyvale, CA, USA) with emphasis placed on the intraoperative identification of vascular and oncological tissue. Notably, ICG is a cheap consumable, and infrared endoscopic equipment is widely available in contemporary dVSS systems (Firefly® in the DaVinci ** for nephron sparing [20], and precise dissection of the prostatic neurovascular bundle [21].

In healthy renal parenchyma the transporter bilitranslocase binds ICG and healthy tissue appears isofluorescent when perfused with ICG laden blood [22]. Renal tumour is deficient in bilitranslocase and therefore appears hypofluorescent [19]. Tobis et al. noted hypofluorescent renal tumours in the presence of ICG during robotic partial nephrectomy (RAPN). The investigators were guided by ICG in selective arterial clam** of a handful of cases. Following this, Manny et al. subsequently calculated hypofluorescent tissue had a sensitivity of 84 % and positive predictive value of 87 % for malignant renal lesions in 100 RAPN cases [23]. Borofsky et al. then described super selective renal artery clam** in 27 patients undergoing partial nephrectomy, with ICG, thus avoiding clamps to healthy renal parenchyma. Maximal loss of eGFR at 3 month follow-up was 1.6 versus 14.0 % in the non-ICG study arm [24]. Bjurlin et al. also reported a comparatively favourable 6.2 % decrease in glomerular filtration rate at 2 weeks postoperatively through the use of super-selective arterial clam** assisted with ICG [20, 25]. Patients undergoing super selective clam** showed significantly improved renal function at follow-up, however, these studies were underpowered and did not examine the impact of the technology upon intraoperative decision making [25].

Robot-assisted radical prostatectomy (RARP) may be supplemented by ICG imaging to identify the prostatic neurovascular bundle. In 2015, Patel et al. at demonstrated 30 % of prostatic neurovascular dissections were revised during nerve sparing RARP [21]. Given the degree of nerve spare correlates to functional outcomes [26], ICG holds promise for improving post-prostatectomy continence and erectile function. Robot-assisted sentinel node harvest is performed at RARP to determine metastatic nodal status. A hybrid fluorescent ICG radiotracer was optimised to detect sentinel lymph nodes in robot-assisted lymph node dissection, and improved in vivo identification through fluorescence to 93.5 % versus 50.0 % in non-optimised samples (n = 38; p = 0.005) [18]. In another series, robotic ICG assisted sentinel node harvest during RARP yielded a sensitivity of 100 % and negative predictive value of 100 % (n = 50), however, this method was non-specific. With larger sample sizes, it is likely the diagnostic coefficients would fall to below 100 % [27]. In spite of this, the potential utility of ICG for metastatic node detection and differentiation of oncological tissue remains encouraging.

Early work by Moore in revising surgical margins based on fluorescence, and the radiolabelling of fluorescent dye to localise oncological tissue was visionary. Recent technological advances in cancer biomarkers and immunology have prompted the hybridisation of ICG with cancer selective ligands, to localise tumour by fluorescence. Prostate specific membrane antigen (PSMA) is upregulated in prostate cancer by 100 to 1000-fold [28, 29]. Although at the pre-clinical stage, ICG bound PSMA-ligand was demonstrated to detect PSMA positive prostatic tumours in mice by Nakajima et al. using the humanised monoclonal antibody specific to PSMA, J591 [30]. Notably the J591 antibody has been delivered in several human clinical trials at high doses with a favourable safety profile [31–33]. In mice with human cell line prostate cancer, tumour was visually identifiable through fluorescence from 1 to 10 days after administration [30]. The accurate identification of previously indiscernible cancer tissue should be expected to transform positive surgical margin rates, and delay biochemical recurrence in prostate cancer. Similar advances in renal and bladder cancer immunology may also allow transfer of the technology to treat these pathologies.

In summary ICG combined with the dVSS is an incumbent game changer, and has been described as a “hammer looking for a nail” [25]. Evidence is emerging that ICG is useful in the robotic surgeon’s arsenal and yields tangible clinical benefit from selective arterial clam** in RAPN. Well-designed randomised controlled trials are urgently needed to quantify improvements in outcomes following ICG-augmented robotic procedures such as sentinel node excision, and identification of oncological tissue. It remains to be proven whether ICG brings small, incremental procedure-specific advances, or clinically substantial benefits to overall survival and functional recovery to these procedures. A revolutionary step in operative imaging though the labelling of fluorescent molecules to enhance en bloc tumour resection was espoused nearly 70 years ago by neurosurgeon Dr George Moore. Although at the pre-clinical stage, it is only a matter of time before Moore’s pioneering work is assimilated into routine robotic surgical practice.