The aim of our study was to establish the usefulness of ICG for mapping the superficial lymphatic network of the breast. The results of our work, which shows lymphatic diffusion of ICG in 53 of the 60 injections performed, demonstrated the feasibility of identifying mammary lymphatic vessels using ICG injections.
We assumed that the images obtained corresponded to the lymphatic network and not to the arteriovenous system because of previous anatomical studies showing that the latter does not overlap with the lymphatic network [21–24]. Thus, the mapping of the mammary arteriovenous system would be significantly different from the lymphatic system and would not produce superimposable images [25–27].
ICG is now a common tool for the exploration of the arteriovenous system [28–30]. Its use for lymphatic network exploration benefits from the cadaveric study of Shinaoka et al [31] who mapped the lymphatic network of the limbs after subcutaneous injection of the dye in the hands and feet. This anatomical study was then confirmed in vivo by the work of Unno et al. [32, 33]. However, ICG has a low radiation intensity which limits its use for the detection of deep lymphatic vessels because infrared cameras cannot detect radiation deeper than 2 cm [32]. The results obtained in this work are therefore limited to the superficial lymphatic network. Subsequently, the use of ICG seems to be of less interest in obese subjects or in areas with a large amount of fat such as the breast. Therefore, it doesn’t have any use for the detection of deep lymph nodes when searching for the sentinel lymph node.
In order to overcome these limitations, Yamashita et al suggest to apply a transparent material to the study area to enhance the detection of radiation by vitro pressure [34]. Schaafsma et al. suggest the combination of ICG and technetium to form a hybrid tracer providing greater spatial resolution [8].
We chose to include both male and female subjects in this study. Although anatomical studies of the male breast are rare in the literature, they show that the vascularity of the breast is similar between men and women [26, 35]. In addition, axillary prosthetic breast augmentation surgery is equally applicable to women for cosmetic purposes as it is to men during sex reassignment surgery. The topographical analysis of the superficial lymphatic vessels revealed by the ICG injection shows that they are predominantly distributed in the superior-external quadrant, corresponding to the axillary zone. This analysis is consistent with previous anatomical studies of the mammary lymphatic network [1, 11, 14, 36]. Previous anatomical studies describe two lymphatic networks in the breast: superficial and deep [10, 11, 14]. Our findings strongly suggest the predominance of the superficial lymphatic network of the breast, which seems to be the main contributor to lymphatic drainage to the axillary lymphnode, confirming Sappey’s findings. Sappey described a primary centripetal lymphatic drainage to the subareolar plexus via the deep network, then, secondarily, the lymph travels to the superficial network to reach the axillary lymph nodes [36, 37].
We noticed during the experiments that the vessels belonging to the upper external quadrant run in the direction of the axillary fossa, without actually ever reaching it.
This phenomenon confirms various studies [38–41] reporting an anatomical zone located behind the lateral edge of the pectoralis major, near the lower pole of the axillary fossa, where the lymph nodes responsible for breast drainage are concentrated: the "Soft Tissue Triangle" (Fig. 6 and Fig. 7). This area is located at a distance from the axillary approach, which is used for prosthetic augmentation.
As this is an original study of mammary lymphatic circulation in the cadaveric subject, there is no previous work regarding the optimal amount to be injected. We based our study on the results of Murawa et al [6] who showed that the level of fluorescence obtained was directly correlated to the dose injected. Therefore, as this was a cadaveric study, we chose to inject a whole dose of infracyanin, i.e. 25 mg of product, in order to potentiate the fluorescence capabilities of the ICG. This amount may need to be adjusted for in vivo lymph nodes detection.
The study of lymphatic circulation in anatomical subjects suffers from an important limitation related to the absence of tissue perfusion pressure. In vivo, lymph flow through vessels is governed by both intrinsic factor, the spontaneous contractility of endothelial cells and extrinsic factors: muscle contraction, perfusion pressure, respiratory amplitude for thoracic lymph vessels [42]. We based our protocol on various previous works using massage to enhance ICG diffusion in cadavers [17, 18, 31, 33] and described passive diffusion time after dye injection during which the diffusion of ICG within the lymphatic vessels was related solely to the mechanical pressure induced by the syringe. Massaging of the injection site was performed afterwards to theoretically reproduce the effect exerted by the cardiac pump under physiological conditions.
However, several limitations to this study should be noted. Amongst the 60 breast injections performed, only 53 were usable. There was some heterogeneity among the subjects. The sex ratio was unbalanced (10 men and 20 women), and the state of preservation was not identical amongst all subjects. The study included only 8 "fresh" subjects, i.e. deceased for less than 72 hours and never frozen.
The 7 excluded breasts in which it was not possible to detect lymphatic vessels were characterized by the formation of a diffusion "cloud" at the dye injection site (Fig. 4). Two hypothesis can be proposed: a degradation of the mammary lymphatic network in the cadaver or a handling mistake during the injection leading to a deeper penetration of the dye not allowing its absorption by the subareolar plexus of the anatomical subjects.
Interestingly, one of the excluded breasts belonged to a subject in which a satisfactory result was obtained on the contralateral breast, inciting us not to neglect the possibility of a handling mistake.