Near-infrared (NIR) light penetrates relatively deep into skin, but its usefulness for biomedical imaging is constrained by high scattering of living tissue. Previous studies have suggested that treatment with hyperosmotic "clearing" agents might change the optical properties of tissue, resulting in improved photon transport and reduced scatter. Since this would have a profound impact on image-guided surgery, we seek to quantify the magnitude of the optical clearing effect in living subjects. A custom NIR imaging system is used to perform sentinel lymph node mapping and superficial perforator angiography in vivo on 35-kg pigs in the presence or absence of glycerol or polypropylene glycol:polyethylene glycol (PPG:PEG) pretreatment of skin. Ex-vivo, NIR fluorescent standards are placed at a fixed distance beneath sections of excised porcine skin, either preserved in saline or stored dry, then treated or not treated with glycerol. Fluorescence intensity through the skin is quantified and analyzed statistically. Surprisingly, the expected increase in intensity is not measurable either in vivo or ex vivo, unless the skin is previously dried. Histological evaluation shows a morphological difference only in stratum corneum, with this difference being negligible in living tissue. In conclusion, topically applied hyperosmotic agents are ineffective for image-guided surgery of living subjects.