Nerve autofluorescence in near-ultraviolet light markedly enhances nerve visualization in vivo

Background During surgery, surgeons must accurately localize nerves to avoid injuring them. Recently, we have discovered that nerves fluoresce in near-ultraviolet light (NUV) light. The aims of the current study were to determine the extent to which nerves fluoresce more brightly than background and vascular structures in NUV light, and identify the NUV intensity at which nerves are most distinguishable from other tissues. Methods We exposed sciatic nerves within the posterior thigh in five 250-300 gm Wistar rats, then observed them at four different NUV intensity levels: 20%, 35%, 50%, and 100%. Brightness of fluorescence was measured by fluorescence spectroscopy, quantified as a fluorescence score using Image-J software, and statistically compared between nerves, background, and both an artery and vein by unpaired Student's t tests with Bonferroni adjustment to accommodate multiple comparisons. Sensitivity, specificity, and accuracy were calculated for each NUV intensity. Results At 20, 35, 50, and 100% NUV intensity, fluorescence scores for nerves versus background tissues were 117.4 versus 40.0, 225.8 versus 88.0, 250.6 versus 121.4, and 252.8 versus 169.4, respectively (all p < 0.001). Fluorescence scores plateaued at 50% NUV intensity for nerves, but continued to rise for background. At 35%, 50%, and 100% NUV intensity, a fluorescence score of 200 was 100% sensitive, specific, and accurate identifying nerves. At 100 NUV intensity, artery and vein scores were 61.8 and 60.0, both dramatically lower than for nerves (p < 0.001). Conclusions At all NUV intensities >= 35%, a fluorescence score of 200 is 100% accurate distinguishing nerves from other anatomical structures in vivo.