Quick non-invasive confocal screen for diabetic neuropathy


Figure 1a

Figure 1a
Confocal images of human skin from normal (left) and diabetic (right) subjects. Nerves in epidermis appear bright white. Images are a projection of 22 sections of 2µm thickness acquired with the Bio-Rad Laser Scan Confocal Microscope. Image acquisition time approximately 1 minute/image or 22 minutes per stack.


Figure 1b

Figure 1b
Confocal images of human skin from normal (left) and diabetic (right) subjects. Nerves in epidermis appear bright white. Images are a projection of 22 sections of 2µm thickness acquired with the CARV Non-Laser Scan Confocal Microscope. Image acquisition time approximately 250 msec/image or 5.5 seconds per stack.


Figure 2

Figure 2
Confocal image of an immunostained sectioned blister. A. Nerves (protein gene product 9.5 immunoreactivity) are green or yellow, basement membrane (type IV collagen-immunoreactivity) is red. Epidermis has separated from dermis just above the dermo-epidermal basement membrane. Scale bar = 200µm. B. Dermal capillaries and the subepidermal neural plexus remained intact. Scale bar = 100µm. C. Epidermal nerves, severed from their proximal segment, remained in the blister roof. Scale bar = 50µm.
Click here to see a higher resolution image (190K).


Figure 3a

Figure 3a
Blister normal (left) vs. diabetic (right) Images from BioRad Laser Scan Confocal Microscope.


Figure 3b

Figure 3b
Blister normal (left) vs. diabetic (right) Images from CARV Non-Laser Scan Confocal Microscope.


Figure 4

Figure 4
Software is used to trace nerve fibers through a confocal z-series. Each image is comprised of up to 30 optical sections, each 2µm thick. The nerves are traced in three dimensions from the point where they cross the dermal-epidermal basement membrane to their endings in the superficial layers of the epidermis. Data is acquired for the number of nerve fibers and the length and number of branch points for each nerve fiber.


Figure 5

Figure 5
Comparison of clinical neurological evaluation scores with epidermal nerve fiber length per volume of epidermis and sweat gland nerve volume in type I diabetic subjects. The amount of nerve present in calf-epidermis correlates well with the clinical evaluation of diabetic subjects when score values are less than 30. Subjects with the most nerve have the best (i.e. lowest) clinical scores; when scores are greater than 60, nerve is usually absent. Sweat gland innervation remains within the normal range until neuropathy is quite severe (clinical score greater than 60).


Figure 6

Figure 6
Confocal images of blister roof, left low magnification (scale bar = 500µm) survey image of a 3mm blister immunostained for nerve with antibody to protein gene product 9.5. Rectangles indicate the size and ideal locations of areas to be used for nerve counts. Right-higher magnification (scale bar = 100µm) image is used for quantification of epidermal nerve fibers (from right rectangle in A).


[ Return to Kennedy Lab Home Page ]