Page 227 - Vibrational Spectroscopic Imaging for Biomedical Applications

Raman Detection of Car otenoids in Human T issue 203

the carotenoid C=C Raman response and the fluorescence back-
ground is high enough (up to ~0.25) that it is easily possible to quan-
tify the amplitudes of the C=C peak after digital background
subtraction. This step is automatically accomplished by the instru-
ment’s data processing software, which approximates the background
with a fourth-order polynomial, subtracts the background from the
raw spectrum, and displays the final result as processed, scaled
spectrum in the right panel of the computer monitor, as shown in
Fig. 7.5b. MP carotenoid RRS spectra measured for the living human
macula were indistinguishable from corresponding spectra of pure
lutein or zeaxanthin solutions measured with the same instrument.
While the fundus-camera-interfaced Raman instrument is well
suited for measurements of elderly subjects, subjects with macular
pathologies, and also research animals, we found that simplified
instrument versions can be used for healthy human subjects, provided
they have good visual acuity (with or without correcting lenses) and
are able to self-align on a fixation target prior to a Raman measurement.
An example for a particularly simple self-alignment instrument con-
figuration is a version in which the CCD/spectrograph combination is
replaced with a single photomultiplier/filter combination. 22
In order to cross calibrate different instrument versions, we con-
structed a simple tissue phantom consisting of a lens and a thin, 1 mm
path length, cuvette placed in the focal plane of the lens, and measured
the RRS response for preset lutein and zeaxanthin solutions with opti-
cal densities in the range 0.1 to 1.0, a range that at the higher end
exceeds typically encountered physiological concentration levels. An
example of a calibration curve for a particular instrument version is
shown in Fig. 7.5c. It demonstrates a linear RRS response up to a rela-
tively high-optical density of 0.8. Besides for cross calibration purposes,
this calibration method can be used to correlate the RRS response of a
subject’s MP with its corresponding optical density value.
An example for clinical RRS measurements of a relatively young
subgroup (33 eyes), ranging in age from 21 to 29 years, is shown in
Fig. 7.6a. A striking observation is the fact, that the RRS-measured MP
levels can vary drastically between individuals (up to ~10 fold differ-
ence). Since the ocular transmission properties of all anterior optical
media in this age group can be assumed to be very similar, the varia-
tions must be attributed to strongly varying MP levels. Subjects with
extremely low-carotenoid levels may be at higher risk of developing
macular degeneration later in life.
When measuring a large population of normal subjects, none of
whom were consuming nutritional supplements with containing sub-
stantial amounts of lutein or zeaxanthin, we found a striking decline of
average macular carotenoid levels with age, 20,23 as shown in Fig. 7.6b.
Part of this decline can be explained by “yellowing” of the crystalline
lens with age, which would attenuate some of the illuminating and