9:00 - 10:30 Spectroscopy as an Aid in Clinical Prodedures: Clinical Chemistry
Many of the pathophysiological parameters measured in ICU patients can alter quickly and dramatically from very abnormal towards normal during treatment and vise versa during deterioration of the clinical condition. From a practical point of view the following continuous measurements will be of great value to the clinician and of benefit for the patient.
The great interest of SERS in the biomedical applications stems mainly from three facts. One is the very high enhancement of the Raman intensity of the absorbed molecules by a factor of 103-106. The SERS effect becomes even more intense if the frequency of the excitation is in resonance with a major absorption band of the molecule being illuminated (surface-enhanced resonance Raman Scattering; SERRS). Another is marked reduction in fluorescence background which often is a serious obstacle in Raman studies of biomedical materials. Third is the surface selectivity which SERS effect provides; only molecules on or very near the metal surface can contribute to the signal in SERS spectra.
We propose here new enzyme-immunoassay based upon surface-enhanced Raman scattering (SERS). In the proposed system, antibody immobilized on a solid substrate reacts with antigen which binds with another antibody labeled with peroxidase (POD). If this immuno-complex is subjected to the reaction with orthophenylenediamine and hydrogen peroxide at 37 °C, azoaniline is generated as reaction product. This azo-compound is adsorbed on silver colloid. In this system only azo-compound gives strong surface-enhanced resonance Raman (SERRS) spectrum. The spectrum shows intense bands at 1582 and 1442 cm-1 due to the C=C and N=N stretching modes, respectively. A linear relationship was observed between the peak intensity of the N=N stretching band and the concentration of antigen, revealing that one can determine the concentration of antigen by the SERRS mesasurement of the reaction product. The correlation coefficient between the peak intensity and the concentration was calculated to be 0.999 for the concentration range from 0.1 to 5.1 ng/ml. The detection limit of this SERS enzyme-immunoassay method was found to be about 10-15 mol/mL which was lower by one-order than by previously reported method employing SERS.
A novel method for the interpretation of mid-infrared spectra of human blood is presented. The so-called "disease pattern recognition (DPR)" method is based on the idea, that � instead of investigating specific molecules within the spectrum � each spectrum is evaluated as a whole.
For the analysis, one microliter of serum is deposited on a disposable and dried. An infrared spectrum of the dried sample is recorded in the range from 500 cm-1 to 5000 cm-1. The spectra of diseased and healthy volunteers are compared by sophisticated mathematical methods. So far, the regularized discriminant analysis as well as binary classifiers appear to be the most promising evaluation method. In the course of this work we have measured more than 4000 samples and evaluated four different types of patterns, namely diabetes, metabolic syndrom, rheumatoid arthritis and healthy. We find that the spectra have pronounced differences in absorption. Furthermore the variation of the spectra among healthy volunteers on average is much smaller than for diseased volunteers.
As an example, the above picture shows the difference in absorption between the serum spectra of healthy volunteers vs. patients with diabetes (type II). Here, the distinctions - which normally are analyzed by highly specialized mathematical tools � are even directly observable by eye in many regions of the spectra.
For comparison we furthermore have quantified the reliability of the DPR method. The first results are very promising.