A new method for using a non-selectively filled hollow-core photonic crystal fiber (HC-PCF) as a sensitive
Raman spectroscopy platform suitable for biosensing applications is presented. A 1550 HC-PCF was
completely filled with ethanol (core and cladding holes). Using a 785nm excitation laser, the Raman spectrum
of ethanol in the fiber core was obtained and compared with the equivalent Raman spectrum of an ethanolfilled
cuvette. Using a relatively short 9.5cm length of HC-PCF, a Raman signal enhancement factor of 40 over
a bulk solution of ethanol was observed under the same excitation conditions. The small sample volume
utilized and longer interaction length provides the potential for compact, sensitive, and low-power Raman
sensing of biological materials
In a step towards the development of improved long-term prognostic indicators for patients with end-stage renal disease, we utilized absorption spectroscopy to determine the baseline status of whole blood in a cohort of 5 clinically-stable hemodialysis patients. The optical absorption spectrum of pre-dialysis and post-dialysis blood samples in the 400-1700nm wavelength range was measured for the cohort over a four-week period. Absorption spectra were consistent over time, with a maximum coefficient of variation (CV) of absorption under 2% (650-1650nm) for any given patient over the four-week period (pre and post-dialysis). Spectra varied by a greater amount across patients, with a maximum CV of 5% in any given week. Analysis of variance indicated a broad spectral range (650-1400nm) where within-patient spectral variation was significantly less than between-patient variation (p<0.001), providing the potential for development of stable baseline blood status indicators. The spectra were investigated using principal component analysis (PCA) including a further set of whole blood absorption spectra obtained from 4 peritoneal dialysis patients. PCA revealed the fingerprint-like nature of the blood spectrum, an overall similarity of the spectrum within each treatment mode (hemodialysis or peritoneal dialysis), and a distinct spectral difference between the treatment modes.
Visible and near infrared transmission and diffuse reflection spectroscopy were used to monitor changes in whole blood resulting from hemodialysis treatment for end-stage renal disease. Blood samples from 8 patients on chronic hemodialysis therapy were measured in the 500- to 1700-nm wavelength range immediately before and after a single treatment. Principal component scores characteristic of each spectrum were derived, and mean pre- and posttreatment scores of the first principal component indicated a significant treatment-dependent change in both optical transmission (P=0.004) and diffuse reflection (P<0.001). Significant treatment-induced change persisted (P<0.05) when the first four principal components were used to account for >97% of the treatment-dependent spectral variation. Some blood spectral changes expressed in terms of difference spectra (posttreatment – pretreatment) were consisent with standard clinical indicators of weight reduction, urea reduction, and potassium change, with probable origins at a molecular level. The results indicate the feasibility of using optical transmission and diffuse reflection spectroscopy to characterize clinically relevant blood changes for the future development of more comprehensive indicators of hemodialysis efficacy and long-term clinical outcomes. Moreover, the optical techniques employed are adaptable for potential online monitoring of blood changes during the hemodialysis treatment.