We recently reported the first noninvasive, label free measurement of pH in a bodily fluid in vivo using only Raman spectra i.e. in vivo rat model measurements probing the immediate vicinity of a contusive spinal cord injury (SCI) in the first minutes and hours after injury. Calibrated and assigned using Raman spectra of authentic materials, in the rat model we were not able to sample the cerebrospinal fluid (CSF) to allow comparison with an independent measurement of the pH. Swine presents a better model because they allow physical sampling of CSF, although still not ideal for our purposes. We were only able to physically sample CSF from the fourth cerebral ventricle of 2 different animals, before and after all spectral measurements on cords were completed. One measurement each for 2 different animals on physically sampled CSF averaged a pH of 7.001±0.106 (N=2) as per standard laboratory instrumentation. Using a dynamic analysis and the Henderson-Hasselbalch equation, the average of (N=12) noninvasive Raman-based pH measurements of CSF was 7.073±0.156 and at >95% confidence there is no statistically significant difference between the Raman-based and the physically sampled results. We discuss the difference between the dynamic and static analysis, the implications for our understanding of SCI, the accuracy, precision, calibration, general applicability of this approach and future work.
|