3 SPUTUM i BILE
Figure 41. Non-linear map of the pyrolysis mass spectra (only centroids shown) of 20 different Klebsiella isolates from six patients (arabic numerals) in the same hospital. Previous analysis by conventional biotyping failed to differentiate these strains and thus the presence of a "hospital epidemic" involving six patients was concluded. Py-MS analysis, however, indicated possible cross-infections between patients 1 and 2, as well as between patients 3 and 4, only. Re-examination of the isolates of patients 5 and 6 by biotyping, showed the presence of typing errors in the first run. The isolate from patient 5 proved to be a slightly different biotype whereas the isolate from patient 6 was an altogether different species (Enterobaoter).
Fast biological profiling of single drops of urine without complex sample preparation, as shown in Figure 42, is one of the newest and most promising applications of Py-MS in medicine (refs. 158, 159). Even if this approach may not be expected to detect all abnormalities detectable by a combination of other, more specific methods, such as GC or GC-MS analysis of urine extracts (refs. 160, 161), the speed and simplicity of the Py-MS technique in combination with the ease of statistical evaluation may bring biochemical profiling of urine samples as a general screening technique closer to its practical realization. A similar approach has given promising results with other body fluids such as bile (refs. 158, 159) and synovial fluids (ref. 162). Experiments with serum and spinal fluid samples also seem possible and preliminary investigations in chese areas are in progress. Py-GC/MS studies (electrically heated filament) of untreated urine samples for detecting metabolic abnormalities have recently been reported by Roy (ref. 163).
Was this article helpful?