Investigation of applicability of a mid-infrared spectroscopic method using an attenuated total reflection accessory and a new near-infrared transmission method for determination of faecal fat

• Published in: Annals of clinical biochemistry Vol. 38; no. Pt 3; p. 256
• Main Authors: Volmer, M, Kingma, A W, Borsboom, P C, Wolthers, B G, Kema, I P
• Format: Journal Article
• Language: English
• Published: England 01-05-2001
• Subjects: Calibration; Chemistry, Clinical - instrumentation; Chemistry, Clinical - methods; Feces; Humans; Lipids - analysis; Regression Analysis; Reproducibility of Results; Spectrophotometry, Infrared - methods
• ISSN: 0004-5632
• DOI: 10.1258/0004563011900489

In many laboratories, the titrimetric method of Van de Kamer is used for the analysis of faecal fat content of patients suspected of steatorrhoea. We investigated the applicability of a mid-infrared (MIR) spectroscopic method, using an attenuated total reflection (ATR) accessory, and a new near-infrared (NIR) spectroscopic method. For the NIR method, sealed plastic bags containing the stool samples were used as transmission cells. Standardization was obtained using a previously described MIR method, with a NaCl flow-cell, as reference method. Partial least-squares regression was used for the calibration of each method. Full cross-validation of the calibration set was used for the internal validation of each method. Fifteen per cent of the stool samples could not be estimated with the ATR method within reasonable accuracy limits compared with the reference. The standard error of prediction of the NIR method was 1.1 g/dL. We conclude that the new NIR method is a promising technique for routine use. However, further experiments need to be done with triplicate measurements of each sample and the use of an external validation set.