Repeated measurement variation and precision of laser Doppler flowmetry measurements

The repeated measurement variation of laser Doppler flowmetry (LDF) recordings is essential for the precision level of the method in the assessment of tissue perfusion. The objective of this paper was to discuss the practical consequences of this problem by presenting a statistical method which can be used to estimate the number of replicates needed to reach a certain precision standard. By a repeated measures analysis of variance with a single-factor design, the mean variation of repeated measurements and its standard deviation were estimated. This estimate was used for simulation of a 95% confidence interval with length defined as a percentage of the mean of repeated measurements. The analyses were made in LDF samples performed in skin, gastric mucosa, and pig kidney in order to exemplify the use of the method. Paired values gave an unacceptable precision estimate in all tissues, but by increasing the number of replicates, the estimated precision was greatly enhanced. A preliminary recommendation for the practical use of LDF in the assessment of tissue perfusion is to perform at least four to six repeated measurements. Further studies are needed in order to establish methodological standards. The presented statistical considerations could also be relevant for other procedures used in microvascular research.