Comparison of laser speckle contrast imaging with laser Doppler for assessing microvascular function

OBJECTIVE: To compare the inter-day reproducibility of post-occlusive reactive hyperaemia (PORH) and sympathetic vasomotor reflexes assessed by single-point laser Doppler flowmetry (SP-LDF), integrating-probe LDF (IP-LDF) and laser speckle contrast imaging (LSCI), and the spatial variability of PORH assessed by IP-LDF and LSCI. We also evaluated the relationship between IP-LDF and LSCI perfusion values across a broad range of skin blood flows. METHODS: Eighteen healthy adults (50% male, age 27 ± 4 years) participated in this study. Using SP-LDF, IP-LDF and LSCI, indices of skin blood flow were measured on the forearm during PORH (1-, 5- and 10-min occlusions) and on the finger pad during inspiratory gasp and cold pressor tests. These tests were repeated 3-7 days later. Data were converted to cutaneous vascular conductance (CVC; laser Doppler flow/mean arterial pressure) and expressed as absolute and relative changes from pre-stimulus CVC (ΔCVC(ABS) and ΔCVC(REL), respectively), as well as normalised to peak CVC for the PORH tests. Reproducibility was expressed as within-subjects coefficients of variation (CV, in %) and intraclass correlation coefficients. RESULTS: The reproducibility of PORH on the forearm was poorer when assessed with SP-LDF and IP-LDF compared to LSCI (e.g., CV for 5-min PORH ΔCVC(ABS)=35%, 27% and 19%, respectively), with no superior method of data expression. In contrast, the reproducibility of the inspiratory gasp and cold pressor test responses on the finger pad were better with SP-LDF and IP-LDF compared to LSCI (e.g., CV for inspiratory gasp ΔCVC(REL)=13%, 7% and 19%, respectively). The spatial variability of PORH responses was poorer with IP-LDF compared to LSCI (e.g., CV ranging 11-35% versus 3-16%, respectively). The association between simultaneous LSCI and IP-LDF perfusion values was non-linear. CONCLUSION: The reproducibility of cutaneous PORH was better when assessed with LSCI compared to SP-LDF and IP-LDF; probably due to measuring larger skin areas (lower inter-site variability). However, when measuring sympathetic vasomotor reflexes on the finger pad, reproducibility was better with SP-LDF and IP-LDF, perhaps due to the high sensitivity of LSCI to changes in skin blood flow at low levels.