Variability in Sublingual Microvessel Density and Flow Measurements in Healthy Volunteers

Abstract

Objectives: As sublingual microvascular indices are increasingly heralded as new resuscitation end-points, better population data are required to power clinical studies. This paper describes improved methods to quantify sublingual microvessel flow and density in images obtained by sidestream dark field (SDF) technology in healthy volunteers, including vessels under 10 µm in diameter.

Materials and Methods: Measurements of sublingual capillary density and flow were obtained by recording three 15-second images in 20 healthy volunteeers over three days. Two independent observers quantified capillary density by using two methods: total vessel length (mm/mm²) and counting (number/mm). Both intraoral and temporal variabilities within subject and observer reproducibilities were determined by using coefficients of variability and reproducibility indices.

Results: For small (1–10 µm), medium (11–20 µm), and large (21–50 µm) diameter, mean vessel density with standard deviations (SDs) in volunteers was 21.3(±4.9), 5.2 (±1.2), and 2.7 (±0.9) mm/mm², respectively. Also, 94.0±1.4% of small vessels, 94.5±1.4% of medium vessels, and 94.5±4.0% of large vessels had continuous perfusion. Within subjects, the means of all measurements over three days varied less than 13, 22, and 35% in small, medium, and large vessels, respectively. Interobserver reproducibility was good, especially for capillary (1–10 µm) density and flow measurements.

Conclusions: Our methods of microvessel flow and density quantification have low observer variability and confirm the stability of microcirculatory measurements over time. These results facilitate the development of SDF-acquired sublingual microvascular indices as feasible microperfusion markers in shock resuscitation.

• sepsis
• sublingual microcirculation
• SDF technology
• capillaroscopy
• capillary density

Acknowledgements

The authors would like to thank all the volunteers for their participation in this study and are also grateful to Roy Powell, head of the Research and Development Unit (RDSU) at the Royal Devon and Exeter Hospital, for his statistical advice and analysis.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.