Background:
Alterations of endothelial cells characterize endothelial dysfunction (ED) a well-established response to cardiovascular risk factors. ED precedes the development of atherosclerosis including epicardial cardiovascular disease (CVD) and coronary microvascular disease (MVD). However to date, detailed knowledge of differential characteristics of coronary-bed specific endothelial cells (hCAECs) during this process is missing due to the lack of a reliable technique to isolate these cells. Therefore, this study sought to establish a comprehensive ex vivo technique (LiqVasc), which should enable hCAECs isolation and characterization during routine invasive coronary procedures. 

Methods:
50 patients (mean age 71.1 yrs, 28% female), who underwent coronary intervention by PCI or invasive assessment of coronary (micro-) circulation by CFR/IMR measurements, were included into this study. To isolate hCAECs attached to the wire within the coronary circulation, wires were removed from the guiding catheter at the end of the procedure and transferred into a special cell dissociation buffer. Cells were mechanically rinsed off and quantified using a multicolor antibody panel for flow cytometry (FACS). 

Endpoint of the study was the number of successfully isolated hCAECs assessed by FACS. To assess a potential impact of the used coronary wire type on the EC isolation success standard workhorse PCI wires (group PCI: Asahi Sion blue; Boston Scientific SAMURAI) were compared to wires allowing functional coronary assessment by FFR/CFR/IMR (group PHY: Abbott Pressure Wire X; Boston Scientific COMET Pressure guidewire) within this study. 

Results:
LiqVasc allows for successful isolation of a mean of 137 (±84) viable cells with endothelial phenotype (CD146⁺). Interestingly, the hCAEC-yield was significantly (p<0.001) different among the investigated wire categories with a significantly lower number of hCAEC in PCI wires (100 ± 80) as compared to PHY wires (212 ± 79). In a multiple univariable regression model the investigated wire type was the only significant (R²=40.9%; p=0.008) factor on the hCAEC-yield as compared to other patient- or disease specific factors as age, sex, risk factors, vessel time and diameter, presentation as ACS or CCS, that had no impact on the isolated hCAEC numbers. Therefore, these data suggest a wire-based effect on the LiqVasc method, that is further investigated in mechanistical bench studies.

Conclusion:
We successfully established an innovative, easy-to-use model for sampling of hCAECs ex vivo. First results suggest an association between the coronary wire type and the isolated number of hCAECs. After successful establishment, LiqVasc represents a novel method to incorporate hCAEC as a disease-specific biomarker in clinical routine, which has the potential to personalize effective preventive measurements for endothelial protection.