Systemic microvascular rarefaction is correlated with dysfunction of late endothelial progenitor cells in mild hypertension: a substudy of EXCAVATION-CHN1

Abstract

Background: Hypertension often presents with microvascular rarefaction (MVR), which is closely associated with impaired angiogenesis. Early detection of MVR is essential for systemic assessment in patient with hypertension. We aimed to determine the systemic MVR through both optical coherence tomography angiography (OCTA) and intravital capillaroscopy, and to investigate their respective efficacies and related mechanisms associated with late endothelial progenitor cells (LEPCs) dysfunction.

Methods: Seventy-one hypertensive and sixty-nine age-match normotensive subjects were included in this study. All subjects received intravital capillaroscopy for skin capillary density (SCD) and OCTA for retinal capillary density (RCD) and non-perfused areas (R-NPA). Subsequently, correlation of LEPCs activities and microvascular rarefaction were examined.

Results: Compared with normotensive subjects, hypertensive patients had significantly lower RCD [(52.9 ± 2.9)% vs. (57.8 ± 1.6)%, P < 0.01] and higher R-NPA [(0.12 ± 0.07) mm2 vs. (0.053 ± 0.020) mm2, P < 0.01]. SCD correlated positively with RCD but negatively with R-NPA [(RCD: OR = 0.40, 95% CI 0.25-0.67, P < 0.01); (R-NPA: OR = 0.39, 95% CI - 0.0029 to 0.0011, P < 0.01)]. The discriminative powers of RCD performed best (AUC 0.79 versus SCD AUC 0.59, P < 0.001) followed by R-NPA (AUC 0.73 versus SCD AUC 0.59, P < 0.001) for systolic blood pressure. Similar pattern is also found for diastolic blood pressure (RCD AUC 0.80 versus SCD AUC 0.54, P < 0.001; R-NPA AUC 0.77 versus SCD AUC 0.54, P < 0.001). Furthermore, LEPCs tube formation was impaired in hypertensive patients (36.8 ± 2.3 vs. 28 ± 3.7, P < 0.01). After multivariate adjustments, positive correlation existed between RCD or R-NPA with LEPCs tube formation (RCD: β = 0.64, 95% CI 0.34-0.91, P < 0.01; R-NPA: β = - 24.67, 95% CI - 43.14 to - 4.63, P < 0.05) but not with SCD (β = 0.082, 95% CI 0.01-0.18, P = 0.085).

Conclusion: Compared to intravital capillaroscopy, OCTA is a more precise technique for early detection of hypertensive microvascular rarefaction, which is associated with the fall in LEPC-mediated angiogenesis. Both of OCTA and LEPCs function can help identify hypertension-related capillary abnormality. Trail Registration The trial is a substudy of EXCAVATION-CHN1, registered at clinicaltrials.gov as NCT02817204 (June 26, 2016).

Keywords: Hypertension; Late endothelial progenitor cells; Microcirculation; Microvascular rarefaction; Optical coherence tomography angiography.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

Quantification and representative photographs of SCD, RCD, R-NPA in normotension and hypertension. A Quantification of SCD at rest and after 2-min VC in both groups; B representative photographs of SCD at rest and after VC in both groups. Hypertensive SCD (b, d) was reduced than the normotensive (a, c) in the rest state and after 2-min VC; C quantification of RCD in both groups; D representative photographs of retinal vessel map and flow density map. Both hypertensive retinal vessel map (b) and flow density map (d) decreased than that in normotension (a, c); E quantification of R-NPA in both groups; F Representative photographs of R-NPA. Yellow zone represented non-perfused areas. Representative photographs of retinal NPA in hypertension (b) increased than that in normotension (a). (B 0.66 mm2 per field, magnification 136, scale bar 100 μm; D and F optical axial resolution, 5 microns; optical transverse resolution, 15 microns). VC venous congestion, RCD retina capillary density, R-NPA retinal non-perfused areas

Fig. 2

Correlation and receiver operating characteristic (ROC) analysis about results of intravital capillaroscopy and OCTA. A Scatter plots of correlation between RCD and SCD (a) and scatter plots of correlation between R-NPA and SCD (b). B ROC curve for SCD and RCD, R-NPA. ROC was performed for blood pressure (definition of high blood pressure), n = 140. a with SBP analysis, AUC for SCD is 0.68, 0.72 for RCD and 0.79 for R-NPA; b with DBP analysis, AUC for SCD is 0.64, 0.81 for RCD and 0.78 for R-NPA. OCTA optical coherence tomography angiography, SCD skin capillary density, RCD retina capillary density, R-NPA retina non-perfused areas, AUC area under the curve

Fig. 3

Effects of hypertension on LEPCs functions. A Quantification of migration, adhesion and tube formation ability of LEPCs. LEPCs functions decreased in hypertension (P < 0.01 vs. normotension); B representative photographs of LEPCs functions. a, b Representative photographs of LEPCs migration; c, d representative photographs of LEPCs adhesion; e, f representative photographs of LEPCs tube formation (magnification 100, scale bar 100 μm). Number of per hp for migration and adhesion, number of per mm2 for tube. LEPCs late endothelial cells

Fig. 4

Correlation analysis of SCD, RCD/R-NPA and LEPCs tube formation. a Scatter plots of correlation between SCD and LEPCs tube formation; b scatter plots of correlation between RCD and LEPCs tube formation; c scatter plots of correlation between R-NPA and LEPCs tube formation. SCD skin capillary density, RCD retina capillary density, R-NPA retina non-perfused areas, LEPCs late endothelial cells