Retinal Blood Flow and Markers of Vascular Inflammation and Endothelial Dysfunction in Type 2 Diabetes
MetadataShow full item record
Abnormal leukocyte adhesion (i.e. leukostasis) to retinal vascular endothelial cells occurs in early diabetes. The processes of leukostasis have been clearly demonstrated in the vascular endothelium of patients with diabetes. In non-proliferative DR, clinical outcomes are manifested by excessive permeability from inflammatory progression leading to inner blood retinal barrier disruption, endothelial cell damage and widespread capillary nonperfusion. Diabetes promotes vascular leakage in DR by upregulation of adhesion molecules. Moreover, many of the pathological changes in NPDR are related to abnormalities in retinal blood flow. Studies have shown that specific circulating markers of inflammatory activity and endothelial dysfunction are associated with clinical signs of diabetic retinopathy. However, few have found an association between circulating levels of inflammatory and endothelial dysfunctional markers and abnormal retinal hemodynamics in patients with non-proliferative DR. The specific aims of this thesis are as follows: (Chapter 3)To correlate baseline levels of inflammatory and endothelial dysfunction markers and 1) baseline retinal arteriolar hemodynamics and 2) any disturbance in retinal hemodynamics over 6-month time in terms of vessel diameter, blood velocity, maximum-to-minimum velocity ratio and volumetric flow. In Chapter 4: To correlate circulating levels of inflammatory and endothelial dysfunction markers and 1) baseline vascular reactivity and 2) any disturbance in vascular reactivity after 6-month time in terms of vessel diameter, blood velocity, maximum-to-minimum velocity ratio and volumetric flow in patients with increasing non-proliferative diabetic retinopathy (NPDR) severity. Methods for Chapter 3: Diabetes subjects were stratified into either mild-to-moderate (Group 2) or moderate-to-severe (Group 3) NPDR based on their retinopathy status. Age-matched non-diabetics were recruited as controls (Group 1). Forearm blood sample was collected to determine baseline levels of inflammatory and endothelial dysfunctional markers. At visit 1, baseline retinal hemodynamics was acquired using Canon Laser Blood Flowmeter. Patients returned for a visit 2 (6 month follow-up visit) and retinal hemodynamics was reassessed. Baseline levels of inflammatory and endothelial dysfunctional markers compared between groups and correlated with both baseline and change in retinal hemodynamic parameters over 6-month time. For Chapter 4: Diabetes subjects were stratified into either mild-to-moderate NPDR or moderate-to-severe NPDR based on their retinopathy status. Age-matched non-diabetics were recruited as controls. At visit 1, forearm blood sample was collected to determine levels of inflammatory and endothelial dysfunctional markers and baseline vascular reactivity response was acquired. Retinal blood flow data was acquired while subjects breathed air. Retinal blood flow measurements were then acquired after exposure to isocapnic hyperoxic stimuli. At visit 2 (6 month follow-up), retinal vascular reactivity was reassessed. Baseline levels of inflammatory and endothelial dysfunctional markers compared between groups and correlated with both magnitude of baseline and change in vascular reactivity in terms of retinal hemodynamics. Results of Chapter 3: Maximum-to-minimum velocity ratio (max: min) was found to be significantly elevated in the group 3 compared to group 1 at baseline (0.72 vs. 0.49, after Bonferroni correction P<0.01). Both sICAM-1 and sE-selectin were significantly elevated as a function of group (ANOVA p=0.02 and p=0.04). A post hoc Bonferroni test showed that Group 3 had significantly higher in both sICAM-1 and sE-selectin levels compared to Group 1 (234.0 vs. 151.5 ng/ml, P=0.02 and 53.4 vs. 27.6 ng/ml, P<0.01, respectively). Hemoglobin A1c was significantly elevated across the groups (ANOVA p<0.01). A post hoc Bonferroni test showed that Group 3 had significantly higher hemoglobin A1c level compared to Group 1 (7.9 vs. 5.6 % , P<0.01). There were no significant associations found between baseline markers of inflammation and baseline retinal hemodynamics across all groups. The Δ velocity was correlated with the baseline sICAM-1 (r=0.42, p=0.02) and A1c levels (r=0.37, p=0.04) in patients with NPDR. After adjustment for all other variables (A1c, hsCRP and vWF), Δ velocity, sICAM-1 and A1c were found not to be reliable predictors of baseline retinal hemodynamics. For Chapter 4: There were no significant differences in magnitude of retinal vascular reactivity in hemodynamic parameters between groups at visit 1 or visit 2. Over 6 months time, compliance was found to be significantly reduced in patients of Group 3 compared to Group 2 (-0.4 vs. 0.1, t-test p<0.01). Both sICAM-1 and sE-selectin were significantly elevated as a function of group (ANOVA p=0.02 and p<0.01). A post hoc Bonferroni test showed that Group 3 had significantly higher in both sICAM-1 and sE-selectin levels compared to Group 1 (243.4 vs. 157.3ngml, P<0.01 and 57.0 vs. 29.3 ng/ml, P<0.01, respectively). Hemoglobin A1c was significantly elevated across the groups (ANOVA p<0.01). A post hoc Bonferroni test showed that Group 3 had significantly higher hemoglobin A1c level compared to Group 1 (8.8 vs. 5.6 % , P<0.01). Baseline VR in blood velocity weakly correlates with sE-selectin (r=0.31, p=0.04) across all groups while sVCAM-1 was associated with VR in terms of blood flow (r=-0.62, p<0.01) in patients with mild-to-moderate NPDR. The ∆ blood flow after 6 months was found to be weakly associated with sE-selectin (r=0.46, p=0.03) across all groups. Finally, the ∆ blood velocity after 6 month time was found to be moderately correlated with baseline vWF Ag level (r=-0.78, p=0.02). Multiple regression analysis found that vascular inflammatory and endothelial function markers had weak predictive power for Δ hemodynamic parameters. Conclusions Chapter 3: We found weak associations between circulating markers and baseline or the disturbance in retinal hemodynamics after 6 months time. Overall, we found both an increase in rigidity of the arteriolar circulation and elevated inflammatory adhesion markers (sICAM-1 and sE-selectin) within the same population sample. Change in velocity over the follow-up period was correlated with sICAM-1 and A1c levels in patients with NPDR but the level of association was such that neither sICAM-1 nor A1c proved to reliably predict retinal hemodynamics. Finally, in Chapter 4 we demonstrated two important characteristics in early NPDR; 1) a disturbance in vascular reactivity in terms of compliance and 2) an increase in systemic markers of inflammation were found in patients with NPDR. Although systemic markers of vascular inflammation and endothelial dysfunction are not predictive of hemodynamic parameters, our study found moderate associations between baseline and disturbances in VR after 6 months time. Therefore, there is evidence that inflammation and vascular function may be related with respect to their development in NPDR.