The degree of vascular dysfunction and contributing physiological mechanisms can be assessed in patients with chronic kidney disease by measuring brachial artery flow-mediated dilation, aortic pulse-wave velocity, and vascular endothelial cell protein expression.
Patients with chronic kidney disease (CKD) have significantly increased risk of cardiovascular disease (CVD) compared to the general population, and this is only partially explained by traditional CVD risk factors. Vascular dysfunction is an important non-traditional risk factor, characterized by vascular endothelial dysfunction (most commonly assessed as impaired endothelium-dependent dilation [EDD]) and stiffening of the large elastic arteries. While various techniques exist to assess EDD and large elastic artery stiffness, the most commonly used are brachial artery flow-mediated dilation (FMDBA) and aortic pulse-wave velocity (aPWV), respectively. Both of these noninvasive measures of vascular dysfunction are independent predictors of future cardiovascular events in patients with and without kidney disease. Patients with CKD demonstrate both impaired FMDBA, and increased aPWV. While the exact mechanisms by which vascular dysfunction develops in CKD are incompletely understood, increased oxidative stress and a subsequent reduction in nitric oxide (NO) bioavailability are important contributors. Cellular changes in oxidative stress can be assessed by collecting vascular endothelial cells from the antecubital vein and measuring protein expression of markers of oxidative stress using immunofluorescence. We provide here a discussion of these methods to measure FMDBA, aPWV, and vascular endothelial cell protein expression.
Chronic kidney disease (CKD) is a major public health concern that has reached epidemic proportions, affecting ~11.5% of the population in the United States alone1. The risk of cardiovascular death or a cardiovascular event in patients with CKD is significantly increased compared with the general population2-4. Although patients with CKD exhibit a high prevalence of traditional cardiovascular risk factors, this only explains part of their increased incidence of cardiovascular disease (CVD)5. Vascular dysfunction is an important nontraditional cardiovascular risk factor gaining increased recognition in the field of nephrology6-9.
While many changes likely contribute to the development of arterial dysfunction, among those of greatest concern are the development of vascular endothelial dysfunction, most commonly assessed as impaired endothelium-dependent dilation (EDD), and stiffening of the large elastic arteries10. Various techniques exist to assess EDD and large elastic artery stiffness, but the most commonly used are brachial artery flow-mediated dilation FMDBA and aortic pulse-wave velocity (aPWV), respectively. Another commonly used technique to assess EDD is measuring forearm blood flow response to pharmacological agents such as acetylcholine using venous occlusion plethysmography11,12. However, this methodology requires catheterization of the brachial artery, which is more invasive than FMDBA and may be contraindicated in patients with CKD. An alternate technique to assess arterial stiffness is to measure the local arterial compliance (the inverse of stiffness) of the carotid artery, although this is not as widely used or validated with clinical endpoints as aPWV13 .
Patients with CKD demonstrate both impaired FMDBA14-16 and increased aortic pulse-wave velocity aPWV13,17,18, even prior to needing dialysis. Importantly from a clinical perspective, both of these noninvasive measures of vascular dysfunction are independent predictors of future cardiovascular events and mortality both in patients with CKD19-21, as well as in other populations22-26. These techniques can be applied to studying various populations at risk of CVD, including patients with CKD.
The exact mechanisms by which arterial dysfunction develops in CKD are incompletely understood; however, reduced nitric oxide (NO) bioavailability is a critical contributor27-30 and a common mechanism of both impaired EDD and increased arterial stiffness10,31. In CKD, oxidative stress is increased and contributes to the reduction in NO bioavailability32-34. Oxidative stress is defined as excessive bioavailability of reactive oxygen species (ROS) relative to antioxidant defenses. Physiological stimuli, including inflammatory signaling, promote oxidant enzyme systems (e.g., the oxidant enzyme NADPH oxidase) to produce ROS, including superoxide anion (O2●–)35. Production of superoxide ultimately leads to reduces bioavailability of nitric oxide (NO).
Impaired NO bioavailability may in turn contribute to the development of CKD, as endothelial dysfunction is an independent predictor of incident CKD36. This is consistent with animal data indicating that eNOS inhibition induces hypertension (systemic and glomerular), glomerular ischemia, glomerulosclerosis, and tubulo-interstitial injury37. Indeed, reduced NO bioavailability appears necessary for the development and progression of experimental kidney disease that mimics human disease, suggesting a key role for endothelial dysfunction in human CKD38,39.
Markers of vascular oxidative stress can be assessed in vascular endothelial cells collected from human research subjects, using a technique originally developed by Colombo et al.40 and modified Seals et al41-43. Using 2 sterile J-wires, cells are collected from the antecubital vein, recovered, fixed, and later positively identified as endothelial cells and analyzed for expression of proteins of interest using immunofluorescence.
We provide here a discussion of this methodology that can be used to a) measure FMDBA; b) measure aPWV; c) measure vascular endothelial cell protein expression of markers of oxidative stress. The focus is on patients with CKD, not requiring chronic dialysis.
This protocol follows the guidelines of the Colorado Multiple Institutional Review Board (COMIRB).
1. Preparation for Testing Session
2. Collection and Processing of Vascular Endothelial Cells
3. Assessment of FMDBA and aPWV
4. Preparing Human Umbilical Vein Endothelial Cell (HUVEC) Control Slides
5. Staining of Vascular Endothelial Cells
6. Imaging and Analysis of Vascular Endothelial Cells
FMDBA is quantified as the peak change in diameter of the brachial artery following reactive hyperemia. Thus, the diameter at rest is compared to the diameter following the end of a 5 min blood pressure cuff occlusion period (Figure 1). Panel A shows a representative ultrasound image of the brachial artery, and Panel B displays a graph of the R-wave gated change in diameter from cuff release to 2 min following, as obtained using commercially available software. As the change is often quite minimal (In Figure 1 the change is 4.8%), small differences in measurement can have large impacts on results. Use of commercially available automated edge detection software is highly recommended to minimize bias and potential error in measurement44,45. As the stimulus for dilation during reactive hyperemia may differ between groups or conditions being compared, shear rate should be calculated using the Doppler blood flow velocities, and FMDBA should be adjusted for differences when applicable46,47.
aPWV is calculated with minimal operator input by most commercially available systems, including the NIHem used in our research. The R-wave of the ECG is compared to "foot" of the waveform at a given site and the time difference is calculated (Figure 2) for the carotid artery (Panel A) and the femoral artery (Panel B). The distance measurements are used in conjunction with the time differences to calculate a velocity. aPWV refers to velocity between the carotid artery to the femoral artery (i.e. along the aorta).
Immunofluorescent analysis of vascular endothelial cells can provide cellular evidence of the level of oxidative stress. To account for differences in staining intensity between staining sessions, the level of fluorescence of a given protein for each individual subject (representative images shown in Figure 3 Panel A) is compared to the fluorescence of the HUVEC control slide (representative images shown in Figure 3 Panel B). Thus, differences in protein expression can be compared either between groups or across conditions (e.g., during an intervention study).
Figure 1. Representative baseline brachial artery diameter obtained during assessment of brachial artery flow-mediated dilation (FMDBA). A) In a patient with chronic kidney disease (CKD). B) R-wave gated change in diameter from cuff release to 2 min following is shown graphically, as obtained using commercially available software. Please click here to view a larger version of this figure.
Figure 2. Representative results print-out from assessment of aPWV in a patient with CKD. A) Time delay from the R-wave of the ECG to the foot of the carotid artery, B) time delay from the R-wave of the ECG to the foot of the femoral artery (Tfoot), overlaid with the carotid waveform. Both panels also show the inputted distances from the suprasternal notch to the respective sites (represented by the letter D; in cm). The calculated aPWV value is shown in panel B (represented by the letters PWV; in cm/sec). Please click here to view a larger version of this figure.
Figure 3. Representative images of protein expression. A) DAPI (nuclear integrity; blue), VE Cadherin (positive endothelial cell identification; green) the oxidant enzyme NADPH oxidase (protein of interest; red) from cells collected from a patient with CKD B) and for a human umbilical vein endothelial cell (HUVEC) control slide.
Obtaining accurate results for FMDBA and aPWV requires acquiring high quality ultrasound images and pressure waveforms, respectively. Central to this is appropriate and continued training and use of each technique by the operator44. In addition, it is critical to control for as many external variables that may influence the results as possible by standardizing the testing session (e.g., prior 12 hr fast, climate controlled room, etc.)44,45. As mentioned above, the use of commercially R-wave gated acquisition software and edge-detection software is highly recommended to minimize bias and potential error in measurement44,45. When FMDBA is impaired, this could be either due to impaired NO release from the endothelium or due to impaired responsiveness of the vascular smooth muscles to the NO released. Sublingual nitroglycerin is administered to control for the responsiveness of the smooth muscle cell layer to an exogenous nitric oxide donor, in order to conclude that any impairment in FMDBA is specific to the capability of the vascular endothelium to produce nitric oxide44,45.
As the measurement is a velocity, accurate measurements of both distance and time are critical. The protocol we have described is based on the methodology employed in the Framingham Heart study24. Use of raised calipers rather than a tape measure improves accuracy of measurement of distance from the suprasternal notch to the femoral artery by taking a direct path rather than potential measuring over abdominal obesity. A clear "foot" of a clean waveform is absolutely necessary for calculation of the time difference from the R-wave of the ECG to the impulse at the measurement site (see Figure 2).
While alternate techniques are available to assess both endothelial function and arterial stiffness, FMDBA and aPWV are both commonly used in clinical research because they are non-invasive and well established as intermediary outcomes. In addition, they are well validated across various populations and are independently predictive of cardiovascular events and mortality19-26. Thus, they can be used as surrogate endpoints in clinical studies assessing the efficacy for an intervention to reduce cardiovascular risk in a given population, such as patients with CKD. Modification of these techniques are not required to specifically study patients with CKD, as compared to other populations at risk of CVD.
However, there are important limitations to both FMDBA and aPWV that merit discussion. FMDBA assesses vascular endothelial function of a large conduit artery (the brachial artery), thus does not provide an index of microvascular endothelial function. A separate technique using venous occlusion plethysmography is better suited to assess the latter. However, this methodology requires catheterization of the brachial artery, which is more invasive than FMDBA and may be contraindicated in patients with CKD. In addition, measurement of FMDBA requires lengthy and specific training in order to be performed well. aPWV provides an index of large elastic artery stiffness, which may differ from local arterial stiffness (such as the carotid artery). An alternate technique to assess arterial stiffness is to measure the local arterial compliance (the inverse of stiffness) of the carotid artery, although this is not as widely used or validated with clinical endpoints as aPWV13. In addition, the contribution of NO as a determinant of aortic stiffness may vary by vascular bed48. Last, there are potential confounds to the interpretation of both FMDBA and aPWV that need be measured and statistically adjusted for as appropriate, including baseline diameter and shear rate for FMDBA45, and heart rate and blood pressure for aPWV49.
An important consideration in the collection of vascular endothelial cells is minimizing blood on the J-wires and subsequently on the slides, such that the endothelial cells can be identified with minimal red blood cells overlapping in images. This can be achieved with training for proper technique as well as adequate washing when recovering the cells. When analyzing the slides, it is critical that fluorescence can be objectively quantified and the images are clear, without much background or overlap with other cells. Optimization of dilutions for staining and technique for microscopy analysis prior to analysis of study samples are key steps. Of note, the cell yield of this technique is ~600 vascular endothelial cells per collection, an insufficient amount of total mRNA is available to measure gene expression, thus limiting our probe to immunofluorescent staining of proteins of interest.
In addition to the techniques presented for assessing vascular oxidative stress, circulating or urine markers can be used to assess oxidative stress12,50. However, they may be less reflective of the level of oxidative stress specific to level of the vascular endothelium. Using these markers in conjunction with the presented techniques may provide the best indication of the overall level of oxidative stress.
We have provided an overview of methods that can be used to measure FMDBA, aPWV, and vascular endothelial cell protein expression. These techniques are appropriate not only for patients with CKD, but also in other populations at increased risk of cardiovascular disease. Collectively, they provide insight into vascular endothelial dysfunction, large elastic artery stiffness, and contributing physiological mechanisms, including oxidative stress.
The authors have nothing to disclose.
The authors thank Nina Bispham for her technical assistance. This work was supported by the American Heart Association (12POST11920023), and the NIH (K23DK088833, K23DK087859).
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
J-wire | St. Jude | 404584 | 2 per collection |
Disposable shorts (MediShorts) | Quick Medical | 4507 | |
Non-invasive hemodynamic workstation (NIHem) | Cardiovascualr Engineering | N/A | Includes custom ruler. An alternate system is the Sphygmocor |
Ultrasound | G.E. | Model: Vivid7 Dimension | We use a G.E., but there are many companies and models |
Vascular software (Vascular Imager) | Medical Imaging Applications | N/A | |
R-wave trigger box | Medical Imaging Applications | N/A | custom made |
Rapid Cuff Inflation System | Hokanson | Model: Hokanson E20 | |
Forearm blood pressure cuff | Hokanson | N/A | custom cuff with 6.5 x 34 cm bladder |
HUVECs | Invitrogren | C-015-5C | |
Donkey serum | Jackson | 017-000-121 | |
Pap pen | Research Products International | 195505 | |
VE Cadherin | Abcam | ab33168 | |
AF568 | Life Technologies | A11011 | depends on specifications of microscpe |
AF488 | Life Technologies | A11034 | depends on specifications of microscpe |
Nitrotyrosine antibody | Abcam | ab7048 | |
NADPH oxidase antibody | Upstate | 07-001 | |
DAPI | Vector | H-1200 | |
Delicate task wipe (Kimwipe) | Fisher Scientific | 06-666-A | |
Plastic paraffin film (parafilm) | Fisher Scientific | 13-374-10 | |
Confocal microscope | Olympus | Model: FV1000 FCS/RICS | many options exist |