Abnormal Flow-Mediated Vasodilation in Normal-Tension Glaucoma Using a Noninvasive Determination for Peripheral Endothelial Dysfunction

Abstract

purpose. To assess peripheral vascular endothelial function in patients with normal-tension glaucoma (NTG) by using a noninvasive method: endothelium-dependent flow-mediated vasodilation (FMD).

methods. Forty patients with NTG and 40 healthy age- and sex-matched normal control subjects underwent measurement of FMD and endothelium-independent nitroglycerin-mediated vasodilation (NMD) via high-resolution, two-dimensional (2D) ultrasonographic imaging of the brachial artery. The patients also underwent blood sampling for biochemistry, lipid profile, and high-sensitivity C-reactive protein (hsCRP) analysis.

results. The NTG group exhibited significant impairment of FMD compared with the control group (NTG: 2.64% ± 2.22%, control: 5.96% ± 2.50%, P < 0.001). Multivariate analysis demonstrated that the presence of NTG was the strongest independent predictor of FMD. The lipid profile and hsCRP did not differ significantly between the two groups.

conclusions. This study demonstrated impaired FMD in patients with NTG and the results provide evidence of a generalized peripheral vascular endothelial dysfunction in patients with NTG.

Glaucoma is a group of ocular diseases characterized by progressive thinning of the neuroretinal rim of the optic nerve head and loss of the retinal nerve fiber layer, together with a particular pattern of visual field loss. Although elevated intraocular pressure (IOP) is the most common risk factor for glaucomatous damage, the concept that IOP is the only relevant risk factor for glaucoma has been increasingly challenged. It was reported that approximately one third to one half of patients with primary open-angle glaucoma (POAG) consistently have IOPs within the normal range of less than 22 mm Hg. ^{1 2 3 4} These patients are defined as having normal tension glaucoma (NTG). ⁵ The pathogenesis of NTG remains uncertain. Vascular dysregulation and blood flow disturbances have been reported, as NTG is often accompanied by widespread cerebrovascular and systemic cardiovascular diseases. ⁶ Vascular endothelium, a barrier to interactions between plasma and vessels, is a complex organ with autocrine, paracrine, and endocrine properties that maintain vascular homeostasis. ⁷ The vascular dysfunction observed in patients with NTG can be a consequence of vascular endotheliopathy. ⁸ Several patients with glaucoma have increased plasma and aqueous levels of endothelin-1. ^{9 10} Glaucomatous optic neuropathy (GON)–like damage can also be experimentally induced by endothelin-1 application. ¹¹ Nitric oxide (NO) and endothelin-1 are potent vascular regulators that play pivotal roles in the maintenance of vascular tone and reactivity. ^{12 13 14} Ischemia or vascular dysregulation occurs as a consequence of an imbalance between NO and endothelin-1. Maintaining such a balance requires intact endothelium. Isolated reports have identified compromised peripheral endothelial cell function in patients with NTG. ^{15 16} Endothelial function has been assessed with a variety of invasive and noninvasive methods. Brachial artery ultrasound assessment of endothelium-dependent, flow-mediated vasodilation (FMD) is a well established and widely used noninvasive measure of endothelial function that is frequently applied to assess coronary artery disease (CAD) risk factors such as diabetes mellitus, hypertension, hypercholesterolemia, homocysteinemia, and heart failure. ^{17 18} The precise mechanism of FMD is not completely understood; however, it is generally believed that FMD is mediated by NO produced by the endothelial cells. This study assesses brachial artery FMD as an indicator of vascular endothelial function in patients with NTG. The purpose of this study was to describe the relationship between FMD and NTG. 

Material and Methods

Patient Group

Forty patients with NTG and 40 age- and sex-matched normal control subjects were recruited in the study. The diagnostic criteria for NTG were as follows: untreated IOP not exceeding 22 mm Hg measured at different times of the day from 8 AM to 5 PM, open anterior chamber angles on gonioscopy, glaucomatous optic disc cupping (cup-to-disc ratio >0.7 with thinning or notching of the neural rim), and characteristic optic nerve-related visual field loss on perimetry (Humphrey perimetry, using the 30-2 SITA standard program; Carl Zeiss Meditec, Dublin, CA). The definition of characteristic optic nerve-related visual field loss is as follows: (1) a pattern SD (PSD) worse than the normal 1% level, (2) a glaucoma hemifield test (GHT) “outside normal limits,” (3) one hemifield cluster worse than the normal 1% level, (4) two hemifield clusters worse than the normal 5% level, (5) four or more adjacent abnormal (P < 0.05) locations on the pattern deviation probability plot. ¹⁹ For all criteria, confirmation on a second visual field was required. Reliability should be <20% fixation losses, <33% false positives, and <33% false negatives. Exclusion factors included history of ocular trauma, eye surgery before diagnosis of glaucoma, or ocular manifestations other than glaucoma. Patients with systemic diseases such as hypertension, congestive heart failure, hypercholesterolemia, diabetes mellitus, cerebral vascular accident, or autoimmune disease were also excluded. None of the subject patients had migraine or Raynaud’s phenomenon. The 40 normal control subjects were recruited from a routine physical check-up group with normal ocular examinations. Both groups were referred to the vascular laboratory for vascular ultrasound. All patients gave informed consent to the study procedures, which were reviewed and approved by the institutional review board of Chang Gung Memorial Hospital. The study conformed to the tenets of the Declaration of Helsinki. 

Study Protocol

All patients underwent blood sampling for fasting serum glucose, lipid profile, creatinine, alanine aminotransferase (ALT), uric acid, and high-sensitivity C-reactive protein (hsCRP) before being referred to vascular ultrasound study. The FMD of the brachial artery was examined noninvasively by two-dimensional (2-D), high-resolution ultrasonic imaging as previously described. ¹⁸ 2-D images of the left brachial artery and pulsed-Doppler flow velocity signals were obtained (model L7 7.5-MHz linear array transducer on an Aspen ultrasound system; Acuson, Mountain View, CA). Imaging was performed in a dimly lit, quiet room with a temperature of 22°C to 25°C. Patients rested supine for at least 10 minutes before the first scan and remained supine until the final recording was acquired with electrocardiogram (ECG) continuous monitoring. Blood pressure was taken from the right arm before imaging. Images were obtained approximately 3 to 5 cm above the antecubital fossa. The transmit zone was set to the depth of the near wall with a depth field of 3 cm. The images were magnified by a resolution box function, so that the interface between the media and adventitia “m” line was well demonstrated. First, baseline 2-D images were acquired followed by pulsed-Doppler blood flow velocity with the signal at a 65° to 70° angle to the vessel lumen and the 1.0-mm-wide gate positioned at the center of the artery to record flow velocity and volume. To induce hyperemia, a 5.6-inch wide blood pressure cuff was inflated at the forearm to 250 mm Hg. Arterial occlusion was kept for 5 minutes with the transducer carefully maintained in an identical position (Fig. 1) . The cuff was then rapidly deflated and pulsed-Doppler velocity signals were recorded at 5 to 10 seconds after the deflation. The reactive hyperemia was determined by the mosaic change in the color-flow imaging and the increase in flow volume. At 60 seconds after cuff deflation, 2-D images of the brachial artery were recorded for 5 seconds. 

To determine the endothelium-independent, nitroglycerin-mediated vasodilation (NMD), sublingual nitroglycerin spray (400 μg) was administered after another 10 minutes of rest. Brachial artery scans were performed again at the same position 4 minutes later, after the nitroglycerin spray was administered. 

Image Analysis

Brachial artery diameter was measured by two observers who were unaware of the patients’ clinical details. Special care was taken to analyze identical segments by identifying anatomic landmarks. Measurements were taken from the anterior to the posterior “m” line at the end diastole, which coincided with the R-wave on ECG (Fig. 2) . The diameter was calculated as the average of three end-diastolic frames. Moreover, FMD was calculated as the percentage change in brachial artery diameter in response to hyperemia (FMD = [(VD _hyperemia − VD _rest)/VD _rest] × 100%, VD: vessel diameter). Similarly, NMD was calculated as the percentage change in brachial artery diameter in response to nitroglycerin (NMD = [(VD _{nitroglycerin} − VD _rest)/VD _rest] × 100%). Brachial artery blood flow volume at rest, during hyperemia, and after nitroglycerin spray was determined by the time average velocity of the four beats. 

Reproducibility

To determine the reproducibility of the imaging studies, 30 subjects were assessed in this study. The intraobserver and interobserver variability of brachial artery diameter determinants was obtained by comparing the baseline arterial diameter measurements. Linear regression analysis of the vessel diameter determinants revealed correlation coefficients of 0.996 and 0.992 between intraobserver and interobserver. The average arterial diameters were 3.99 ± 0.73 and 3.98 ± 0.73 mm, respectively. The mean differences between the measurements were 0.04 ± 0.06 and 0.06 ± 0.07 mm (1.13% ± 1.49% and 1.71% ± 1.88% of the vessel diameter). 

Statistical Analysis

Data are expressed as the mean ± SD for the continuous variables and percentage for the categorical variables. Continuous variables were compared between the NTG and control groups by unpaired Student’s t-test. Categorical variables between groups were compared by χ² test. In addition, multivariate analysis employed stepwise multiple linear regression models to identify individual and joint factors that influenced FMD. Correlations between variables were calculated by Pearson’s coefficient. A P < 0.05 was considered statistically significant. 

Results

Table 1lists the demographics of the NTG and control groups. Table 2lists the vascular parameters in the two groups. The percentage change in FMD was 2.64 ± 2.22 and 5.96 ± 2.50 in the NTG and control groups, respectively (P < 0.001). Meanwhile, the percentage change in NMD was 13.35 ± 4.80 in the NTG group and 14.60 ± 4.36 in the control (P = 0.230; Fig. 3 ). No statistically significant changes were observed in age, sex distribution, IOP, body mass index (BMI, body weight/(body height in m)²), heart rate, systolic and diastolic blood pressure, biochemistry data, lipid profile, and hsCRP. Multivariate analysis demonstrated that the presence of NTG was the strongest independent predictor of FMD (P < 0.001). 

Discussion

This study showed impaired FMD in the patients with NTG compared with the normal control subjects. To our knowledge, this is the first study to describe peripheral vascular endothelial dysfunction in patients with NTG using a noninvasive method (FMD). Henry et al. ¹⁵ and Buckley et al. ¹⁶ previously reported endothelial dysfunction in NTG. The former used venous occlusion plethysmography (VOP) and the latter used cutaneous arteries biopsies to analyze endothelial function. The findings of this study support the existence of an underlying generalized vascular endothelial dysfunction in the patients with NTG. 

The vascular endothelium plays a vital role in the control of blood flow. The short-term control of arterial tone and coagulation status and the long-term control of smooth muscle cell proliferation and extracellular matrix production are achieved via an intact vascular endothelium. Injury to the vascular endothelium is a frequent preliminary event in most, if not all, vascular disease. Two main methodologies are currently being used for assessment of in vivo endothelial function: VOP and FMD. VOP is a highly accurate and reproducible procedure that has been used for years. However, it involves arterial cannulation for vasoactive drug delivery, which is considered invasive, and thus has limited clinical application. FMD uses high-frequency ultrasonographic imaging of the brachial artery to assess endothelium-dependent flow-mediated vasodilation. Hyperemia induced by transient ischemia of the forearm increases shear stress of the brachial artery vessel wall, which provokes endothelial NO release and subsequently causing vasodilation. This process can be quantitated as an index of vasomotor function. ²⁰ The noninvasive nature of the technique permits repeated measurements and has high patient acceptance. FMD can be used as an indicator for clinical follow-up and response to medical treatment. However, FMD is vulnerable to criticisms of reproducibility and intra- and interobserver variability. An International Brachial Artery Reactivity Task Force has been appointed to investigate and minimize these problems. ¹⁸ The theory and methodology behind FMD measurement is now well established, and FMD becomes a widely used method of determining peripheral vascular function. ^{21 22 23 24} All the FMD examinations in this study were performed by an experienced cardiologist (WJH). The intra- and interobserver variability was minimal and comparable to that in previous reports. ^{23 24}  

This study excluded patients with systemic diseases that might impede FMD results. None of the patients studied had received cardiovascular medication. Both groups had similar age, gender, and BMI. The presence of NTG was the strongest predictive factor for FMD. This study conducted blood sampling for hsCRP and found no significant difference between the two groups. This finding implied that the FMD impairment in the patients with NTG did not result from vascular inflammation. 

The measurement of peripheral FMD has been reported to be a good indicator of coronary endothelial function. ²⁵ FMD was reduced in patients with CAD risk factors and improved after risk-reduction treatment. Certain drugs such as hydroxymethylglutaryl coenzyme-A reductase inhibitors (HMG–CoA reductase inhibitors, statins) and calcium channel blockers (CCBs) can improve FMD. ²⁶ Nifedipine has been reported to improve forearm vascular endothelial function, and the same effect has been observed in coronary vessels. ^{27 28} Patients with NTG share various risk factors with patients who have CAD. ^{29 30 31 32} In glaucoma, CCBs have been reported to be beneficial in patients with NTG because of their vasodilatory effect, which improves ocular hemodynamics and visual function. ^{33 34 35 36} This beneficial effect of CCBs and their role in improving vascular endothelial function is unclear. Furthermore, whether improvement of FMD influences the course of NTG requires further study. 

Direct evidence of local ocular endothelial dysfunction is difficult to obtain. The impairment of endothelial function of the brachial artery in patients with NTG observed in this study indicated a systemic rather than a local vascular effect. Whether such a dysfunction exists in ocular circulation and its relationship to the progression of the disease remain uncertain. 

In conclusion, the data presented demonstrated significant FMD impairment in patients with NTG. However, the relationship between peripheral vascular endothelial dysfunction and NTG presence and progression is unclear. The noninvasive and reproductive nature of forearm FMD evaluation provides an opportunity to assess and follow-up vascular endothelial function. A prospective, randomized, case-controlled study is needed to clarify whether medical treatment that improves FMD can change the course of NTG. 

 Submitted for publication January 10, 2006; revised February 24, 2006; accepted May 25, 2006.

 Disclosure: W.-W. Su, None; S.-T. Cheng, None; T.-S. Hsu, None; W.-J. Ho, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Ultrasound imaging of the brachial artery with lower cuff placement and transducer positioned above the antecubital fossa.

Figure 2.

 

2-D high-resolution ultrasound image of the brachial artery at (A) baseline, (B) 60 seconds after cuff release, and (C) 4 minutes after nitroglycerin sublingual spray (FMD = 2.13% and NMD = 17.02% in this case). Vessel lumen was measured between the arrows.

Table 1.

 

Baseline Characteristics in Patients with NTG and Control Subjects

	NTG	Control	P
Number of cases	40	40
Age (y)	50.15 ± 12.08	48.80 ± 11.92	0.616
Sex (male, %)	62.5	50.0	0.367
IOP (mm Hg)	14.16 ± 3.07	13.78 ± 3.35	0.591
Body mass index (kg/m2)	24.20 ± 3.48	24.92 ± 3.14	0.332
Heart rate (beats/min)	70 ± 12	73 ± 11	0.284
Systolic pressure (mm Hg)	117 ± 13	121 ± 15	0.180
Diastolic pressure (mm Hg)	74 ± 9	77 ± 10	0.211
Serum glucose (mg/dL)	94 ± 11	94 ± 9	0.718
Creatinine (mg/dL)	0.94 ± 0.18	0.93 ± 0.19	0.866
ALT (U/L)	21 ± 10	18 ± 8	0.277
HDL (mg/dL)	53 ± 14	52 ± 11	0.766
LDL (mg/dL)	107 ± 28	114 ± 29	0.323
Cholesterol (mg/dL)	183 ± 32	180 ± 31	0.721
Triglyceride (mg/dL)	114 ± 48	108 ± 41	0.525
Uric acid (mg/dL)	5.81 ± 1.60	6.10 ± 1.37	0.419
hsCRP (mg/dL)	1.27 ± 1.73	1.33 ± 2.01	0.888

 Data are the mean ± SD. ALT, alanine aminotransferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; hsCRP, high-sensitivity C-reactive protein.

Table 2.

 

Vascular Parameters in Patients with NTG and Control Subjects

	NTG	Control	P
Number of cases	40	40
Baseline vessel diameter (mm)	4.12 ± 0.74	4.08 ± 0.71	0.842
Baseline TAV (m/s)	0.44 ± 0.29	0.46 ± 0.31	0.739
Baseline flow volume (mL/min)	306.31 ± 146.04	329.84 ± 187.87	0.534
Reactive hyperemia (%)	233.44 ± 83.00	232.23 ± 73.85	0.945
Hyperemic vessel diameter (mm)	4.22 ± 0.73	4.32 ± 0.68	0.511
Hyperemic TAV (m/s)	0.68 ± 0.45	0.59 ± 0.37	0.369
Hyperemic flow volume (mL/min)	478.71 ± 277.30	489.71 ± 298.73	0.865
FMD (%)	2.64 ± 2.22	5.96 ± 2.50	<0.001
Nitroglycerin vessel diameter (mm)	4.64 ± 0.73	4.67 ± 0.74	0.856
Nitroglycerin TAV (m/s)	0.76 ± 0.48	0.66 ± 0.45	0.379
Nitroglycerin flow volume (mL/min)	676.89 ± 337.14	655.93 ± 481.79	0.822
NMD (%)	13.35 ± 4.80	14.60 ± 4.36	0.230

 Data are the mean ± SD. TAV, time average velocity.

Figure 3.

 

The percentage changes of endothelium-dependent FMD and endothelium-independent NMD in NTG and control.