What Is the Most Appropriate Methodology for Detection of Conduit Artery Endothelial Dysfunction?
Background— Use of upper-arm arterial occlusion to induce reactive hyperemia, and endothelium-dependent flow-mediated dilation (FMD) of the brachial artery, induces greater conduit vessel dilatation than lower-arm occlusion. However, brachial artery ischemia after upper arm arterial occlusion may make this approach unreliable. We studied whether upper or lower arm occlusions differ in their ability to detect endothelial dysfunction in cigarette smokers, and its improvement with an antioxidant strategy.
Methods and Results— Ten cigarette smokers with a >20 pack year history and 10 age- and gender-matched healthy controls participated in a 2-phase randomized controlled study of xanthine oxidase inhibition, using a 600-mg oral dose of allopurinol administered beforehand. Endothelium-dependent dilatation was assessed using ultrasound-Doppler after lower and upper arm occlusion. After lower arm occlusion, FMD was significantly impaired in smokers compared with controls (3.8±1.1% versus 8.7±2.2%; P=0.001). However, after upper arm occlusion, brachial artery dilatation in smokers was higher (11.8±2.7%; P<0.0001 versus lower arm) and did not differ from controls (9.4±2.9%; P=0.3). There was no difference in endothelium-independent dilatation to sublingual nitroglycerin between smokers and controls. Inhibition of xanthine oxidase with allopurinol improved lower arm FMD (3.8±1.1 to 10.1±1.9%; P<0.0001), but did not improve upper arm FMD (11.8±2.7 to 14.1±3.7%; P=0.4).
Conclusions— Although upper arm occlusion induces robust brachial vasodilatation, it cannot detect endothelial dysfunction induced by smoking or its improvement by inhibition of xanthine oxidase. The increase in brachial artery diameter with upper arm occlusion may be confounded by ischemia of the artery. Conduit artery FMD after release of lower arm occlusion appears to be a more valid method for assessment of endothelial function in humans.
Reactive hyperemia after lower arm (LA) arterial occlusion has been an established method for studying endothelium-dependent flow-mediated dilatation of the brachial artery in humans. However, the brachial artery dilatation induced by reactive hyperemia from lower arm occlusion is often modest, and alternative methods have been proposed to induce a more robust response. Use of upper arm (UA) occlusion to produce brachial artery dilatation was found to be comparable to resistance vessel endothelium-dependent responses in a validation study using strain gauge plethysmography.1 Subsequent studies showed that UA occlusion produces a greater degree of brachial vasodilatation than LA occlusion, and was thus proposed as a better means of evaluating endothelial function.2,3 In subjects with a single risk factor for atherosclerosis, UA occlusion was found to better distinguish subjects with risk factors than LA flow-mediated dilatation (FMD).4 The method has, therefore, been used in multiple studies to study endothelial function. However, in a study by Doshi et al in healthy subjects, inhibition of nitric oxide synthesis only partially inhibited brachial artery dilatation to UA occlusion, but abolished brachial dilatation to wrist occlusion, suggesting that lower arm occlusion may be a more valid method to study NO-dependent endothelial function.5 The International Brachial Artery Reactivity Task Force was unable to reach consensus as to which technique provided the most accurate or precise data.6 Given the large number of studies now using the FMD technique, new information on the best methodology for inducing reactive hyperemia is urgently needed. No previous study has explored which approach is more sensitive at detecting changes in endothelium-dependent dilatation induced by therapeutic intervention.
We performed this study to compare LA and UA occlusion to induce brachial artery FMD in evaluation of endothelial function in smokers. We have previously shown that xanthine oxidase contributes to endothelial dysfunction in both resistance and conduit vessels in heavy smokers.7 We therefore also explored whether there were differences between LA and UA occlusion in their ability to detect the effect of xanthine oxidase inhibition on endothelial function.
This study was performed in a subset of 28 human subjects who participated in a previously published study examining xanthine oxidase inhibition in resistance vessels and conduit vessels.7,8 The last 20 subjects recruited to the study (10 cigarette smokers and 10 age- and gender-matched nonsmoking control subjects), had lower followed by upper arm occlusion for testing of brachial artery flow mediated dilatation on each study session. No subject had clinical evidence of atherosclerosis. Age- and gender-matched healthy subjects who never had a history of tobacco use were selected for the control group. Inclusion criteria for the subjects were age 18 to 85 years, low-density lipoprotein cholesterol <130 mg/dL, total cholesterol <250 mg/dL, systolic blood pressure <130 mm Hg, diastolic blood pressure <80 mm Hg, and fasting blood glucose <126 mg/dL. Smokers were included if they had a >20 pack-year smoking history. The study was approved by the local Institutional Review Board and written informed consent was obtained from each subject.
Subjects participated in a single-blinded, randomized, two-phase crossover study. On one study day, subjects received allopurinol; the other day was a control session with no intervention. All subjects, both controls and smokers, were admitted to the General Clinical Research Center (GCRC) at least 12 hours before the study, to have a rigorous control over both diet and prevent acute smoking. None of the subjects were permitted to smoke from the time of admission to the GCRC. Studies were performed in a quiet room maintained at a constant temperature between 22 and 25°C. None of the subjects received vasoactive drugs in the week before the study.
Oral allopurinol was used to inhibit xanthine oxidase. On one day, a single dose of allopurinol (600 mg) was administered orally at 5:00 am and endothelial function assessed at 12:00 noon. Subjects were not blinded to the medication and took either allopurinol or nothing, and were instructed not to reveal their treatment allocation. GCRC research staff not involved in performing the vascular studies prepared the randomization code and administered the drug. The investigators who performed the vascular study and scored the data were blinded to the treatment until completion of the entire data collection. Treatment allocation was revealed to the investigators only after completion of studies on all subjects.
Allopurinol is converted by xanthine oxidase to oxypurinol, which has a much longer half-life (t½=17 to 21 hours) and mediates the majority of the effects of allopurinol.9 A 600-mg oral dose of allopurinol would achieve a mean oxypurinol concentration of 9 to 10 μg/mL, sufficient to inhibit >95% of xanthine oxidase.10 We performed the studies at least a week apart because 99.8% of oxypurinol is metabolized within 1 week. Normal xanthine oxidase activity is apparent 4 days after halting a one-week treatment course of allopurinol 300 mg.11
Conduit-vessel endothelial function was assessed by using ultrasound measurement of brachial artery diameter during changes in brachial artery flow (10 MHz linear array transducer, Biosound Esaote AU5). A 5-cm length of the brachial artery was imaged in a longitudinal section above the antecubital fossa and the optimal probe site on the skin marked. Baseline images of brachial artery diameter and Doppler velocities from the center of the vessel were recorded on videotape. While images for brachial artery diameter were being continuously recorded, a distal occluding forearm cuff placed just below the antecubital fossa was inflated to 50 mm Hg above systolic pressure for 5 minutes. The brachial artery scan was then obtained for 15 seconds before and 120 seconds after cuff deflation, including a repeat flow velocity recording for the first 15 seconds after cuff release. After 10 minutes, once basal diameter and flow were restored, a second baseline scan was obtained for 30 seconds, after which a proximal occluding upper arm cuff was inflated to 50 mm Hg above systolic blood pressure for 5 minutes and measurements taken. After a further 10 minutes of rest, nitroglycerin 400 μg was administered sublingually and vessel measurements made for 6 minutes. To exclude the possibility that always performing upper arm occlusion second might confound the results, we performed a validation study in 6 healthy nonsmoking control subjects. Flow-mediated dilatation in response to lower arm occlusion was performed twice in these subjects, separated by 10 minutes. There was almost identical brachial artery dilatation to the first (4.9±2.0%) and second lower arm occlusions (5.6±1.7%). The stimulus to dilatation (magnitude of the increase in blood velocity) was also similar between the two occlusions (85±20 versus 76±19%).
Brachial artery diameter and blood velocity measurements were analyzed by a trained sonographer. Digitized images of the brachial artery were used, and analysis performed using an automated neural network based method for complete near and far wall border detection (Brachial Analysis Tool Version 3.2.6, Medical Imaging Applications). For flow-mediated dilatation, a mean diameter was obtained from diameter measurements at 25 to 35, 50 to 60, and 110 to 120 seconds after cuff deflation, and a mean percentage change from baseline calculated. Peak Doppler flow velocity for hyperemia was measured in the first 15 seconds after cuff deflation. It was averaged over 5 consecutive cardiac cycles and a percentage change from baseline was calculated. Velocity rather than flow was calculated because the stimulus for conduit vessel dilatation is shear stress at the endothelial cell surface. For nitroglycerin responses, mean diameter and percentage change was obtained from measurements at 1, 2, 4, and 6 minutes after nitroglycerin administration; velocity was obtained at 6 minutes. Arterial pressure was measured in duplicate using a noninvasive automated oscillometric sphygmomanometer (Lifestat 200, Physio-Control). Heart rate was measured continuously using a lead II ECG.
All the laboratory assays were performed in the clinical chemistry laboratory of the University of Iowa Hospitals and Clinics. Blood uric acid levels were measured using the uricase reaction and quantified by a colorimetric technique. Blood levels of allopurinol and oxypurinol were measured 7 hours after oral allopurinol administration and were measured using high performance liquid chromatography with ultraviolet detection.
Power calculations were done with responses to reactive hyperemia as the primary end point, and we calculated that a sample size of 10 gave 84% power to detect a change of 50% in flow-mediated dilatation (estimated effect of cigarette smoking) at a significance level of 0.05 (SD=6%). To test differences between baseline variables between the two groups, we used Students two-sample t test. We used repeated-measures ANOVA to test for differences in brachial artery diameters, velocities, and percent change from baseline between the 2 groups. Data are expressed as mean±SE; and P<0.05 was taken as statistically significant.
Baseline variables were similar in both groups (Table 1), except for triglycerides and high-density lipoprotein cholesterol. Studies have shown that smoking independently decreases HDL and increases triglycerides significantly. Thus, we believe low HDL was likely a consequence of smoking than an independent additional risk factor.
Plasma oxypurinol levels 7 hours after oral allopurinol administration were 10.3±1.6 μg/mL in smokers and 8.5±0.6 μg/mL in controls. Allopurinol did not affect blood pressure or heart rate in both the groups. Baseline levels of uric acid were not significantly different between smokers and controls. Uric acid levels decreased significantly 7 hours after oral allopurinol in both the groups.
There was no significant difference in either peak blood flow velocity during hyperemia (115±15% with UA, 101±27% with LA in smokers) or in controls (Table 2). After LA occlusion, FMD was significantly impaired in smokers compared with controls (3.8±1.1% versus 8.7±2.2%; P=0.001). In contrast, after UA occlusion, FMD in smokers was significantly higher compared with LA FMD (11.8±2.7%; P<0.0001, Figure 1) and did not differ from controls (9.4±2.9%; P=0.3 versus smokers). There was no difference in endothelium-independent dilatation to sublingual nitroglycerin between smokers (9.4±2.4%) and controls (7.7±2.9%; P=0.3). Inhibition of xanthine oxidase with allopurinol improved LA FMD (10.1±1.9%; P<0.0001 versus control day), but did not improve UA FMD in smokers (14.1±3.7%; P=0.4 versus control day, Figure 2). No statistically significant effect of xanthine oxidase inhibition was seen in non-smoking control subjects, though FMD tended to be lower after allopurinol (Table 2).
Flow mediated dilatation of the brachial artery is being used in many studies, some very large with long-term prospective follow-up. Many investigators have been using upper arm occlusion because it produces a larger brachial dilatation, even though the procedure renders the brachial artery ischemic. We performed this study to test the validity of upper versus lower arm occlusion as a method to evaluate baseline and post-treatment conduit artery endothelial function. We found that lower arm occlusion was superior in detection of endothelial dysfunction, and also its improvement with an antioxidant strategy.
There is general agreement that brachial artery dilatation after UA occlusion is significantly greater than after LA occlusion, a finding that we confirmed. The larger brachial artery dilatation with UA occlusion is possibly secondary to higher wall shear stress and a larger territory involved below the occluded segment.12 Whether this represents endothelial NO-dependent generation or perhaps brachial artery ischemia has been controversial. In our study, use of LA occlusion demonstrated impaired brachial artery dilatation in smokers, but the dilatation induced by UA occlusion was not significantly different between smokers versus controls.
No previous study has compared UA and LA occlusion in detection of change in endothelial function with a therapeutic intervention. Inhibition of xanthine oxidase has been shown to improve both conduit and resistance vessel endothelial dysfunction.7,8,13,14 Previous conduit vessel studies testing xanthine oxidase inhibition used LA occlusion to induce reactive hyperemia. Importantly, we demonstrate here that LA occlusion is clearly superior to UA occlusion in detecting the conduit vessel endothelial benefits of xanthine oxidase inhibition. Thus, use of UA occlusion to induce reactive hyperemia and flow mediated dilatation failed to demonstrate both endothelial dysfunction and its improvement.
A few other studies have compared upper and lower arm occlusion as a way of inducing flow mediated dilatation of the brachial artery. In two studies done only in healthy subjects, FMD after UA occlusion was found to be significantly greater compared with LA occlusion and thus proposed as a better means of evaluating endothelial function.2,3 These two studies did not attempt to measure differences in NO-dependent endothelial function in subjects with atherosclerosis risk factors.
Two studies have been done comparing UA to LA occlusion in subjects with risk factors for atherosclerosis. In a study by Vogel et al, in subjects with a single risk factor for atherosclerosis, UA occlusion was found to produce a significantly greater hyperemia and FMD compared with LA occlusion. Compared with healthy controls, those with risk factors had a significantly lower FMD after UA occlusion.5 Another study by Dalli et al in subjects with multiple risk factors and acute myocardial infarction, UA occlusion produced a larger increase in FMD and was also significantly higher compared with LA FMD in healthy controls.15 However, this study did not test NTG-induced vasodilatation and hence it is uncertain how much of the dilatation after upper arm occlusion was entirely endothelium-mediated. In contrary to these reports, we did not demonstrate endothelial dysfunction in smokers using brachial dilatation after UA occlusion.
In another study in healthy subjects by Doshi et al using L-NMMA to inhibit NO synthase, though brachial artery dilatation after UA occlusion was greater than after LA, brachial dilatation could not be abolished with L-NMMA, whereas FMD after LA occlusion was abolished.5 This suggests the possibility that brachial artery dilatation after UA occlusion is not entirely attributable to increased shear stress, which should be NO dependent. The findings in our study support this observation, although in a different subject group. Our subjects (smokers) had a baseline endothelial impairment, in whom we tried to demonstrate improvement; whereas Doshi et al induced endothelial dysfunction in subjects with normal endothelial function.
It has also been suggested that increased brachial artery dilatation after UA occlusion is because of increased hyperemia.4 However, in our study blood velocity with hyperemia increased to similar extent in smokers and controls with both LA and UA occlusions, and thus it is unlikely that the results were confounded by differences in stimulus intensity. One possible limitation is that UA occlusion was performed after LA occlusion. This is unlikely to have affected the results because the arm was rested for an adequate period of time between occlusions with very similar baseline diameters and velocities. In addition, in validation studies we demonstrated that performing a second lower arm occlusion 10 minutes after the initial one resulted in an almost identical flow mediated dilatation of the brachial artery. Thus, it is unlikely that there was fatigue or preconditioning of the brachial artery following the first episode of flow-mediated dilatation.
Though UA occlusion is being used to evaluate endothelial function, our study shows that brachial artery dilatation after UA occlusion does not accurately detect either endothelial dysfunction or its improvement with xanthine oxidase inhibition. Use of more distal occlusion provides a more accurate assessment of endothelial function. The current widespread use of upper arm occlusion for testing endothelial function of the brachial artery needs to be reconsidered.
Sources of Funding
The authors’ research is supported by grants from the National Institutes of Health (NHLBI: HL58972; NCRR General Clinical Research Centers program: RR00059).
Consulting Editor for this article was Alan M. Fogelman, MD, Professor of Medicine and Executive Chair, Departments of Medicine and Cardiology, UCLA School of Medicine, Los Angeles, Calif.
Original received September 1, 2006; final version accepted December 21, 2006.
Berry KL, Skyrme-Jones AP, Meredith IT. Occlusion cuff position is an important determinant of the time course and magnitude of human brachial artery flow-mediated dilation. Clinical Science. 2000; 99: 261–267.
Doshi SN, Naka KK, Payne N, Jones CJ, Ashton M, Lewis MJ, Goodfellow J. Flow-mediated dilatation following wrist and upper arm occlusion in humans: the contribution of nitric oxide. Clinical Science. 2001; 101: 629–635.
Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D, Charbonneau F, Creager MA, Deanfield J, Drexler H, Gerhard-Herman M, Herrington D, Vallance P, Vita J, Vogel R. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery: A report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol. 2002; 39: 257–265.
Guthikonda S, Sinkey C, Barenz T, Haynes WG. Xanthine oxidase inhibition reverses endothelial dysfunction in heavy smokers. Circulation. 2003; 107: 416–421.
Murrell GAC, Rapeport WG. Clinical pharmacokinetics of allopurinol. Clin Pharm. 1986; 11: 343–353.
Cardillo C, Kilcoyne CM, Cannon RO 3rd, Quyyumi AA, Panza JA. Xanthine oxidase inhibition with oxypurinol improves endothelial vasodilator function in hypercholesterolemic but not in hypertensive patients. Hypertension. 1997; 30: 57–63.
Butler R, Morris AD, Belch JJ, Hill A, Struthers AD. Allopurinol normalizes endothelial dysfunction in type 2 diabetics with mild hypertension. Hypertension. 2000; 35: 746–751.
Dalli E, Segarra L, Ruvira J, Esteban E, Cabrera A, Lliso R, Lopez E, Llopis E, Sotillo JF. Brachial artery flow-mediated dilation in healthy men, men with risk factors, and men with acute myocardial infarction. Importance of occlusion-cuff position. Rev Exp Cardiol. 2002; 55: 928–935.