Conduit Artery Constriction Mediated by Low Flow

Transcription

1 Journal of the American College of Cardiology Vol. 51, No. 20, by the American College of Cardiology Foundation ISSN /08/$34.00 Published by Elsevier Inc. doi: /j.jacc Assessing Vascular Function Conduit Artery Constriction Mediated by Low Flow A Novel Noninvasive Method for the Assessment of Vascular Function Tommaso Gori, MD, PHD,* Saverio Dragoni, MD,* Monica Lisi, MD,* Giuseppe Di Stolfo, MD,* Serena Sonnati, MD,* Massimo Fineschi, MD,* John D. Parker, MD Siena, Italy; and Toronto, Ontario, Canada Objectives Background Methods Results Conclusions We describe and validate a novel noninvasive method that complements the data from traditional flowmediated dilation (FMD) studies. The study of peripheral vascular reactivity provides important diagnostic and prognostic information in patients with (or at risk for) cardiovascular disease. High-resolution ultrasound and automatic computerized analysis were used to measure the diameter of the radial artery at rest and in conditions of locally decreased and increased shear stress (respectively, low flow-mediated constriction [L-FMC] and flow-mediated dilation [FMD]). A composite end point (L-FMC FMD) was also calculated. A total of 196 studies were performed. When the repeatability of the method was tested, the range of variation across measurements was 1.1% for L-FMC and 1.7% for FMD; the intraclass correlation coefficient was 0.80 and 0.68, respectively. Low flowmediated constriction, FMD, and their composite end point were significantly blunted after acute smoking, in coronary artery disease patients, and in hypertensive patients as compared with that seen in healthy agematched volunteers (p 0.01, analysis of variance). Low flow-mediated constriction, but not FMD, was blunted (p 0.05) after administration of fluconazole (an inhibitor of a cytochrome P450-derived endothelium-derived hyperpolarization factor) and aspirin (an inhibitor of cyclooxygenase). Flow-mediated dilation, but not L-FMC, was blunted (p 0.05) by nitric oxide synthase inhibition. Low flow-mediated constriction is a simple, rapid, and accurate measure of resting arterial tone that does not require further procedures as compared with traditional FMD measurements. While FMD measures endothelial responses to sudden increases in shear stress, L-FMC is a measure of the response to resting shear stress levels, and, therefore, it provides additional information that is complementary to FMD. (J Am Coll Cardiol 2008; 51:1953 8) 2008 by the American College of Cardiology Foundation Flow-mediated dilation (FMD) (1) is a simple and noninvasive measure of the capacity of the endothelium, when stimulated by a sudden increase in shear stress, to cause smooth muscle cell relaxation and vasodilation. Importantly, while FMD measures the recruitability of endothelial vasomotor function (i.e., the capacity of the endothelium to modify its biosynthesis and release of mediators to produce vasodilation), it does not provide information regarding the vascular response to resting levels of shear stress. This study reports a series of studies aimed at developing and validating a new method (which, in From the *Department of Internal, Cardiovascular and Geriatric Medicine, University of Siena, Siena, Italy; and the Division of Cardiology, Department of Medicine, Mount Sinai and University Health Network Hospitals, University of Toronto, Toronto, Ontario, Canada. Dr. Gori is the recipient of a grant from the Italian Ministry of Research. Dr. Parker holds a Career Investigator Award from the Heart and Stroke Foundation of Ontario, Canada. This study was funded by a grant from the Heart and Stroke Foundation of Canada. Manuscript received October 24, 2007; revised manuscript received December 21, 2007, accepted January 6, analogy with FMD, we termed low flow-mediated constriction or L-FMC) that allows quantification of this response. This measurement can be made at the same time as traditional FMD, and it provides information that is both complementary and additive to that of FMD, therefore improving its diagnostic capacities and, potentially, clinical utility. Methods This study was approved by the local ethics committee; written informed consent was obtained in all cases. All studies were conducted between 11 AM and 2 PM. Subjects were in a fasting state for at least 6 h. Vascular function was measured in female subjects at the end of the menstrual cycle. See page 1965 Population. A total of 196 studies were performed. For reproducibility (variability between observers) analysis, 50 stud-

2 1954 Gori et al. JACC Vol. 51, No. 20, 2008 Assessment of Vascular Function May 20, 2008: Abbreviations and Acronyms CAD coronary artery disease EDHF endotheliumderived hyperpolarization factor FMD flow-mediated dilation L-FMC low flow-mediated constriction NO nitric oxide ies were analyzed twice by 2 observers. For repeatability (variability between occasions), 25 healthy young volunteers (age range 25 to 34 years, 7 women) underwent measurement of L-FMC and FMD on 2 occasions separated by 24 h. For the comparison of L-FMC measurements between healthy subjects and patients with risk factors and/or cardiovascular disease, 20 healthy young volunteers, 15 healthy middle-age adults, 15 smokers, 18 hypertensive patients, and 13 patients with known coronary artery disease (CAD) were studied. For mechanistic studies, healthy young volunteers (age 24 to 37 years) were enrolled. All subjects had normal ( 180 mg/dl) cholesterol and triglyceride ( 150 mg/dl) levels and no family history of premature cardiovascular disease. Healthy young and middle-age volunteers were lifelong nonsmokers with systolic blood pressure 130 mm Hg and diastolic blood pressure 80 mm Hg. Smokers had a history of 5 to 10 cigarettes/day for 3 to 10 years and smoked 1 cigarette immediately before L-FMC measurement. Hypertensive patients had undergone 24-h blood pressure monitoring that documented average systolic values 140 and diastolic values 90 mm Hg. No subject was on treatment with any drug (including supplemental vitamins), none had history of diabetes, and none had clinical evidence of CAD or peripheral artery disease. The 13 patients with CAD had at least 1 stenosis 70% at angiography and were clinically stable for the previous month. Seven of them had hypertension, 7 had hypercholesterolemia, and 5 had diabetes mellitus. None of them had systolic or diastolic heart failure. All CAD patients were taking aspirin. Measurement of arterial diameter, L-FMC, and FMD. The radial artery was imaged using high-resolution ultrasound, and its diameter was measured using automatic customdesigned software (2 4). Low flow-mediated constriction was calculated as the percent decrease in arterial diameter in the last 30 s of cuff occlusion as compared with resting diameter. Flow-mediated dilation was calculated as the maximum percent increase in arterial diameter after cuff deflation as compared with resting diameter. Finally, a composite end point was calculated as the sum of the absolute values of FMD and L-FMC (Fig. 1). Radial artery blood flows were measured at rest, 15 s before wrist cuff deflation, and immediately upon cuff deflation using pulsed-wave Doppler. All data were analyzed in a randomized, blinded fashion. Mechanistic studies. THE ROLE OF NITRIC OXIDE (NO). Low flow-mediated constriction and FMD were measured during intra-arterial infusion of the NO synthase inhibitor levo-n-monomethyl arginine (8 mol/min) or normal saline in 9 healthy male subjects (5). Infusions were started 10 min before L-FMC and FMD measurements. THE EFFECT OF FLUCONAZOLE AND ASPIRIN. Fourteen healthy volunteers (4 women) were enrolled for this doubleblind, placebo-controlled, cross-over trial. In 3 separate visits, subjects were randomized to receive placebo, fluconazole (150 Figure 1 Representative Plot of FMD and Low-FMC Radial artery diameter changes throughout the study of low flow-mediated constriction (FMC) and flow-mediated dilation (FMD). After 1 min of resting radial artery diameter measurement, the wrist cuff is inflated (arrowhead) to suprasystolic pressure. At this point, the radial artery diameter progressively decreases in healthy volunteers until it reaches a plateau (low-fmc) at approximately 3 30 after wrist cuff inflation and does not further constrict for longer occlusion times. Upon wrist cuff deflation (double arrowhead), the increased shear stress causes radial artery dilation (FMD). Therefore, L-FMC is a negative number, while FMD is a positive one. Doppler blood flows are reported at each time point. Image from the analysis software (4).

3 JACC Vol. 51, No. 20, 2008 May 20, 2008: Gori et al. Assessment of Vascular Function 1955 Figure 2 Repeatability of FMD and L-FMC Correlation analysis across consecutive (24 h) measurements of L-FMC (A) and FMD (B). The intraclass correlation coefficient (ICC) was higher for L-FMC. Abbreviations as in Figure 1. mg orally, an inhibitor of the isoform 2C9 of the cytochrome P450, which is believed to be involved in the synthesis of an endothelium-derived relaxing factor [EDHF]) (6,7) or aspirin (an inhibitor of cyclooxygenase products such as prostaglandins, 500 mg orally). This dosage of fluconazole is expected to achieve plasma concentrations in the range of 3 g/ml (8), which has been previously shown to inhibit EDHF production (7). Two hours later, L-FMC and FMD were measured as previously described. Figure 3 Bland-Altman Plots for Consecutive Measurements of L-FMC and FMD The plots show a higher concordance for L-FMC (left) measurements. The dotted lines represent 1.96 SD. Abbreviations as in Figure 1. THE ROLE OF BLOOD PRESSURE CHANGES AND DISTEND- ING PRESSURE. A 21-gauge intra-arterial cannula was placed in the left brachial artery of 11 male subjects (5). Invasive systolic, mean, and diastolic blood pressures were measured at rest and during inflation of a pneumatic cuff placed around the homolateral wrist. Another group of 15 subjects underwent the same measurements with the pneumatic cuff placed around the contralateral wrist. THE ROLE OF SYSTEMIC REFLEXES. A wrist cuff was placed around the right wrist, and arterial diameter was measured contralaterally in 5 subjects (2 women). Statistical analysis. Data are presented as mean SD. For reproducibility and repeatability analysis, the appropriate intraclass correlation coefficient and range of variation (mean of the absolute differences between consecutive measurements) was calculated. Bland-Altman plots (9) were constructed for both L-FMC and FMD, and the coefficient of repeatability was calculated (1.96 times the standard deviation of the difference between consecutive measurements). Both 1- and 2-way analysis of variance (ANOVA) were employed for analysis of differences within and between patient and control groups. For mechanistic studies, t tests or repeated measures ANOVA were employed (Statview version 5, SAS Institute Inc., Cary, North Carolina). Post-hoc comparisons were done with the Fisher test. The threshold for significance was p Results In 2 subjects, scans were of insufficient quality to assess FMD. Low flow-mediated constriction was assessed in all subjects. Reproducibility of L-FMC and FMD. The mean range of variation between 2 sets of arterial diameter measurements using the software employed here is 0.03 mm. The mean range of interobserver variation for the measurement of percent L-FMC and FMD was, respectively, 0.45% (0.02% to 2.9%) and 0.60% (0.04% to 2.9%) (n 50 studies in arteries with a mean diameter of approximately 2.5 mm, n 2 observers). The corresponding intraclass coefficient was 0.98 for both L-FMC and FMD. Repeatability of L-FMC and FMD. The intraclass correlation coefficient for repeated measurements of baseline diameter was 0.85; for L-FMC it was 0.80; for FMD it was The Role of NOS Table 1 The Role of NOS Normal Saline L-NMMA p Value Resting diameter, mm NS during cuff inflation NS after cuff deflation Resting blood flow, ml/min Blood flow during cuff inflation NS Blood flow after cuff deflation NS Radial artery diameters at rest and change during cuff inflation and after cuff deflation. Levo-Nmonomethyl arginine (L-NMMA) blunted flow-mediated dilation but did not modify low flowmediated constriction. NOS nitric oxide synthase.

4 1956 Gori et al. JACC Vol. 51, No. 20, 2008 Assessment of Vascular Function May 20, 2008: Figure 4 Mechanistic Hypothesis (Left) Low flow-mediated constriction (FMC) and FMD measurements during infusion of normal saline or levo-n-monomethyl arginine (L-NMMA) (8 mol/min). Low FMC was not modified by L-NMMA, suggesting that nitric oxide plays a lesser role in regulating resting arterial tone. *p (Right) Low-FMC and FMD measurements after placebo, fluconazole (150 mg orally), and aspirin (500 mg orally). Low FMC was significantly blunted after fluconazole and aspirin. *p Abbreviations as in Figure (Fig. 2). The range of variation for each of these variables was, respectively, 0.15% (0.0% to 0.32%), 1.1% (0.0% to 2.7%), and 1.7% (0.1% to 5.2%). Bland and Altman plots for repeated L-FMC and FMD measurements are presented in Figure 3. The coefficient of repeatability was 1.5% for L-FMC and 3.2% for FMD. Mechanistic studies. THE ROLE OF NO. While blunting blood flow, levo-n-monomethyl arginine (L-NMMA) did not modify resting radial artery diameter or L-FMC. As compared with vehicle, both arterial diameter during reactive hyperemia and FMD were significantly blunted during L-NMMA (Table 1, Fig. 4). THE ROLE OF EDHF AND CYCLOOXYGENASE PRODUCTS. Neither fluconazole nor aspirin modified resting radial artery diameter, resting blood flow, FMD, or blood flow during inflation and deflation of the cuff. In contrast, L-FMC was significantly attenuated after both agents (Table 2, Fig. 4). The Role of EDHF and COX-Derived Products Table 2 The Role of EDHF and COX-Derived Products THE ROLE OF BLOOD PRESSURE CHANGES. Inflation of a wrist cuff placed around the homolateral wrist caused a nonsignificant (p 0.1) increase in systolic blood pressure measured invasively at the level of the left brachial artery while not changing mean and diastolic blood pressure or heart rate (Table 3). Inflation of a pneumatic cuff placed on the right wrist did not affect these parameters nor did it modify the arterial diameter of the left radial artery (resting diameter mm; during cuff occlusion mm, p NS). L-FMC in health and disease. In healthy volunteers, L-FMC varied from 0.1% to 12.2%; FMD varied from 1.4% to 14.8%. Flow-mediated dilation and L-FMC were similar in healthy young and middle-age volunteers and were significantly blunted after acute smoking in hypertensive and CAD patients (p 0.01 for all). The composite end point was also significantly blunted after smoking in Placebo Aspirin Fluconazole p Value Resting diameter, mm NS during cuff inflation * after cuff deflation NS Resting blood flow, ml/min NS During cuff inflation NS After cuff deflation NS Radial artery diameters at rest and the change in diameter during cuff inflation and after cuff deflation. Both aspirin and fluconazole blunted low flow-mediated constriction but did not modify flow-mediated dilation. *p 0.05 for low flow-mediated constriction expressed as percent change from baseline. COX cyclooxygenase; EDHF endothelium-derived hyperpolarization factor.

5 JACC Vol. 51, No. 20, 2008 May 20, 2008: Gori et al. Assessment of Vascular Function 1957 Lack of a Role of Distending Pressure in L-FMC Table 3 Lack of a Role of Distending Pressure in L-FMC Homolateral Cuff Contralateral Cuff Systolic blood pressure, mm Hg At rest During cuff inflation Diastolic blood pressure At rest During cuff inflation Mean blood pressure At rest During cuff inflation Homolateral and contralateral wrist cuff inflation did not modify blood pressure. L-FMC low flow-mediated constriction. hypertensive and CAD patients (Table 4, Fig. 5, right). Resting blood flow, blood flow during cuff inflation, and reactive hyperemia were similar across groups. A weak but significant positive correlation was found between L-FMC and percent decrease in blood flow during cuff occlusion (R , p 0.01). There was no correlation between L-FMC and FMD in the whole study population (R , p 0.1) or in any of the subgroups. Discussion Among the methods that allow measurement of endothelial function in humans in vivo, FMD has rapidly gained popularity because of its simplicity, reproducibility, and noninvasiveness. One important limitation of FMD is that although this technique measures how much the endothelium modifies its biosynthetic activity in response to sudden increases in shear stress (i.e., endothelial recruitability ), it does not provide information concerning the effect of shear stress in resting conditions (i.e., shear stress-dependent basal tone or resting endothelial tone ). To address this, we developed and validated a novel measure that is also noninvasive, reproducible, and does not require further procedures as compared with traditional FMD. The important additional information brought by introduction of L-FMC is that this variable provides information concerning the control of arterial tone at rest, therefore FMD and L-FMC in Health and Disease Table 4 FMD and L-FMC in Health and Disease complementing (and not overlapping with) the information provided by FMD. This point is further emphasized by the lack of a correlation between L-FMC and FMD; the absence of such correlation highlights the incremental value of L-FMC and is a stimulus for further studies investigating the relationship between basal and recruitable responses to shear stress. Furthermore, introduction of the measurement of L-FMC allows calculating a composite end point, expressed as the sum of the absolute values of FMD and L-FMC. This end point provides a comprehensive measurement of shear stress-dependent vasomotion, which combines both resting tone and endothelial recruitability. While mechanisms underlying FMD have been described in detail, less is understood concerning the mediators involved in modulating resting conduit artery tone. Our data show that inhibition of the synthesis of NO with L-NMMA does not modify L-FMC, suggesting that NO has a lesser role in maintaining resting conduit artery tone. In contrast, the significant blunting of L-FMC observed after fluconazole and aspirin administration allows the hypothesis that cytochrome P450-derived EDHF and/or cyclooxygenase products (such as prostaglandins) are actively secreted in resting conditions by the endothelium of conduit arteries (7) and that this production is decreased in conditions of reduced shear stress. Finally, Spieker et al. (10) reported that inhibition of type A receptors of endothelin-1 reduces L-FMC. Therefore, we hypothesize that L-FMC might be mediated by a coordinated effect of vasodilator (a P450-derived EDHF and cyclooxygenase products) inhibition and increased endothelin-1 production. This emphasizes the importance of L-FMC as an adjunct to FMD measurement, as, taken together, these 2 measures provide more comprehensive information on the multiple molecular pathways involved in the control of vascular homeostasis. Future studies will have to identify the exact nature of these mediators. Conclusions In sum, the reproducibility and repeatability of L-FMC is comparable to that of traditional FMD; L-FMC, as well as Healthy Young Healthy Age-Matched Young After Smoking Coronary Artery Disease Hypertension Age, yrs (range) 29 3 (25 30)* 53 3 (37 65)* 29 1 (25 32)* 65 7 (54 70)* (35 65)* Resting diameter, mm during cuff inflation, mm after cuff deflation, mm Blood flow, ml/min At rest During cuff inflation Reactive hyperemia Resting heart rate, beats/min There was no difference across groups in heart rate and blood flow at any time point. Differences among groups in resting blood flow were due to 3 outliers (1 each in the healthy young, the smoking, and the hypertensive groups) and were not significant (p 0.67). The median was not different across groups. *p 0.01 analysis of variance among groups; p 0.01 compared with younger subjects; p 0.05 for the comparison with corresponding control subjects. FMD flow-mediated dilation; L-FMC low flow-mediated constriction.

6 1958 Gori et al. JACC Vol. 51, No. 20, 2008 Assessment of Vascular Function May 20, 2008: Figure 5 L-FMC, FMD, and the Composite End Point in Health and Disease All 3 end points are blunted in the setting of cardiovascular disease. CAD coronary artery disease; Htn hypertension; other abbreviations as in Figure 1. FMD, is blunted after smoking in patients with CAD and in hypertensive patients. Because it can be obtained at the same time as FMD measurements and it yields information that is both complementary and additive to that of FMD, the introduction of L-FMC provides a simple yet more comprehensive assessment of vascular function for pathophysiological and pharmacologic studies as well as for clinical prevention programs. Reprint requests and correspondence: Dr. Tommaso Gori, Department of Medicine 2, University Hospital of Siena, Siena, Italy. Tommaso.gori@utoronto.ca. REFERENCES 1. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992;340: Gori T, Sicuro S, Donati G, Dragoni S, Forconi S, Parker JD. Sildenafil prevents endothelial dysfunction induced by ischemia and reperfusion via opening of ATP-sensitive K channels. A human in vivo study. Circulation 2005;111: Gori T, Di Stolfo G, Sicuro S, et al. Nitroglycerin protects the endothelium from ischaemia and reperfusion: human mechanistic insight. Br J Clin Pharmacol 2007;64: HeartWorks. Available at: Accessed May 1, Gori T, Burstein JM, Ahmed S, et al. Folic acid prevents nitroglycerin-induced nitric oxide synthase dysfunction and nitrate tolerance: a human in vivo study. Circulation 2001;104: Fisslthaler B, Popp R, Kiss L, et al. Cytochrome P450 2C is an EDHF synthase in coronary arteries. Nature 1999;401: Bellien J, Joannides R, Iacob M, Arnaud P, Thuillez C. Evidence for a basal release of a cytochrome-related endothelium-derived hyperpolarizing factor in the radial artery in humans. Am J Physiol Heart Circ Physiol 2006;290:H Al Gaai E, Lockyer M, Al Digither S, Hammami MM. Bioequivalence evaluation of two formulations of fluconazole 150 mg capsule in healthy Arab men. Biopharm Drug Dispos 2005;26: Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1: Spieker LE, Luscher TF, Noll G. ETA receptors mediate vasoconstriction of large conduit arteries during reduced flow in humans. J Cardiovasc Pharmacol 2003;42:315 8.