Embed Size (px)
Effect of Captopril and Enalapril on EndothelialFunction in Hypertensive Patients
Mark A. Creager, Mary-Anne Roddy
Abstract Endothelium-dependent vasodilation is impairedin patients with essential hypertension. The objective of thisstudy was to determine whether long-term treatment withangiotensin-converting enzyme inhibitors improves endotheli-um-dependent vasodilation in forearm resistance vessels ofpatients with hypertension. Furthermore, since tissue thiolsmay be relevant to nitric oxide-mediated vasodilation, wequeried whether an angiotensin-converting enzyme inhibitorwith a sulf hydryl group preferentially augments endothelium-dependent vasodilation in these individuals. The study in-cluded 24 patients with essential hypertension (mean age,45±2 years) and 20 normotensive subjects (mean age, 47±1years). Methacholine chloride (0.3 to 10 jtg/min) was infusedvia the brachial artery to assess endothelium-dependent vaso-dilation in forearm resistance vessels. Nitroglycerin (1 to 30/ig/min) was administered to evaluate endothelium-indepen-dent vasodilation. Forearm blood flow was determined byvenous occlusion strain-gauge plethysmography. Forearm vas-cular function studies were performed in hypertensive patientsbefore and 7 to 8 weeks after randomization to either captoprilor enalapril, angiotensin-converting enzyme inhibitors withand without a sulf hydryl moiety, respectively. Normotensivesubjects were studied on only one occasion. Before treatment,the forearm vasodilative response to methacholine was atten-
The endothelium synthesizes a variety of vasoac-tive substances that regulate vascular tone,including the endothelium-derived relaxing fac-
tor (EDRF) nitric oxide.1-2 It is well established thatendothelium-dependent vasodilation is impaired in con-duit and resistance vessels of experimental models ofhypertension and in patients with essential hyperten-sion.312 This has led some to speculate that abnormal-ities in endothelium-dependent vasodilation may con-tribute to the pathogenesis of hypertension by offsettingthe balance between vasoconstrictor and vasodilatorforces on vascular tone.1315 Others have implied thatabnormalities in endothelial function occur after hyper-tension develops.1620 Thus, reduced EDRF activitywould be a consequence rather than a cause of hyper-tension. In either case, reduced endothelial release ofEDRF/nitric oxide could not only exacerbate hyperten-sion but perhaps contribute further to vascular injury.
Received October 29, 1993; accepted in revised form May 16,1994.
From the Vascular Medicine and Atherosclerosis Unit of theCardiovascular Division, Brigham and Women's Hospital, Har-vard Medical School, Boston, Mass.
Reprint requests to Mark A. Creager, MD, CardiovascularDivision, Brigham and Women's Hospital, 75 Francis St, Boston,MA 02115.
© 1994 American Heart Association, Inc.
uated in hypertensive compared with normotensive subjects(P< .01). The effects of nitroglycerin on forearm blood flow didnot differ significantly between the two groups. Both captopriland enalapril reduced mean blood pressure in the hyperten-sive subjects (12±2 versus 15±3 mm Hg, respectively; P=NS).The forearm vasodilative response to methacholine was thesame during the placebo and captopril treatment periods(P=NS) and also during placebo and enalapril treatmentperiods (P=NS). Even when combining both treatment groups(captopril and enalapril), no significant difference in theresponse to methacholine was found between placebo anddrug treatment periods. It is concluded that endothelium-dependent vasodilation is abnormal in forearm resistancevessels in patients with essential hypertension, thus confirmingobservations made previously by others. The new finding isthat antihypertensive therapy for up to 7 to 8 weeks with anangiotensin-converting enzyme inhibitor does not improveendothelium-dependent vasodilation in hypertensive humans,regardless of whether or not a sulfhydryl group is present.(Hypertension. 1994^4:499-505.)
Key Words • endothelium-derived relaxing factor •angiotensin-converting enzyme inhibitors • captopril •enalapril • hypertension, essential • forearm • bloodflow
Several groups of investigators have sought to de-termine whether antihypertensive treatment restoresendothelium-dependent vasodilation in patients withessential hypertension. Panza et al15 reported that long-term treatment with antihypertensive therapy did notimprove abnormal endothelium-dependent vasodilationin forearm resistance vessels of patients with hyperten-sion. Hirooka et al,21 however, recently reported thatadministration of a single dose of the angiotensin-converting enzyme (ACE) inhibitor captopril improvedendothelium-dependent vasodilation in hypertensivesubjects. Supporting the latter findings are studies invessels obtained from experimental models of hyperten-sion. Clozel et al22 reported that the ACE inhibitorscilazapril and captopril, but not hydralazine, improvedendothelium-dependent relaxation in aortic rings ofspontaneously hypertensive rats. One explanation forthe disparate findings is that ACE inhibitors, in contrastto other antihypertensive medications, preferentiallyaffect endothelium-dependent vasodilation. Several in-vestigators have suggested that ACE inhibitors increasethe concentration of bradykinin, a known stimulus forEDRF release, thereby promoting endothelium-depen-dent vasodilation.23'24 Moreover, tissue thiols are rele-vant to nitric oxide-mediated vasodilation, raising thepossibility that an ACE inhibitor with a sulfhydrylmoiety may enhance endothelium-dependent vasodila-
by guest on April 22, 2018
500 Hypertension Vol 24, No 4 October 1994
tion by inducing formation of S-nitrosothiol adducts toactivate guanylate cyclase.2527
Accordingly, the objectives of this study were (1) todetermine whether treatment with ACE inhibitors im-proves endothelium-dependent vasodilation in forearmresistance vessels of patients with hypertension and (2)to determine whether an ACE inhibitor with a sulfhy-dryl group preferentially improves endothelium-depen-dent vasodilation in these individuals.
The subject population in this study included 24 patientswith essential hypertension (18 men, 6 women; age range, 26 to67 years; mean [±SE], 45±2 years). The mean blood pressurein the untreated hypertensive subjects was 109±3 mm Hg. Theduration of hypertension ranged from 1 to 25 years (mean,8±1 years). None of the patients had clinical evidence ofatherosclerosis. This was determined by the absence of symp-toms of angina, claudication, or cerebrovascular ischemia orany physical evidence of arterial occlusive disease, as would besuggested by decreased pulses, asymmetrical blood pressure,or bruits. The study included 20 normotensive subjects (13men, 7 women; age range, 35 to 60 years; mean, 47±1 years).The ages of the patients with hypertension and the normoten-sive subjects were not significantly different. Normalcy wasdetermined by careful history, physical examination, and lab-oratory analysis to exclude individuals with hematologic, renal,or hepatic dysfunction. The mean blood pressure in thenormotensive subjects was 79±2 mm Hg. This study wasapproved by the Human Research Committee of Brigham andWomen's Hospital, and each subject gave written informedconsent.
Experimental ProtocolAfter an initial screening period to confirm eligibility re-
quirements, hypertensive patients were withdrawn from allantihypertensive medications and followed for 2 weeks. Ifblood pressure did not increase to values exceeding 180/120mm Hg, patients were allowed to continue their participationand were administered placebo, in a single-blind manner, for 4weeks. Forearm vascular function studies, described below,were then performed. Thereafter, patients were randomized ina double-blind manner to treatment with one of two drugs:captopril, an ACE inhibitor with a sulfhydryl moiety, orenalapril, an ACE inhibitor that does not have a sulfhydrylgroup. Over the next 3 to 4 weeks, the dose of each drug wastitrated to achieve normotension, ie, a blood pressure less than140/90 mm Hg, or a maximal dose of each drug, which was 75mg PO twice dairy for captopril and 15 mg PO twice daily forenalapril. After 4 more weeks of treatment at a stable dose ofthe respective ACE inhibitor, the forearm vascular functionstudies were repeated. Normotensive subjects were studied ononly one occasion.
Vascular Function StudiesEach subject was studied in a 23°C temperature-controlled
room in the postabsorptive state. Alcohol, caffeine, and ciga-rettes were all prohibited within 12 hours of the study. Underlocal anesthesia and sterile conditions, a polyethylene catheterwas inserted into a brachial artery of each subject for deter-mination of blood pressure and for infusion of drugs. Thevascular research laboratory was quiet, and lights weredimmed. All subjects rested at least 30 minutes after catheterplacement to establish a stable baseline before data collection.
During the control period, measurements of forearm bloodflow and blood pressure were repeated every 10 minutes untilstable. Dextrose 5% was infused intra-arterially at a rate of 0.4mL/min during the control period. To determine the maximalvasodilative potential of the resistance vessels, forearm blood
flow was measured in the basal state and during peak reactivehyperemia after 5 minutes of an ischemic stimulus. Ischemiawas induced by inflation of a sphygmomanometric cuff on theupper arm to suprasystolic pressure. Measurements of peakreactive hyperemic blood flow were made within 10 seconds ofcuff deflation. Abnormalities in reactive hyperemic blood flowoften imply structural problems in the resistance vessels,preventing maximal vasodilation. Forearm blood flow mea-surements were repeated until basal conditions werereestablished.
The protocol used to examine forearm vascular reactivity inpatients with hypertension and normal subjects has beenreported previously.28 To specifically assess endothelium-de-pendent vasodilation, methacholine chloride (a congener ofacetylcholine) was administered via the brachial artery. Fore-arm blood flow was measured during infusion of methacholinechloride at concentrations of 0.3, 1, 3, and 10 /xg/min, each for3 minutes, delivered at a rate of 0.4 mL/min. To distinguishabnormalities in endothelial function from abnormalities ofvascular smooth muscle, subjects received an intra-arterialinfusion of nitroglycerin. This agent, which acts directly onvascular smooth muscle by stimulating soluble guanylate cy-clase and inducing hyperpolarization, was given at doses of 1,3, 10, and 30 jtg/min, at a rate of 0.4 mL/min, each for 3minutes. The order of administration for methacholine chlo-ride and nitroglycerin was randomized for each subject. Basalconditions were reestablished before each intervention. Thedoses of each drug were chosen to achieve decreases inforearm vascular resistance without causing systemic effects.Dose-response curves were generated for each drug infusion.
Hemodynamic MeasurementsBilateral forearm blood flow was determined by venous
occlusion strain-gauge plethysmography, using calibrated mer-cury-in-silastic strain gauges, and expressed as milliliters per100 mL tissue per minute (D.E. Hokanson, Inc).29 Each armwas supported above heart level. The mean venous occlusionpressure was 34±1 mm Hg. Circulation to the hand wasprevented by inflating a wrist cuff to suprasystolic pressurebefore each forearm blood flow determination. Determinationof forearm blood flow comprised at least five separate mea-surements performed at 10- to 15-second intervals. By mea-suring blood flow in the infused arm, one can determine thedirect effect of the vasoactive drug. By measuring blood flow inthe noninfused arm, one can be assured that systemic effectshave not occurred if no change in blood flow developed duringthe drug infusion. Forearm vascular resistance was calculatedas a ratio of mean blood pressure to forearm blood flow andexpressed as units reflecting millimeters of mercury per milli-liter per 100 mL tissue per minute.
Blood pressure was measured via an intra-arterial cannulathat was attached to a pressure transducer aligned to anamplifier on a physiological recorder (Gould, Inc). Heart ratewas determined from the simultaneously obtained electrocar-diographic signal and calculated from the RR interval.
Statistical AnalysisResults are presented as mean±SE. A between-within split
plot design analysis was employed to compare the response ofthe drug infusions between the hypertensive and normalsubjects. Single-factor repeated-measures ANOVA followedby a Newman-Keuls post hoc test was used to compare theeffect of each vasoactive drug in the hypertensive patientsbefore and after randomization. Student's f test was used tocompare paired data only. Statistical significance was acceptedat the 95% confidence interval at P<.05.
ResultsThe pertinent hemodynamic characteristics of the
hypertensive and normotensive subjects are provided inthe Table. Basal forearm blood flow was significantly
by guest on April 22, 2018
Creager and Roddy Endothelial Function in Hypertension 501
Hemodynamlc Characteristics of Normotenslve andHypertensive Subjects
BP, mm Hg
RHFBF, (mL/100 mL)/min
BP indicates mean blood pressure; HR, heart rate; bpm, beatsper minute; FBF, forearm blood flow; FVR, forearm vascularresistance; RHFBF, peak reactive hyperemic FBF; and MFVR,minimal FVR. Values are expressed as mean±SE.
higher in the hypertensive subjects. Peak reactive hyper-emic blood flow tended to be somewhat less in patientswith hypertension than in normotensive subjects, butthe difference between groups did not achieve statisticalsignificance. Basal forearm vascular resistance was com-parable between groups. Minimal forearm vascular re-sistance, obtained during peak reactive hyperemia, wassignificantly greater in patients with hypertension thanin normotensive subjects, indicating an impairment inmaximal vasodilative capacity.
Vasodilative Responses in Hypertensive VersusNormotensive Subjects
Intra-arterial infusion of methacholine chloridecaused a dose-dependent and significant increase inforearm blood flow and decrease in forearm vascularresistance in both hypertensive and normotensive sub-jects. In the hypertensive subjects, however, cholinergicvasodilation was attenuated compared with normoten-sive subjects (Fig 1). The differences between thegroups were significant at each dose of methacholine.At the highest dose of methacholine, forearm bloodflow increased 414±58% in hypertensive subjects and629±100% in normotensive subjects (P<.01). Similarly,the decrease in forearm vascular resistance was signifi-cantly less in hypertensive subjects than in normotensivesubjects at doses of 0.3 fig/min ( -47±5% versus-62±2%), 1.0 Mg/min (-56±5% versus -73±2%), 3.0
ao ^o •m c
0 0.3 1 3 10
METHACHOLINE DOSE (/ig/mln)
FIG 1. Line graph shows the effect of methacholine chloride onforearm blood flow in normotensive and hypertensive subjects.Endothelium-dependent vasodilation was significantly reducedin hypertensive subjects. Data are expressed as mean±SE.
. O O NORMOTENSIVE• • HYPERTENSIVE
% 300do§ 1 200CO C
2iQ H 100
O O NORMOTENSIVE (n=20)• • HYPERTENSIVE (n=24)
o 1 3 10 30NITROGLYCERIN DOSE ( / i f l /mln)
FIG 2. Line graph shows the effect of nitroglycerin on forearmblood flow in normotenslve and hypertensive subjects. Therewas no significant difference in response to this endothellum-independent vasodilator between the two groups.
jig/min (-66±4% versus -81±1%), and 10.0 jig/min(-74±4% versus -85±1%), respectively (all P<.01).No changes in forearm blood flow occurred in thenoninfused arm in either subject group. In addition,intra-arterial infusion of methacholine chloride causedno change in blood pressure or heart rate in eithergroup of subjects.
Infusion of nitroglycerin also increased forearm bloodflow and decreased forearm vascular resistance in hy-pertensive and normotensive subjects. In contrast to theattenuated response to methacholine chloride observedin the hypertensive subjects, the effects of nitroglycerinon forearm blood flow and forearm vascular resistancedid not differ significantly between the two groups (Fig2). At a dose of 30 jig/min, forearm blood flow in-creased 205±26% and 232±31% and forearm vascularresistance decreased -61±4% and -65±4% in hyper-tensive and normotensive subjects, respectively (each/>=NS). No changes in forearm blood flow or forearmvascular resistance occurred in the noninfused arm ofeither group of subjects. Taken together, these dataconfirm that endothelium-dependent vasodilation is im-paired in patients with essential hypertension.
Effect of ACE Inhibition on Endothelium-Dependent Vasodilation
Twenty-two hypertensive patients were treated withan ACE inhibitor and completed placebo and treatmentforearm vascular function studies. Two patients with-drew before the second study. Each drug reduced meanblood pressure (Fig 3). Captopril decreased mean bloodpressure from 109±5 mm Hg on placebo, determined atthe first vascular function study, to 98±9 mm Hg, deter-mined at the second vascular function study (Z'<.001).Similarly, enalapril reduced mean blood pressure from110±4 mm Hg to 95±3 mm Hg (P<.001). There was nosignificant difference in the magnitude of blood pres-sure reduction between the two treatment groups (12±2versus 15 ±3 mm Hg for captopril and enalapril, respec-tively). Mean blood pressure in the treated hypertensivegroups remained higher than that in nonnotensivesubjects (/)<.01).
Eleven hypertensive patients were treated with cap-topril. Neither basal forearm blood flow (3.3±0.4 to2.9±0.4 mL/100 mL per minute; />=NS) nor forearmvascular resistance (40±5 to 38±4 U; P=NS) changedduring captopril treatment. The forearm blood flow
by guest on April 22, 2018
502 Hypertension Vol 24, No 4 October 1994
CD PLACEBOESS ACE-INHIBITION
T • p<0.001
FIQ 3. Bar graph shows the effect of captopril and enalapril onmean blood pressure in hypertensive subjects. Each drugsignificantly reduced blood pressure. The reduction in bloodpressure was not significantly different between the captopriland enalapril treatment groups.
oo •Zj CDIS C
• • ENALAPRILO O PtACEBO
100 0.3 1 3METHACHOUNE DOSE (pg/mln)
Fia 5. Line graph shows the forearm blood flow response tomethacholine chloride during the placebo and enalapril treat-ment periods in hypertensive subjects. The forearm blood flowresponse curves were similar.
response to methacholine chloride was the same duringthe placebo and captopril treatment periods (Fig 4). Atdoses ranging from 0.3 to 3 /ig/min, captopril tended tocause a mild augmentation in blood flow, but this didnot reach statistical significance. Similarly, there was nosignificant difference in the forearm blood flow responseto intra-arterial nitroglycerin between the placebo andcaptopril treatment periods (data not shown). Elevenhypertensive patients were treated with enalapril. Nei-ther basal forearm blood flow (3.1±0.3 to 28±0.4mL/100 mL per minute; P=NS) nor forearm vascularresistance (32 ±1 to 29 ±1 U; P=NS) changed duringenalapril treatment. The forearm blood flow responsecurves to methacholine chloride during the placebo andenalapril treatment periods were virtually identical (Fig5). Similarly, there was no significant difference in theforearm blood flow response to nitroglycerin betweenthe two treatment periods (data not shown).
To ascertain further that 7 to 8 weeks of treatmentwith an ACE inhibitor did not substantially affectendothelium-dependent vasodilation in forearm resis-tance vessels, the results from both the captopril andenalapril treatment groups were combined. The designallowed us to address the issue of whether ACE inhib-itors, as a class, restore endothelium-dependent vasodi-lation. Furthermore, the larger number of subjectsincreases statistical power, thus reducing the risk of a/3-error. Nonetheless, even when combining both treat-ment groups, we found no significant difference in the
SIi j zooce o
CAPTOPRILO O PLACEBO
0.3 1 3UETHACHOUNE DOSE (/ig/mln)
FIG 4. Line graph shows the forearm blood flow response tomethacholine chloride during the placebo and captopril treat-ment periods in hypertensive subjects. Captopril did not signif-icantly improve endothelium-dependent vasodilation.
forearm vasodilative response to methacholine betweenthe placebo and drug treatment periods (Fig 6). Simi-larly, the forearm blood flow response to nitroglycerinduring the placebo and ACE inhibitor treatment peri-ods was comparable (Fig 7).
DiscussionThe results of this study enable us to conclude that
endothelium-dependent vasodilation is abnormal inforearm resistance vessels of patients with essentialhypertension, thus confirming observations made previ-ously by others.6-7-9 The principal new finding is thatantihypertensive therapy with an ACE inhibitor doesnot improve endothelium-dependent vasodilation inhypertensive humans, regardless of whether a sulfhy-dryl group is present. This discussion will briefly reviewendothelium-dependent vasodilation in hypertensionand discuss our reasoning for failing to confirm thehypothesis that ACE inhibitors improve endothelium-dependent vasodilation in patients with hypertension.
Endothelium-Dependent Vasodilation Is Abnormalin Hypertension
Most, but not all, studies in experimental models ofhypertension have found that endothelium-dependentrelaxation is abnormal in both conduit and resistancevessels.35-30 Similarly, studies in humans have shown
• • ACE-INHIBITION
O O PLACEBO
0.3 1 3METHACHOUNE DOSE (pfl/mln)
FIG 6. Line graph shows the effect of chronic angiotensin-converting enzyme (ACE) inhibition on the forearm blood flowresponse to methacholine chloride in all hypertensive subjects.ACE inhibition did not affect endothelium-dependent vasodila-tion in forearm resistance vessels, even when captopril andenalapril treatment groups were combined.
by guest on April 22, 2018
Crcager and Roddy Endothelial Function in Hypertension 503
0 1 3 10 30NrrROGLYCEJBN DOSE (/ifl/min)
FIG 7. Line graph shows the effect of angiotensin-convertingenzyme (ACE) inhibition on the forearm blood flow response tonitrogrycerin in all hypertensive subjects. Endothelium-indepen-dent vasodilation was comparable during placebo and ACEinhibitor treatment.
abnormal endothelium-dependent vasodilation in fore-arm resistance vessels, coronary resistance vessels, andepicardial coronary arteries.612 In this study the vasodi-lative response to the highest dose of methacholinechloride was approximately 33% less in hypertensivethan in normotensive subjects, consistent with abnormalendothelium-dependent vasodilation. The vasodilativeresponse to the endothelium-independent agent nitro-glycerin was comparable between the two groups. Thesedata not only confirm those of previous studies butdemonstrate that the technique used in our laboratoryis valid.
Mechanisms of Endothelial DysfunctionWe assume that abnormalities in endothelium-depen-
dent vasodilation in human hypertension are secondaryto reduced release or activity of EDRF/nitric oxide.This assumption is based on the following information.The forearm vasodilative response to cholinergic ago-nists is inhibited by AfG-monomethyl-L-arginine, thenitric oxide synthase antagonist, and not by aspirin,indomethacin, or phentolamine.6-28-31 A^-monomethyl-L-arginine inhibits forearm vasodilation more in nor-motensive than in hypertensive subjects.8-9 In animalmodels of hypertension, however, the mechanism ofabnormal endothelium-dependent vasodilation is lessclear. It has been attributed variably to abnormalities inthe EDRF/nitric oxide pathway, decreased endotheli-um-derived hyperpolarizing factor, or increased releaseof vasoconstrictor prostanoids.3236 Different models ofhypertension, vessels of interest, and agents used toinduce endothelium-dependent relaxation may account,in part, for the variability among studies. One mightanticipate, therefore, that endothelium-dependent re-sponses to antihypertensive therapy may differ amonganimal studies per se and between animal models andpatients with hypertension.
Effect of Antihypertensive TreatmentOur findings are consistent with those of Panza et al,15
who reported that withdrawal of antihypertensive medica-tion did not affect endothelium-dependent vasodilation offorearm resistance vessels in patients with hypertension.Patients in that study were being treated with a variety ofantihypertensive medications, and most were not takingan ACE inhibitor. Our findings are also consistent with a
preliminary report by Kiowski et al,37 who reported thatlong-term treatment with the ACE inhibitor perindoprildid not affect endothelium-dependent vasodilation offorearm resistance vessels of hypertensive patients. Thefindings, however, appear to conflict with those of Hirookaet al,21 who reported that administration of a single dose ofcaptopril, but not nifedipine, improved endothelium-de-pendent vasodilation in forearm resistance vessels of hy-pertensive patients. Both drugs reduced blood pressure toa similar extent, suggesting that it was the ACE inhibitorand not simply the reduction in blood pressure that wasresponsible for the observation. The study by Hirooka etal differed from all other studies in humans in that theforearm vasodilative response to acetylcholine was im-paired in hypertensive patients only at low doses ofacetylcholine but not at the higher doses. It is noteworthy,then, that captopril had no effect on endothelium-depen-dent vasodilation at the highest doses of acetylcholine.Nonetheless, these findings cannot be discounted becausemultiple studies in experimental models of hypertensionalso have found that ACE inhibitors improve endotheli-um-dependent relaxation.22-23'27'38
We had thought that an ACE inhibitor with a sulf hy-dryl moiety would be particularly likely to enhanceendothelium-dependent vasodilation in hypertensivesubjects. The potential of such an agent would be notonly to augment endothelium-dependent vasodilationvia bradykinin potentiation but also to enhance endo-thelium-dependent vasodilation by inducing formationof 5-nitrosothiol adducts or by preventing inactivationof nitric oxide by scavenging superoxide anions. Theselatter properties have previously been attributed tocaptopril but not to enalapril.2527-39 Recent preliminarystudies from our laboratory, however, do not supportthis postulate. We found that n-acetylcysteine, a sulf hy-dryl donor, potentiated the vasodilator response tonitroglycerin but not methacholine chloride in forearmresistance vessels of normotensive humans subjects,suggesting that thiol is not a limiting factor for EDRFactivity.1*0
In the present study the forearm blood flow responseto methacholine chloride was similar during the placeboand captopril treatment periods. It was noted, however,that captopril tended to cause a mild augmentation ofblood flow at low doses of methacholine, but this did notreach statistical significance. The dose-response curvesto methacholine chloride during placebo and enalapriltreatment periods were virtually superimposable. Thus,neither ACE inhibitor significantly affected endotheli-um-dependent vasodilation in our hypertensive pa-tients. In addition, we combined both treatment groupsto increase statistical power and address the issue ofwhether ACE inhibitors, as a class, restore endotheli-um-dependent vasodilation. No significant difference inendothelium-dependent vasodilation was found be-tween the placebo and drug treatment periods, indicat-ing that up to 7 to 8 weeks of treatment with an ACEinhibitor does not substantially affect endothelium-de-pendent vasodilation in patients with hypertension.
Limitations of the StudyNegative findings suggest that the original hypothesis
was not correct or that the methods used to test thehypothesis were inadequate. The variability of repeatedforearm blood flow measurements during a 2-month
by guest on April 22, 2018
504 Hypertension Vol 24, No 4 October 1994
period may have precluded our ability to detect a smallchange in endothelium-dependent vasodilation duringthe treatment phase. It is possible that the mild, albeitstatistically insignificant, improvement in endothelialfunction observed during captopril treatment wouldhave become significant if a larger number of subjectshad been enrolled. One might argue also that we did nottreat our hypertensive patients long enough or wellenough to realize a beneficial effect on endothelialfunction. Luscher et al20 found that antihypertensivetreatment improved endothelium-dependent vasodila-tion in Dahl salt-sensitive rats after 2 weeks of treat-ment. Clozel et al22 reported that 4 days of cilazapriladministration improved endothelium-dependent relax-ation in spontaneously hypertensive rats. Furthermore,Hirooka et al21 reported their observations after only asingle dose of captopril. Thus, we doubt, but cannot becertain, that endothelium-dependent vasodilationwould be restored after a longer treatment period. It isalso possible that the magnitude of the blood pressurereduction was insufficient to affect endothelium-depen-dent vasodilation. We did strive to reduce blood pres-sure to normotensive levels, ie, less than 140/90 mm Hg.However, the posttreatment blood pressure in our hy-pertensive patients still exceeded that of our normoten-sive control subjects. It is possible that we did not reachthe threshold at which a restoration of endothelialfunction would have occurred. We think this unlikelysince there was no overall tendency to improve endo-thelium-dependent relaxation during ACE inhibitortreatment in the group as a whole or even in thosepatients with the greatest reduction in blood pressure.
In conclusion, our data confirm observations madepreviously by others that endothelium-dependent vaso-dilation is abnormal in forearm resistance vessels ofpatients with hypertension. The new findings are that 2months of antihypertensive therapy with an ACE inhib-itor does not improve endothelium-dependent vasodila-tion in these patients, regardless of whether or not asulfhydryl group is present. It remains possible, albeitunlikely, that longer or more aggressive ACE inhibitortherapy may have favorably affected endothelium-de-pendent vasodilation in these individuals. Our data donot enable us to make conclusions about other antihy-pertensive agents. Thus, ACE inhibitors are effectiveantihypertensive agents with many favorable properties;however, augmentation of endothelium-derived vasodi-lation does not appear to be one of them.
AcknowledgmentsThis study was supported by a grant from the Bristol-Myers
Squibb Company. Dr Creager is a recipient of a NationalHeart, Lung, and Blood Institute academic award in systemicand pulmonary vascular disease (HL-02663). We gratefullyacknowledge Sharon Coleman for technical assistance andJoanne Normandin for manuscript preparation.
References1. Vane JR, Angaard EE, Botting RM. Regulatory function of the
vascular endothelium. N Engl J Med. 1990-,323:27-36.2. Furchgott RF, Zawadzki JV. The obligatory role of endothelial
cells in the relaxation of arterial smooth muscle by acetylcholine.Nature. 1980;288:373-376.
3. Luscher TF, Raij L Vanhoutte PM. Endothelium-dependentvascular responses in normotensive and hypertensive Dahl rats.Hypertension. 1987;9:157-163.
4. Mayhan WG, Faraci FM, Heistad DD. Impairment of endotheli-um-dependent responses of cerebral arterioles in chronic hyper-tension. Am J Physiol. 1987;253:H1435-H1440.
5. Carvalho MH, Scivoletto R, Fortes ZB, Nigro D, Cordellini S.Reactivity of aorta and mesenteric microvessels to drugs in spon-taneously hypertensive rats: role of the endothelium. J Hypertens.1987;5:377-382.
6. Under L, Kiowski W, Buhler FR, Luscher TF. Indirect evidencefor release of endothelium-derived relaxing factor in humanforearm circulation in vivo: blunted response in essential hyper-tension. Circulation. 1990;81:1762-1767.
7. Panza JA, Quyyumi AA, Brush JE Jr, Epstein SE. Abnormalendothelium-dependent vascular relaxation in patients withessential hypertension. N Engl J Med. 1990-,323:22-27.
8. Carver A, Collier J, Moncada S, Vallance P. Effect of local intra-arterial N^monomethyl-L-arginine in patients with hypertension:the nitric oxide dilator mechanism appears abnormal. J Hypertens.1992;10:1025-1031.
9. Panza JA, Casino PR, Kilcoyne CM, Quyyumi AA. Role of endo-thelium-derived nitric oxide in the abnormal endothelium-dependent vascular relaxation of patients with essential hyper-tension. Circulation. 1993;87:1468-1474.
10. Treasure CB, Manoukian SV, Klein JL, Vita JA, Nabel EG,Renwick GH, Selwyn AP, Alexander RW, Ganz P. Epicardialcoronary artery responses to acetylcholine are impaired in hyper-tensive patients. Ore Res. 1992;71:776-781.
11. Treasure CB, Klein JL, Vita JA, Manoukian SV, Renwick GH,Selwyn AP, Ganz P, Alexander RW. Hypertension and left ven-tricular hypertrophy are associated with impaired endothelium-mediated relaxation in human coronary resistance vessels.Circulation. 1993;87:86-93.
12. Brush JE Jr, Faxon DP, Salmon S, Jacobs AK, Ryan TJ. Abnormalendothelium-dependent coronary vasomotion in hypertensivepatients. J Am Coll Cardiol. 1992;19:809-815.
13. Shepherd JT. Franz Volhard lecture: increased systemic vascularresistance and primary hypertension: the expanding complexity.J Hypertens. 199O;8:S15-S27.
14. Taddei S, Virdis A, Mattei P, Arzilli F, Sarvetti A. Endothelium-dependent forearm vasodilation is reduced in normotensivesubjects with familial history of hypertension. / CardiovascPharmacol. 1992;20:S193-S195.
15. Panza JA, Quyyumi AA, Callahan TS, Epstein SE. Effect of anti-hypertensive treatment on endothelium-dependent vascularrelaxation in patients with essential hypertension. J Am CollCardiol. 1993;21:1145-1151.
16. Lockette W, Otsuka Y, Carretero O. The loss of endothelium-dependent vascular relaxation in hypertension. Hypertension. 1986;8(suppl II):II-61-II-66.
17. Nigro D, Fortes ZB, Carvalho MH, Scivoletto R. Chronic but notacute treatment with hydralazine reverses the decreased endothe-lium-dependent responses in spontaneously hypertensive rats. CUnExp Hypertens. 1989;ll:573-586.
18. Shirasaki Y, Kolm P, Nickols GA, Lee TJ. Endothelial regulationof cyclic GMP and vascular responses in hypertension. J PharmacolExp Ther. 1988;245:53-58.
19. Dohi Y, Criscione L Luscher TF. Renovascular hypertensionimpairs formation of endothelium-derived relaxing factors and sen-sitivity to endothelin-1 in resistance arteries. BrJ Pharmacol. 1991;104:349-354.
20. Luscher TF, Vanhoutte PM, Raij L. Antihypertensive treatmentnormalizes decreased endothelium-dependent relaxations in ratswith salt-induced hypertension. Hypertension. 1987;2(suppl III):III-193-III-197.
21. Hirooka Y, Imaizumi T, Masaki H, Ando S, Harada S, MomoharaM, Takeshita A. Captopril improves impaired endothelium-dependent vasodilation in hypertensive patients. Hypertension.1992;20:175-180.
22. CJozel M, Kuhn H, Hefti F. Effects of angiotensin convertingenzyme inhibitors and of hydralazine on endothelial function inhypertensive rats. Hypertension. 1990; 16:532-540.
23. Mombouli JV, Illiano S, Nagao T, Scott-Burden T, Vanhoutte PM.Potentiation of endothelium-dependent relaxations to bradykininby angiotensin I converting enzyme inhibitors in canine coronaryartery involves both endothelium-derived relaxing and hyperpo-larizing factors. Ore Res. 1992;71:137-144.
24. Wiemer G, Sch61kens BA, Becker RHA, Busse R. Ramiprilatenhances endothelial autacoid formation by inhibiting breakdownof endothelium-derived bradykinin. Hypertension. 1991;18:558-563.
by guest on April 22, 2018
Creager and Roddy Endothelial Function in Hypertension 505
25. van Gilst WH, de Graeff PA, de Leeuw MJ, Scholtens E, WesselingH. Converting enzyme inhibitors and the role of the sulf hydrylgroup in the potentiation of exo- and endogenous nitrovasodi-lators. J Cardiovasc Pharmacol. 1991;18:429-436.
26. Lahcra V, Khraiba AA, Romero JC. Sulfhydryl group donorspotentiate the hypotensive effect of acetylcholine in rats.Hypertension. 1993;22:156-160.
27. Shultz PJ, Raij L. Effects of antihypertensive agents on endothe-lium-dependent and endothelium-independent relaxations. Br JClin Pharmacol. 1989;28:151S-157S.
28. Creager MA, Cooke JP, Mendelsohn MD, Gallagher SG, ColemanSM, Loscalzo J, Dzau VJ. Impaired vasodilation in forearmresistance vessels in hypercholesterolemic humans. J Clin Invest.1990;86:228-234.
29. Hokanson DE, Sumner DS, Strandness DE Jr. An electricallycalibrated plethysmograph for direct measurement of limb bloodflow. IEEE Trans Biomed Eng. 1975;22:25-29.
30. Angus JA, Wright CE, Cocks TM, Satoh K, Campbell G.Endothelium-dependent relaxation is unaltered by hypertension,cholesterol or intimal thickening. Clin Exp Pharmacol Physwl.1986;13:289-293.
31. Vallance P, Collier J, Moncada S. Effects of endothelium-derivednitric oxide on peripheral arteriolar tone in man. Lancet. 1989;2:997-1000.
32. Mayhan WG. Impairment of endothelium-dependent dilatation ofbasilar artery during chronic hypertension. Am J Physiol. 199O;259:H1455-H1462.
33. Diederich D, Yang ZH, Buhler FR, Luscher TF. Impaired endo-thelium-dependent relaxations in hypertensive resistance arteries
involve cyclooxygenase pathway. Am J Physiol. 1990;258:H445-H451.
34. Hayakawa H, Hirata Y, Su2uki E, Sugimoto T, Matsuoka H,Kikuchi K, Nagano T, Hirobe M, Sugimoto T. Mechanisms foraltered endothelium-dependent vasorelaxation in isolated kidneysfrom experimental hypertensive rats. Am J Physiol. 1993;264:H1535-H1541.
35. Fujii K, Tominaga M, Ohmori S, Kobayashi K, Koga T, Takata Y,Fujishima M. Decreased endothelium-dependent hyperpolar-ization to acetylcholine in smooth muscle of the mesenteric arteryof spontaneously hypertensive rats. Ore Res. 1992;70:660-669.
36. Yang ST, Mayhan WG, Faraci FM, Heistad DD. Endothelium-dependent responses of cerebral blood vessels during chronichypertension. Hypertension. 1991;17:612-618.
37. Kiowski W, Under L, Neusch R, Martina B. Effects of angiotensin-converting enzyme inhibition on endothelial function and vasculardesign in hypertension. Circulation. 1992;86(suppl I):I-559.Abstract.
38. Mombouli JV, NephtaJi M, Vanhoutte PM. Effects of the con-verting enzyme inhibitor cilazaprilat on endothelium-dependentresponses. Hypertension. 1991;18(suppl II):II-22-II-29.
39. Goldschmidt JE, Tallarida RJ. Pharmacological evidence that cap-topril possesses an endothelium-mediated component of vasodi-lation: effect of sulfhydryl groups on endothelium-derived relaxingfactors. J Pharmacol Exp Ther. 1991;257:1136-1145.
40. Creager MA, Roddy M-A, Coleman SM, Stamler JS. Effect ofN-acetylcysteine on endothelium-dependent vasodilation inhumans. Circulation. 1992;86(suppl I):I-62O. Abstract.
by guest on April 22, 2018
M A Creager and M A RoddyEffect of captopril and enalapril on endothelial function in hypertensive patients.
Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 1994 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/01.HYP.24.4.499
http://hyper.ahajournals.org/content/24/4/499World Wide Web at:
The online version of this article, along with updated information and services, is located on the
is online at: Hypertension Information about subscribing to Subscriptions:
http://www.lww.com/reprints Information about reprints can be found online at: Reprints:
document. Permissions and Rights Question and Answer process is available in the
Request Permissions in the middle column of the Web page under Services. Further information about thisOffice. Once the online version of the published article for which permission is being requested is located, click
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialHypertension Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:
by guest on April 22, 2018