Heart 1996;75:261-266

Variable coronary vasomotor responses toacetylcholine in patients with normal coronaryarteriograms: evidence for localised endothelialdysfunction

D Tousoulis, G Davies, D C Lefroy, AW Haider, T Crake

AbstractObjective-The vasomotor responses ofthe epicardial coronary arteries to acetyl-choline were examined in patients withnormal coronary arteries and chest pain.Design-Quantitative angiography wasused to measure minimum lumen diame-ter ofproximal and distal coronary arterysegments at baseline, during intracoro-nary infusion of acetylcholine (10-'-10-3moMl), and following an intracoronarybolus (2 mg) of isosorbide dinitrate.Patients-Coronary arteriograms wereobtained in 15 patients (mean (SEM) age48 (10) years) with normal coronaryarteries and chest pain.Main results-In response to the low con-centrations of acetylcholine (10-7-10-6moMl) 20 (61%) distal and 11 (41%) proxi-mal segments showed dilatation (group1), whereas 13 (39%) distal segments and14 (52%) proximal segments showed con-striction (group 2) (P < 005 v group 1). Ingroup 1, the maximum dilatation inducedby acetylcholine in the proximal and dis-tal segments was 7-83 (1-19)% and 11-6(2.2)% respectively. In group 2, the maxi-mum constriction at higher concentrationwas 16-55 (3'3)% and 33-11 (11.63)% inthe proximal and distal segments respec-tively. The two different patterns of thevasomotor response coexisted in eight(53%) of the 15 patients. Intracoronaryisosorbide dinitrate caused a greaterincrease in the coronary luminal diame-ter of distal segments than in proximalsegments in group 1 (25.63 (5.16)% v12-43 (3.48)%, P < 0.01) but not in group 2(12-65 (2.53)% v 10*82 (3-33)%.Conclusions-Constriction and dilatationmay occur in proximal and distal coro-nary artery segments, suggesting localareas of endothelial dysfunction, inresponse to acetylcholine in patients withchest pain and angiographically normalcoronary arteries.

(Heart 1996;75:261-266)

Keywords: acetylcholine; endothelium; vasoconstric-tion; vasodilation

In patients with angiographically normal epi-cardial coronary arteries intracoronary acetyl-choline causes coronary artery dilatation. 1-3Experimental and clinical studies have shown

that epicardial coronary artery dilatation is dueto the release of nitric oxide in response toacetylcholine.46 In patients with atheromatouscoronary arteries acetylcholine causes coro-nary artery constriction even at low doses3 7-9

and the coronary artery responses to acetyl-choline are dependent on the patient's charac-teristics and the extent of coronaryatherosclerosis.101' Experimental studies havesupported these findings and it has conse-quently been argued that dilatation of a coro-nary artery in response to acetylcholine isconsistent with normal endothelial functionand constriction with endothelial dysfunc-tion.'2-14

Other studies have shown that in patientswith chest pain and normal coronary arteriesthe vasodilator response to acetylcholine maybe attenuated or vasoconstriction may evenoccur.' 15-16 Furthermore, El-Tamimi et al. 17have recently reported that in patients withcoronary artery disease both localised dilata-tion and constriction at a given dose of acetyl-choline can be observed in the sameatheromatous vessel suggesting that in thesearteries there are localised areas with normalendothelial function.The effects of acetylcholine in normal coro-

nary arteries seem to be dose dependent, con-striction occurs at high dose in vessels dilatedat lower doses and the effects of a particulardose seem to vary between patients.26Localised dilatation and constriction at a givendose in the same coronary artery has neverbeen reported in patients with angiographi-cally normal coronary arteries. In this study weexamined the segmental coronary vasomotoreffects of acetylcholine in patients with chestpain and normal coronary arteriograms.

Patients and methodsFifteen (six men; mean (SEM) (range) age 48(10) (32-68) years) consecutive patients witha history of chest pain in exertion and angio-graphically normal coronary arteries werestudied. Four patients had a positive treadmillexercise test (> 0 1 mV ST segment depres-sion between 5 and 7 METS using the modi-fied Bruce protocol) (table 1). Patients wereallowed to use only sublingual glyceryl trini-trate as necessary, but no study was performedwithin 3 h of administration of the drug.

Patients were excluded fro the study if theyhad diabetes mellitus, left ventricular hyper-trophy (as assessed by echocardiography), leftventricular dysfunction (left ventricular ejec-

Cardiology Unit, RoyalPostgraduate MedicalSchool, HanmmersmithHospital, LondonD TousoulisG DaviesD C LefroyAW HaiderSt Bartholomew'sHospital, WestSmithfield, LondonT CrakeCorrespondence to:Dr G Davies, CardiologyUnit, HammersmithHospital, Du Cane Road,London W12 OHS.Accepted for publication20 September 1995

261

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from

Tousoulis, Davies, Lefroy, Haider, Crake

Table 1 Clinical characteristics of the patients

Age Family history Exercise testPatient no (years) Sex Hypertension Smoking ofCAD Hypercholesterolaemia result

1 45 F + - + - +ve2 60 F + - - - -ve3 41 F - + + - -ye4 49 M - + - -ye5 46 M - - + + -ve6 48 M - - + - -ve7 58 F - + - - -ve8 59 F - - + - -ve9 33 M - - + - -ve10 43 M - + + - -ve11 32 M - - + - -ve12 42 M + - - - +ve13 68 M - + - - +ve14 46 F - + - - -ye15 47 F - + - + +ve

CAD, coronary artery disease; +, present; -, absent; +ve, positive; -ye, negative.

tion fraction < 50%), or valvular heart disease.Patients with a clinical history that suggestedcoronary artery spasm or with ergonovine-induced coronary artery spasm were excluded.Patients had at least one of the followingrisk factors for coronary artery disease:(a) hypercholesterolaemia: defined as a fast-ing serum cholesterol > 5-69 mmol/l (> 220mg/dl) before treatment; (b) hypertension:requiring antihypertensive drug treatment bythe primary physician; (c) cigarette smoking:patients who were currently smoking or hadstopped less than 3 months before the study;and (d) family history: a parent or siblingunder the age of 60 years of the patient whohad documented evidence of coronary arterydisease such as a history of myocardial infarc-tion, coronary artery bypass surgery, anginapectoris, sudden death, or angiographicallydocumented coronary artery disease. 18

The study protocol was approved by theResearch Ethics Committee of theHammersmith Hospital and informed writtenconsent was obtained from all patients.

STUDY PROTOCOLTwo electrocardiogram leads were monitoredcontinuously throughout the study. Coronaryarteriography was performed through the rightfemoral artery using 8 French Judkinscatheters. Femoral arterial blood pressure andheart rate were recorded during the last 30 s ofeach infusion period. Following the diagnosticstudy, an optimal radiographic projection was

chosen to visualise the coronary arteries to bestudied, and the position of the image intensi-fier was subsequently kept constant. The leftcoronary artery only was studied in ninepatients, the right coronary artery only in four,the right and left coronary arteries in two.Infusions were administered through theJudkins catheters at room temperature at a

rate of 1 ml/min using a syringe pump

(Perfusor; Braun-Melsungen, Germany). Allpatients received a single 2 min infusion ofvehicle solution (0 9% saline), followed by 2min infusions of acetylcholine (1 ml/min) inincremental intracoronary concentrationsfrom 10-I to 10-5 mol/l in two patients, to10-4 mol/l in four, and to 10-3 mol/l in nine,followed by an intracoronary bolus of isosor-bide dinitrate (2 mg in 2 ml saline).

Coronary angiography was performed with

a hand injection of non-ionic contrast medium(6-8 ml) at baseline, after each infusion, and 2min after isosorbide dinitrate. The catheterwas emptied before each angiogram to avoidbolus administration of infused solutions.

QUANTITATIVE CORONARY ANGIOGRAPHYThe arterial segments in an end diastolic framefrom each arteriogram were analysed in ran-dom order using quantitative computerisedanalysis with an automated edge contourdetection analysis system (ComputerisedAngiographic Analysis System, Version 2V2;Pie Data Medical, Maastricht). 19 20 The angio-graphic catheter was used as a scaling deviceand this together with the pincushion distor-tion correction allowed the diameters to berecorded as absolute values. The diameter ofangiographically proximal and distal segmentswas measured from the arteriograms recordedat baseline and after saline, acetylcholine andnitrate administration. The proximal left ante-rior descending coronary artery diameter wasmeasured just beyond the origin of the arteryand the distal diameter was measured just dis-tal to its second diagonal branch; the proximalleft circumflex coronary artery diameter wasmeasured just beyond the origin of the arteryand the distal diameter just beyond the origin ofthe second marginal branch; the proximalright coronary artery diameter was measuredjust beyond the origin of the artery and thedistal diameter just beyond the origin of poste-rior descending branch. In the 15 patients, 1 1left and six right coronary arteries were stud-ied. Eleven proximal and distal left anteriordescending coronary artery segments wereanalysed. Ten proximal left circumflex (in onepatient the image quality of the left circumflexwas not suitable for analysis) and 14 distal leftcircumflex segments were analysed; in threepatients with a dominant left circumflex coro-nary artery an additional distal segment in thesecond marginal branch was analysed. Sixproximal and eight distal right coronary arterysegments were analysed; in two patients withpronounced dominant right coronary arteriesone additional distal segment was analysed.

Quantitative analysis of coronary arteri-ograms was carried out by two independentobservers, who blindly reanalysed the films ata remote time for reproducibility of themethod. No significant intra- or inter-observer

262

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from

Variable coronary vasomotor responses to acetylcholine in patients with normal coronary arteriograms: evidence for localised endothelial dysfunction

Table 2 Reactivity ofproximal and distal segments in patients with normal coronaryarteries to intracoronary administration of acetylcholine and nitrates

Minimum luminal diameter (mm)

Group 1 Group 2

Proximal Distal Proximal Distalsegments segments segments segments

Baseline 3-53 (0-2) 1-32 (0-07) 3-15 (0-1) 1 51 (0-06)Saline 3-58 (0-2) 1-31 (0-07) 3-19 (0-1) 1.53 (0-05)Acetylcholine (mol/l)

10-7 3 55 (0-2) 1-39 (0 07)t 2-94 (0 1)* 1-35 (0 05)*10-6 3 71 (0-2)t 1-48 (0 08)* 2-95 (0 1)* 1 36 (0-05)*10-s 3-81 (02)* 1-41 (0-08)t 2-87 (0-1)* 1-28 (006)*10-4 3-67 (0-2) 1-29 (0-08) 2-83 (02)* 1-26 (005)*10-3 3-60 (0-3) 1-04 (0-13)t 2-63 (02)* 0-97 (0 19)*

ISDN 3-93 (0-3)* 1-62 (0 10)* 3-52 (0-2)* 1-68 (0-07)*

Values are mean (SEM). *P < 0-01 versus baseline; tP < 0-05 versus baseline. ISDN, isosorbidedinitrate.

variability was found (analysis of variance F =0-34, P = 0 75). Coronary artery segmentswere divided into two groups according totheir response () 5%) to low dose acetyl-choline ((10-7-10-6 mol/l).

STATISTICAL ANALYSISData are expressed as mean (SEM). A oneway analysis of variance with an allowance forrepeated measures was performed for compar-isons of haemodynamic data and coronaryartery diameters. A two tailed Student's t testfor paired observations with the Bonferronicorrection was used to test differences amongmeans when an F value was significant. AP value < 0-05 was considered significant.

ResultsMean aortic pressure and heart rate remainedunchanged throughout intracoronary adminis-tration of acetylcholine compared with that ofbaseline (92 (5) mm Hg and 72 (4) beats/min v89 (5) mm Hg and 69 (5) beats/min respec-tively). In three patients ST segment depression(> 0.1 mV), which was accompanied by chestpain in two, occurred at infused concentrationsof acetylcholine of 10-5, 10-5, and 10-4 mol/lrespectively and this was relieved by intracoro-nary administration of isosorbide dinitrate.

All patients had at least one risk factor forcoronary artery disease and five had two. Twopatients had hypercholesterolaemia, sevenwere smokers, eight had a relevant family his-tory, and three had hypertension (table 1).

CORONARY ARTERY DIAMETER CHANGES INRESPONSE TO ACETYLCHOLINE AND NITRATEDuring saline infusion no significant changeswere observed in the proximal and distal seg-ments (table 2).

Three patterns of response to the low dosesof acetylcholine (10-7-10-6 mol/l) wereobserved; constriction only in three patients,dilatation only in four, and constriction anddilatation in eight (in the same vessel in sixand in different vessels in two). At these lowconcentrations of acetylcholine 33 distal and27 proximal segments were studied and coro-nary artery segments were divided into twogroups, according to their response to lowdoses of acetylcholine (table 2). Twenty(61%) of 33 distal and 11 (41%) of 27 proxi-mal segments dilated > 5% (group 1), and 13(39%) of 33 distal segments and 14 (52%) of27 proximal segments constricted > 5%(group 2). Two proximal segments did notchange. The frequency of dilatation of distalvessels was greater than constriction (P <0 05). The maximum magnitude of dilatationin group 1 in the proximal segments (7-83(1.19)%) was observed at 10-5 mol/l and thatin the distal segments (11 -6 (2 2)%) at 10-6mol/l (figs 1 and 2).

At maximum acetylcholine concentration(10-3 mol/l) in group 1 segments, eight (73%)of 11 distal and two (40%) of five proximalsegments constricted (P < 0 001 comparedwith the low concentration response) and theremainder dilated or did not change. In group2, all distal (n = 9) and proximal (n = 9) seg-ments constricted further at 10-3 mol/l.During the maximum dose of acetylcholineconstriction was greater in group 2 segmentsthan in group 1 (proximal segments - 16-55(3 3)% v - 0 7 (2 63)%, P < 0 05; distal seg-ments -33-11 (11-63)% v - 15-78 (8 2)%)and greater in distal than in proximal segments(P < 0 05) for both groups (fig 2).

Increase in the coronary luminal diametercaused by intracoronary isosorbide dinitratewas greater in distal segments than in proximalsegments in group 1 (25.63 (5.16)% v 12-43(3 48)%, P < 0 01) but not in group 2 (12-65(2.53)% v 10-82 (3-33)%, P = not significant)(fig 2). In three patients focal constriction(> 30% baseline minimum luminal reduction)

Figure 1 Dose dependentresponses to acetylcholine(mean (SEM) changes insegment diameter inabsolute measurements(mm)) of the proximaland distal segments (tP <0 05; *P < 0 01 versusbaseline). Two patterns ofvasomotion were observedin response to lowacetylcholineconcentrations in proximaland distal segments(10-7-10-6 molll):dilatation (group 1) orconstriction (group 2).NS, normal saline; ISDN,isosorbide dinitrate.

4

EE

E

3

2

Proximal segments

t 2.0

1.5 [-

Distal segments

-_--- Group 1|-- Group 2

t

1-o0

0-5 _

NS 10-7 106 10-5 10-4 10-3 ISDNAcetylcholine (mol/l)

NS 10-7 10 10c-5 -4 -3 ISDNAcetylcholine (mol/1)

*

263

*

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from

Tousoulis, Davies, Lefroy, Haider, Crake

Figure 2 Dose dependentresponses to acetylcholine(mean (SEM) per centchange in coronary luminaldiameterfrom baseline) ofthe proximal and distalsegmnents (tP < 0 05; *P <0-01 versus saline). Twopatterns of vasomotion wereobserved in response to lowacetylcholine concentrationsin proximal and distalsegments (10-7-10-6molll): dilatation (group 1)or constriction (group 2).Values in parentheses arenumbers ofsegmentsstudied. Abbreviations asgiven in fig 1.

Proximal segments

NS 10-7 10-6 10 10-4 10-3 ISDN

Acetylcholine (mol/1)

Distal segments

-0-- Group 1

- - Group 2

(20)(20) *

(13) (13)(13) (12)

-40 - T '9'*l

NS 10-7 10-6 10-5 io-4Acetylcholine (mol/1)

reversed by nitrates was observed in five seg-ments (four mid, and one distal) at sites whichwere not initially selected for analysis. At thesesegments the minimum luminal diameter was

2-26 (0 25) mm at baseline, 0-66 (0 27) mm(mean reduction 66 (13)%) after acetylcholine(P < 0-05), and 2-36 (0.25) mm after nitrateadministration.

DiscussionIn patients with chest pain and angiographi-cally normal coronary arteries the effects ofintracoronary acetylcholine are variable.Constriction and dilatation coexisted in mostpatients not only in different coronary arterybranches but also in different segments of thesame branch. All segments dilated in responseto isosorbide dinitrate, a direct acting smoothmuscle dilator. These results suggest that inpatients with chest pain most angiographicallynormal coronary arteries have areas of dys-functional endothelium.'I

EFFECTS OF ACETYLCHOLINE INANGIOGRAPHICALLY NORMAL CORONARYARTERIESAcetylcholine is an endothelium dependentvasodilator as well as a potent vascular smoothmuscle vasoconstrictor.421 Previous clinicalstudies3 have shown that angiographically nor-mal coronary arteries dilate but atheroscleroticcoronary arteries constrict in response to theintracoronary infusion of acetylcholine. Wernset al found that acetylcholine caused dilata-tion of angiographically normal coronaryarteries and constriction of angiographicallynormal segments in patients with coronary

artery disease suggesting that these normalsegments are functionally abnormal. An inter-action between cardiovascular risk factors andendothelial dysfunction has been reported pre-viously. Acetylcholine has been shown toinduce vasoconstriction in patients with nor-mal coronary arteries and hypercholestero-laemia with increased plasma concentrationsof low density lipoproteins,'° 1122 althoughthese authors did not investigate the segmentaleffects of acetylcholine in their patients.Experimental studies have shown that choles-

terol fed animals develop selective attenuationof endothelium dependent vasodilatationbefore and after developing histological evi-dence of frank atherosclerosis.'2 13 Acute andchronic increase in blood pressure leads toimpairment of endothelium dependent vasodi-latation, perhaps due to endothelial injury.23-25Patients with angiographically normal arteriesmay have intimal thickening of variable degreeand therefore will produce a variable responseto vasoactive substances.

Endothelial dysfunction appears very earlyin the development of coronary artery dis-ease." Our observations are consistent with apatchy impairment of endothelial function dueto early atherosclerosis, so that the directsmooth muscle constrictor effect predomi-nates in some segments. Such atherosclerosiscannot be detected by coronary angiography.In our study it is likely, therefore, thatthe smooth coronary segments that constrictedin response to acetylcholine had early athero-sclerosis which was not detectable by angiog-raphy.

PROXIMAL AND DISTAL SEGMENT RESPONSESTO ACETYLCHOLINE AND ENDOTHELIALFUNCTIONAlthough in the present study acetylcholinecaused vasodilatation and vasoconstriction inproximal and distal segments, the percentageresponse was significantly greater in distalsegments. These results are consistent withprevious studies showing enhanced distalvasoconstriction in response to serotonin,20acetylcholine,9 and ergonovine.26 The mecha-nism underlying this variable magnitude ofresponse of the coronary arteries of differentsizes to certain stimuli is unclear. This differ-ential response may also be modified by thedistribution of atherosclerosis which is more

prevalent and severe in proximal segments.The observed response to acetylcholine is

consistent with a patchy impairment ofendothelial function due to atherosclerosis so

that the direct smooth muscle constrictioneffect predominates in some segments. It canalso be explained by an increased sensitivity ofthe vascular smooth muscle in the region ofatherosclerosis to acetylcholine, a defect in the

(20)*

(20)*

1o-3 ISDN

264

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from

Variable coronary vasomotor responses to acetylcholine in patients with normal coronary arteriograms: evidencefor localised endothelial dysfunction

coupling mechanisms between endotheliumderived relaxing factor and smooth musclecells, or an increased release of endotheliumderived vasoconstricting factors.

CLINICAL IMPLICATIONSThe segmental heterogeneity of the responseof apparently normal coronary arteries may bea means of detecting early atherosclerosis oran alteration of endothelial function whichprecedes the development of atherosclerosis,as suggested by other studies.11 The techniqueis safe and rapidly performed and does notrequire instrumentation of the coronary arteriesas do other techniques for detecting early ath-erosclerosis such as intracoronary ultrasoundand therefore the risk of vascular trauma isreduced.2728 It also permits examination of theentire artery including the distal branches.Although detection of early atherosclerosismay not be immediately relevant to a patient'ssymptoms, it could be of prognostic signifi-cance. Angiographically normal coronaryarteries are frequently found in patients withchest pain undergoing cardiac catheterisa-tion.'9 Our study has shown that at least someof these patients may have coronary artery seg-ments which behave abnormally in response toacetylcholine; indeed, these segments have thesame response as atheromatous coronaryartery segments.'7 The clinical implications ofthis finding remain speculative at the presenttime and follow up studies are required toascertain whether such patients are atincreased risk of developing flow limiting coro-nary stenosis or coronary thrombosis perhapsat the sites of abnormal endothelial behaviour.Further studies will determine whether riskfactor modification results in restoration ofnormal endothelial function.

STUDY LIMITATIONSOur study population was relatively small andrepresents a selected group of patients withchest pain syndromes referred for cardiologicalassessment who had angiographically normalcoronary arteriograms. The coronary arterieswere assessed by coronary arteriography andthey were therefore angiographically "nor-mal". It must be emphasised that coronaryarteriography has important limitations in thedetection of early coronary atherosclerosis andpossibly the pattern of response to acetyl-choline that we observed was due to localisedareas of early atherosclerosis. In this respectintravascular ultrasound may have been a use-ful method to identify such areas of earlyatheroma.

ConclusionsIn patients with chest pain and normal coro-nary arteriograms the response to acetyl-choline is heterogeneous, not only betweenpatients, but between vessels in an individualpatient and between segments in an individualvessel. This evidence of patchy endothelialdysfunction may be indicative of early under-lying atherosclerosis which cannot be detectedby coronary angiography.

1 Yasue H, Matsuyama K, Matsuyama K, Okumura K,Morikami Y, Ogawa H. Responses of angiographicallynormal human coronary arteries to intracoronary injec-tion of acetylcholine by age and segment. Possible role ofearly atherosclerosis. Circulation 1990;81:482-90.

2 Egashira K, Inou T, Hirooka Y, Yamada A, Urabe Y,Takeshita A. Evidence of endothelium-dependentvasodilation in patients with angina pectoris and normalcoronary angiograms. N EnglJ Med 1993;328: 1659-64.

3 Ludmer PL, Selwyn AP, Shook TL, Wayne RR, MudgeGH, Alexander RW, et al. Paradoxical vasoconstrictioninduced by acetylcholine in atherosclerotic coronaryarteries. NEnglJMed 1986;315:1046-51.

4 Furchgott RF, Zawadzki JV. The obligatory role of theendothelial cells in the relaxation of arterial smooth mus-cle by acetylcholine. Nature 1980;288:373-6.

5 Palmer RM, Ferrige AG, Moncada S. Nitric oxide releaseaccounts for the biological activity of endothelial-derivedrelaxing factor. Nature 1987;327:524-6.

6 Quyyumi AA, Cannon RO, Panza JA, Diodato JG, EpsteinSE. Endothelial dysfunction in patients with chest painand normal coronary arteries. Circulation 1992;86:1864-71.

7 Hodgson JMcB, Marshall J. Direct vasoconstriction andendothelium-dependent vasodilation: mechanisms ofacetylcholine effects on coronary flow and arterial diame-ter in patients with nonstenotic coronary arteries.Circulation 1989;79: 1043-51.

8 Werns SW, Walton JA, Hsia HH, Nabel EG, Sans ML,Pitt B. Evidence of endothelium dysfunction in angio-graphically normal coronary arteries in patients withcoronary artery disease. Circulation 1989;79:287-91.

9 Newman CM, Maseri A, Hackett D, El-Tamimi HM,Davies GJ. Response of angiographically normal and ath-erosclerotic left anterior descending coronary arteries toacetylcholine. Am Jf Cardiol 1990;66: 1070-6.

10 Vita JA, Treasure CB, Nabel EG, McLenachan JM, FishDR, Yeung AC, et al. Coronary vasomotor response toacetylcholine relates to risk factors for coronary arterydisease. Circulation 1990;81:491-7.

11 Zeiher AM, Drexler H, Wollschlager H, Just H.Modulation of coronary vasomotor tone in humans.Progressive dysfunction with different early stages ofcoronary atherosclerosis. Circulation 1991;83:391-401.

12 Andrews HE, Bruckdorfer KR, Dunn RC, Jacobs M. Low-density lipoproteins inhibit endothelium-dependentrelaxation in rabbit aorta. Nature 1987;327:237-9.

13 Venbeuren TJ, Jordaens FH, Zonnekeyn LLO, Van HoveCE, Coene MC, Herman AG. Effects of hypercholes-terolemia on vascular reactivity in the rabbit. Circ Res1986;58:552-64.

14 Shimokawa H, Vanhoutte PM. Impaired endothelium-dependent relaxation to aggregating platelets and relatedvasoactive substances in porcine coronary arteries inhypercholesterolemia and atherosclerosis. Circ Res 1989;64:900-14.

15 Motz W, Vogt M, Rabenau 0, Scheler S, Luckhoff A,Strauer BE. Evidence of endothelial dysfunction in coro-nary resistance vessels in patients with angina pectorisand normal coronary arteries. Am Jf Cardiol 1991;68:996-1003.

16 Vrints CJM, Bult H, Hitter E, Herman AG, Snoeck J.Impaired endothelium-dependent cholinergic coronaryvasodilation in patients with angina and normal coronaryarteriograms. JAm Coil Cardiol 1992;19:21-31.

17 El Tamimi H, Mansour M, Wargorich TJ, Hill JA,Kerensky RA, Conti RC, et al. Constrictor and dilatorresponses to intracoronary acetylcholine in adjacent seg-ments of the same coronary artery in patients with coro-nary artery disease. Endothelial function revisited.Circulation 1994;89:45-51.

18 Barrett-Connor E, Khaw KT. Family history of heartattack as an independent predictor of death due to car-diovascular disease. Circulation 1984;69: 1065-9.

19 Reiber JHC, Serruys PW, Koouman C, Wijns W, Slager C,Gerbrands JJ, et al. Assessment of short-, medium, andlong-term variations in arterial dimensions from com-puter-assisted quantitation of coronary cineangiograms.Circulation 1985;71:280-8.

20 Tousoulis D, Davies G, McFadden E, Clarke J, Kaski JC,Maseri A. Coronary vasomotor effects of serotonin inpatients with angina: relationship to coronary stenosismorphology. Circulation 1993;88: 1518-26.

21 Furchgott RF. Role of endothelium in responses ofvascularsmooth muscle. Circ Res 1983;53:557-73.

22 Casino PR, Kilcoyne CM, Quyyumi AA, Hoeg JM, PanzaJA. Investigation of decreased availability of nitric oxideprecursor as a mechanism responsible for impairedendothelium-dependent vasodilation in hypercholes-terolemic patients. JAm Coil Cardiol 1994;23:844-50.

23 Panza JA, Quyyumi AA, Bruch JE Jr, Epstein SE.Abnormal endothelium-dependent vascular relaxation inpatients with essential hypertension. N Engl Jf Med1990;323:22-7.

24 Panza JA, Casino PR, Kilcoyne CM, Quyyumi AA. Role ofendothelium-derived nitric oxide in the abnormalendothelium-dependent vascular relaxation of patientswith essential hypertension. Circulation 1993;87:1468-74.

25 Luscher TF, Vanhoutte PM, Raij L. Antihypertensivetreatment normalizes decreased endothelium-dependentrelaxations in rats with salt-induced hypertension.Hypertension 1987;9(suppl III):III193-7.

26 Tousoulis D, Kaski JC, Bogaty P, Crea F, Gavrielides S,Galassi AR, et al. Reactivity of proximal and distal angio-

265

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from

Tousoulis, Davies, Lefroy, Haider, Crake

graphically normal and stenotic coronary segments inchronic stable angina pectoris. Am J Cardiol 1991;67:1195-200.

27 Nissen SE, Gurley JC, Grines CL, Booth DC, McClure R,Berk M, et al. Intravascular ultrasound assessment oflumen size and wall morphology in normal subjects andpatients with coronary artery disease. Circulation 1991;84:1087-99.

28 St Goar FG, Pinto FJ, Alderman EL, Fitzgerald PJ, StadiusML, Popp RL. Intravascular ultrasound imaging ofangiographically normal coronary arteries as in vivo com-parison with quantitative angiography. J Am Coil Cardiol1991;18:952-8.

29 Cannon RO, Epstein SE. "Microvascular angina" as acause of chest pain with angiographically normal coro-nary arteries. AmY Cardiol 1988;61:1338-43.

SHORT CASES IN CARDIOLOGY

Interventricular septal hydatid cyst presenting as

complete heart block

Dinesh K Agarwal, Reshma Agarwal, Satish P Barthwal

Department ofCardiology and PGDepartment ofMedicine, MLNMedical College,Allahabad, IndiaD K AgarwalR AgarwalS P BarthwalCorrespondence to:Dr D K Agarwal, 7/5A/4CY Chintamani Road,Darbhanga Colony,Allahabad 211 001,Uttar Pradesh, India.Accepted for publication11 September 1995

A 25 year old man presented with Stokes-Adams attacks, which he had had for 18months. His pulse rate was 40 per minute andhis blood pressure was 140/60 mm Hg. Cardiacexamination showed a grade 2/6 mid-ejectionsystolic murmur at the left sternal border withno signs of cardiac failure. Pulmonary andneurological examinations were normal. Theelectrocardiogram showed complete heartblock. The chest x ray and haemogram werenormal. He was treated with temporaryventricular pacing. Colour Doppler echocardio-graphy (Vingmed CFM 725) showed ahomogeneous cystic mass (diameter 2-8 cm)with smooth rims located in the middle third ofthe interventricular septum (figure). There wasno gradient across left ventricular or rightventricular outflow tracts. Abdominalultrasound showed a cyst (diameter 5-3 cm)with internal echogenic shadows in theposterior segment of right hepatic lobe. Acomputed tomogram of the abdomen showedsimilar findings.The indirect haemagglutination test for

Apicalfour chamber echocardiogram showing a rounded cystic mass in the interventricularseptum.

echinococcus was positive (1/640). The hepaticcyst was drained and injected with hypertonicsaline. The cyst contained scolices ofechinococcus. He had been treated with alben-dazole for 2 weeks before the hepatic cyst wasdrained. This dose (5 mg/kg twice daily) wascontinued for another 2 weeks. Two morecourses were repeated after an interval of 3weeks.The patient refused cardiac surgery. A per-

manent pacemaker (VVI multiprogrammable)was implanted because he was dependent onthe temporary pacemaker. When the patientwas reassessed after 6 months and after 2 yearshe was symptom free. The patient wasreassessed at one year when he was still pace-maker dependent and there was no change inthe size of the cyst on the echocardiogram.

Echinococcosis in the heart is uncommonand a cyst in the interventricular septum is rarerstill, accounting for only 2-9% of all cardiaccases.'-2 They occur in the pericardium or freewall of left ventricle in 50%-77% cases.'3 In ourpatient the hydatid cyst was located in the inter-ventricular septum and caused syncopal attacksowing to complete heart block. Another similarcase has been reported.2 We agree with the ear-lier reports3 4 that cross sectional echocardiogra-phy is the most reliable method of diagnosingan intraseptal hydatid cyst. In our patient thescolices found in the fluid from the hepatic cystwas further evidence that the cyst in the inter-ventricular septum was hydatid. This tended toexclude the other possibility of a mesothelioma.

Surgical removal of the cardiac cyst remainsthe best treatment.2 Where this is not feasible,however, patients can be treated satisfactorilywith repeated courses of albendazole and thesupportive measures used in our patient.

1 Desnos M, Brochet E, Cristofini P, et al. Polyvisceralechinococcosis with cardiac involvement imaged by twodimensional echocardiography, computed tomographyand nuclear magnetic resonance imaging. Am J Cardiol1987;59:383-5.

2 Ottino G, Villani M, De Paulis R, Trucco G, Viara J.Restoration of atrioventricular conduction after surgicalremoval of a hydatid cyst of the intraventricular septum. .TThorac Cardiovasc Surg 1987;93: 144-7.

3 Lanzoni AM, Barrios V, Moya JL, Epeldegui A, CeleminD. Dynamic left ventricular outflow obstruction causedby cardiac echinococcosis. Am HeartJ 1992;124:1083-5.

4 Kontopoulos AG, Avramides MJ, Athyros VG. Diagnosis,treatment, and long-term follow up of a patient with ahydatid cyst of the left ventricle. BrHeartr_ 1994;72:592.

266

on Decem

ber 27, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.3.261 on 1 March 1996. D

ownloaded from