Heart 1996;75:50-54

Myocardial /B adrenoceptors and left ventricularfunction in hypertrophic cardiomyopathyLubna Choudhury, Stefano Guzzetti, David C Lefroy, Petros Nihoyannopoulos,William J McKenna, Celia M Oakley, Paolo G Camici

AbstractObjective-To assess the relation betweenleft ventricular function and myocardial I)adrenoceptor density.Methods-17 patients with hypertrophiccardiomyopathy, six with and 11 withoutheart failure, were studied. Left ventricu-lar function was assessed by echocardiog-raphy, and myocardial II adrenoceptorsby positron emission tomography. Patientdata were compared with those obtainedin normal controls.Results-Myocardial ,8 adrenoceptor den-sity in the 17 patients was 7*00 (SD 1.90)pmollg v 11-50 (2.18) pmollg in normalcontrols (P < 0.01). fi Adrenoceptor den-sity in the six patients with left ventricularfailure was 5*61 (0.88) pmollg v 7*71 (1.86)pmollg in the 11 patients with normalventricular function (P < 0.05), and therewas a significant correlation (r = 0*52;P < 0.05) between left ventricular frac-tional shortening and myocardial ,adrenoceptor density. A positive correla-tion (r = 0-51; P < 0.05) was also foundbetween myocardial fi adrenoceptor den-sity and the E/A transmitral flow ratio, anindex ofleft ventricular diastolic function.Conclusions-There is myocardial /1adrenoceptor downregulation in patientswith hypertrophic cardiomyopathy withor without signs ofheart failure.

(Heart 1996;75:50-54)

Keywords: hypertrophic cardiomyopathy; myocardialadrenoceptors; left ventricular function; positronemission tomography

MRC Clinical SciencesCentre and RoyalPostgraduate MedicalSchool, HammersmithHospital, LondonL ChoudhuryS GuzzettiD C LefroyP NihoyannopoulosC M OakleyP G CamiciDepartment ofCardiologicalSciences, St George'sHospital MedicalSchool, LondonW J McKennaCorrespondence to:Prof P Camici, MRCCyclotron Unit,Hammersmith Hospital,Du Cane Road, LondonW12 OHS.

Accepted for publication14 August 1995

Hypertrophic cardiomyopathy is characterisedby inappropriate myocardial thickness, mostoften involving the interventricular septum ofa non-dilated left ventricle.' The disease istransmitted as an autosomal dominant trait inup to 50% of patients.2 Recently various muta-tions of the ,B myosin heavy chain gene havebeen described in affected families.5However, this gene has been excluded as thecause of the disease in some families6 andanalysis of other candidate genes is awaited.

It has been hypothesised that other factorscould play a role in the development and nat-ural history of this disease, including anincreased sympathetic activity in the heart.7Accordingly, a raised myocardial noradrena-line content8 as well as increased cardiacspillover of noradrenaline9 have been found inpatients with hypertrophic cardiomyopathy.

Moreover, a downregulation of myocardial fiadrenoreceptors has recently been observed inthese patients by means of positron emissiontomography,'0 which is presumably a result oflocally increased levels of noradrenaline."I

Although ventricular systolic function isusually hyperdynamic in hypertrophic cardio-myopathy, a minority of patients (5-10%)develop progressive dilatation of the left ven-tricular chamber and systolic dysfunction laterin the course of the disease.'2-'5 To investigatethe relation between left ventricular functionand changes in neural control of the heart, weassessed left ventricular function by echocar-diography and myocardial f, adrenoceptordensity by positron emission tomography in agroup of patients with hypertrophic cardiomy-opathy with and without heart failure. Theresults were compared with those obtained ina group of normal controls.

MethodsSTUDY POPULATIONSeventeen patients with hypertrophic cardio-myopathy were studied (11 males and sixfemales; table). Patients were excluded fromthe study if they were on fi blocker or amio-darone treatment or if they had asthma or anyother systemic disease, and women of childbearing potential were also excluded. Elevenpatients (mean age 37 (SD 10) years) had noevidence of heart failure and had preserved leftventricular systolic function with normal frac-tional shortening. In these 11 patients themyocardial fi adrenoceptor density has previ-ously been reported.'0 Six patients (mean age53 (10) years, P < 0 01 v patients with pre-served left ventricular function) had clinicalevidence of heart failure (New York HeartAssociation functional class III and IV),received treatment with diuretics, and hadreduced fractional shortening (table). All ofthese six patients had a family history of hyper-trophic cardiomyopathy and had documentedevidence of typical disease previously.Coronary angiography was normal in these sixpatients.

Eight normal volunteers (mean age 28 (7)years; all males) whose myocardial ,B adreno-ceptor density had been reported previously,'0served as control subjects. All the normalvolunteers were selected on the basis of theirclinical history and physical examination,which indicated a low risk of coronary arterydisease. All had normal resting electrocardio-grams and negative exercise tests in responseto a high workload.

50 on A

pril 1, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.1.50 on 1 January 1996. Dow

nloaded from

Myocardialf, adrenoceptors and left ventnicularfunction in hypertrophic cardiomyopathy

Clinical and echocardiographic characteristics ofpatients

Echocardiographic findingsSymptoms

Age ivs pw LVEDD LVESD Fractional LVOTgradient E.ASex (years) Dyspnoea Orthopnoea Medication (mm) (mm) (mm) (mm) shortening (%) (mm Hg) ratio

Patients with preserved systolic function1 M 33 yes no Verapamil 21 13 51 30 41 32 1-672 M 33 yes no None 23 13 47 31 34 5 1-63 F 47 yes no Diltiazem 17 14 40 20 50 121 1-134 M 32 no no Verapamil 23 12 36 20 44 3 1-65 M 36 no no None 31 14 40 25 37 60 0 396 M 51 no no Verapamil 17 14 47 30 36 14 0 737 M 31 no no None 16 13 50 24 52 26 1-258 M 20 no no None 27 11 44 27 39 4 1-279 M 50 no no None 15 11 46 34 26 9 1-6210 M 27 no no Verapamil 15 10 46 32 30 0 1-8911 M 48 no no None 17 14 54 38 30 9 0 43Mean(SD) 37(10) 19(6) 12(2) 45(5) 28(6) 38(8) 26 1-23(0-16)

Patients with systolic dysfunction12 M 57 yes yes Diuretics 13 13 62 56 10 9 0-313 F 68 yes yes Diuretics 15 15 48 38 21 9 1-5214 F 52 yes yes Diuretics 12 12 47 38 19 0 2-3915 F 39 yes yes Diuretics 10 10 52 42 19 016 F 58 yes no Diuretics 8 7 48 44 8 0 1-417 F 47 yes no None 17 17 51 39 24 0 0 33Mean(SD) 53(10)** 12(3)* 12(4) 51(6)* 43(7)** 17(6)** 3** 1-19(0-40)

pnd, paroxysmal nocturnal dyspnoea; ivs, interventricular septum; pw, posterior wall; LV, left ventricle; LVESD, LV end systolic dimension; LVEDD, LV enddiastolic dimension; LVOT, left ventricular outflow tract.*P < 0*05, **P < 0-01 v patients with preserved systolic function.

All subjects gave written informed consentto the protocol which was approved bythe Hammersmith Hospital research ethicscommittee and the United KingdomAdministration of Radioactive SubstancesAdvisory Committee.

STUDY PROTOCOLEchocardiographyEchocardiographic studies were performed inall patients, and the left ventricular wall thick-ness and internal dimensions were measuredat the level of the mitral valve leaflet tips.16The fractional shortening of the left ventriclewas calculated as the percentage differencebetween the end systolic dimension and theend diastolic dimension normalised to the enddiastolic dimension. Continuous waveDoppler velocities from the apical projectionwere used to calculate the peak intracardiacgradient (PG), as PG = 4 x Vmax2, where Vm isthe left ventricular outflow tract velocity.Transmitral early (E wave) and late (A wave)peak inflow velocities were measured by sam-pling with pulsed Doppler proximal to themitral leaflet tips and recording in midexpira-tion; an average of three values was taken.

Positron emission tomographyThe subject was positioned on the bed of anECAT 931-08/12, 15 plane positron tomo-graph (Siemens/CTI) so that the left ventriclelay as close as possible to the centre of theaxial and transaxial fields of view. Before per-forming the emission study, a circular ringsource filled with about 2 mCi of 68Ge wasused for the blank and transmission dataacquisition. Initially, a rectilinear transmissionscan was recorded and used as a low resolu-tion x ray film to identify the heart profile. Atransmission scan was then recorded for a 20min period in order to measure the attenua-tion correction coefficients to be used for eachline of response of the emission sinogram.A blood volume scan was obtained using

inhaled oxygen-15 labelled carbon monoxide

(C'50). Four serial venous samples were takenduring the C150 scan and counted in a cali-brated well counter.'7

Regional myocardial blood flow was subse-quently measured using oxygen-15 labelledwater (H,150) from inhaled oxygen-15 labelledcarbon dioxide (C'50,) as described else-where. '7

Measurement of myocardial ,B adrenoceptordensity was performed according to a modifi-cation of the double injection method ofDelforge et al."' The carbon-i 1 labelled S-enantiomer of the non-selective, hydrophilic /3blocker CGP 12177 (S-["C]CGP 12177) wasprepared as previously described'9 and used asa high affinity radioligand to quantify myocar-dial ,B adrenoceptor density. A first dose of S-["lC]CGP 12177 with high specific activity(7-2 (1.9) ,ug S-["C]CGP 12177; 155 (33)MBq; specific activity 6A4 (1 9) GBq/umol)was infused intravenously over 2 min, fol-lowed 30 min later by a second dose with lowspecific activity (34 5 (4 5) ,ug S-["C]CGP12177; 324 (148) MBq; specific activity 2-4(0.7) GBq/pmol) infused intravenously over 2min. A 45 frame dynamic emission scan wasused to define the temporal and spatial distrib-ution of the tracer in vivo. A single 30 s back-ground frame was acquired before theintravenous infusion of the first dose ofS-["C]CGP 12177. During the 30 minperiod following the start of the first infusionof S-["C]CGP 12177, 22 time frames (8 x15s,4 x 30s,2 x 60s,2 x 120s,4 x 150s, and 2 x 300 s) were acquired. The secondinfusion of S-["C]CGP 12177 was then givenand scan data were acquired according to anidentical sequence of time frames. Nineteenvenous blood samples were collected duringthe S-["C]CGP 12177 scan for later correc-tion for vascular activity of the tracer.

Sinograms were normalised, corrected forattenuation, and then reconstructed to provideimages with a transaxial spatial resolution of8-4 mm full width at half maximum (FWHM)and a slice thickness of 6-6 mm FWHM. Data

51 on A

pril 1, 2020 by guest. Protected by copyright.

http://heart.bmj.com

/H

eart: first published as 10.1136/hrt.75.1.50 on 1 January 1996. Dow

nloaded from

Choudhury, Guzzetti, Lefroy, Nihoyannopoulos, McKenna, Oakley, et al

collection and initial processing were per-formed using dedicated array processors on aMicro Vax 2 computer (Digital EquipmentCorporation). The final images were trans-ferred to SUN 3/60 workstations (SunMicroSystems) for further analysis by use ofAnalyze image analysis (Mayo Foundation)20and Pro-Matlab (MathWorks, Inc) mathemat-ical software packages.

Regional values of blood volume (VB, mlblood/ml region of interest (ROI)) wereobtained by relating equilibrium images of theC'50 distribution to the radioactive concentra-tions ofvenous blood samples during the scan.Corrections were made for radioactive decayand blood density (1 06 g/ml). In each of thefive adjacent scan planes being analysed,which encompassed most of the left ventricu-lar myocardium, four different ROI corre-sponding to anterior, lateral, infero-posterior,and septal myocardium were selected. Themean left ventricular myocardial blood flowwas obtained by defining a further ROI whichencompassed the whole left ventricle withinthe five adjacent scan planes. Myocardialblood flow (ml/min/g) for each ROI was calcu-lated by fitting the arterial input (obtainedfrom a left atrial ROI) and tissue time-activitycurves from the blood flow scan to a single tis-sue compartment tracer kinetic model, as pre-viously described.'7 This model includescorrections for the underestimation of tissueactivity due to the partial volume effect andthe spillover of activity from the left ventricu-lar chamber into the myocardial ROI.The same ROI which had been used to cal-

culate myocardial blood flow were thenapplied to the S-["C]CGP 12177 scan. Themyocardial tracer time-activity curves were

Figure 1 Mean leftventricular myocardialbloodflow (panelA) and(3 adrenoceptor density(panel B) in the normalsubjects, and in thepatients with hypertrophiccardiomyopathy withpreserved systolic function(HC) or with leftventricular dysfunction(HC-LVD)

1.5

1-0 H

0.5 H

A

T

HC HC-LVD0-0

Normals

1-0

0.5

D

P<0.001I-

P<0-05

T

0.0' V//////////Normals HC

corrected for the radioactive decay and for thevascular activity using the regional values ofblood volume and the radioactive concentra-tions of blood samples taken throughoutthe dynamic scan. The sections of the curvecorresponding to the two slow phases, whichrepresent S-["C]CGP 12177 bound to ,Badrenoceptors, were exponentially extrapo-lated on the y axis back to the start of the infu-sions. The /3 adrenoceptor density was derivedfrom the maximum number of available spe-cific S-["]CGP 12177 binding sites per g oftissue (Bmax) in the ROI. Bm. values were cal-culated using a modification of the equationderived by Delforge et alP8 to take account ofthe molar content of S-CGP 12177 in bothinjections.'0 In addition, the values of Bmaxwere corrected for the partial volume effect bynormalising to the extravascular tissue volume(ml tissue per ml of ROI), obtained by sub-tracting vascular density image from the nor-malised transmission scan.'7 To convert themyocardial blood flow and ,B adrenoceptordensity values from units per ml of tissue tounits per gram tissue, all final values weredivided by 1-04 (myocardial tissue density).

STATISTICSAll values are expressed as means (SD). Twotailed unpaired Student's t tests were used forall between group comparisons except wherenot indicated. The left ventricular outflowtract gradients between the patients with pre-served and impaired systolic function werecompared using Wilcoxon's rank sum test.Analysis of variance (ANovA) was used to com-pare regional myocardial blood flow andregional /3 adrenoceptor density in the four dif-ferent ROI within each group, and to compareperfusable tissue index values among the twopatient groups and control subjects. Linearregression analysis was used to examine therelation between myocardial ,B adrenoceptordensity and left ventricular systolic and dias-tolic function in the patients with hypertrophiccardiomyopathy. A P value < 005 was consid-ered statistically significant.

ResultsVENTRICULAR ANATOMY AND FUNCTIONAll our patients had a clinical history of hyper-trophic cardiomyopathy; at the moment of thestudy some significant differences were evidentin the echocardiographic indices betweenpatients with and without preserved systolicfunction (table). Note the lower thickness ofinterventricular septum, the increased left ven-tricular diastolic diameter, and the lower frac-tional shortening in patients with systolicdysfunction.

MYOCARDIAL BLOOD FLOWThe myocardial blood flow was homoge-neously distributed throughout the differentregions of the left ventricular wall in the twogroups of patients and control subjects, andthe mean myocardial blood flow did not differsignificantly among the three groups (fig 1A).The perfusable tissue index was 1 05 (0-08) in

0

00

- CM1-:5C-

EC_Cu5)

CoC)

oXa)00

c-50

C._

"E'a

C-,0

52

on April 1, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.75.1.50 on 1 January 1996. D

ownloaded from

Myocardial f, adrenoceptors and left ventricularfunction in hypertrophic cardiomyopathy

Figure 2 Linearregression analysis of therelation between myocardialP adrenoceptor density andleft ventricularfractionalshortening in the patientswith hypertrophiccardiomyopathy. The filledsymbols represent thepatients with heartfailureand the empty symbolsrepresent the patients withpreserved systolic function

Figure 3 Linearregression analysis of therelation betweenmyocardial ,B adrenoceptordensity and the EIA(transmitral early and lateflow velocity) ratio. Thefilled symbols represent thepatients with ,Badrenoceptor density< 7-00 pmollg

75

._E4)

0

0

Cr.,C)

0)IL)

a1)

0.j

0o50

2.5

2-0

.° 1.54C

wi 1-0

0-5

n.n

* r= 0.52; P = 0.03

4 6 8 10Myocardial 0 adrenoceptors (pmol/g)

0

0

r= 0-51; P < 0.05

2 4 6 8 10 12Myocardial ,B adrenoceptors (pmol/g)

controls, 1 08 (0 05) in patients with pre-served left ventricular function and 1-05(006) in patients with heart failure (P = NS).

MYOCARDIAL f/ ADRENOCEPTOR DENSITYThere was no significant difference in themyocardial adrenoceptor densities amongthe four left ventricular regions within eachgroup studied. Myocardial adrenoceptordensity in all the 17 patients with hypertrophiccardiomyopathy was 700 (1 90) v 11-50(2-18) pmol/g in the control subjects (P <001). adrenoceptor density in the sixpatients with hypertrophic cardiomyopathyand left ventricular failure was 5-61 (0 88) v

7-71 (1 86) pmol/g in the 11 patients with pre-served systolic function (P < 0 05: fig 1B).These latter values are 49% and 33% lowerthan in normal subjects respectively.

MYOCARDIAL ADRENOCEPTOR DENSITY ANDLEFT VENTRICULAR FUNCTIONIn the patients with hypertrophic cardiomy-opathy, there was a significant correlation(r = 0-52; P < 0 05) between left ventricularshortening and myocardial adrenoceptordensity (fig 2). In addition, a positive correla-tion (r = 0-51; P < 0 05) was detectedbetween the adrenoceptor density and theE/A transmitral flow ratio (fig 3), an index ofleft ventricular diastolic function.2'

DiscussionThese results lend further support to thehypothesis of an increased sympathetic activity

in the heart of patients with hypertrophic car-diomyopathy.7 Moreover, they underline thestrict relationship between the dysfunction ofthe autonomic control of the heart and systolicand diastolic function of the left ventricle inthese patients.

Although it is unlikely that the autonomicdysfunction is the prime mover in the patho-genesis of hypertrophic cardiomyopathy,which seems to be genetically transmitted,2-5 itmight be an important cofactor in the develop-ment of the disease. There is strong experi-mental evidence that myocardial hypertrophycan be induced by direct stimulation of a and /3adrenergic receptors on cardiac myocytes, anaction which seems to be independent of thehaemodynamic load.22 Furthermore, there isnow evidence that a higher level of localmyocardial catecholamines is an importantdeterminant of the downregulation of myocar-dial ,B adrenoceptors."I It has recently beenfound that myocardial /3 adrenoceptor densityis reduced in patients with hypertrophic car-diomyopathy who have preserved left ventricu-lar systolic function.'0 The results of thepresent investigation show a more profoundreduction of myocardial /3 adrenoceptor den-sity in patients with hypertrophic cardiomy-opathy who have developed impairment of leftventricular systolic function and clinical evi-dence of heart failure. It is noteworthy that thedensity of /3 adrenoceptor is correlated withthe systolic and diastolic function of the leftventricle. Reductions in the myocardial /3adrenoceptor density of approximately 50%have been reported in patients with heart failure

2324 nof different causes,2 using in vitro ligandbinding to homogenised myocardial biopsysamples. In a recent study, Merlet et al,25 usingS-["C]CGP 12177 and positron emissiontomography, reported a 53% decrease inmyocardial /3 adrenoceptor density in patientswith dilated cardiomyopathy compared to nor-mal controls. In the latter study, the absolutevalue of myocardial /3 adrenoceptor densitymeasured in normal controls is about 40%lower than that measured in our study. Twomain technical reasons could account for thisdifference: (1) the data of Merlet et al are cor-rected for partial volume using echocardiogra-phy, which does not account for partialvolume effect arising from movement; (2) theequation used by the French group has beenmodified by us'0 to take into account theamount of cold CGP 12177 which is co-injected with S-["C]CGP 12177 and results inan additional underestimation of receptor den-sity of the order of 30%. However, severalother potential problems may limit the accu-racy of the measurement of myocardial /3adrenoceptors with S-["C]CGP 12177 andpositron emission tomography. These includethe following: (1) Non-specific receptor bind-ing. Non-specific binding of S-["C]CGP12177 in rat myocardium is approximately10%.26 We have calculated non-specific bind-ing of S-["C]CGP 12177 to human myocard-ium (measurements performed during controlconditions and following administration ofpropranolol) to be in the range of 8% to 12%

53

I'd

0

00

0 00

0 00

00

25

on April 1, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.75.1.50 on 1 January 1996. D

ownloaded from

Choudhury, Guzzetti, Lefroy, Nihoyannopoulos, McKenna, Oakley, et al

(unpublished results). It is worth noting, how-ever, that the same problem of non-specificbinding would also be encountered with invitro techniques. (2) Potential transmural dif-ferences in receptor densities which cannot beassessed because of the limited spatial resolu-tion of the present positron cameras. (3)Enhanced tissue fibrosis in patients with car-diomyopathy, which might lead to erroneouslylower receptor density compared with con-trols. To account for this problem, the perfus-able tissue index (an index of viable tissue)was measured in the three groups (normalsand patients with and without heart failure)and found to be comparable, suggesting a lackof significant interstitial fibrosis.

In conclusion, we have confirmed the pres-ence of myocardial ,B adrenoceptor downregu-lation in patients with hypertrophiccardiomyopathy with or without signs of heartfailure. This study is the first to show that,although a certain degree of /3 adrenoceptordownregulation can be associated with a pre-served systolic function, values lower than6-5-7T0 pmollg are almost invariably accom-panied by heart failure.

We are grateful to the radiochemists of the MRC Cyclotronunit for preparation of the radioligands for PET scanning. Wethank the radiographers for their invaluable help in performingthe studies.

1 Maron BJ, Bonow RO, Cannon RO, Leon MB, EpsteinSE. Hypertrophic cardiomyopathy: interrelations of clini-cal manifestations, pathophysiology and therapy. N EnglJ Med 1987;316:780-9.

2 Maron BJ, Nichols PF, Pickle LW, Wesley YE, MulvihillJJ. Patterns of inheritance in hypertrophic cardiomyopa-thy: assessment by M-mode and two-dimensionalechocardiography. Am J Cardiol 1984;53: 1087-94.

3 Tanigawa G, Jarcho JA, Kass S, Solomon SD, Vosberg HP,Seidman JG, et al. A molecular basis for familial hyper-trophic cardiomyopathy: an alpha/beta cardiac myosinheavy chain hybrid gene. Cell 1990;62:991-8.

4 Geisterfer-Lowrence AA, Kass S, Tanigawa G, VosbergHP, McKenna W, Seidman CE, et al. A molecular basisfor familial hypertrophic cardiomyopathy: a beta cardiacmyosin heavy chain missense mutation. Cell 1990;62:999-1006.

5 Watkins H, Rosenzweig A, Hwang DS, Levi T, McKennaW, Seidman CE, et al. Characteristics and prognosticimplications of myosin missense mutations in familialhypertrophic cardiomyopathy. N Engl J Med 1992;326:1108-14.

6 Schwartz K, Beckmann J, Dufour C, Faure L, FougerousseF, Carrier L, et al. Exclusion of cardiac myosin heavychain and actin gene involvement in hypertrophic car-diomyopathy of several French families. Circ Res1992;71:3-8.

7 Pearse AGE. The histochemistry and electron microscopyof obstructive cardiomyopathy. In: Cardiomyopathies[Ciba Foundation Symposium No 81]. London:Churchill, 1964:132.

8 Kawai C, Yoshili Y, Hoshino T, Sasayama S, MatsumoriA. Myocardial catecholamines in hypertrophic and

dilated (congestive) cardiomyopathy: a biopsy study. JAm Coll Cardiol 1983;2:834-40.

9 Brush JE, Eisenhofer G, Garty M, Stull R, Maron BJ,Cannon RO, et al. Cardiac norepinephrine kinetics inhypertrophic cardiomyopathy. Circulation 1989;79:836-44.

10 Lefroy DC, de Silva R, Choudhury L, Uren NG, Crake T,Rhodes CG, et al. Diffuse reduction of myocardial betaadrenoceptors in hypertrophic cardiomyopathy: a studywith positron emission tomography. Jf Am Coll Cardiol1993;22: 1653-60.

11 Bristow MR, Minobe W, Rasmussen R, Larrabee P, SkeriL, Klein JW, et al. Beta adrenergic neuroeffector abnor-malities in the failing human heart are produced by localrather than systemic mechanisms. J Clin Invest1992;89:803-15.

12 Spirito P, Maron BJ, Bonow RO, Epstein SE. Occurrenceand significance of progressive left ventricular wall thin-ning and relative cavity dilatation in hypertrophic car-diomyopathy. Am Jf Cardiol 1987;59: 123-9.

13 Oakley CM. Hypertrophic obstructive cardiomyopathy.Patterns of progression. In: Wolstenholme GEW,O'Connoe M, eds. Hypertrophic obstructive cardiomyopa-thy [CIBA Foundation Study Group, No 31]. London:Churchill, 1971:9-29.

14 Fighali S, Krajcer Z, Edelman S, Leachman R. Progressionin hypertrophic cardiomyopathy into a hypokinetic leftventricle: higher incidence in patients with midventricu-lar obstruction. JAm Coil Cardiol 1987;9:288-94.

15 Corr LA, French J, Oakley CM, Nihoyannopoulos P.Predictors and prognosis of left ventricular dilatation inpatients with hypertrophic cardiomyopathy [abstr]. JAmColl Cardiol 1993;21:123A.

16 Tajik AJ, Seward JB, Hagler DJ, Mair DD, Lie JT. Two-dimensional real time ultrasonic imaging of the heart andgreat vessels. Mayo Clin Proc 1978;53:271-303.

17 Araujo LI, Lanmertsma AA, Rhodes CG, McFalls EO,Iida H, Rechavia E, et al. Noninvasive quantification ofregional myocardial blood flow in coronary artery diseasewith oxygen-15-labeled carbon dioxide inhalation andpositron emission tomography. Circulation 1991;83:875-85.

18 Delforge J, Syrota A, Lanqon JP, Nakajima K, Loch C,Janier M, et al. Cardiac beta-adrenergic receptor densitymeasured in vivo using PET, CGP 12177 and a newgraphical method. J Nucl Med 1991;32:739-48.

19 Brady F, Luthra SK, Tochon-Danguy H, Steel CJ, WatersSL, Kensett MJ, et al. Asymmetric synthesis of a precursorfor the automated radiosynthesis of s[11C]CGP 12177as a preferred radioligand for the beta-adrenergic recep-tors. Appl Radiat Isotopes 1991;42:621-8.

20 Robb RA, Hanson DP. A software system for interactiveand quantitative visualization of multidimensional bio-medical images. Aust Phys Eng SciMed 1991;14:9-30.

21 Maron BJ, Spirito P, Green KJ, Wesley YE, Bonow RO,Arce J. Noninvasive assessment of left ventricular dias-tolic function by pulsed Doppler echocardiography inpatients with hypertrophic cardiomyopathy. J Am CollCardiol 1987;lO:733-42.

22 Laks MM, Morardy F, Swan HJC. Myocardial hypertro-phy produced by chronic infusion of subhypertensivedoses of norepinephrine in the dog. Chest 1973;64:75-8.

23 Bristow MR, Ginsburg R, Minobe W, Cubiciotti RS,Sageman WS, Lurie K, et al. Decreased catecholaminesensitivity and beta-adrenergic receptor density in failinghuman hearts. NEnglJ Med 1982;307:205-1 1.

24 Bristow MR. Pathophysiologic and pharmacologic ratio-nales for clinical management of chronic heart failurewith beta-blocking agents. Am J Cardiol 1993;71:12-22C.

25 Merlet P, Delforge J, Syrota A, Angevin E, Maziere B,Crouzel C, et al. Positron emission tomography with 11CCGP-12177 to assess beta-adrenergic receptor concen-tration in idiopathic dilated cardiomyopathy. Circulation1993;87:1 169-78.

26 Law MP. Demonstration of the suitability of CGP 12177for in vivo studies of 3-adrenoceptors. Br J7 Pharmacol1993;109:1101-9.

54

on April 1, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Heart: first published as 10.1136/hrt.75.1.50 on 1 January 1996. D

ownloaded from