UMEÅ UNIVERSITY MEDICAL DISSERTATIONS New Series No 451 - ISSN 0346-6612
A
o r t ic
S
t e n o s is
:
D
ia g n o s t ic
U
se
a n d
H
e m o d y n a m ic
E
f f e c t s
o f
D
ip y r id a m o l e
Akademisk avhandling
som för avläggande av medicine doktorsexamen med vederbörligt tillstånd av rektorsämbetet vid Umeå Universitet offentligen kommer att försvaras i Rosa Salen, Tandläkarhögskolan,
Norrlands Universitetssjukhus i Umeå, lördagen den 16 december 1995, kl 09.15.
av
Peter Rask leg. läkare
Fakultetsopponent Med dr Lars-Ake Brodin
Institutionen för Kirurgi Karolinska Sjukhuset
Stockholm
UMEÅ UNIVERSITY MEDICAL DISSERTATIONS
New Series No 451 - ISSN 0346-6612 ISBN 91 -7191 -110-3
Aortic Stenosis:Diagnostic Use and Hemodynamic Effects of Dipyridamole
Peter Rask, Departments of Clinical Physiology and Internal Medicine, Umeå University, 901 85 Umeå, Sweden
Valvular aortic stenosis is today the most frequently occurring heart valve lesion in the adult W estern population. The degree of outflow obstruction, the presence and severity of accompanying valve lesions as well as left ventricular function can be assessed noninvasively using echocardiography - including Doppler, 2-D and colour flow imaging. As concomitant coronary artery disease has significant impact on patient management, coronary angiography is usually performed as a part of the preoperative evaluation. Reliable noninvasive methods for determination of the presence or absence of coronary artery disease would be valuable and may reduce the need for coronary angiography in these patients.
In a total of 129 adult patients with aortic stenosis dipyridamole was infused intravenously (0.56 mg/kg dipyridamole dissolved in 250 ml of saline given over a 4 minute period). There were no serious adverse effects. Patients were examined using cardiac catheterisation, echocardiography, 20,T1 SPECT and coronary angiography. No patient was excluded due to severe aortic stenosis. The smallest aortic valve area observed was 0.3 cm2 and the average valve area was approximately 0.7 cm2 in all studies.
During a dipyridamole stress test with the subject in the supine position, patients with aortic stenosis increased their cardiac output, stroke volume, left ventricular work and myocardial oxygen demand and showed a slight drop in blood pressure. Infusion of dipyridamole according to the present protocol appeared to be safe and may be used as a diagnostic tool in patients with aortic stenosis.
The aortic valve area has been considered to be essentially independent of transvalvular flow. However, in the present study both invasive and noninvasive measurements of the size of the valve area were found to be flow dependent. Increases in valve areas of up to 24 % were observed with increased transvalvular flow. This flow dependency of the aortic valve area has to be considered in clinical situations with altered transvalvular flow.
The present study establishes the gender specific normal distribution of 20IT1 uptake in patients with aortic stenosis given dipyridamole to increase coronary blood flow. Prospective computer assisted evaluation showed a high sensitivity (100%), specificity (75%), and positive (94%) and negative (100%) predictive values for significant coronary artery stenoses in men using the mean - 2.5 SD curve as the discriminating threshold. In women, however, this method showed a considerably lower diagnostic accuracy.
In patients with aortic stenosis increased myocardial oxygen demand is likely to be an important factor for development of wall motion abnormalities when dipyridamole is used in echocardiography stress testing. Using 2-D echocardiography and the combined criteria of a segmental wall motion abnormality at baseline or a new segmental wall motion abnormality after dipyridamole administration as a sign of coronary artery disease resulted in a high sensitivity for detection of multivessel or left anterior descending coronary artery disease (94%).
UMEÂ UNIVERSITY MEDICAL DISSERTATIONS
New Series N o 451 ISSN 0346-6612
From the Departments of Clinical Physiology and Internal Medicine Umeå University, Umeå Sweden
A
o r t ic
S
t e n o s i s
:
D
ia g n o s t ic
U
se
a n d
H
e m o d y n a m ic
E
f fe c t s
o f
D
ip y r id a m o l e
by
Peter
Rask
» i C/5 A .V1995
New Series No 451 ISSN 0346-6612 ISBN 91-7191-110-3 Printed by Solfjädern Offset AB
Umeå, Sweden 1995
I röreke
!Ven mätta dagen, den är addrig störst.
Den Bästa dagen är en dag av törst.
9dpgfinns det måC och mening i vår fä rd -
men det är vägen, som är mödan värd.
Det Bästa måCet är en nattCång rast,
där eCden tänds och Brödet Bryts i hast.
Tå stöden, dår man sover B Bott engång,
B hr sömnen trygg och drömmenfu.CC av sång.
Tryt upp, Bryt upp! Den nya degengryr.
Oändhgt är vårt stora äventyr.
CONTENTS
ABBREVIATIONS...6
ABSTRACT... 7
ORIGINAL PAPERS...8
INTRODUCTION... 9
AIMS...12
PATIENTS... 13
METHODS... 14
Ph a r m a c o l o g ic a lstr e ss (St u d ie s I - V ) ... 14 Ca r d ia c c a t h e t e r iz a t io n (St u d y I ) ... 14 Ec h o c a r d io g r a p h y (St u d ie s I I - V ) ...15 Th a l l iu m-201 s p e c t (St u d ie s III a n d I V ) ...15 Co r o n a r ya n g io g r a p h y (St u d ie s I-V )...15 St a t is t ic s...16RESULTS AND DISCUSSION...17
He m o d y n a m ice f f e c t so fd ip y r id a m o l e... 17
De p e n d e n c eo fa o r t icv a l v ea r e aont r a n sv a l v u l a rf l o w... 18
No n in v a s iv ea s s e s s m e n to fc o r o n a r ya r t e r yd is e a s e... 20
GENERAL SUMMARY AND CONCLUSIONS... 26
ACKNOWLEDGEMENTS... 27
REFERENCES... 28
ABBREVIATIONS
2-D Two-dimensional
ALVSWI Aortic left ventricular stroke work index (J x b e a t1 x (m2BSA) ‘)
A oPm ean Mean pressure in ascending aorta (mmHg)
AS Aortic stenosis AVA Aortic valve area (cm2)
A V A con t Aortic valve area calculated according to the continuity equation (cm2)
AV A G orlin Aortic valve area calculated according to the Gorlin equation (cm2)
BSA Body surface area (m2) CAD Coronary artery disease CO Cardiac output (1 x min'1) ET Ejection time (s)
HR Heart rate (beat x m in'')
LV O Tarea Left ventricular outflow tract area LVsm Mean left ventricular systolic
pressure obtained by planimetry of the area under the left ventricular pressure curve during ejection (mmHg)
LVSWI Left ventricular stroke work index (J x b e a t1 x (m2BSA) ') NLVSWI Net left ventricular stroke work
index (J x b e a t1 x (m2BSA) ’) n.s. Not significant
NYHA Functional class according to the New York Heart Association PTM Pressure time per minute (mmHg
x s x min ')
AP Pressure difference (mmHg) APmean The mean pressure difference
between the left ventricle and the aorta during ejection (mmHg) APpeak The peak pressure difference
between the left ventricle and the aorta during ejection (mmHg) SD Standard deviation
SI Stroke index (ml x b e a t1 x (m2B S A )‘)
SPECT Single photon emission computed tomography SV Stroke volume (ml)
SVRI Systemic vascular resistance index (mmHg x l 1 x min x m2BSA)
20lrjij
Thallium-201 V Velocity (m x s ‘)
voc
Vitium organicum cordis VTI Velocity time integral (cm) V T Ia s Velocity time integral across thestenotic aortic valve (cm) V T Il v o t Velocity time integral measured
in the left ventricular outflow tract (cm)
ABSTRACT
Valvular aortic stenosis is today the most frequently occurring heart valve lesion in the adult Western population. The degree of outflow obstruction, the presence and severity of accompanying valve lesions as well as left ventricular function can be assessed noninvasively using echocardiography - including Doppler, 2-D and colour flow imaging. As concomitant coronary artery disease has significant impact on patient management, coronary angiography is usually performed as a part of the preoperative evaluation. Reliable noninvasive methods for determination of the presence or absence of coronary artery disease would be valuable and may reduce the need for coronary angiography in these patients.
In a total of 129 adult patients with aortic stenosis dipyridamole was infused intravenously (0.56 mg/kg dipyridamole dissolved in 250 ml of saline given over a 4 minute period). There were no serious adverse effects. Patients were examined using cardiac catheterisation, echocardiography, 2()IT1 SPECT and coronary angiography. No patient was excluded due to severe aortic stenosis. The smallest aortic valve area observed was 0.3 cm2 and the average valve area was approximately 0.7 cm2 in all studies.
During a dipyridamole stress test with the subject in the supine position, patients with aortic stenosis increased their cardiac output, stroke volume, left ventricular work and myocardial oxygen demand and showed a slight drop in blood pressure. Infusion of dipyridamole according to the present protocol appeared to be safe and may be used as a diagnostic tool in patients with aortic stenosis.
The aortic valve area has been considered to be essentially independent of transvalvular flow. However, in the present study both invasive and noninvasive measurements of the size of the valve area were found to be flow dependent. Increases in valve areas of up to 24 % were observed with increased transvalvular flow. This flow dependency of the aortic valve area has to be considered in clinical situations with altered transvalvular flow.
The present study establishes the gender specific normal distribution of 20lTl uptake in patients with aortic stenosis given dipyridamole to increase coronary blood flow. Prospective computer assisted evaluation showed a high sensitivity (100%), specificity (75%), and positive (94%) and negative (100%) predictive values for significant coronary artery stenoses in men using the mean - 2.5 SD curve as the discriminating threshold. In women, however, this method showed a considerably lower diagnostic accuracy.
In patients with aortic stenosis increased myocardial oxygen demand is likely to be an important factor for development of wall motion abnormalities when dipyridamole is used in echocardiography stress testing. Using 2-D echocardiography and the combined criteria of a segmental wall motion abnormality at baseline or a new segmental wall motion abnormality after dipyridamole administration as a sign of coronary artery disease resulted in a high sensitivity for detection of multivessel or left anterior descending coronary artery disease (94%).
ORIGINAL PAPERS
This thesis is based on the following publications, which will be referred to by their Roman numerals.
I.
Rask P, Karp K, Teien D. Hemodynamic changes during dipyridamole stress in patients with aortic stenosis. J Heart Valve Dis 1994;3:510-5.II.
Rask P, Karp K, Eriksson P. Flow dependence of the aortic valve area in patients with aortic stenosis: Assessment by application of the continuity equation. J Am Soc Echocardiogr; In press.III.
Rask P, Karp K, Edlund B, Eriksson P, Mooe T, Wiklund U. Computer-assistedevaluation of dipyridamole thallium-201 SPECT in patients with aortic stenosis. J Nucl Med 1994;35:983-8.
IV .
Rask P, Karp K, Eriksson P, Mooe T. Dipyridamole thallium-201 single-photonemission tomography in aortic stenosis: gender differences. Eur J Nucl Med; In press.
V .
Karp K, Rask P, Teien D. Non-in vasi ve detection of coronary artery disease in patients with aortic stenosis using baseline and dipyridamole 2-D echocardiography. Manuscript.INTRODUCTION
The relative frequency of different valve le sions in the western population has changed during the last decades. The decline of rheu matic fever has led to a reduction in the prevalence of mitral stenosis, while the increasing longevity of the general popu lation has led to an increase in the prevalence of aortic stenosis. Valvular aortic stenosis is now the most frequent valve lesion in the adult population [Hall 1989]. The prevalence of at least moderate stenosis in a population between 55 and 86 years of age showed to be approximately 5% in a recent study [Lindroos et al 1993]. The most common aetiologies are congenital malformations and degenerative calcific disease, while rheumatic disease is now a rare cause of aortic stenosis.
Regardless of aetiology, calcification of the valves occurs in most longstanding cases. The expression congenital aortic stenosis is used when there is an obstruction at birth, but excludes cases with malformations of the valves without obstruction. Consequently, malformation of the valves at birth (usually bicuspid) with later development of stenosis is classified as calcified aortic stenosis.
The exact mechanism of obliteration and calcification is not fully understood. Tension, trauma to and turbulence around the mal formed valves are thought to play important roles. Infections affecting the valves with organisation of micro-thrombus may also be involved in the aetiology of calcification [Hall 1989]. It has recently been suggested that immune reactions against antigens in the aortic valve leaflets may be crucial to the formation of valvular fibrosis and subsequent calcification [Olsson et al 1994].
The prognosis of aortic stenosis is good as long as there are no symptoms [Pellikka 1990]. After the onset of symptoms however, the prognosis is poor and the mortality rate is higher than in many malignant neoplasms [Rahimtoola 1989]. The time from onset of symptoms to the time of death is approxi mately 2 to 5 years [Selzer 1987, Braunwald
1988].
Angina pectoris is more common in aortic than in other valve lesions [Olofsson et al. 1985] and it has been reported to occur in 40 to 70% of adults with aortic stenosis. Only half of these patients, however, have coronary artery disease [Exadactylos et al 1984]. Syncope is present in about 25% of patients with symptomatic aortic stenosis [Hall 1989]. Dyspnea has been reported as the most frequent symptom (up to 80%) [Cullhed 1964, Nylander et al 1986]. However, severe dyspnea is a late symptom and associated with poor left ventricular function and a poor prognosis [Seltzer 1987, Braunwald 1988, Hall 1989].
No medical treatment has yet been found for aortic stenosis. Open heart surgery with replacement of the aortic valve is the treatment of choice. Balloon valvuloplasty has been tried in patients with calcified aortic stenosis. Although temporary hemodynamic improvement can be achieved, the long-term results are not overwhelmingly positive and therefore the use of valvuloplasty in the treatment of aortic stenosis is controversial [Seltzer 1987, Hall 1989, McKay 1991].
Although the prognosis of untreated aortic stenosis is poor after the onset of symptoms, prognosis can be improved dramatically by
surgical treatment. Perioperative mortality varies between 2 and 5% in patients without frank left ventricular failure, and the 5-year postoperative survival rate is 85 to 90%. Even in an elderly population, surgical treatment improves prognosis and the quality of life [Braunwald 1988, Lindblom et al 1990, Olsson 1994]. In general, surgery is advisable when the patient has symptoms attributable to the valve lesion. However, patients in whom objective measurements indicate severe stenosis may be asymptomatic, while patients with symptoms which could be attributed to aortic stenosis may have only a mild stenosis according to objective measure ments [Hall 1989, Danielsen et al 1991].
Patients with aortic stenosis are routinely examined noninvasively with echocardio graphy. Echocardiography - including Dop pler, 2-D and colour flow imaging - makes it possible to determine the degree of valvular stenosis, the presence and severity of concomitant valve lesions as well as left ventricular function [Skjaerpe et al 1985, Helmcke et al 1987, Karp et al 1989, Holm et al 1992].
In adults, the aortic valve area is normally > 3.0 cm2 [Greenberg 1987]. Different criteria have been used to define severe aortic stenosis. An aortic valve area of < 0.75 cm2 is generally considered to indicate severe stenosis which is likely to produce symptoms. An obstruction of this degree has also been considered as a limit for intervention [Braunwald 1988]. In the presence of normal cardiac output and normal flow across the aortic valve, a peak to peak gradient across the valve of 50 mmHg has also been considered a limit for intervention [Rahimtoola 1989]. Regardless of which criteria are used, the obtained measurements must be
interpreted together with the clinical presentation.
While transvalvular pressure differences are dependent upon transvalvular flow, the valve area is considered to be essentially independ ent of flow. There is, however, evidence that the valve area calculated by the Gorlin equation increases with increasing flow [Bache et al 1971]. The Gorlin equation gives an estimate of the anatomical valve area and contains an empirical constant, which may have different values at different flow rates [Cannon et al 1985]. This could explain the apparent increase in valve area with increasing flow. Another possibility is that the valve area actually increases with increased flow [Burwash et al 1994].
The prevalence of coronary artery disease increases with increasing age and is relatively common in an elderly population with aortic stenosis. As the presence of coronary artery disease has significant impact on patient management, coronary angiography is usu ally performed as part of the preoperative evaluation. A beneficial effect of simultane ous valve replacement and coronary by-pass grafting in patients with aortic stenosis and coronary artery disease has been reported [Czers et al 1988, Jones et al 1989], and by-pass grafting of hemodynamically significant coronary lesions should be done together with valve replacement. Coronary angio graphy, however, involves a significant cost and also a small risk for catheterization complications. Reliable noninvasi ve methods for determining the presence or absence of coronary artery disease may reduce the need for coronary angiography in these patients.
Dipyridamole is frequently used in pharma cological stress testing. It is a potent vasodilator in most vascular beds [Sollevi et al 1984]. Dipyridamole causes near maximal dilatation of the coronary arteries and increases the coronary blood flow in vessels without coronary artery disease [Gould et al 1978]. The dilation of coronary arteries can lead to steal phenomena [Flameng et al 1974]. The exact mechanisms behind the effects of
dipyridamole are as yet unknown. Dipyri damole delays inactivation of adenosine [Pfleger et al 1969] and inhibits cellular uptake of adenosine [Knabb et al 1984]. An increased plasma concentration of adenosine is thought to be responsible for the main effect on coronary blood flow. This effect can be blocked by an adenosine antagonist, e.g. theophylline [Sollevi et al 1984].
AIMS
The present series of investigations were undertaken
-— to examine hemodynamic changes during a dipyridamole stress test in patients with aortic stenosis and to asses the safety of the procedure,
— to determine whether or not the aortic valve area in patients with aortic stenosis varies with transvalvular flow, and
— to examine if noninvasive methods, dipyridamole 201T1 SPECT and/or dipyridamole 2-D echocardiography, can demonstrate the presence or absence of concomitant coronary artery disease in patients with aortic stenosis.
PATIENTS
All patients included in the present series of studies were adults, between 44 and 80 years of age (mean 68 years), and referred for consideration for aortic valve replacement due to aortic stenosis. A total of 129 patients were examined. The patients were in functional class II, E l or IV according to the NYHA classification, with the overwhelming majority in functional class III. The prevalence of angina on effort was approximately 70%. General exclusion criteria were obstructive lung disease, other hemodynamically predominant valve lesions, e.g. mitral insufficiency, aortic insufficiency or mitral stenosis. No patient was excluded due to severe aortic stenosis. The smallest aortic valve area observed was 0.3 cm2 and the average valve area was approximately 0.7 cm2 in all studies.
Study I.
Six men and four women.
Study II.
Sixteen men and eighteen women.
Study III.
Twenty-eight men and twenty-four women.
Study IV.
Fifty-seven men and fifty-two women.
Study V.
Twenty-three men and twenty females.
n=43
n=52
n=10
METHODS
Pharmacological stress (Studies l-V)
In all patients a low dose dipyridamole protocol was used. Dipyridamole (0.56 mg/kg body weight dissolved in 250 ml saline) was infused intravenously over a 4 minute period with the patients in a supine position [Younis et al 1990]. No additional stress was used. Four to 8 minutes after completion of the dipyridamole infusion, 115 mg theophylline was administered slowly intravenously to all patients to reverse the dipyridamole effect.
Cardiac catheterization (Study I)
Right heart catheterization was performed using a 7F thermodilution catheter. Left heart catheterization was performed via the femoral artery using a 8F dual port high fidelity catheter allowing simultaneous recording of left ventricular and ascending aortic pressures. The pressure recordings were traced using a digitizing tablet interfaced with a Macintosh computer, allowing calculation of pressure differences.
The equation
was presented by Daniel Bernoulli in 1738. Using the ”simplified Bernoulli equation”, equation 1, instantaneous pressure differ ences were transformed into velocities [Hatle et al. 1980]. Systemic vascular resistance index was calculated using equation 2, and left ventricular stroke work index using equation 3. Pressure time per minute (an index of myocardial oxygen demand) was calculated by equation 4 [Yang et al 1978]. Equations 5 and 6 were derived for comparison between the work required to overcome the resistance of the aortic stenosis and the net left ventricular work (LVSWI = NLVSW I + ALVSWI). Aortic valve area was calculated according to Gorlin and Gorlin, equation 7 [Gorlin & Gorlin 1951], and according to the continuity equation, equation 8 [Skjerpe et al 1985].
Equation 1. AP = 4.0 X V 2
Equation 2. SVRI = AoPmean X CO'1 X BSA
Equation 3. LVSW I = SI X LVsm X 0.0136 X 9.8 x 1000'1
Equation 4. PTM = LVsm x ET x HR
Equation 5. ALVSW I = SI x APmean x 0.0136 x 9.8 x 1000'1 Equation 6. NLVSW I = SI x (LVsm - APmean) x 0.0136 x 9.8 x 1000' Equation 7. AVAGor = SV x (ET x 44.3 x APmean1'2)'1
Equation 8. AVACont = SV x VTIas'1
Echocardiography (Studies II- V)
Thallium-201 SPECT (Studies III and IV)
The presence and severity of concomitant aortic or mitral regurgitation was assessed using colour flow imaging. Left ventricular wall motion was assessed using 2-D echocardiography. Myocardial mass was calculated according to the cube formula of Troy [Troy et al 1972]. A myocardial mass index >150 and >120 g/m2 in men and women, respectively, was regarded as a sign of myocardial hypertrophy [Levy et al 1987]. Aortic valve area was calculated using the continuity equation,
A V A = V T I lv o t X LVOTarea x V T Ia s \ In Study II the flow and valve area were calculated before and during pharmacological stress with dipyridamole. The percent change in flow was calculated as the percent change in V T Ilvotdivided by the ejection time from
baseline to the post-dipyridamole study. The ejection time was measured as the interval between the onset and end of the transvalvular flow recorded by Doppler. The percent change in valve area was calculated as the change in the fraction V T Ilvot/V T Ias
from baseline to the study immediately after dipyridamole.
In Study V, extended evaluation of wall motion abnormalities was undertaken before and during dipyridamole. Left ventricular wall motion was assessed using a 16 segment model [American Society of Echocardiography Committee on Standards 1989]. Depressed wall motion at baseline in at least one segment or a new segmental wall motion abnormality after dipyridamole was considered to indicate the presence of significant coronary artery disease.
Two minutes after the pharmacological stress with dipyridamole, 74 MBq 201T1 was injected intravenously. Image acquisition started within 10 minutes of the injection of 201T1. The images were acquired on a Gamma 11 computer system and tomographic reconstructions were performed. The reconstructed short axis images were transferred to a Macintosh Usi computer for further evaluation. One basal, one mid- ventricular and one apical short-axis slice were selected manually. For each selected slice the outer boundary of the myocardium was defined by the operator. Starting at the “3 o ’clock” position and moving clockwise the computer program divided the slice into 6° segments and the highest activity in each segment was normalised to the highest activity in any segment in the slice (Figures 2 and 3). The relative activity in each segment was then plotted. Vascular territories were assigned for each slice as follows: The 147°- 315° segment was assigned to the left anterior descending coronary artery (LAD), the 316°-63° segment was assigned to the left circumflex coronary artery (LCX) and the 64°-146° segment to the right coronary artery (RCA).
Coronary angiography (Studies i-V)
Coronary angiography was performed in all patients via the femoral artery according to Judkins, with multiple angulations and amplification technique. A visually judged area reduction of at least 75% in at least one of the major coronary arteries or in a major branch was considered as a significant
coronary lesion. An area reduction of at least 75% in the main stem was considered as 2- vessel disease, regardless o f the presence or absence of stenoses in the left anterior descending or the left circumflex coronary artery.
Statistics
Linear regression analyses were performed and Student's two-tailed paired and unpaired t-tests were used where appropriate. In all statistical tests, the null hypothesis was rejected at the 5% level (p<0.05).
Figure 2. Quantitative analysis techniques. After operator determination of the outer boundary of the myocardium, the computer divided the cardiac image into sixty 6° segments and the maximum activity within each segment was measured.
1 0 0 % -r o 80% 20% « -0% 360° Anterior Posterior 180° 0° Location
Figure 3. Thallium-201 distribution for the short-axis slice in Figure 2.
RESULTS AND DISCUSSION
Hemodynamic effects of dipyridamole
The dipyridamole infusion caused a signifi cant increase in heart rate and a decrease in both systolic and diastolic blood pressure. This was observed in all studies. In Study IV, which involved 109 patients, there was an increase in heart rate from a mean of 68 (SD 13) beats/min to a mean of 82 (SD 13) beats/min (p <0.001) and a decrease in systolic and diastolic blood pressure from 142 (SD 21) mmHg to 135 (SD 22) mmHg (p<0.001) and from 84 (SD 13) mmHg to 81 (SD 14) mmHg (p<0.001), respectively. There was thus only a slight decrease in blood pressure after the dipyridamole infusion, which means that the influence on cerebral blood flow ought to be minimal as long as the patient remains in the supine position.
Stroke volume and flow increased after dipyridamole in all patients in Study I (thermodilution method) and in 29 out of 34 patients in Study II (echo-Doppler). The hemodynamic response in both studies thus compared favourably (Table 1). After theophylline administration there was no difference in stroke volume or flow compared with baseline (Study I).
More detailed analysis of the hemodynamic changes was possible in Study I where invasive measurements were made. No significant change in left ventricular systolic or end-diastolic pressure was found. A significant decrease in systemic vascular resistance index was found in all patients after dipyridamole, from 44(SD 8.8) to 31(SD 7.0) mmHg x I'1 x min x m2BSA (p<0.01). The systemic vascular resistance increased after theophylline but it did not return to baseline. Left ventricular stroke work index increased after dipyridamole, from 0.83(SD 0.13) to 0.95(SD 0.16) J x b e a t1 x m2B S A 1 (p<0.0001). The increase in the left ventricular work was due to increased work to overcome the aortic outflow resistance and due to increased net left ventricular work.
Comparison of the aortic left ventricular stroke work index (the work required to overcome the resistance of the stenotic valve) and the net left ventricular stroke work index (the work required if, hypothetically, there is no pressure gradient across the valve) indicates that, even in patients with severe aortic stenosis, less than one-third of the ventricular work is explained by the resistance over the stenotic valve. The
CO (1 x m in'1) SV (1) Flow (ml x s'1) ET (s) Rest 4.39(0.80) 70(17) 213(54) 0.329(0.042) Study I Dipyridamole 5.72(1.12)*** 81(17)*** 241(59)** 0.341(0.038) Study II Rest 4.26(1.49) 64(23) 224(75) 0.285(0.035) Dipyridamole 5.63(1.92)*** 70(24)*** 243(80)*** 0.287(0.031)
Table 1. Means (SD) for hemodynamic parameters in Studies I and II before and after dipyridamole. Statistically significant differences between rest and dipyridamole are indicated (paired t-test: *p<0.05, **p<0.01, ***p<0.001).
systemic vascular resistance appears to be more important than the resistance over the aortic valve for determination of cardiac output.
There was also a significant increase in pressure time per minute after dipyridamole. Even if it is not possible to calculate the intrinsic left ventricular work with this method, the increase in both left ventricular stroke work index and pressure time per minute clearly suggest an increase in myocardial oxygen demand. However, after theophylline, there was no significant change compared with baseline.
Four mechanisms have previously been suggested to explain dipyridamole's useful ness in stress testing 1. shortened coronary perfusion time, 2. lowering of the coronary perfusion pressure, 3. coronary dilatation, and 4. steal phenomena [Picano 1992]. A fifth important reason could be added in patients with aortic stenosis - increased left ventricular oxygen demand. In previous studies in patients and animals without aortic stenosis, increased myocardial oxygen demand has not been considered a major factor for dipyridamole-induced ischemia [Picano et al 1986, Fung et al 1987].
In Study II there was an increase in APpeak
from 74(29) to 78(29) mmHg (p<0.05) and
APmean from 50(21) to 54(22) (pcO.Ol) when dipyridamole was given. In Study I the transvalvular peak to peak gradient had a tendency to increase (from 52 to 61 mmHg), not statistically significant.
As there were only minor differences in the hemodynamic measurements when baseline data were compared with measurements after theophylline, theophylline appears to be a
suitable antidote to dipyridamole also in patients with aortic stenosis. Approximately 35% of the patients reported chest pain or chest tightness during the dipyridamole infusion and approximately 5% reported mild headache. After theophylline the symptoms were relieved in all patients except in a few who were given sublingual nitroglycerin. There were no serious adverse effects.
Dependence of aortic valve area on
transvalvular flow
In Study I (invasive study) flow and aortic valve area were assessed at baseline, after dipyridamole infusion and after theophylline. In Study II (Doppler study) flow and aortic valve area were assessed at baseline and after dipyridamole. Measurement of stroke volume is critical in the evaluation of AVA. In these studies, two entirely different methods were used, thermodilution and Doppler, which gave very similar results.
In Study I a new methodology was intro duced, AVA was calculated not only according to the Gorlin formula but also by application of the continuity equation. The use of the simplified Bernoulli equation to transform velocity into gradients has been validated in a number of studies [Hatle et al 1980, Stamm et al 1984, Berger et al 1984, Hegrenaes et al 1985] and the reverse, the transformation of gradients to velocity, should be equally accurate. When the velocity time integral is obtained, the aortic valve area can be calcu lated according to the continuity equation, whereby the need for a constant in the equation is eliminated. The effective orifice area is obtained as opposed to the anatomic area, which the original Gorlin equation is considered to estimate.
The aortic valve area was found to be larger after dipyridamole (Table 2). A correlation was found between flow and aortic valve area both in Study I and Study H However, there was a wide variation of the flow dependence of the stenotic valve area between patients. Some patients showed a large increase in valve area with increased flow while others showed virtually no change at all (Figure 4). One possible explanation is that some stenotic valves are stiffer and less affected by
changes in flow than others. It is thus difficult to predict the change in valve area from the change in flow in an individual patient.
In Study II flow decreased after dipyridamole infusion in four patients who all showed a reduction of the AV A. The reduction of flow was presumably caused by induction of myocardial ischemia during dipyridamole.
Study I Study II
Rest Dipyridamole Rest Dipyridamole
AVAGorlin 0.75(0.39) 0.83(0.41)** — —
AVAcont 0.69(0.36) 0.77(0.38)** 0.67 (0.26) 0.71(0.26)***
Table 2. Means (SD) for calculated AVA (cm2) in Studies I and II before and after dipyridamole. Statistically sig nificant differences between rest and dipyridamole are indicated (paired t-test: *p<0.05, **p<0.01, ***p<0.001).
y = 1.10 + 0.5 6*
r = 0.72
<
>
<
0) O)c
CO SZo
-30
-20
-10
0
10
20
30
40
% Change flow
Figure 4. Patients in Study II. A linear correlation between per cent change in aortic valve area (% Change AVA) and per cent change in transvalvular flow (% Change flow) was found (p<0.001).
The severity of a valvular aortic stenosis is often expressed as the size of the valve area. W e observed increases in valve areas of up to 24 % with increased flow. If a strict limit for severe stenosis (e.g. <0.75 cm2) is used without regard to the hemodynamic situation, the severity of a stenosis can be underestimated in patients with increased transvalvular flow (aortic regurgitation, inotropic stimulation, etc.), while the opposite is true in patients with reduced flow (left ventricular dysfunction, left to right shunt, etc.).
Noninvasive assessm en t of coronary
artery disease.
The purpose of Studies HI and IV was to determine whether or not SPECT can be used to detect or exclude CAD.
In Study III the specificity could not be calculated due to the fact that patients with normal coronary angiograms were compared only with patients having both insignificant and significant lesions according to angiography. Specific limits with regard to gender were not used. Nevertheless, the sensitivity for significant lesions was 88% when the lowest relative activity in each segment in the group without CAD was used as the lower limit of normal. The sensitivity was thus in at least the same magnitude as reported previously in small studies using planar imaging in patients with aortic stenosis [Huikuri et al 1987, Aubry et al 19911. Encouraged by these results, Study IV was initiated.
In Study IV, 10 men and 10 women, none of whom had angiographic signs of coronary artery disease, a history of myocardial infarction, left bundle branch block or
localised hypokinesia (echocardiography), were selected as a reference group and the other eighty-nine patients were evaluated prospectively. In both the reference and the prospective groups, 70% of the patients had a history of angina pectoris. The frequency of chest pain provoked by dipyridamole was 15% in the reference group and 35% in the prospective group. The patients in the reference group showed a mean aortic valve area of 0.70 (SD 0.25) cm^ compared with 0.75 (SD 0.37) cm2
[n
^ prospective group (n.s.). Ninety-one percent of the men and 92% of the women had myocardial hypertrophy. In the prospective group of 89 patients, 19 patients had no angiographic signs of coronary disease, 13 patients had coronary lesions considered non-significant (< 75 % area reduction), and 57 patients had significant coronary artery disease (>75% area reduction).In the evaluation of scintigraphic data, the mean activity curve, the lower limit of range curve, the mean minus 2 SD and minus 2.5 SD curves in each slice were calculated separately for men and women in the reference group (Figure 5).
For men, the mean-2.5 SD curves yielded a sensitivity of 100% and a specificity of 75%, and positive and negative predictive values of 94% and 100%, respectively (Table 3). Using mean-2.0 SD or the minimum (range) as the lower limit of normal resulted in a decrease in specificity to 50% and 38%, respectively, while the sensitivity remained at 100%.
Evaluation of different vessel territories was done at the mean-2.5 SD level (Table 4). The highest sensitivity (87%) and specificity (83%) were found for lesions in the left anterior descending coronary artery. The
sensitivity was lowest for the right coronary artery, 69%. No attempt was made to adjust for individual variation in coronary anatomy. Patients with non-significant coronary lesions (5 men and 8 women) were evaluated
separately. Four of 5 men with coronary lesions with an area reduction of less than 75% fell below the lower limit of normal regardless of whether -2.5 SD, -2 SD or range was used as discriminating thresholds.
100% c 60% 2; 4 0% Mean * Range »- -2 sd * -2.5 sd 0 % Anterior 360° 180c Posterior 100% c o c c 60% o O ^ 40% Mean » Range <> -2 sd * -2.5 sd | Anterior Posterior Location Location 100% o 80% Mean Range -o -2 sd * -2.5 sd | 20% 0% Anterior 360“ Posterior 0° 100% c 60% 2 40% -2.5 sd I Mean Range Posterior 180° Anterior 100% 80% 60% 40% Range -o- -2 sd -2.5 sd — Mean
%
20% 0% Anterior 360° 0C Posterior 100% 25-c o JO c c o Ü o C\J1 ♦ Mean -» Range -o- -2 sd * -2.5 s d |
Posterior
0° 180° Anterior
f
Figure 5 The distribution of thallium-201 uptake for the ten men (a-c) and the ten women (d-f) in the reference groups. Figures a and d represents basal planes, b and e the mid-ventricular planes and c and f the apical planes.
Men Scintigraphy Sensitivity 100%
+ Specificity 75%
Angiography + 34 0 Positive predictive value 94%
2 6 Negative predictive value 100%
Women Scintigraphy Sensitivity 61%
+ Specificity 64%
Angiography + 14 9 Positive predictive value 78%
4 7 Negative predictive value 44%
Table 3. Results obtained when the mean-2.5 SD curves derived from the reference groups was applied prospectively to a new group of patients.
Men
LAD Scintigraphy Sensitivity 87%
+ Specificity 83%
Angiography + 26 4 Positive predictive value 93%
2 10 Negative predictive value 71%
LCX Scintigraphy Sensitivity 81%
+ Specificity 63%
Angiography + 21 5 Positive predictive value 78%
6 10 Negative predictive value 67%
RCA Scintigraphy Sensitivity 69%
+ Specificity 69%
Angiography + 18 8 Positive predictive value 78%
5 11 Negative predictive value 58%
Women
LAD Scintigraphy Sensitivity 47%
+ Specificity 65%
Angiography + 8 9 Positive predictive value 57%
6 11 Negative predictive value 55%
LCX Scintigraphy Sensitivity 38%
+ Specificity 57%
Angiography + 5 8 Positive predictive value 36%
9 12 Negative predictive value 60%
RCA Scintigraphy Sensitivity 36%
+ Specificity 80%
Angiography + 5 9 Positive predictive value 56%
4 16 Negative predictive value 64%
Table 4. Extended evaluation at the mean- 2.5 SD level, with the heart divided into different vessel territories.
It was not possible to find a cut-off point which resulted in high sensitivity and high specificity for the female patients, for whom the diagnostic accuracy of the test was poor. The -2.5 SD curves resulted in a sensitivity of 61% and specificity of 64% (Table 3). Most previous studies with myocardial scin tigraphy have been undertaken in men. Myocardial scintigraphy is more difficult to evaluate in women than in men [DePuey et al 1989]. Breast attenuation artefacts are well known. An additional problem in the present study was that the majority of the women were overweight, 56% of them had a body mass index greater than 25 kg/m2 and 17% had a body mass index greater then 30 kg/m2. The range in body mass index was larger in females (18-44 kg/m2) than in males (17-34 kg/m2). M ost likely, this variation in body mass index is associated with a greater varia tion in attenuation in the female group. The female heart is usually smaller than the male heart, which may also cause interpretation difficulties. In this study more than 90% of the women had myocardial hypertrophy and the left ventricular cavity on the scintigraphic images was usually small. The results in females may be improved by a combination of appropriate attenuation correction, use of a technetium-99m-labelled perfusion agent and EKG-triggered gated acquisition.
Important gender differences may have been overlooked in the past. The need for gender specific reference groups for 201T1 distribution also in patients without aortic stenosis has been acknowledged, but there is a lack of large prospective studies evaluating test performance in women compared to men [Friedman et al 1982, Rabinovitch et al 1986, Eisner et al 1988, Kong et al 1992, Van Train et al 1993, Wenger 1994]. SPECT is performed more often in men than in women even if groups with
the same incidence of typical angina and cardiac risk factors are compared [Shaw et al 1994].
Despite the gender difference in test perform ance, the overall results of dipyridamole single photon emission computed tomo graphy in Study IV in men and women with aortic stenosis compare well with those of previous studies in patients without aortic stenosis [Leppo 1989]. If all 109 patients (48% females) are included in the evaluation, the overall sensitivity is 84% and the specificity is 75%, using -2.5 SD as the discriminating threshold.
In Study V, 2-D echocardiography at rest and after dipyridamole infusion was used to detect CAD (Figure 6). Using the combined criteria of either a segmental wall motion abnormality at base line or a new segmental wall motion abnormality after dipyridamole, the sensitivity for CAD was 85% and the specificity 70%. The sensitivity for single vessel disease was 71% (5/7) and for multivessel disease 92% (12/13). Sixteen of 17 patients (94%) with significant left anterior descending or multivessel disease were detected. All 6 patients with triple vessel disease were correctly identified as having significant coronary lesions.
An exclusion criterion in Study V was suboptimal echocardiographic image quality. This is an inherit limitation of transthoracic echocardiography and thus also a limitation for dipyridamole echocardiography in assess ing the presence of coronary artery disease. However, the proportion of patients excluded for this reason should be less than 10%. Patients with suboptimal transthoracic acoustic windows could be examined with transesophageal echocardiography.
A wall motion abnormality at rest is likely to al 1993]. These factors may cause resting wall be present in patients with a previous motion abnormalities. Thus there is a infarction but it may also be present in rationale for including not only a new or patients with severe coronary artery disease worsening segmental wall motion abnorma-without previous myocardial infarction, lity but also a resting segmental wall motion Aortic stenosis increases myocardial oxygen abnormality as a marker for myocardial demand and can also result in a systolic flow ischemia, indicating angiographically signifi-reversai in the coronary arteries [Yoshikawa et cant coronary lesions.
Figure 6 a-d. End diastolic and end systolic apical 2-chamber views of the left ventricle at baseline and post dipyridamole in a patient with significant coronary lesions: a) end diastolic baseline, b) end systolic baseline, c) end diastolic post-dipyridamole, d) end systolic post-dipyridamole. Post-dipyridamole there is a new wall motion abnormality in the apical region (denoted by arrows).
In order for dipyridamole echocardiography to be able to detect coronary artery disease coronary blood flow must be reduced suffi ciently enough to produce myocardial ischemia and segmental wall motion abnormalities. This flow reduction may be caused by redistribution of coronary artery blood flow from a stenotic to a nonstenotic vessel [Flameng et al 1974], or by redistribution of blood flow from the subendocardial to the subepicardial region [Picano et al 1986]. Sub endocardial to subepicardial steal may be more common in patients with aortic stenosis because of the high intraventricular pressure and left ventricular hypertrophy.
As there is evidence that dipyridamole causes an increase in myocardial oxygen demand in patients with aortic stenosis (Study I), dipyridamole echocardiography may be a more sensitive test in patients with aortic stenosis than in those without. Factors that tend to increase the sensitivity of a test may also decrease the specificity. This can explain the somewhat lower specificity for signifi cant lesions in our patients compared to the specificity in patients without aortic stenosis [Picano et al 1986].
Angiography is the gold standard in defining coronary artery anatomy, but the inter observer variability is a problem, especially in moderately severe lesions [De Rouren 1977]. More important is the fact that coronary angiography gives an estimate of the ana tomical narrowing of the arteries but the effect on blood flow is only assumed, and both under- and overestimation of the hemodynamic effect is possible. Coronary
angiography has been reported to be a poor predictor of coronary blood flow reserve [White 1984].
The scintigraphic method reflects the relative coronary perfusion reserve and thus gives a functional measure of the hemodynamic sig nificance of a coronary lesion. Also dipyri damole 2-D echocardiography addresses the significance of coronary lesions in a functional manner. Thus, complete agreement between coronary angiography and dipyridamole scintigraphy or 2-D echocardiography cannot be expected. One patient with an insignificant coronary stenosis judged by angiography but with abnormal scintigraphy developed ST- elevation after valve replacement and required reoperation and grafting of the left anterior descending artery.
If SPECT is performed in men with aortic stenosis approximately 50 % of the coronary angiographies in this patient group can be avoided. Since SPECT preferably can be performed in outpatients the overall cost of an examination is much less than the cost of an angiographic examination, consequently, the overall cost will be reduced. If 2-D echocardiography is used the cost reduction will be even greater, if the relatively low rate of detection of single vessel right or left circumflex coronary artery disease can be accepted. However, the performance of 2-D echocardiography is highly dependent on the skill of the echocardiographer whereas SPECT evaluation using the present method is almost operator independent.
GENERAL SUMMARY AND CONCLUSIONS
i.
During a dipyridamole stress test (0.56 mg/kg dipyridamole dissolved in 250 ml of saline infused and given over 4 minutes) with subjects in the supine position, patients with aortic stenosis increase their cardiac output, stroke volume, left ventricular work and myocardial oxygen demand and show a slight drop in blood pressure. This indicates that dipyridamole infusion according to the present protocol is not likely to impair cerebral blood flow and may be used as a diagnostic tool in patients with aortic stenosis. Intravenous dipyridamole in patients with aortic stenosis seems to be safe and theophylline is a suitable antidote.
3.
The present study establishes the gender specific normal distribution of 201T1 uptake in patients with aortic stenosis using dipyri damole to increase coronary blood flow. Computer assisted evaluation results in high sensitivity, specificity, and positive and negative predictive values for significant coronary artery stenoses in men using the mean-2.5 SD curve as the discriminating threshold. In women, however, this method has a considerably lower diagnostic accuracy. If SPECT is performed as part of the pre operative evaluation, a substantial number of men can be excluded from coronary angio graphy prior to aortic valve replacement.
2.
The size of the aortic valve area is flow dependent. In the present study, increases in valve area of up to 24 % were observed with increased trans valvular flow. This flow dependency of the aortic valve area has to be considered in the evaluation of patients with aortic stenosis.
4.
Using 2-D echocardiography and the combined criteria of a segmental wall motion abnormality at baseline or a new segmental wall motion abnormality after dipyridamole resulted in a high sensitivity for detection of multivessel or left anterior descending coronary artery disease (94%) with a moderate specificity (70%). Patients with a negative test might be excluded from coronary angiography if the relatively low rate of detection of single vessel right or left circumflex coronary artery disease can be accepted. Increased myocardial oxygen demand is likely to be an important factor when dipyridamole is used in echo cardiography stress testing in patients with aortic stenosis.
ACKNOWLEDGEMENTS
My sincere thanks to colleagues and personnel at the Departments of Clinical Physiology, Internal Medicine and Radiology for stimulating and encouraging co-operation and skilful aid and assistance.
In particular I want to thank:
Acting Professor Kjell Karp, my mentor and friend.
Dr Peter Eriksson, my co-tutor and friend.
Dr Dag Teien, my close co-worker and friend.
My wife Eva, for sharing life with me and Dr Per Bjerle, the head of the Department of our children Gunilla, Karolin, Björn and Clinical Physiology and my friend.
Erik.
This study was supported by grants from the Norrland Heart Fund, the Swedish National Association for Heart and Lung Patients and the Umeå University.
REFERENCES
American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two- Dimensional Echocardiograms: Schiller NB, Shah PM, Crawford M et al. Recommendations for quantification of the left ventricle by two- dimensional echocardiograhy. J Am Soc Echo 1989;2:358-67
Aubry P, Assayag P, Faraggi M, Haddad A, Akesbi A, Massian O, Brochet, E Bok, B Valére PE. Detection d’une maladie coronarienne avant chirurgie valvulaire. Apport de la scintigraphie myocardique sous dipyridamole. Ann Cardiol Angéiol 1991;40:9-13
Bache RJ, Wang Y, Jorgensen CR. Hemodynamic effects of exercise in isolated aortic stenosis. Circulation 1971; 44:1003-13
Berger M, Berdoff RL, Gallerstein PE, Goldberg E. Evaluation of aortic stenosis by continuous wave ultrasound. J Am Coll Cardiol 1984:3:150-6
Braunwald E. In Braunwald E. Heart disease. A textbook of cardiovascular medicine. 3ed W.B. Saunders Philadelphia 1988: 1052-60
Burwash IG, Pearlman AS, Kraft CD, Miyake-Hull C, Healy NL, Otto CM. Flow dependence of measurements of aortic stenosis severity during exercise. J Am Coll Cardiol 1994;24:1342-50
Cannon SR, Richards KL, Crawford M. Hydraulic estimation of stenotic orifice area: A correction of the Gorlin formula. Circulation 1985;71:6:1170-8
Cullhed I. Aortic stenosis: A clinical study of aortic stenosis, isolated or in combination with aortic insufficiency, with special reference to haemodynamic and angiographic observations. Uppsala, University hospital. 1964, Thesis
Czer LS, Gray RJ, Stewart ME, DeRobertis M, Chaux A, Matloff JM. Reduction in sudden late death by concomitant revascularization with aortic valve replacement. Thorac Cardi ovasc Surg 1988; 95:390-401
Danielsen R, Nordrehaug J, Vik-Mo H. Clinical and haemodynamic features in relation to severity of aortic stenosis in adults. Eur Heart J 1991 ; 12:791- 5
DePuey EG, Garcia EV. Optimal specificity of thallium-201 SPECT through recognition of imaging artifacts. J Nucl Med 1989;30:441-9
DeRouren T, Murray J, Owen W. Variability in the analysis of coronary arteriograms. Circulation 1977;55:324-28
Eisner R, Tamas M, Cloninger K, Shonkoff D, Oates J, Gober A, Dunn D, Malko J, Churchwell A, Patterson R. Normal SPECT thallium-201 Bull's- eye display: Gender differences. J Nucl Med 1988;29:1901-9
Exadoctulus N, Sugrue D, Oakley C. Prevalence of coronary arteryu disease in patients with isolated aortic stenosis. Br Heart J 1984;51:121-4
Flameng W, Wuster B, Schaper W. On the distribution of myocardial blood flow II. Effects of arterial stenosis and vasodilatation. Basic Res Cardiol 1974;69:435-46
Friedman TD, Greene AC, Iskadrian AS, Hakki AH, Kane SA, Segal BL. Exercise thallium-201 myocardial scintigraphy in women: correlation with coronary arteriography. Am J Cardiol
1982;49:1632-7
Fung AY, Gallagher KP, Buda AJ. The physiologic basis of dobutamine as compared with dipyridamole stress interventions in the assessment of critical coronary stenosis. Circulation 1987; 76:943-51
Gorlin R, Gorlin SG. Hydraulic formula for calculation of the areas of mitral valve, other cardiac valves, and central circulatory shunts. Am Heart J 1951;41:1-29
Gould KL, Westcott RJ, Albro PC, Hamilton GW. Noninvasive assessment of coronary stenoses by myocardial imaging during pharmacological coronary vasodilatation. II. Clinical methodology and feasibility. Am J Cardiol 1978;41:279-87
Greenberg B. Aquired aortic valve disease. In: Greenberg B, Murphy E ed. Valvular Heart Disease, Littleton: PSG Publishing Company 1987:147-73
Hall R. Aortic stenosis. In: Julian D, Camm A, Fox K, Hall R, Poole-Wilson P ed. Diseases of the heart, Balliére Tindall: W B Saunders 1989: 707-30
Hatle L, Angelsen BA, Tromsdal A. Non-invasive assessment of aortic stenosis by Doppler ultrasound. Br Heart J 1980;43:284-292
Hegrenaes H, Hatle L. Aortic stenosis in adults: noninvasive estimates of pressure drop with Doppler. Br Heart J 1985; 54:396-404
Helmcke F, Nanda N, Hsiung M, Soto B, Adey C, Goyal R, Gatewood R. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75,175-83
Holm S, Eriksson P, Karp K, Österman G, Teien D. Quantitative assessment of aortic regurgitation by combined two-dimensional, continous wave and colour flow Doppler measurements. J Intern Med 1992;231:115-21
Huikuri HV, Korhonen UR, Ikäheimo MJ, Heikkilä J, Takkunen JT. Detection of coronary artery disease by thallium imaging using a combined intravenous dipyridamole isometric handgrip test in patients with aortic valve stenosis. Am J Cardiol 1987; 59:336-40
Jones M, Schofield P M, Brooks N H, et al. Aortic valve replacement with combined myocardial revascularisation. Br Heart J 1989; 62:9-15
Karp K, Teien D, Eriksson P. Doppler echocardiographic assessment of the valve area in patients with atrioventricular valve stenosis by application of the continuity equation. J Intern Med 1989;225:261-6
Knabb R, Gidday J, Ely S, Rubio R, Berne R. Effects of dipyridamole on myocardial adenosine and active hyperemia. Am J Physiol 1984;247:804-10
Kong B, Shaw L, Miller D, Chaitman B. Comparison of accuracy for detecting coronary artery disease and side-effect profile of dipyridamole thallium- 201 myocardial perfusion imaging in women versus men. Am Coll Cardiol 1992;70:168-73
Leppo J. Dipyridamole-thallium imaging: The lazy man’s stress test. J Nucl Med 1989;30:281-7
Levy D, Savage DD, Garrison RJ, Anderson KM, Kannel WB, Castelli WP. Echocardiographic criteria for left ventricular hypertrophy: The Framingham Heart Study. Am J Cardiol 1987; 59:956-60
Lindblom D, Lindblom U, Qvist J, Lundström H. Longterm relative survival rates after heart valve replacement. J Am Coll Cardiol 1990;15:566-73
Lindroos M, Kupari M, Heikkilä I, Tilvis R. Prevalence of aortic valve abnormalities in the elderly: An echocardiographic study of a random population sample. J Am Coll Cardiol 1993; 21:1220-5
McKay R. The Mansfield scientific aortic valvulu- plasty registry:Overview of acute hemodynamic results and procedural complications. J Am Coll Cardiol 1991;17:485-91
Nylander E, Ekman I, Marklund T, Sinnerstad B, Karlsson E, Wranne B. Severe aortic stenosis in elderly patients. Br Heart J 1986;55:480-7 Olofsson B-O, Bjerle P, Åberg T, Österman G,
Jacobsson K-A. Prevalence of coronary artery disease in patients with valvular heart disease. Acta Med Scand 1985;218:365-71
Olsson M. Valvular aortic stenosis: studies on etiology, mortality and quality of life. Stockholm, Karolinska institute, 1994. Thesis
Olsson M, Dalsgaard C-J, Haegrestrand A, Rosenqvist M, Rydén L, Nilsson J. Accumulation of T-lymphocytes and expression of interleukin-2 receptors in nonrheumatic stenotic aortic valves. J Am Coll Cardiol 1994;23:1164-70
Pellika P, Nishimura R, Bailey K, Tajik J. The natrual history of adults with asymptomatic, hemo- dynamically significant aortic stenosis. J Am Coll Cardiol 1990;15:1012-7
Pfleger K, Niederau D, Volkmer I, Stock H. Ein beitrag zum Wirkungsmechanismus von dipyrida mole: Hemmung der adenosinaufnahme in erythrocyten durch dipyridamole. Naunyn Schmeidebergs Arch Pharmacol 1969;265:118- BO
Picano E, Lattanzi F, Masini M, Distante A, 1 / Abbate A. High dose dipyridamole echocardio graphy test in effort angina pectoris. J Am Coll Cardiol 1986;8:848-54
Picano E, Simonetti I, Masini M et al. Transient myocardial dysfunction during pharmacologic vasodilatation as an index of reduced coronary reserve: a coronary hemodynamic and echocardio- graphic study. J Am Coll Cardiol 1986;8:84-90
Picano E. Anatomic and functional targets of stress testing. In Stress echocardiography. Springer- Verlag, Berlin Heidelberg 1992: 9-15
Rabinovitch M, Suissa S, Elstein J, Standoff H, Tang A, Rush C, Aldis A, Tannous R, Turek M, Addas A, O'Brien C, Burgess J, Rosenthall L. Sex- specific criteria for interpetation of thallium-201 myocardial uptake and washout studies. J Nucl Med 1986;27:1837-41
Rahimtoola S. Perspective on valvular heart disease: An update. J Am Coll Cardiol 1989;14:1-23
Seltzer A. Changing aspects of the natrual history of valvular aortic stenosis. N Engl J Med 1987; 317:91-8
Shaw L, Miller D, Romies J, Kargl D, Younis L, Chaitman B. Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease. Ann Inter Med 1994;120:559-66
Skjaerpe T, Hegrenaes L, Hatle L. Noninvasive estimation of valve area in patients with aortic stenosis by Doppler ultrasound and two-dimensional echocardiography. Circulation 1985;72:810-18
Sollevi A, Östergren J Fagrell B, Hjelmdahl P. Theophylline antagonizes cardiovascular responce to dipyridamole in man without affecting increases in plasma adenosine. Acta Physiol Scand 1984;121:165-71
Stamm BR, Martin BR. Quantification of pressure gradients across stenotic valves by Doppler ultrasound. J Am Coll Cardiol 1984:2:707-18. Troy BL Pombo J, Rackley CE. Measurements of left
ventricular wall thickness and mass by echocardiography. Circulation 1972;45:602-11
Van Train K, Areeda J, Garcia E, Cooke D, Maddahi J, Kiat H, Germano G, Silagan G, Folks R, Berman D. Quantitative same-day rest-stress technetium-99m-sestamibi SPECT: Definition and validation of stress normal limits and criteria for abnormality. J Nucl Med 1993;34:1494-502
Wenger N. Coronary Heart disease in women: Gender differences in diagnostic evaluation. JAMWA
1994;49:181-97
White CW, Wright CB, Doty DB et al. Does visual interpretation of the coronary angiogram predict the physiologic importance of a coronary stenosis? N Engl J Med 1984;310:819-24
Yang SS. Assessment of ventricular function. In Yang SS, ed. From Cardiac Catheterisation Data to Hemodynamic Parameters. 2nd ed. Philadelphia: F.A. Davis Company 1978:233-358
Younis LT, Chaitman BR. Update in intravenous dipyridamole cardiac imaging in the assessment of ischemic heart disease. Review. Clin Cardiol 1990;13:3-10
Yoshikawa J, Akasaka T, Yoshida K, Takagi T. Systolic coronary flow reversal and abnormal diastolic flow patterns in patients with aortic stenosis: Assessment with an intracoronary Doppler catheter. J Am Soc Echo 1993;6:516-24
