- Email: [email protected]
Prognostic value of dipyridamole echocardiography early after myocardial infarction in elderly patients
Individuals X3 years of age account for 12%of the total population in the United States, twice the proportion existcted to increase by 20% 0 constitute >20% of the poptdation by the year 2030. Coronary heart disease achm the lstituto di Fisiologia Clinica, CNR, Piss, Italy. *A list of the main inwstip,ators and itmstitutionsinvolved in the EPIC group appears in the Append-.. Manuscript rweived March 25, 1W3; revised manuscript received July 6, 199 : Dr. Eugenio Picnno, Istituto di Fisiologia Clinica, CNR, Via Pactlo Savi, 8,561OQPisa, Italy. 01993 by the American Coke
of Cardiolow
no~i~vasiv~~ test inclini ary
arterydisease and it
tsw
CAMERIEIUETAL. DIP~IDAMOLEECWOCARE)IOGRAPZIYIN'F~EeLDERLY
provides important prognostic information (5). Unfortunately, it has important limitations in elderly patients because cf the physical stress employed and the KG auks that &em the induced myocardial ischemia. Dynamic exercise may be unfeasibleor inadequate in elderly patients because of the high prevalence of severe arthritis, V~S&N disease, chronic pulmonary disease, neur or generalizedweakness and physical &conditioning. Moreover, -50% ofpatients >65 years of age have abnormal rest
tory quality and availabilityo had a history of systemic differed from the younger (< general popuIation of the EP ischemia with respect to a p 9 MD), gender distribution(23%in women compnred with 12%in the yo and rest wall motion score index (1
myacardial~ofarctio~a
tic evaluation (f4-17) of corocular, the prognostic value of ventricular dysfunction was clearly established in patients evaluated early after acute
stems were used, In the
consisted of 262patients it for a first episade of fkrction, Of these 262
e Study 8rQupconsisted of 190patients 4 * 3-3 years, range 65 to 78, 147men cIinicaUyuncomplicatedacute myo~ar&line echocardiographicstudy of satisfac-
studies 8s welt as
obtained when possible. During the test, new areas of abnormal wall motion were identified in multiple views whenever possible. Test positiviry was based ou the detection of a transient asynegy ofcontraction that was absent or of a lesser degree at the baseline examination. Regions that were akinetic or dyskinetic at rest were excluded from the analysis. The videotapes were analyzed by a cardiologistechocardiographerwho had no knowledgeof the clinicaland angiographicdata and met quality control criteria for stress ech~~i~~a~hic readings required to enter the residual ischemia subprqject of the EPIC study (20). A wail motion score index was derive for rest and peak di~y~~amo~~ ech~ardiograms in each patient. The left ventricle was divided into t I segments. The 11 left ventricular segments considered for the analysis were: apex: ~~P&MI and distal anterior septum; proximal and distal inferior septum; proximal and distal anterior wall; proximaland distal lateral wall, and proximal and distal inferior wall (20). Segmentat wall motion was graded as: normal (normal
angina, coronary revascularization proc
access either to chaical and a~g~ogra~hicdata or to the
concerned. The second criterion was based on 20 consecutive studies from each cant 20 studies were examined in a blinde rienced cardiologist-echocardjogra~berof the coordinating center, whose reading was arbitrarily assumed to be the reference standard. It was assumed that the minimalthreshahty control bad to be old of concordance to pass Bowerthan that for the ~80%. The concordance cut first type of reading because nd set of tapes was not selected on the basis of superior quality but randomly sampled from each center in a consecutive fashion. All 11 enrolled centers met the minimal requirements for cllu data were obtained from at least one of the fohowing sources: review hospital record, personal communication w physician and review of the patient’s chart, view with the patient conducted by trained personnel and interview of the patient by a staff physician at regular intervals in the outpatient clinic. The outcome events were
artery bypass surgery and c~~~~~~ya
Student t test (two-tailed). Statistical eier life table estimates of survival, hard event-free survival, spontaneously occurring event-free survival and all event-free survival rates were used to su follow-upexperience in these patients and to clarifypreseniation. When event-free survival was stratified on the basis o resultant curves
s The dipyridamole ec~ocard~o~raQ echocardiographicfindings in patients are listed it-tTable 1.
1812
CAMERlERl ET AL. DIPYRDAMOLE ECHOCARDIOGRAPHY IN THE ELDERLY
All Patients
14 7 21 42 62
DET Results --Positive Negative (II = 105) (n = 85) 11 3 14 31 44
3 4 7 11 18
Events During the Follow-Up
With Events (n = 622)
Patients Without nts (n =
68.4 f 2.97
68.4 ” 3.5
d&M 19(30.6)
101R7 41 (321
Patients
p Value 0.018 0.775 0.0% O.OMl 0.00
Peri
Age (sr)
MaMfemale Arterialhypertension
to the results ofdipyridhe 120patientswho ufl) 43 (39%) krnda positive
fn nli ~~ti~~ts~the ~~~~~~d~~~ ~~t~~r~t~~~~ during the dipyridaherefore suit&le for ana%ysis.No mqjor et such U infmtion, death, asystoicand vantnculattachycurdiaoccurreddu ghy. However, in four ot be administeredbecau lower dose (hypotensionand symptomtic bndyctii~ in three md headachein on@)* The test resultsin these four
9s.Z% NEG
__.____82.9
0.0
I.0
8.0
MONTHS
12.0
16.0 _
a.0
% POS
ZLO
mu ----91.4
%
I 7..
^---.
._.
_..-.-__-.
r94 %
bIEP Pi”;
EM
----__.
I
/!I
%
hiG
I
1.0
1.0
___~~%!I 96 PO2
rates in I%) or pmiti
survival
with a negative (NEG) or p echocardiography.There was the two grBsups(p < -9.W. 100.0
NLG 81.0
60.0
POS
40.0
20.0.
Positive predictfve valueofpasitive DET t%l/a) Negative predictive v&w 0.0
I1 0.0
I
4.0
8.0
MONTl-tS
17.0
16.0
ill.0
24.0
of ac@e
DET (%I
Relative risk ratio Relative risk ratio for
death
DET = dipyridamole
echocardiography.
1gt4
CAMERIER! ET AL. DIPYRIDAMOLE ECHOCARDIBGRAPHY
JACC Vol. 22, No. 7 December 1993:l8W45
IN THE ELDERLY
to address the issues of either diagnostic accuracy for angjographicallyassessed coronary artery disease or comparison with exercise electrocardiography, we found that consistent with previous experiences (24), a positive dipyridamole echocardiographictest result was more frequently sso&ted with a positive exercise stress test (24).Our data further substantiate the prognostic value of dipyridamole echocardiography, as previously shown in a variety of patient subsets, rmging from those with chronic coronary artery disease (14), early postinfarction patients (1620) to ajar noncardiac vascular su s evaluated after coronary a dipyridamole cchocardio ostic ~t~ti~cation of elde one was c~nside~d, The low positive predictive value of wp0aitive test result v cardiac cvcnts shsuld be considarc ~~v~~~~c~of cardiac events in pa infarction who have no contrainion n we ex8mincd rei ant uld not And any s ivc and negative findingson dipyrardiography. This finding may be due to the of nonfatal myocardial infarctions (only sevbut it is dso not uncxpcctcd because even a “physio” approach cannot predict phenomena such as fissuraembslization, ulceration and thrombosis, which are largely unrelated to the hemodynamic severity of plaque. It is also obvious that the ologic mechanisms of death, reinfarction and un na are not likely to be the sww and no sin& test can simultaneously address all
echocardiography was 113 (Zg%) of 40‘7patients in the overall study group but only 13 (IS%) of 85 in the eP group. in our previoudy reported (20)overall analysis of the EPIC study patients, we showed that interventions dra icallyreduce the risk in the patients w amole echocardiography result (with vascularization showing an 1I-fold comparison with those without r~vascu~a~~~tio~).There-
rices in survival can bc easily identified by too oldfor risk str~ti~cation. stratificationcan be safely an dose di~y~damo~eechocardiography~
mortality rate das, as can be t.hoelderly group (‘IS>than in @spitethe similar follow-up time achieved by the response to Srt&iy ~~~t~~~~~~?ts~ tDr. Anplo Camerieri,OspedQle Geneva
f-
Semi Ponente, Genovrz; Milan; Dr. Giamni SWEXI, Prof. Pada hssi, Ospedak Maggiore, Novara; Dr. Stefarm Maei, Dr. Monica Baroni, Dr. Andrea Biagini, DEvisione di Dr.
tion Wdures
in the pup
with a
positive dipyridamole
hmcesco Faletra, Ospedale Niguarda,
*Unless otherwise indicated, dl instiutious are in Italy. tThis list does not include the cmthors
listed on the title page.
the patients unable lo exercise: alternatives eo exercise stress eesriog. Neart J 2%9;117:1344-65. IO. Armstrong WF. Stress echocardiography for detection of coronapv artery disease. Circulation i99J;g4~SupplJ):J-~44-8. II. ini istante A, L’Abbate A. in effort angina pectotis. phY
12.
+AtreteP, et d. ~ipy~da~o~e
echocardiography as in the assessment of coronary artery disease iu the elderly.J Am Geriatr SOS 1999B;39:993-9.
13. Agati L, Reazi M, Sciomer S, et al. ~ra~~esop~~~~~~~ dipyridamo~e ec~~ca~~~~~~a~hy for diagnosis of coronary artery diirase. J Am Coil Cardial ~~~2;1~:76~-7~. 14. Picauo E, Severi S, hiichelassi C, et al. Prognostic rmportance of diI’lyridamose=echocaPdiog~~~lny test iu coronary artery oii$ease. Circulaet al. Predictiou of major eveuls I; using d~~y~idamo~e ec~~~cardio~ra” S, et al. Stress lrsliug in the ~enod after 17. Cobu PF. Edilorial Comment. Silent left ventrictdar ~y~f~~c~~oa during rdiography: a uew proguostic marker. J Am Cdl
r acme Ol)KHXdiai 1. Smith SC Jr, Gilpiu E. Ahuve S, et al. Outlook tienls 3ged6%to 7! infarction in the very elderly COlll~ilKd with that i years. 3 Am Coil Cardiol 1990:16:784-92. 2. Stone PM. Turi ZG, Muller JE, et al. Prognostic significance of the treadmill exercise test pcrformauce 6 months after myocardial infarckm. J Am Coil Cardiol 1%6;8:1037-17. 3. Waters DD, Bosch X, Bouchard A, et al. Comparison of clinicalvariables and variables derived from limited predischarge exercise test as predictor of e&y aod late mortality aftermyocardial infarction. J Am Coil Cardiol 1985;5: I-13. 4. Berning 8, Steensgaard-Hansen F. Early estimation of risk by cchocardiographic determinaiion of wall motion index in an unselected population with acme myocardial infarction. Am J Cardiol 1990;65:567-76. 5. Sheffield LT. Exercise stress lest. In: Braunweld E, editor. Heart Disease: A lextbook of' Cardiovascular Medicine. Philadelphia: Saunders, 1988:X%-41 I 6. Mihalick MJ, Fisch C. Electrocardiographic findings in the aged. Am Heart J 1974;g7: I 17-9.
ossi L, Sarasso 6, et al. Silent versus symplomatic dipyridamole-induced ischemia after myocardial infarc prognostic significance. J Am Coll Cardiol 1992;19:95319. Picano E, Marini C Pirelli S. et al, Safety of intravenous high dose dipy~~damo~e-ec~~ocardio~rap~~y. Am J Cardio 20. Picano E. Landi P, Bolognese L. et al. dipyridamole-echocardiography early after uncomplicated myocardial infarction: a large scale rn~~tice~~e~ trial. Am 3 21. Fletcher GF, Froelicher VF, Hartley LH, Haskell WL, Pollock ML. Exercise standards: a statement for health professionals from the American Heart Associarion. Special report. CircuBation 1 ~~~;~~~-~2~, ?2. Lam NT, Chaimun BR. Glacnzer M, et 81. Safety and diagao\tic accuracy of dipyridamole-thallium imaging in the elderiiy. J Am Co!! Cindi0l 1988; 1I :585-9. 23. Shaw I_.,Chaitman BR. !Hilton ‘IT, et al. ~ro~ao~~~c value ofdipyri thallium~201 imaging in elderly patients. J Am COB Cardiol B 1390-9. 24. Picano E, L&tab F, Dipyridamole-echocardiogphy test: a new diagnostic window on coronary heart disease. Circutarion 1991;83(suppJ. I%J):JJ1~19-26. 25. Massa D, Pirclli S, Gara E, et al. Exercise testing and dipyridamoleechocardiography test before and 48 hours after successful angioplasly: tic implication. Em Heart J 1989;10:13-7. J, Simpfendorfer C. ArnoldAZ, PiedmonleM, EyUeBW. Early and late results of coronary angioplasty and bypass in octogenarians. Am J Cardiol 1991;68:1316-20. 27. Belier GA. Are you too old to he risk stratified? J Am Coil Cardiol 1992;19:1399-401.
of Cardiolow
no~i~vasiv~~ test inclini ary
arterydisease and it
tsw
CAMERIEIUETAL. DIP~IDAMOLEECWOCARE)IOGRAPZIYIN'F~EeLDERLY
provides important prognostic information (5). Unfortunately, it has important limitations in elderly patients because cf the physical stress employed and the KG auks that &em the induced myocardial ischemia. Dynamic exercise may be unfeasibleor inadequate in elderly patients because of the high prevalence of severe arthritis, V~S&N disease, chronic pulmonary disease, neur or generalizedweakness and physical &conditioning. Moreover, -50% ofpatients >65 years of age have abnormal rest
tory quality and availabilityo had a history of systemic differed from the younger (< general popuIation of the EP ischemia with respect to a p 9 MD), gender distribution(23%in women compnred with 12%in the yo and rest wall motion score index (1
myacardial~ofarctio~a
tic evaluation (f4-17) of corocular, the prognostic value of ventricular dysfunction was clearly established in patients evaluated early after acute
stems were used, In the
consisted of 262patients it for a first episade of fkrction, Of these 262
e Study 8rQupconsisted of 190patients 4 * 3-3 years, range 65 to 78, 147men cIinicaUyuncomplicatedacute myo~ar&line echocardiographicstudy of satisfac-
studies 8s welt as
obtained when possible. During the test, new areas of abnormal wall motion were identified in multiple views whenever possible. Test positiviry was based ou the detection of a transient asynegy ofcontraction that was absent or of a lesser degree at the baseline examination. Regions that were akinetic or dyskinetic at rest were excluded from the analysis. The videotapes were analyzed by a cardiologistechocardiographerwho had no knowledgeof the clinicaland angiographicdata and met quality control criteria for stress ech~~i~~a~hic readings required to enter the residual ischemia subprqject of the EPIC study (20). A wail motion score index was derive for rest and peak di~y~~amo~~ ech~ardiograms in each patient. The left ventricle was divided into t I segments. The 11 left ventricular segments considered for the analysis were: apex: ~~P&MI and distal anterior septum; proximal and distal inferior septum; proximal and distal anterior wall; proximaland distal lateral wall, and proximal and distal inferior wall (20). Segmentat wall motion was graded as: normal (normal
angina, coronary revascularization proc
access either to chaical and a~g~ogra~hicdata or to the
concerned. The second criterion was based on 20 consecutive studies from each cant 20 studies were examined in a blinde rienced cardiologist-echocardjogra~berof the coordinating center, whose reading was arbitrarily assumed to be the reference standard. It was assumed that the minimalthreshahty control bad to be old of concordance to pass Bowerthan that for the ~80%. The concordance cut first type of reading because nd set of tapes was not selected on the basis of superior quality but randomly sampled from each center in a consecutive fashion. All 11 enrolled centers met the minimal requirements for cllu data were obtained from at least one of the fohowing sources: review hospital record, personal communication w physician and review of the patient’s chart, view with the patient conducted by trained personnel and interview of the patient by a staff physician at regular intervals in the outpatient clinic. The outcome events were
artery bypass surgery and c~~~~~~ya
Student t test (two-tailed). Statistical eier life table estimates of survival, hard event-free survival, spontaneously occurring event-free survival and all event-free survival rates were used to su follow-upexperience in these patients and to clarifypreseniation. When event-free survival was stratified on the basis o resultant curves
s The dipyridamole ec~ocard~o~raQ echocardiographicfindings in patients are listed it-tTable 1.
1812
CAMERlERl ET AL. DIPYRDAMOLE ECHOCARDIOGRAPHY IN THE ELDERLY
All Patients
14 7 21 42 62
DET Results --Positive Negative (II = 105) (n = 85) 11 3 14 31 44
3 4 7 11 18
Events During the Follow-Up
With Events (n = 622)
Patients Without nts (n =
68.4 f 2.97
68.4 ” 3.5
d&M 19(30.6)
101R7 41 (321
Patients
p Value 0.018 0.775 0.0% O.OMl 0.00
Peri
Age (sr)
MaMfemale Arterialhypertension
to the results ofdipyridhe 120patientswho ufl) 43 (39%) krnda positive
fn nli ~~ti~~ts~the ~~~~~~d~~~ ~~t~~r~t~~~~ during the dipyridaherefore suit&le for ana%ysis.No mqjor et such U infmtion, death, asystoicand vantnculattachycurdiaoccurreddu ghy. However, in four ot be administeredbecau lower dose (hypotensionand symptomtic bndyctii~ in three md headachein on@)* The test resultsin these four
9s.Z% NEG
__.____82.9
0.0
I.0
8.0
MONTHS
12.0
16.0 _
a.0
% POS
ZLO
mu ----91.4
%
I 7..
^---.
._.
_..-.-__-.
r94 %
bIEP Pi”;
EM
----__.
I
/!I
%
hiG
I
1.0
1.0
___~~%!I 96 PO2
rates in I%) or pmiti
survival
with a negative (NEG) or p echocardiography.There was the two grBsups(p < -9.W. 100.0
NLG 81.0
60.0
POS
40.0
20.0.
Positive predictfve valueofpasitive DET t%l/a) Negative predictive v&w 0.0
I1 0.0
I
4.0
8.0
MONTl-tS
17.0
16.0
ill.0
24.0
of ac@e
DET (%I
Relative risk ratio Relative risk ratio for
death
DET = dipyridamole
echocardiography.
1gt4
CAMERIER! ET AL. DIPYRIDAMOLE ECHOCARDIBGRAPHY
JACC Vol. 22, No. 7 December 1993:l8W45
IN THE ELDERLY
to address the issues of either diagnostic accuracy for angjographicallyassessed coronary artery disease or comparison with exercise electrocardiography, we found that consistent with previous experiences (24), a positive dipyridamole echocardiographictest result was more frequently sso&ted with a positive exercise stress test (24).Our data further substantiate the prognostic value of dipyridamole echocardiography, as previously shown in a variety of patient subsets, rmging from those with chronic coronary artery disease (14), early postinfarction patients (1620) to ajar noncardiac vascular su s evaluated after coronary a dipyridamole cchocardio ostic ~t~ti~cation of elde one was c~nside~d, The low positive predictive value of wp0aitive test result v cardiac cvcnts shsuld be considarc ~~v~~~~c~of cardiac events in pa infarction who have no contrainion n we ex8mincd rei ant uld not And any s ivc and negative findingson dipyrardiography. This finding may be due to the of nonfatal myocardial infarctions (only sevbut it is dso not uncxpcctcd because even a “physio” approach cannot predict phenomena such as fissuraembslization, ulceration and thrombosis, which are largely unrelated to the hemodynamic severity of plaque. It is also obvious that the ologic mechanisms of death, reinfarction and un na are not likely to be the sww and no sin& test can simultaneously address all
echocardiography was 113 (Zg%) of 40‘7patients in the overall study group but only 13 (IS%) of 85 in the eP group. in our previoudy reported (20)overall analysis of the EPIC study patients, we showed that interventions dra icallyreduce the risk in the patients w amole echocardiography result (with vascularization showing an 1I-fold comparison with those without r~vascu~a~~~tio~).There-
rices in survival can bc easily identified by too oldfor risk str~ti~cation. stratificationcan be safely an dose di~y~damo~eechocardiography~
mortality rate das, as can be t.hoelderly group (‘IS>than in @spitethe similar follow-up time achieved by the response to Srt&iy ~~~t~~~~~~?ts~ tDr. Anplo Camerieri,OspedQle Geneva
f-
Semi Ponente, Genovrz; Milan; Dr. Giamni SWEXI, Prof. Pada hssi, Ospedak Maggiore, Novara; Dr. Stefarm Maei, Dr. Monica Baroni, Dr. Andrea Biagini, DEvisione di Dr.
tion Wdures
in the pup
with a
positive dipyridamole
hmcesco Faletra, Ospedale Niguarda,
*Unless otherwise indicated, dl instiutious are in Italy. tThis list does not include the cmthors
listed on the title page.
the patients unable lo exercise: alternatives eo exercise stress eesriog. Neart J 2%9;117:1344-65. IO. Armstrong WF. Stress echocardiography for detection of coronapv artery disease. Circulation i99J;g4~SupplJ):J-~44-8. II. ini istante A, L’Abbate A. in effort angina pectotis. phY
12.
+AtreteP, et d. ~ipy~da~o~e
echocardiography as in the assessment of coronary artery disease iu the elderly.J Am Geriatr SOS 1999B;39:993-9.
13. Agati L, Reazi M, Sciomer S, et al. ~ra~~esop~~~~~~~ dipyridamo~e ec~~ca~~~~~~a~hy for diagnosis of coronary artery diirase. J Am Coil Cardial ~~~2;1~:76~-7~. 14. Picauo E, Severi S, hiichelassi C, et al. Prognostic rmportance of diI’lyridamose=echocaPdiog~~~lny test iu coronary artery oii$ease. Circulaet al. Predictiou of major eveuls I; using d~~y~idamo~e ec~~~cardio~ra” S, et al. Stress lrsliug in the ~enod after 17. Cobu PF. Edilorial Comment. Silent left ventrictdar ~y~f~~c~~oa during rdiography: a uew proguostic marker. J Am Cdl
r acme Ol)KHXdiai 1. Smith SC Jr, Gilpiu E. Ahuve S, et al. Outlook tienls 3ged6%to 7! infarction in the very elderly COlll~ilKd with that i years. 3 Am Coil Cardiol 1990:16:784-92. 2. Stone PM. Turi ZG, Muller JE, et al. Prognostic significance of the treadmill exercise test pcrformauce 6 months after myocardial infarckm. J Am Coil Cardiol 1%6;8:1037-17. 3. Waters DD, Bosch X, Bouchard A, et al. Comparison of clinicalvariables and variables derived from limited predischarge exercise test as predictor of e&y aod late mortality aftermyocardial infarction. J Am Coil Cardiol 1985;5: I-13. 4. Berning 8, Steensgaard-Hansen F. Early estimation of risk by cchocardiographic determinaiion of wall motion index in an unselected population with acme myocardial infarction. Am J Cardiol 1990;65:567-76. 5. Sheffield LT. Exercise stress lest. In: Braunweld E, editor. Heart Disease: A lextbook of' Cardiovascular Medicine. Philadelphia: Saunders, 1988:X%-41 I 6. Mihalick MJ, Fisch C. Electrocardiographic findings in the aged. Am Heart J 1974;g7: I 17-9.
ossi L, Sarasso 6, et al. Silent versus symplomatic dipyridamole-induced ischemia after myocardial infarc prognostic significance. J Am Coll Cardiol 1992;19:95319. Picano E, Marini C Pirelli S. et al, Safety of intravenous high dose dipy~~damo~e-ec~~ocardio~rap~~y. Am J Cardio 20. Picano E. Landi P, Bolognese L. et al. dipyridamole-echocardiography early after uncomplicated myocardial infarction: a large scale rn~~tice~~e~ trial. Am 3 21. Fletcher GF, Froelicher VF, Hartley LH, Haskell WL, Pollock ML. Exercise standards: a statement for health professionals from the American Heart Associarion. Special report. CircuBation 1 ~~~;~~~-~2~, ?2. Lam NT, Chaimun BR. Glacnzer M, et 81. Safety and diagao\tic accuracy of dipyridamole-thallium imaging in the elderiiy. J Am Co!! Cindi0l 1988; 1I :585-9. 23. Shaw I_.,Chaitman BR. !Hilton ‘IT, et al. ~ro~ao~~~c value ofdipyri thallium~201 imaging in elderly patients. J Am COB Cardiol B 1390-9. 24. Picano E, L&tab F, Dipyridamole-echocardiogphy test: a new diagnostic window on coronary heart disease. Circutarion 1991;83(suppJ. I%J):JJ1~19-26. 25. Massa D, Pirclli S, Gara E, et al. Exercise testing and dipyridamoleechocardiography test before and 48 hours after successful angioplasly: tic implication. Em Heart J 1989;10:13-7. J, Simpfendorfer C. ArnoldAZ, PiedmonleM, EyUeBW. Early and late results of coronary angioplasty and bypass in octogenarians. Am J Cardiol 1991;68:1316-20. 27. Belier GA. Are you too old to he risk stratified? J Am Coil Cardiol 1992;19:1399-401.