Survival of Medically Treated Patients in the
Coronary Artery Surgery Study (CASS) Registry
MICHAEL B. MOCK, M.D., IVAR RINGQVIST, M.D., LLOYD D. FISHER, PH.D.,
KATHRYN B. DAVIS, PH.D., BERNARD R. CHAITMAN. M.D., NICHOLAS T. KOUCHOUKOS, M.D.,
GEORGE C. KAISER, M.D., EDWIN ALDERMAN, M.D., THOMAS J. RYAN, M.D..
RICHARD 0. RUSSELL. JR., M.D., SUZANNE MULLIN, R.N., M.P.H., DAVID FRAY, B.A.,
THOMAS KILLIP III, M.D., AND PARTICIPANTS IN THE CORONARY ARTERY SURGERY STUDY
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
SUMMARY The objective of this study was to evaluate the impact on survival of the anatomic extent of
obstructive coronary artery disease and of two measures of left ventricular (LV) performance. This study is
based on 20,088 patients without previous coronary artery bypass graft surgery who were enrolled in the
registry of the National Heart, Lung, and Blood Institute Coronary Artery Surgery Study from 1975 to
1979. The cumulative 4-year survival of medically managed patients was analyzed to determine the survival
of specific subsets of patients with obstructive coronary disease. The vital status of 99.8% of the patients was
known. The 4-year survival of medically treated patients with no significant obstructive disease was 97%, in
contrast to 92%, 84% and 68% in patients with one-, two- and three-vessel disease, respectively. The
presence of left main coronary artery disease decreased survival significantly. The 4-year survival decreased
from 70% to 60% in patients with three-vessel disease when significant obstruction of the left main
coronary artery was also present. Patients with significant coronary artery disease who had an ejection
fraction of 50-100%o, 35-49%, and 0-34% had a 4-year survival of 92%, 83% and 58%, respectively. The
systolic contraction pattern was assessed in five selected segments and given a score of 1-6, with a score of I
for normal function, increasing to 6 if an aneurysm was present. In a patient with normal LV contraction in
all five segments of the LV ventricular angiogram, the LV score would equal 5. Patients with an LV score of
5-11, 12-16 and 17-30 had 4-year survivals of 90%, 71 % and 53 %, respectively. Patients with good LV
function (a score of 5-11) had a 4-year survival of 94%, 91% and 79% for one-, two- and three-vessel
disease, respectively. Patients with poor left ventricular function (score of 17-30) had a 4-year survival rate
of 67%, 61% and 42% in one-, two- and three-vessel disease, respectivelv. Thus, LV function is a more
important predictor of survival than the number of diseased vessels.
THE CUMULATIVE survival of patients with coronary artery disease who receive medical management,
as reported in studies since Herrick and Nuzum's'
clinical description of 1918, is extremely variable. Before the development of coronary angiography, the
extent of coronary artery disease could not be reliably
determined. Therefore, it is not surprising that the
5-year survival rates reported ranged from 15% to
75%.'1-"
The poor survival rates of medically treated patients
reported in earlier studies has influenced the increasing
frequency with which surgical management is recommended for patients with coronary artery disease. It is
important to obtain as precise a classification of patients with coronary artery disease as possible when
considering therapy on the basis of long-term survival.
The frequent use of the bypass surgery to treat coronary artery disease makes it difficult to study the natural history of patients with obstructive coronary artery
disease in this country. Many patients suspected clinically of having coronary artery disease have been
studied by angiography in the National Heart, Lung,
and Blood Institute Coronary Artery Surgery Study
(CASS)."2 A detailed analysis of the coronary arterio-
grams and left ventriculograms makes it possible to
characterize certain patient subgroups. The complete
follow-up in the CASS group has made possible an
accurate study of the survival of medically treated patients within these subgroups.
Methods
The CASS registry contains 24,959 patients who
were admitted to the 15 cooperating centers with suspected or proved coronary artery disease. This study is
based on an analysis of 20,088 registry patients with
no previous cardiac surgery who were enrolled from
August 1975 through May 1979. The details of the
CASS study design, methods and baseline data have
been described.'2
Seventy-five percent of the patients were males. The
average age was 53 years, and 58% were employed
full-time at the initial evaluation. Thirty-three percent
were currently smoking cigarettes and 42% had formerly smoked cigarettes. Thirty-one percent had a
cholesterol level of more than 250 mg/dl. Seventythree percent presented with typical anginal chest pain,
and of these, 6% had Canadian Heart Association class
I angina, 30% class 1I, 36% class III and 16% class IV;
12% had angina at rest. Forty-eight percent of the
patients had a history of myocardial infarction. Ten
percent had diabetes mellitus.
All of the patients underwent coronary angiography,
performed by either the brachial or femoral technique,
and multiple views of each vessel were obtained. The
extent of arterial stenosis, defined as the percentage of
From the Coordinating Center for Collaborative Studies in Coronary
Artery Surgery, University of Washington, Seattle, Washington.
Address for correspondence: Coordinating Center for Collaborative
Studies in Coronary Artery Surgery, JD 30. University of Washington,
Seattle, Washington 98105.
Received July 6, 1981; revison accepted February 10, 1982.
Circulation 66, No. 3, 1982.
562
MEDICAL PATIENTS IN CASS/Mock et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
maximal narrowing of the luminal diameter, was recorded for each of 27 coronary artery segments. The
criterion for clinically significant one-, two- or threevessel coronary artery obstruction was either 70% or
more reduction in the internal diameter of the right or
left anterior descending or left circumflex coronary
artery or 50% or more reduction in the internal diameter of the left main coronary artery.
The ejection fraction was calculated from left ventricular (LV) angiograms and provided a good global
measure of LV function. Not all patients had determinations of ejection fraction, primarily because of technical and logistic reasons. By evaluation of the segmental LV performance pattem, calculated as an LV
score, LV function could be determined in 97% of the
patients. This LV score provides a quantitative assessment of the segmental abnormalities of the left
ventricle. 12
With the use of a 300 right anterior oblique view of
the LV angiogram, the ventriculogram was divided
into five segments: anterobasal, anterolateral, apical,
diaphragmatic and posterolateral. The systolic contraction pattern of each of the five segments was numerically scored: 1 = normal, 2 = moderate hypokinesis, 3 = severe hypokinesis, 4 = akinesis, 5 =
dyskinesis and 6 = aneurysm present. The LV score is
the sum of the points of these five segments. Thus, in a
patient with normal systolic contraction in all five segments of the LV angiogram, the LV score would be 5
(fig. 1).
The decision for surgical or medical therapy in registry patients, was made by the study physicians in
conjunction with the referring physicians and the patient. Although some patients with significant coronary disease were treated medically because they were
anatomically or hemodynamically unsuitable for surgery, 93% of all patients in this study had at least one
operable vessel. The proportion of each subgroup
treated medically varied from site to site. Since patients were not randomized in this registry study, the
decision to treat a patient medically or surgically was
determined by prevailing clinical opinion. Considering
1. Anterobasal
I.
Anterolateral
V. Posterobasal
IV. Diaphragmatic
FIGURE 1. Terminology for the five left ventricular segments
of the left ventriculogram analyzed in the right anterior oblique
view. LV = left ventricle; LA = left atrium.
563
the expected differences of such opinions between institutions and the variability of patient populations, the
data presented in this study are viewed as being representative of clinical practice in the United States and
Canada during this period.
Patients who were chosen for surgical therapy were
initially regarded as medically treated. Many patients
who were initially treated medically later had coronary
bypass graft surgery. Most often, subsequent therapy
was performed because of the failure of medical therapy to provide adequate control of the patient's angina.
At the time of coronary bypass surgery, patients were
withdrawn from further analysis. Exposure time was
calculated from the time of enrollment in the CASS
registry; early deaths before scheduled surgery were
considered a failure of medical therapy.
Survival curves were calculated by the life-table
method and were compared by log-rank statistics.'3
For 99.8% of these patients, the vital status was
known.
Results
In this group of patients with significant disease who
were referred for coronary angiography, 5804 patients
were found not to have significant obstruction of the
coronary arteries. The 4-year survival of patients with
no significant obstructive coronary disease was 97%.
The anatomic extent of disease in the coronary vessels
was directly related to survival. The 4-year survival
rates were 92%, 84% and 68% for patients with one-,
two- and three-vessel disease, respectively (fig. 2).
Patients with at least one operable vessel did not differ
significantly with regard to survival from patients with
inoperable vessels; therefore, these groups were considered together for this study. In patients with twovessel disease, the presence of left main coronary artery disease did not significantly (p = 0.89) alter the
4-year survival - 80% compared with 84% (fig. 3).
However, in patients with three-vessel disease, the
presence of an obstructed left main coronary artery
decreased the 4-year survival to 60%, compared with
70% in patients with three-vessel disease without left
main coronary artery disease (p < 0.001).
The LV ejection fraction had a major effect on survival. Patients who had at least one-vessel disease,
without 50% or more left main coronary artery obstruction and an ejection fraction of 50-100%, 3549% and less than 35%, had a 4-year survival of 92%,
83% and 58%, respectively (p < 0.0001) (fig. 4).
Within the one-, two- and three-vessel disease subgroups, ejection fraction was a very important predictor of survival. For each ejection fraction category,
survival decreased according to the number of vessels
obstructed (p < 0.001) (fig. 4).
A good correlation was also found between LV performance based on the LV score and cumulative survival. This is expected, since the LV ejection fraction
and the LV score are correlated (r = - 0. 74). Patients
with an LV score of 5-11, 12-16 and 17-30 had different 4-year survivals - 90%, 71% and 53%, respectively (p < 0.0001) (fig. 5). Analysis of survival was
564
z. °~DIS
CIRCULATION
VOL 66, No 3, SEPTEMBER 1982
DISEASED VESSELS
-100
80
:D
(X
Z)
2,
..............................
I
.
a
a
X
.........
0
C-P
4
.................-
60.
6
40
m
ob.
...........
:.... .
.......
.
100
80
6~ 0
.........
...
L20
LEGEND
,
3VES
VES
VES
VES
DIS
.................
0
1
2
..
3
...............
i
.140
........
20
..........
I
I
O
1
2
3
4
N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL
99
5804 100
5429
98
5695
4529
98
3081
97
97
4138 100
2521
2716
96
2026
94
1301
92
95
- 4482 100
1743
91
1370
88
1943
850
84
88
1301
80
X- 5629 100
1004
74
1576
614
68
P < .0001
Log Rank Stat = 1247.301
YEAR
0
0
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
FIGURE 2. Cumulative 4-year survival of all of the medically treated CASS registry patients. Four groups are
shown on the basis of angiographic extent of coronary obstructive diseases. DISVES = diseased vessels.
also made on the basis of both extent of obstruction and
LV performance. Within the three subgroups of LV
function, survival decreased according to the number
of vessels diseased (p < 0.001). The effects of the
extent of vessels diseased in the subgroups of patients
with good LV function (LV score 5-11), moderate
impairment of LV function (LV score 12-16) and poor
LV function (LV score 17-30) are listed in figure 5.
Patients with good LV function had a 4-year survival
of 94%, 91% and 79% for one-, two- and three-vessel
100'
disease, respectively. The 4-year survival rates of the
patients with one- and two-vessel disease and poor
ventricular performance were 67% and 61%, respectively. The 4-year survival of patients with three-vessel disease and poor ventricular performance was 42%.
This is significantly lower than that for the one- and
two-vessel disease groups with poor ventricular performance (p = 0.001).
The percentage of patients being treated medically
at a fixed time after angiographic study is presented in
..................................................-
80
80
D
60
40
LEGEND
c-
LLJJ
A-
z
X-E-
C)
w
YF-AR
100
40
....
2 VES DIS, NO LM
2 VES DIS, LM
3 VES DIS, NO
3 VES D!S, LM
LM
..
2
0
0
60
.........
1
1.
...
I1
...
20
0
3
4
70SURVIVAL N %SURVIVAL N %SURVIVAL
91
1330
88
821
84
89
40
85
80
29
82
879
76
541
70
74
125
66
73
60
2
N %SURVIVAL N %SURVIVAL N
4195 100
1885
95
1692
287 100
58
94
51
x4445 100
1383
89
1140
1184 100
193
84
161
P
< .0001
Log Rank Stat = 1 60.1 30
FIGURE 3. Cumulative 4-year survival of CASS registry patients with two- and three-vessel disease (VES DIS)
without left main coronary artery disease (LM) compared with the survival of patients with two- and three-vessel
disease with left main coronarv artery disease.
565
MEDICAL PATIENTS IN CASS/Mock et al.
EJECTION FRACTION
100
>D
(1)
1 DISEASED VESSEL
80
60
40
...
LEGEND
I- EJECFR 50-100
-
20
oIflL
YEAR
.EJECFR
X
D
.
'½
(I)
60 F
..1 40
z
3
4
N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL
92
2802
96
2232
94
1403
98
3081
6791 100
83
702
86
416
988
94
893
89
1977 100
58
63
410
73
318
66
183
513
909 100
=
P < .0001
Log Rank Stat = 481.1 99
60
40
20
20
O
.,
0
YEAR
0
4
3
2
1
N %SURVIVAL N %SURWVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL
761
95
1201
96
1492
98
1601
99
100
a.- 2517
92
184
91
307
390
95
416
97
535 100
57
74
92
80
120
85
140
92
172 100
P < .0001
Log Rank Stat = 78.308
3 DISEASED VESSEL
2 DISEASED VESSEL
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
100
100
100
J
.
40
LEGEND
Ie-- EJECFR 50-100
35-49
I EJECFR
EJECFR 0-34
0l
LLi
0
E
2
80
60
20
1
80
r-
35-49
0-34
-EJECFR
y-J
>
a-
I........:.......
Q1
100
1004
80
....................................
z
LQi
a-
100
...
hi
__...........
100
.........................................
-j
80 _..
...
..
80
80
80
Y-
()
z
Li
0
LLi
ara.
YEAR
40
_
_
z
LXi
C)
40
LEGEND
n
I
01.
0
,, 1
..t
1
,
2241
60
60
~~~~~~~~~~~~~~~~~.
40
40
LEGEND
LiJ
.--EJECFR 50-100
EJECFR 3 -49
0-34
EJECFR
.2
N %SURVIVAL
97
91
73
3
20 . ..
20
.........................
N %SURVIVAL N %SURVIVAL
846
100
925
98
657 100
95
310
335
84
135
172
294 100
P < .0001
Log Rank Stat - 168.040
9-
D
60
60
IO
4
N %SURVIVAL N %SURVIVAL
93
95
415
672
144
83
87
242
57
57
64
102
'
YEAR
O
°- EJECFR 50-100
~-~- EJECFR 35-49
EJECFR
'
'
*
=
''
*
*
1
N %SURVIVAL
N %SURVIVAL
556
2033 100
237
785 100
201
443 100
P < .0001
Log Rank Stat = 110.644
a
95
88
77
*
*
'
patients with left main
50% or more received
1
*
2
3
4
N XSURVIVAL N %SURVIVAL N %SURVIVAL
82
87
227
90
359
464
71
88
76
153
194
80
50
58
69
124
67
155
FIGURE 4. Four-year survival data for patients with at least one-vessel disease, less than 50% left main coronanr
and a measured ejection fraction (EJEC FR).
figure 6. The percentage of patients being treated
medically was higher among those with less severe
angina (p < 0.0001) and those with fewer vessels
diseased. There was more surgery for patients with
proximal vessels diseased when the analysis was restricted to CASS patients with three-vessel disease.
The same findings (not presented) hold true for patients with one- and two-vessel disease. Among patients with three-vessel disease, a higher proportion of
patients were treated medically when they had poor LV
function compared with patients who had three-vessel
disease and better LV function. The number of patients
receiving a coronary bypass graft operation was influenced by angina class, extent of coronary disease,
number of proximal obstructions greater than 70%,
and LV performance. Figure 6 shows the rate and time
until surgery for patients on the basis of these four
important variables.
Figure 7 shows the percentage of patients who received coronary artery bypass surgery within 6 months
of their enrollment in CASS. This figure reinforces in
more detail the findings of figure 6. The percentage
treated surgically increased as the number of vessels
diseased increased; the percentage also increased with
a better ventricular performance for a fixed number of
vessels diseased. Among the CASS participants, most
20
........................
0-34
artery
obstruction,
coronary artery obstruction
surgery within 6 months.
of
Discussion
In the CASS registry, both the extent of obstructive
coronary disease and the status of LV function were
evaluated in a standardized manner with routine quality control of the angiographic readings. The 99.8%
follow-up of this large group of patients has made
possible an evaluation of the relationship between both
the number of obstructed major coronary arteries and
LV performance and the subsequent survival of medically treated patients in a number of subgroups.
The cumulative survival of medically treated patients in the CASS group is better than that reported in
most earlier studies, particularly in patients with threevessel disease. The survival of medically treated patients in the Duke University data bank is similar to
that in CASS experience.9
Many factors may have influenced the difference.
One major factor may be the use of different techniques of statistical analysis. 12. 14 Another major difference between this study and some earlier reported studies is the medical treatment provided. In this study,
B-blocking drugs were prescribed frequently for medically treated patients. The earlier studies may also
= ~ ~ ~ ~ ~ ~10
CIRCULATION
566
VOL 66, No 3. SEPTEMBER 1982
LV SCORE
LV SCORE 5- 11
100
80
U)
80
60
[
wL-
40
...
...........................
LEGEND
a:
LVSCORE 5-Il1
LVSCORE 12-16
LVSCORE 17-30
uLJ
0a-
20
80
80
60
U)
60~
40
z
LLJ
C)
40
z
0
100
100
...
o
oL
0
1
2
3
4
N %SURVI/VAL N %SURVPVAL N %SURVIVAL N %SURVIVAL N XSURVIVAL
100
4478
97
4063
95
3214
92
2048
90
o-10533
2545 100
1232
89
1069
82
853
77
513
71
718 100
381
78
308
67
242
59
150
53
P < .0001
Log Rank Stat = 657.773
YEAR
a-
YEAR
40
LEGEND
LLI
20
:60
DIS VES
VES
...231 DIS
DIS VES
20
20
................
O
0
0
1
2
3
4
N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURVIVAL N %SURV!VAL
2149
99
2000
97
1602
96
3390 100
1033
94
1398
98
1272
95
1004
93
636
3387 100
91
94
782
89
602
922
84
374
79
3745 100
P < .0001
Log Rank Stat = 175.150
LV SCORE 12-16
L 'I SCORE 17-30
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
100'
100
100
100'
J
80
_..
80
....................
80
80
D.SE.SED
80
U)
.......
60
.60
LLI
40
40
LEGEND
LJ
YEAR
+
=
60
60
~~~~~~.
F-
z
1
DIS
20 .........2.DIS.
3 DIS
0
1
0
I/ES
ES....
I/ES
C)
40
2
3
4
88.225
'40
LEGEND|
r
20
I
P
=
LLI
20
N %SURVIVAL N %SURVIVAL N %SURVITVAL N %SURVIVAL N %SURIVAL
92
89
198
371
87
401
95
306
502 100
91
327
85
259
80
151
370
71
733 100
1300 .100
458
85
370
74
287
66
163
59
< .0001
Log Rank Stat
z
VES
1-It3IVES
VES
DISEASED
20
DISEASED
DISEASED 1
o
YEAR
0
1
2
N %SURVIVAL N %SURVIVAL N %SURVIVAL
89
86
83
95
129 100
83
102
72
127
204 100
157
69
118
56
382 100
P < .0001
Log Rank Stat = 27.464
3
N %SURVNAL
67
76
80
63
94
49
i
o
4
N XSURI/VAL
40
67
50
61
59
42
FIGURE 5. Four-year survival data for patients with one-, two- or three-vessel disease. The upper left panel contains all the cases
placed into three subgroups according to left ventricular (LV) score; see textfor definition. The other three panels present data for
each subgroup.
have included more patients with advanced congestive
failure and ventricular aneurysm. Still another difference between our study and others is the availability of
ventricular performance data on most patients. The
CASS registry has permitted the delineation of subgroups of patients within the one-, two- and threevessel stratification disease categories on the basis of
LV performance.
LV contraction patterns have an important influence
on subsequent survival, particularly in patients with
more extensive anatomic involvement of their coronary vessels. '5 Again, this study shows that ventricular
performance may be a more important predictor of
subsequent survival than the number of diseased
vessels.
The 4-year cumulative survival of medically treated
patients in the CASS registry is lower than that reported for the medically assigned patients in the Veterans
Administration coronary artery surgery study16: onevessel, 91% vs 96%; two-vessel, 83% vs 87%; and
three-vessel, 68% vs 74%.16
The European Coronary Artery Surgery Study enrolled only patients who had at least two-vessel disease
and good LV function. The 4-year survival of the
CASS patients with good ventricular function shows
close correlation with the 4-year cumulative survival
of patients in the European study assigned to medical
therapy. The comparison between CASS and the European Randomized Study for patients with two-vessel
and three-vessel disease is, respectively, 94% vs 93%
and 82% vs 84%. 4
It is difficult, under the best of circumstances, to use
medical survival in a registry such as CASS for medical-surgical comparisons. In patients with three-vessel
disease, the 4-year survival is 42-79%, depending on
the performance of the left ventricle. Thus, to have
reasonable validity in such comparisons, one must
have comparable data on both medically and surgically
treated patients. This means that the same clinicians
should record data at the same time period for both
medically and surgically treated cases. It is rare that
two published papers in the literature contain sufficient
data to allow a sophisticated adjustment to another data
set. In addition, such data from two sources are recorded by different clinicians, so the comparability of the
data is suspect. An additional complicating factor is
surgery occurring at variable intervals among those
who were initially treated therapeutically by medical
treatment alone. The decision for surgery depends differentially on many characteristics. For example, with
MEDICAL PATIENTS IN CASSIMock et al.
CLASS
D'SEASED VESSELS
100
'SO,
r..
567
au
o
>Z0 -40--. 410uuy
100
4
Li
tY
.,
80
80
0
d
60
L-
.
A
0
1
I
I
2
3
100
100
52'1
2135
648
1895
576
42
30
40
1498
462
41
29
959
29
Iog
Rank
Stat
28
20
4138
4482
1
2722
67
45
5629
1
1575
31
1294
1 00
Log Rank Stat
CHCLAS
2
1951
3
CHCLAS 11
CHCLAS III
4
N %MEDICAL
N %MEDICAL
N %MfDICAl
99
4526
99
3081
98
65
2030
63
1301
62
41
43
1369
42
849
5425
2523
1742
996
29
.0001
<
28
28
606
LGN
e 6729.854
1
EGEND
I
3 DIEASECHLS rV
DISVES -02
DISVES
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
3
40
-,.
"--
20 .-
o~
P
I
-818.749
I
1
YEAR
0
N %MEDICAL
N %MEDICAL
_ 5804 100
5690
99
00
40
323
0
00
CL
N %MEDICAL
N %MEDICAL
N %MEDICAL
N %kOEDICAL
Nl %MEDICAL
66
287
67
460
65
618
69
575
902 '00
2805
64
2597
63
2142
61
1426
60
4419 100
2337
6
40
,
4
60
0
LJ
.0001
<
60
LJ
20
o L.
P
z
z
40
20
80
0
60
40
L
P
80
D
I
-3
DISEASED VESSELS
3 DISEASED VESSELS
100
LJ
100
0
:
a)
80
80
0
0
60
z
60
z
t-
Li
40
C)
z
LA
LAJ
a-
40
LiJ
a.
0
NEAR
_
I
0
II
,
1
2
N %MEDICAL
N %MEDICAL
100
335- 43
I 62
00
558 36
1
48
28
792
1
00
1351
23
N %MEDICAL
41
34
467
348
260
804
292
189
100
3
226 %MEPICAL
39
N
363
24
33
26
2
149
.0001
P <
Lag Rank Stat
.
JO
4
N %MEDICAL
135
38
223
32
156
25
20
93
I
0'
1
0
YEAR
N %MEDICAL
N %MEDICAL
929
26
_ 3745 100
454
1300
P
LEGN
111.612
Io-
PRXVES
IPRXES
;PRXVES
-
<
100
154
.
2
3
N %MEDICAL
763
24
365
39
40
57
116
N %MEDICAL
92
IV
2
LV SCURE
5-11
12-16
LVSCORE
a
17-3m
XME0ICAL
56
22
37
55
SCORE
5-11
RE 12-16
=
Canadian Heart
given number of vessels diseased, the surgical cases
a
higher proportion
of
patients
who have
good
ventricular performance than in the overall CASS registry. Mathematical adjustment in response to this fac-
100
80
80
71
60 W
40S
-J O
4
LV SC0RE 17-30
3I
include
:D
N
373
1 60
55
LGN
LEGEND
LV
.0001
FIGURE 6. Percentage ofpatients treated without coronary arterv bvpass surgery is presented by time. CHCLAS
Association functional classification; LV = left ventricular.
LV SCORE
23
600
28
55
LOg ROnk Stat - 207.640
0
1
PRXVES
100
382
EGEND
-
20
56
S ,
}
a
tor depends on fine subdivision for its validity or on
Xassumptions about a mathematical model. All of this
again implies that data such as those presented here
not profitably be used for medical/surgical survival comparisons when surgical survival curves from
151may
another data base are used.
The high cumulative survival
34
20L
S cu1519M ES
a |N
Ventr1CUlar
plain why
the
LMCA
50%
FIGURE 7. Percentage of patients receiving coronary artery
bvpass surgery within 6 months ofangiographic study. Percentage is presented for patients with one-, two- and three-vessel
disease, including left main coronary artery dis'ease (LMCA)
and patients with 50% or more LMCA disease.
European
patients
failed
to
show
a
with
disease
Randomized
Surgery
2 ence Artery
in survival between
nary
3
2
NUMBER OF DISEASED VESSELS
1
of
function and two-vessel
Study
good
may exof Coro-
significant
differ-
medically and surgically
treated patients with two-vessel disease, since this
study randomized only patients with good LV
performance.
The survival of these medically treated patients canbe used for direct comparison with the survival of
patients with comparable disease treated surgically,
not
568
CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
since not all medically treated patients were surgical
candidates. Also, many patients who received medical
treatment initially were subsequently treated surgically. Nevertheless, the poor cumulative survival of
medically treated patients with two- and three-vessel
disease and poor ventricular function may suggest earlier consideration of surgical treatment for these patients. However, patients with evidence of ventricular
dysfunction have been reported to have a higher operative mortality. The operative risk of coronary bypass
surgery at the 15 CASS clinical units and the factors
that influence it have been reported.'7 The operative
mortality for all coronary artery bypass graft surgery
was 2.3%. In patients with a history of congestive
heart failure, the operative mortality was 4.7%, and in
patients with pulmonary rales at baseline it was 7.5%.
The results of the CASS show better or similar cumulative survival of medically treated patients with
one-, two- and three-vessel disease than previously
reported. In addition, LV perfornance has an important influence on subsequent survival. Even when the
probability of survival is already high, such as in the
patients with only one or two obstructed coronary arteries, the effect of poor ventricular performance is
significant. When the severity of the arterial disease
increases, the impact of LV performance is even
greater.
References
l.Herrick JB, Nuzum FR: Angina pectoris: clinical experience with
two hundred cases. JAMA 70: 67, 1918
2. Sigler LH: Prognosis of angina pectoris and coronary occlusion:
follow-up of 1,700 cases. JAMA 146: 998, 1951
3. Zukel WJ, Cohen BM, Mattingly TW, Hrubec Z: Survival following first diagnosis of coronary heart disease. Am Heart J 78: 159,
1969
VOL 66, No 3, SEPTEMBER 1982
4. Pell S, D'Alonzo CA: Immediate mortality and five-year survival
of employed men with a first myocardial infarction. N Engl J Med
270: 915, 1964
5. Block WJ Jr, Crumpacker EL, Dry TJ, Gage RP: Prognosis of
angina pectoris: observations in 6,882 cases. JAMA 150: 259,
1952
6. Proudfit WL, Bruschke AVG, Sones FM Jr: Natural history of
obstructive coronary artery disease: ten-year study of 601 nonsurgical cases. Prog Cardiovasc Dis 21: 53, 1978
7. Burggraf GW. Parker JO: Prognosis in coronary artery disease:
angiographic, hemodynamic, and clinical factors. Circulation 51:
146, 1975
8. Oberman A, Jones WB, Riley CP, Reeves Ti, Sheffield LT, Turner ME: Natural history of coronary artery disease. Bull NY Acad
Med 48: 1109, 1972
9. Harris PJ, Harrell FE Jr, Lee KL, Behar VS, Rosati RA: Survival
in medically treated coronary artery disease. Circulation 60: 1259,
1979
10. Webster JS, Moberg C, Rincon G: Natural history of severe proximal coronary artery disease as documented by coronary cineangiography. Am J Cardiol 33: 195, 1974
11. Reeves TJ, Oberman A, Jones WB, Sheffield LT: Natural history
of angina pectoris. Am J Cardiol 33: 423, 1974
12. Killip T (ed): The National Heart, Lung, and Blood Institute Coronary Artery Surgery Study (CASS). Circulation 63 (suppl I): 1-1,
1981
13. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure
Time Data. New York, Wiley & Sons, 1980
14. European Coronary Surgery Study Group: Prospective randomised
study of coronary artery bypass surgery in stable angina pectoris.
Lancet 2: 491, 1980
15. Mundth ED, Austen WG: Surgical measures for coronary heart
disease. N Engl J Med 293: 13, 1975
16. Takaro T. and Participants in the V.A. Cooperative Study of Surgery for Coronary Arterial Occlusive Disease: Results of a randomized study of medical and surgical management of angina pectoris.
World J Surg 2: 797, 1978
17. Kennedy JW, Kaiser GC, Fisher LD, Fritz JK, Myers W, Mudd
JG. Ryan TJ: Clinical and angiographic predictors of operative
mortality from the Collaborative Study in Coronary Artery Surgery
(CASS). Circulation 63: 793, 1981
Survival of medically treated patients in the coronary artery surgery study (CASS)
registry.
M B Mock, I Ringqvist, L D Fisher, K B Davis, B R Chaitman, N T Kouchoukos, G C Kaiser,
E Alderman, T J Ryan, R O Russell, Jr, S Mullin, D Fray and T Killip, 3rd
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. 1982;66:562-568
doi: 10.1161/01.CIR.66.3.562
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1982 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/66/3/562
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/