Download "False Negative" Treadmill Exercise Test

The "False Negative" Treadmill Exercise Test
and Left Ventricular Dysfunction
NEIL KRAMER, M.D., ARMANDO SUSMANO, M.D.,
AND RICHARD B. SHEKELLE, PH.D.
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
SUMMARY One hundred and fifteen consecutive symptomatic patients undergoing graded exercise testing, selective coronary angiography and left ventriculography were retrospectively evaluated. The
sensitivity, specificity, and false negative response rates of the exercise tests were 79%, 81%, and 21%, respectively. Although the magnitude of a positive ST-segment response was related to more extensive
vascular disease, the frequency of false negative responses was nearly
identical in patients with single, double, or triple vessel disease (22%,
21%, 19%). Analysis of the false negative group demonstrated significant ventriculographic and hemodynamic abnormalities when com-
pared to the true positive responders. Five out of six patients with the
most serious motion disorders in the study fell into the false negative
group. There were no significant differences in the extent, distribution and severity of vascular involvement, or in the development of collateral circulation in the two groups. However, occluded vessels
supplied abnormal ventricular segments more frequently in the false
negative group (88% vs 38%); the absence of an "ischemic response"
and the presence of segments of abnormal myocardium may be related.
Left ventricular dysfunction appears to be an important reason
for a false negative response to exercise.
ELECTROCARDIOGRAPHIC EXERCISE STRESS
TESTING has gained widespread acceptance as a noninvasive diagnostic tool in the evaluation of patients suspected
of having ischemic heart disease. With the introduction of
the multistage protocols, improved sensitivity over single
stage methods was expected. Recent studies, however, have
continued to demonstrate a high percentage of false negative
responses in symptomatic patients.'"4 These reports attributed such responses to less extensive vascular disease.
Others have suggested that left ventricular function may be a
contributing factor.' 1-2 The present study was designed to
assess further the relative influence of these parameters as
determinants of the false negative response to exercise in
symptomatic patients.
II, III, V4-6 were recorded at control, during and immediately after exercise and then at 3 and 6 min thereafter.
Exercise was considered complete if either an 85% maximal
target rate was achieved or a symptomatic end point was
reached (progressive chest pain, presyncope, fatigue or
dyspnea) or if the patient developed serious arrhythmia.
A positive test response consisted of either depression or
elevation of the ST segment 1 mm from baseline with
straightening or downsloping extending for 0.08 sec in any
monitored lead. In patients with resting ST-segment abnormalities, a change of at least 1 mm beyond the resting level
constituted a positive test.'3 14 The test was considered
negative if the patient achieved an 85% target heart rate
without significant ST changes. A false negative test consisted of the presence of significant coronary obstruction (as
defined below) associated with a negative treadmill response.
Patients failing to achieve a predicted target heart rate in the
absence of significant ST changes were regarded uninterpretable responses. The development of exercise-induced
chest pain was noted but not included in the definition of test
responses. The data were subjected to Chi-square analysis.
In those instances in which the number of patients was
small, a test on a 2 X 2 contingency table was also done.
Selective coronary arteriography was performed by the
percutaneous femoral technique or using the brachial approach. Significant coronary artery disease was considered
present if there was luminal narrowing of 60% or greater in
at least one major vessel. Left main coronary artery stenosis
was considered functionally equivalent to double vessel disease. Significant lesions were further subdivided into
moderate (60-80%), severe (80-99%), and occluded
categories (100%). Collateral circulation was defined as
present if an obstructed artery was opacified by antegrade
flow via bridge vessels or through the conus branch, and
when filling of distal vessels was noted via retrograde flow
through distal anastomosis. The RAO projection for selective left ventriculography was utilized to detect contraction
abnormalities as well as to determine the ejection fraction
according to the modified method of Greene,15 utilizing the
formula
Methods
One hundred fifty patients referred to Rush-Presbyterian
St. Luke's Medical Center for evaluation of chest pain syndromes during the period of January 1974 to January 1975
were retrospectively evaluated. The patients were chosen
from a consecutive group after the exclusion of valvular or
congenital heart disease, hypertension, previous cardiac surgery, left ventricular hypertrophy, bundle branch block, or
inotropic drugs. Each patient underwent clinical evaluation,
graded multistage submaximal exercise testing, selective
coronary arteriography and left ventriculography. The data
were analyzed for sex, age, presence of pathologic Q waves
on electrocardiogram (ECG), exercise-induced angina, and
ST-segment changes on the treadmill; degree of coronary
artery disease, presence of collateral circulation, left ventricular wall motion, LV end-diastolic pressure and ejection
fraction. Both treadmill and angiographic data were
evaluated by two independent observers.
The patients were exercised according to an uninterrupted protocol at I mph at 0% inclination for 1 min, then at
3 'mph at 4, 8, and 12% inclination for 3 min each. Leads I,
From the Section of Cardiology, Department of Medicine, and the Department of Preventive Medicine, Rush Medical College, Chicago, Illinois.
Presented in part at the 48th Annual Meeting of the American Heart
Association, November 1975, Anaheim, California.
Address for reprints: Neil Kramer, M.D., Director of Non-Invasive
Laboratory, Division of Adult Cardiology, Cook County Hospital, 1835
West Harrison Street, Chicago, Illinois 60612.
Received July 12, 1977; revision accepted November 21, 1977.
EF=-
EDV - ESV X 100
EDV
Pressures were measured with Statham P23db strain gauges.
763
VOL 57, No 4, APRIL 1978
CIRCULATION
764
Results
General Characteristics
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
There were 150 consecutive patients evaluated for the
study. From this group 115 patients (77%) were selected for
having attained target heart rate or symptomatic end point
and had interpretable treadmill tests. These comprise the actual study population. The age range for the 87 males was
32-68 years (mean age 47) and for the 28 females it was
29-64 (mean age 48). Their resting ECGs revealed: a normal
pattern in 44 patients, nonspecific changes in 43 patients,
pathologic Q waves in 22 patients (anterior 9, inferior 13)
and minor conduction disturbances in 6 patients. There were
53 patients without significant coronary artery disease, 18
patients with single vessel, 28 patients with double vessel and
16 patients with triple vessel disease. Of the 35 excluded patients (see Methods) the incidence of single, double and triple vessel disease was identical to that of the study population. Ventriculographic contraction abnormalities were
noted in 25 (40%) of the patients with significant coronary
artery disease. These occurred in four patients (22%) with
single vessel, 13 patients, (46%) with double vessel and eight
patients (50%) with triple vessel disease.
Treadmill Test Correlations with Coronary Angiography
The overall treadmill sensitivity
true positive
true positive + false negative
for the study
was
triple vessel disease, the frequency of false negative exercise
nearly identical 22%, 21%, and 19% respectively.
The distribution of treadmill responses as correlated with
ventriculography is shown in figure 1. Of the 56 negative exercise responses there were 13 patients with false negative
tests (12 males, 1 female) and 12 of them (92%)
demonstrated ventriculographic motion disorders. Of the 59
positive ECG exercise responses there were 49 patients with
true positive tests and 13 of them (26%) showed similar ventriculographic abnormalities (P < 0.0001). There were no
significant differences in the incidence of single, double, or
triple vessel disease in the false negative and true positive
groups. None of the ten patients in the false positive group (8
females, 2 males) demonstrated ventriculographic motion
disorders.
tests were
In the 62 patients with significant coronary artery disease
the presence of either an abnormal ventriculogram (25 patients) or an electrocardiographic pattern of infarction (22
patients) were associated with positive exercise ischemic
response rates of 52 and 59% respectively. Contrariwise
either a normal ventriculogram (37 patients) or the lack of
an electrocardiographic pattern of infarction (40 patients)
was associated with positive exercise response rates of 97
and 90% respectively.
Determinants of Treadmill Test Responses
X
100)
The false negative and true positive groups were next
analyzed in detail in order to ascertain the relative influence
of vascular and left ventricular determinants on the ischemic
response to exercise.
79% and the overall specificity
true negative
true negative + false positive
-
x
100
/
81%. The rate of false positive responses was 19% and
the rate of false negative responses was 21%.
Table 1 relates the maximal ST-segment exercise
responses with the extent of significant coronary artery disease in the 115 patients studied. As shown, progressive
ST-segment change (1 mm 2 mm) was associated with a
greater occurrence of combined double and triple vessel disease. No relationship existed between the maximal ST-segment response and the presence of collateral circulation. The
distribution of single, double, and triple vessel disease was
similar in the false negative group (first column). Additionally, among those patients with either single, double or
was
-
Electrocardiographic Patterns
Diagnostic resting electrocardiographic patterns of infarction were observed in nine patients (69%) of the false
negative group and in 13 patients (27%) of the true positive
group (P < 0.01, fig. 2). The sites of infarction were similar
in both groups, with inferior wall patterns noted in 44% of
the false negative group and 69% of the true positive group
(NS). Anterior infarction patterns were present in 56% and
31% of the two respective groups (NS). The incidence of
nonspecific ST wave abnormalities was 15% in the false
negative and 28% in the true positive group (NS, fig. 2).
During exercise a mean maximal heart rate of 158
beats/min (140-175) was attained in the false negative group
and 141 beats/min (85-210) in the true positive group.
TABLE 1. Correlation of Maximal ST-Segment Change with the Arteriographic Extent of Significant Coronary
Artery Disease (115 patients)
ST-Segment Response
No. vessels involved
with SCAD
0
1
2
3
Total
Negative
43 (77%)
4 (7%)
6 (11%)
3 (5%)
56 (100%)
Positive
1
mm
4 (27%)
5 (33%)
4 (27%)
2 (13%)
15 (100%)
1.5
2 mm
mm
1 (9%)
4 (37%)
3 (27%)
3 (27%)
11 (100%)
(25%)
(15%)
(45%)
(15%)
20 (100%)
5
3
9
3
>2
mm
0
2 (15%)
6 (46%)
5 (39%)
13
(Total 59 pt)
(100%)
SCAD
=
significant coronary artery disease.
765
FALSE NEGATIVE TESTS AND LV DYSFUNCTION/Kramer, Susmano, Shekelle
TREADMILL G.X.T. ANCIOGRAPIIC RESPONSES
115 PATIENTS
POSITIVE G.XT.
NEGATIVE G.X.T.
56
iRJ NiATIVE
L.V.
N L.
39
59
rFM.SE NBSATIVE
NO CAD
SCAD
43
13
ABN.
AN
4
|
POSITIVE
NO CAD
10
ANN.
I'
Tl.
IFALSE POSI
(2
NL.
10
FIGURE 1. Distribution of treadmill responses
as correlated with angiography in 115 patients,
G.X. T. = graded exercise test; SCAD = significant coronary artery disease; L V = left ventriculography; NL = normal; ABN = abnormal.
SCAD
49
ANN.
°
ANN.
NL.
36
13
(26S)
(92%)
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
Although there was not a significant difference, patients in
the false negative group experienced exercise-induced pain
less often as compared to the true positive group. The mean
duration of exercise prior to discontinuing the test was 5.7
min and 3.1 min respectively.
Significant Coronary Artery Disease
There were no significant differences in specific vessel involvement in either the false negative or true positive groups
(fig. 3). The left anterior descending artery was the most frequently involved vessel in both groups (85% vs 82%)
followed by the right coronary artery (62 vs 65%). The left
circumflex artery was involved in 46% of the patients with
false negative responses and in 49% of the patients with true
positive responses. Left main artery disease was not
observed in the false negative group and seen only in 4% of
the true positive group. Similarly, no significant differences
were observed in the frequency of single, double, and triple
vessel disease (fig. 3) nor in the severity of vascular involvement (moderate, severe or occluded categories, fig. 3). In the
false negative group there were eight occluded vessels occurring in as many patients. Seven of these vessels (88%)
supplied abnormally contracting ventricular segments. In
the true positive group, 21 occluded vessels occurred in 18
-patients. Eight of these occluded vessels (38%) supplied abnormal ventricular segments.
Table 2 presents the distribution and location of wall motion abnormalities as they occurred in the false negative (12
of 13 patients) and true positive groups (13 of 49 patients).
As shown on the left panel, dyskinetic segments were
significantly more prevalent in the false negative group while
lesser motion disorders were more often observed in the true
positive group. There was no significant difference in the
specific location of these segmental abnormalities in the two
groups.
An abnormally high left ventricular end-diastolic pressure (> 15 mm Hg) was noted in five patients (42%) in the
false negative group and in four patients (9%) of the true
positive group (P < 0.01, fig. 4). The left ventricular enddiastolic pressure was not determined in five patients (1 false
negative, 4 true positive). All of these patients demonstrated
normal ventriculographic patterns.
The ejection fraction was determined in 12 of 13 patients
in the false negative group and 41 of 49 patients in the true
positive (fig. 4). In the former group, the ejection fraction
was abnormal in five patients (42%) and in the latter group it
was abnormal in three patients (7%, P < 0.01). The ejection
fraction could not be determined in nine patients (I false
negative, 8 true positive) as a result of catheter-induced arrhythmias or unavailability of material for review. However,
EKG PATTERNS
Collateral Circulation
The presence of collateral circulation was uniformly
associated with severely obstructed or occluded vessels in
both the false negative and true positive groups. Eight patients in the false negative group (62%) demonstrated collateral circulation while 20 patients in the true positive group
(41%) showed similar findings. This difference was not
significant. Of the patients with left ventricular dysfunction
the incidence of collateral circulation was nearly identical
for both groups (58% vs 62%).
Left Ventricular Performance
The incidence of ventriculographic wall motion disorders
was 92% in the false negative group and 27% in the true
positive group (P < 0.0001, fig. 4). When corrected for the
extent of vascular involvement (single, double or triple vessel
disease) ventriculographic abnormalities were likewise more
prevalent in the false negative group at each level of vascular
disease.
FALSE
TRUE
_
-
+
L
13
49
P < . 01
-C
Co
L6
HX .1'
N.S.
NORMAL
MI
NSSTWA
FIGURE 2. Electrocardiographic responses in false negative and
true positive groups. F- = false negative group; T+ = true
positive group; MI = myocardial infarction; NSSTWA = nonspecific ST-T wave abnormalities; NS = not significant.
CIRCULATION
766
VOL 57, No 4, APRIL 1978
SIGNIFICANT CORONARY ARTERY DISEASE
FALSE ®3
TRUE ®
_
13
100
=
49
90
80
70
60
N S.
50
c:
-40
30
FIGURE 3. Coronary angiographic findings in
the falve negative and true positive groups. LMA,
LAD, LCA, RCA = left main, anterior descending, circumflex and right coronary arteries,
respectively; MOD = moderate; SEV= severe;
OCC= occluded; VD = vessel disease.
20
10
IYD 2VD 3VD
SEVERITY'
DISTRIBUTION
EXTENT
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
the ventriculograms of all nine patients were reported as
normal. An inverse relationship was observed between the
ejection fraction and the occurrence of false negative
responses. Thus, in the groups with low ( < 55%), normal
(55% - 75%), and high (> 75%) ejection fraction, the incidence of false negative responses was 63%, 37% and 0%
respectively (P < 0.001, fig. 5).
Discussion
A "false negative" electrocardiographic response to exercise may theoretically arise when myocardial ischemia is
prevented, masked, mild in degree, or absent entirely. Thus,
certain drugs (eg., nitroglycerin or propranolol) may prevent
ischemia while certain conduction disturbances (bundle
branch block) may mask its presentation on the surface electrocardiogram. The observation that false negative
responses frequently occur in association with single vessel
LEFT VENTRICULAR FUNCTION
100
FALSE
P < .0001
90
-
TRUE +
m
80
vq
70
-cc
60
50
P < 0.01
P < 0.01
40
c>:
30
20
10
in]
-ABN L.V.
ANGIO
NO.
13
49
ABN
LVEDP
ABN
E .F
12
41
12
45
FIGURE 4. Parameters of left ventricular function in true positive
and false negative groups. EF = ejection fraction; L VEDP = left
ventricular end-diastolic pressure; N = number of patients.
involvement1"8 has been taken to indicate that a milder and
perhaps undetectable degree of ischemia exists. Likewise, it
has been suggested that a negative exercise test can be
demonstrated in patients with left ventricular dysfunction,`0-`2 usually resulting from a prior myocardial infarction. The implication here is that when occluded vessels
supply infarcted tissue a nonischemic response is identified.
In order to clarify this issue further, we retrospectively
evaluated the relationship of several parameters as determinants of the false negative response to exercise.
Analysis of our data in general showed an overall sensitivity and specificity in keeping with the results of prior
studies in symptomatic patients.4-7 As has been reported,3 6, 16
the magnitude of ST-segment change in this study was
related to more extensive vascular disease (table 1) and unrelated to the presence of collateral circulation. In those patients with significant vascular disease and no ST segmental
changes on exercise (false negative group), however a spectrum of vascular involvement rather than single vessel disease was noted. Similarly, the frequency of false negative
tests in patients with either single, double or triple vessel disease was nearly identical (22, 21, and 19%, respectively).
A detailed comparative analysis of the false negative and
true positive groups indicated that no significant differences
existed in coronary vascular disease. Specifically, the distribution, severity and extent of arterial disease were nearly
identical in both groups. Accordingly, there was no specific
vessel or number of vessels involved which were apt to
produce either a negative or positive response to exercise;
nor was the presence of collateral circulation a major determinant of the electrocardiographic responses to exercise.
In this study the sole difference observed between the false
negative and true positive groups was the extraordinary incidence of myocardial dysfunction associated with infarction patterns in the former group. Angiographically, motion disorders occurred much more frequently (figs. 1 and 4)
and were qualitatively more severe (table 2) in the false
negative responders. Both an increased left ventricular end
diastolic pressure and a reduced ejection fraction
characterized the hemodynamics in the false negative group.
An inverse relationship between the ejection fraction and
frequency of false responses (fig. 5) was also noted.
Early in the development of exercise testing Fitzgibbon'2
FALSE NEGATIVE TESTS AND LV DYSFUNCTION/Kramer, Susmano, Shekelle
767
TABLE 2. Distribution of Ventriculographic Abnormalities in the False Negative and True Positive Groups
Severity of
Left Ventricular (LV) Dysfunction
Dyskinesia
False (12)
Negative
True (13)
Positive
5
(42%)
1
(8%)
P = 0.001
Location of
Abnormal (LV) Wall Motion
Ant/Apical
5
(42%)
6
A-Hypokinesia
7
(58%)
12
(92%)
P <0.05
(46%)
NS
Posterior
Inferior
Diffuse
6
(50%)
5
(39%)
NS
1
(8%)
2
(15%)
NS
Abbreviations: A = akinesia; Ant = anterior; NS = not significant.
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
noted that "positive step tests in subjects with coronary
artery disease are commoner in those not sustaining
myocardial infarction." Kattus'0 subsequently suggested
that the vast majority of false negative responses occur in
patients who have survived myocardial infarction. The Seattle Heart Watch project17 extended this concept by noting a
20% greater frequency of exercise-induced ST-segment
deviations in their "angina only" subgroup, in comparison
to their postinfarction group. These results appear comparable to ours in which a 31% greater frequency of exercise
test positivity was observed in the noninfarction group.
An association between the false negative test and left
ventricular hemodynamics was described by Murray et al.11
Thus, this type of response was noted to occur commonly in
association with 1) a raised left ventricular end-diastolic
pressure, 2) a larger left ventricular dimension, 3) a reduced
ejection fraction, and 4) the presence of wall motion abnormalities.'8 The above findings concur with our overall
experience. Bruce'9 postulated that the spectrum of motion
disorders (akinesis, dyskinesis) resulting from extensive infarction could alter the ST-segment responses upward in a
manner that would cancel the expected ischemic depression.
The net result would then be that "false negative ST
responses are observed with other manifestations of severe
impairment of cardiac function." In this regard five of six
patients (83%) with the most serious motion abnormalities
in our study appeared in the false negative group (table 2).
To what then may we attribute the false negative response
to exercise? From our data it would appear that similar if
not identical vascular anatomy can elicit either response.
Left ventricular dysfunction, however, occurred
predominantly, although not exclusively, in the false
negative group. On a metabolic basis, the above may be interpreted to mean that the postmyocardial infarction state
possesses less potentially ischemic myocardium which is
available to participate in the response to exercise. Accordingly, lactate studies performed in similar patients with combined extensive vascular and myocardial disease have shown
the absence of lactate production under conditions of
stress.'0 It is conceivable that the lower incidence of exerciseinduced pain, enhanced chronotropic response, and increased duration of exercise, as noted in our false negative
group, was caused by a similar underlying mechanism.
Occluded vessels supplying abnormal Ventricular myocardial segments appeared much more frequently in the false
negative than in the true positive responders (88% vs 38%), a
finding which supports our hypothesis. Unfortunately,
because lactate, oxygen utilization, and work determinants
were not performed in this retrospective study, such a conclusion must await definitive evaluation.
While our data suggest that extensive myocardial disease
is contributing to the lack of ST-segment response, other
reports have not noted this association."s 3 6, The explanation for these observed differences may reflect patient selection, varying criteria for significant coronary disease, or
variations in exercise testing methodology. It is also possible
that had exercise been performed at the time of catheterization and angiography, differences between groups in
myocardial performances would not have appeared.
Although the purpose of this report was to investigate the
pathophysiology of the false negative exercise response, certain clinical points deserve mention. It should be apparent
from our data that the surface resting electrocardiogram, by
demonstrating an infarction pattern, distinguished the false
negative from true negative responders in the majority of instances. Thus, a negative stress test in this context would not
negate the recognition of significant coronary artery disease
especially in symptomatic patients. However, such a
response may provide additional information regarding the
status of left ventricular performance.
100
90
z
80-
0
C.
LU70
IX70 o
60-
,
50 -
-J_
coo
P < . 001
40
La
<
30
La
X
20-
10
<55
55-74
>7 5
NO.
8
19
26
FIGURE 5. Incidence of false negative response (%) in patients
with SCAD according to their ejection fraction (< 55% = low,
55-74% = normal, and > 75% = high).
CIRCULATION
768
Therapeutic implications of our study suggest that patients in the false negative subset may respond to a lower
dose of antianginal therapy and demonstrate an improved
prognosis as suggested by their better exercise tolerance.
Their poor left ventricular function, however, may be an
offsetting factor both medically and surgically. Obviously,
such conclusions must await definitive prospective studies.
In conclusion, we have evaluated the determinants of the
false negative exercise test in symptomatic patients. A
strong association of this response with left ventricular dysfunction rather than a lesser extent of vascular disease was
observed. The mechanisms involved may relate to complex
metabolic and electrocardiographic relationships resulting
from extensive segmental disease.
Our data suggest that in the clinical setting of symptomatic coronary artery disease a negative electrocardiographic exercise response may not always exclude extensive coronary disease and may signal the presence of
associated moderately severe myocardial dysfunction.
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
Acknowledgment
We would like to express our gratitude to Dr. M.S. Rosenberg for his help
in the interpretation of electrocardiographic abnormalities and to Mrs. Susan
Burin, Ms. Susan Adams, and Ms. Patricia Houston for their technical and
secretarial assistance.
References
Tonkon MJ, Miller RR, DeMaria AN, Vismara LA, Amsterdam EA,
Mason DT: Multifactor evaluation of the determinants of ischemic electrocardiographic response to maximal treadmill testing in coronary disease. Am J Med 62: 339, 1977
2. Schweitzer P, Jelinek VM, Herman MV, Gorlin R: Comparison of the
two-step and maximal exercise tests in patients with coronary artery disease. Am J Cardiol 33: 797, 1974
3. Bartel AG, Behar VS, Peter RH, Orgain ES, Kong Y: Graded exercise
stress tests in angiographically documented coronary artery disease. Circulation 49: 348, 1974
1.
VOL 57, No 4, APRIL 1978
4. Rios JC, Hurwitz LE: Electrocardiographic responses to atrial pacing
and multistage treadmill exercise testing: Correlation with coronary
arteriography. Am J Cardiol 34: 661, 1974
5. Kaplan MA, Harris CN, Aronow WS, Parker DP, Ellestad MH:
Inability of the submaximal treadmill stress test to predict the location of
coronary disease. Circulation 47: 250, 1973
6. Martin CM, McConahay DR: Maximal treadmill exercise electrocardiography: Correlations with coronary arteriography and cardiac
hemodynamics. Circulation 46: 956, 1972
7. McHenry PL, Phillips JF, Knoebel SB: Correlation of computer treadmill exercise electrocardiogram with arteriographic location of coronary
artery disease. Am J Cardiol 30: 747, 1972
8. Goldschlager N, Selzer A, Cohn K: Treadmill stress tests as indicators of
presence and severity of coronary artery disease. Ann Intern Med 85:
277, 1976
9. Goldschlager N, Sakai FJ, Cohn KE, Selzer A: Hemodynamic abnormalities in patients with coronary artery disease and their relationship to
intermittent ischemic episodes. Am Heart J 80: 610, 1970
10. Kattus AA: Exercise electrocardiography: Recognition of the ischemic
response, false positive and negative patterns. Am J Cardiol 33: 721, 1974
11. Murray JA, Hamilton G, Kennedy JW, Bruce RA: Disparities between
coronary arteriography, resting left ventricular function and maximal exercise performance in ischemic heart disease patients. Circulation (suppl
II) 44: 11-204, 1971
12. Fitzgibbon GM, Burggraf GW, Groves TD, Parker JO: A double
master's two-step test: Clinical, angiographic and hemodynamic correlations. Ann Intern Med 74: 509, 1971
13. Kansal S, Roitman D, Sheffield LT: Stress testing with ST-segment
depression at rest: An angiographic correlation. Circulation 54: 636, 1976
14. Linhart JW, Turnoff HB: Maximal treadmill exercise test in patients with
abnormal control electrocardiograms. Circulation 49: 667, 1974
15. Greene DG, Carlisle R, Grant C, Bunnell IL: Estimation of left ventricular volume by one-plane cineangiography. Circulation 35: 61, 1967
16. Helfant RH, Vokonas PS, Corlin R: Functional importance of the
human coronary collateral circulation. N Engl J Med 284: 1277, 1971
17. Bruce RA, Gey GO, Cooper MN, Fisher LD, Peterson DR: Seattle heart
watch: Initial clinical, circulatory and electrocardiographic responses to
maximal exercise. Am J Cardiol 33: 459, 1974
18. Bruce RA: Exercise electrocardiography. In The Heart, ed 3. Edited by
Hurst JW. New York, McGraw, Hill, 1974, pp 315
19. Bruce RA, Cooper MN, Gey GO, Fisher LD, Peterson DR: Variations in
responses to maximal exercise in health and in cardiovascular disease:
Initial findings of Seattle heart watch. Angiology 24: 691, 1973
20. Cohen LS, Elliott WC, Klein MD, Gorlin R: Coronary heart disease:
Clinical, cinearteriographic and metabolic correlations. Am J Cardiol 17:
153, 1966
The "false negative" treadmill exercise test and left ventricular dysfunction.
N Kramer, A Susmano and R B Shekelle
Downloaded from http://circ.ahajournals.org/ by guest on April 29, 2017
Circulation. 1978;57:763-768
doi: 10.1161/01.CIR.57.4.763
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1978 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/57/4/763
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/