The accuracy of the electrocardiogram during exercise stress test based on heart size.
ABSTRACT Multiple studies have shown that the exercise electrocardiogram (ECG) is less accurate for predicting ischemia, especially in women, and there is additional evidence to suggest that heart size may affect its diagnostic accuracy.
The purpose of this investigation was to assess the diagnostic accuracy of the exercise ECG based on heart size.
We evaluated 1,011 consecutive patients who were referred for an exercise nuclear stress test. Patients were divided into two groups: small heart size defined as left ventricular end diastolic volume (LVEDV) <65 mL (Group A) and normal heart size defined as LVEDV ≥65 mL (Group B) and associations between ECG outcome (false positive vs. no false positive) and heart size (small vs. normal) were analyzed using the Chi square test for independence, with a Yates continuity correction. LVEDV calculations were performed via a computer-processing algorithm. SPECT myocardial perfusion imaging was used as the gold standard for the presence of coronary artery disease (CAD).
Small heart size was found in 142 patients, 123 female and 19 male patients. There was a significant association between ECG outcome and heart size (χ(2) = 4.7, p = 0.03), where smaller hearts were associated with a significantly greater number of false positives.
This study suggests a possible explanation for the poor diagnostic accuracy of exercise stress testing, especially in women, as the overwhelming majority of patients with small heart size were women.
- [Show abstract] [Hide abstract]
ABSTRACT: Significant gender differences have been found in performance of the exercise ECG for the identification of coronary artery disease. However, identical exercise ECG ST segment criteria have been used in men and women, which might contribute to the lower accuracy of these methods in women than in men. To assess the effect of gender-specific test partitions on relative performance of standard and heart rate-adjusted ST segment depression criteria in men and women, the exercise ECGs of 143 women and 477 men were examined. Non-gender-specific test partitions, selected to have matched specificities of 96% for each test method, were determined in all 283 normal subjects, and gender-specific test partitions with identical specificity were determined separately in the 52 normal women and 231 normal men; sensitivity of these criteria was then examined in the 91 women and 246 men with coronary disease. Standard ST segment depression criteria (0.1 mV of additional horizontal or downsloping ST segment depression at end exercise) with identical 96% specificity in the entire group of normal subjects and separately in women and men had a significantly lower sensitivity of 51% in women compared with 67% in men (P < .01). Among women, performance of the ST segment/heart rate (ST/HR) slope was more improved than that of the ST/HR index by the use of gender-specific criteria. Compared with the performance of non-gender-specific criteria, application of gender-specific ST/HR slope partitions with matched specificity of 96% resulted in a significant increase in sensitivity in women from 84% to 91% (P < .01), with no significant change in sensitivity in men (89% to 88%) and with no residual difference in sensitivity between men and women. Although the use of gender-specific ST/HR slope criteria significantly improved sensitivity in both men and women with respect to standard criteria (each P < .0001), the relative increase in sensitivity provided by heart rate adjustment was significantly greater in women than in men (40% versus 21%, P < .001). Similar gender differences in improvement in performance using gender-specific criteria for the ST/HR slope were observed when analysis of test performance was restricted to the detection of three-vessel coronary disease (50% versus 9%, P = .002). At high specificity, gender-specific test partitions improve sensitivity of the ST/HR slope for the identification of coronary disease in women, with no decrease in sensitivity in men. In contrast, gender-specific partitions do not change performance of standard test criteria, which is lower in women than in men. Accordingly, the relative benefit of heart rate adjustment by the ST/HR slope method is greater in women than in men. These findings support use of the ST/HR slope with use of gender-specific partitions for the identification and quantification of coronary artery disease in both men and women.Circulation 09/1995; 92(5):1209-16. · 14.95 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: This study was designed to evaluate the incremental prognostic value over clinical and exercise variables of rest thallium-201/exercise technetium-99m sestamibi single-photon emission computed tomography (SPECT) in women compared with men and to determine whether this test can be used to effectively risk stratify patients of both genders. To minimize the previously described gender-related bias in the evaluation of coronary artery disease in women, there is a need to identify a noninvasive testing strategy that is able to accurately and effectively risk stratify women. We identified 4,136 consecutive patients (2,742 men, 1,394 women) who underwent dual-isotope SPECT. The incremental value of nuclear testing was determined using both a stepwise Cox proportional hazards model and Kaplan-Meier survival analysis. Receiver operating characteristic curve analysis was performed to determine test discrimination for high risk patients in men and women. The patient population was followed up for 20 +/- 5 months for events (cardiac death or nonfatal myocardial infarction). During this time, 63 myocardial infarctions and 32 cardiac deaths occurred in the men, and 31 myocardial infarctions and 14 cardiac deaths occurred in the women. Nuclear testing significantly stratified both men and women irrespective of their rest electrocardiogram. Cox proportional hazards analysis revealed that nuclear testing added incremental prognostic value in both men and women after inclusion of the most predictive clinical exercise variables (overall chi-square 89 in men vs. 120 in women, p < 0.005). Kaplan-Meier survival analysis demonstrated that nuclear testing further stratified men and women with both intermediate to high and low prescan likelihoods of coronary artery disease (p < 0.005 for all). Receiver operating characteristic curve analysis demonstrated superior discrimination for the nuclear scan results in identifying high risk women than men (area under the curve: 0.84 +/- 0.03 vs 0.71 +/- 0.03 in men, p < 0005). The odds ratio comparing event rates in patients with women than in men, suggesting superior stratification using nuclear testing in women. Dual-isotope myocardial perfusion imaging yields incremental prognostic value in both men and women. This modality identifies low risk women and men equally well but relatively high risk women more accurately than relatively high risk men and, thus, is able to stratify women more effectively than men.Journal of the American College of Cardiology 08/1996; 28(1):34-44. · 15.34 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: We attempted to formally compare the diagnostic accuracy of thallium-201 single-photon emission computed tomographic (SPECT) myocardial perfusion imaging in men and women and the effect of chamber size on accuracy. The diagnostic accuracy of conventional exercise testing has been shown to be lower in women. Less is known about the relative accuracy of perfusion imaging. Because of smaller body size, women have a smaller heart size than men, a factor that may reduce accuracy. We identified 323 patients undergoing thallium-201 SPECT myocardial perfusion imaging who either had < 5% probability of coronary artery disease (CAD) by Bayesian analysis or who underwent cardiac catheterization within 60 days of stress testing. Patients with documented history of infarction, coronary artery bypass grafting, pathologic Q waves on the electrocardiogram, left bundle branch block or nonischemic cardiomyopathy were not included. We performed strict quantitative analysis, and receiver operating characteristic (ROC) curves were generated and the area under the curve was calculated for men and women. A size index was generated from the number of short-axis slices and average radius of each slice, and the group was classified as having a large or a small chamber size. The ROC areas of men and women with a large and a small chamber size were then compared. Diagnostic accuracy was lower in women than in men (ROC are 0.82 vs. 0.93, p < 0.05) despite similar values for peak heart rate and rate-pressure product and similar severity of CAD. There was a greater difference in accuracy between patients with a large versus a small chamber size (ROC area 0.94 vs. 0.73, p < 0.01) despite similar levels of exercise and severity of CAD. When we compared men and women in groups stratified by chamber size, we could not detect a significant difference between ROC area values of men and women (large: 0.94 men, 0.93 women, p = 0.77, power to detect difference in area of 0.15 = 91%; small: 0.79 men, 0.72 women, p = 0.58, power to detect difference in area of 0.15 = 35%). The diagnostic accuracy of thallium SPECT myocardial perfusion imaging is lower in women than in men. Most of the difference appears to be due to smaller left ventricular chamber size in women, although a small residual gender effect in smaller heart sizes cannot be entirely excluded. It is proposed that the most likely cause for this difference is the relatively greater effect of imaging blurring on smaller hearts.Journal of the American College of Cardiology 12/1996; 28(5):1214-9. · 15.34 Impact Factor
The Accuracy of the Electrocardiogram during Exercise
Stress Test Based on Heart Size
Jason C. Siegler1, Shafiq Rehman2, Geetha P. Bhumireddy2, Raushan Abdula2, Igor Klem5, Sorin J.
Brener2, Leonard Lee4, Christopher C. Dunbar2,3, Barry Saul2, Terrence J. Sacchi2, John F. Heitner2*
1Department of Sport, Health & Exercise Science, University of Hull, Hull, United Kingdom, 2Division Cardiology, New York Methodist Hospital, Brooklyn, New York,
United States of America, 3Brooklyn College of the City University of New York, New York, New York, United States of America, 4Division of Cardiothoracic Surgery, New
York Methodist Hospital, Brooklyn, New York, United States of America, 5Division of Cardiology, Duke Medical Center, Durham, North Carolina, United States of America
Background: Multiple studies have shown that the exercise electrocardiogram (ECG) is less accurate for predicting ischemia,
especially in women, and there is additional evidence to suggest that heart size may affect its diagnostic accuracy.
Hypothesis: The purpose of this investigation was to assess the diagnostic accuracy of the exercise ECG based on heart size.
Methods: We evaluated 1,011 consecutive patients who were referred for an exercise nuclear stress test. Patients were
divided into two groups: small heart size defined as left ventricular end diastolic volume (LVEDV) ,65 mL (Group A) and
normal heart size defined as LVEDV $65 mL (Group B) and associations between ECG outcome (false positive vs. no false
positive) and heart size (small vs. normal) were analyzed using the Chi square test for independence, with a Yates continuity
correction. LVEDV calculations were performed via a computer-processing algorithm. SPECT myocardial perfusion imaging
was used as the gold standard for the presence of coronary artery disease (CAD).
Results: Small heart size was found in 142 patients, 123 female and 19 male patients. There was a significant association
between ECG outcome and heart size (x2=4.7, p=0.03), where smaller hearts were associated with a significantly greater
number of false positives.
Conclusions: This study suggests a possible explanation for the poor diagnostic accuracy of exercise stress testing,
especially in women, as the overwhelming majority of patients with small heart size were women.
Citation: Siegler JC, Rehman S, Bhumireddy GP, Abdula R, Klem I, et al. (2011) The Accuracy of the Electrocardiogram during Exercise Stress Test Based on Heart
Size. PLoS ONE 6(8): e23044. doi:10.1371/journal.pone.0023044
Editor: Conrad P. Earnest, Pennington Biomedical Research Center, United States of America
Received April 1, 2011; Accepted July 5, 2011; Published August 17, 2011
Copyright: ? 2011 Siegler et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: No current external funding sources for this study.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
Cardiovascular disease is the leading cause of death in the
United States claiming the lives of over 230,000 individuals each
year . Although the mortality from cardiovascular disease has
recently been decreasing in men, it has been increasing for women
. Inaccurate diagnosis and ensuing management inefficiencies
may contribute to the increased mortality .
Numerous studies have indicated that noninvasive cardiac stress
tests have a lower diagnostic accuracy in women [2,3]. In addition,
diagnostic accuracy in women also varies depending on the test
administered (i.e. stress echocardiography, stress myocardial
perfusion imaging, pharmacologic or exercise electrocardiogram)
[4–6]. Although sensitivity and specificity vary greatly between
studies, as reported values depend widely upon patient selection
criteria and methodological construct, studies using cross-gender
comparisons consistently report lower diagnostic accuracy in
female populations [4,6–9]. The lower accuracy has been
attributed to lower ECG voltage, smaller size of the coronary
vessels, smaller heart size, hormonal factors (premenopausal
relationship with endogenous estrogen levels), breast attenuation
and functional impairment [2,5,8].
Specific to ECG diagnosis and ischemia, reports have indicated
a higher number of false positive results in female patients when
compared to male patients [6–8]. The purpose of this study was to
evaluate and compare the diagnostic accuracy of the ECG during
exercise stress test based on left ventricular (LV) cavity size.
The study included 1,011 consecutive male and female patients
who were referred for an exercise nuclear stress test at New York
Methodist Hospital. All patients were referred for evaluation of
chest pain (CP), dyspnea or other associated risk factors for CAD
by their primary care physician or cardiologist. Patients with
resting ECG’s unsuitable for stress interpretation were excluded
(pathologic Q waves, left bundle branch block (LBBB), left
ventricular hypertrophy with strain pattern (LVH), Wolff-Parkin-
son-White (WPW) syndrome, or other significant ($1 mm)
PLoS ONE | www.plosone.org1 August 2011 | Volume 6 | Issue 8 | e23044
downward displacement of the ST segment) . Additionally,
patients who did not reach a minimum of 85% predicted maximal
heart rate during the exercise stress test were excluded from data
analysis to standardize and ensure sufficient myocardial stress.
All patients underwent a thorough history and physical exam
with data collected on presenting symptoms, past medical history,
cardiac risk factors, as well as medications. The baseline ECG’s
were analyzed by a certified exercise physiologist prior to
undergoing the stress test. All patients were instructed to hold
their beta-blockers and calcium channel blockers for 24 hours
prior to the stress test.
The study was approved by the Institutional Review Board
(IRB) of New York Methodist hospital. An informed consent was
not required because the study data was obtained and analyzed
anonymously. The IRB of New York Methodist Hospital
specifically waived the need for consent.
SPECT201Tl and Tc-99m Imaging
The gold standard for the diagnosis of ischemia was myocardial
single photon emission computed tomography (SPECT) imaging.
Initial resting myocardial SPECT images were acquired while the
patient was in a supine position with shoulders flexed to 180u,
using a GE Millennium MyoSIGHT rotating gamma camera
(General Electric Company, Milwaukee, WI). Resting images were
performed after receiving thallium (201Tl) (predetermined using a
weight-based algorithm) for patients weighing less than 300 lbs
using the following parameters:201Tl dosing up to 4mcI; imaging
time: 18 min; 36 views/ 30 sec per view; matrix 64664; circular;
Collimator: LEHR. Patients weighing greater than 300 lbs
underwent resting images with technetium (Tc-99m), but followed
the same parameters with the exception of a dosing rate of
Stress imaging for patients less than 300 lbs was performed with
Tc-99m and used the following parameters: Tc-99m dosing range
30–40mcI; imaging time: 15 min; 36 views/25 sec per view;
matrix 64664; circular; 16 frames/RR interval. Again, the same
parameters were used for patients weighing over 300 lbs, only the
dosing ranged from 35–40mcI.
Treadmill Exercise Testing
All treadmill testing was conducted using either the Bruce or
Modified Bruce protocols (GE Medical Systems CASE Stress
System Version 5 with Series 2000 Marquette Treadmill, General
Electric Company, Milwaukee, WI). End points of exercise were
predetermined according to absolute and relative indications for
terminating exercise testing . One minute prior to peak
exercise, patients received Tc-99m (weight based) and were
imaged 15 to 20 min post exercise. All 12 leads of the standard
ECG were monitored and used for analysis. ST measurements
were assessed visually 80 msec post J-point during exercise and
recovery with the PR segments used as the baseline. The criteria
for determining a positive exercise ST-segment response were as
follows: $1.0 mm horizontal or downsloping depression 80 msec
post J-point for at least 3 consecutive beats .
Image Interpretation & Heart Size Criteria
The horizontal and vertical planes and the short-axis views were
reviewed to detect the presence of defects. A 17-segment semi-
quantitative method was utilized for visual interpretation of
perfusion defects . In addition, a semi-quantitative scoring
system using the five-point model was utilized to assess myocardial
perfusion (0–4: 0=normal perfusion; 1=mild reduction in counts;
2=moderate reduction in counts; 3=severe reduction in counts;
4=absent uptake) . An image was considered positive for
ischemia if there was $1 segment perfusion defect seen at stress
which was not seen at rest.
The left ventricular cavity size was determined by a computer
algorithm that assesses the left ventricular cavitary borders and
computes the size of the cavity in milliliters (mL). An end diastolic
cavitary size less than 65 mL was used as a cutoff for small vs.
normal cavitary size. This cut off was determined prior to
initiation of the study by doing a retrospective analysis of data
consisting of 200 consecutive patients to determine the smallest
(20%) ventricular cavitary size in our patient population. The first
200 patients were performed on the same clinical grounds as the
patients in our current study, using the new GE scanner and
computer database system. This population was selected as it best
estimated the population that we analyzed in the current study.
The diagnostic accuracy of the ECG with nuclear imaging as
the reference standard was determined by calculating sensitivity,
specificity, and predictive values . Associations between
demographic data and between ECG outcome (false positive
versus no false positive) and heart size (small versus normal) were
analyzed using the Chi square test for independence, with a Yates
continuity correction. A multivariate analysis was performed to
assess which pertinent variables (diabetes, age, body mass index
(BMI), hypertension, coronary artery disease, and gender) were
significantly associated with a false positive test. Both a C statistic
and Goodness of fit were performed to assess the validity of the
Two-tailed significance was accepted at p,0.05 and all
statistical analyses were conducted using SPSSH for Windows
software, release 17.0 (SPSS Inc., Chicago, IL).
A total of 1,011 consecutive patients were evaluated in the study
sample and baseline characteristics provided in Table 1. Of the
1,011 patients, 482 were females, 359 had normal heart size and
123 had small heart size. There were 529 men, 510 with normal
heart size and 19 with small heart size. The female patients
accounted for the majority of small heart patients (86% vs. 13%).
Patient’s presenting complaints were most commonly chest pain
(33.4%), angina (33.1%), dyspnea (28.3%) or palpitations (26.3%),
although no differences were determined based on heart size (chest
pain: x2=0.01, p=0.92; angina: x2=1.8, p=0.18; dyspnea:
x2=0.3, p=0.57; palpitations: x2=0.2, p=0.63). Interestingly,
there were a significantly higher percentage of smokers in the
normal heart size patients (x2=7.8, p,0.01). The most commonly
prescribed medications throughout the sample population includ-
ed ACE inhibitors (29.3%), lipid lowering agents (35.0%), aspirin
(35.1%), beta-blockers (28.8%) and digoxin (,1%). Again, no
significant difference in medications prescribed was observed
between small heart size and normal heart size patients (ACE
inhibitors: x2=0.01, p=0.93; lipid lowering agents: x2=0.02,
p=0.89; aspirin: x2=0.5, p=0.46; beta-blockers: x2=0.9,
p=0.33; digoxin: x2=0.5, p=0.48). Additionally, hypertension
(52.2%), dyslipidemia (54.4%) and diabetes (21.5%) were the most
commonly observed risk factors, and again, no difference was
noted based on heart size (hypertension: x2=0.1, p=0.74;
dyslipidemia: x2=3.7, p=0.06; diabetes: x2=1.0, p=0.32).
Corresponding with the higher incidence of smokers in the
Electrocardiogram during Exercise Stress Test
PLoS ONE | www.plosone.org2August 2011 | Volume 6 | Issue 8 | e23044
normal heart size population, there was a significantly higher
degree of lung disease in this group (x2=5.0, p,0.05). The female
patients with small heart size also were older (p,0.001, Table 1)
and had a lower BMI (p,0.01, Table 1) than those women with
normal heart size. During the treadmill stress test, an overwhelm-
ing majority of patients did not experience any chest pain or any
abnormal blood pressure responses (Table 2).
ECG diagnostic accuracy
All false positive ECG results for small heart size patients
(n=31) and normal heart size patients (n=91) and all false
negative results for small heart size (n=13) and normal heart size
(n=146) are depicted graphically in Figure S1. Twenty-nine
women with small heart size were determined to have false positive
ECG, 12 false negative, 16 true positive and 66 true negative
interpretations. Forty-one women with normal heart size were
determined to have false positive ECG, 42 false negative, 66 true
positive and 210 true negative interpretations. There was a
significant association between ECG outcome and heart size
(x2=4.7, p=0.03), where smaller hearts were associated with a
significantly greater number of false positives. On multivariate
analysis, diabetes was found to be significantly associated with a
false positive test (p=0.03). Both age and heart size had a trend
towards significance with p=0.05 and p=0.08, respectively. The
goodness of fit (P=0.08) indicated very poor accuracy of this
multivariate analysis and the C statistic (P=0.65) indicating only
Within the male patients, 19 were categorized with small hearts.
Only two of the 19 patients demonstrated false positive ECG
readings, while only one was false negative, three were true
positive and the remaining 13 were true negative for ischemia. For
the 510 males categorized as normal hearts, 50 were false positive,
104 false negative, 96 true positive and 260 true negative. The
sensitivity, specificity and predictive values of exercise ECG in the
total study population, female patients and male patients are
demonstrated in Table 3.
The unique finding of this study was that patients with smaller
LV chamber sizes were more likely to display a false positive ECG
result during a maximal exercise stress test than patients with a
normal chamber size. The overwhelming majority of patients with
small heart size were women.
Several studies have addressed the lack of diagnostic accuracy in
ECG stress testing in women [7,8,15,16]. Sensitivity and specificity
results from these studies range from 31–80% and 41–68%,
respectively [17,18]. The current hypothesis’ in the literature for
the lower diagnostic accuracy in women include: breast attenu-
ation, coronary vessel size, hormonal factors (pre-menopausal
estrogen or post menopausal hormone replacement therapy), and/
or a lowered exercise tolerance/capacity [7,8,15,16]. One
anatomical mechanism that has not been addressed in detail,
however, is the physical dimension of chamber size and the
potential relationship to stress ECG tracings. In the current study,
small chamber size (LVEDV,65 mL) was indicative of higher
rates of false positive ECG diagnosis in women.
To our knowledge, only Hansen and colleagues have addressed
diastolic cavitary size and the relationship between nuclear stress
testing . Others have found correlations between large body size
and reduced cardiac workload . However, this would not explain
the high rate of false positives observed in the current study as all of
the patients included in the analysis, regardless of heart size, attained
a high workload (as indicated by 85% of predicted HRmax).
There are other possible explanations as to why left
ventricular chamber size may affect the diagnostic accuracy of
the exercise ECG. Small heart size may have associated smaller
coronary arteries leading to diffuse subendocardial balanced
ischemia which may not be picked up on nuclear SPECT
imaging. This would imply that the gold standard is inaccurate
and not the exercise stress ECG itself. A third possible
explanation may be that patients with a small heart size need
to have a different lead placement than patients with normal
heart size as perhaps the position of the leads relative to the
myocardium can affect the ST segment. In addition, small heart
Table 1. All patient characteristics (n=1,011) including
medications, symptoms and risk factors prior that completed
the stress testing (treadmill) and categorized as small and
normal heart size.
Small Heart Patients
Normal Heart Patients
Total Number% of TotalTotal Number % of Total
ACE42 29.6 254 29.2
B Blockers 3625.4 25529.3
Clopidogrel 107.0 8810.1
Digoxin0 0.03 0.3
Diuretics 25 17.6 10612.2
Lipid Lower 4934.5 305 35.1
Nitrates4 2.837 4.3
Hypertension 7653.5 45252.0
Dyslipidemia 8761.3 46353.3
Diabetes 3524.6 18220.9
Syncope3 2.1 101.2
Lung disease 19 13.467 7.7
Renal disease1 0.730.3
PVD1 0.78 0.9
Arrhythmia2 1.4 182.1
Chest pain 4833.8 29033.4
ACE: angiotensin-converting enzyme inhibitors
B Blockers: Beta Blockers
Ca2+Channel: calcium channel inhibitors
Lipid Lower: Lipid lowering agents
PVD: peripheral vascular disease
BMI: Body mass index
Electrocardiogram during Exercise Stress Test
PLoS ONE | www.plosone.org3August 2011 | Volume 6 | Issue 8 | e23044
size changes the dynamics of the relationship between the R
wave and the ST segment which could potentially be another
mechanism for the increase rate of false positive ECG changes
in patients with small heart size.
In our study, we used automated gated SPECT for the
quantification of left ventricular (LV) volumes. Gated SPECT
has become the state-of-the-art technique for myocardial perfusion
imaging and it combines the evaluation of both myocardial
Table 2. Stress test results for all patients and categorized under small and normal heart size.
Small Heart (n=123)Normal Heart (n=359)
Stress Variables: Total Number% of Total Total Number % of Total
Normal BP response11492.7 32189.4
No CP 11593.5319 88.9
CP (atypical)2 1.64 1.1
CP (typical)3 2.412 3.3
CP (unspecified)3 2.4246.7
Small Heart (n=19) Normal Heart (n=510)
Stress Variables: Total Number% of Total Total Number% of Total
Normal BP response16 84.2442 86.7
No CP18 94.744487.1
CP (atypical)0 0.09 1.8
CP (typical)0 0.0265.1
CP (unspecified)1 5.231 6.1
BP: blood pressure.
CP: chest pain.
Table 3. Comparison of ECG diagnostic accuracy of: a) all exercise stress tests; b) all female patients (n=482); c) male patients
TreadmillSmall Heart (n=142) Normal Heart (n=869)
Positive Predictive Value38.0%64.3%
Negative Predictive Value85.9% 76.3%
TreadmillSmall Heart (n=123)Normal Heart (n=359)
Positive Predictive Value 34.9%*58.5%
Negative Predictive Value 84.0%84.1%
Treadmill Small Heart (n=19)Normal Heart (n=510)
Positive Predictive Value60%** 65.8%
Negative Predictive Value 92.9%**71.4%
*=significantly different from normal heart size (p,0.001).
**=unable to conduct statistical analysis due to insufficient sample size.
Electrocardiogram during Exercise Stress Test
PLoS ONE | www.plosone.org4August 2011 | Volume 6 | Issue 8 | e23044
perfusion and function in one setting. Studies conducted on a large
cohort of patients have not only demonstrated a high reproduc-
ibility of LV volume quantification by automated gated SPECT,
but also a good correlation with traditional techniques like
echocardiography, contrast angiography or pooled blood studies
[20–24]. For example, in a study conducted by Maruyama et al,
the LV volumes and LV mass index calculated by QGS SPECT
were consistent with those quantified by M-mode echocardiogra-
phy, by Devereux’s method . Even though few smaller studies
looking at the older gated SPECT in patients with small hearts
postulated that LV volumes can be over-estimated due to limited
spatial resolution, larger studies using appropriate image acquisi-
tion parameters and quality control methods validated gated
SPECT in such patients [26,27]. In addition, the newer and
commonly used software for Cedar-Sinai QGS with a SPET filter,
high cut-off frequency and appropriate zooming, and Emory tool
box or layer of maximum count method are well-correlated
[28–31]. Furthermore, in studies done by Gimelli et al, the authors
demonstrated a superior predictive value of gated SPECT derived
volumes compared to those derived from echocardiogram [32,33].
A potential limitation of this study was using nuclear imaging as
the criterion measure and not coronary angiography. However,
other studies have justified the accuracy and predictive capacity of
nuclear SPECT imaging and using myocardial perfusion imaging
rather than anatomical angiography may be more relevant in
determining ischemia [10,11]. Hansen et al., described a lower
diagnostic accuracy of SPECT imaging based on small heart size
. However, using coronary angiography as a gold standard has
its own problems as one is then using an anatomical imaging
modality rather than a functional imaging modality. The range of
angiographic stenosis, which results in ischemia, is quite broad and
dependent not just on the severity of a stenosis but also on the
length of stenosis and the degree of atherosclerotic disease
proximal and distal (e.g. 2 tandem 50% stenosis may be more
hemodynamically significant than one 70% stenosis). In addition,
determining the presence of CAD based on a secondary or tertiary
branch with significant stenosis can also lead to problems in
determining the accuracy of a diagnostic test for the presence of
ischemia. Ideally, obtaining fractional flow reserve on all patients
referred for angiography would help alleviate many of these
problems; however, this is quite burdensome for both the patient
and the catheterization laboratory.
Even though it is ideal to correlate the stress ECG and nuclear
stress results with coronary angiographic data, we were interested
in investigating the functional strategy for the diagnosis of
myocardial perfusion using non-invasive modalities, instead of
anatomic evaluation. In addition, we are unable to retrospectively
review this data as our archive for angiography during this time
period has been corrupted and the data is lost.
It is possible that a larger body mass index would be more prone
to attenuation artifacts due to either breast attenuation or other
soft tissue attenuation, which would lead to more positive test
results in the gold standard. Outcome data in future studies would
be required to better assess the contribution of larger body mass
index on the accuracy of SPECT imaging.
Another potential limitation is that hormonal status i.e. post
menopausal (in the female population) were not accounted for in
the patient population.
It is well-known that plain exercise-ECG test has high false
positivity. Our study strongly suggests that certain groups of
population like older women and particularly women with lower
BMI who have smaller body size may have smaller heart and
probably smaller coronary arteries. Patients with these character-
istics on physical examination may probably benefit more being
referred to imaging stress tests when indicated, compared to the
conventional exercise ECG stress test because of high false positive
results. In addition, patients with small left ventricular volumes
and mass by echocardiography also should be considered for stress
imaging instead of exercise ECG.
The exercise ECG has a higher false positive rate in patients
with small heart size when compared to patients with normal heart
size. As the overwhelming majority of patients with small heart size
were women, the results offer additional insight into a potential
explanation for the poor diagnostic accuracy of stress testing in
Female patients; (b) Male patients; (c) all patients.
False Positive and False Negative comparison for (a)
Conceived and designed the experiments: IK JFH. Performed the
experiments: CCD. Analyzed the data: SR GPB RA SJB LL BS JFH.
Wrote the paper: JCS GPB. Final approval of the article: IK SJB LL BS
TJS JFH. Critical revision of the article: TJS JFH.
1. American Heart Association Scientific Statement (2005) Consensus Statement:
Role of noninvasive testing in the clinical evaluation of women with suspected
coronary artery disease. Circ 111: 682–696.
2. Mieres JH, Shaw LJ, Hendel RC (2003) A report of the American Society of
Nuclear Cardiology Task Force on Women and Heart Disease. J Nucl Cardiol
3. Okin PM, Kligfield P (1995) Gender-specific criteria and performance of the
exercise electrocardiogram. Circ 92: 1209–1216.
4. Hachamovitch R, Berman DS, Kiat H (1996) Effective risk stratification using
exercise myocardial perfusion SPECT in women: gender-related differences in
prognostic nuclear testing. J Am Coll Cardiol 28: 34–44.
5. Hansen CL, Crabbe D, Rubin S (1996) Lower diagnostic accuracy of thallium-
201SPECT myocardial perfusion imaging in women: an effect of smaller
chamber size. J Am Coll Cardiol 28: 1214–1219.
6. Iskandrian AE, Jaekyeong H, Nallamothu N (1997) Detection of coronary artery
disease in women with use of stress single-photon emission computed
tomography myocardial perfusion imaging. J Nucl Cardiol 4: 329–335.
7. Barolsky SM, Gilbert CA, Faruqui A (1979) Differences in electrocardiographic
response toexercise ofwomen and men:a Non-Bayesian factor.Circ5:1021–1027.
8. Morise AP, Diamond GA (1995) Comparison of the sensitivity and specificity of
exercise electrocardiography in biased and unbiased populations of men and
women. Am Heart J 130: 741–747.
9. Santana-Boado C, Candell-Riera J, Castell-Conesa J (1998) Diagnostic accuracy
of technetium-99m-MIBI myocardial SPECT in women and men. J Nucl Med
10. Miller TD, Roger VL, Milavetz JJ (2001) Assessment of the exercise
electrocardiogram in women versus men using tomographic myocardial
perfusion imaging as the reference standard. Am J Cardiol 87: 868–873.
11. Pancholy SB, Fattah AA, Kamal AM (1995) Independent and incremental
prognostic value of exercise thallium single-emission computed tomographic
imaging in women. J Nucl Cardiol 2: 110–116.
12. American College of Sports Medicine (2006) In ACSM’s Guidelines for Exercise
Testing and Prescription. 7th ed. Philadelphia, PA: Lippincott Williams &
13. Hansen CL, Goldstein RA, Akinboboye O (2007) Myocardial perfusion and
function: Single photon emission computed tomography. J Nucl Cardiol 14:
14. Metz C (1978) Basic principles of ROC analysis. Semin Nucl Med 8: 283–298.
Electrocardiogram during Exercise Stress Test
PLoS ONE | www.plosone.org5 August 2011 | Volume 6 | Issue 8 | e23044
15. Sketch MH, Mohiuddin SM, Lynch JD (1975) Significant sex differences in the
correlation of electrocardiographic exercise testing and coronary arteriograms.
Am J Cardiol 36: 169–173.
16. Val PG, Chaitman BR, Waters DD (1982) Diagnostic accuracy of exercise ECG
lead systems in clinical subsets of women. Circ 65: 1465–1473.
17. Bokhari S, Bergmann SR (2002) The effect of estrogen compared to estrogen
plus progesterone on the exercise electrocardiogram. J Am Coll Cardiol 40:
18. Kwok Y, Kim C, Grady D (1999) Meta-analysis of exercise testing to detect
coronary artery disease in women. Am J Cardiol 83: 660–666.
19. de Simone G, Devereux RB, Kimball TR (1998) Interaction between body size
and cardiac workload; influence on left ventricular mass during body growth and
adulthood. Hypertension 31: 1077–1082.
20. Germano G, Kavanagh PB, Joseph T, Kavanagh MD (1998) Repeatability of
automatic left ventricular cavity volume measurements from myocardial
perfusion SPECT. J Nucl Cardiol 5: 477–83.
21. Nichols K, Lefkowitz D, Faber T (2000) Echocardiographic Validation of Gated
SPECT Ventricular Function Measurements. J Nucl Med 41: 1308–1314.
22. Cwajg E, Cwajg J, Keng F (2000) Comparison of global and regional left
ventricular function assessed by gated-SPECT and 2-D echocardiography. Rev
Port Cardiol Feb 19 Suppl 1: I39–46.
23. Nichols K, Tamis J, DePuey EG, Mieres J, Malhotra S, Rozanski A (1998)
Relationship of gated SPECT ventricular functional parameters to angiographic
measurements. J Nucl Cardial 5: 295–303.
24. Maruyama K, Hasegawa S, Nakatani D (2003) Left ventricular mass index
measured by quantitative gated myocardial SPECT with 99mTc-tetrofosmin: a
comparison with echocardiography. Annals of Nuclear Medicine 1: 31–39.
25. Ford PV, Chatziioannou SN, Moore WH (2001) Overestimation of the LVEF
by Quantitative Gated SPECT in Simulated Left Ventricles. J Nucl Med 42:
26. Toba M, Kumita S, Cho K (2000) Comparison of Emory and Cedars-Sinai
methods for assessment of left ventricular function from gated myocardial
perfusion SPECT in patients with a small heart. Ann Nucl Med 14: 421–6.
27. Nakajima K, Taki J, Higuchi T (2000) Gated SPET quantification of small
hearts: mathematical simulation and clinical application. Eur J Nucl Med 27:
28. Magdy MK, Abdelhamid E, Wafaa K (2005) Assessment of left ventricular
ejection fraction by four different methods using
SPECT in patients with small hearts: correlation with gated blood pool. Nuclear
Medicine Communication 26: 885–893.
29. Pratt JP, Lear JL, Trujillo N (1999) An automated program for LVEF
calculation from gated SPECT perfusion data that significantly reduces partial
volume errors (abstract). J Nucl Med 40(suppl): 176P.
30. Feng B, Sitek A, Gullberg GT (2002) Calculation of the Left Ventricular
Ejection Fraction Without Edge Detection: Application to Small Hearts. J Nucl
Med 43: 786–794.
31. Schwartz R, Mixon L, Germano G (1999) Gated SPECT reconstruction with
zoom and depth dependent filter improves accuracy of volume and LVEF in
small hearts (abstract). Journal of Nuclear Cardiology 6: S17.
32. Sharir T, Germano G, Kavanagh PB (1999) Incremental prognostic value of
post-stress left ventricular ejection fraction and volume by gated myocardial
perfusion single photon emission computed tomography. Circulation 100:
33. Gimelli A, Landi P, Marraccini P (2008) Left ventricular rejection fraction
measurements: accuracy and prognostic implications in a large population of
patients with known or suspected ischemic heart disease. Int J Cardiovasc
Imaging 24: 793–801.
99mTc tetrofosmin gated
Electrocardiogram during Exercise Stress Test
PLoS ONE | www.plosone.org6 August 2011 | Volume 6 | Issue 8 | e23044