Skip Navigation
Department of Health and Human Services www.hhs.gov
  • Home
  • Search for Research Summaries, Reviews, and Reports
 
 

Executive Summary – Jun. 12, 2012

Noninvasive Technologies for the Diagnosis of Coronary Artery Disease in Women

Formats

Table of Contents

Background

Cardiovascular disease is the leading cause of mortality for women in the United States.1 Coronary heart disease—which includes coronary artery (or atherosclerotic) disease (CAD), myocardial infarction (MI), acute coronary syndromes, and angina—is the largest subset of this mortality.1 According to the American Heart Association (AHA), approximately one in three female adults has some form of cardiovascular disease. Since 1984, the number of deaths attributed to cardiovascular disease in women has exceeded that in men, reaching 454,613 in 2005—more than deaths from all forms of cancer combined.2 It is estimated that 8.1 million women alive today have a history of heart attack, angina pectoris (chest pain or discomfort caused by reduced blood supply to the heart muscle), or both, and experts predict that in 2010 alone an estimated 370,000 women will have a new or recurrent MI. Overall, women who have had an acute MI—particularly those older than 55 years of age—have a worse prognosis than men, with a greater recurrence of MI and higher mortality.1 More women (5.5 million) than men (4.3 million) have angina in total numbers. However, the prevalence of CAD in women with chest pain is about 50 percent, compared with 80 percent in men, which complicates diagnosis in women.3

The AHA suggests there is evidence showing that women at risk for CAD are less often referred for the appropriate diagnostic test than are men.1 Coronary anatomy and pathology have traditionally been defined and identified by invasive, catheter-based x-ray angiography, also referred to as coronary angiography.4 In this invasive procedure, a catheter is inserted into the femoral, brachial, or radial artery and passed up through the aorta to directly engage the right and left coronary arteries; an iodinated contrast agent is then injected into each artery while digital x-ray images are recorded.4 The major benefits of invasive coronary angiography over noninvasive techniques are that the use of a catheter makes it possible to see the coronary arteries with greater anatomic precision and resolution and to combine diagnosis and treatment in a single procedure. The limitations of the procedure include the invasive nature of the test and the limited data on the functional impact of a luminal obstruction. These limitations are generally considered to be minor when compared with the benefits of the procedure, and coronary angiography is now the reference (gold) standard for clinical care of patients who have chest pain suggestive of CAD.

Coronary angiography, however, is not risk-free.4 Arterial bleeding can occur at the access site, and manipulation of the catheter within the aorta and coronary arteries may cause an atherosclerotic embolus that, in turn, could result in stroke or heart attack. Separation of material from the inner lining of the artery may also cause a blockage downstream of the catheter tip. The contrast agent used during the procedure to visualize the coronary arteries may cause anaphylaxis, renal impairment, or injury, and there is radiation exposure during the digital x-ray imaging. Although it is a rare occurrence, the catheter can puncture an artery and cause internal bleeding.

Coronary angiography is generally indicated in patients who have chest pain and are at high risk for CAD. For intermediate-risk patients, clinicians have a wide range of noninvasive diagnostic modalities to choose from, with wide variability in reported sensitivities and specificities. Noninvasive technologies (NITs) are especially important options for patients who have contraindications to invasive catheterization, or for those who would be put at higher risk for complications with invasive screening.4

Types of Noninvasive Technologies

NITs can assess functional status (i.e., ischemia or no ischemia) or visualize anatomic abnormalities (i.e., no CAD, nonobstructive CAD, or obstructive CAD). Types of NITs include the following:

  • Functional modalities:
    • Exercise/stress electrocardiography (ECG) exercise/stress or resting
    • Exercise/stress echocardiography (ECHO) with or without a contrast agent
    • Exercise/stress radionuclide myocardial perfusion imaging, including single proton emission computed tomography (SPECT) and positron emission tomography (PET)
  • Anatomic modalities:
    • Stress myocardial perfusion and wall motion magnetic resonance imaging (CMR)
    • Coronary computed tomography angiography (coronary CTA)

The AHA and the American College of Cardiology (ACC) recommend that women with suspected CAD should be classified as either symptomatic or asymptomatic and further classified as being at low, intermediate, or high risk for the disease to guide the decision about which diagnostic test to use first.1 In 2005, the AHA developed a consensus statement on the role of noninvasive testing in the clinical evaluation of women with suspected CAD. In this statement, the AHA recommended that women who are symptomatic and at intermediate to high risk of having CAD should undergo noninvasive diagnostic studies (i.e., exercise electrocardiography and cardiac imaging studies) and that those who are asymptomatic and at low risk of CAD should not undergo cardiac imaging studies.1 The AHA consensus statement was a thorough synopsis of the extant literature regarding the diagnosis of CAD in women with expert-guided recommendations for the workup of symptomatic women but did not include a comparative effectiveness review of the accuracy of the various NIT modalities in women.

Objectives

The goal of this comparative effectiveness report was to conduct a systematic review of the peer-reviewed medical literature assessing (1) the accuracy of different NITs for diagnosing CAD in women with symptoms suspicious of CAD, (2) the predictors affecting test accuracy, (3) the ability to provide risk stratification and prognostic information, inform decisionmaking about treatment options, and affect clinical outcomes, and (4) the safety concerns and risks to women undergoing these tests. The following Key Questions (KQs) were considered in this comparative effectiveness review:

  • KQ 1. What is the accuracy of one NIT in diagnosing obstructive and nonobstructive CAD when compared with another NIT or with coronary angiography in women with symptoms suspicious for CAD?
    • Exercise ECG stress test, including resting ECG technology (e.g., multifunctional cardiogram)
    • Exercise/stress ECHO with or without a contrast agent
    • Exercise/stress radionuclide myocardial perfusion imaging, including SPECT and PET
    • CMR imaging
    • Coronary CTA
  • KQ 2. What are the predictors of diagnostic accuracy (e.g., age, race/ethnicity, body size, heart size, menopausal status, functional status, stress modality) of different NITs in women?
  • KQ 3. Is there evidence that the use of NITs (when compared with other NITs or with coronary angiography) in women improves:
    • KQ 3a. Risk stratification/prognostic information?
    • KQ 3b. Decisionmaking regarding treatment options (e.g., revascularization, optimal medical therapy)?
    • KQ 3c. Clinical outcomes (e.g., death, myocardial infarction, unstable angina, hospitalization, revascularization, angina relief, quality of life)?
  • KQ 4. Are there significant safety concerns/risks (i.e., radiation exposure, access site complications, contrast agent-induced nephropathy, nephrogenic systemic fibrosis, anaphylaxis, arrhythmias) associated with the use of different NITs to diagnose CAD in women with symptoms suspicious for CAD?

Analytic Framework

Figure A shows the analytic framework for the systematic review of NITs for the diagnosis of CAD in women.

Figure A. Analytic framework

Abbreviations: CAD = coronary artery disease; KQ = Key Question; NSF = nephrogenic systemic fibrosis

Methods

Input From Stakeholders

The Evidence-based Practice Center (EPC) followed AHRQ’s recommended methodology, described in Methods Guide for Effectiveness and Comparative Effectiveness Reviews,5 for literature search strategies, inclusion/exclusion of studies, abstract screening, data abstraction and management, assessment of methodological quality of individual studies, data synthesis, and grading of evidence for each KQ.

During the topic refinement stage, we solicited input from Key Informants, representing clinicians (cardiology, primary care, cardiac imaging), patients, scientific experts, and Federal agencies to help define the KQs. The KQs were then posted for public comment for 30 days, and the comments received were considered in the development of the research protocol. We next convened a Technical Expert Panel (TEP), comprising clinical, content, and methodological experts, to provide input in defining populations, interventions, comparisons, or outcomes as well as identifying particular studies or databases to search. The Key Informants and members of the TEP were required to disclose any financial conflicts of interest greater than $10,000 and any other relevant business or professional conflicts of interest. Any potential conflicts of interest were balanced or mitigated. Neither Key Informants nor members of the TEP did analysis of any kind and did not contribute to the writing of the report.   

Data Sources and Selection

We included studies published in English from January 1, 2000, through September 12, 2011. Search strategies were specific to each database in order to retrieve the articles most relevant to the KQs. Our search strategy used the National Library of Medicine’s medical subject headings (MeSH) keyword nomenclature developed for MEDLINE® and adapted for use in other databases. We used PubMed®, Embase®, the Cochrane Database of Systematic Reviews, and the Cochrane Central Registry of Controlled Trials for our literature search. We also searched the grey literature of study registries and conference abstracts for relevant articles from completed trials, including Clinicaltrials.gov; metaRegister of Controlled Trials; ClinicalStudyResults.org; WHO: International Clinical Trials Registry Platform Search Portal; CSA Conference Papers Index; and Scopus. The exact search strings used in our strategy are given in Appendix A of the full report. Reference lists of articles applicable to the relevant KQs of previous AHRQ reports on this topic6,7 and from identified systematic reviews and meta-analyses were manually hand-searched and cross-referenced against our library, and additional manuscripts were retrieved. All citations were imported into an electronic bibliographic database (EndNote® Version X4; Thomson Reuters, Philadelphia, PA).

We developed a list of article inclusion and exclusion criteria for the KQs (Table A). Using the prespecified inclusion and exclusion criteria, titles and abstracts were examined independently by two reviewers for potential relevance to the KQs. Articles included by any reviewer underwent full-text screening. At the full-text screening stage, two independent reviewers read each article to determine if it met eligibility criteria. At the full-text review stage, paired researchers independently reviewed the articles and indicated a decision to “include” or “exclude” the article for data abstraction. When the paired reviewers arrived at different decisions about whether to include or exclude an article, we reconciled the difference through a third-party arbitrator. Articles meeting our eligibility criteria were included for data abstraction. Relevant systematic review articles, meta-analyses, and methods articles were flagged for hand-searching and cross-referencing against the library of citations identified through electronic database searching.

Table A. Summary of inclusion and exclusion criteria
Study Characteristic Inclusion Criteria Exclusion Criteria

Study design

An article was included if the following two criteria were met:
Original data or related methodology paper of an included article
Randomized controlled trial, prospective or retrospective observational study, or registry

Editorial
Not a systematic review
Letter to editor
Case series
Review article, meta-analysis, or methods paper of an excluded article
Not peer reviewed

Population

Study included adult women (age ≥ 18 years of age) who present symptoms of symptoms suspicious for CAD (e.g., exertional dyspnea, shortness of breath, and/or angina) with or without a known diagnosis of CAD; data for women must be presented separately from data for men

All subjects were < 18 years of age, or some subjects were under < 18 but results were not broken down by age
No patients had symptomatic chest pain (i.e., an asymptomatic population), or some of the patients had symptomatic chest pain but results were not reported separately for this subgroup
All patients were known to have CAD and were not being tested for chest pain symptoms (e.g., postrevascularization testing to assess for persistent ischemia)

Interventions

NITs for the diagnosis of obstructive and nonobstructive CAD included:
Exercise electrocardiogram stress test
Resting electrocardiogram technology
Exercise/stress echocardiography with or without a contrast agent
Exercise/stress radionuclide myocardial perfusion imaging, including single proton emission computed tomography and positron emission tomography
Stress myocardial perfusion and wall motion magnetic resonance imaging
Coronary computed tomography angiography

Coronary artery calcium scoring by electron beam computed tomography since this modality is often used to screen asymptomatic patients for CAD

Comparators

Another NIT or coronary angiography

Study did not compare one NIT with another, or with coronary angiography

Outcomes

Primary outcome—accurate diagnosis of obstructive and nonobstructive CAD

Outcomes not related to diagnostic accuracy for detecting CAD

KQ 1 patient-level outcomes:
Sensitivity
Specificity
True positive, false negative, true negative, false positive
Indeterminate or technically inadequate results
Prevalence

Vessel-based outcomes
Outcomes of women not reported separately from total population

KQ 2 outcomes: Predictors of diagnostic accuracy—age, race/ethnicity, body size, heart size, menopausal status, functional status, stress modality

Outcomes of women not reported separately from total population

KQ 3 outcomes:
Risk stratification/prognostic information
Treatment —none, medical therapy, percutaneous coronary intervention, or coronary artery bypass surgery
Clinical outcomes—death, myocardial infarction, unstable angina, hospitalization, revascularization, angina relief, quality of life

Outcomes of women not reported separately from total population

KQ 4 outcomes: Safety and adverse events—radiation exposure, access site complications, contrast agent-induced nephropathy, nephrogenic systemic fibrosis, anaphylaxis, arrhythmias—and how these events varied by demographic factors

Outcomes of women not reported separately from total population

Setting

Inpatient or outpatient settings, primarily primary care and cardiology clinics

None

Publication languages

English only

Given the high volume of English-language publications (including the majority of known important studies), non-English articles were excluded

Abbreviations: CAD = coronary artery disease; KQ = Key Question; NIT = noninvasive technology

Data Extraction and Quality Assessment

The investigative team created forms for abstracting the data elements for the KQs. Based on their clinical and methodological expertise, two researchers were assigned to abstract data from the eligible articles pertaining to the research questions. One researcher abstracted the data, and the second overread the article and the accompanying abstraction form to check for accuracy and completeness. Disagreements were resolved by consensus or by obtaining a third reviewer’s opinion if consensus was not reached by the first two researchers. To aid in both reproducibility and standardization of data collection, researchers received data abstraction instructions directly on each form created specifically for this project with the DistillerSR data synthesis software program (Evidence Partners Inc., Manotick, ON, Canada). We designed these forms to collect the data required to evaluate the specified eligibility criteria for inclusion in this review as well as to collect demographics and data needed to determine outcomes (intermediate outcomes, health outcomes, and safety outcomes). Appendix B of the full report lists the elements used in the data abstraction forms.

Appendix C of the full report contains a bibliography of all studies included in this review, organized alphabetically by author. When appropriate, methods articles providing additional detail were considered when abstracting data for an included study.

The studies included in this comparative effectiveness review were assessed on the basis of the quality of their reporting of relevant data. We evaluated the quality of individual studies using the approach described in AHRQ’s Methods Guide for Effectiveness and Comparative Effectiveness Reviews (hereafter referred to as the Methods Guide).5 To assess study quality, we (1) classified the study design, (2) applied predefined criteria for quality and critical appraisal, and (3) made a summary judgment of the study’s quality. To evaluate methodological quality, we applied criteria for each study type that were derived from the core elements described in the Methods Guide5 and from QUADAS,8 a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. To indicate the summary judgment of the quality of the individual studies, we used the summary ratings of Good, Fair, and Poor based on the study’s adherence to well-accepted standard methodologies (such as QUADAS) and adequate reporting standards.

We used data abstracted on the population studied, the intervention and comparator, the outcomes measured, settings, and timing of assessments to identify specific issues that may have limited the applicability of individual studies or a body of evidence as recommended in the Methods Guide.5,9 We used these data to evaluate the applicability to clinical practice, paying special attention to study eligibility criteria, demographic features of the enrolled population in comparison with the target population, the intervention used in comparison with technologies currently in use, and clinical relevance and timing of the outcome measures. We summarized issues of applicability qualitatively. Appendix D of the full report summarizes our assessment of the quality and applicability for each included study as well as the assessed QUADAS quality scores for diagnostic accuracy.

Data Synthesis and Analysis

We summarized the primary literature by abstracting relevant continuous data (e.g., age, sensitivity, specificity, event rates) and categorical data (e.g., race/ethnicity, presence of CAD). Data for patients with no known diagnosis of CAD were collected and analyzed separately from data for mixed CAD populations that included patients with and without known CAD. We then determined the feasibility of completing a quantitative synthesis (i.e., summary receiver operating characteristic [SROC] curves for diagnostic accuracy or meta-analysis for other outcomes). The feasibility of a meta-analysis or SROC curve depended on the volume of relevant literature, the homogeneity of the studies in terms of the populations studied, the interventions included or the outcomes assessed, and the completeness of the results reporting. For each SROC calculation, we ran separate analyses of the accuracy of each NIT modality compared with coronary angiography on the no-known CAD and mixed CAD populations using random-effects models to quantitatively synthesize the available evidence. In our primary analyses, we evaluated these performance characteristics in the population of women who had no previously known CAD. In secondary analyses, we explored a broader patient population by including those studies that had women from mixed populations of known and no known CAD. We also assessed the impact on our findings if, in each population, we restricted our analyses to those studies that were assessed to be good quality. We then compared the performance characteristics of the NIT modalities with each other in a generalized linear mixed model. In a final exploratory analysis, we evaluated the test performance of the modalities in women compared with men in a similar generalized linear model with sex as a covariate. We presented summary estimates and confidence intervals (CIs).

For synthesizing the accuracy data for studies included in our assessment of KQ 1, we used the following approach as advocated by Leeflang,, et al.10 This approach allows the paired nature of sensitivity and specificity and randomness between studies to be taken into account. The analyses are based on true positive (TP), false negative (FN), false positive (FP), and true negative (TN) frequencies abstracted from relevant publications. Estimated study specific sensitivity (TP/[TP+FN]) and specificity (TN/[TN+FP]) values are displayed in paired forest plots together with exact 95% CIs.11 The fixed-effects estimates and their variance–covariance matrix provided (after reverse logit transformation) summary sensitivity and specificity values and a joint confidence region (dotted oval shape on figures) as well as separate CIs for summary sensitivity and specificity as presented on figures and forest plots in the report. We used the Rutter and Gatsonis12 SROC curve as described by Arends,, et al.,13 and it is presented in figures as a solid line over the range of the available data.

Grading the Body of Evidence

The strength of evidence for each Key Question was assessed using the approach described in AHRQ’s Methods Guide on Medical Test Reviews for grading the evidence related to the diagnostic accuracy of the NITs (KQ 1),14 and the Methods Guide for Effectiveness and Comparative Effectiveness Reviewsfor grading the evidence related to the other Key Questions (KQs 2–4).5,15 The evidence was evaluated using the four required domains: risk of bias (low, medium, or high), consistency (consistent, inconsistent, or unknown/not applicable), directness (direct or indirect), and precision (precise or imprecise). Additionally, when appropriate, the studies were evaluated for the presence of confounders that would diminish an observed effect, the strength of association (magnitude of effect), and publication bias. The strength of evidence was assigned an overall grade of High, Moderate, Low, or Insufficient according to the following four-level scale:

  • High—High confidence that the evidence reflects the true effect. Further research is very unlikely to change our confidence in the estimate of effect.
  • Moderate—Moderate confidence that the evidence reflects the true effect. Further research may change our confidence in the estimate of effect and may change the estimate.
  • Low—Low confidence that the evidence reflects the true effect. Further research is likely to change the confidence in the estimate of effect and is likely to change the estimate.
  • Insufficient—Evidence either is unavailable or does not permit estimation of effect.

Results

The flow of articles through the literature search and screening process is depicted in Figure B. Of the 8,231 citations identified by our searches, 634 were duplicates. A manual search identified an additional 445 citations for a total of 8,042 citations. After applying inclusion/exclusion criteria at the title-and-abstract level, 1,772 full-text articles were retrieved and screened. Of these, 1,662 articles were excluded at the full-text screening stage. Of these, we excluded 1,376 (83 percent) for not reporting data on women and 615 (37 percent) for looking only at a population with known CAD. (Note that an article may have been excluded for more than one reason.) The final set comprised 110 articles representing 104 studies.

Of the 104 studies, 1 was an RCT, 79 were prospective observational, and 24 were retrospective observational with study cohorts comprising individuals who presented for NIT testing and received diagnostic coronary angiography (100 studies) or another NIT modality only (4 studies). The four studies without coronary angiography compared ECHO with ECG16,17 or ECG with SPECT.18,19 Three of these studies were applicable to KQ 3,16-18 and one was applicable to KQ 2.19 Of the 94 studies included in the KQ 1 results, 5 reported NIT versus NIT comparisons in addition to coronary angiography.20-24

Figure B. Literature flow diagram

Figure B. Literature flow diagram

Abbreviations: CAD = coronary artery disease; KQ = Key Question; NIT = noninvasive technology; SR = systematic review

Summary of Key Findings

We analyzed the results by study population (no known CAD and mixed CAD populations) and by study quality (good quality rating). Table B and Figure C show the summary sensitivities and specificities for each NIT modality. Table C summarizes our key findings.

Table B. Summary of accuracy of NITs compared with coronary angiography for diagnosing CAD in women
Modality Population Quality of Included Studies Number of Studies Number of Women Summary Sensitivity
(95% CI)
Summary Specificity
(95% CI)

ECG

No known CAD

All

Good

29

10

3,392

1,410

62% (55%–68%)

70% (58%–79%)

68% (63%–73%)

62% (53%–69%)

Mixed population

All

Good

41

13

4,879

1679

61% (54%–67%)

65% (52%–76%)

65% (58%–72%)

60% (52%–68%)

ECHO

No known CAD

All

Good

14

5

1,286

561

79% (74%–83%)

79% (69%–87%)

83% (74%–89%)

85% (68%–94%)

Mixed population

All

Good

22

8

1,873

807

78% (73%–83%)

77% (65%–85%)

86% (79%–91%)

89% (76%–95%)

SPECT

No known CAD

All

Good

14

4

1,000

394

81% (76%–86%)

83% (52%–95%)

78% (69%–84%)

72% (37%–92%)

Mixed population

All

Good

30

10

2,146

982

82% (77%–87%)

82% (72%–88%)

81% (74%–86%)

79% (66%–87%)

CMR

No known CAD

All

Good

5

5

501

501

72% (55%–85%)

72% (55%–85%)

84% (69%–93%)

84% (69%–93%)

Mixed population

All

Good

6

5

778

610

78% (61%–89%)

76% (55%–89%)

84% (74%–90%)

84% (72%–91%)

Coronary CTA

No known CAD

All

Good

5

3

474

124

93% (69%–99%)

85% (26%–99%)

77% (54%–91%)

73% (17%–97%)

Mixed population

All

Good

8

4

690

201

94% (81%–98%)

83% (58%–94%)

87% (68%–96%)

77%% (40%–94%)

Abbreviations: CAD = coronary artery disease; CI = confidence interval; CMR =cardiac magnetic resonance; CTA = computed tomography angiography; ECG = exercise/stress electrocardiogram; ECHO = echocardiogram; SPECT = single proton emission computed tomography

Figure C. Summary of accuracy of NITs compared with coronary angiography for diagnosing CAD in women with no known CAD (all studies)

Figure C. Summary of accuracy of NITs compared with coronary angiography for diagnosing CAD in women with no known CAD (all studies)

Table C. Summary of key findings
Key Question Strength of Evidence Conclusions

KQ 1: Diagnostic accuracy of NITs in women

ECG: High
ECHO: High
SPECT: High
CMR: Low
Coronary CTA: Low

94 studies described the diagnostic accuracy of NITs in comparison to another NIT or coronary angiography in women. Of these 94 studies, 78 studies included sufficient data to estimate the sensitivity and specificity of the NIT compared with coronary angiography.
Summary from all studies with no known CAD:
41 studies (13 good quality, 22 fair, 6 poor) of exercise ECG showed a summary sensitivity of 62% and specificity of 68%
22 studies (8 good quality, 13 fair, 1 poor) of exercise/stress ECHO showed a summary sensitivity of 79% and specificity of 83%
30 studies (10 good quality, 15 fair, 5 poor) of exercise/stress radionuclide perfusion imaging (SPECT, PET) showed a summary sensitivity of 81% and specificity of 78%
6 studies (5 good quality, 1 fair) of CMR imaging showed a summary sensitivity of 72% and specificity of 84%
8 studies (4 good quality, 4 fair) of coronary CTA showed a summary sensitivity of 93% and specificity 77%
Overall, within a given modality, the summary sensitivities and specificities were similar for both types of populations (unknown CAD and mixed known and no known CAD) and for all studies when compared with good-quality studies. For the newer technologies (i.e., coronary CTA and CMR), more studies in women would be needed to support these findings since the 95 % CIs were quite wide.
In testing for a statistically significant difference between the diagnostic accuracy of testing modalities in women, our analyses determined that for women with no previously known CAD, there were differences between the performance of the available modalities (p < 0.001). The sensitivity of ECHO and SPECT was significantly higher than that of ECG. Specificity of ECG was less than that of CMR (borderline) and of ECHO. In the subset of studies that were good-quality and where there was no known CAD in the included population, our analyses again demonstrated differences between performance of tests (p = 0.006) with the specificity of ECG being less than that of CMR and ECHO.

KQ 1 (continued)

 

Sensitivity analyses exploring mixed populations of women with known and no known CAD showed no statistically significant difference in the sensitivities and specificities from our primary analysis. An analysis exploring the prevalence of CAD across the different NIT modality studies also showed no statistically significant difference. In addition, there were very few studies (1 SPECT, 1 ECHO, and 3 ECG) that did not complete a coronary angiography in all patients who underwent the NIT; therefore the results are minimized for verification bias. Finally we found no evidence of publication bias across the different modalities in our 4 populations of interest (studies of women with no known CAD, good quality studies of women with no known CAD, studies of women from mixed populations, and good quality studies of women from mixed populations).

KQ 2: Predictors of diagnostic accuracy in women

Insufficient

11 studies (4 good quality, 5 fair, 2 poor) described diagnostic accuracy, and 9 of these examined predictors of diagnostic accuracy of different NITs in women.
Summary:
The predictors assessed included (1) postmenopausal women ages 55 to 64 (1 study), (2) race/ethnicity (2 studies), (3) heart size (4 studies), (4) pretest probability (3 studies), and (5) use of beta blocker medications (1 study).
We identified no studies examining the influence of age alone, functional status, or body size on diagnostic accuracy in women.
In terms of the NIT modality, we found four studies of stress ECHO, six studies of stress ECG, two studies of CMR, and four studies of SPECT that reported these predictors.
Insufficient evidence was available to draw definitive conclusions about predictors given the small number of studies for each predictor and for each modality, as well as the combination of predictor by modality.

KQ 3: Improving risk stratification, decisionmaking, and outcomes in women

Insufficient

13 studies (3 good quality, 9 fair, 1 poor) reported prognostic, outcome, or decisionmaking data comparing one NIT with another NIT or with coronary angiography in women with symptoms suspicious for CAD.
Summary:
We found 8 studies assessing risk stratification and prognostic information, 2 studies assessing decisionmaking for treatment options, and 4 studies that provided comparative clinical outcomes.
There were insufficient data to demonstrate that the use of specific NITs (compared with coronary angiography) routinely provided incremental risk stratification, prognostic information, or other meaningful information to improve decisionmaking and improve patient outcomes.
Most findings reported in the literature would require significant confirmation and replication in larger studies with women.

KQ 4: Safety concerns

Insufficient

13 studies (9 good quality, 4 fair) reported data pertinent to safety concerns or risks associated with the use of NITs to diagnose CAD in women with symptoms suspicious for CAD.
Summary:
Safety data were reported on the following modalities: (1) stress ECG (4 studies), (2) ECHO (6 studies), (3) SPECT (3 studies), (4) CMR (2 studies), and (5) coronary CTA (4 studies).
Data specific to women on access site complications, contrast agent-induced nephropathy, nephrogenic systemic fibrosis, or anaphylaxis associated with NITs were not reported in any of the studies included in this report.
Other than higher mean effective radiation doses for coronary CTA studies for women compared with men (from 3 out of 4 studies reporting radiation exposure levels), the extant literature does not provide sufficient evidence to conclude whether safety concerns, risks, or radiation exposure associated with different NITs to diagnose CAD in patients with suspected CAD differ significantly between women and men.

Discussion

In summary, the findings of this comparative effectiveness review provide evidence for the accuracy of exercise/stress ECG, ECHO, SPECT, CMR, and coronary CTA compared with coronary angiography used for diagnosing CAD in women. The diagnostic accuracy appears to be consistent over time except for the sensitivity of CMR, which appears to be increasing over time (although the large confidence intervals reflect the underlying uncertainty in this measure). We are confident that the summary statistics for ECG, ECHO, and SPECT are robust and unlikely to change with the addition of new studies based on both the number of good-quality trials comparing these modalities with coronary angiography and the tight confidence intervals. More good-quality studies comparing CMR or coronary CTA with coronary angiography in the no-known CAD population and reporting sex-based results are needed to strengthen the summary statistics for those modalities.

Decisions around performing tests (either noninvasive or invasive) in patients with symptoms suspicious for CAD revolve around first understanding the pretest probability and testing/action thresholds for patients from the AHA/ACC stable angina guidelines and appropriate use criteria for the various NIT modalities.25-28 Specifically, clinicians faced with patients who have a guideline-defined low-to-intermediate pretest probability of CAD may decide to obtain a noninvasive test, ideally with a high negative predictive value in this population and low risk of adverse events, in order to “rule out” disease. These may be patients with atypical chest pain (e.g., reflux or musculoskeletal disease) who are concerned about a heart problem and who require reassurance that their symptoms are not cardiac in origin. In contrast, in patients with high pretest probability of CAD (greater than 90 percent chance), a test with very high positive predictive value in this population and potentially more risk may be chosen since the disease of interest is thought to be present; in these cases, invasive angiography—the gold standard—is recommended by the current clinical practice guidelines. Finally, it is the spectrum of intermediate probability between 10 and 90 percent for which the clinicians must choose noninvasive tests that provide the right balance of sensitivity, specificity, and clinical risk to warrant testing. The choice of NIT may differ by clinician preference, availability, or setting (outpatient vs. chest pain unit of an emergency department).

It is in this context that the findings of this report on the effectiveness of NITs in women must be considered. First, women are thought to be at lower pretest probability of CAD when evaluated in comparison with men of the same age. When comorbidities or risk factors are taken into account, the pretest likelihood increases with a higher number of comorbidities. Second, women susceptible to some of the adverse effects of testing may have poor test performance or have higher rates of complication from invasive arterial access. Third, because of body shape and limited functional capacity, women may not obtain the same test performance that men do from noninvasive testing. Finally, because of the lack of full representation of women across the spectrum of disease, the available literature may not provide data on performance at the ends of the probability spectrum. Spectrum bias may be present since the studies we evaluated had potentially varied populations and varied disease definitions.

While readers may assume that requiring coronary angiography as the comparator would bias this report toward a higher risk CAD population, we found that the mean CAD prevalence ranged from 0.26 to 0.44; thus there was a broad spectrum of CAD prevalence in these studies. In fact, the range of CAD prevalence in this review is similar to a recent analysis of a large administrative database of patients referred for coronary angiography in which the prevalence of significant obstructive disease was 38 to 40 percent.29 The patient population that does not require coronary angiography can be characterized as having symptoms with low suspicion for CAD or pretest probability of less than 10 percent (note that all included studies enrolled patients with “suspected CAD”). Thus, results from this review would not apply to patients with low pretest probability of disease (e.g., gastroesophageal reflux, musculoskeletal pain, or panic attacks) where an NIT may be performed for clinical reassurance that their symptoms are noncardiac in origin.

In general, because there are few patients with high pretest probability, most clinicians would prefer to have patients undergo one NIT prior to determining a treatment choice or referral to coronary angiography. More than one NIT test is often required when the initial test results are equivocal. Our review did not identify studies that discussed the order in which different NITs were used for evaluating CAD. In fact, multiple testing or layered-testing strategies are areas where significant research is needed.

The current data suggest that NITs with higher sensitivity include coronary CTA and SPECT, and stress ECHO may represent an NIT with higher specificity. Stress CMR shows emerging data that may be in the upper range for both sensitivity and specificity. Additionally, the findings also demonstrate that NIT performance in women is not as good as in men, likely due to the reasons addressed above. The accuracy may also be location or operator dependent, and thus the results of published studies conducted at highly specialized centers may not uniformly apply to those seen in routine practice. Choice of NIT—and whether to use exercise or pharmacological stress imaging—may be influenced by functional capacity, which tends to be lower in women compared with men. Of note, the accuracy data for NIT modalities in men appeared a little higher than expected given previous meta-analyses of diagnostic accuracy data in the total population, which is likely because the published literature combined the accuracy data for men and women. Taken in context, these findings give support to the current ACC/AHA recommendations and studies on noninvasive testing in women.

Women are more likely than men to have false-positive stress tests; i.e., abnormal stress imaging with nonobstructive CAD on coronary angiography. In fact, up to 9 percent of women presenting with acute coronary syndrome will not have obstructive CAD when they undergo coronary angiography for potential PCI.30 Some experts suggest that these phenomena are due to the presence of microvascular obstruction, the incidence of which is hard to determine since there is no clear diagnostic test used to establish the diagnosis.

Currently, there is debate on whether NITs that measure heart function abnormalities (ECG abnormalities, wall motion abnormality, ischemia), including exercise ECG, stress ECHO, and cardiac nuclear imaging, are equivalent or inferior to NITs that measure anatomic abnormalities (detection of CAD) by CMR or coronary CTA. Will knowing the coronary anatomy (nonobstructive or obstructive) in symptomatic patients lead to better implementation of secondary measures—control of blood pressure, diabetes, and hyperlipidemia—to reduce future cardiac events? Or is it more important to intervene with medications and/or revascularization when ischemia is present? Though this review does not answer these important questions, we describe this evidence gap in the Future Research section.

Limitations of This Review

Despite identifying 104 studies (110 articles) that met the inclusion criteria, this systematic review has several limitations. First, our search focused on comparator studies of the various NITs with a gold standard of coronary angiography for establishing the diagnosis of CAD in symptomatic patients. While this focus was adequate for identifying studies to assess the diagnostic accuracy of the NIT modalities in women, we found very few comparative studies that reported the influence of clinical characteristics or patient demographics on diagnostic accuracy. Few comparative studies (NIT vs. coronary angiography, or NIT vs. NIT) provided information on incremental risk stratification, prognostic information, or meaningful information regarding decisionmaking, and few reported the significant risks in women. Study results on these issues were reported for the total patient population and did not separate the effects by sex. Many of the included studies were single-sex (women) studies and limited our ability to fully evaluate sex differences. Also, by focusing on symptomatic patients, this report did not review the use of coronary artery calcium scoring for asymptomatic, high-risk populations.

We are aware that there are several noncomparator studies of each of the NIT modalities that address these issues in women since routine clinical care does not require two NIT modalities or an NIT modality plus coronary angiography for the diagnostic workup of suspected CAD. Given the focus on comparative effectiveness, we did not include these noncomparator studies in our review. By focusing the review on comparative studies, however, we are reducing the bias that is inherent in noncomparative studies. Noncomparative studies have selection, spectrum, and intervention biases for the following reasons: The choice of NIT is determined by the treating provider; only a subset of patients with indeterminate or positive results are referred for further NIT testing or coronary angiography; and the clinical outcomes may be influenced by the medical treatments or revascularization options that are offered. Second, the sample size and low representation of women in most of the comparator studies may affect the authors’ ability to analyze the results by sex, therefore reducing the number of studies reporting these findings separately. Third, most studies lacked long-term followup of the patient population, which affected our ability to find studies that reported prognostic information on how the different NITs influenced clinical outcomes. Finally, our summary of the harms and risks of NITs is limited by the lack of disclosure of periprocedural and postprocedural complications in most of the studies.

Conclusions

This systematic review has provided evidence for the summary sensitivities and specificities of exercise/stress ECG, ECHO, SPECT, CMR, and coronary CTA compared with coronary angiography in women. There was limited or insufficient evidence on the influence of clinical and demographic factors on comparative diagnostic accuracy, risk stratification, prognostic information, treatment decisions, clinical outcomes, and harms from different NITs specifically in women. Modifying the search criteria to include noncomparator studies of NIT modalities may increase the number of studies that address this limitation.

Glossary

AHA American Heart Association
ACC American College of Cardiology
CAD coronary artery disease
CMR cardiac magnetic resonance imaging
CTA coronary computed tomography angiography
ECG electrocardiogram, electrocardiography
ECHO echocardiogram, echocardiography
KQ Key Question
MeSH Medical Subject Heading
NIT noninvasive technology
NSF nephrogenic systemic fibrosis
PET positron emission tomography
MI myocardial infarction
SPECT single proton emission computed tomography
TEP Technical Expert Panel

References

  1. Mieres JH, Shaw LJ, Arai A,, et al. Role of noninvasive testing in the clinical evaluation of women with suspected coronary artery disease: Consensus statement from the Cardiac Imaging Committee, Council on Clinical Cardiology, and the Cardiovascular Imaging and Intervention Committee, Council on Cardiovascular Radiology and Intervention, American Heart Association. Circulation 2005;111(5):682-96. PMID: 15687114
  2. Lloyd-Jones D, Adams RJ, Brown TM,, et al. Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 2010;121(7):e46-e215. PMID: 20019324
  3. Stangl V, Witzel V, Baumann G,, et al. Current diagnostic concepts to detect coronary artery disease in women. European Heart Journal 2008;29(6):707-17. PMID: 18272503
  4. Matchar DB, Mark DB, Patel MR,, et al. Non-invasive imaging for coronary artery disease. Technology Assessment. Rockville, MD: Agency for Healthcare Research and Quality, October 2006. http://www.cms.gov/determinationprocess/downloads/id34TA.pdf. Accessed December 7, 2010.
  5. Agency for Healthcare Research and Quality. Methods Guide for Effectiveness and Comparative Effectiveness Reviews. Rockville, MD: Agency for Healthcare Research and Quality. http://www.effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?pageaction=displayproduct&productid=318. Accessed August 22, 2011.
  6. Grady D, Chaput L, Kristof M. Results of Systematic Review of Research on Diagnosis and Treatment of Coronary Heart Disease in Women. Evidence Report/Technology Assessment No. 80. (Prepared by the University of California, San Francisco-Stanford Evidence-based Practice Center under Contract No 290-97-0013.) AHRQ Publication No. 03-0035. Rockville,MD: Agency for Healthcare Research and Quality. May 2003. http://archive.ahrq.gov/downloads/pub/evidence/pdf/chdwom/chdwom.pdf. Accessed July 14, 2011.
  7. Grady D, Chaput L, Kristof M. Diagnosis and Treatment of Coronary Heart Disease in Women: Systematic Reviews of Evidence on Selected Topics. Evidence Report/Technology Assessment No. 81. (Prepared by the University of California, San Francisco-Stanford Evidence-based Practice Center under Contract No 290-97-0013.) AHRQ Publication No. 03-E037. Rockville, MD: Agency for Healthcare Research and Quality. May 2003. http://www.ahrq.gov/downloads/pub/evidence/pdf/chdwomtop/chdwmtop.pdf. Accessed July 14, 2011.
  8. Whiting P, Rutjes AW, Reitsma JB,, et al. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003;3:25. PMID: 14606960
  9. Atkins D, Chang SM, Gartlehner G,, et al. Assessing applicability when comparing medical interventions: AHRQ and the Effective Health Care Program. J Clin Epidemiol 2011;64(11):1198-207. PMID: 21463926
  10. Leeflang MM, Deeks JJ, Gatsonis C,, et al. Systematic reviews of diagnostic test accuracy. Ann Intern Med 2008;149(12):889-97. PMID: 19075208
  11. Blyth CR, Still HA. Binomial Confidence-Intervals. Journal of the American Statistical Association 1983;78(381):108-116. PMID: ISI:A1983QG01500024
  12. Rutter CM, Gatsonis CA. Regression methods for meta-analysis of diagnostic test data. Acad Radiol 1995;2 Suppl 1:S48-56; discussion S65-7, S70-1 pas. PMID: 9419705
  13. Arends LR, Hamza TH, van Houwelingen JC,, et al. Bivariate random effects meta-analysis of ROC curves. Med Decis Making 2008;28(5):621-38. PMID: 18591542
  14. Agency for Healthcare Research and Quality. Methods Guide for Medical Test Reviews [posted November 2010]. Rockville, MD. http://archive.ahrq.gov/downloads/pub/evidence/pdf/chdwomtop/chdwmtop.pdf. Accessed August 22, 2011.
  15. Owens DK, Lohr KN, Atkins D,, et al. AHRQ series paper 5: Grading the strength of a body of evidence when comparing medical interventions—Agency for Healthcare Research and Quality and the Effective Health-Care Program. J Clin Epidemiol 2010;63(5):513-23. PMID: 19595577
  16. Dodi C, Cortigiani L, Masini M,, et al. The incremental prognostic value of pharmacological stress echo over exercise electrocardiography in women with chest pain of unknown origin. Eur Heart J 2001;22(2):145-52. PMID: 11161916
  17. Sanfilippo AJ, Abdollah H, Knott TC,, et al. Stress echocardiography in the evaluation of women presenting with chest pain syndrome: a randomized, prospective comparison with electrocardiographic stress testing. Can J Cardiol 2005;21(5):405-12. PMID: 15861257
  18. Shaw LJ, Mieres JH, Hendel RH,, et al. Comparative Effectiveness of Exercise Electrocardiography With or Without Myocardial Perfusion Single Photon Emission Computed Tomography in Women With Suspected Coronary Artery Disease: Results From the What Is the Optimal Method for Ischemia Evaluation in Women (WOMEN) Trial. Circulation 2011. PMID: 21844080
  19. Siegler JC, Rehman S, Bhumireddy GP,, et al. The Accuracy of the Electrocardiogram during Exercise Stress Test Based on Heart Size. PLoS One 2011;6(8):e23044. PMID: 21857990
  20. Doyle M, Fuisz A, Kortright E,, et al. The impact of myocardial flow reserve on the detection of coronary artery disease by perfusion imaging methods: an NHLBI WISE study. J Cardiovasc Magn Reson 2003;5(3):475-85. PMID: 12882078
  21. Lehmkuhl HB, Siniawski H, Lehmkuhl E,, et al. Value and limitations of dobutamine stress echocardiography in women with suspected coronary artery disease. Zeitschrift fur Herz-, Thorax- und Gefasschirurgie 2007;21(6):250-258. PMID: 2008012568
  22. Lu C, Lu F, Fragasso G,, et al. Comparison of exercise electrocardiography, technetium-99m sestamibi single photon emission computed tomography, and dobutamine and dipyridamole echocardiography for detection of coronary artery disease in hypertensive women. Am J Cardiol 2010;105(9):1254-60. PMID: 20403475
  23. Miller TD, Roger VL, Milavetz JJ,, et al. Assessment of the exercise electrocardiogram in women versus men using tomographic myocardial perfusion imaging as the reference standard. Am J Cardiol 2001;87(7):868-73. PMID: 11274942
  24. Raman SV, Donnally MR, McCarthy B. Dobutamine stress cardiac magnetic resonance imaging to detect myocardial ischemia in women. Prev Cardiol 2008;11(3):135-40. PMID: 18607148
  25. Hendel RC, Patel MR, Kramer CM,, et al. ACCF/ACR/SCCT/SCMR/ASNC/NASCI/SCAI/SIR 2006 appropriateness criteria for cardiac computed tomography and cardiac magnetic resonance imaging: a report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group, American College of Radiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, American Society of Nuclear Cardiology, North American Society for Cardiac Imaging, Society for Cardiovascular Angiography and Interventions, and Society of Interventional Radiology. J Am Coll Cardiol 2006;48(7):1475-97. PMID: 17010819
  26. Taylor AJ, Cerqueira M, Hodgson JM,, et al. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 Appropriate Use Criteria for Cardiac Computed Tomography. A Report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Cardiovasc Comput Tomogr 2010;4(6):407 e1-33. PMID: 21232696
  27. Douglas PS, Garcia MJ, Haines DE,, et al. ACCF/ASE/AHA/ASNC/HFSA/HRS/SCAI/SCCM/SCCT/SCMR 2011 Appropriate Use Criteria for Echocardiography. A Report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, American Society of Echocardiography, American Heart Association, American Society of Nuclear Cardiology, Heart Failure Society of America, Heart Rhythm Society, Society for Cardiovascular Angiography and Interventions, Society of Critical Care Medicine, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance Endorsed by the American College of Chest Physicians. J Am Coll Cardiol 2011;57(9):1126-66. PMID: 21349406
  28. Bonow RO, Douglas PS, Buxton AE,, et al. ACCF/AHA methodology for the development of quality measures for cardiovascular technology: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Performance Measures. Circulation 2011;124(13):1483-502. PMID: 21875906
  29. Patel MR, Peterson ED, Dai D,, et al. Low diagnostic yield of elective coronary angiography. [Erratum in: N Engl J Med. 2010 Jul 29;363(5):498.] N Engl J Med 2010;362(10):886-95. PMID: 20220183
  30. Patel MR, Chen AY, Peterson ED,, et al. Prevalence, predictors, and outcomes of patients with non-ST-segment elevation myocardial infarction and insignificant coronary artery disease: results from the Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines (CRUSADE) initiative. Am Heart J 2006;152(4):641-7. PMID: 16996828

Full Report

This executive summary is part of the following document: Dolor RJ, Patel MR, Melloni C, Chatterjee R, McBroom AJ, Musty MD, Wing L, Coeytaux RR, Ross AK, Bastian LA, Anderson M, Kosinski AS, Sanders GD. Noninvasive Technologies for the Diagnosis of Coronary Artery Disease in Women. Comparative Effectiveness Review No. 58 (Prepared by the Duke Evidence-based Practice Center under Contract No. 290-2007-10066-I.). AHRQ Publication No. 12-EHC034-EF. Rockville, MD: Agency for Healthcare Research and Quality. June 2012. www.effectivehealthcare.ahrq.gov/reports/final.cfm.

For More Copies

For more copies of Noninvasive Technologies for the Diagnosis of Coronary Artery Disease in Women: Comparative Effectiveness Review Executive Summary No. 58 (AHRQ Pub. No. 12-EHC034-1), please call the AHRQ Publications Clearinghouse at 1-800-358-9295 or email ahrqpubs@ahrq.gov.

Return to Top of Page