Stress Testing, Nuclear Imaging, Coronary CT Angiography, Cardiac MRI, Cardiopulmonary Exercise Testing

Stress Testing, Nuclear Imaging, Coronary CT Angiography, Cardiac MRI, Cardiopulmonary Exercise Testing

I. Indications for stress testing

Stress testing is indicated for diagnostic and, more importantly, prognostic evaluation of CAD. Its sensitivity is reduced in single-vessel CAD and in non-LAD disease.

A. Stable chest pain presentation

  1. Stress testing is useful for:

    • The diagnostic and the prognostic assessment of patients with intermediate CAD probability
    • The prognostic assessment of patients with high CAD probability

  2. Stable angina is defined by three features: (1) chest pain or angina equivalent, (2) occurs with exertion or stress, and (3) resolves with rest or with NTG, within 30 seconds to 5 minutes.
  3. The presence of all three features implies that angina is typical or definite. If only two features are present, angina is possible or atypical. Chest pain without any other feature is non-anginal pain. The combination of angina features, age, sex, and risk factors establishes the probability of CAD (Forrester and Duke classifications, combined in Table 33.1).
  4. According to Bayes’ theorem, stress testing is particularly useful in patients with an intermediate pre-test probability of CAD, in whom the stress test result significantly increases or reduces the probability of CAD.

    If the pre-test probability is high, stress testing, regardless of the results, will only a have minor effect on the probability of CAD; thus, stress testing is not indicated for diagnostic purposes, but may be performed for prognostic purposes and for deciding if the patient may benefit from revascularization. Extensive ischemia may warrant coronary angiography and revascularization, whereas medical therapy is an acceptable first-line option for a low- or intermediate-risk stress test.

    If the pre-test probability of CAD is low (<15%), stress testing is often not needed. Even if the stress test is positive, the probability of CAD will only increase from <15% up to 20%, which means that the stress test is likely falsely positive. However, if judged necessary, stress ECG may be performed (class IIa).

B. Acute chest pain presentation, with non-ischemic ECG and negative troponin 3 hours after symptom onset, and with no typical exertional angina at low levels of exertion.

C. Recent STEMI that was not urgently reperfused with PCI, whether thrombolysis was acutely administered or not

In the absence of angina recurrence or severe HF (conditions that would require a coronary angiogram), submaximal or symptom-limited stress testing may be performed at 4 days (before discharge) to see if there is any residual ischemia within or around the infarcted territory, whether reperfused or not, or in other areas.

Table 33.1 Clinical probability of CAD.

High probability of CAD (>85%)

  • Typical angina in older patients (age ≥40 for men, ≥ 60 for women)
  • Typical angina in patients with a combination of multiple risk factors (diabetes, smoking, hyperlipidemia, especially when all three are present)
Intermediate probability of CAD (15–85%)

  • Typical angina in younger patients (age <40 for men, < 60 for women)
  • Possible angina or non-anginal pain in older patients (age ≥40 for men, ≥ 60 for women) or in patients with multiple risk factors
Low probability of CAD (<15%)

  • Possible angina in younger patients without a combination of multiple risk factors

D. Known CAD of borderline significance on the coronary angiogram (45–70% stenosis)

In this case, stress imaging may be performed after coronary angiography to determine if the stenosis is hemodynamically significant. A significant stenosis leads to ischemia in the correspondent territory on the echo or nuclear images. Alternatively, fractional flow reserve (FFR) may be performed during coronary angiography to determine whether the stenosis is significant.

II. Contraindications to all stress testing modalities

  • Recent STEMI ≤ 2 days
  • NSTEMI (perform coronary angiogram)
  • Active HF
  • Severe symptomatic AS
  • Arrhythmias: uncontrolled AF, uncontrolled VT, second-degree Mobitz II or third-degree AV block
  • HTN >180/110 mm Hg

Additional contraindication for adenosine nuclear testing: history of severe asthma or current, decompensated COPD.

Risk of death from stress testing is <1/10,000. Morbidity (MI, arrhythmias) is 5/10,000. The risk is greater when the test is performed in the early post-MI setting.

III. Stress testing modalities

Three stress testing modalities are available:

  1. Treadmill stress ECG
  2. Treadmill stress imaging (echo, nuclear). Echo assesses ischemia (exercise-induced hypokinesis), while nuclear imaging assesses the difference in coronary flow reserve between the area subtended by a stenosis and the areas subtended by non-obstructed coronary arteries.
  3. Pharmacological stress imaging (echo, nuclear, MRI). Dobutamine, often with atropine, is used for stress echocardiography; adenosine is used to assess the difference in coronary flow reserve with nuclear imaging. Dobutamine may be used for nuclear imaging, but is less sensitive than adenosine in nuclear imaging. The cardiac workload is less with dobutamine than with exercise, and thus dobutamine echocardiography is also less sensitive than maximal exercise echocardiography or exercise nuclear imaging (See Appendix 1).

Since exercise achieves the highest cardiac workload and provides prognostic information independent of imaging (exercise time, symptoms, ECG response, BP response, and heart rate response), exercise stress testing is preferred to pharmacological testing. The selection of a stress test modality depends on three factors (Figure 33.1): (i) the patient’s ability to walk; (ii) the presence of baseline ECG abnormalities that preclude ischemic assessment on stress ECG, mainly LBBB or resting ST-segment depression >1 mm; (iii) compelling indication for treadmill stress imaging (for its higher sensitivity and specificity), such as intermediate-to-high pre-test probability of CAD or prior revascularization.

The Bruce treadmill protocol is the exercise protocol most commonly used. It consists of 3-minute stages, with an increase of speed and inclination at each stage (Table 33.2).

Schematic illustration of stress testing modality.

Figure 33.1 Stress testing modality.

*Baseline ECG abnormalities precluding stress ECG interpretation:

  • LBBB, paced rhythm, or non-specific intraventricular conduction delay ≥120 ms
  • Any ST depression ≥1 mm
  • LVH or digoxin therapy with any ST depression, even if <1 mm
  • Pre-excitation (WPW)

These conditions will routinely have worsening of the ST depression during stress and during tachycardia, regardless of ischemia (false-positive result).

Other ECG abnormalities and clinical situations:

  • RBBB or resting ST depression <1 mm ↓ specificity of stress ECG but do not preclude interpretation.
  • LVH without any resting ST depression: stress ECG remains reasonably specific for ischemia.
  • AF: patients may become quickly tachycardic or may not increase their heart rate enough if they are on rate-controlling medications. Exercise testing, is, however, appropriate and assesses the adequacy of rate control.
  • History of PCI or CABG: stress imaging is preferred to stress ECG. Stress ECG is less sensitive in this context and cannot localize ischemia. Stress ECG is poorly sensitive for the diagnosis of single-vessel restenosis.

Table 33.2 Standard Bruce treadmill protocol.

Stage Inclination (°) Speed (mph) MET
1 10 1.7 5
2 12 2.5  7
3 14 3.4 10
4 16 4.2 13
5 18 5 16

In patients who are not able to cope with the workload of a standard Bruce protocol (e.g., elderly, weak), a modified Bruce protocol may be performed. Stages 1, 2, and 3 of the modified Bruce protocol consist of an inclination of 0, 5, and 10° respectively, for a speed of 1.7 mph. Thus, stage 3 of the modified Bruce protocol is equivalent to stage 1 of the Bruce protocol. The energy expenditure achieved with the modified Bruce protocol is much lower than the energy expenditure of the standard Bruce protocol. A modified Bruce protocol is also called “submaximal stress testing.”

The metabolic equivalent (MET) corresponds to the energy expenditure for an activity level. The reference, 1 MET, is the resting metabolic rate obtained during quiet sitting and corresponds to the consumption of 3.5 ml/kg/min of O2. METs and O2 consumption increase with the peak exercise achieved.

IV. Diagnostic yields and pitfalls of stress ECG and stress imaging

A. Diagnostic yield of exercise stress ECG

For the exercise stress ECG or imaging to be valid, the patient should:

  1. Reach target HR, which is 85% of the maximal HR (=85% of [220 – age])


    Achieve a good workload ≥6–7 METs, which means finish 5 minutes of the Bruce protocol.

  2. Or attain high-risk criteria before this goal.

A sufficient workload is necessary to produce ischemia or nuclear hypoperfusion in the presence of a hemodynamically significant CAD. Otherwise, the test is less sensitive, and pharmacologic stress imaging may be performed. Note, however, that with exercise stress imaging the sensitivity is still acceptable if the patient partially achieves the heart rate or workload goals, because stress imaging is more sensitive than stress ECG. One study suggested that even with stress ECG, a negative test at a submaximal heart rate was still predictive of a low event risk if ≥7 METs are achieved.2

Schematic illustration of during sinus tachycardia, atrial repolarization becomes accentuated and extends over the ST segment, leading to PR depression + junctional ST depression.

Figure 33.2 During sinus tachycardia, atrial repolarization becomes accentuated and extends over the ST segment, leading to PR depression + junctional ST depression. Referencing the ST segment to the PR junction accounts for the effect of atrial repolarization.

Achieving the appropriate heart rate is even more important with dobutamine stress echocardiography than exercise stress echocardiography, as dobutamine induces less cardiac workload and thus less ischemia than exercise.

B. Positive stress ECG

A positive stress ECG is defined as descendant (downsloping) or horizontal ST depression ≥1 mm measured at 60–80 ms past the J point, during exercise or within 3 minutes of recovery in at least one lead. While ST displacement should generally be measured relative to the TP segment, PQ or PR junction is chosen as the isoelectric point during exercise (Figure 33.2). If baseline ST depression is present, the exertional worsening of ST depression, rather than the absolute ST depression during exercise, is used to assess ischemia. If baseline ST elevation is present, exertional ST depression is measured in absolute values.

The lateral precordial leads (especially V5) are the most specific and most sensitive for exercise-induced ischemia. Isolated inferior changes are less specific and are often a false-positive result.

ST depression occurring during exercise usually persists ≥3 minutes into recovery. 3 If it does not, the ST depression is considered less specific; in particular, quick recovery of ST depression in <1 minute was predictive of a low risk of cardiac events and CAD, as low as a negative ECG result.2

About 10% of patients develop the diagnostic ST depression only during recovery, most typically and most specifically during the first 3 minutes of recovery,3 sometimes preceded by equivocal or upsloping ST depression at peak exercise. This recovery-only ST depression is as predictive of CAD and coronary events as ST depression that starts at peak exercise.4

Slow upsloping ST depression (≥1.5-2 mm at 80 ms) that does not become horizontal or downsloping in recovery is non-specific; its diagnostic yield differs between studies, yet it has a value in patients with a high pre-test likelihood of CAD. In the landmark DTS study, only horizontal or downsloping ST depression was used in DTS calculation.

C. High-risk positive stress ECG (see Table 33.3)

Table 33.3 High-risk stress ECG and Duke Treadmill Score (DTS).

  1. Early ST depression or severe angina at stages 1 or 2 (at <5 METs)
  2. ST depression ≥2 mm, especially if more than five leads
  3. ST elevation in leads without prior Q waves (→ signifies transmural ischemia, very high risk)
  4. Sustained reduction of SBP >10 mmHg in comparison with the baseline, inability to increase SBP beyond the baseline with progressive exercise
  5. Arrhythmias:

    • Sustained VT
    • Non-sustained VT or complex PVCs (couplets, triplets, polymorphic) associated with other ischemic signsa

  6. Heart rate:

    • A decrease in HR of <12 bpm in the first minute of recovery is a strong predictor of cardiovascular mortality (implies lack of vagal reactivation soon after exercise)
    • Inability to achieve 85% of maximal heart rate (chronotropic incompetence) not only reduces stress test sensitivity, but also implies an increased mortality (because of autonomic dysfunction and an association with twice the risk of CAD and perfusion defects).5 A similar prognostic value is found for a low chronotropic reserve index: (peak HR – resting HR)/([220-age] – resting HR) (low if <0.8, or <0.62 if on β-blockers)
Duke Treadmill Score b

= Exercise time on Bruce protocol – 5 × (the most severe ST depression) – 4 × (angina score)

  • Score ≥+5: Low cardiovascular risk (<1% mortality and cardiac events per year)
  • Score +4 to –10: Intermediate cardiovascular risk (1–3% mortality, 1–5% cardiac events per year)
  • Score ≤ –11: High cardiovascular risk (>3% mortality, > 5% cardiac events per year)

a Exercise-induced NSVT or complex PVCs portend an increased mortality in the case of underlying CAD or associated ischemic ST changes or multifocal PVCs on mild exertion. Exertional PVCs are common in normal subjects (up to 5%), and even NSVT is benign in subjects with no heart disease. Monomorphic VT may be a form of idiopathic VT initiated by exercise (e.g., RVOT VT), ARVD, or may in fact be SVT with rate-related bundle branch block.

Post-exertional frequent PVCs may be seen with or without heart disease but portend a stronger prognostic value than exertional PVCs in heart disease, as they imply a lack of vagal reactivation soon after exercise.

b Exercise time is the time spent on a standard rather than modified Bruce protocol; 9 minutes on a modified Bruce protocol does not give a score of 9, but is equivalent to 3–4 minutes on a standard Bruce protocol. Angina score is 0 if no angina, 1 if non-exercise-limiting angina, 2 if exercise-limiting angina

Note: blood pressure response during exercise

Normally, SBP increases with exercise (high stroke volume) while DBP remains unchanged or decreases (vasodilatation). SBP may decrease during progressive exercise after an initial peaking from the catecholaminergic surge, but it does not decrease below baseline.

Exertional hypotension may be due to:

  • Extensive ischemia.
  • Cardiomyopathy: the lack of contractile reserve in systolic HF, or the lack of filling reserve in diastolic HF, prevents the increase in stroke volume. If cardiac output cannot increase and fill the dilated systemic space induced by exercise, BP decreases (BP = [CO × SVR]: if SVR ↓ but CO cannot ↑, BP ↓).
  • AS or HOCM: LVOT obstruction prevents the exertional increase in stroke volume.
  • Hypovolemia or antihypertensive drugs.

On the other hand, post-exertional hypotension is usually benign and is related to the sudden reduction of venous return and cardiac output while the systemic space is still dilated; also, the empty, hypercontractile ventricle may trigger a vasovagal response. Post-exertional hypotension may also be seen with severe AS or HOCM upon sudden cessation of activity (i.e., sudden decrease in venous return, AS and HOCM being very sensitive to preload reduction).

In hypovolemic patients, dobutamine may induce hypotension as it reduces diastolic time and preload; this paradoxically reduces stroke volume in patients who are preload dependent, on the upslope of the Starling curve. Conversely, exercise increases venous return and preload and does not cause hypotension, except in cardiomyopathies or extensive CAD.

An increase in SBP to >214 mmHg, or an increase in DBP, may imply abnormal systemic vasoreactivity and a risk of future HTN.

D. Limitations of DTS and exercise stress ECG: value of stress imaging

Exercise stress ECG has a low sensitivity for CAD detection, especially in patients with single-vessel non-LAD disease. While it is more sensitive in patients with more extensive CAD, a significant proportion of these patients is still missed. In the Duke stress testing database of symptomatic patients, most of whom had definite or possible angina (two or three angina features), a low-risk DTS was still associated with a ~10% risk of left main and/or three-vessel CAD, and a ~10% risk of two-vessel or proximal LAD disease.6 Thus, a high-risk subgroup is concealed within the low DTS group and is often picked up by nuclear imaging.79 Also, ~12% and 55% of patients with three-vessel CAD had a low and intermediate DTS score, respectively. However, a very low-risk DTS with over 10 METs of exercise capacity (8 minutes of Bruce protocol) and no ST depression has been associated with an extremely low prevalence of moderate or severe ischemia on nuclear imaging (<2%).10

Importantly, the high-risk DTS had a definitive diagnostic value, with a 99% risk of significant CAD (slightly less in women), and a 75% risk of left main or three-vessel CAD.6

Table 33.4 Sensitivity and specificity of various stress tests

Test Sensitivity Specificity
Stress ECG 65% 70% (lower in women)
Stress echo 75–80% (lower with dobutamine) 85%
Stress nuclear 80–85% 70% (lower in women)

For all these tests, the sensitivity is higher in left main or three-vessel CAD. Also, the prognostic value is superior to the diagnostic value.

In addition to sensitivity, stress ECG has specificity limitations, particularly in women. In symptomatic patients, ST depression has a good positive predictive value of ~75% in men, but only ~50% in women.11 In fact, exertional ST depression is common in women in general, including asymptomatic women. In asymptomatic women, ST depression is encountered on ~5% of stress tests and does not, per se, affect the long-term prognosis, even when ST is depressed >2 mm.1214 This is particularly related to the lower pre-test probability, but also the higher prevalence of microvascular dysfunction in women, and a digoxin-like effect of estrogen on the ECG. Yet, the lack of ST depression retains a good rule-out value in women, similar to men (~75-80%); the presence of ST depression retains a prognostic and diagnostic value in symptomatic women with an intermediate to high pre-test probability. Also, DTS and exercise parameters maintain a prognostic value in women similar to that in men.

E. Stress echocardiography

Normally, the myocardial segments become hypercontractile with stress, meaning the myocardial thickening and excursion increase with stress. An ischemic response is characterized not only by a lack of hyperkinesis, but by a paradoxical worsening of contraction in comparison to baseline: a normal myocardial segment at rest becomes hypokinetic or akinetic with stress, a hypokinetic segment becomes akinetic or dyskinetic. The change from akinesis to dyskinesis is the only deterioration that does not, by itself, imply ischemia. Also, isolated hypokinesis of the inferobasal or inferoseptal segments is frequently seen in normal individuals and is not diagnostic of ischemia.

Two other responses are taken into account but are less specific for ischemia (their value depends on the pre-test probability of CAD):

  1. Lack of hypercontractile response. This may be related to ischemia, but is often related to inappropriate workload, severe HTN, underlying cardiomyopathy, imaging in early recovery after heart rate has declined, prior β-blocker therapy, or older age.
  2. Tardokinesis of a myocardial segment, meaning delayed contraction of a segment.

The following cases scenarios complicate ischemic assessment with stress echo:

  1. Baseline hypokinesis: ischemia may manifest as worsening of hypokinesis or extension of hypokinesis to adjacent segments. However, this adjacent hypokinesis may be missed if the adjacent segments are tethered by the normal myocardium, or it may be overcalled if the adjacent segments are tethered by the dysfunctional myocardium. Nuclear testing is preferred in patients with baseline wall motion abnormality because of its higher sensitivity.
  2. LBBB or abnormal postoperative septal motion: the assessment of myocardial thickening, rather than excursion, of the anterior wall and apex provides appropriate diagnostic yield.1,19

F. Nuclear myocardial perfusion imaging (MPI), using single photon emission computed tomography (SPECT) (see also Appendices 1 and 2)

While having an excellent prognostic value, superior to stress ECG and overruling low or intermediate stress ECG results, SPECT MPI is flawed by a small but significant risk of missing multivessel disease. The brightness of the nuclear uptake is not an absolute radioactive count, it is comparative to the pixel with the highest radioactive count: resting segments are compared to other resting segments, while stress segments are compared to other stress segments. This means that in diffuse balanced ischemia, in which all the myocardium is equally ischemic (severe triple-vessel CAD), nuclear counts are equally low, which makes all pixels appear bright and the images appear normal. Since one area is usually more ischemic than the others, the more common caveat is that only one area looks ischemic, while the others look normal despite being ischemic. Thus, nuclear testing may miss assessing the true extent of ischemia in multivessel disease. A patient with severe LAD stenosis, but more severe RCA stenosis, may appear to have only an inferior defect, because the anterior wall is less ischemic and may thus look bright.

In one study, despite angiographically severe three-vessel CAD, SPECT MPI showed no defect in 18% of patients and a single-vessel disease pattern in 36% of patients.20 A nuclear substudy of the FAME trial found that in patients with angiographic multivessel disease, ~50% of vessels with FFR <0.80 were not identified on nuclear imaging, and 34% of patients with ischemia by FFR had a negative nuclear scan.21 Thus, in multivessel disease, the lack of defect in one territory on nuclear imaging does not imply the lack of ischemia in that territory, and a completely normal scan is not uncommon. A high pre-test probability of CAD, chest pain during exercise testing, abnormal ECG or BP response, transient ischemic dilatation (TID), and EF on gated SPECT allow the diagnosis.

TID is the ratio of LV volume at stress compared to rest, the LV volume used being time-averaged from both systole and diastole using the non-gated perfusion images (ECG-gated LV volumes may be used with more weight provided for the systolic volume than the diastolic volume).22,23 A post-exertional TID (ratio >1.2) proved highly specific (95%) for severe and extensive CAD, with a higher sensitivity than perfusion imaging in extensive CAD (71% vs. 33%).22

Nov 27, 2022 | Posted by in CARDIOLOGY | Comments Off on Stress Testing, Nuclear Imaging, Coronary CT Angiography, Cardiac MRI, Cardiopulmonary Exercise Testing

Full access? Get Clinical Tree

Get Clinical Tree app for offline access