The quality, character, intensity and duration of chest pain induced by exercise may be helpful in establishing a diagnosis even in the absence of ST-segment depression. It is important to obtain an accurate account of symptoms, in particular, chest pain that occurs during or after exercise testing.

Cardiac examination immediately after exercise may reveal a precordial bulge or third heart sound resulting from left ventricular dysfunction. Papillary muscle dysfunction due to transitory ischemia may result in a mitral regurgitation murmur.

Blood pressure is dependent on cardiac output and peripheral resistance. Systolic blood pressure rises to 160–220 mmHg at peak exercise, whereas diastolic blood pressure does not usually change by more than ±10 mmHg from the resting value in normal subjects. An inadequate rise (<120 mmHg) or a progressive fall in systolic blood pressure (>10 mmHg) during exercise may be associated with severe CAD and ischemic dysfunction (Table 8.3). However, some normal subjects may have a transient drop in systolic blood pressure shortly after maximal exercise due to a mismatch between the rapid changes in cardiac output and autonomic tone that occur in the immediate post-exercise phase.

An inappropriate increase in the heart rate with exercise may be observed in patients with impaired left ventricular dysfunction. Alternatively, this response may be observed in patients who are physically deconditioned, hypovolemic, anemic, or in atrial fibrillation. A blunted heart rate response may be a sign of chronotropic incompetence, a variable associated with an increased risk of cardiac events in higher clinical risk patients (Fig. 8.1). Chronotropic incompetence is usually defined by inability to achieve 85 % of age-predicted maximum in the absence of beta-blocker therapy. Alternatively, chronotropic incompetence can be defined by the percent of heart rate reserve achieved (peak heart rate-rest heart rate)/220-age-rest heart rate). Values <80 % are considered abnormal [7]. Increased heart rate is the largest contributor to the increase in peak with aerobic exercise. Thus, chronotropic incompetence often leads to symptomatic exercise intolerance [8]. The prognostic value of chronotropic incompetence has been assessed in different patient populations. A peak exercise HR >1 SD below the predicted mean was associated with decreased 5 year survival in a 5,000 patient series of asymptomatic women (Fig. 8.2) [8, 9]. Several different formulae have been proposed to estimate age-predicted maximal heart rate, a pre-requisite to assess chronotropic incompetence. For example, the Tanaka formula (APMHR = 208–0.7 × age) has been proposed for healthy subjects [8]. The Brawner equation has been proposed for patients with coronary artery disease on beta blocker therapy (APMHR = 164–0.7 × age) [8]. Another formula to assess inadequate heart rate reserve is the Wilkoff chronotropic index, an index that can be used when the patient does not attain 80 % of APMHR. The formula is as follows:

Correction of chronotropic incompetence to reduce cardiac events using therapy such as rate-adaptive pacing has not been adequately studied.

Occasionally, the heart rate response in the early post-exercise phase takes longer to decelerate. This form of abnormal heart-rate recovery is defined by a heart rate that fails to decrease by at least 12 beats per minute (bpm) in the first minute postexertion using a 2-min cool-down or by 18 bpm if the exercise test is stopped abruptly. Abnormal heart–rate recovery (HRR) is associated with increased long-term mortality even after adjustment for exercise–induced ST segment depression, angina, coronary disease extent, left ventricular dysfunction, and Duke treadmill score (DTS) (Fig. 8.3) [10, 11]. Some patients with abnormal HRR can be restored to normal HRR with conditioning (e.g. cardiac rehabilitation programs) resulting in survival rates similar to those with a normal baseline HRR response [12].

During exercise, there are changes in virtually all components of the ECG in normal subjects, including progressive shortening of the PR, QRS and QT intervals. The amplitude of the P wave increases and the amplitude of the T wave decreases during exercise. There is also a progressive rightward shift in the QRS axis. R wave amplitude changes have also been noted to occur during exercise. J-point depression may occur, particularly in older subjects, but the ST segment is rapidly upsloping (Fig. 8.4). Myocardial ischemia most often manifests itself as ST-segment shifts. To determine the magnitude of ST segment shift, the ST level should be measured relative to the PR segment. The PQ junction is often chosen as the point of reference (Fig. 8.5). The J point is then identified and the ST segment level is measured 60–80 ms after the J point, depending on heart rate (ST 60 is usually measured at heart rates >130–140/min). When the baseline ST-segment is depressed, an additional 0.1 mV (1 mm) of flat or downsloping ST segment depression is required to call the test “abnormal”. The standard criteria for abnormal ST segment depression are horizontal or downsloping ST-segment depression of 0.10 mV (1 mm) ≥60–80 ms after the J point in three consecutive beats (Fig. 8.6). In patients with early repolarization on the rest ECG manifest by J-point elevation, the magnitude of exercise-induced ST-segment depression should be determined using the PQ junction as the reference baseline. Thus, if the J-point at rest in a particular lead was +0.05 mV above the PQ junction, an additional −0.15 mV during or post-exercise would be required (i.e. –0.1 mV below the PQ junction), to be considered abnormal. A slow, upward sloping (>0.7–1 mV/s) ST-segment depressed 0.15 or more at 80 ms after the J point in three consecutive beats is also considered by some to be a manifestation of exercise-induced myocardial ischemia, particularly in patients with a high pre-test risk of disease (Fig. 8.5).

In most laboratories, the exercise ECG is printed out in a 3 × 4 format or 2.5 s/3 lead group. To maximize the opportunity of meeting the above criteria, some laboratories have adopted a strategy of recording a full 10 s of ECG data with leads II, aVF, V5 when the tracing starts to show abnormal beats close to the ischemic threshold. A 10 as opposed to a 2.5 s tracing increases the likelihood of capturing consecutive abnormal beats with a stable baseline (Fig. 8.7). It will require future study to determine if 3 consecutive abnormal beats for the majority of beats in a 10 s study has the same diagnostic accuracy as 3 consecutive beats in a 3 × 4 format. Coronary artery disease severity is usually greatest in the presence of downsloping ST segment depression, followed by horizontal ST segment depression and finally slow-upsloping ST segment depression (Figs. 8.5 and 8.8). The magnitude of the ischemic response is related to coronary disease severity and extent. Patients with early onset of myocardial ischemia manifest as typical angina or an ST segment shift at low exercise workloads with an “ischemic” pattern that persists late into the recovery phase, have an increased risk of multivessel disease and future cardiac events. Rapid recovery of ST-segment depression (defined as recovery to baseline within 1 min in the recovery phase) is associated with less severe and extensive myocardial perfusion abnormalities compared to patients that have more prolonged recovery of ST-segment depression [13].

ST-segment elevation may also indicate myocardial ischemia. Exercise-induced ST-segment elevation ≥1 mm occurring in a non-infarct territory usually indicates transmural ischemia and is associated with severe proximal coronary disease or coronary vasospasm. When ST-segment elevation >0.1 mV occurs in lead aVR, the likelihood of left-main or proximal left anterior descending (LAD) disease is increased [14]. While ST-segment depression does not localize the site of coronary ischemia, the localization of ST-segment elevation is relatively specific for the distribution of the coronary artery involved. ST-segment elevation ≥1 mm during exercise in Q wave leads from a prior myocardial infarction (MI) is often associated with more extensive wall motion abnormalities than patients without this finding; in most patients, this type of pattern does not represent an ischemic response (Fig. 8.9).

T wave inversion, usually noted in the post-exercise phase, is not diagnostic of myocardial ischemia in the absence of ST segment depression criteria. The same is true for pseudonormalization of T waves (i.e. inverted T waves that become upright with exercise).

A major indication for exercise testing is to estimate the prognosis and determine the need for subsequent workup that may include cardiac catheterization and coronary revascularization. If the posttest likelihood of a cardiac event, estimated using clinical and exercise test variables cannot be reduced by a coronary revascularization procedure (e.g. estimated 5-year risk of a cardiac event after exercise testing <0.8 %/year), then there is little need to proceed further. Exercise test variables associated with an increased likelihood of multivessel disease and future cardiac events include early onset of ischemia (Figs. 8.6 and 8.8), angina, or complex ventricular arrhythmias at low exercise workloads, inability to complete low levels of work (i.e., functional capacity <5 METs), extensive and profound exercise-induced ST segment abnormalities, hemodynamic instability such as a drop in systolic blood pressure during exercise, abnormal heart rate response (Fig. 8.1), and persistence of ischemic changes late into the recovery phase (Fig. 8.8, Table 8.3).

Functional capacity is one of the most important prognostic exercise test variables since it provides a global estimate of cardiac and pulmonary reserve. Patients that have the onset of ST segment depression at high exercise workloads (e.g., >10 METs) have a favorable prognosis regardless of whether or not abnormal ST segment depression is present [2]. Myers et al. [18] reported a 12 % increase in survival for every 1 MET increment in exercise capacity in a 6,213 patient cohort followed for a mean of 6 years (Fig. 8.14). Marshall et al. [19] reported an annual cardiac event rate of <0.5 %/year in 516 patients that were able to complete 9 min on a Bruce protocol. Cardiac event rates at these high workloads were similar regardless of the results of the exercise ECG or myocardial perfusion scan. The risk gradient for future cardiac events is greatest in individuals who cannot perform exercise, followed by those that are unable to complete the exercise protocol. The lower the exercise capacity, estimated in METs, the greater the risk. Patients with exercise-induced ischemia or hemodynamic compromise (e.g., hypotension) at low workloads (i.e., <5 METs) should be referred for cardiac catheterization and revascularization if there are no clinical contraindications. The likelihood of angiographic high-risk anatomy is high, and if present, would be an indication for coronary revascularization regardless of symptomatic status.

Exercise testing should never be categorized as positive or negative based solely on the ST segment response. The entire variable set that includes final exercise capacity, hemodynamic and heart rate responses, symptoms, and workload at which symptoms or the ischemic response become manifest is critical in order to provide the best prognostic estimates possible. Mark and colleagues [20] devised a treadmill score and applied it to 2,842 symptomatic inpatients who were being evaluated by treadmill exercise testing and cardiac catheterization. The treadmill score was calculated on the basis of the duration of exercise, the amount of maximal electrocardiographic depression, and the presence or absence of angina occurring during exercise. The authors found that the treadmill score was a better discriminator than clinical assessment alone and appeared to better distinguish patients who subsequently died from those who lived. Approximately one third of patients had scores indicating low risk (≥ +7) with a 93 % 5-year survival rate (i.e., annual mortality rate of 1.4 %). High-risk scores (≤ −11) occurred in 13 % of patients; the mortality rate in this group was 3 times greater, with a 67 % 5-year survival rate (i.e., annual mortality rate of approximately 7 %).