Treadmill exercise testing is a standard exam for the detection of myocardial ischemia and ST-segment depression during exercise is a positive criterion for ischemia. However, this exam presents low sensitivity (68%) and low specificity (71%) for the detection of significant epicardial coronary artery disease.1 False positive tests are generally explained by mechanisms such as abnormal repolarization on the baseline ECG (low serum potassium levels, left ventricular [LV] hypertrophy [LVH], valvular disease and digitalis effect, among others) or excessive increase in blood pressure during exercise.2–4

Another possible explanation is the development of intraventricular gradients (IVG) during exercise.5 This phenomenon can be identified during stress echocardiography,6 a test with higher sensitivity (85%) and specificity (86%) for the detection of epicardial coronary artery disease (CAD).1 IVG develop during dobutamine stress echocardiography (DSE) in 8-38% of non-selected populations and are often described as having no clinical impact.7–15 However, the prevalence of IVG has never been assessed in a selected population of patients with positive treadmill exercise test and without significant CAD.

• 1.

  To assess the prevalence of IVG during DSE in a selected population of patients with a false positive treadmill exercise test;

• 2.

  to determine whether this prevalence is different from that described for non-selected populations;

• 3.

  to determine whether patient characteristics (gender, age, risk factors for CAD, treatment with beta-blockers or calcium antagonists and significant valvular disease or LVH) are associated with the presence or absence of IVG;

• 4.

  to assess the relation between the presence of IVG and the occurrence of ECG changes, symptoms and blood pressure response to stress.

The clinical data of 50 patients included in the study are shown in Table 1. The population included 36 men and 14 women, aged between 31-79 years (mean 55) with a mean body mass index of 24.8. In this population only 18% of patients were asymptomatic; 66% presented fatigue, 32% had typical chest pain and 42% presented atypical chest pain. According to our data, 78% of the patients had risk factors for CAD: 24% had one risk factor, 36% two risk factors and only 4% (two patients) had three risk factors. Only three patients were under treatment with beta-blockers and none were taking calcium channel antagonists (Table 1).

Table 2 shows the results of the ECG, conventional echocardiogram and dobutamine stress echo.

The baseline ECG was normal in all but one patient (patient 3), who presented LVH (Sokolow-Lyon index 37mm). The standard echocardiogram (performed before the dobutamine stress echo) also did not show relevant abnormalities in most patients (Table 2). Remarkably, in all patients the LVOT diameter was normal; none of the patients had SAM or LVOT gradient at rest; none of the patients had septal hypertrophy and only one (patient 37) showed concentric LVH (LV mass index 141g/m2, relative wall thickness 0.46, interventricular septum thickness/posterior wall thickness 1.1); and in all patients LV cavity dimensions and volumes were normal, as were conventional indices of global systolic function. Estimation of LV filling pressures, as well as of right atrial pressure (inferior vena cava diameter and respiratory variation), showed normal results in all patients.

During dobutamine stress echocardiography none of the 50 analyzed patients developed echocardiographic signs of myocardial ischemia and the dobutamine stress test was considered negative for ischemia in all of them. However, 34 patients (68%) developed a new IVG in the LVOT, while the other 16 (32%) had no stress-induced IVG (p<0.05). The mean gradient developed during DSE was 99.32mmHg, ranging between 50-200mmHg (Table 2). The prevalence of IVG in our study (68%) was significantly higher than that described for non-selected populations (8-38%) (p<0.001).

We also evaluated whether certain factors (gender, age, risk factors for CAD, and significant valvular disease or LVH), usually considered predictors of the development of IVG, were associated with their development during DSE. According to our data, none of these factors was statistically associated with the development of IVG (p>0.05 for all of them). Moreover, we assessed the relation between new IVG and the occurrence of symptoms (dyspnea or angina) and blood pressure levels during DSE. We found that only three patients had symptoms, and no patient had hypotension, during DSE. Therefore neither the occurrence of symptoms nor blood pressure levels showed any significant relation with the development of LVOT gradients (p>0.05) during DSE.

However, the occurrence of ECG changes with positive criterion for myocardial ischemia during DSE was significantly associated with the development of IVG (p<0.05).

The presence of ventricular gradients during DSE has traditionally been considered a dobutamine-specific effect, with no physiological or clinical significance,19 and not reproducible during physical exercise.7,20

Our study shows that, in a selected population of false positives on treadmill stress testing, a very large proportion of patients (more than two-thirds) develop IVG during DSE. Though this prevalence (68%) is significantly higher than that described for non-selected populations (8-38%), the reasons for this difference and a plausible mechanism to explain it are not easy to find.

However, taking into account the results of our study, it seems reasonable to presume that the two phenomena – ST depression in the treadmill exercise test and IVG during DSE – may be linked, perhaps by a cause-effect relation, because during DSE the presence of IVG was also associated with ST-segment depression.

So, in our opinion, the development of intraventricular gradients during stress may not be only a dobutamine-specific effect but may also occur during exercise, which would explain the results of the treadmill exercise tests. The development of IVG during stress (dobutamine or exercise) would increase afterload, with subsequent pressure overload within the LV cavity and increased wall stress. This increased wall stress would decrease subendocardial perfusion, with ECG changes characteristic of subendocardial ischemia21–25 and in some cases with symptoms of dyspnea or chest pain.26

Thus, we propose that the intraventricular gradient itself causes subendocardial ischemia22,27,28 with ST-segment depression, both in the treadmill stress test and during DSE, without new wall motion abnormalities.

Some unexpected findings of our study are the lack of a significant relationship between induced outflow tract gradients and symptoms and blood pressure reduction during DSE. However, the decubitus (non-orthostatic) position of patients during DSE (helping to maintain adequate preload and cardiac output) and the typical late systolic occurrence of these gradients (when most of the LV stroke volume has been ejected from the left ventricle into the aorta) may explain these results.

Finally, the lack of a significant relation between the conventional predictors of IVG29–31 and the development of LVOT gradients in our study will need to be confirmed in future studies.

Our previous explanation for the association between IVG and ST depression in the exercise test, though conceptually attractive, should be confirmed in clinical settings, with patients undergoing exercise stress echocardiography to assess IVG before, during and after exercise.32,33 This will show whether the IVG induced during DSE is reproducible during exercise, and if so, it will confirm our theoretical explanation for the ST depression. Preliminary and still unpublished results from our center suggest that dobutamine-induced IVG are in some cases reproducible during exercise, though the magnitude of the gradient is lower than with dobutamine.

The reasons for this quantitative difference undoubtedly lie in the fact that dobutamine and exercise share some common features but also have different hemodynamic effects. Though they both have positive chronotropic and inotropic effects,34 dobutamine induces peripheral vasodilatation,27 while physical exercise induces vasodilatation of the arteries of the exercising muscular territory but visceral arterial vasoconstriction, which may prevent the development of IVG.7,20 On the other hand, orthostatism itself leads to a decrease in venous return and preload with a subsequent increase in intraventricular gradients, mainly during the recovery period (when the muscular pump of the lower limbs is absent, no longer preventing the decrease in preload).35

Finally, the decision to perform another diagnostic test after a negative scintigraphic study or after CT coronary angiography without significant stenosis was taken because most of our patients were still symptomatic and without diagnosis after these tests. As the sensitivity of these methods is not 100% we felt it clinically appropriate to perform another exam that is inexpensive, non-invasive and radiation-free and that could not only rule out ischemia but also provide another explanation for the clinical and treadmill test findings: dobutamine stress echocardiography.

Our definition of a false positive treadmill exercise test, based on the normality of different tests (anatomical and functional), contributed to the non-homogeneity of the false positive population and is a limitation of this study. Future studies on this topic should define false positive patients based only on a single functional or anatomical test (even though none of them has 100% sensitivity and specificity).

Another limitation lies in the fact that we did not study a control group of individuals with negative exercise treadmill test in order to compare the proportion of patients with IVG in these patients with those with positive exercise test. However, we used the figures of non-selected populations taken from a number of studies in the literature.7–15

Finally, and as discussed above, exercise echocardiography should be performed in these patients, to assess the reproducibility of our results under physiological conditions.

This study shows that LV outflow tract gradients during dobutamine stress echocardiography in a selected population of false positives of treadmill stress test occur in more than two-thirds of patients, a significantly higher prevalence than that described in non-selected populations.

The presence of these gradients is associated with the occurrence of ischemic ST-segment changes, but not with symptoms or with blood pressure reduction.

The development of intraventricular pressure gradients during stress may therefore explain some of the false positive results of treadmill exercise tests. To confirm this theory, our results must be reproducible in physiological conditions, during exercise stress echocardiography.

The authors have no conflicts of interest to declare.