INTRODUCTION

A number of authors have demonstrated that patients with chest pain and negative myocardial perfusion scintigraphy results have a good prognosis with <1% per year mortality and incidence of infarction.1-10 However, exceptions do appear in the literature and others report a history of ischemic heart disease,11,12 diabetes in women,12 low exercise heart rate,11 cardiomegaly,13 and sometimes a correlation with positive exercise electrocardiogram (ECG).14

To determine the prevalence of negative myocardial perfusion single-photon emission computerized tomography (SPECT) results associated with positive exercise ECG and its significance in terms of coronary angiography, we reviewed all SPECT studies performed in our center over a period of 8 years.

PATIENTS AND METHODS

We conducted a retrospective evaluation of 7350 myocardial perfusion SPECT studies performed in our center between 1994 and 2002. In total, 294 (4%) patients had normal SPECT and positive exercise ECG (horizontal or descending ST-segment depression >=1 mm at 0.08 s after the J-point) (Table 1). We excluded patients with possible false-positive ECG (76 with abnormal baseline repolarization, 34 with intraventricular conduction disturbances, 18 with pacemakers, and 41 taking digitalis) and revascularized patients (n=59), leaving 66 with normal baseline ECG, normal SPECT and positive exercise ECG. For the present study, we selected 33 patients referred for coronary angiography by the clinical cardiologist.

Myocardial Perfusion SPECT With Combined Techniques

All patients underwent symptom-limited bicycle exercise test. A short protocol was followed (exercise-rest on one day only) with 99mTc-tetrofosmin. Tomographic images of short, horizontal, and long vertical axis projections were obtained with an Eslcint SP4 gamma camera with high resolution collimator. All SPECT studies were considered normal after qualitative evaluation in a nuclear cardiology session involving 2 experts from each specialty. All images were subsequently reviewed for possible presence of multiple vessel disease indicated by pulmonary uptake and/or left ventricular ischemic dilatation.

Cardiac Catheterization

All patients underwent coronary angiography using Seldinger's technique at ≤6 months after SPECT, assuming no previous complications. At least 2 projections were obtained to evaluate left and right main coronary arteries.

RESULTS

Results of SPECT assessment and exercise ECG for all 7350 patients are in Table 1. We enrolled 33 patients after excluding those with possible false-positive ECG, revascularization, without coronary angiography or with coronary angiography at >6 months after SPECT, and with no complications during this period. Of these 33 patients, 26 were women (P<.000); 23 (70%) had at least one risk factor (15, hypertension; 11, dyslipidemia, 6, type 2 diabetes, and 4, smoking). Prevalence of risk factors was somewhat lower (59%) among the 33 patients without coronary angiography but this was not statistically significant.

Exercise test and coronary angiography findings for these patients appear in Tables 2 and 3. Pulmonary uptake or transient ischemic dilation of the left ventricle were not found in any patient. At exercise testing, 10 women and 1 man were not taking anti-angina medication; 11 patients (7 women) were taking beta-blockers; 11 patients (10 women), calcium antagonists; 11 patients (10 women), nitrates, and 16 patients (10 women), antiplatelet drugs. Ten of the 26 women (38.5%) and 2 of the 7 men (28.6%) presented angina during exercise testing.

Ejection fraction was normal in all patients. A total of 21 of the 26 women (81%) had normal coronary angiography, compared with 1 of the 7 men (14%) (P=.004). One of the 5 women with coronary heart disease had 40% stenosis of the left main coronary artery and 1 had 2-vessel disease. One of the 6 men with coronary heart disease had 50% stenosis of the left main coronary artery and 4 had multivessel disease. None of the patients presented collateral circulation.

DISCUSSION

Prevalence of normal myocardial perfusion SPECT and positive exercise ECG is low: in our experience around 4%, similar to the 2%-3.5% range described elsewhere10,15,16. When patients with possible false-positive ECG and revascularization are excluded, prevalence is <1%,14,16 coinciding with our findings (0.9%).

As indicated previously14,15, we have demonstrated that the pattern of "normal SPECT with positive exercise ECG" is principally found in women (26/33, 79.6%), in most of whom it is associated with angiographically normal coronary arteries (81% vs 14% of men; P=.004). These results are similar to findings reported by He et al16 in the only publication specifically to evaluate coronary angiography results in 52 patients with normal 201Tl SPECT scans and very positive exercise ECG (horizontal or downsloping ST-segment depression ≥2 mm). He et al found 80% of men had significant coronary stenosis by comparison with only 24% of women (P<.0001).

One possible cause of false-negative myocardial perfusion SPECT studies may be the presence of diffuse coronary heart disease with homogeneous ischemia of the entire left ventricle, perfusion images of which fail to identify any single region with comparatively more reduced uptake. We found most patients (7/11) with coronary heart disease had multivessel disease or disease of the left main coronary artery, whereas the remaining 4 presented non-critical one-vessel disease. These results also coincide with He et al16 who report 50% of patients with coronary heart disease presented significant stenosis of more than 1 vessel.

The low prevalence of patients with normal SPECT and positive ECG, greater presence of women, and high percentage of normal coronary angiographic findings in women, explain the generally good prognosis for patients with negative myocardial perfusion SPECT, independently of exercise ECG results. However, it is clear that severe coronary heart disease may be present when normal SPECT is associated with positive exercise ECG, suggesting this should be considered an indication for coronary angiography.

Correspondence: Dr. J. Candell Riera.
Servicio de Cardiología. Hospital Universitari Vall d'Hebron.
Pg. Vall d'Hebron, 119-129. 08035 Barcelona. España.
E-mail: jcandell@vhebron.net