Myocardial perfusion scintigraphy: interpretation of a normal scan

Although a number of imaging modalities are available to non-invasively evaluate patients with suspected or known coronary artery disease (CAD), myocardial perfusion scintig-raphy (MPS) with single photon emission computed tomog-raphy (SPECT) has been the workhorse for this purpose for over decades [1]. Notwithstanding its high sensitivity to detect ischaemia (85–90 %), a caveat of a normal perfusion scintigram is the fact that it can also be compatible with balanced ischaemia due to multi-vessel or left main disease [2]. (Fig. 1). A normal scan must therefore be interpreted with care, and a further refinement in risk stratification is warranted even in the presence of a seemingly reassuring result. In this issue of the Netherlands Heart Journal, Bom et al. report on the prognostic value of a normal MPS during a 2-year follow-up period in 762 patients without a prior history of CAD [3]. The event rate was low (4.2 %) and predominantly driven by revascularisation. Cardiac death and non-fatal myocardial in-farction occurred in only nine patients (1.2 %). These results are consistent with pooled analysis from large databases comprising close to 40,000 patients, which yield an annualised event rate of 0.6 % following a normal perfusion SPECT scan [4]. Multivariate analysis by Bom and colleagues revealed that male gender, a positive stress ECG, and a reduced left ven-tricular (LV) ejection fraction (<45 %) had a negative impact on prognosis. Some other well-documented prognostic risk factors such as age, type of stressor (adenosine vs. exercise), renal failure, and diabetes failed to reach statistical significance , which is probably attributable to the relatively limited sample size of the current study. These data highlight that a normal test result must be interpreted in light of the clinical risk profile of the patient, i.e. the pre-test likelihood of disease. According to Bayes' theorem, a negative test should alert us when pre-test probability is high as it does not rule out disease as it does when pre-test probability is low. Besides clinical risk factors, alternative imaging parameters may also aid in further risk stratification of a normal MPS. Transient ischaemic dilation (TID) has been linked to extensive CAD due to post-stress LV dysfunction, as a result of stunning and/or diffuse subendocardial hypoperfusion, which mimics LV enlargement by reduced subendocardial tracer uptake during stress. Abidov et al. explored the prognostic value of TID in patients with a completely normal MPS and identified TID (with an optimal stress-to-rest …