In the present study we investigated the accuracy for determining the need for a rest study when using NC images or NC + AC images. We found that adding AC images were superior to using only NC images. When using NC + AC compared to NC images, 214 more patients (17% of the study population) could avoid a rest study. The advantages of being able to reduce the number of rest studies are substantially reduced radiation exposure, lower costs by eliminating unnecessary imaging time and radiopharmaceutical doses, and improved laboratory efficiency by freeing up camera time to study additional patients.
In this material, when using NC images, 1.2% of the patients would have been sent home without a rest study, when a rest study was needed according to the final report. The number for AC images was 2.6% (no statistically significant difference). The reasons for this could be either that the final report was not always correct or that the physicians who evaluated the stress studies did not have access to the clinical information (as stated in the study limitations section) or that can be difficult to see small perfusion abnormalities on stress-only images in some cases. The number of missed certain ischemia/infarction due to a normal interpretation on stress-only images was very low in this study (less than 1%), but the goal is not to miss any patients with ischemia or infarction. In a study by Johansson et al. , trying to decide whether nuclear medicine technologists are able to determine the necessity for a rest study, 2.6% of the patients would have been sent home without a rest study, when a rest study was needed according to gold standard (ischemia or infarction on combined stress-rest interpretation). Thus, their results were similar to the ones found in the present study.
The stress-only approach has been investigated in several studies. Worsley et al.  demonstrated that rest images were not required if normal imaging findings had been obtained after exercise or pharmacologic stress. This was later confirmed by Schroeder-Tanka et al.  in a larger study. Heller et al.  found that AC applied to studies with stress-only 99mTc MPS significantly increased the ability to interpret studies as definitely normal or abnormal and reduced the need for rest imaging. In their study, ten experienced nuclear cardiologists independently interpreted 90 stress-only MPS in a sequential fashion: MPS alone, MPS plus ECG-gated data, and AC MPS with ECG-gated data. Images were interpreted for diagnostic certainty (normal, probably normal, equivocal, probably abnormal, abnormal, and perceived need for rest imaging). Adding AC data increased the number of studies characterized as definitely normal or abnormal to 84% (37% for MPS data alone) and reduced the perceived need for rest imaging from 77% for MPS data only to 43% for the addition of AC data. Heller et al., however, included far fewer patients than the present study, and only included patients with either known coronary artery disease or patients with a 5% or lower likelihood of coronary artery disease. The ten independent readers also interpreted the studies in a sequential manner. The fact that only patients with known coronary artery disease or patients with a very low likelihood of coronary artery disease in combination with the sequential interpretation could be the reason for the larger difference in the number of rest studies needed when adding AC images compared to NC images only found in their study compared to the present study.
Current guidelines also recommend the stress study to be performed first, since the rest study can be omitted if the stress study is interpreted as normal . Thus a rest study should only be performed in patients with equivocal or clearly abnormal studies. Chang et al.  investigated whether a normal stress-only MPS confers the same prognosis as a normal MPS on the bases of evaluation of stress and rest images. They found that patients who had a normal MPS on the basis of stress imaging alone have a similar mortality rate as those who have a normal MPS on the basis of evaluation of both stress and rest images.
In the present study the final reports according to clinical routine was used to compare the results from the stress-only assessment by the three study physicians. This might not be the optimal since physicians with different levels of experience interpreted the studies, and because there is always inter-observer variability when interpreting studies.
The physicians who evaluated the NC and AC stress studies did not have any clinical information about the patients. It is possible that fewer patients would have been falsely regarded as “no-rest-study-required” if the clinical information was available, as it is in the true clinical setting.
A third limitation is that in clinical routine at our department, physicians already evaluate the need for a rest study after the stress study (based on NC and AC stress images, gated images and clinical information). In 44% of the cases, no rest study is performed, and “no ischemia or infarction” is stated on the final report. If any mistakes were done in the clinical assessment, the final report used in the present study could be wrong.