We found that nuclear medicine technologists were able to determine when a rest study was required. There was very good agreement between the technologists and the physicians, and the proportion of rest studies did not change after the new routine was introduced.
If the nuclear medicine technologist who acquires the stress image is able to determine the necessity for a rest study, this will improve clinical workflow. The technologist does not have to wait for a decision made by the physician, and the physician will have more time to correctly interpret the images and write the final report to the referring clinician. The present study indicates that it is possible to delegate this assessment to the nuclear medicine technologist. It is, however, important to state that the physicians are still responsible for the final evaluation of the studies. There is always the opportunity to call the patient back for a rest study, if the physician who interprets the images and writes the final reports so desires. All patients that were sent home after the stress study during the follow-up period were informed that a rest study might still be needed and that they would be contacted if a rest study was desired by the physician. Thus, in our opinion this approach increases laboratory efficiency without increasing the risk for false negative interpretations.
In this study, one patient would have been sent home by the nuclear medicine technologist without a rest study, when the study was interpreted as ischemic on the final report. It is not clear whether this was a typing error made by this experienced technologist or if the technologist was not properly trained. If this would have been a patient sent home by a technologist, the patient would have been contacted for a subsequent rest study when the physician responsible for interpreting the study evaluated the images. This patient was not missed by the two technologists who assessed inter- and intra-observer variability.
Inter- and intra-observer variability was higher for technologists than for physicians. This is probably due to less experience of interpreting images for nuclear medicine technologists, and will probably improve over time. In this study, the option of “consulting a physician” was not possible for the inter- and intra-observer variability assessment, which probably also lowered the kappa value.
A similar study was performed by Johansson et al [10
] in 2008. In their study, visual interpretation of 532 patients of the complete stress and rest images by 1 experienced physician was used as gold standard. All cases categorized as infarction or ischemia present were categorized as the group requiring a rest study, whereas all other cases were categorized as the group not requiring a rest study. A total of 3 nuclear medicine technologists and 3 physicians independently classified each of the stress studies as rest-study-required or no-rest-study-required. They found that the nuclear medicine technologists were able to assess whether a rest study was needed. The risk that this assessment would be incorrect was not higher for the technologists than it was for the physicians. Their gold standard differed from the one used in the present study. In the clinical routine, more patients than those with ischemia or infarction onthe final interpretation will undergo a rest study (i.e. patients with equivocal images). We therefore believe that our approach is more suitable as a gold standard. Our study with only the technologist and physician responsible for the patient evaluating the need for a rest study also reflects a more “accurate” clinical situation compared to using only 3 technologists and 3 physicians evaluating all images evaluating images only in a retrospective study.
There are both advantages and disadvantages for using a stress-only approach for stress studies interpreted as normal or probably normal. In most cases, the approach will improve work flow in the nuclear medicine department and, more importantly, reduce the radiation dose to the patient. However, there is a possibility that small defects might be overlooked at the stress study, thus giving false negative interpretations. There is also a possibility that balanced 3-vessel disease only presenting with transient ischemic dilatation and reduction in ejection fraction on stress images are missed when no rest studies are available for comparison. Current guidelines, however, recommend that rest studies should not be performed if the stress study is considered normal [6
The physicians who evaluated the NC and AC stress studies for the gold standard did not have any clinical information about the patients. It is possible that there would have been higher agreement between gold standard and technologists/physicians if the clinical information would have been available when deciding the gold standard.