Author + information
- John J. Mahmarian, MD* ( and )
- Su Min Chang, MD
- ↵*Methodist DeBakey Heart and Vascular Center, 6550 Fannin Street, Suite 677, Houston, Texas 77030
We thank Drs. Bhalodkar and Blum for their interest in our report (1). They raise concerns regarding the adoption of a stress-only single-photon emission computed tomography (SPECT) imaging protocol based on our study results. Although ours was a single-center, retrospective trial, it represents data on almost 17,000 consecutive patients who had similar baseline characteristics and event rates as reported from other large centers that perform stress SPECT. In this regard, we feel our study results are applicable to most patients who are referred for SPECT imaging.
The perceived conundrum in event rates between the stress-only and stress/rest groups is probably explained by adjustment for baseline characteristics that included clinical variables, stress electrocardiogram results, and the type of stressor used (please see the Statistical Analysis section of Chang et al. ). The stress/rest group did have a significantly higher crude annual event rate than the stress-only group (2.92% vs. 2.57%), which was probably related to their significantly greater number of cardiac risk factors, higher incidence of coronary artery disease (CAD), and higher frequency of pharmacologic stress testing. Statistical significance was lost after adjustment for these variables.
We acknowledge the retrospective design of our study. However, the criteria used for defining a normal stress study and deciding who should undergo additional rest imaging were prospectively implemented in our laboratory before the start date of our study. In addition, only 2 cardiologists (J.J.M. and M.S.V.) interpreted all of the SPECT studies and only 1 (J.J.M.) from years 2001 through 2007. In this regard, our study had many features of a prospective trial with consistency in image interpretation.
Stress-only studies were not restricted to patients weighing <200 lbs. Rather, weight was used only to determine the initial isotope dose and not the imaging protocol. As shown in Table 2 of Chang et al. (1), 3,086 patients weighing >200 lbs underwent a stress-only procedure, which represented 38% of all stress-only patients and 53% of all patients weighing >200 lbs. Thus, our study results are applicable to patients of all body weights.
Only a small percentage of patients (1.7%) did not have assessment of all-cause mortality by the Social Security Death Index, and lack of follow-up was evenly distributed between stress-only (n = 137 or 1.7%) and stress/rest (n = 165 or 1.8%) groups. It is highly unlikely that the inclusion of these subjects would have altered our findings.
We agree that the incidence of nonfatal end points are important and can add to health care costs. Although we did not specifically address nonfatal events, our overall mortality results are strikingly similar to those reported by other investigators (2–6). Based on these similarities, there is no reason to assume that nonfatal end points would have differed between our stress-only and stress/rest cohorts.
Regarding the issue of transient ischemic left ventricular dilation, we recognize that this can, at times, be the only abnormality seen in patients with multivessel CAD. For this reason, all patients undergoing stress-only imaging had normal end-diastolic and -systolic volumes by gated SPECT and no evidence of post-stress left ventricular dilation on the initial perfusion images. We hope that all of these explanations have clarified our study results in a satisfactory manner.
We also thank Drs. Kim and Bokhari for their remarks regarding our paper (1). We agree that there are inherent benefits to a stress-only imaging protocol in view of growing shortages of technetium (Tc) 99m and concerns over radiation exposure. Stress-only imaging should also reduce imaging time in a large percentage of patients and decrease imaging costs.
We recognize that performing stress imaging as the first test could potentially underestimate detection of ischemia if the rest injection is not delayed for several hours. However, this is more of a theoretical concern than a practical one, because there has not been a study to convincingly demonstrate a reduction in ischemia detection based on a stress/rest imaging protocol. Conversely, a rest/stress protocol may potentially decrease the sensitivity of SPECT because of contamination by the rest dose. In this regard, there are inherent limitations to any same-day stress/rest or rest/stress protocol that uses separate injections of a Tc 99m radiopharmaceutical. Performing a 2-day protocol in patients with abnormal or equivocal studies is an optimal approach, although admittedly inconvenient to the patient.
We agree that it is important to evaluate all raw, perfusion, and gated SPECT image information before deciding whether a patient should have rest imaging. This is exactly what was done in our study in addition to reviewing left ventricular cavity size, regional wall motion, and quantitative polar plot information. We respectfully disagree with performing rest imaging based on other findings such as increased lung uptake, right ventricular uptake, evidence of coronary artery calcification on a computed tomography attenuation image, or an abnormal stress test result. We agree that all of these findings are clinically important and may indicate the presence of significant CAD in a patient with an otherwise normal gated SPECT study. However, based on our results, rest imaging will not add any additional diagnostic or prognostic information if the stress images are already normal. Rather than performing rest imaging, we generally recommend additional testing, such as computed tomography coronary angiography, in our patients who have disparate perfusion and stress test results or in those whose symptoms are highly suggestive of angina despite a normal gated SPECT study.
In our study, 27% of patients had known CAD and 58% of the remaining patients had at least an intermediate pre-test likelihood. Although increasing age, male sex, history of CAD, diabetes mellitus, and history of chest pain all predicted a higher mortality rate, the addition of rest imaging added little to assessing risk when the initial stress images were normal. In this regard, we believe that acquiring stress images first is appropriate in most patients referred for SPECT imaging except in those who will clearly require rest imaging (i.e., patients with prior myocardial infarction). False reassurance from a normal stress-only or stress/rest perfusion study can best be avoided by further evaluating patients who have high clinical suspicion of CAD based on symptoms or abnormal stress test results.
- American College of Cardiology Foundation
- Chang S.M.,
- Nabi F.,
- Raza U.,
- Mahmarian J.J.
- Berman D.S.,
- Kang X.,
- Hayes S.W.,
- et al.
- Hachamovitch R.,
- Hayes S.,
- Friedman J.D.,
- et al.
- Chang S.M.,
- Nabi F.,
- Xu J.,
- et al.