DSE vs. PET for Viability
abstract & commentary
Synopsis: PET is more sensitive for detecting viability but DSE is more specific.
Source: Pasquet A, et al. Am J Cardiol 1999;84:58-64.
There are few direct comparison studies of dobutamine stress echo (DSE) to position emission tomography (PET) for the detection of myocardial viability as validated by return of left ventricular (LV) function after surgical revascularization. Thus, Pasquet and colleagues studied 94 consecutive patients with coronary artery disease (CAD) and severe LV dysfunction (ejection fraction < 35%). PET imaging was done before and after dipyridamole handgrip stress using rubidium-82 for perfusion imaging and 18-fluorordeoxyglucose to assess metabolism. DSE was done with atropine, if necessary, to achieve 85% of predicted heart rate. Both DSE and PET were done before consideration of revascularization in all patients. At about three months postsurgery, LV functional recovery was assessed at rest in 68, and at rest and stress in 29. Ischemia was defined as an induced perfusion defect on PET or wall motion abnormality on DSE. Viable myocardium was defined as perfusion-metabolism mismatch on PET or low-dose DSE wall motion augmentation. Using a 16-segment model, concordance of PET and DSE for identifying viable myocardium before surgery was 63%. Most of the discordant segments were nonviable by DSE and viable by PET; 50% of the discordant segments were in anterior or septal segments. Surgical revascularization was performed in 75 patients. Prediction of improved resting function after surgery was 83% by PET and 69% by DSE (P < 0.001), but the specificity of DSE was higher (78% vs 37%; P < 0.001), as was accuracy (75% vs 53%; P < 0.001). The accuracy of DSE was enhanced by consideration of the postsurgery stress images (86%), but PET was not. Pasquet et al conclude that PET is more sensitive for detecting viability, but DSE is more specific.
Comment by Michael H. Crawford, MD
Since most centers do not have PET available, this study may seem moot. However, there are few studies of DSE using the clinical gold standard of functional recovery after revascularization. In this study, all the patients had surgical revascularization because most patients with severe LV dysfunction and CAD have multivessel disease. Of the 75 patients who had surgery, seven were dropped because of major perioperative events, so the analysis focuses on the 90% who did well with surgery. These 68 patients had 1088 myocardial segments; one-fourth were normal, half showed no improvement (non-viable), one-fourth improved (viable), and a few deteriorated (1%). Surgery seems to be a big intervention to improve the function of one-fourth of the LV. Also, the study did not report the postoperative ejection fraction values, which suggests they were unchanged. Of course, there are other reasons to do surgery. Patients may get symptom relief, experience fewer arrhythmias, and reduce remodeling after revascularization. Hence, it would have been interesting to know the long-term functional and clinical outcomes of these patients and how they related to the viability studies.
The discordance between PET and DSE is understandable since they are completely different techniques (apples/oranges) where alignment of imaging planes is likely to be inaccurate. Also, the distinction between severe hypokinesis and akinesis of a segment is difficult to determine by echocardiography. In fact, the main added value of evaluating the poststress echo images to determine recovery of segmental function was in the interpretation of hypokinetic segments. A major category of discordance was segments with PET viability but no function at rest or stress on echo. Perhaps these were stunned segments, but ordinarily stunned myocardium would augment with dobutamine. More likely, these segments had islands or peninsulas of viable myocardium large enough to affect the PET results but too small to affect segmental contractility. This would argue that DSE gives more reliable results in terms of what can be expected from surgery and this is reflected in the higher accuracy of DSE in this study. Thus, if your center does not have PET, this application is not a good reason to buy one.
The real issue is when and whether to do DSE to detect viability. Patients with low ejection fraction and multivessel CAD with good targets should have revascularization surgery if feasible and DSE is unlikely to influence this decision, nor should it. DSE is accurate enough that it can be used where the clinical decision is not as obvious (i.e., if only partial revascularization is possible or the patient is at higher risk for surgery). DSE is especially useful if it shows viability (high specificity and positive predictive value). Nonviable segments on DSE may or may not recover.
For the detection of surgical revascularization validated myocardial viability, which of the following is most correct?
a. PET is less sensitive than SDE
b. DSE is less specific than PET
c. DSE is more clinically accurate than PET
d. All of the above
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content