Myocardial Perfusion Imaging Using Gated SPECT in Heart Failure Patients Undergoing Cardiac Resynchronization Therapy

MATERIALS AND METHODS

We studied 20 consecutive patients with advanced heart failure (New York Heart Association [NYHA] class ≥ III, LVEF < 35%) and left bundle branch block with QRS duration > 120 ms, scheduled for CRT because of persisting symptoms despite optimal medical therapy. Table 1 summarizes patient characteristics on enrollment. Resting gated SPECT was performed before implantation and was repeated 3 mo later, together with clinical evaluation, including 6-min walking distance and quality of life (QOL) assessment with the Minnesota Living with Heart Failure questionnaire (13). The optimized medical treatment of each patient remained unchanged during the study. Our Ethics Committee approved the protocol, and patients gave informed consent to participate.

RESULTS

Before implantation, LVEF was 22.4% ± 7.2%; LVED volume index (VI), 163 ± 62 mL; LVESVI, 130 ± 61 mL; and WMSI, 2.6 ± 0.3. Implantation was successful and without complication in all patients. QRS width decreased from 139 ± 15 ms to 122 ± 8 ms (P < 0.00001). A significant perfusion defect (mean, 24% ± 12.5% of the LV wall) was detected in 11 patients, 7 with a diagnosis of ischemic cardiomyopathy. The location of the defect was anteroseptal or anteroapical in 5 patients and inferior or inferolateral in 6. One patient with ischemic cardiomyopathy and an extensive perfusion defect died 2 mo after implantation because of pump failure and was therefore excluded from further analysis. Of the remaining 19 patients who could undergo the 3-mo follow-up study, 10 had a significant perfusion defect (group A) and 9 had a normal myocardial perfusion pattern (group B) in the baseline evaluation before CRT. Table 2 compares the main characteristics of the 2 groups, which differed only in terms of quality of life (worse in group A).

At the 3-mo follow-up, an improvement in NYHA class was registered in 6 of 10 patients of group A and in 8 of 9 of group B. Both groups showed a significant improvement in the QOL score, a significant increase in the 6-min walking distance, and a significant improvement in regional wall motion (Fig. 1). The QOL score of group A remained significantly higher and the 6-min walking distance became significantly shorter than the corresponding values of group B. The reduction in WMSI was clearly larger in group B than in group A (Fig. 1). In group A, the LVEF, LVEDVI, and LVESVI did not change significantly during CRT (Fig. 2). Conversely, in group B, the LVEF increased from 23.1% ± 8% to 27.1% ± 11% (P < 0.03), and LVEDVI and LVESVI decreased from 159 ± 70 mL to 135 ± 68 mL (P < 0.02) and from 127 ± 67 mL to 104 ± 65 mL (P < 0.01), respectively (Fig. 2).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Evolution of QOL score, 6-min walking distance, and WMSI in group A (□) versus group B (▪).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Evolution of LVEF, LVEDVI, and LVESVI in group A (□) versus group B (▪).

DISCUSSION

Although the value of CRT in advanced heart failure has been demonstrated, a proportion of patients responds suboptimally to treatment. To identify those patients, electrophysiologic characterization of the potential candidates was emphasized when CRT was proposed. Preimplantation QRS duration and its shortening under CRT were considered most important parameters for predicting response (5,17), but their value has been questioned (6). Other factors, such as LV dimensions and concomitant mitral regurgitation, were considered potential predictors of patient outcome under CRT (4–6). Also, the possible role of the cause of congestive heart failure in determining the response to CRT was examined. In a study on acute hemodynamic changes induced by CRT, Mansourati et al. did not register any difference between patients with dilated cardiomyopathy and those with ischemic cardiomyopathy (18). In a prospective study involving a patient population with heart failure of mixed causes who underwent CRT, Lunati et al. did not identify any epidemiologic or clinical variable able to predict the patients who would not have a satisfactory response to treatment (19). Other studies, however, suggested a possible role of cause on the basis of the observation that ischemic cardiomyopathy occurs slightly more frequently in patients without improvement under CRT (5,7). Reuter et al. found that ischemic heart disease, together with older age and absence of mitral regurgitation, was a significant predictor of absence of response to CRT (6). In a recent study, Duncan et al. demonstrated significantly less extensive LV reverse remodeling in patients with ischemic cardiomyopathy than in those with idiopathic dilated cardiomyopathy (8). So far, however, only scanty data are available about myocardial perfusion in candidates for CRT, and the data demonstrate mainly that CRT does not cause major changes in myocardial blood flow in patients with dilated cardiomyopathy but causes homogenization of glucose metabolism (20). To our knowledge, our study is the first that evaluates the relationship between resting perfusion pattern and response to CRT.

According to our results, severe resting uptake defects can be encountered in a series of consecutive patients with severe heart failure of mixed cause. Almost all patients with a history of ischemic cardiomyopathy had a clear resting-uptake defect, likely caused by prior infarction. Extensive hypoperfusion was observed in 4 other patients with nonischemic cardiomyopathy. This finding has various possible explanations: concomitant undiagnosed coronary artery disease cannot be excluded. Alternatively, both nonischemic fibrotic changes and abnormalities of resting perfusion have been reported in patients with dilated cardiomyopathy (21,22). Furthermore, an apparent decrease in tracer uptake because of the partial-volume effect could play a role in regions with severe regional dysfunction and wall thinning (23). Also, the presence of abnormal wall motion because of left bundle branch block could contribute to the finding of decreased tracer uptake, although no isolated septal defect was observed in our series. Independently of the actual cause, the finding of a severe uptake defect could reasonably be related to more severe disease in these patients. The main result of our study, however, is that in patients with a significant resting perfusion defect, the clinical improvement under CRT seems more limited and, most important, is not accompanied by a significant improvement in LV volumes or in LVEF, with just a limited reduction of regional asynergy. Conversely, with a single exception, the patients without abnormal findings in the preimplantation resting perfusion scan showed noteworthy clinical improvement, which was confirmed by a significant decrease in LV volumes and an increase in global LV function. If confirmed, this observation would have major clinical implications, because although epidemiologic data indicate that the combination of severe LV failure and advanced bundle branch block is relatively rare among patients with coronary artery disease (24), ischemic cardiomyopathy must be considered the most important cause of heart failure, particularly in older patients (25).

A second interesting result of our study is that gated SPECT appears a feasible technique for the assessment of baseline LV function and of its long-term changes in patients undergoing CRT. For this aim, echocardiography has, so far, been the most frequently used method (8,26–28), and its role was reinforced by the introduction of tissue Doppler imaging to select the optimal pacing site and to evaluate changes in the synchronicity of LV contraction (16,29,30). However, the feasibility of echocardiography may be limited because of acoustic window problems and the reproducibility of volumetric measurements may be inadequate. Thus, radionuclide angiography was used for LVEF measurement in another recent multicenter study about CRT (31). On the other hand, simultaneous measurement of LV volumes and function is possible and easily feasible using gated SPECT (9). With the exception of patients with severe arrhythmia, all patients can undergo the technique, and the robustness and reproducibility of LV volume measurements using gated SPECT have been demonstrated (10). A single examination could both evaluate abnormalities in the perfusion pattern, which could be relevant to the outcome of CRT, and assess the functional status of the LV to objectively demonstrate its evolution.

In our population, gated SPECT allowed reliable evaluation of both global and regional changes in LV function during CRT. The improvement in LVEF and decrease in LV volumes agree with prior reports and are closely related to clinical evolution. Volumetric reduction was registered in all patients with clinical improvement. These data suggest that gated SPECT effectively evaluates LV functional evolution in patients undergoing CRT and confirms the presence of reverse remodeling in patients who respond to treatment (8,30).

The small number of patients and the lack of comparison with other established imaging methods are the most important limitations of this study. The threshold used to define a significant perfusion defect was selected to identify probably nonviable myocardium (14,15). Therefore, the possible interactions between the presence of viable hibernating myocardium and CRT were not considered. In heart failure patients in whom myocardial hibernation is the likely cause of LV dysfunction, coronary revascularization is the treatment of choice (32). Evaluation of myocardial perfusion to exclude inducible ischemia or hibernation should be considered in heart failure patients with suspected coronary artery disease before CRT is considered. Further studies are needed to clarify this issue.

CONCLUSION

Severe resting perfusion defects encountered in candidates for CRT apparently do not preclude a positive clinical response to CRT but are related to a more limited objective improvement of LV volumes and function. Gated SPECT functional data can characterize the baseline status of patients with severe heart failure and ensure an objective basis for their follow-up during CRT.