Although many of the studies on autonomic imaging in CHD mainly focused on patients after myocardial infarction, cardiac sympathetic nerve terminals are more sensitive to ischemia than myocardial cells as described earlier (Matsunari et al. 2000). Therefore, it is conceivable that ischemic sympathetic neuronal damage would occur even in the absence of infarction. This was confirmed by clinical studies demonstrating [¹²³I]-MIBG defects exceeding perfusion defects in patients without a history of myocardial infarction (Guertner et al. 1993; Hartikainen et al. 1997). Similar results were observed using [¹¹C]-mHED PET (Bulow et al. 2003). This concept was also supported by an experimental study by Luisi et al. using a pig model of chronic hibernation without infarction showing extensive defects in [¹¹C]-mHED uptake in the area of hibernating but viable myocardium (Luisi et al. 2005).

Vasospastic angina is a form of angina characterized by a sudden spasm of one of the coronary arteries without significant stenosis. It is known that vasospastic angina is often associated with altered autonomic nervous activity (Ricci et al. 1979). Because coronary vasospasm causes a transient ischemia in that vascular territory, such an ischemic episode may cause reduction in [¹²³I]-MIBG uptake despite preserved perfusion. The results of meta-analysis using Meta-DiSc software (Zamora et al. 2006) from seven published studies (Ha et al. 1998; Hirano et al. 2002; Inobe et al. 1997; Sakata et al. 1997; Takano et al. 1995; Taki et al. 1998; Watanabe et al. 2002a) that reported the diagnostic value of [¹²³I]-MIBG imaging are summarized in Fig. 14.4. The pooled estimates of sensitivity and specificity were 77 % (95 % confident intervals: 70–82 %) and 77 % (95 % confident intervals: 70–83 %), respectively. The area under the summary receiver operating characteristic analysis (SROC) curve and its Q*-point were 89 ± 5% and 82 ± 5 %, respectively. However, both sensitivities and specificities had significant heterogeneity (P < 0.0001 for both), as is evident from the wide range (42–100 % for sensitivity and 40–100 % for specificity) and the high inconsistency index (90.8 % for sensitivity and 80.8 % for specificity). The heterogeneity in diagnostic performance is also evident from the wide 95 % confidence intervals in SROC curves. In a study by Sakata et al., the use of lower [¹²³I]-MIBG washout rate rather than reduced [¹²³I]-MIBG uptake as a marker of vasospastic angina significantly improved sensitivity and specificity from 42 to 76 % and 72 to 87 %, respectively, indicating that cardiac sympathetic nerve activity in vasospastic angina is suppressed probably because of the enhanced parasympathetic nerve activity (Sakata et al. 1997). Furthermore, a subsequent study from the same authors has demonstrated that high H/M ratio or lower washout rate was a marker of future cardiac events (Sakata et al. 2005). By contrast, Inobe et al. have demonstrated that the patients with higher disease activity were frequently associated with either the uptake reduction or the abnormally high washout of [¹²³I]-MIBG (Inobe et al. 1997). Although these inconsistent published results could be explained in part by different patient characteristics such as disease severity and duration and by different methodologies among the studies (Ha et al. 1998; Hirano et al. 2002; Inobe et al. 1997; Sakata et al. 1997; Takano et al. 1995; Taki et al. 1998; Watanabe et al. 2002a), the exact underlying mechanisms are still unclear. Thus, the clinical use of autonomic imaging in vasospastic angina remains to be established by further work.

Silent myocardial ischemia is defined as evidence for ischemia without symptoms such as chest pain (Gutterman 2009). It is important to note that patients with silent myocardial ischemia often have a poor prognosis as compared with those with painful ischemia. Therefore, early detection and monitoring of patients with silent myocardial ischemia before hard cardiac events is clinically relevant. Although the exact mechanism is yet to be determined, impairment in cardiac sympathetic innervation is likely to be linked with the occurrence of silent myocardial ischemia. In this regard, autonomic imaging may play a role for characterization of silent ischemia. However, published results focusing on this topic using [¹²³I]-MIBG imaging are inconclusive so far. In a study by Shakespeare et al. using delayed (4 h) [¹²³I]-MIBG SPECT, no evidence of denervation was observed in silent ischemia (Shakespeare et al. 1993). Hartikainen et al. found that the patients with silent ischemia revealed smaller areas of viable but denervated myocardium than those with chest pain (Hartikainen et al. 1994). Studies by other groups (Langer et al. 1995; Matsuo et al. 1996; Shimonagata et al. 1995), however, have suggested a link between silent ischemia and abnormalities on [¹²³I]-MIBG SPECT. A study by Langer et al. have demonstrated that diabetic patients with silent myocardial ischemia had evidence of a diffuse abnormality in [¹²³I]-MIBG uptake (Langer et al. 1995). A subsequent study by Matsuo et al. showed decreased [¹²³I]-MIBG uptake particularly in the inferior wall in patients with silent ischemia (Matsuo et al. 1996). All these studies were based on relatively small patient sample size, which may have contributed to the inconclusive results among the studies. It should also be noted that all these studies mainly relied on interpretation of [¹²³I]-MIBG SPECT images, where accurate assessment of regional abnormalities may sometimes be difficult due to poor image quality (Matsunari et al. 2010). Thus, much work still needs to be done to determine the value of autonomic imaging to characterize silent myocardial ischemia.

There is an increasing body of evidence that autonomic imaging using [¹²³I]-MIBG provides important prognostic information in patients with heart failure (Agostini et al. 2008; Jacobson et al. 2010; Kuwabara et al. 2011; Merlet et al. 1992; Nakata et al. 1998; Verberne et al. 2008). Patients with a reduced late H/M ratio or increased washout rate are likely to be associated with poor prognosis. Although there are only a few data available that have specifically looked at patients with coronary heart disease (Kasama et al. 2011; Wakabayashi et al. 2001), many published studies regarding prognostic value of [¹²³I]-MIBG imaging included a considerable number of ischemic heart failure patients. In a large prospective study (ADMIRE-HF) by Jacobson et al. involving 961 patients, for example, 61 % of total study population had ischemic heart failure (Jacobson et al. 2010). Additionally, it is known that myocardial adrenergic nerve activity measured by [¹²³I]-MIBG imaging is accelerated in proportion to severity of heart failure, independent of the underlying cause (Imamura et al. 1995). Furthermore, in a study by Kasama et al. that focused on patients with ST-segment elevation myocardial infarction, increased [¹²³I]-MIBG washout rate was a significant predictor of cardiac events, independent of left ventricular ejection fraction (LVEF) (Kasama et al. 2011). Thus, it is possible that [¹²³I]-MIBG imaging predicts prognosis in patients with ischemic heart disease as does in general heart failure population. However, it should be noted that the cutoff thresholds of H/M ratio to identify high-risk patients may not be identical between ischemic and nonischemic heart failure, as demonstrated in a study by Wakabayashi et al. (2001).

At present, data on the prognostic value of autonomic PET imaging in CHD is still limited. Pietila et al. performed [¹¹C]-mHED PET imaging in 46 chronic heart failure (CHF) patients (72 % of total population were CHD) and found that a low [¹¹C]-mHED retention was a predictor of poor prognosis (Pietila et al. 2001). Gaemperli et al. have found that reduced myocardial β-adrenergic receptor density measured by [¹¹C]-CGP-12177 PET was associated with development of CHF during a median follow-up period of 12.7 years (Gaemperli et al. 2010). Thus, autonomic PET imaging also seems to be promising for risk stratification in CHD patients.