Splenic switch-off as a predictor for coronary adenosine response: validation against 13N-ammonia during co-injection myocardial perfusion imaging on a hybrid PET/CMR scanner

BACKGROUND Inadequate coronary adenosine response is a potential cause for false negative ischemia testing. Recently, the splenic switch-off (SSO) sign has been identified as a promising tool to ascertain the efficacy of adenosine during vasodilator stress cardiovascular magnetic resonance imaging (CMR). We assessed the value of SSO to predict adenosine response, defined as an increase in myocardial blood flow (MBF) during quantitative stress myocardial perfusion 13 N-ammonia positron emission tomography (PET). METHODS We prospectively enrolled 64 patients who underwent simultaneous CMR and PET myocardial perfusion imaging on a hybrid PET/CMR scanner with co-injection of gadolinium based contrast agent (GBCA) and 13N-ammonia during rest and adenosine-induced stress. A myocardial flow reserve (MFR) of > 1.5 or ischemia as assessed by PET were defined as markers for adequate coronary adenosine response. The presence or absence of SSO was visually assessed. The stress-to-rest intensity ratio (SIR) was calculated as the ratio of stress over rest peak signal intensity for splenic tissue. Additionally, the spleen-to-myocardium ratio, defined as the relative change of spleen to myocardial signal, was calculated for stress (SMR$_{stress}$) and rest. RESULTS Sixty-one (95%) patients were coronary adenosine responders, but SSO was absent in 18 (28%) patients. SIR and SMR$_{stress}$ were significantly lower in patients with SSO (SIR: 0.56 ± 0.13 vs. 0.93 ± 0.23; p < 0.001 and SMR$_{stress}$: 1.09 ± 0.47 vs. 1.68 ± 0.62; p < 0.001). Mean hyperemic and rest MBF were 2.12 ± 0.68 ml/min/g and 0.78 ± 0.26 ml/min/g, respectively. MFR was significantly higher in patients with vs. patients without presence of SSO (3.07 ± 1.03 vs. 2.48 ± 0.96; p = 0.038), but there was only a weak inverse correlation between SMR$_{stress}$ and MFR (R = -0.378; p = 0.02) as well as between SIR and MFR (R = -0.356; p = 0.004). CONCLUSIONS The presence of SSO implies adequate coronary adenosine-induced MBF response. Its absence, however, is ...