Can Spleen T1 Mapping Be Used as a Marker of Adenosine Response in Cardiac MRI?

ABSTRACT Splenic switch‐off, defined as the visible reduction in splenic perfusion during adenosine stress, can be used as a marker of adequate adenosine response in cardiac stress MRI studies, which can be assessed after the study. Moreover, there is some evidence regarding the use of T1 mapping for the evaluation of splenic switch‐off during the study. The aim of this work is to further investigate splenic T1 mapping as a method for quantitatively assessing splenic switch‐off, given the limited existing data. Patients with suspected coronary artery disease were scanned at 1.5 T with the routine clinical protocol for stress perfusion cardiac magnetic resonance (CMR), including a T1 mapping sequence at rest and during adenosine stress. Visual assessment of first‐pass images was used as the reference standard for detecting splenic switch‐off. Semi‐quantitative analysis of splenic perfusion was performed based on the signal intensity curves and signal changes (ΔSI) were calculated. T1 maps were used to obtain T1 value changes in the spleen from rest to stress (ΔT1 spleen ). Statistical analysis was assessed using an align rank transform followed by a mixed analysis of variance, with Bonferroni corrected post hoc tests. A total of 74.5% ( n = 35/47) of analyzed patients experienced splenic switch‐off according to the visual assessment. Statistically significant differences were found in splenic T1 between rest and stress only in the switch‐off group ( p < 0.001), with lower T1 values during stress. There was significant positive weak correlation (ρ s = 0.37, p = 0.01) between splenic ΔT1 and ΔSI based on first‐pass. ROC curves demonstrated that ΔT1 spleen classified the failed switch‐off and positive switch‐off groups with a specificity of 75% and a sensitivity of 80% (AUC: 81.2%, 95% CI: 66.7%–95.8%). This work indicates that the splenic T1 value change has the potential to identify splenic switch‐off during adenosine stress CMR in a clinical population.