Disagreement between splenic switch-off and myocardial T1-mapping after caffeine intake

Abstract

Caffeine is an adenosine receptor antagonist and a possible cause of inadequate stress perfusion. Splenic switch-off (SSO) and splenic rest-stress T1-mapping have been proposed as indicators of stress adequacy during perfusion cardiac magnetic resonance (CMR). We compared myocardial rest-stress T1-mapping with SSO and splenic rest-stress T1-mapping in patients with and without recent coffee intake. We analyzed 344 consecutive patients suspected of myocardial ischemia with adenosine perfusion CMR. All 146 normal CMR studies with a normal T1-rest of the myocardium, used as standard of reference, were included and divided in two groups. 22 patients accidentally ingested coffee < 4 h before CMR, compared to control group of 124 patients without self-reported coffee intake. Two independent readers graded SSO visually. T1-reactivity (ΔT1) was defined as percentual difference in T1-rest and T1-stress. Follow-up data were extracted from electronic patients records. In patients with recent coffee intake SSO was identified in 96%, which showed no significant difference with SSO in controls (94%, p = 0.835), however event rates were significantly different (13.6 and 0.8%, respectively (p < 0.001), median FU 17 months). Myocardial ΔT1 in the coffee group (- 5.2%) was significantly lower compared to control (+ 4.0%, p < 0.001), in contrast to the splenic ΔT1 (- 3.7 and - 4.0%, p = 0.789). The splenic T1-mapping results failed to predict false negative results. SSO and splenic rest-stress T1-mapping are not reliable indicators of stress adequacy in patients with recent coffee intake. Therefore, the dark spleen sign does not indicate adequate myocardial stress in patients with recent caffeine intake. Myocardial rest-stress T1-mapping is an excellent indicator of stress adequacy during adenosine perfusion CMR.

Keywords: Adenosine; Cardiac magnetic resonance; Splenic switch-off; Stress CMR; T1-mapping.

Fig. 1

Illustrative presentation of T1-mapping results and SSO assessment based on a case of a 75-year old female that underwent multiple adenosine perfusion CMR evaluations within a short period of time due to: a technical failure during the first examination with leakage of adenosine through the extension set of the intravenous line causing inadequate adenosine induced hyperemia, b a second examination (14 days later) in which ingestion of coffee < 4 h examination was retrospectively determined and c a third final examination 7 days after the second examination without protocol conflicts. For illustrative purposes we present (from left to right) myocardial and splenic native T1-mapping analysis with endo- and epicardial contours graphically visualized (green lines) and regions of interest placed in the inferior septal region of the myocardium (pink) and in the spleen (yellow) with the use of dedicated software, a graphical illustration of the ΔT1 color maps and adenosine perfusion images. a Inadequate myocardial stress as shown by a very low ΔT1 of the myocardium (+ 0.2%, purple), minor stress changes of the spleen (purple) and no SSO. b Inadequate myocardial stress due to coffee intake in the hours prior to the examination as indicated by the inverted ΔT1 of the myocardium (− 8.3%, red). Strongly negative ΔT1 in the spleen (red) and presence of SSO. c Adequate myocardial stress with a positive ΔT1 of the myocardium (blue), adequate stress changes of the spleen (red) and SSO

Fig. 2

Boxplot of myocardial and splenic T1-reactivity (ΔT1) in patients with coffee intake < 4 h prior to the examination and patients with normal perfusion CMR and no recent coffee intake. There was a significant difference between myocardial and splenic ΔT1 in patients with normal perfusion CMR. No statistical difference was present in patients with coffee intake < 4 h prior to the examination

Fig. 3

Kaplan–Meier survival curve comparing patients with recent coffee intake to patients in the control group. MACE was defined as myocardial ischemia, myocardial infarction, heart failure or cardiac death. Log-rank test p-value < 0.001