Author + information
- Hidenari Matsumoto, MD, PhD∗ (, )
- Shunpei Ushimaru, MD,
- Tetsuya Matsuda, MD, PhD,
- Toshihiko Shimada, MD,
- Mikiko Mikuri, MD,
- Haruya Takahashi, PhD,
- Nobuyuki Takahashi, PhD,
- Teruo Kawada, PhD and
- Taketoshi Yamazaki, MD, PhD
- ↵∗Cardiovascular Center, Iseikai Hospital, 6-2-25, Sugahara, Higashiyodogawa-ku, Osaka 533-0022, Japan
Caffeine antagonizes the pharmacological actions of adenosine by blocking adenosine receptor activity (1). A protocol for adenosine stress myocardial perfusion imaging recommends that caffeine-containing products be withheld for 12 h before the test (2). However, there has been no widely accepted consensus for the need of caffeine abstention before fractional flow reserve (FFR) measurement. Conflicting results have been reported in the literature concerning the effects of caffeine on FFR measurement (3,4). Thus, we designed this study to determine if caffeine abstention is required before FFR measurement and if high-dose intracoronary adenosine overcomes the caffeine antagonism.
This prospective, single-center study enrolled 76 patients who underwent clinically indicated FFR assessment. Of these patients, 19 patients in each group were asked to refrain from caffeine-containing products for 12, 24, and 48 h before the test and 19 patients were allowed to have caffeine. Exclusion criteria included acute myocardial infarction, severe arrhythmia, previous coronary artery bypass grafting, any contraindications for adenosine or papaverine, and the presence of pressure drift (>0.03) after pullback. Hyperemia was induced by central intravenous adenosine at a dose of 140 μg/kg/min (ADN-IV), by intracoronary adenosine at varying doses (60 μg [ADN-IC60], 150 μg [ADN-IC150], 300 μg [ADN-IC300], by 600 μg [ADN-IC600]), and by papaverine (10 to 12 mg in the right coronary artery or 15 to 20 mg in the left coronary artery), as a reference standard. The next higher doses of intracoronary adenosine were not administered in the event of a pause in the heartbeat of >3 seconds.
The median caffeine levels were 1.424, 0.452, 0.240, and 0 mg/l (interquartile ranges: 0.876 to 4.764, 0.291 to 1.054, 0.060 to 0.303, and 0 to 0.002 mg/l) in the no caffeine, and 12-, 24-, and 48-h caffeine abstention groups, respectively. The protocol of ADN-IV was completed in all patients. Figure 1A shows scatterplots of FFR between papaverine and ADN-IV. Compared with papaverine, ADN-IV overestimated FFR in the no caffeine group (p < 0.001) and in the 12-h (p < 0.001) and 24-h caffeine abstention (p < 0.01) groups, whereas the difference in FFR was not significant in the 48-h abstention group (p = 0.61). The overestimation in FFR by ADN-IV was correlated with caffeine levels (r = 0.48; p < 0.001) (Figure 1B). ADN-IV overestimated FFR in 91% (29 of 32) and 94% (16 of 17) of the patients who had caffeine levels of >0.5 and >1.0 mg/l, respectively.
ADN-IC600 was feasible in 84% (51 of 61) of the patients for the left coronary artery and in only 13% (2 of 15) of the patients for the right coronary artery. Incremental ADN-IC produced progressively lowered FFR values in all the groups. However, even increased doses of intracoronary adenosine up to 600 μg still overestimated FFR compared with papaverine in the no caffeine group (p < 0.001) and the 12-h (p < 0.01) and 24-h caffeine abstention (p < 0.01) groups. In the 48-h caffeine abstention group, the mean FFR value did not differ significantly between papaverine and ADN-IC600 (p = 0.31), whereas FFR was overestimated by ADN-IC60 (p < 0.01) and tended to be overestimated by ADN-IC150 (p = 0.07) and ADN-IC300 (p = 0.05).
We found that the currently recommended central ADN-IV overestimated FFR in the presence of caffeine in a dose-dependent manner, and the increased doses of ADN-IC up to 600 μg could not fully surmount the antagonism. A review article suggested that a caffeine level of <3 to 4 mg/l was acceptable for adenosine stress myocardial perfusion imaging (5). However, the remaining caffeine in the blood could be clinically relevant for FFR measurement even after 24 h of caffeine abstention.
The results of this study suggest prolonged caffeine abstention (>24 h) would be safer to avoid incomplete hyperemia. Otherwise, we would have to choose alternative drugs that are unaffected by caffeine, such as papaverine and nicorandil, when an FFR value obtained with adenosine stays above 0.80.
In conclusion, caffeine abstention is necessary before adenosine-induced FFR measurement, regardless of the administration route.
Please note: The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation