Author + information
- Published online June 3, 2019.
- Christina Byrne, MD∗ (, )
- Andreas Kjaer, MD, DMSc,
- Mads Wissenberg, MD, PhD,
- Preetee Kapisha Hurry, MD,
- Anne Schmedes, MSc, PhD,
- Julie Lyng Forman, MSc, PhD and
- Philip Hasbak, MD, DMSc
- ↵∗Copenhagen University Hospital, Rigshospitalet, Department of Cardiology, 9841, Blegdamsvej 9, 2100-Cph, Denmark
Caffeine is a nonselective competitive antagonist of the A2A-adenosine receptor that mediates coronary vasodilation (1,2). However, the effect of interaction between caffeine and adenosine on myocardial blood flow is not clarified (3). Therefore, we investigated whether caffeine changed the quantitative flow measurements of adenosine stress rubidium-82 positron-emission tomography/computed tomography (82Rb-PET/CT).
From September 2016 to March 2017, 41 healthy young volunteers underwent 3 rest and adenosine stress 82Rb-PET/CT scans separated by at least 24 h obtaining the myocardial blood flow (MBF) reserve (stress flow/rest flow) measuring vasomotor function. Inclusion criteria were: age >18 years, drinking coffee every day, no known medical condition, no medicine consumption, and no use of tobacco or euphoric substances (except alcohol) within the last 3 months. Volunteers were excluded according to standard adenosine-stress-PET/CT exclusion criteria. Informed oral and written consents were obtained, and the Scientific Ethics Committee of the Capital Region of Denmark and the Danish Data Protection Agency approved the protocol (protocol number H-15009293).
One-half of the participants received the following caffeine dosage regime orally in random order 1 h before scan: 0 mg, 100 mg, and 300 mg. The rest received 0 mg, 200 mg, and 400 mg. We collected 4 blood samples at each scan: before caffeine administration (T = 0 min), before the rest scan (T = 60 min), after adenosine administration (T = 75 min), and at the end of the scan (T = 90 min). Participants were instructed not to consume substances containing caffeine and theophylline for 24 h before scans. PET/CT imaging and analyses were performed as previously described (4). SAS version 9.4 (SAS Institute, Cary, North Carolina) were used for statistical analyses. The effects of caffeine dosing and serum concentrations were assessed in linear mixed models with random subject effects.
Median age was 23 years (interquartile range [IQR]: 22 to 25 years) and 51% were men. Median daily coffee consumption was 2 cups (IQR: 1.5 to 3 cups). Median time from first to third scan was 28 days (IQR: 14 to 38 days). Mean serum caffeine levels at time T = 0 min to T = 90 min are shown in Figure 1A.
MBF reserve was significantly lower in volunteers receiving 200 mg, 300 mg, or 400 mg of caffeine compared with 0 mg (3.16 vs. 4.05; p < 0.001; 3.03 vs. 4.05; p < 0.0001; and 2.75 vs. 4.05; p < 0.0001), respectively. We found no significant difference in MBF reserve between receiving 100 mg and 0 mg (3.80 vs. 4.05; p = 0.31). Furthermore, using caffeine levels at T = 60 min, we found that MBF reserve was significantly negatively associated with increasing serum caffeine concentration (−0.12/mg/l, 95% confidence interval [CI]: −0.15 to −0.08; p < 0.0001) (Figure 1B). Similar results were found for MBF at stress. A significant positive linear association was found between coronary vascular resistance at stress and caffeine concentration (1.6 mm Hg/(ml/g/min)/(mg/l), 95% CI: 0.8 to 2.4; p < 0.001). In addition MBF reserve was significantly negatively associated with increasing concentration of serum paraxanthine (−1.4/mg/l, 95% CI: −1.9 to −1.0; (p < 0.0001).
Our flow results for caffeine doses ≥200 mg were in accordance with a previous study using administration of 3 mg/kg caffeine 1.5 h before scan with a mean plasma caffeine level of 3.1 mg/l. Lower caffeine levels were not investigated in that study (5). The linear correlation between serum caffeine and MBF reserve has to our knowledge not been shown previously, and positive correlation between coronary vascular resistance at stress and caffeine concentration supports that caffeine inhibits the vasodilator effect of adenosine and thereby affects MBF reserve. It is possible that one could find a significant difference in MBF reserve between 0 mg and 100 mg by adding more patients to a study. Also, results may differ in patients with coronary artery disease.
In conclusion, in healthy volunteers, MBF reserve decreased linearly with increase in serum caffeine and serum paraxanthine.
Please note: The authors have reported that they have no relationships relevant to the content of this paper to disclose.
- 2019 American College of Cardiology Foundation
- Belardinelli L.,
- Shryock J.C.,
- Snowdy S.,
- et al.
- Byrne C.,
- Hasbak P.,
- Kjær A.,
- Thune J.J.,
- Køber L.
- Kubo S.,
- Tadamura E.,
- Toyoda H.,
- et al.