Key Points:
- Conventional medical wisdom warns against excessive coffee and/or caffeine consumption due to the perceived risk of arrhythmia.
- Most evidence against coffee consumption is derived from observational data.
- CRAVE is the first randomized trial to assess the effects of coffee consumption on cardiac arrhythmias in real time, using data from wearable technology. Participants who owned a smartphone and did not have a history of cardiac arrhythmias were enrolled into the trial, in which they continuously wore a FitBit Flex to track steps and sleep, a Ziopatch to monitor their cardiac rhythm, and a continuous glucose monitor to track coffee metabolism.
- Caffeine consumption was not associated with an increase in atrial arrhythmias but was associated with more PVCs, especially in faster metabolizers.
It is common practice to ask patients with palpitations to avoid caffeine consumption to prevent cardiac arrhythmias. This advice, while backed by guidelines from professional societies, is not rooted in evidence from randomized clinical trials. Furthermore, observational studies have been across the map with regard to coffee consumption and health effects, with some even associating improved serum glucose and overall mortality with higher caffeine consumption.
During a Late Breaking Session at the 2021 American Heart Association Scientific Sessions on Sunday, Dr. Gregory Marcus (University of California San Francisco) presented the results of the CRAVE trial, a first-of-its-kind randomized trial to assess the effects of caffeine consumption on cardiac arrhythmias in real time, using the assistance of several wearable devices. Enrollment consisted of healthy, volunteer adults who owned a smartphone and did not have a history of cardiac arrhythmias. Participants were given a Fitbit Flex wearable wrist band to track total steps taken and sleep activity, a continuous glucose monitor to monitor serum glucose and coffee metabolism, and a Ziopatch heart monitor for continuous ECG tracking.
On any given evening, participants were sent a text message with their daily assignment: avoid caffeine or consume at least one cup of coffee. Participants were instructed to trigger the Ziopatch when consuming coffee, and were texted questions to query their true caffeine consumption that day. Primary outcomes of interest were total daily counts of premature atrial contractions (PACs) and premature ventricular contractions (PVCs). 100 participants were randomized, and compliance was assessed using geo-location software for proximity to coffee shops, as well as time-stamped receipts of coffee purchase. In the intention to treat analysis, no significant difference was seen in the number of PACs per day based on coffee consumption, but coffee was associated with a higher number of PVCs (RR 1.54, 95% CI 1.19-2.0, p = 0.001).
Secondary outcomes of interest were total daily step counts, which were approximately 1,000 steps per day more in those who drank coffee than those who did not, minutes of sleep, which were approximately 36 fewer in those who drank coffee, and daily average glucose, which was unchanged between the two groups. After the presentation, discussant Dr. Sana Al-Khatib of Duke University applauded the investigators, but warned that the small sample size of the study and young population limit its generalizability. “I think it’s going to be critically important for future investigators to try and validate these results and look at harder endpoints that you follow over time,” she stated.