Coffee consumption linked to more PVCs, more physical activity and less sleep

The Coffee And Real-time Atrial And Ventricular Ectopy (CRAVE) Trial

News - Nov. 14, 2021

Presented at the American Heart Association’s Scientific Sessions 2021 by: Prof. Gregory M Marcus, MD - San Francisco, CA, USA.

Introduction and methods

Aim of the study

This randomized trial aimed to assess the real-time effects of coffee consumption on cardiac ectopy, physical activity, sleep and glucose levels.

Study design

A hundred adult volunteers (mean age 38±13 years, 51% women) who consumed coffee were randomized on a daily basis for 14 days to either consume as much coffee as they liked or avoid coffee and all caffeinated products. The participants received texts and reminders with the daily random assignments. Participants were fitted with a wrist-worn Fitbit device which tracked physical activity and sleep, a wearable continuous ECG device and a continuous glucose monitor. In addition, participants provided a saliva sample which was used to genotype DNA for caffeine metabolism-related SNPs. They were instructed to press a button on the ECG patch to time-stamp every coffee drink and filled in daily surveys about the actual coffee consumption on the previous day. In addition, geolocation data was collected to geofence visits to coffee shops. Costs for coffee purchases were reimbursed during the study.


Primary outcomes were daily premature atrial contraction (PAC) and premature ventricular contraction (PVC) counts. Secondary outcomes included daily supraventricular tachycardia (SVT) and ventricular tachycardia (VT) counts, daily mean step counts, nightly mean sleep duration and daily mean glucose levels.

Main results

  • Random assignment to coffee consumption was associated with more PVCs (RR 1.54, 95%CI 1.19-2.00, P=0.001). Individuals with fast caffeine metabolism were at higher relative risk of PVCs.
  • There was no significant relationship between random assignment to coffee consumption PAC counts, SVT episodes and VT episodes.
  • In per-protocol analyses, coffee consumption (per real-time coffee drink) was related to less SVT episodes.
  • Random assignment to drink coffee was associated with 1,058 more steps per day (95% CI 441-1675, P=0.0010) and 36 less minutes sleep per night (95%CI 22-50, P<0.001), compared to randomization to avoid caffeinated products. Individuals with slow caffeine metabolism experienced a more potent effect of coffee on sleep.
  • There was no significant relationship between random assignment to coffee consumption or per-protocol coffee consumption and daily mean glucose levels.


This study gave daily random assignments to a hundred volunteers to either drink as much coffee as they liked or to avoid all caffeinated products and found that coffee consumption resulted in more daily PVC counts, more physical activity as assessed by steps per day and less sleep duration. Coffeee consumption was not related to atrial arrhythmias. Prof. Gregory M Marcus, MD said that these results suggest that individuals prone to PVCs or ventricular arrhythmias may benefit from coffee abstinence, but genetic differences influence risk.


The discussant Prof. Elaine Hylek, MD (Boston, MA, USA) who admitted to be a coffee drinker, felt somewhat reassured by this small study that showed that coffee consumption does not appear to induce atrial arrhythmias. She was glad to see that coffee consumption appears to increase physical activity and hopes that this translates to better CV health. Regarding the finding that coffee consumption was related to increased PVC counts, she said that among healthy individuals with no CVD we should not get too worried about the PVCs.

-Our reporting is based on the information provided at the American Heart Association’s Scientific Sessions 2021-

Watch a video about the CRAVE trial

Facebook Comments


We’re glad to see you’re enjoying PACE-CME…
but how about a more personalized experience?

Register for free