The trial will have two study arms: active tVNS vs. sham tVNS. Patients will be randomized to active tVNS vs. sham tVNS and will receive optimal post-op care in both arms. Active tVNS (Parasym device, Parasym Health, Inc, London, UK) will be performed with a clip attached to the ear at 20 hertz (Hz), 250 microseconds (ms) at a current just below discomfort threshold for one hour twice a day, starting on post-day 0. For sham tVNS, Parasym device will be attached to the ear twice a day, turned on but current set to 0 milliamp (mA), starting on post-op day 0. Stimulation will continue until 5 days post-op or discharge. Discomfort threshold will be determined in both arms pre-operatively in the conscious state. This current will be used for stimulation for this patient until they are awake and extubated after surgery. The stimulation threshold may be reassessed once the patient is able to provide feedback.
Patients will be approached and recruited prior to their scheduled cardiac surgery. Recruited patients who give informed consent will have the discomfort stimulation threshold (current that leads to discomfort at the tragus) determined prior to surgery. Post-operatively, stimulation will be performed in the tVNS group at just below discomfort threshold. In the sham group, the stimulator will be turned on but current set at 0 mA. Stimulations will be performed within 12 hours of arrival to the ICU after surgery, and then twice a day between the hours of 7:00-9:00 am and 6:00-8:00 pm. If POAF develops in either arm, stimulation will be continued for the full 5 days. Ten ml of blood will be drawn within 12 hours of arrival to the ICU after surgery and on day 3 post-op for measurement of biomarkers. Serum will be stored at -80 Celsius and processed in batches of 10-15 samples.
Sample Size:
The investigators expect cardiac surgery to be associated with 40% incidence of POAF. The investigators expect tVNS to reduce this incidence by 40%. A sample size of 133 subjects per arm will be able to achieve 80% power at alpha of 0.05. If interim analysis is planned, Pocock method will be used and a p value of 0.03 will be used for interim and 0.03 for final analysis. Data will be analyzed according to the intention-to-treat principle.
Randomization:
A 1:1 randomization ratio for the tVNS vs. sham will be utilized. Patients who meet all of the inclusion and none of the exclusion criteria will be randomized in the order of their enrollment. After completing the Informed Consent process, the subject is then randomized following completion of the baseline and demographics information case report forms. Randomization should occur prior to any study-related tests or procedures. The subjects will be considered enrolled in the study once randomization has occurred.
Randomization will be stratified by clinical center and post-operative amiodarone use. A computer-generated randomization list with random permuted block of a variable will be produced for each clinical center. Investigators and other study staff members should not be able to identify the study assignment until this time. If a randomization assignment is inadvertently disclosed prior to use, the assignment will never be used.
A report of randomization compliance will be generated at the conclusion of the study.
