The objective of this study is to test the effects of caffeine on neurocognitive and clinical recovery after major surgery. Specifically, this trial tests the primary hypothesis that caffeine will reduce the incidence of postoperative delirium.

Delirium is a syndrome characterized by failure of basic cognitive functions that affects approximately 20-50% of older surgical patients. Delirium during surgical recovery is associated with increased mortality, cognitive and functional decline, and prolonged hospitalization. In fact, 3-year survival rates for acutely hospitalized patients with delirium, and subsyndromal delirium, are both less than 50%. Older age is predictive of delirium after surgery, and with aging surgical populations, the incidence of postoperative delirium and related complications are likely to increase in the coming years.

Caffeine represents a novel, neurobiologically informed candidate intervention for reducing risk of early postoperative delirium. Caffeine promotes arousal via adenosine receptor antagonism and improves cognitive function concurrent with increased cortical cholinergic tone. Our preliminary data suggest that caffeine reduces risk of postanesthesia care unit (PACU) delirium in adult non-cardiac surgery patients by optimizing cortical dynamics for cognition. Caffeine also optimizes key neurocognitive processes that support information processing and may improve other, related aspects of clinical recovery, such as rebound headache in habitual caffeine users. The objective of this trial is to thus test the effects of caffeine on neurocognitive and clinical recovery after major surgery. Specifically, the primary hypothesis is that caffeine will reduce the incidence of postoperative delirium. The secondary objectives are to (1) test whether caffeine positively impacts the quality of postoperative recovery via validated patient-reported measures and (2) identify neural correlates of delirium and Mild Cognitive Impairment via advanced electroencephalographic (EEG) analysis.

Inclusion Criteria:

1. Adult (>/= 70 years old) undergoing non-cardiac, non-intracranial neurologic, non-major vascular surgery requiring general anesthesia with a planned admission for at least 48 hours.
2. Stated willingness to comply with all study procedures and availability for the duration of the study.
3. Provision of signed and dated informed consent form.

Exclusion Criteria:

1. Emergency surgery
2. Outpatient surgery
3. Severe cognitive impairment precluding the capacity for informed consent
4. Seizure disorder history
5. Intolerance or allergy to caffeine (based on subjective reporting or objective documentation)
6. Weight >130 kg (as a 3 mg/kg dose would approach the upper limit of daily intake recommended by the FDA)
7. Enrollment in conflicting research study
8. Patients in acute liver failure
9. Acute kidney injury preoperatively
10. Diagnosis of pheochromocytoma
11. Active carcinoid syndrome
12. Severe audiovisual impairment
13. Non-English speaking

Clinical and electroencephalographic data will be preserved and made available for sharing. Data will be saved in various formats, such as comma-separated values (CSV), open document format (ODF), and European Data Format (EDF) for waveform data. Software programs designed to accommodate electroencephalographic waveform data (e.g., MATLAB) will be required for data access and analysis. Specific software and coding recommendations can be made available upon data sharing request.

Any data not permitted to be shared, as specified in written informed consent documents, will be withheld. Likewise, no data will be shared that may compromise the safety or privacy of participants, and no data will be shared that would violate local, state, or federal laws or policies (including the Health Insurance Portability and Accountability Act). All data will be de-identified.

Scientific data will be made available after primary trial manuscripts (for both clinical and electroencephalographic data) are accepted for publication. Data will be made available for as long as possible in compliance with National Institutes of Health and University of Michigan policies.

Data will be shared upon execution of Data Use Agreements in compliance with University of Michigan policies and procedures. Interested researchers will need to contact the study PI (Dr. Vlisides) for obtaining data via Data Use Agreement. All data will be shared upon reasonable request after primary trial manuscripts are published.

This will be a single-center, quadruple-blinded, randomized control trial at Michigan Medicine