The investigators are studying the use of Melatonin in non-ventilated patients over the age of 65. Primary endpoint will be assesments of delirium, with secondary endpoints to include length of stay, use of anti-psychotic medications, and mortality

Summary Melatonin is a naturally produced hormone involved in the synchronization of circadian rhythms and the sleep-wake cycle in humans. It is often given as a sleep aid and is generally well tolerated, with very few drug-drug interactions. Recent studies have shown marginal benefit in increasing amount of sleep, but have proven benefit in decreasing delirium in elderly patients on medical wards.

Delirium is a serious problem in hospitals and Intensive-Care Units (ICUs), contributing to longer hospital stays as well as morbidity and mortality. Studies have shown that each additional day of delirium increases mortality by 10%1. It is estimated that delirium costs the US healthcare system $38-$152 billion annually, and the cost of treating a delirious patient is 2.5 times higher than treating one without delirium. It is estimated that up to one third of cases of delirium are preventable. Additionally, many of the medications given in the ICU can worsen delirium, including narcotics, antipsychotics, and benzodiazepines.

Treatment of delirium consists mostly of prevention, by frequently re-orienting the individual, letting in as much natural light as possible, frequent touching, and providing a quiet place for rest. Pharmacological treatment consists mostly of antipsychotics, which often have adverse effects and can worsen delirium.

Incidents of delirium are much higher in elderly patients, up to 80% in some cohorts, as well as ICU patients. Multiple theories for this exist, including the lack of a true rest time at night, multiple sedatives that can affect sleep-wake cycle, and large amounts of noise. It is also known that elderly patients have less total sleep and more nighttime arousals, and that ICUs in particular are quite loud, especially at night. The increase in ambient noise paired with frequent interruptions in sleep for monitoring and blood draws exacerbates this problem and likely leads to a delirious state.

To our knowledge, no study has looked at melatonin's effect on elderly, non-ventilated patients in the ICU, which is the group most at risk for developing delirium. Recent studies have indicated that delirium is decreased with melatonin use in a simulated ICU environment.2 This randomized, controlled study would aim to confirm these results in an actual ICU environment.

Additionally, it is unknown what doses are most effective for treatment of delirium. Literature has shown improvement with both 3 mg and 5 mg, but recent meta-analysis have not shown benefit of one over the other. 3 Participants Multiple studies done on critically ill, non-ventilated patients in the ICU show an incidence of delirium of at least 60%, and in some cases up to 80%4. At Parkview, the highest incidence recorded was 55%, but this was with all patients. It is likely that this percentage is much higher in the elderly. Assuming an average incidence of 70%, and assuming the investigators would want to be able to detect a 20% decrease in this incidence with our intervention, the investigators would aim to enroll 600 patients (200 in each arm), with an interim analysis to be done at 300 patients to asses for harm or efficacy. This will be done by analyzing the primary outcome for a P-value of <0.05.

Exclusion Criteria Patients with head trauma or Neurosurgical intervention Patients <65 years of age Patients with an expected life expectancy <48 hours Blind patients Patients with a seizure history Patients with uncontrolled hypertension Patients with a supratheraputic (>3.0) INR Patients on strong CYP1A2 inhibitors: ciprofloxacin, fluvoxamine, methoxsalen, ofloxacin, primaquine Patients who do not speak English or Spanish

Design Upon admission to the ICU, all consecutive, eligible patients or their POA will be presented with an informed consent describing the trial (Appendix A). All patients over 65 will be screened. Attending physicians as well as Critical Care Fellows will be allowed to consent the patients. At the point of deciding to join the trial, an order will be put in the computer for "melatonin protocol". At this point, pharmacy will assign that patient to either the placebo, 3mg, or 5mg arm of the study in a 1:1:1 fashion. Physicians, pharmacists, and nursing staff will be blinded to the protocol. Melatonin dosages of 3 and 5mg will be used for the study, as recent work with higher doses has been shown to have a "carryover" effect into the following day. The placebo pill will be a small white tablet of similar size to the 3 and 5mg tablets of melatonin. It is sucrose and is regulated by the FDA as safe for human consumption. Either placebo or melatonin will be administered at 9pm with evening medications.

The treatment group will be given either the assigned dose or placebo nightly. Either PO or PT would be acceptable. The protocol will run until the patient is discharged from the hospital. Delirium will be assessed using the CAM-ICU evaluation tool in the ICU and bCAM on the floor according to pre-existing protocols. These will be assessed every 12 hours per a protocol already in place. These tools are the most widely used in the US, with multiple studies showing reliable detection of delirium. The CAM scale is attached for reference.

If a patient deteriorates while in the ICU and requires intubation, they will be continued on the protocol.

Melatonin is generally well-tolerated5, but the two most common side effects are headaches and confusion. We hope to mitigate the confusion by using a lower dose of melatonin. Incidence of headache will be tracked, and if deemed necessary by the attending physician, melatonin therapy can be stopped. Melatonin has minor interactions with Warfarin and dihydropyridine calcium channel blockers, but patient's INR (if they are on Warfarin) and BP will be tracked routinely. Melatonin is metabolized by the CYP1A2 system, and so patients on medications that are potent inhibitors of this enzyme will also be excluded from the study. There is an interaction between Melatonin and Olanzipine, which is sometimes used for treatment of delirium. There is the potential for increased sedation with this combination, but patients will be monitored in an ICU setting while receiving these medications.

Primary outcome will be incidence of delirium (using the CAM-ICU and bCAM scores) in the study population

Secondary outcomes will be:

ICU days Total Hospital days 30 and 90-day mortality Use of anti-psychotic medications for treatment of delirium (Haloperidol, Ziprasidone, Olanzapine, Quetiapine) Average Richmond Agitation and Sedation Scale (RASS) scores when delirium assessed

Hypothesis The investigators' hypothesis is that the use of melatonin will significantly decrease the incidence of delirium in the elderly, non-ventilated patient. The investigators will use an accepted P-value of < 0.05 to determine significance. If the p-value doesn't meet significance, the null hypothesis that melatonin does not decrease the incidence of delirium in the elderly non-ventilated patient will be accepted.

Collection/Retention of information:

Patients' CAM-ICU scores are already documented in Meditech. Other variables (when patient moves to floor, when the patient is discharged) are already routinely documented. Following completion, these variables will be placed in a database and de-identified. The data will be kept for an additional 3 years to allow for further analysis if required.

Nursing staff as well as house staff will undergo a brief in-service on the study principles as well as a reminder of how to use the CAM-ICU scores.

Confidentiality:

Only the PI will see the patient's identified information before deidentifying it. To ensure patient privacy, the informed consent discussion will happen in the patient's private room

Risks:

Melatonin has been proven to safe in every patient population by multiple studies. A consensus statement by the World Sleep Foundation in 2000 confirms that there have been no significant side-effects in healthy adults. Specifically, the use of melatonin in seizure disorders is controversial, with some data showing detriment and some showing benefit. For this reason, patients with a seizure disorder have been excluded from the study. The investigators will be using USP (United States Pharmacopeia)- certified melatonin as well. USP is a non-profit organization that regulates over-the-counter (OTC) supplements in the US. Its seal verifies that the manufacturer adheres to good manufacturing practices.

The attending physician on the case can choose to withdraw the patient from the study at any time if they feel that participation is adversely affecting the patient.

Compensation:

There will be no compensation provided for participation in this study.

Benefits:

Considering the multiple studies that have suggested safety and efficacy, critically ill patients stand to benefit from a lack of delirium and possibly shorter hospital stays. This would benefit Parkview Medical Center as well.

Inclusion Criteria:

• Non-ventilated Patients over the age of 65

Exclusion Criteria:

• Patients with head trauma or Neurosurgical intervention
• Patients <65 years of age
• Patients with an expected life expectancy <48 hours
• Blind patients
• Patients with a seizure history
• Patients with uncontrolled hypertension
• Patients with a supratheraputic (>3.0) INR
• Patients on strong CYP1A2 inhibitors: ciprofloxacin, fluvoxamine, methoxsalen, ofloxacin, primaquine
• Patients who do not speak English or Spanish