The goal of this clinical trial is to learn the affect of melatonin on sleep, cognitive function, and quality of life (QoL) in patients with cirrhosis and a complication called hepatic encephalopathy (HE). The main questions this study aims to answer are:

• Does taking melatonin increase REM sleep, an important part of healthy sleep that is reduced in cirrhosis?
• Does taking melatonin improve cognitive function and reported QoL?

This is a pilot study, where participants will:

• take one month of melatonin, followed by one month of thiamine, which is another supplement but is not suspected to impact sleep significantly.
• Undergo cognitive testing and take surveys
• Wear a commercial wearable sleep tracker
• Have a formal sleep study and salivary melatonin collection at the end of taking each supplement at our sleep center Participants will be blinded, and neither they nor the researchers will know which supplement they are taking first and which they are taking second. They will also be randomized, with half starting with melatonin and the other half starting with thiamine.

Study Objectives

This project aims to determine the impact of melatonin supplementation on sleep physiology in patients with cirrhosis and HE.

Hypothesis: Melatonin supplementation will improve sleep physiology, and REM specifically.

Background and Significance

Hepatic encephalopathy (HE) affects approximately 50% of patients with cirrhosis, causing lasting cognitive and quality of-life impairments even with therapy. Most HE cases are covert, significantly increasing risks of hospitalization, mortality, and progression to overt HE (OHE) within three years (>50%). Guidelines recommend screening all patients for covert HE (CHE) using neurocognitive tools such as the Psychomotor Hepatic Encephalopathy Score (PHES). However, low screening and treatment rates persist due to the need for specialized training, equipment, and significant time commitments. Even in patients receiving HE therapy, clinicians are hesitant to escalate treatment absent overt confusion, largely due to the cost and poor tolerability of rifaximin and lactulose. These limitations contribute to frequent hospital readmissions and high short-term mortality in those with prior OHE. Improved recognition and management strategies are critical to advancing HE outcomes.

Sleep changes in HE are frequently described, often as disturbances in the sleep-wake cycle. The pathophysiology is likely multifactorial, including adenosine mediated hyperammonemic effects on arousal and alterations in endogenous melatonin metabolism in cirrhosis with resultant circadian rhythm dysfunction. Prior research confirms that patients with CHE likewise suffer from suboptimal sleep with associated quality of life reduction, and our research and others have noted reduced REM sleep, and increased sleep fragmentation. There may also be a bi-directional relationship between sleep decline and cognitive impairment in progressive OHE. Traditional gold-standard assessment of sleep physiology utilizes polysomnography (PSG), which requires on-site monitoring, application of multiple cumbersome leads, and costs up to $2000 for a night's study, making it impractical for routine assessment of disturbed sleep in cirrhosis. Consumer wearables have drastically increased in popularity for personal sleep assessment. Their ease of use, affordability, and promising performance in validation studies with PSG make them an attractive target for research. Utilization of wearables in cirrhosis may make sleep a more approachable target for monitoring in CHE. Melatonin is an endogenous pineal hormone with a known role in circadian maintenance with reduced hepatic metabolism and higher baseline serum levels in cirrhosis, perhaps contributing to circadian dysfunction. Prior research suggests that supplementing melatonin in patients with REM sleep disorders, poor subjective sleep, and cirrhosis improves subjective or objective sleep and wellbeing, but have never assessed the impact on cognitive function or CHE. Data suggests melatonin can improve REM sleep duration, sleep latency, and sleep disturbances, while being safe and affordable with high tolerability, but objective data is lacking in cirrhosis and large-scale randomized trials have not been done to date.

Overall Design

This study is a pilot randomized double-blinded, crossover study of melatonin. This design allows participants to serve as their own controls, reducing sample size needed, minimizing inter-individual sleep variability and permitting 1:1 randomization, provided proper washout and analysis for period effect, addressed below. Randomization will be performed in Stata, with a single unblinded coordinator. Each participant will have a 2-week run in for baseline home sleep assessment, 4 weeks on 3 mg nightly melatonin or 100 mg thiamine, a 1-week washout, and a 4-week period on the alternate therapy for a total of 11 weeks in the study. The washout period was chosen from previously published washouts of melatonin. Thiamine was chosen to approximate placebo due to its identical taste/appearance, affordability and lack of hepatotoxicity. After a two-week lead in, the investigational pharmacy will give participants an identical 30-day supply of either 3 mg melatonin tablets or 100 mg thiamine tablets who will be instructed to take it nightly 30-60 min before intended bedtime. A MEDLINE search does not identify any published reports of thiamine impacting melatonin secretion, sleep physiology or sleep quality. Both melatonin and thiamine are dietary supplements and do not require an investigational new drug application (IND) for assessment of sleep physiologic effects. After the first sleep study is completed the bottle will be collected and the alternative therapy will be given, with instructions to start after a one-week washout. Sleep will be tracked throughout the entire study period via the Oura ring.

Inclusion Criteria:

• Cirrhosis with clinically significant portal hypertension or decompensation defined by Baveno VII criteria [de Franchis R et al 2022]
• Adults over age 18
• CHE (defined by PHES≤ -4) or previously diagnosed HE
• Disturbed sleep, with Pittsburgh Sleep Quality Index (PSQI) ≥5
• Possession of a "smart phone" with Bluetooth capability and ability to download the Oura application (Apple iOS version 14.0 or greater or Android version 8.0 or higher)

Exclusion Criteria:

• Use of melatonin regularly (3x per week) if unable/unwilling to discontinue for the study

• Inability provide informed consent

• Heavy current alcohol use (>7 drinks weekly for women and 14 drinks weekly for men)'

  -- Body mass index >40

• Known prior sleep disorder including obstructive sleep apnea

• Use of other prescription neuromodulating sleep aides

• Self-reported pregnancy during study screening, as sleep physiology is different in this population

de-identified study data that underlie results in a publication.

Beginning 3 months and ending 5 years following article publication

Researchers who provide a methodologically sound proposal, to achieve aims in the approved proposal. Proposals should be directed to apb9012@med.cornell.edu. To gain access, data requestors will need to sign a data access agreement. Data will be available for 5 years after publication at a third party website.