Melatonin for delirium prevention in acute medically ill, and perioperative geriatric patients

Abstract Delirium is a challenging neuropsychiatric ailment that has a negative impact on morbidity and mortality and is difficult to treat once it has developed. The purpose of this review was to analyze the efficacy of melatonin in the prevention of delirium in hospitalized geriatric patients in the acute medically ill and perioperative wards. The databases searched included PubMed (1946 to February 12, 2020), CINAHL (1982 to February 12, 2020), EMBASE (1974 to February 12, 2020), and Web of Science (1900 to February 12, 2020) using search terms related to melatonin, delirium, and prevention. Meta‐analyses, randomized controlled trials, and observational studies were included. We excluded publications pertaining to the intensive care unit or oncology, case reports/series, and those not in English. Seven full‐text publications were included for qualitative analysis. Patient comorbidities, patient medications, melatonin dosing, dosing regimens, and duration of treatment varied between the studies, which yielded heterogeneous results. Overall, this literature review yielded four studies that showed positive results and three that showed negative results for delirium prevention. The current data for the use of melatonin in delirium is conflicting. This area requires further research of more homogeneous studies with larger sample sizes.

there has been little success in finding an effective agent for the prevention of delirium in the hospital setting.
Hospitalized patients often have disruption in their circadian rhythm upon admission to hospital. The hospital environment is always active with noise and light and can be stressful or unsettling for a patient, which can be detrimental to a patient's sleeping cycle. Disruption of sleep and circadian rhythm has been found to be a risk factor for the development of delirium. 10 Melatonin is an endogenous hormone synthesized from tryptophan, and is an important regulator of the circadian rhythm. Melatonin is normally released later in the evening when stressors are low and there is less sunlight exposure. 10 It is postulated that supplementation of melatonin in the hospital setting to help mimic the circadian rhythm may aid in preventing delirium in patients at risk. 11 In addition, there have been observations that elderly patients have disturbed melatonin secretion patterns, and that patients with high risk of delirium may have lower levels of melatonin. 12,13 All of these factors point to the potential role of melatonin in delirium. The purpose of this review was to summarize the current evidence for melatonin on its efficacy, use, and dosing for preventing delirium in hospitalized elderly patients.

| Search strategy
A literature search was performed using PubMed (1946 to February 12, 2020), CINAHL (1982 to February 12, 2020), EMBASE (1974 to February 12, 2020), and Web of Science (1900 to February 12, 2020). The search terms used were "melatonin," "delirium," and "prevention." The search was limited to English-language publications only. Articles yielded from the search were then analyzed against this review's inclusion criteria by their titles, abstracts, and full texts.
A manual review of the references of the available literature was performed with relevant literature included.

| Study selection and data extraction
For the purpose of this review, only original articles on geriatric patients in the acute medically ill and perioperative patient populations were included. The geriatric population included any individual who was ≥ 65 years old. The acute medically ill included any patient with a medical condition that had presented with a severe and abrupt onset that was expected to resolve within approximately 6 months. Lastly, "perioperative" was defined as the time period of a patient's surgical procedure, including admission, anesthesia, surgery, and recovery. Meta-analyses, randomized controlled trials (RCTs), and observational studies were included in this review. Studies conducted in the intensive care unit and oncology units were excluded. The study inclusion process is outlined in the PRISMA flow diagram in Figure 1.

| RE SULTS
A total of one meta-analysis, 11 five RCTs, [14][15][16][17][18] and one observational study 19 were included in this analysis. A summary of the studies is presented in Table 1.
The studies were grouped into either acute medically ill or perioperative, given the heterogeneity of these two populations. The number of patients in each study varied from 161 patients to 500 patients, who were split into intervention and comparator groups.
All of the studies compared melatonin to placebo with the exception of Sultan et al, 17 which compared melatonin to midazolam 7.5 mg, clonidine 100 µg, and a placebo group. Delirium was assessed with a number of different assessment tools, with the most common being the Confusion Assessment Model (CAM). 11,15,16,18 There were varying results reporting the efficacy of melatonin in the prevention and treatment of delirium. Three studies 14,15,18 had negative results, suggesting melatonin is not associated with the prevention of delirium in hospitalized and postoperative patients. In contrast, four studies 11,16,17,19 yielded positive results, showing a decrease in delirium rates in melatonin intervention groups.

| Dosing of melatonin
Doses of melatonin used in studies ranged between 0.5 mg and 5 mg once daily. There was variability of when the melatonin was administered: For patients undergoing hip arthroplasty, melatonin was administered the night before the operation, again at 90 minutes before the surgery, and then for 5 consecutive days postoperatively. 14 In contrast, for patients admitted to the internal medicine ward in Al-Aama et al, 16 melatonin was administered prior to sleep between 6:00 pm and midnight for 14 days.

| Acute medically ill geriatric patients
There was conflicting evidence on the association between melatonin administration and delirium prevention in the acute medically ill geriatric population. In a study on geriatric patients admitted to an internal medicine ward, Jaiswal et al 15  Scale (MDAS) was used to assess delirium severity, and there was no statistically significant difference between groups. 16 Chen et al 11 carried out a meta-analysis of four RCTs, three of which are included in this review (de Jonghe et al, 14 Al-Aama et al, 16 and Sultan et al 17 ) and one of which met our exclusion criteria due to its analysis of intensive care unit patients and assessment of ramelteon's efficacy. 20 Chen et al 11 analyzed Al-Aama et al 16

| Perioperative geriatric patients
De Jonghe et al 14

| D ISCUSS I ON
The geriatric population is at higher risk of developing delirium as well as other complications, including fall risk, frailty, and sensitivity to medications, given their pharmacokinetics. In the studies reviewed, the DSM-5 criteria for delirium, the CAM, the MDAS, and the Abbreviated Mental Test were used solely or in different combinations to assess presence of delirium in patients. The DSM-5 is considered to be the gold standard for assessment and diagnosis of delirium in medical patients; however, it requires a trained psychiatric professional and is too cumbersome for most situations. 21 The CAM, which can be completed in 5 minutes by non-psychiatric trained professionals, is the most widely accepted alternative to the DSM-5 in assessing delirium, given its ease of use at the bedside and has been well validated in multiple studies. The Abbreviated Mental Test is considered a part of the delirium assessment but not a complete assessment of the condition. 22 The MDAS was used to assess properly detect the effect of melatonin on delirium incidence. In current practice, melatonin is not routinely used as a delirium prevention agent, but it is used for other purposes, such as a sleep aid.
This may enable a larger study sample to be assessed with one of the standardized diagnostic tools. In addition, an effective dose should be established, as well as a dosing interval and a time interval for evaluating delirium with a diagnostic tool.

| Ramelteon
Ramelteon is a melatonin receptor agonist that has similar action to exogenous melatonin. Ramelteon is found to have a high affinity and selectivity for melatonin receptors 1 and 2 versus melatonin. 23 Ramelteon has been approved by the US Food and Drug Administration for the treatment of insomnia and has been studied as an option for preventing delirium in a case series. 24 In a randomized placebo-controlled trial, ramelteon was associated with a statistically significant decrease in incidence of delirium from 32%

| Why should we continue to study melatonin?
Delirium can have mortality and morbidity consequences in the geriatric population admitted to hospital, given their comorbidities and frailty. These patients are likely to be subjected to polypharmacy and are at high risk of medications' adverse effects and drug-drug interactions. 25,26 Many of the medications studied in the prevention of delirium have not been found to be effective, and oftentimes they are on the American Geriatrics Society Beers Criteria list as having a high potential of complications in geriatrics. 27 Also, once delirium develops, geriatric patients are often given medications such as benzodiazepines and antipsychotics, which have been associated with worse delirium and increased morbidity and mortality. [28][29][30] Melatonin, with its relatively high safety and tolerability profile, has been used widely for sleep disorders. While its role for delirium prophylaxis remains unclear, it remains an attractive research subject for delirium, especially in the geriatric population.

| CON CLUS ION
Overall, our results suggest that melatonin possibly plays a role in delirium prevention in hospitalized and postoperative geriatric patients; however, the results are conflicting. Inconsistencies in methodologies, assessment tools, and dosing result in much heterogeneity in the evidence. It may be reasonable to direct research efforts in a more focused subset of patients who are at higher risk of developing delirium. Regardless, in order to determine the efficacy of melatonin in this population, more RCTs with larger sample sizes and more sophisticated study designs are needed.

CO N FLI C T S O F I NTE R E S T
We have no conflicts of interest for conducting this review.

AUTH O R CO NTR I B UTI O N S
Both authors: writing of the paper, substantially contributed to conception or design, gave final approval. Demi R. Asleson: drafted the manuscript, contributed to acquisition and interpretation of data.
Ada W. Chiu: critically revised the manuscript for important intellectual content.