This randomised double-blind study was carried out at the Department of Anaesthesiology, King George Medical University, Lucknow, Uttar Pradesh. After approval by the Institutional Ethics Committee (89^th ECM II B Thesis/P14), patients age more than 65 years, admitted to Gandhi Memorial and Associate Hospitals, King George’s Medical University, Lucknow, undergoing surgery at various surgical departments were recruited for this study.

Patients belonging to the American society of Anesthesiologists (ASA) physical status Grade 1 and 2 and planned for surgery requiring neuraxial anaesthesia for duration longer than one hour were included. Patients with history of dementia, severe infections, intracranial bleed, having any acute cardiac event and taking any psychiatric medications were excluded from the study. Hundred patients were randomly distributed in two groups with each group containing 50. It was calculated to be 50 in each group on the basis of sensitive and specification of CAM [17] using the formula:

where Sn=0.8 sensitive; Sp=0.959 specificity CAM; d=2, effect for multiple assessment parameters; Type I error α=5%; Type II error β=20% for 80% power of study.

Randomisation was done according to a computer-generated list. Group I (Ramelteon group): Received oral ramelteon; one tablet of ramelteon (8 mg) 12 hours before surgery and 1 tablet of ramelteon (8 mg) 1 hour before surgery.

Group II (Control group): Received placebo tablets; one tablet 12 hours before surgery and 1 tablet 1 hour before surgery [Table/Fig-1].

[Table/Fig-1]:

CONSORT Flow Diagram.

After obtaining written informed consent from patients, the complete medical history, physical examination and mini-mental examination were done for all patients [Table/Fig-2]. Study drug was given in a sealed envelope to all patients by the anaesthetist who was not involved in the observation of the patients and asked to take 1^st dose 12 hours before surgery and 2^nd dose 1 hour before surgery. It was a double-blind study as observer was not involved in giving the study drug to patients and patients were also not aware which study drug was given to them. No sedative drug was given to the patient preoperatively. After arrival in an operation theatre, standard monitors were attached which include pulse oximeter (SpO₂), ECG, non-invasive blood pressure monitoring. All these parameters were recorded and monitoring continued throughout intraoperatively. On arrival in operation theatre, patients Ramsay sedation score was noted and thereafter every hour till completion of surgery. All patients were given combined spinal and epidural anaesthesia. On the 1^st day of surgery, epidural infusion was used for post-operative analgesia. On 2^nd day and 3^rd day, Inj. Paracetamol 1 gm IV 6 hourly was used for post-operative analgesia.

Delirium was assessed by CAM on 1^st, 2^nd and 3^rd day [Table/Fig-3]. Intraoperative Ramsay sedation score [Table/Fig-4] was observed. Intraoperative haemodynamic (BP, Heart Rate, SpO₂) were recorded at baseline to 30 minutes at every 5 minutes and subsequently every 15 minutes. Post-operative Ramsay sedation score was observed immediately after surgery, after 6 and 24 hours.

All the data for continuous variables (age, disease duration, HR, MAP, SpO₂ and MMSE) are shown as the mean±standard deviation and the categorical variable (gender) is shown as a percentage. The total scores on the MMSE were calculated as the sum of single items. The Chi-square test was used to compare the categorical variables between the groups. Unpaired t-test was used to compare the continuous variables between the groups. The SPSS Version 21.0 statistical Analysis Software was used for the statistical analyses. p<0.05 was considered to indicate a statistically significant difference.

The demographic variables such as age, gender, ASA grade and baseline hemodynamic parameters were comparable between Group l and Group ll as shown in [Table/Fig-5]. This study included patients of age between 65-82 years. Mean age of patients was comparable in between group I (69.30±4.05 years) and Group II (70.64±3.76 years) and male preponderance was observed. Pre-operative heart rate, systolic BP and diastolic BP of patients were found to be comparable in both groups (p>0.05).

At the time of enrolment none of the patients in the study was found to be suffering from dementia, severe infection, intracranial events, MI, CHF, arrhythmias, COPD or Pulmonary embolism. Associated illness in the population was higher in Group II as compared to group I for prevalence of hypertension (22.0% vs. 20.0%) and diabetes (18.0% vs. 16.0%) but these differences were not found to be statistically significant. Whereas prevalence of Asthma was higher in group I as compared to Group II (6.0% vs. 4.0%) but this difference was also not found to be statistically significant.

Mental status of the patients was assessed using Mini-Mental Status examination. MMSE score was found to be higher in group I (27.22±1.33) as compared to that of Group II (27.19±1.82) but this difference was not statistically significant.

Patients were admitted for surgical intervention of different ailments and were admitted in the respective speciality wards. Among patients of Group I admissions to Urology wards were most common followed by Orthopedics while among patients of Group II admissions to Urology and Orthopedics wards were most common. Proportional difference in patients of above two groups admitted to different speciality wards was not found to be statistically significant (p=0.474).

Mean Heart rate of patients of Group I was found to be higher as compared to that of Group II except at 25 minutes and 150 minutes. Difference in mean heart rate of patients of above two groups was not found to be statistically significant at any of the periods of observation [Table/Fig-6].

[Table/Fig-6]:

Comparison of heart rate (Beats/minute).

Systolic BP of patients of group I was found to be higher as compared to that of Group II at baseline, 5 minutes, 20 minutes, 25 minutes, 30 minutes while at rest of the periods of observation, SBP of Group II was higher as compared to that of group I, but difference in mean systolic BP of patients of two groups was found to be statistically significant only at 25 minutes (119.46±11.68 vs. 111.92±5.15 mmHg) [Table/Fig-7].

[Table/Fig-7]:

Comparison of Systolic Blood Pressure (mm of Hg).

Mean Diastolic BP of patients of Group II was found to be higher as compared to that of Group I at all the periods of observation except at 25 minutes, and differences in mean diastolic BP of patients of above two groups was found to be statistically significant at all the periods of observation except at baseline, 5 minutes, 20 minutes and 30 minutes [Table/Fig-8].

[Table/Fig-8]:

Comparison of Diastolic Blood Pressure (mm of Hg).

Oxygen saturation level of all the patients during the period of observation was maintained above 95% hence this difference has no clinical relevance.

Sedation score of all the patients just before surgery was 2.00. Mean sedation score of patients of group I (2.32±0.62) was found to be higher as compared to that Group II (2.08±0.63). Difference in Ramsay sedation score of patients of above two groups was not found to be statistically significant at any of the period of observation [Table/Fig-9].

Confusion assessment method was used to assess the post-operative delirium on follow-up at Day 1, Day 2 and Day 3. Total incidence of delirium was 8%; lower in group I as compared to Group II (4% vs 12%). On day 2, one patient developed delirium in group I while 3 patients developed delirium in Group II but difference was not found to be statistically significant. On day 3, one patient developed delirium in group I and 3 patients in Group II but difference was not statically significant. None of the patients, irrespective of any of the groups was positive for CAM on follow-up at Day 1 [Table/Fig-10].

Post-operative delirium is one of post-operative neurological complication, which is seen, in elderly patients who are of age more than 65 years. This leads to prolonged hospital stay and an increasing need for rehabilitation and home care. So, this study was conducted to evaluate the effect of Ramelteon on post-operative delirium as primary objective and its effect on haemodynamic responses and sedation scores in elderly patients as secondary objectives.

Demographic variables were compared between two groups. The aetiology of post-operative is multifactorial and some of the predisposing factors are like pulmonary disease, pre-operative cognitive dysfunction, patient on antiarrhythmic and use of opioids [18,19]. These factors were comparable between two groups in this study. Post-operative delirium was determined using CAM on day 1,2,3. CAM has more than 90% sensitivity and specificity in elderly patients [20,21]. Wei LA et al., found CAM scale to be a reliable tool to evaluate post-operative delirium if used by trained personnel [22].

In this study, it was observed that the incidence of delirium in Ramelteon group was lower in comparison with placebo group but it was not statistically significant; similar to other studies that reported that Ramelteon prolongs the time to development of delirium and decreases the frequency of delirium [12,16]. Hatta K et al., did a multicenter randomised placebo-control trial in ICU and observed that there was decrease in delirium after Ramelteon administration (3% vs 32%, p=0.003) [12]. Miyata R et al., did a retrospective analysis of patients of age 70 years or more undergoing anatomical pulmonary resection for lung cancer and observed that incidence of delirium was lower in the Ramelteon group in comparison to control group (p=0.061) [15]. Similarly, Booka E et al., studied 65 patients who underwent pharyngo-larygectomy with oesophagectomy and observed that Ramelteon was associated with significantly decreased incidence of delirium in post-operative period [16].

In this study, post-operative delirium developed on day second and day third. Whitlock EL et al., also observed that post-operative delirium usually peaks between first and third day [19]. The total incidence of delirium in this study was 8% which is lesser than previously reported studies [15,16]. The difference may be due to inclusion of patients undergoing surgery only under regional anaesthesia. Although perioperative anaesthesia can affect the development of post-operative delirium, all patients in this study received similar perioperative anaesthesia technique and analgesic administration under the same perioperative anaesthetic policy.

Intra-operatively, the haemodynamic variable such as heart rate was comparable in both Ramelteon and placebo groups but there were more changes in placebo group. These findings are supported by other studies, that reported that from the baseline values, the change in heart rate between the study groups were found to be non-significant at any given time interval [23,24]. Although no significant decrease in MAP was noted in the Ramelteon group but melatonin was found to have a mild hypotensive effect due to the direct action on vascular melatonin receptors and the reduction of the adrenergic outflow and catecholamine levels [25]. This property of melatonin was utilised in eye surgeries where 10 mg melatonin given orally 90 minutes before the operation in cataract surgery patients provided significant sedation and lowering of intraocular pressure [26]. In the present study, the overall mean arterial pressures were found to be on higher side in Ramelteon as compared to placebo groups although it was not statistically significant. The difference in SpO₂ was not found to be statistically significant among different study groups at any given time intervals. Throughout the intraoperative period, the SpO₂ was maintained at 96-100% with no episode of fall in saturation at any point of time.

The difference in Ramsay sedation score of patients of the two groups was not statistically significant at any of the period of observation. Patients in the Ramelteon group had higher sedation scores as compared to placebo group but none had respiratory depression. Mallick S et al., compared the effect of Ramelteon and placebo on perioperative sedation in cases of laparoscopic cholecystectomies and they also reported that the sedation score was significantly higher in Ramelteon groups compared to the placebo group after 30 minutes and 60 minutes of administration of drug [27]. Borazan H et al., and Naguib M et al., also reported that the sedation score was markedly higher in the melatonin group than in control group [28,29]. Various studies suggested that the Ramelteon increased sedation effects in a dose-independent manner [30-32].

The main limitation of this study is that it is a single-centre study. Also, all surgical specialties were included, which could have effected the total incidence of delirium so studies focusing on single speciality patients are required.

Ramelteon decreases the incidence of post-operative delirium with higher sedation scores. So Ramelteon can be recommended in patients having high risk of developing post-operative delirium. The scope of further research on Ramelteon includes the use of Ramelteon in higher doses and post-operative continuation of Ramelteon, and its effects on the delirium, sedative effects and hemodynamic variables can be studied.