Study design and patients

This randomised, double-blind, placebo-controlled trial will be conducted at Weifang People’s Hospital. The study flow chart is presented in figure 1. Our study protocol is reported following the Standard Protocol Items: Recommendations for Interventional Trials statement (figure 2). The research protocol has received approval from the local biomedical research ethics committee on 23 May 2024 and registered with the Chinese Clinical Trial Registry on 4 June 2024 (ChiCTR2400085253). The study commenced with the enrolment of the first patient on 6 June 2024 and is projected to conclude by 6 June 2026. All participants will sign a written informed consent prior to their inclusion in the trial (see online supplemental file 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

The schematic flow chart for this study. MoCa, Montreal Cognitive Assessment; NRS, Numeric Rating Score; PSQI, Pittsburgh Sleep Quality Index; RASS, Richmond Agitation-Sedation Scale; RCSQ, Richards–Campbell Sleep Questionnaire.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Schedule of enrollment, main intervention and assessment (Standard Protocol Items: Recommendations for Interventional Trials figure). GSD-15, Geriatric Depression Scale .

Inclusion criteria

Patients who meet the following criteria will be recruited: (1) age 65 years or older, but younger than 90 years; (2) conform to the American Society of Anesthesiologists grades I–III; (3) scheduled for elective TKA under general anaesthesia; (4) transferred to a general ward before 19:00 on the day of surgery and (5) consente to patient-controlled intravenous analgesia (PCIA) for postoperative pain management.

Exclusion criteria

In this protocol, patients will be excluded based on the following criteria: (1) emergency surgery; (2) a diagnosis of sleep apnoea syndrome or a STOP-Bang score≥3; (3) severe cardiac dysfunction (preoperative Left Ventricular Ejection Fraction <30% or New York Heart Association class IV), severe liver dysfunction (Child–Pugh class C) and severe renal dysfunction (necessitating preoperative dialysis); (4) expected to be transferred to the intensive care unit (ICU) after surgery; (5) allergy to any components of ketamine, esketamine and dexmedetomidine, as well as currently undergoing treatment with these medications; (6) sleep disturbances within the past month requiring medical intervention, such as the use of hypnotic drugs; (7) preoperative conditions, including sick sinus syndrome, severe sinus bradycardia with a heart rate<50 beats per minute, atrioventricular block of second degree and higher without the presence of a pacemaker; (8) an inability to communicate and cooperate with the study due to severe dementia and language barriers; and (9) a preoperative history of schizophrenia, epilepsy, parkinsonism and myasthenia gravis.

Randomisation and blinding

Random numbers are generated using SAS V.9.3 software (SAS Institute, Cary, North Carolina, USA) with a 1:1 allocation ratio and block sizes of 4. Throughout the study, the research coordinator (CL) will be responsible for assigning enrolled patients to one of the two groups randomly and preparing the study medications—both the test drug and the placebo—20 min before the end of surgery. Both the test drug and the placebo are stored in 5 mL syringes from the same brand and batch. The syringes are labelled only with random identifiers (eg, ‘STUDY-001’) and expiration dates, without any indication of group assignment (eg, ‘Group A’ or ‘Group B’). The test drug and placebo are both colourless and odourless transparent liquids. The PCIA pump, a drug infusion device, consists of a pump unit (including a display screen, driving device, processing module, battery), a liquid storage bag, a pipeline and a self-control drug administration button. The PCIA pump is driven by a micro-motor to ensure accurate drug delivery, with the infusion rate adjustable within a specified range. Apart from variations in patient demographic information, the appearance of the prepared PCIA pumps is consistent. A strict blinding protocol is implemented, in which the preoperative visit is conducted by the anaesthesiologist (YL), while anaesthesia is administered by another two anaesthesiologists (XW and JL) according to the established protocol. The postoperative assessment is performed by the anaesthetic nurse (XJ), who is blinded to the research protocol and is solely responsible for postoperative follow-up. All the participating patients, healthcare team members and research personnel, except the research coordinator (CL), are unaware of the trial group assignments. Group assignments will be announced by the coordinator (CL) at the end of the trial. In the event of an emergency, such as the occurrence of a serious adverse event or when guidance is needed for subsequent treatment steps based on the blinding results, researchers have the option to reveal the group allocation or to halt medication administration. Any such instances will be meticulously documented and securely preserved.

Study intervention

Based on the results of the randomised group, interventions will be carried out according to the research plan. Patients in the experimental group receive a 2 mL test solution, consisting of esketamine 25 mg, dexmedetomidine 100 µg and sufentanil 1.5 µg/kg, which is diluted to a final volume of 100 mL with normal saline. Patients in the control group are given a placebo of 2 mL normal saline along with sufentanil at the same dosage, which is also diluted to 100 mL with normal saline. The treatment regimen involves an initial loading dose of 2 mL, followed by a continuous infusion at a rate of 1 mL per hour. A single top-up dose of 2 mL is permitted, with a lockout interval set at 15 min. The primary objective of this protocol is to maintain the Numeric Rating Score (NRS) for pain within or below the threshold of 3 points, ensuring effective postoperative analgesia.

Rescued analgesia protocol

In the recovery room, patients with an NRS pain score of 4 or higher are enabled to press the PCIA pump. If additional analgesia is required and the PCIA is insufficient, a single dose of sufentanil will be administered to alleviate symptoms until the NRS score is reduced to 3 or below. It is important to note that the use of esketamine and dexmedetomidine is prohibited in this context.

Perioperative management

Standard preoperative fasting and water deprivation are conducted for all patients the evening prior to surgery. On their arrival in the operating room, peripheral venous access is established immediately. Comprehensive monitoring, including ECG, non-invasive blood pressure, pulse oxygen saturation (SpO₂), bispectral index and end-tidal carbon dioxide partial pressure (PetCO₂), is initiated. Invasive arterial blood pressure monitoring is selectively deployed according to patient-specific needs. Before induction, patients will undergo routine ultrasound-guided adductor canal block with 30 mL of 0.33% ropivacaine to optimise comfort and analgesia.

The anaesthetic regimen employed is total intravenous anaesthesia, in accordance with the standard practice of our unit. For the induction of anaesthesia, patients receive an intravenous bolus injection of sufentanil at a dose of 0.3 µg/kg, propofol at 2 mg/kg or etomidate at 0.3 mg/kg, rocuronium bromide at 0.6 mg/kg and dexamethasone at 5 mg. Throughout the surgery, anaesthesia is sustained via a continuous intravenous infusion of propofol and remifentanil. The objectives of our anaesthesia management include maintaining BIS within a range of 40–60, keeping heart rate and blood pressure fluctuations within 20% of the patient’s baseline values, and regulating the core body temperature between 36°C and 37°C. PCIA pump is connected 20 min before the end of surgery. With the absence of any contraindications, ondansetron at a dose of 8 mg and flurbiprofen axetil at 50 mg are administered intravenously. Ondansetron will be administered as a prophylactic measure against postoperative nausea and vomiting, and flurbiprofen axetil will be used for patient analgesia. Lornoxicam (8 mg) is administered routinely by orthopaedic surgeons postoperatively for 2 days, two times per day. No other pharmacological or non-pharmacological treatment measures are used for sleep disorders that occurred during the postoperative period in the ward.

Additionally, vasoactive medications, including ephedrine, atropine, norepinephrine, nitroglycerin and esmolol, are made available during the surgery as necessary based on the patient’s condition.

Vital signs, including blood pressure, heart rate and pulse oximetry, are closely monitored in the anaesthesia recovery room for a minimum of 30 min prior to the patients being transferred back to the ward. On the patient’s return to the ward, continuous cardiac monitoring is maintained, and blood pressure assessments are conducted every 30 min to an hour, based on the individual’s condition, until 08:00 the following day after surgery or until the patient’s condition is deemed stable by the clinical team. At that point, a long-term medical prescription is issued by the clinician and executed by the nurse on the ward.

Data collection and outcome assessment

Baseline variables, including demographic data, comorbidities and major laboratory tests, will be collected. Based on exclusion criteria, these tests include assessment of liver and kidney function, cardiac assessments such as ECG and echocardiography, and other routine laboratory tests such as complete blood count, electrolyte levels, blood glucose and coagulation. Sleep quality within the recent 1 month is evaluated using the Pittsburgh Sleep Quality Index (PSQI), which ranges from 0 to 21, with higher scores indicating poorer sleep quality.34 Pain intensity, both at rest and during movement, is assessed using NRS (an 11-point scale, where 0 indicates no pain and 10 represents the most severe pain). Depressive symptoms are assessed with the Geriatric Depression Scale (GDS-15), which ranges from 0 to 15 (a score of 1–4 indicating no depression, 5–9 suggesting possible depression and 10 or above indicating depression). Anxiety levels are evaluated using the Self-Rating Anxiety Scale (SAS), which ranges from 20 to 80, with higher scores indicating greater anxiety.35 Cognitive function is assessed using the Montreal Cognitive Assessment (MoCA). Intraoperative data such as the amount of anaesthesia administered, duration of anaesthesia, length of the surgical procedure, use of tourniquets, fluid infusion rates, estimated blood loss and transfusion requirements, and the use of any vasoactive drugs during surgery will also be collected.

Primary outcome

The primary outcome measure is the subjective sleep quality on the night of surgery, which is assessed by the Chinese version of the Richards–Campbell Sleep Questionnaire (RCSQ) between 08:00 and 10:00 on the first postoperative morning.36 37 Five domains, including sleep depth, sleep latency, number of awakenings, ability to return to sleep and overall sleep quality, are encompassed in RCSQ. The overall RCSQ sleep score, calculated as the mean of these five domains, is categorised into four groups: very poor sleep (scores of 1–25), poor sleep (scores of 26–50), good sleep (scores of 51–75) and very good sleep (scores of 76–100). The Chinese version of the RCSQ has demonstrated good reliability and validity in evaluating sleep quality for patients in the ICU and elderly inpatients.38 In addition, to ensure the consistency of sleep assessments, the investigators for the RCSQ evaluation will receive training twice prior to and during the study.

Secondary outcomes

The RCSQ on the first and second postoperative nights will be evaluated according to the primary outcome measures. Pain intensity assessed using the NRS and the consumption of analgesic medications, along with the Richmond Agitation-Sedation Scale (RASS), are monitored during 0.5 hours after surgery and the first two postoperative days at specific times (08:00–10:00 and 16:00–17:00). The NRS includes both movement pain and resting pain. The total 48-hour analgesic consumption via the PCIA pump is also recorded. The incidence of nausea and vomiting within the first 24 hours and between 24 and 48 hours postoperatively is also recorded. Additionally, the MoCA is assessed on the day of discharge to evaluate cognitive function, and the PSQI will be used to measure sleep quality at 30 days following surgery.

Other outcomes

Self-administered attempts and total self-administered dose of PCIA in postoperative 48 hours. The duration of postoperative in-hospital stay is defined as the period spanning from the day of surgery to the day of hospital discharge. Major complications necessitating medical interventions, such as those classified as Clavien–Dindo grade 2and above,7 within the 30-day postoperative period will also be collected.

Safety outcomes

We conducted surveillance for hypotension defined as a systolic blood pressure<90 mm Hg or a decrease exceeding 30% from the baseline, hypertension indicated by a systolic blood pressure>180 mm Hg or an increase of more than 30% from the baseline, bradycardia characterised by a heart rate falling below 45 beats per minute or a reduction surpassing 30% from the baseline, tachycardia with a heart rate>100 beats per minute or an increase exceeding 30% from the baseline and pulse oxygen saturation<90%. These vital signs are monitored from the onset of anaesthesia until 08:00 the following day after surgery or until the patient reaches a stable condition.

In addition, psychiatric symptoms, including nightmares, hallucinations, restlessness, confusion, diplopia and blurred vision, will be systematically monitored. Hallucinations and nightmares are documented through standardised self-report questionnaires incorporating structured prompts such as “Did you see or hear something that didn’t exist?” and “Did you have frightening dreams?” Restlessness is evaluated using RASS. Confusion assessment is examined using the delirium assessment scale. Specifically, confusion is assessed during the Post-Aanesthesia Care Unit (PACU) using the Confusion Assessment Method ICU, while assessments on the first and second postoperative days used the 3D Confusion Assessment Method for the Chinese population. For the evaluation of diplopia and blurred vision, a simplified bedside version of the Snellen visual acuity chart is employed for quantitative assessment of visual acuity, and the red glass test is used to determine the type of diplopia. All the aforementioned mental symptoms of the patients are evaluated at 30 min after admission to the PACU, as well as the first two postoperative days at specific times (08:00–10:00 and 16:00–17:00).

Sample size calculation

In a prior observational study of 20 elderly patients, the average RCSQ score on the night of surgery was 50, with an SD of 15.39 We hypothesise that the application of esketamine–dexmedetomidine in the PCIA will elevate the RCSQ score to 60 on the night of surgery. To achieve statistical significance at the 0.05 level and a power of 90%, the required sample size to detect this difference is calculated to be 49 patients per group using PASS 15.0 software. Accounting for a dropout rate of 10%, we aim to enrol 55 patients in each group.

Statistical analysis

Statistical analysis will follow the intention-to-treat (ITT) principle, where all randomised participants will be analysed according to their original group assignment, regardless of any protocol deviations or withdrawals after randomisation. For confirmation purposes, a per-protocol analysis will also be performed, but ITT analysis will serve as the primary analytical framework.

Data distribution normality will be evaluated through the Kolmogorov-Smirnov (α=0.05) and Q–Q plot visual inspection. Normally distributed data will be expressed as mean (SD) and compared between groups using an independent-samples t-test. Non-normally distributed data will be expressed as the median (IQR) and analysed using the Mann-Whitney U test. Categorical data will be shown as frequencies (percentages) and compared using the χ² test or Fisher’s exact test.

The primary outcome measure is the total sleep score derived from the RCSQ on the day of surgery. This score will be evaluated by using either the inter-group t-test or the Mann-Whitney U test. Additional analyses, including the RCSQ score, RCSQ subitem scores, total sleep duration, sleep efficiency, sleep latency, wake time after sleep onset, pain scores at various postoperative time points, self-administered attempts and total self-administered dose of PCIA, postoperative hospital stay and inter-group comparisons, will also be analysed using either t-tests or the Mann-Whitney U test. Missing data will be processed by multiple imputation and maximum likelihood method. According to possible confounding factors (such as preoperative anxiety, depression, painful reaction), sensitivity analysis of the research results is carried out to ensure the stability of the research results. Repeated measurement data will be assessed using a mixed linear model. For ordinal data, the Mann-Whitney U test will be employed for inter-group comparisons, while mixed linear models will be used to analyse repeated measurements. Time-to-event data will be analysed using Kaplan-Meier survival analysis to compare postoperative hospital stays between groups. The log-rank test will be applied to determine the differences between groups. These differences will be quantified as hazard HRs and 95% CIs.