Trial design
We conducted a single-center, double-blind, randomized, placebo-controlled study at Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. Participants were recruited between August 2018 and December 2018. The study was approved by the certified Medical Ethics Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University. Generic quetiapine was supplied by the pharmaceutical department of Ramathibodi Hospital.
The hospital pharmacy prepackaged the study drug and placebo into identical capsules, carried out the randomization, and blinded the investigators and subjects. An independent data and safety monitoring committee evaluated all potentially serious adverse events. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice guidelines. The trial was retrospectively registered with the Thai clinical trials registry (TCTR) at clinicaltrials.in.th (TCTR20190927001). The study adheres to CONSORT guidelines.
Participants
Eligible participants were patients aged ≥ 65 years and acutely hospitalized in a medical specialty. All patients were recruited from the emergency medical observation unit and general medical wards. Patients provided written informed consent before participation. Exclusion criteria included patient or family refusal or patients who were already diagnosed with delirium, dementia, or severe Parkinson’s disease, patients who were critically ill, unable to take medication, unable to communicate, expected to be discharged within 24 hours, needed emergency surgery, had terminal illness, were currently taking antipsychotics, or patients who had active cardiac conditions, history of epilepsy, substance dependence or abuse, a blood potassium level ≤ 3.0 mEq/L or a corrected QTc ≥ 500 msec from EKG.
Randomization and interventions
Eligible and consenting patients were randomly assigned to quetiapine 12.5 mg tablets or placebo once daily at 9 pm for a maximum of seven consecutive days. The dosage of quetiapine was based on the recommended initial dose for delirium treatment in older patients.(15) Quetiapine was given once daily at bedtime due to sedating effects to aid sleep and cover the night time, when delirium usually developed.(19-22) The duration of the intervention was 7 days, which is the time during which most patients develop new-onset delirium.(23) Patients were randomized into the intervention or placebo group using fixed randomization schemes per site with a block size of 4 (1:1) according to a computer-generated randomization list.
Placebo and quetiapine were identical in appearance and packaged in identical medical envelopes containing 7 tablets with sequentially numbered labels, each with a unique study identification number. The medication was given by the nurse. Emergency unblinding was possible via 24-hour contact with an on-call pharmacist, in case delirium developed or assigned intervention might affect patient care. Study staff, clinicians and participants were to remain blinded throughout the study.
Outcomes
The primary outcome was delirium incidence within seven consecutive days after the intervention was initiated. Secondary outcomes were delirium duration, hospital length of stay, ICU admission, rehospitalization, and 30- and 90-day mortalities.
Data collection
All patients were interviewed and assessed by trained clinicians and investigators. Baseline demographic and health-related characteristics were recorded. Delirium was assessed at baseline by experienced clinicians or investigators. Patients who were diagnosed with delirium at baseline or had medical records suspecting delirium during admission were excluded from the study. During the study intervention period, all patients were assessed daily by clinicians or investigators. Patients were also observed by nursing and medical staff. A session of delirium assessment training based on the confusion assessment method (CAM)(24) was provided for the nurses in every participating ward. A document consisting of a questionnaire based on CAM was attached to the patient’s medical chart and filled in by the investigator or nurses to monitor and record the symptoms. Medical and nursing records were also reviewed for evidence of delirium symptoms. Delirium diagnosis was confirmed by a clinician or investigator according to the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-V) criteria for delirium.(25)
When delirium was diagnosed, the intervention was unblinded. The patient was further investigated for all potential causes and received standard delirium treatment. The patient remained under follow-up until discharge. Medical records from every patient were reviewed after discharge and at 3 months to collect data on primary diagnosis, comorbidities, complications, ICU admission, readmission and mortalities. Safety was monitored throughout the study by daily observation, examination by clinicians and investigators and by patient report. The study would be immediately terminated and unblinded in the case of suspected adverse events.
Sample size
Sample size was calculated based on the assumptions that the incidence of delirium in the placebo group would be 30% and that quetiapine prophylaxis would result in an absolute risk reduction of 20%. The estimate of the incidence of delirium in the placebo group was based on a previous study in a geriatric ward, which reported incidence rates ranging from 20 to 29%.(3) To detect a significant difference between groups, we sought to randomize 118 patients into 2 groups of 59 patients per treatment arm to give 80% statistical power at a two-sided 5% significance level (alpha).
Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics version 23. Descriptive statistics [mean ± standard deviation (SD)], frequency and percentage, or median and interquartile range (IQR) were used to describe baseline patient characteristics. Intention-to-treat (ITT) analysis was performed. Patients who did not develop delirium during the 7-day intervention period or left the study in case of early termination or hospital discharge were regarded as negative on the primary outcome.
The incidence of delirium was based on the number of participants who developed at least one delirium episode within the first 7 days after study initiation. The incidence of delirium was compared between groups using the Chi-squared test. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported as effect sizes using placebo as a reference group. Secondary outcomes were compared by using the chi-square test or Fisher’s exact test for dichotomous and nominal outcomes, the independent-samples t-test for normally distributed continuous outcomes, the Mann-Whitney U-test for ordinal outcomes and continuous outcomes that were not normally distributed and the Hodges-Lehmann estimator for confidence intervals for the difference between 2 medians. P-values <0.05 were considered statistically significant. Survival analyses presented by Kaplan-Meier curves were used for graphical demonstration. Cox proportional hazard regression analyses were performed to estimate the hazard ratios (HRs) for 30- and 90-day survival for the quetiapine and placebo groups. Sensitivity analyses were performed with age and length of hospital stay groups. Age groups were stratified by those aged younger than 75 years and 75 years and older according to the mean age (75.3 ± 7.1 years) from our study baseline characteristics. Length of hospital stay were stratified by hospital stay less than or equal to 5 days and over 5 days groups according to the median length of hospital stay (5 days) from our study outcomes. Interaction analysis was tested by subgroup.