Pilot Clinical Trial of an Integrative Intervention to Improve Critically Ill Patients' Delirium and Related Outcomes

Overview

Methodology: Pilot randomized controlled trial with 1 intervention and 1 parallel standard care group to assess the feasibility and treatment effect of a multimodal integrative intervention for delirium prevention and associated detrimental effects. Duration: 2 years, two months. Study Center(s): Multi-center (2 sites) Aim: To assess feasibility and measures of effect of a multimodal intervention consisting of relaxation with guided imagery (RGI), and moderate pressure massage on physiological and psychological outcomes of critically ill patients. Objectives: a) assess clinical trial feasibility with pre-defined goals (enrollment, randomization, adherence, timing of intervention, workload), b) calculate estimates and variance of treatment effect across outcome measures, c) calculate confidence intervals (CI) of incidence proportions, means and standard deviation (SD) of outcome measures in study groups., d) explore underlying physiological mechanisms of efficacy Number of Subjects: 60 participants per arm. Total: 120 participants Diagnosis and Main Inclusion Criteria: Critical illness Inclusion criteria: a) Age over 18 years, b) ICDSC:0-3. Exclusion Criteria: Patients: a) with expected ICU LOS\< 72 hours, b) with acute neurological illness/ trauma, persistent sedation or coma, c) with current history of severe mental health problems and dementia, as per history and psychiatrist assessment, c) with hearing impairment or conditions not permitting use of headphones, e) on neuro-muscular blockers, f) with substance/ alcohol withdrawal, g) enrolled in trials of sedatives, antipsychotics. Intervention: REPOSE intervention (a multimodal relaxation intervention) It includes: a) relaxation and guided imagery (RGI) (40 min, headphones), b) a brief moderate pressure massage session (massage: 15 min). RGI involves: a) guided relaxation, b) a structured guided imagery script and c) music for 15 min (Mozart piano sonata KV283, G major (2 3) II Andante). Duration of administration: Up to 5 days while participants still in the ICU. Reference therapy: Standard care plus placebo to maintain blinding. Statistical Methodology: Outcomes will be analyzed longitudinally over 5 days by logistic regression model based on generalized estimating equations (GEE) with AR(1) correlation structure. Confidence intervals will be presented with estimated effects. Primary analysis will be based on all available data utilizing data from all assessments.

1. Background 1.1 Risk/Benefits Risks: Anticipated risks from participation in this study are minimal. This and similar relaxation interventions have been tested before and no inconveniences or side-effects have been reported. Adverse events, irrespective of causal relationship, will be collected for all participants Benefits: The direct benefits to participants include the opportunity to receive a relaxation intervention that has been shown helpful in terms of decreasing pain ratings and anxiety and improving quality of sleep in similar settings before. 1.2 Treatment/ Dose Rationale The evidence-based multimodal intervention (duration: 55min) is based on a literature review and the recommendations of the American Holistic Nurses Association (31) and a successful pilot (10). Multimodal integrative interventions demonstrate superior efficacy compared to uni-dimensional ones (32). It includes: a) relaxation and guided imagery (RGI) (40 min, headphones), b) a brief moderate pressure massage session (massage: 15 min). RGI involves: a) guided relaxation, b) a structured guided imagery script and c) music for 15 min (Mozart piano sonata KV283, G major (2 3) II Andante). Moderate pressure, low velocity (4 N, 1-5cm/s) massage consists of squeezing movements with wide area of contact will be applied in the following order and then reverse order, for 2-3 min at each site: over trapezius muscles, lateral arms, forearms, feet and then over the temple and forehead area. Moderate pressure massage is used as it elicits parasympathetic activation, as opposed to light pressure (33). The intervention will be administered once daily (10-11am) for up to 5 consecutive days by trained intervention nurses not involved in patient care. The intervention will start as soon as possible after ICU admission, and will be terminated upon a patient's transfer or discharge from the ICU. Selection of timing of the interventions is based on the daily schedule of the unit, to minimize interference with routine care. A 5-day intervention was selected as most patients develop delirium within the first 5 days of ICU admission (28). 1.3 Trial Conduct This study will be conducted in compliance with the protocol approved by the University of Alberta Health Research Ethics Board (HREB), and according to Good Clinical Practice standards. No deviation from the protocol will be implemented without the prior review and approval of the HREB. Participants: Adult critically ill patients meeting the eligibility criteria. Recruitment: Consecutive patients admitted into 2 academic teaching ICUs with Intensive Care Delirium Screening Checklist (ICDSC): 0-3 will be screened by research assistants for admission to the protocol and will be recruited by a research nurse. 2. Trial Objectives Goal: to assess clinical trial feasibility and treatment effect estimates of a multimodal intervention incorporating RGI and moderate pressure massage for prevention of ICU delirium and for improvement of physiological and psychological outcomes in critical illness. 3. Trial Design 3.1 Primary Study Endpoints/Secondary Endpoints Please see below 3.2 Study Design/Type Design: Pilot feasibility, randomized, controlled, double-blinded (clinicians, outcome assessors) trial with 2 parallel groups (intervention and standard care group). Accounting for major risk factors of delirium (28), stratified randomization according to age (\<65, ≥65) and surgical or injury trauma (Trauma/ Surgery, non-Trauma/ surgery) with blocking and 1:1 allocation to assure balance in numbers per group will be employed. 3.3 Randomization Allocation-sequence generation and concealment: Participants will be randomly assigned to either control or intervention (1:1 allocation) as per a computer generated randomization schedule, generated by the Epidemiology Coordinating Research Centre (EPICORE), University of Alberta (UoA), stratified by site, age (\<65, ≥65), and presence of surgical or injury trauma using permuted blocks of random sizes. Blinding: Clinicians, outcome assessors, research assistants, nurses and laboratory technicians will be blinded to group allocation. Participants cannot be blinded. A sham intervention with silent headphones to mask the audio component, and then presence of an intervention nurse at the bedside with drawn curtains will be used. Blood samples will be in code-identified vials to maintain blinding. Assessments regarding outcomes will be conducted by assessors blinded to treatment allocation. 3.4 Duration Duration of participation: The total duration of participation is approximately 14 weeks. Duration of participating will vary depending on the length of a patients' stay in the ICU and hospital. Time points include: baseline measurements, follow-up while in the ICU for up to 5 days, follow up 48-96 h after ICU discharge, follow-up 3 months after hospital discharge. Participant timeline: Based on patient population characteristics in the participating ICUs, eligibility rate is expected to be 40% and the recruitment rate 30% of eligible patients, yielding an estimated recruitment period 6-8 months. 3.5 Discontinuation Intervention discontinuation: For a given participant the intervention may be discontinued (or a session may be cancelled) if a participant or legal surrogate so wishes. If a participant skips one or more sessions but remains in the protocol, the remaining sessions will be administered as per study protocol. Reasons for missed sessions will be reported and analyzed. Sessions may be cancelled or the intervention discontinued in case of reported harms, such as abrupt worsening of delirium and/ or patient anxiety. Adverse events, irrespective of possibility of causal relationship, will be collected for all participants after recruitment and for as long as a participant remains in the protocol, and will be reported. 4. Selection and Withdrawal of Subjects 4.1 Inclusion Criteria Please see below 4.2 Exclusion Criteria Please see below 4.3 Subject Withdrawal Criteria for subject withdrawal: 1\) Active withdrawal by the study participant. 2) Loss of contact. 3) Evidence of trial-related harms. 4) Assessment that premature termination of the trial related intervention is indicated, e.g. because of a belatedly identified violation of the criteria for inclusion and/or exclusion that imply the premature termination of the therapy 5) Non-compliance of the subject which indicates the premature termination of the trial related intervention. 1. Documentation of the clinical trial will continue even if the trial related therapy is terminated prematurely, expect if the participants specifies that documentation should be terminated. 2. Subjects will be replaced if the duration of participation was less than 24 hrs. 3. Adverse events will be observed, irrespective of a withdrawal of a trial. 4.4 Medication No restrictions with regard to concomitant medication. Standard care will be continued concomitantly for all participants as per unit protocols and clinicians' decisions. Pain will be treated when present and sedatives will be titrated to maintain a lightly sedated state (RASS= -1- -3). 4.5 Monitoring for subject compliance To assure adherence, interventions will be administered by trained research nurses, not involved in participants' care, who will be randomly audited by the TSC to ensure protocol compliance, especially with respect to touch massage procedure. Deviation from the intervention will be recorded in detail by intervention nurses immediately after each intervention session. 5 Assessment of Safety 5.1 Safety Parameters Any physiological/ behavioral alteration during interventions will be recorded and analysed. Adverse events, irrespective of causal relationship, will be collected for all participants. 5.2 Adverse Event Reporting \& Follow-up The time period for adverse event (AE) and serious AE (SAE) reporting will begin with each beginning of intervention, continue throughout the intervention, and end 30 min after the end of each intervention session. All adverse events occurring during the reporting period will be recorded on an AE/SAE Report Form, and categorized as to the degree of expectedness, relatedness and severity.. The clinical course of each adverse event will be followed until resolution, stabilization, or until it has been determined that the study treatment or participation is not the cause. 6 Statistical Plan 6.1 Statistical Methods Statistical methods: Demographic/ clinical characteristics of patients and all outcomes will be presented by treatment group using descriptive statistics- mean (SD), median (IQR) or proportion. Outcomes will be analyzed longitudinally over 5 days by logistic regression model based on generalized estimating equations (GEE) with AR(1) correlation structure. ANCOVA, t-test or Mann-Whitney test, as appropriate, will be conducted for the continuous outcomes that are not longitudinal. Chi-square or exact test, as appropriate, will be used for categorical outcomes. Confidence intervals will be presented with estimated effects. \\A significance level alpha=0.05 will be employed. 6.2 Subject Population(s) for Analysis 60 subjects per arm/ 60 subjects per site. Total 120 subjects. Sampling of consecutive ICU patients according to inclusion/ exclusion criteria. This pilot is not powered to determine a difference in a primary outcome, since the aim is to assess estimates of effect. Participants will include 120 patients (60 per site) to attain high probability of equivalence at baseline, precision of estimates, to be able to employ appropriate repeated measurement procedures for estimation of effects and to accrue experience for the larger clinical trial (34). 6.3 Termination Criteria 1. Alterations in current clinical practice that make trial procedures no longer feasible. 2. Reaching a positive or negative statistical end point earlier than anticipated. 3. Clear statistical evidence of superiority of treatment that raises ethical concerns with regard to randomization to a control group. 4. Evidence that the risks outweigh the potential benefits of the trial.+ 6.4 Accountability Procedure Reasons for missing data will be analyzed. Primary analysis will be based on all available data. A sensitivity analysis will be conducted based on inverse probability weighted GEE (IPWGEE) (36). 6.5 Deviation Reporting To account for noncompliance, protocol deviations and missing outcomes intention-to-treat (ITT) analysis will be employed. Additionally, per protocol analysis will also be employed. 7 Direct Access to Source Data/Documentation The investigators and participating institutions will permit trial-related monitoring, audits, Institutional review (IHREB) and regulatory inspections by providing direct access to source data/documentation. 8 Quality Control and Quality Assurance This pilot will be supervised by an independent trial steering committee (TSC). Study personnel will be trained for standardized processes. Clinical data will be retrieved from the quality-controlled clinical information system of the units. Detailed electronic data collection forms with embedded quality controls will be used. Day-to-day running of the trial will be overseen by a trial monitoring group (TMG). 9 Ethical Considerations This study will be conducted according to Canadian and international standards of Good Clinical Practice for all studies. Applicable government regulations and University of Alberta research policies and procedures will also be followed. This protocol and any amendments will be submitted to the University of Alberta HREB for formal approval to conduct the study. The decision of the HREB concerning the conduct of the study will be made in writing to the investigator.