Background: Delirium during intensive care unit (ICU) stay is frequent and associated with significant morbidity, mortality and healthcare-related costs. International guidelines suggest its prevention. However, curative treatment remains unclearly established. Despite contradictory and ambiguous academic literature, international guidelines suggest the use of second-generation (atypical) antipsychotics over haloperidol. However, haloperidol remains the most widely used neuroleptic worldwide as a first-line treatment of agitation and/or delirium. Dexmedetomidine, an alpha2- adrenergic receptors agonist, has shown its efficiency in the treatment of delirium in intubated patients but also in its prevention. Dexmedetomidine represents a widely used alternative to haloperidol. Only few studies have compared the efficacy of dexmedetomidine in non-intubated ICU patients as a first-line curative treatment of delirium. The main objective of the 4D trial is to demonstrate that dexmedetomidine decreases delirium duration compared to placebo. Methods/design: The 4D trial is an investigator-initiated, prospective, multicenter, randomized, double-blinded, two-arm trial, randomizing 300 non-intubated ICU patients with a diagnosis of agitated delirium to receive dexmedetomidine or placebo as a cure. In case of agitation (RASS≥ + 2), immediate haloperidol administration will be allowed, to protect patient and staff in charge, while waiting for study treatment action. The primary outcome measure is a composite of duration of agitation or delirium or the use of intubation with deep sedation and mechanical ventilation. Secondary outcomes include mortalities at 7 and 30 days, ICU length of stay and occurrence of adverse effects related to dexmedetomidine use (bradycardia or hypotension requesting any treatment; or haloperidol use (neuroleptic malignant syndrome, extrapyramidal syndrome, prolonged QTc). The sample size will allow the detection of a 50% decrease of agitation duration (120 min), of an absolute reduction of delirium duration (1 day) and of a 50% relative decrease of intubation and mechanical ventilation, with a type 1 error rate of 1.8% (error risk inflation due to components of composite) and power of 90%, assuming a 15% incidence of intubation and mechanical ventilation requirements, an agitation duration of 240 min and a delirium duration of 3 days. One hundred and ten patients by group will be needed. An intermediate analysis is scheduled and requires the inclusion of 150 patients. Discussion: The 4D trial may provide important data on the safety of commonly used sedative dexmedetomidine and could have a significant impact on future treatment of non-intubated ICU patients presenting with agitated delirium. Trial registration: ClinicalTrials.gov,ID: NCT 03317067. Registered on 23 October 2017. Keywords: Dexmedetomidine, Haloperidol, Delirium, ICU patients, Sedation, Mechanical ventilation, Intubation * Correspondence: firstname.lastname@example.org Département de Médecine Périopératoire (MPO), Centre Hospitalier Universitaire (CHU) Clermont-Ferrand, 63003 Clermont-Ferrand, France GReD; UMR/CNRS6293; Université Clermont-Auvergne; INSERM U1103, 63003 Clermont-Ferrand, France Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Louis et al. Trials (2018) 19:307 Page 2 of 11 Background Haloperidol, a centrally acting dopamine antagonist, Delirium, whatever its presentation (agitated or not), is a is the most frequently used drug for delirium treat- frequently observed condition in intensive care unit ment because of its antipsychotic and sedative proper- (ICU) patients. According to recent studies, incidence ties without anticholinergic effects . Despite ease ranges between 16 and 89% [1, 2]. Diagnosis requires a of use based on titration opportunity, a short delay of clinical examination, which includes changes in con- action, and intravenous administration, haloperidol sciousness (altered concentration, decreased attention), carries potential adverse effects such as neuroleptic in cognitive function (disorganization of thought, tem- malignant syndrome, dyskinesia, extrapyramidal syn- poral disorders, memory problems) or in perception drome, excessive somnolence and deep sedation with (hallucinations). Delirium appearance is brutal and its potential apnea. However, the efficacy of this drug natural evolution fluctuates over time. Different scores has not yet been demonstrated in the treatment  can help physicians to diagnose delirium, the most com- or the prevention  of delirium in ICU compared mon being the Confusion Assessment Method for the to placebo. In addition, the rate of non-responders to ICU (CAM-ICU) [3–5] and the Richmond Agitation and haloperidol may sometimes be high and requires the Sedation Scale (RASS) [6, 7]. The CAM-ICU scale is a use of other drugs to jugulate agitation. In this latter readily available and reliable instrument to monitor case, other sedative neuroleptics or benzodiazepines delirium in ICU patients. Delirium is defined in terms of can be used, even if these latter drugs might be four diagnostic features, and is deemed positive when directly responsible for maintaining delirium. In some Feature 1 (acute change or fluctuating course of mental cases, deep sedation and mechanical ventilation may status) and Feature 2 (inattention) and either Feature 3 be required to cure agitation and delirium. (altered level of consciousness) or 4 (disorganized think- Dexmedetomidine, an alpha2-adrenergic receptors ing) are present . RASS allows the evaluation of agita- agonist, could be an alternative to haloperidol [16, 17]. tion and sedation of ICU patients and is included in This drug has proven efficacy in the treatment of Feature 3 of the CAM-ICU. This validated 10-point Vis- delirium in intubated ICU patients as well as delirium ual Numeric Scale Numeric Rating Scale ranges from − prevention [18, 19]. Dexmedetomidine combines several 5 to 4. The score definitions are as follows: − 5, unarou- advantages: a moderate sedative action, a wide thera- sable; − 4, deep sedation; − 3, moderate sedation; − 2, peutic index, the absence of infusion rates adaptation light sedation; − 1, drowsy; 0, alert and calm; 1, restless; outside hepatocellular insufficiency and a substantial 2, agitated; 3, very agitated; 4, combative. analgesic effect. In 2009, Reade et al. showed that using Although incidence overestimation is likely, delirium dexmedetomidine to treat delirium in ICU reduced de- is difficult to diagnose, especially in its hypoactive (calm) lays to extubation and decreased the use of additional presentation. Delirium remains a public health issue sedatives as well as ICU length of stay compared to because its presence is associated with long-term adverse haloperidol . In the DahLIA study, the same group outcomes: decreased survival , increased ICU and demonstrated that the addition of dexmedetomidine to hospital lengths of stays which are associated with standard care in ventilated ICU patients increased increased healthcare-associated costs . Adverse out- ventilator-free hours at day 7 in patients with agitated comes are even more important when delirium is severe delirium compared to placebo . and prolonged. However, few studies have compared the efficacy of Several risk factors have been identified to date . dexmedetomidine to treat delirium in non-intubated Patients’ characteristics (age), habits (current smoking patients. To our knowledge, only one non-randomized status, drug addiction or alcohol abuse ), postop- prospective controlled trial  has compared dexmede- erative periods, prolonged hospitalization, treatment tomidine to haloperidol in non-intubated patients. In including benzodiazepines, patient isolation and ICU this Spanish study, patients were initially treated with hospitalization have been associated with delirium oc- haloperidol and secondarily with dexmedetomidine in currence. The North American guidelines proposed case of haloperidol-resistant delirium. Patients who strategies to prevent delirium . Once detected, the received both treatments had a shorter delirium time curative treatment of delirium remains uncertain to date. associated with fewer disorders of consciousness. A non-pharmacological approach seems mandatory, in- Benefits related to the use of dexmedetomidine to cluding early rehabilitation, limited number of caregivers treat delirium in non-intubated ICU patients are in charge of the patient or sleep-disorder prevention potentially important. Therefore, the definitive aim of . These non-pharmacological approaches are usu- the 4D trial is to investigate reduction of delirium ally insufficient and a pharmacological approach is duration related to dexmedetomidine use in non-intu- often necessary, especially for agitated (hyperactive) bated ICU patients, compared to placebo, in a multicenter, delirium. randomized, controlled, double-blind study. Louis et al. Trials (2018) 19:307 Page 3 of 11 Methods/design Interventional Trials (SPIRIT) Figure and Checklist are Ethics available as Fig. 2 and Additional file 1, respectively. Due to the specific medical condition of recruited pa- tients (agitated delirium), an emergency inclusion pro- Selection of patients cedure will be possible. Written pursuit consent will be Patients will be included in the 4D trial if they comply obtained from all participants (once medical condition with the indicated inclusion and exclusion criteria as resolved) or their next-of-kin. The Institutional Review below. Board of the University Hospital of Clermont-Ferrand (France) approved the trial. By 7 June 2017, the study Inclusion criteria had been approved for all centers by a central ethics com- For inclusion, adult patients must meet all the following mittee (Comité de Protection des Personnes Sud-Est V, criteria: Grenoble, France, 17-CHCF-02) with the registration number EudraCT 2017–000731-14. Any further add- 1. Age > 18 years itional important protocol modification will require 2. Patient hospitalized in an ICU the approval of a central ethics committee (Comité de 3. Presenting a productive delirium according to the Protection des Personnes Sud-Est V, Grenoble, France, following criteria: 17-CHCF-02). The 4D trial was registered on 23 (a) Acute onset (< 2 h) and fluctuating course October 2017 at http://www.clinicaltrials.gov with trial during the same day identification number NCT 03317067. (b) Alteration of cognitive functions: disorganization of thought (delirium of persecution, inability to reason logically), abnormal perceptions Trial design (hallucinations), memory impairments, temporal The 4D trial is an investigator-initiated, national, multi- disorientation, non- or misrecognitions, center, double-blind, parallel, randomized, controlled, difficulties in naming objects or writing) two-armed trial with concealed allocation of non- (c) In whom a simple cropping and non-medicated intubated ICU patients with delirium, 1:1 to treatment therapeutics are not sufficient to allow symptom of delirium using dexmedetomidine (Orion Corporation, resolution for a few hours Espoo, Finland) or placebo (0.9% sodium chloride). (d) CAM-ICU (Annex 1, Additional file 1) positive and a RASS > + 1 (Annex 2, Additional file 1) CONSORT Diagram and SPIRIT Checklist 4. Non-intubated or extubated (> 24 h) The Consolidated Standards of Reporting Trials 5. Tracheotomized without pressure support (> 24 h) (CONSORT) Diagram of 4D is presented in Fig. 1. 6. No contraindication to dexmedetomidine or The Standard Protocol Items: Recommendations for haloperidol use Assessed for eligilility (n 300) Excluded (n=?) Not meeting inclusion criteria (n=?) Declined to participate (n=?) Other reasons (specify) (n=?) Randomization (n = 300) Allocated to 0.9% saline Allocated to Allocation based placebo dexmedetomidine (n = 150) (n = 150) Lost to follow-up, in hospital, at Lost to follow-up, in hospital, at Day 30, at 1 year (give reasons) Follow-up Day 30, at 1 year (give reasons) Discontinued intervention (specify why) Discontinued intervention (specify why) Analysis Analyzed (n=?) Analyzed (n=?) Fig. 1 Consolidated Standards of Reporting Trials (CONSORT) flowchart illustrating the randomization and flow of patients in the study Louis et al. Trials (2018) 19:307 Page 4 of 11 Enrolment Post-allocation Close-out TIMEPOINT T0 T2 T3 T4 T5 etc. Tend ENROLMENT: Eligibility screen Informed consent X Medical history and physical examination Pregnancy test for women with childbearing potential Previous psychotropic medications Study medications contra-indications Allocation INTERVENTIONS: Randomization Dexmedetomidine Placebo ASSESSMENTS: Review of concurrent XX medication Agitation (RASS) XX X X X X X Delirium (CAM-ICU) XX X X X X X Physical restraints XX X X X X X Intubation and XX XX X X mechanical ventilation Occurrence of bradycardia or XX XX X X hypotension Occurrence of unplanned catheter, XX XX X X tube or drain removal UEs, SUEs, SUSARs XX XX X X Fig. 2 Patients’ schedule of activities according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) Figure. CAM-ICU Confusion Assessment Method-Intensive Care Unit, RASS Richmond Assessment Sedation Scale, SUE serious unexpected events, SUSAR severe unexpected serious adverse event, UE unexpected events Non-inclusion criteria oro-pharyngeal dysfunction, arterial hypotension or Patients will not be included for any of the following bradycardia, QTc interval prolongation, and hepatic reasons: or renal dysfunction), as mentioned in the Summaries of Product Characteristics 1. Administration of dexmedetomidine and/or clonidine 3. Neuropsychiatric pathology judged by the and/or haloperidol during the 72 h before inclusion investigator as a potential source of bias (in 2. Contraindication to the use of dexmedetomidine particular: active drug addiction, psychosis, etc.) and haloperidol (history of allergy, Parkinson’sdisease, 4. Parturient or breast-feeding woman Louis et al. Trials (2018) 19:307 Page 5 of 11 Fig. 3 Study protocol diagram. Trial medication administration according to patient’s RASS. ICU intensive care unit, RASS Richmond Agitation Sedation Scale 5. Protected major (guardianship) participating centers that are anticipated to be recruiting 6. Patient’s or relative’s refusal to participate will be coordinated by the pharmacy of the coordinating center. The receipt, storage and dispensing of the Randomization and blinding blinded trial bags will be conducted by the pharmacy Randomization will be conducted over a dedicated, department in each individual trial site. Each trial site password-protected, SSL-encrypted website (Clinsight will have a sufficient number of sets of trial drugs to be Software) to allow immediate and concealed allocation. allocated to the included patients. The initial and any Each patient will be given a unique patient number and subsequent allocation of trial drugs will be determined a randomization number. The allocation sequence will by the web-based randomization system at each site. be generated by permutation blocks of 2. The participant This will ensure that the patient only receives the trial allocation will be carried out by local investigators who drug that they were randomized to receive. The informa- will log into the randomization system using a personal tion regarding which codes correspond to what treat- ID code and will enter any relevant information (includ- ment will be maintained in a secure location at the ing weight to calculate infusion rates of study drugs). coordinating center. All staff, excluding the only person Trial drugs (dexmedetomidine and 0.9% sodium sa- who prepared drug administration, at the participating line) are not visually identical. However, to ensure the trial sites and the coordinating center will be blinded to blinding of study drug administration, opaque bags will the treatment allocation. By the end of study protocol, be available, each of those containing a medication vial. unused medication bags as well as used medications Trial bags will be blinded and identified only by a unique bags and vials will be returned to the pharmacy of co- number. The initial allocation of trial drugs will be ordinating center. determined by the web-based randomization system No data monitoring committee (DMC) is programmed. through the allocation of a single bag number. Prepar- In the present clinical study, patients are treated for ation of the study drug syringe will be conducted by a a relatively short time and the drugs under investiga- nurse and/or a physician independent of the study tion are well characterized and known for not harming protocol and not in charge of the included patient. patients . Further needs for study drugs (new syringe preparation) will be obtained from the web-based randomization Trial interventions system and the new allocation of a bag number. The After inclusion and randomization, patients presenting logistics of the trial bags’ distribution to each of the 13 with a RASS ≥ + 2, will be intravenously administered Louis et al. Trials (2018) 19:307 Page 6 of 11 boluses of haloperidol (2.5 mg), to promptly protect the (d).Duration of mechanical restraint prescribed and patient against self-inflicted physical damages. Indeed, carried out dexmedetomidine requires infusion of at least 1 h to (e). Occurrence of bradycardia or rhythm disorders or obtain complete efficacy, incompatible with possibly myocardial ischemia or tachycardia or prolonged aggressive and unsafe delirious patients. corrected QTc interval on the electrocardiogram, After this bolus administration, or if RASS ≤ + 1, all respiratory distress or apnoea included patients will be allocated to one of the following (f). Occurrence of hypotension requiring any two study groups (Fig. 3): vasopressor administration or hypertension (g). Occurrence of dyskinesia, extrapyramidal symptoms 1. Dexmedetomidine: patients in this group will or neuroleptic malignant syndrome receive dexmedetomidine infusion starting with a (h).All-cause mortality at day 7 and day 30 − 1 − 1 dose of 0.2 to 0.5 μg.kg .h (i). Self-evaluation of sleep quality on a digital scale 2. Placebo: patients randomized to placebo group will (0 to 10) be administered an identical infusion of 0.9% (j). SF36 score at 1 year sodium saline at an equivalent rate (k). Occurrence of delirium-related complications such as unplanned catheter, tube or drain removal Study medication will be adjusted by the bedside nurse − 1 − 1 or clinicians between 0.2 and 1.4 μg.kg .h depending Patient withdrawal on the control of agitation. Continuous infusion will be Studied drugs are to be used during ICU stay and for continued at least 36 h after delirium disappears, or 36 h after delirium control. Nevertheless, a participant until the patient leaves the ICU if necessary. or a patient’s relative who no longer agrees to partici- If study medications are not sufficient to control delir- pate in the clinical trial can withdraw the informed ium beyond the maximal dose, new boluses of haloperi- consentatany time withoutneed of furtherexplan- dol will be repeatable every 10 to 30 min up to the ation. Patients who are withdrawn from the protocol resolution of agitation (with a maximum dose of 30 mg). will be followed up and analysed as with the If maximal doses are reached, the use of benzodiaze- remaining patients. In order to conduct intention-to- pines will be left to the discretion of physicians in charge treat analyses with as little missing data as possible, it of the patient (promoting clorazepate). Otherwise, the is in the interest of the trial to collect as much data use of other psychotropic drugs will be documented. from each participant as possible. Therefore, the in- vestigator may ask the participant and/or relatives which aspects of the trial they wish to withdraw from Summary of outcomes (participation in the remaining follow-up assessments The primary outcome measure will be a composite of: or use of already collected data) and, whenever pos- sible, the participant will be asked for permission to 1. Duration of agitation (in hours), defined by a obtain data for the primary outcome measure. If this RASS ≥ +1 person declines, no more data will be collected, and 2. Duration of delirium (in days), defined by a positive new patients will be randomized to obtain the full CAM-ICU sample size. All randomized patients will be reported, 3. Requirement of intubation to control delirium with and all data available with consent will be used in the deep sedation and mechanical ventilation analyses. If appropriate, missing data will be handled in accordance with multiple imputation procedures if In addition, each component of the primary outcome missing data are greater than 5% [22–24]. measure will be analysed separately. RASS will be evaluated at H1, H2, H3, H4, H5, and H6, after randomization and then every 12 h after inclusion; Safety CAM-ICU will be evaluated at H0 and then every 12 h; All adverse events thought to be related to the trial If mechanical ventilation is necessary, the time be- medications will be reported to the trial coordinating tween inclusion and intubation will be noted. center. According to the French Public Health Code, all Secondary outcomes will be as follows: suspected unexpected serious adverse events will be reported to the ANSM. The investigator can use, after (a). Length of ICU stay (in days) consultation of the Steering Committee, at any moment (b).Number of ventilator-free days at day 30 and in any situation that seems necessary (occurrence of (c). Adverse effects such as the occurrence of pneumonia severe adverse events), the unblinding procedure by (following the ATS definitions) and/or septicemia consulting the web-based randomization table. Louis et al. Trials (2018) 19:307 Page 7 of 11 Statistics model to take into account: (1) fixed effects covariates de- Sample size estimation termined according to univariate results and to clinical The sample size estimation is based on work proposed relevance and (2) center as random effects (to measure by O’Brien concerning composite endpoints : between and within-center variability). Results will be weighted summation of single endpoints with standard expressed as regression coefficients and 95% confidence procedures leads to asymptotically normal statistics. The intervals. Other continuous parameters outcomes will be average z-score extends this approach to include con- analysed as described previously. tinuous, ordinal, dichotomous, and time-to-event end- Categorical parameters (as proportion of patients points. Specifically, continuous, ordinal, and with bradycardia and rhythm disorders, proportion of dichotomous variables are converted to z-scores by patients with hypotension requiring treatment with subtracting an individual’svalue from the overall vasopressors, mortality) will be analysed using chi or mean and dividing by the SD of the pooled group; Fisher’s exact tests in univariable analysis. Multivariable time-to-event variables are first transformed to log- analysis will be performed using generalized linear mixed- rank scores and then converted to z-scores by sub- model analysis (logistic for dichotomous dependent out- tracting the mean and dividing by the SD of the comes). The censored data (survival at day 30) will be esti- pooled data. The z-scores are then aligned to the same mated by the Kaplan-Meier method and compared by the direction so that worse outcomes have smaller scores. The log-rank test in the univariate situation and the marginal z-scores are then averaged across endpoints for each pa- Cox model in the multivariate situation. The covariates tient. Treatment groups will be compared with respect to defined above for the main analysis will be considered in a this average z-score, the primary efficacy outcome. similar way, in addition to the center effect as random According to the academic literature concerning non- effect. Longitudinal analyses concerning repeated mea- intubated patients in ICU for the duration of agitation sures will be studied using random-effect models (linear (240 min, relative expected difference of 50%), the dur- or generalized linear), to take into account patient as ation of delirium (3 days, absolute difference of 1 day) and random effect (slope and intercept), nestled in center proportion of patients requiring deep sedation with mech- random effect. According to clinical relevance and to anical ventilation (15%, relative expected difference of European Medicines (EMA) and Consolidated Stan- 50%), n = 110 patients per group are required for a type I dards of Reporting Trials (CONSORT) recommenda- error at 0.018 (correction due multiple components of tions, subgroup analyses will be proposed after the composite outcome) and a statistical power at 90%. study of randomization group interaction in regres- Finally, we have decided to include 150 patients per group, sion models (for repeated data or not). Secondarily, a i.e., 300 patients. per-protocol analysis will be considered. A post hoc As sample size estimation was based on a composite cri- subgroup analysis of elderly patients (above 65 years of teria for which each of the expected effect sizes associated age) will be conducted to investigate age influence. This to each component could be re-evaluated, and considering threshold of 65 years was based on a recently published that the correlation between components is not known, it review on delirium in older patients . In the event appears reasonable and relevant to consider an interim that investigators or reviewers introduce analyses in analysis after the inclusion of 150 patients (75 by group). addition to those described above, these will be clearly A difference between randomization groups will be delimitated as post hoc and will be considered hypothesis considered statistically significant for a type I error at 0.01 generating. (Kim-DeMets correction). Finally, a sensitivity analysis will be performed and the nature of missing data will be studied (missing at Statistical analysis random or not). According to this, the most appro- Statistical analyses will be conducted using Stata soft- priate approach to the imputation of missing data will ware (version 13, StataCorp, College Station, TX, US). A be proposed. two-sided p value of less than 0.05 will be considered to indicate statistical significance (except interim analysis). Data registration Concerning the primary outcome, the comparison Data will be entered into the web-based electronic between groups will be analysed using Student’s t case report form (eCRF) using Clinsight electronic test or the Mann-Whitney U test. Normality will be data capture tools hosted at Centre Hospitalier studied by the Shapiro-Wilk test and homoscedastic- Universitaire (CHU) Clermont-Ferrand by trial or ity using the Fisher-Snedecor test. Intention-to-treat clinical personnel under the supervision of the trial analysis will be considered for the primary outcome. site investigators at each participating center. Data Then, the analysis of the primary outcome will be collection will be monitored by trained research completed by multivariable analysis using a linear mixed coordinators. Louis et al. Trials (2018) 19:307 Page 8 of 11 The following data will be registered: clearance, plasma lactate, C-reactive protein (CRP), bilirubin, liver enzymes 1. Pre-randomization and baseline characteristics: 3. RASS score at H1, H2, H3, H4, H5 and H6 after demographic data (age, height, weight, gender, randomization and then RASS score every 12 h and Body Mass Index); co-morbidities after randomization (hypertension 4. CAM ICU score every 12 h after randomization (Y/N), renal dysfunction (Y/N), chronic heart 5. BPS NI score or VAS every 12 h failure (Y/N), heart-rhythm disorder (Y/N), 6. Cumulative dose of haloperidol at H1, H2, H3, H4, diabetes mellitus (Y/N), malnutrition (Y/N), H5, H6 and then every 12 h chronic alcoholism (Y/N) and active smoking 7. Minimum and maximum dose of dexmedetomidine − 1 − 1 (Y/N)); and routine biological data, including or placebo infusion in μg.kg .h every day − 1 baseline serum creatinine and liver enzymes and 8. Total volume of study drug administration (μg.kg ) metabolic acidosis. per day 2. Simplified Acute Physiology Score (SAPS II) score 9. Time to achieve RASS = 0, date, interval between  and Sequential Organ Failure Assessment randomization and achievement in hours (SOFA) score  10. Time to achieve a negative CAM ICU score, date, 3. Date of ICU admission (day, hour) interval between randomization and achievement in 4. Reasons for ICU admission: postoperative period or hours medical or trauma-related 11. Occurrence of intubation to treat delirium (Y/N), 5. Delirium characteristics: delirium of persecution (Y/N), and date of diagnosis inability to reason logically (Y/N), abnormal 12. Occurrence of hypotension (Y/N), hypertension perceptionsorhallucinations (Y/N),temporal (Y/N), bradycardia (Y/N), tachycardia (Y/N), disorientation (Y/N), non- or misrecognitions rhythm disorders (Y/N), myocardial ischemia (Y/N) (Y/N), difficulties in naming objects (Y/N) or writing after study drug onset (Y/N), fluctuating course during the same day (Y/N) 13. Occurrence of hypotension requiring any 6. Date/hour of symptom onset vasopressor administration, date, vasopressor drug 7. Time between admission to ICU and enrollment used, maximal dose, total dose and duration) after (hours) study drug onset 8. CAM-ICU score  14. Occurrence of respiratory distress (Y/N) or apnoea 9. RASS score  (Y/N) after study drug onset 10. BPS NI score  or Visual Analogic Scale (VAS) 15. Patients receiving rescue drug (clorazepate) because score for pain of maximum tolerated dose is reached (Y/N, type, dose, duration) At randomization, the following data will be collected: 16. Daily self-evaluation of sleep quality on digital scale if possible (Y/N and 0 to 10) 1. Date/hour of randomization 17. Length of ICU stay 2. Time between admission to ICU and 18. Date of hospital discharge randomization (min) 19. Death (Y/N and date) 3. CAM ICU and RASS scores at H0 4. Behavioural Pain Scale in Non-intubated patients Thirty days after randomization: (BPS NI) score or VAS score at H0 5. Patient’s treatment and use of pharmacotherapy: 1. Length of ICU stay (in days) morphine or opioids (Y/N), vasopressor drugs 2. Date of hospital discharge (as obtained from (Y/N): type and infusion rates, antipsychotic (Y/N), hospital notes) benzodiazepines (Y/N), antidepressant drug (Y/N) 3. Survival status (If the patient is deceased, date of death) Daily from randomization (08.00) to day 7 (or hospital discharge): If the patient is still present on day 30, follow-up will continue until hospital discharge: 1. Daily lowest values for heart rate, blood pressure, peripheral O saturation, respiratory rate, 1. Occurrence of intubation to treat delirium (Y/N temperature and date) 2. Daily blood glucose level, blood pH, results of 2. Duration of mechanical restraint prescribed and samples of plasma creatinine and creatinine carried out if it occurs Louis et al. Trials (2018) 19:307 Page 9 of 11 3. Number of ventilator free days at day 30 Publication plan 4. Survival status: alive (Y/N), if the patient is The trial is registered at http://www.clinicaltrials.gov. deceased, date of death Upon trial completion, the main manuscript will be sub- mitted to one of the major clinical journals regardless of At 1 year after randomization: the results. All trial sites, including patients, will be acknowledged, and all investigators at these sites will 1. Survival status (if the patient is deceased, date of appear with their names under “the 4D investigators” in death) an appendix to the final manuscript. The 4D trial Steer- 2. Short Form Health Survey, 36 items (SF36) score ing Committee will grant authorship in adherence to the Vancouver guidelines and number of patients enrolled Data handling and retention by the individual investigator. If a trial site investigator is Data will be handled according to French law. All ori- to gain authorship, the site has to include 10 patients or ginal records (including consent forms, reports of sus- more. If the site includes 20 patients or more, two pected unexpected serious adverse events, and relevant authorships will be granted. A writing committee will be correspondences) will be archived at the trial sites for composed of members of the Steering Committee and 15 years. The clean trial database file will be anonymized investigators to define the order of authors of any and maintained for 15 years. publications. The listing of authors will be as follows: T Godet Enrollment and timeline (Clermont-Ferrand site investigator) will be first author, The patients are expected to be included from 13 French C Louis (Clermont-Ferrand site investigator) will be university and non-university hospitals during a 1-year second author, then other members of the Steering period starting in December 2017. Committee and trial site investigators depending on the Listing of trial centers at study initiation is presented number of included patients per site and per month below: (trial site investigator with the greater number of inclusion per month will be third author), B Pereira CHU de Clermont-Ferrand (biostatistician) will be third to last author, G Hospices Civils de Lyon, service d’anesthésie- Chanques (associated investigator) will be second to réanimation last author, JM Constantin (principal investigator) will CHU de Montpellier, hôpital St-Éloi, service de be responsible for the writing of the manuscript and réanimation chirurgicale DAR B will appear as the last author and then ‘for AZUREA AP-Hôpitaux de Marseille, service de réanimation Network’ will be added. CHU de Nîmes, service de réanimation chirurgicale CHU de Nantes, service d’anesthésie-réanimation Finances chirurgicale The 4D trial is funded by an institutional grant from Centre Hospitalier Le Puy-en-Velay, service de Clermont-Ferrand University Hospital and a grant from réanimation AZUREA network (www.azurea.org). Centre Hospitalier de Montluçon, service de Management and logistic of the trial drug distribution réanimation to each of the 13 participating centers that are antici- Centre Hospitalier d’Aurillac, service de réanimation pated to be recruiting will be coordinated using the CHU de Nice, service d’anesthésie-réanimation web-based randomization system by the pharmacy of CHU de Grenoble, service d’anesthésie-réanimation the Clermont-Ferrand University Hospital. The receipt, CHU de Saint-Etienne, service d’anesthésie-réanimation storage and dispensing of the blinded trial drugs will be CHU de Tours, service d’anesthésie-réanimation the responsibility of the pharmacy department in each individual trial site. This will be performed in accord- Each participating center has to include five patients ance with accredited standards for routine pharmacy per month (holidays excluded) to complete inclusions in practice. less than 1 year. Funding sources have no influence on trial design, trial 2016–2017: protocol, approvals from the Ethics Com- conduct, data handling, data analysis or writing of the mittee, and trial tool development (eCRF, randomization manuscript. system). 2018 to 2019: inclusion of patients. 2020: cleaning and closure of the database Perspectives Mid-2020: data analyses and writing of the manuscript, Millions of patients undergo ICU admission world- and submission for publication wide each year. Delirium occurrence is frequent and is Louis et al. Trials (2018) 19:307 Page 10 of 11 associated with increased patients’ short- and long- significant impact on future treatment of non- term morbidity and mortality as well as healthcare- intubated ICU patients presenting with delirium. associated costs. Few studies have examined the effects of dexmedetomidine in the treatment of agi- Trial status tated delirium in non-intubated ICU patients and The 4D trial is currently recruiting patients. much of the available data are extrapolated from mechanically ventilated patients. As far as the investi- Additional file gators are aware, no other large RCTs are assessing the efficacy and safety of dexmedetomidine or placebo Additional file 1: Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 Checklist: recommended items to with the goal of delirium treatment in non-intubated address in a clinical trial protocol and related documents. (DOCX 571 kb) ICU patients. Abbreviations Discussion BPS NI: Behavioural Pain Scale in Non-intubated patients; CAM Performing the 4D trial is in line with conclusions ICU: Confusion Assessment Method for the ICU; CHU: Centre Hospitalier from the 2013 recommendations of the American Universitaire (University Hospital); eCRF: Electronic (web-based) case report form; ICU: Intensive care unit; RASS: Richmond Agitation Sedation Scale; College of Critical Care Medicine on the clinical prac- SAPS: Simplified Acute Physiology Score; SOFA: Sequential Organ Failure tice guidelines for the management of pain, agitation Assessment; VNS: Visual Numeric Scale and delirium in adult patients in the ICU , espe- Acknowledgements cially in those not requiring tracheal intubation and We wish to thank the following: clinical staff at all trial sites and data mechanical ventilation. management team at Clermont-Ferrand University Hospital. Dexmedetomidine is widely used in ICU patients as a component of sedation therapy for delirium treatment Funding The 4D trial is funded by Clermont-Ferrand University Hospital and a grant or prophylaxis. However, haloperidol remains the most from AZUREA research consortium. widely used treatment of agitation and delirium in ICU [30, 31], but may reveal ineffective and associated with Availability of data and materials Data and material will be made available on request after discussion with potential cardiovascular complications. The low level of the Steering Committee. evidence of the few studies on the subject encouraged us to conduct the 4D trial, and to compare dexmedetomi- Authors’ contributions dine to placebo since haloperidol, and perhaps second- CL drafted the manuscript, designed the study and is a member of the Steering and Management Committee. TG drafted the manuscript, designed generation antipsychotics, should not remain first-line the study and is a member of the Steering and Management Committee. GC delirium treatments in non-intubated patients as stated designed the study. NB is a member of the Management Committee. DM is in 2013 guidelines on agitation, pain and delirium . a member of the Management Committee. BP planned the statistical analysis, is member of the Steering and Management Committee. JMC is the Indeed, recent trials have reported inefficiency of halo- principal investigator of 4D, drafted the manuscript, designed the study and peridol, when compared to placebo, to cure delirious is member of the Steering and Management Committee; All authors read ICU patients  and to prevent delirium appearance in and approved the final manuscript. high-risk patients . Carrasco and colleagues  Ethics approval and consent to participate investigated dexmedetomidine for the treatment of Due to the specific medical condition of recruited patients (agitated hyperactive delirium refractory to haloperidol in non- delirium), an emergency inclusion procedure will be possible. Written pursuit consent will be obtained from all participants (once the medical condition has intubated ICU patients. The trial concluded to a poten- resolved) or their next-of-kin. The Institutional Review Board of the University tial cost-benefit of dexmedetomidine over haloperidol in Hospital of Clermont-Ferrand (France) approved the trial. By 7 June 2017, the this non-RCT. However, this trial compared two groups: study had been approved for all centers by a central ethics committee (Comité de Protection des Personnes Sud-Est V, Grenoble, France, 17-CHCF-02) with the haloperidol alone versus dexmedetomidine plus halo- registration number EudraCT 2017–000731-14. Any further additional important peridol (since the mean elimination half-life is 21 h) with protocol modification will require the approval of a central ethics committee dexmedetomidine as a second-line treatment. Important (Comité de Protection des Personnes Sud-Est V, Grenoble, France, 17-CHCF-02). The 4D trial was registered on 23 October 2017 at http://www.clinicaltrials.gov information from the trial by Carrasco et al. is the excel- with trial identification number NCT 03317067. lent tolerance of co-administration of dexmedetomidine and haloperidol. As reported by these authors, dexmede- Competing interests tomidine could be the medication of choice to treat JMC received consulting fees from Baxter. TG reports receiving lecture fees and support for meeting (travel delirious patients due to specific properties: absence of reimbursement) from Baxter Gambro. excessive sedation, easy titration, fewer side effects than The remaining authors declare that they have no competing interests. neuroleptics and rare interactions with other drugs. Whatever the result of the 4D trial, this will provide Publisher’sNote needed and new data on the efficacy and safety of Springer Nature remains neutral with regard to jurisdictional claims in commonly used medications, which could have a published maps and institutional affiliations. Louis et al. Trials (2018) 19:307 Page 11 of 11 Author details 16. Teegarden BM, Prough DS. What to do when haloperidol fails to treat Département de Médecine Périopératoire (MPO), Centre Hospitalier agitated delirium: is dexmedetomidine the next step? Crit Care Med. 2016; Universitaire (CHU) Clermont-Ferrand, 63003 Clermont-Ferrand, France. 44:1426–8. GReD; UMR/CNRS6293; Université Clermont-Auvergne; INSERM U1103, 17. Pichot C, Longrois D, Ghignone M, Quintin L. Dexmedetomidine and 63003 Clermont-Ferrand, France. Département d’Anesthésie-Réanimation B, clonidine: a review of their pharmacodynamy to define their role for Hôpital Saint-Eloi, Centre Hospitalier Universitaire (CHU) Montpellier, 34090 sedation in intensive care patients. Ann Fr Anesth Reanim. 2012;31:876–96. Montpellier, France. Centre Hospitalier Universitaire (CHU) Clermont-Ferrand, 18. Reade MC, O’Sullivan K, Bates S, Goldsmith D, Ainslie WR, Bellomo R. Délégation à la Recherche Clinique et à l’Innovation (DRCI), 63000 Dexmedetomidine vs. haloperidol in delirious, agitated, intubated patients: Clermont-Ferrand, France. Département de Médecine Périopératoire (MPO), a randomised open-label trial. Crit Care. 2009;13:R75. Hôpital Estaing, Centre Hospitalier Universitaire (CHU) Clermont-Ferrand, 1 19. Reade MC, Eastwood GM, Bellomo R, Bailey M, Bersten A, Cheung B, Davies place Lucie Aubrac, 63003 Clermont-Ferrand, France. A, Delaney A, Ghosh A, van Haren F, et al. Effect of dexmedetomidine added to standard care on ventilator-free time in patients with agitated Received: 21 November 2017 Accepted: 28 April 2018 delirium: a randomized clinical trial. JAMA. 2016;315:1460–8. 20. Carrasco G, Baeza N, Cabre L, Portillo E, Gimeno G, Manzanedo D, Calizaya M. Dexmedetomidine for the treatment of hyperactive delirium refractory to haloperidol in non-intubated ICU patients: a nonrandomized controlled trial. Crit Care Med. 2016;44:1295–306. References 21. European Medicines Agency. Guideline on data monitoring committees. 1. Ely EW, Girard TD, Shintani AK, Jackson JC, Gordon SM, Thomason JW, Pun 2005 http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_ BT, Canonico AE, Light RW, Pandharipande P, Laskowitz DT. Apolipoprotein guideline/2009/09/WC500003635.pdf. Accessed 9 Oct 2017. E4 polymorphism as a genetic predisposition to delirium in critically ill 22. Bennett DA. How can I deal with missing data in my study? Aust N Z J patients. Crit Care Med. 2007;35:112–7. Public Health. 2001;25:464–9. 2. Bergeron N, Dubois MJ, Dumont M, Dial S, Skrobik Y. Intensive Care 23. Schafer JL. Multiple imputation: a primer. Stat Methods Med Res. 1999;8:3–15. Delirium Screening Checklist: evaluation of a new screening tool. Intensive 24. Little R, Rubin D. Statistical analysis with missing data. 2nd ed: Wiley- Care Med. 2001;27:859–64. Interscience; 2002. 3. Ely EW, Inouye SK, Bernard GR, Gordon S, Francis J, May L, Truman B, Speroff 25. O'Brien PC. Procedures for comparing samples with multiple endpoints. T, Gautam S, Margolin R, et al. Delirium in mechanically ventilated patients: Biometrics. 1984;40:1079–87. validity and reliability of the confusion assessment method for the intensive 26. Oh ES, Fong TG, Hshieh TT, Inouye SK. Delirium in older persons: advances care unit (CAM-ICU). JAMA. 2001;286:2703–10. in diagnosis and treatment. JAMA. 2017;318:1161–74. 4. Chanques G, Garnier O, Carr J, Conseil M, de Jong A, Rowan CM, Ely EW, 27. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Jaber S. The CAM-ICU has now a French "official" version. The translation Score (SAPS II) based on a European/North American multicenter study. process of the 2014 updated Complete Training Manual of the Confusion JAMA. 1993;270:2957–63. Assessment Method for the Intensive Care Unit in French (CAM-ICU.fr). 28. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, Anaesth Crit Care Pain Med. 2017;36:297–300. Reinhart CK, Suter PM, Thijs LG. The SOFA (Sepsis-related Organ Failure 5. Confusion Assessment Method for the ICU (CAM-ICU) The Complete Assessment) score to describe organ dysfunction/failure. On behalf of the Training Manual. 2014. http://www.icudelirium.org/docs/CAM_ICU_training. Working Group on Sepsis-Related Problems of the European Society of pdf. Accessed 1 Oct 2017. Intensive Care Medicine. Intensive Care Med. 1996;22:707–10. 6. Ely EW, Truman B, Shintani A, Thomason JW, Wheeler AP, Gordon S, Francis 29. Chanques G, Payen JF, Mercier G, de Lattre S, Viel E, Jung B, Cisse M, Lefrant J, Speroff T, Gautam S, Margolin R, et al. Monitoring sedation status over JY, Jaber S. Assessing pain in non-intubated critically ill patients unable to time in ICU patients: reliability and validity of the Richmond Agitation- self report: an adaptation of the Behavioral Pain Scale. Intensive Care Med. Sedation Scale (RASS). JAMA. 2003;289:2983–91. 2009;35:2060–7. 7. Chanques G, Jaber S, Barbotte E, Verdier R, Henriette K, Lefrant JY, Eledjam 30. Mo Y, Zimmermann AE, Thomas MC. Practice patterns and opinions on JJ. Validation of the French translated Richmond Vigilance-Agitation Scale. current clinical practice guidelines regarding the management of delirium Ann Fr Anesth Reanim. 2006;25:696–701. in the intensive care unit. J Pharm Pract. 2017;30:162–71. 8. Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell FE Jr, Inouye SK, 31. Kotfis K, Zegan-Baranska M, Zukowski M, Kusza K, Kaczmarczyk M, Ely EW. Bernard GR, Dittus RS. Delirium as a predictor of mortality in mechanically Multicenter assessment of sedation and delirium practices in the intensive ventilated patients in the intensive care unit. JAMA. 2004;291:1753–62. care units in Poland - is this common practice in Eastern Europe? BMC 9. Milbrandt EB, Deppen S, Harrison PL, Shintani AK, Speroff T, Stiles RA, Anesthesiol. 2017;17:120. Truman B, Bernard GR, Dittus RS, Ely EW. Costs associated with delirium in mechanically ventilated patients. Crit Care Med. 2004;32:955–62. 10. Dubois MJ, Bergeron N, Dumont M, Dial S, Skrobik Y. Delirium in an intensive care unit: a study of risk factors. Intensive Care Med. 2001;27: 1297–304. 11. O’Connor PG, Schottenfeld RS. Patients with alcohol problems. N Engl J Med. 1998;338:592–602. 12. Barr J, Fraser GL, Puntillo K, Ely EW, Gelinas C, Dasta JF, Davidson JE, Devlin JW, Kress JP, Joffe AM, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;41:263–306. 13. Schrijver EJ, de Graaf K, de Vries OJ, Maier AB, Nanayakkara PW. Efficacy and safety of haloperidol for in-hospital delirium prevention and treatment: a systematic review of current evidence. Eur J Intern Med. 2016;27:14–23. 14. Page VJ, Ely EW, Gates S, Zhao XB, Alce T, Shintani A, Jackson J, Perkins GD, McAuley DF. Effect of intravenous haloperidol on the duration of delirium and coma in critically ill patients (Hope-ICU): a randomised, double-blind, placebo-controlled trial. Lancet Respir Med. 2013;1:515–23. 15. van den Boogaard M, Slooter AJC, Bruggemann RJM, Schoonhoven L, Beishuizen A, Vermeijden JW, Pretorius D, de Koning J, Simons KS, Dennesen PJW, et al. Effect of haloperidol on survival among critically ill adults with a high risk of delirium: The REDUCE randomized clinical trial. JAMA 2018, 319:680–690.
– Springer Journals
Published: Jun 4, 2018