DANish DELIrium Study On Neurointensive Care Patients (DANDELION)

Background It has been established that in medical and surgical intensive care units (ICUs) that delirium increases mortality and length of stay (LOS), as well as the risk of dementia and cerebral atrophy after discharge. There is a lack of studies into delirium in the neurointensive care unit (N-ICU) population, and no studies have addressed the association between delirium, quality of life, dementia symptoms and cerebral atrophy in patients with acquired brain injury after discharge from N-ICU. A recent study has demonstrated a correlation between delirium symptoms in stroke patients and quality of life, and functional levels at 3, 6 and 12 months after N-ICU discharge, suggesting that delirium during the acute phase effects the long-term outcome.

Delirium screening has not been introduced systematically in N-ICUs in Denmark or internationally, and validated instruments are lacking in this population. The Confusion Assessment Method for ICU (CAM-ICU) and the Intensive Care Delirium Screening Checklist (ICDSC) are internationally most recommended and best validated delirium screening tools for non-psychiatric health care staff. Neither of the two instruments, however, has been fully validated in patients with acquired brain injury.

Several studies and guidelines suggest that delirium is potentially preventable by minimizing sedation, and ensuring systematic pain management, early mobilization and adequate sleep. The present study therefore aims to investigate if delirium is preventable in an N-ICU setting by introducing a multi-modal intervention bundle optimizing management of sedation, pain, mobilization and sleep.

Aim

The primary aim of the study:

To measure the effect of the intervention bundle on prevalence, duration and type of delirium symptoms using the Intensive Care Delirium Screening Checklist (ICDSC) and Confusion Assessment Method for the ICU (CAM-ICU) in N-ICU patients with acute acquired brain injury

Secondary aim: To validate ICDSC and CAM-ICU for adult N-ICU with acute acquired brain injury, and investigate if delirium symptoms are negatively correlated with quality of life, cognitive function and dementia symptoms at 12-month follow-up post N-ICU discharge.

Method Study design: A controlled two-phased before-and-after trial. Phase I: Baseline study (4 months), Phase II: Intervention study (4 months). This design was chosen as the most appropriate, because a randomized clinical trial would have a large spill-over effect.

The primary outcome is duration of delirium symptoms with ICDSC and CAM-ICU in the N-ICU based on a hypothesis that it is possible to decrease time with delirium in the N-ICU 20% or more. Power calculation: 56 patients will be enrolled during a 4-month period, in the baseline and the intervention group respectively, yielding a total of 112 patients (20% drop-out rate, 5% significance, 80% power).

Inclusion criteria All Danish speaking adults (18 years or above) with an expected N-ICU length of stay (LOS) more than 48 hours and admitted with acute brain injury from spontaneous or traumatic bleeding, such as subarachnoid hemorrhage (SAH), intracranial hemorrhage (ICH) or traumatic brain injury (TBI).

Exclusion criteria Patients not expected to survive 48 hours, or continues Richmond Agitation-Sedation Score (RASS) of -4 or -5 during the stay in the N-ICU.

Data collection Demographics: age, gender, primary diagnosis, marital status and living situation, education, BMI, and alcohol consumption. Clinical data include: Glasgow Coma Score (GCS), Richmond Agitation and Sedation Scale (RASS), ICDSC, CAM-ICU, Acute Physiology and Chronic Health Evaluation II (APACHE II), Simplified Acute Physiology Score (SAPS II), intubation status, sedation agent and dose, type of mobilization and pain status. Follow-up includes EuroQoL-5D (EQ-5D), Repeatable Battery for Assessment for the Neuropsychological Status (RBANS) and Mini Mental State Examination (MMSE) questionnaire at 12 months after discharge.

Phase I Baseline study (4 months). A baseline investigation will be conducted to establish the prevalence, duration and type of delirium symptoms in patients with acute acquired brain injury. Usual care will be provided at this time. For validation a psychiatric team will rate included patients usingThe Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) and The International Classification of Diseases (ICD-10) as the gold standard. The psychiatric team assesses patients the first and second day after the patient has a RASS -2 or above for 24 hours.

Phase II Intervention study (4 months). After full implementation of the intervention bundle, the study will be conducted to establish the prevalence, duration and type of delirium symptoms in patients with acute acquired brain injury. The psychiatric team assesses patients the first and second day after the patient has a RASS -2 or above for 24 hours.

Implementation of intervention An all day workshop introducing the multi-modal intervention bundle will be held for all nurses in the unit. The study nurses will be specially trained to maintain study protocol. Physicians will be informed and trained on a weekly morning meeting during phase II, and some will participate in the all day workshop. Journal audits will be used as monitoring the implementation. The Intervention bundle consist of a sedation, sleep, mobilization and pain treatment regimen based on the newest available evidence in the four areas.

Sedation: Sedatives will be kept to a minimum by daily orders for target RASS sedation level and titration of sedatives by primary nurse. Dose and agent will be recorded.

Sleep: Sleep will be optimized by systematic use of earplugs at night, dim lighting, visitation restrictions, and midday rest periods. Procedures (e.g. blood samples) will be carried out in the day time instead of night time. Hours of activity/rest will be assessed by an actiwatch (actigraph).

Pain: A strategy of multi-modal pain management combining pharmacological and non-pharmacological interventions will be implemented. Pain will be assessed by Critical-Care Pain Observation Tool (CPOT) for unresponsive patients and Numeric Rating Scale (NRS) for patient-reported pain.

Mobilization: Systematic early mobilization will be implemented by daily orders for mobilization level and a mobilization guideline.

Follow-up: Follow-up will be conducted on all possible patients in the two study groups at twelve months post N-ICU discharge. Patients will come to the hospital for the EQ-5D for self-rated health, RBANS for cognitive function and the MMSE questionnaire for dementia symptoms.

Ethical considerations The study includes an observational and an interventional part introducing optimized treatment and care according to current best international evidence. The intervention items will be implemented in the entire unit as standard care for all patients. Most patients will be unable to give initial consent to the study due to their cerebral status. Relatives and general practitioner will be contacted and informed verbally and in writing to provide consent by proxy. The study is approved by the Regional and national ethics committees and the Danish Data Protection Agency.

Data analysis Prevalence, type and duration of delirium symptoms in the baseline group will be compared to the intervention group. Furthermore the two groups will be compared on LOS and mortality. Patients ICDSC score are compared with EQ-5D, RBANS and MMSE in order to investigate possible associations between delirium and self-rated health, cognitive function and dementia.

All data will be coded and plotted in Excel and SPSS for the statistical data analysis. Hardware and worksheets are handled and stored as prescribed by the Danish Data Protection Agency. There will be used descriptive and analytical statistics. Categorical variables will be described as frequency and percent. Continuous variables will be described as mean and standard deviation (SD). The two groups will be compared by correlation analysis, univariate and multivariate regression analyzes and questionnaires are analyzed by a paired t-test to compare the two groups' average score.