Rationale: Delirium is a frequently encountered problem in ICU patients and leads to increased morbidity and mortality; Delirium in the ICU is associated with sleep deprivation which is among others caused by a disrupted circadian rhythm; Dynamic Light application aims at restoring a proper circadian rhythm by rhythmically alternating light intensity and has shown beneficial effects in sleep quality. Whether DLA improves sleep quality and reduces delirium incidence in ICU patients is not known

Goals/endpoints:

To evaluate the feasibility of dynamic light application in the ICU and to study the effects of dynamic light application on the incidence of delirium, duration of mechanical ventilation, the number of ICU and hospital days, and mortality in a mixed population of medical and surgical ICU patients. In a subgroup of patients with a high risk of developing delirium, markers of circadian rhythm, inflammation and brain damage and post ICU HRQoL will be assessed Study design: prospective randomized single centre trial Study population: adult ICU patients > 18 years old with an expected duration of stay of more than 24 hours Intervention: Patients will be randomized between Standard Care or Standard Care + DLA; When receiving standard care, normal lighting settings will be used in that patient room, which can be controlled by the medical personnel; In the rooms of patients randomized to the DLA group, DL is applied with a changing intensity during the day according to a fixed rhythm, which is regulated centrally. In addition when necessary, an intervention light can be used which can be operated in the patient room.

Study parameters/endpoints: incidence of delirium as measured by the CAM-ICU; duration of mechanical ventilation, ICU and total hospital mortality; ICU and hospital LOS; Serum levels of inflammatory markers and markers of brain damage, urinary levels of markers of circadian rhythm, data of HRQoL questionnaires and total light exposure in both groups

STUDY DESIGN This is a prospective randomized, single centre trial. The proposed starting date is 01-july- 2011 and the duration of the study will be 15 months.

All ICU patients who are expected to have an ICU stay of more than 24 hours are potential eligible. On ICU admission, patients will be screened for eligibility. The patients' demographics, patients' health status, medication use, APACHE II score, SOFA score and admission diagnosis will be reported at baseline. Before recruitment and enrollment in the study, each patient and/or relative will be given full explanation of the study and will be informed that they are free to discontinue their participation in the study at any time.

When considered eligible, patients will be given a unique patient number in consecutive order and will be assigned to the treatment corresponding with this number. Patients will be randomized in a one-on-one fashion according to a computer generated randomization list between Standard Care and Standard Care + DLA. The randomization list will be kept at the data co-ordination centre.

All of the 20 patient rooms on the 2 ICU wards are equipped with a special lighting system. This lighting system can offer the standard basic lighting in the patient room, with a set intensity and light colour, or the DL setting which offers changing light conditions depending on the time of the day, following a rhythmic pattern. In addition, all rooms have a basic low-intensity orientation light, which is turned on during night-time in all rooms.

The switch between the two modes is regulated centrally on a control panel that is located in the nursing station and cannot be altered in the patient-room

1.1 Standard Care On admission, a full physical examination is performed and, when possible, a history taken. An ECG and chest x-ray, when not performed shortly before, will be done. In addition, blood will be drawn from the indwelling arterial catheter for haemoglobin, leucocyte and thrombocyte count, renal and liver function, electrolytes, arterial blood gas analysis, lactate and markers of inflammation (C-reactive protein, procalcitonin). Blood pressure, heart rate, fluid balance, oxygen saturation are measured continuously and automatically registered in our Patient Data Management System (PDMS). When considered necessary, ventilatory support is commenced or continued (post-operative patients) and additional hemodynamic measurements performed.

All patients will receive thromboprophylaxis, when no clear contraindications exist, and ulcer prophylaxis. When patients are intubated, they receive analgesic and often sedative medications for patient comfort. If patients are expected to receive mechanical ventilation for at least 48 hours, Selective Digestive Decontamination (SDD) prophylaxis is commenced, consisting of application of oral and gastrointestinal antibiotic suspension, administration of four days of i.v. antibiotics (cefotaxime) and surveillance cultures.

Twice daily at 0800hrs and 2000hrs, routine blood analysis consisting of arterial blood gas analysis, haemoglobin, leucocyte and thrombocyte electrolytes, renal function and markers of inflammation, is performed. When no clear contraindications exist, sedative infusions are interrupted every morning since this has proven to decrease length of mechanical ventilation (39). In addition, when considered feasible, an attempt is made to "wean" the patient off the mechanical ventilator. Protocol-guided early mobilisation is commenced as soon as possible, directed by dedicated physical therapists.

Level of sedation is monitored as part of daily care by nurses in all patients every two hours using the Richmond Agitation Sedation Scale (RASS). This scale divides levels of consciousness indicated by a number ranging from -5 (unarousable) to +4 (combative, see Appendix). Ideally, patients have a RASS of 0 (alert and calm) and sedation is titrated to achieve a RASS of -2 to 0. In addition, screening on the presence of delirium is done using the CAM-ICU every eight hours by well trained ICU nurses . As has been mentioned before, this method has been validated for use in ICU patients and is easy to use by non-psychiatric personnel. When the CAM-ICU is positive, delirium treatment is started according to our protocol (see Appendix).

Standard measures to prevent delirium among our patients include maintaining a proper day-and- night rhythm by promoting night-time sleep (reduction of light and noise disturbance and reduction of daytime sleep using daily activity programs and mobilisation. When patients have visual or auditive impairments, glasses and hearing aids are being used as much as possible. A clock is present in each ICU room and relatives are asked to bring photographs of them so that patients can have some sense of familiarity. Dosages of deliriogenic medications (anticholinergics, opiates, benzodiazepines) are reduced as much as possible however cannot be ruled out completely since benzodiazepines and opiates are often used for sedation and analgesia.

When a patient is randomized to receive the basic light setting, standard lighting is used, including the possibility of using a high intensity intervention light when performing interventions. The light switch in the room then offers three options: 1) main light on 2) main light off 3) intervention light on/off.

1.2 Standard Care + DLA In addition to standard care, inpatients who are randomized at receiving DLA, the light switch in that room is set to DLA, as soon as possible after the patient is installed and initial investigation and handlings (e.g. placing urinary catheter, a central venous line and/or arterial cannula) have been performed. When interventions are necessary, a high intensity intervention light can be turned on manually by physicians or nurses. The lighting operation panel in this room then offers only one option, i.e. intervention light on/off. A computer will log the use of the dynamic lighting system,

1.3 Assessment During the first 24 hours, the risk of developing delirium will be assessed using the validated PRE-DELIRIC score. When patients with have a risk of developing delirium ≥>40%, they will be defined as a high risk patient.

In all patients, the presence of delirium is assessed three times a day until death or discharge off the ICU or death on fixed times using the CAM-ICU. When patients are unconscious, either because of their illness or due to sedative medications, the CAM-ICU cannot be performed and delirium cannot be assessed. This will be recorded in the case-record form.

Since delirium is associated with the use of sedatives and opioids, 24-hour dosages of these medications will be recorded daily at 8 am. For a list of these medications, see appendix.

In addition, renal function, (plasma creatinine and urea), electrolytes (sodium, potassium, ionized calcium) haemoglobin, white blood cell count, procalcitonin) will be monitored daily until discharge of the ICU. Total hospital length of stay will be registered for all patients. Patients will be analyzed in an intention-to-treat principle.

To assess differences in light exposure between the two groups, light levels are measured in every patient room, close to the patient's head. Since daylight may also influence total light exposure and differences exist in daylight exposure of the different patient rooms (See appendix 3) differences between light exposures in different rooms will be analysed and associated to the primary endpoint.

In patients with a high risk of developing delirium 5 ml of blood will be drawn on day 1,3,5,7, 14,21 and 28 and will be stored at -80 degrees Celsius until analysis. In addition in these high risk patients, circadian rhythm derived biomarkers will be determined by means of urinary excretion collected in 3-hour urinary samples during 24 hours on day 7, 14,21, 28 and then once per 2 weeks. To determine pre-existent quality of life a validated Health Related Quality of Life questionnaire will be used.

1.4 randomisation, blinding and allocation of treatment Patients will be randomized in one-on one fashion according to a computer generated randomization list. The randomization list will be kept at the data co-ordination centre. Since blinding is impossible with dynamic light, patients and relatives, nurses and doctors are aware of the treatment arm.

1.5 Study procedures DLA will be applied in the DLA group. In all patients 5 ml of blood will be drawn within 24 hours after admission and stored. In high risk patients, an extra 5 ml of blood will be collected during the morning laboratory rounds on day 1, 3, 5, 7, 14, 21 and 28. Since all patients have an indwelling arterial line and blood is already collected this is not considered to be an additional burden for the patient. In total a maximum of ca. 35 millilitres of blood during a period of 28 days will be collected. In addition in high risk patients, 3 -hour urinary samples will be collected during 24 hours on day 7, 14, 21, 28 and than every week until discharge off the ICU. On our ICU, all patients have an indwelling urinary catheter. On admission, a validated QoL questionnaire will be handed over to the patient or his/her next of kin to determine the pre-existent QoL; after 3, 6 and 12 months a QoL questionnaire will be sent to the patient to determine the post ICU QoL.

Analysis will be performed on the whole group but also in a post-hoc analysis where the effect of DLA will be analysed per season of admission.

Inclusion Criteria:

• ICU-patients >18 yrs old
• expected duration of stay > 24 hrs

Exclusion Criteria:

• life expectancy of <48 hrs on ICU admission
• necessity of prolonged deep sedation
• blindness
• inability to speak or understand dutch