Innovative ICU Solutions to Prevent and Reduce Delirium and Post–Intensive Care Unit Syndrome

Alawi Luetz, MD, PhD; Julius J. Grunow; Rudolf Mörgeli, MD; Max Rosenthal, MD, PhD; Steffen Weber-Carstens, MD, PhD; Bjoern Weiss; Claudia Spies, MD, PhD


Semin Respir Crit Care Med. 2019;40(5):673-686. 

In This Article


The Role of Light in Critical Illness

Visible light is the part of the electromagnetic spectrum that human eyes can perceive visually, and it exerts various effects on the human body. In this context, targeted medical interventions utilizing light have been described since the beginning of modern medicine.[96,97] More recently, light therapy has been shown to improve mood disorders, such as depression,[98] seasonal affective disorder,[99] and sleep disorders.[100]

Sleep is an essential part of the human life cycle[101] and sleep deprivation leads to impaired neurocognitive function,[102,103] as well as dysregulation of inflammatory pathways.[104] Critically-ill patients are especially susceptible to harmful effects of sleep deprivation, as their mental and physical capacity to withstand it are inherently low. Disturbances in the sleep–wake circle are almost always seen in delirious ICU patients.[105] Recent evidence revealed that sleep-architecture is severely impaired in critically-ill patients in comparison to healthy controls,[106,107] showing atypical sleep and absence of normal polysomnographic sleep characteristics. Moreover, the lack of K-complexes or sleep spindles has been shown to be significantly associated with higher mortality in ICU patients.[108] Poor sleep quality is frequently named as one of the top stressors for ICU patients.[109–112]

Light is the most critical external regulating factor of the circadian clock. Circadian rhythm affects all cells and has internal (genetical and epigenetical) and external regulations. The circadian clock itself is located in the suprachiasmatic nucleus (SCN) which is part of the anterior part of the hypothalamus. As such, the SCN plays an essential role in regulating the sleep–wake cycle, cognition, and brain metabolite clearance.[113] One major factor influencing suprachiasmatic nucleus activity is the influx of blue light reaching melanopsin photoreceptor cells in the retina, leading to suppression of melatonin secretion. Conversely, a decrease in the amount of blue light entering the eye decreases SCN activity and triggers increased release of melatonin.[114,115]

Melatonin supplementation is known to improve sleep quality in healthy persons and noncritically-ill patients.[116–118] However, the effects of melatonin and melatonin receptor agonists on sleep quality and outcome of critically-ill patients have been inconsistent.[119–121] Studies on melatonin release in critically-ill patients observed that patients exhibited either constantly high- or low-melatonin levels. This phenomenon that has been reported in several studies and constantly high-melatonin levels are especially prevalent in patients with a high severity of illness.[122] As abnormally high-melatonin levels were found in many critically-ill patients, it was reasoned that suppression of melatonin levels could lead to improved sleep architecture and quality. Since light is known as a potent suppressor of melatonin levels, its effect in ICU patients was studied accordingly.[123,124]

Using Natural Light

As hospital architecture in the early 20th century focused on efficiency over aesthetics,[14] ICUs were often designed to maximize patient visibility.[125] Such ICUs often consisted of individual patient compartments, separated by curtains and surrounding main observation stations. In general, the need for windows and natural light was not a priority.[1] Current recommendations for the design of ICU rooms include sufficient amounts of natural, as well as artificial light, in an attempt to improve patient's well-being and staff workflow.[23]

Experimental data suggest that environments with an abundance of natural light offer beneficial immunomodulatory effects, as compared with environments with restricted light influx. Alibhai and colleagues showed that the rate of myocardial healing, as well as the type of cellular immune response, following myocardial infarction in a mouse-model was significantly different in mice kept under high and low levels of natural light influx, respectively.[126]

The effect of natural light on non-ICU patients was studied by Joarder and Price by quantifying the level of exposure to natural light in hospitalized patients, reporting a dose–response relationship of natural light influx and decreased hospital length of stay.[127] Additionally, the lack of natural light has been implicated as a contributing factor to the disturbances in the sleep–wake cycle in ICU patients.[128] Wunsch and colleagues analyzed the influence of ICU room windows in patients with traumatic brain injury. The presence of windows in the ICU rooms was viewed as a surrogate for the amount of natural light reaching the patients. In their cohort, the presence of windows in ICU rooms did not have any effect on patient outcomes.[129] However, a major problem when interpreting the findings of Wunsch et al is that, there are not quantifiable data on the level of light actually reaching the patient's retina and, by extension, the photoreceptor cells. Deep sedation levels in critically-ill patients are another confounder that must be taken into consideration when interpreting data from light studies in this population.

Zaal and colleagues analyzed the effect of ICU environment on patient outcomes, with a move from wards with low influx of natural light, to single-patient rooms with large window fronts and high influx of natural light. Patients in the single-patient rooms had a lower duration of delirium, although the incidence of delirium did not differ significantly.[17]

In summary, the inconsistency of study results might be due to the enormous degradation of radiant natural light (99%) entering the room through the window to the ICU bed.[130] Consequently, for most of the scenarios, the circadian effective irradiance, which is a measure of potential circadian efficacy, is not high enough to exert any attributable biological effect in those patients.[131]

Using Artificial Light

In consequence of the poor controllability and effectiveness of natural light sources in the ICU, technical solutions using artificial light therapy are coming into focus. So far, bright-light therapy alone has not been extensively studied in critically-ill patients but was shown to improve delirium in geriatric patients[132] and postsurgical cohorts.[133] A recent randomized-controlled trial using a dynamic light application in 734 ICU patients failed to prove a significant reduction in the cumulative incidence of delirium.[134] Consequently, there are no comprehensive data supporting the use of bright-light therapy in the critically–ill patients.[135,136]

Within a systematic review evaluating the effect of light and darkness on the outcome of ICU patients, the authors emphasize that heterogeneity among interventions and designs is high, which makes interpretation of results and a general conclusion difficult.[137] Major limitations among most of the reviewed studies are that patients were either sedated or that depth of sedation was poorly documented. Sedated patients tend to have more frequent and prolonged closed-eye phases, thus greatly limiting the amount of light hitting the retina. Interventions that target neuronal activity via light application are only effective if patients' eyes are open, as the stimulation of the retinal cells is a required precondition to affect the SCN activity and melatonin secretion.[138] Consequently, deeply sedated critically-ill patients exposed to bright light (>10,000 lux) for 1 hour, showed no discernible changes in melatonin levels.[139]

To ensure circadian efficacy of light interventions, compliance to distinct photometric parameters need to be assessed. The circadian effective irradiance (Ec ) is a measure of the melanopic illuminance. The Ec is calculated by using measured data of spectral irradiance and must be weighted with the action spectrum for melatonin suppression.[140,141] Thresholds for maximal melatonin suppression (MMS) are known for healthy young adults (0.3 W × m−2) and healthy people > 60 years (0.6 W × m−2).[142] However, it is entirely unknown whether these values are applicable for the ICU patient. It is very likely, that thresholds for MMS in critically-ill patients are different from that of healthy adults and probably need to be adjusted depending on the clinical condition of the patient.

The timing of light exposure is another potentially important factor that has not been adequately addressed in studies applying light therapy in the ICU. It is known that the timing of blue-light influx can have a significant effect on circadian rhythmicity. Münch and colleagues found that using light with a higher portion of shorter wavelengths in the early morning can be used as a counterstrategy for low-illuminance levels during the day or for high-light exposure in the evening.[143]

In conclusion, utilizing light as a reliable intervention to stabilize circadian phase or to induce circadian phase-shifting requires adherence to predefined photometric parameters, and securing the site of action, namely, the retina. Results from studies that do not control for these conditions will not add valuable knowledge to this field of research.

Avoiding Harmful Effects of Light

As indicated previously, a high illuminance of >1,000 lux might be necessary to achieve a sufficient Ec at the patient's eye level. Most of the technical lighting solutions with high illuminance use relatively small light-emitting areas (e.g., 1.5 m × 0.4 m) with high-luminous intensity (candela, cd × m−2). Luminance is a measure of how bright a light source is perceived. Corresponding therapy lamps often exceed the threshold level of absolute glare (>10,000 cds), causing significant discomfort and glare.

Concerning patients acceptance of light therapy, the color rendering index (CRI), which quantifies the capability of a light source to illuminate object colors "realistically" and "acceptably" is a likewise important parameter. Daylight, a reference light source in the CRI system, has a maximum CRI of 100%.

To make light therapy feasible in the ICU, the clinician has to consider these parameters as well.[131]

As studies have demonstrated unusually high-illuminance levels during night time in the ICU,[144] the concept of shielding the individual patient from the potentially harmful light sources have also been investigated. Sleep masks as part of a multicomponent intervention have shown effectiveness in terms of improving patients' subjective sleep quality. But again, study methods are very heterogeneous and the results are inconclusive. Since sensory deprivation can lead to disruptions in cognition, as well as hallucinations, even in healthy patients,[145–147] such measures might be feasible in nonsedated patients who can decide whether or not to wear the devices but might be difficult in delirious patients. The perceived loss of agency can be a major source of stress for patients that cannot regulate the level of sensory deprivation.[22] Furthermore, patients suffering from disorientation, delirium, or anxiety might deteriorate when reorientation via external stimuli is mitigated, as reorientation seems to be an effective attenuator in these conditions.[148] In fact, the trend to reduce sedative use in the critically-ill aims to allow an increased interaction with the environment, a principle that is contrary to the use of forced sensory deprivation.

Utilizing carefully selected room features to minimize the illuminance levels reaching the patients' retina during nighttime, might be a gentle alternative to sleep masks, supporting a physiological day–night cycle.

Conceptualizing Light as an Interventional Modality

So far, there are no comprehensive data supporting the use of light therapy as a single-concept intervention in ICU patients. However, considering the significant methodological limitations of the performed studies, this does not exclude potential beneficial effects on outcome, when administered properly using an adequate dosage. Similar to pharmacologic interventions, light can be conceptualized as a therapeutic modality. In order for a targeted intervention to be successfully applied and evaluated, various factors need to be defined and controlled, including a concrete definition of the intended target mechanism. For light therapy, the most popular target mechanism is currently the suppression of melatonin release by exposing retinal photoreceptors to light.

The modality must be delivered in a way that allows proper utilization of the active compound. In the case of light, it must be certain that light actually reaches the photoreceptor cells of the retina. Level of sedation and patient compliance are major potential factors that must be regarded, and the distance of the light source to the patient is another modifiable factor influencing this variable.

Every targeted intervention must be dosed in a way that ensures sufficient levels of the active ingredient reaching the target structure. Analogously, light source technology should be designed in such fashion that illuminance and frequency can be adjusted according to the intended effect.

The timing of therapeutic modalities, in terms of duration and frequency of application, are equally important for light therapy as for any pharmacological intervention. In order for light therapy to be adequately studied, timing and frequency of light application must be reported in future trials.

Finally, the modality should be delivered in a way that maximizes the active compound while minimizing side effects. One of the most frequent adverse events related to light interventions is glare due to high luminance. Glare is a disruptive stressor, which in turn decreases patient compliance with the therapeutic modality, that is, patients look away or shield their eyes from the light source, effectively diminishing the uptake of the active compound, and analogous do not taking a prescribed medication.