Medical hypothesis: Light at night is a factor worth considering in critical care units

Abstract

Exposure to light at night is not an innocuous consequence of modernization. There are compelling data linking long-term exposure to occupational and environmental light at night with serious health conditions, including heart disease, obesity, diabetes, and cancer. However, far less is known about the physiological and behavioral effects of acute exposure to light at night. Among healthy volunteers, acute night-time light exposure increases systolic blood pressure and inflammatory markers in the blood, and impairs glucose regulation. Whether critically ill patients in a hospital setting experience the same physiological shifts in response to evening light exposure is not known. This paper reviews the available data on light at night effects on health and wellbeing, and argues that the data are sufficiently compelling to warrant studies of how lighting in intensive care units may be influencing patient recovery.

1. Introduction to circadian rhythms and health risks associated with light at night

Life on Earth evolved under 24 h solar days. Circadian rhythms evolved as adaptations to respond to predictable daily cycles of light and dark for the synchronization of vital behavioral and biological processes to the external environment. For nonhuman animals, restricting activities to the appropriate temporal niche is crucial to fitness and survival. For humans, temporal organization of physiology is equally important for health and wellness. Over the past century, the boundaries between day and night have been obscured by the widespread adoption of electric lighting devices at night. Disruption of our internal circadian rhythms has become common in developed countries. For humans, the health consequences of chronic circadian disruption by nighttime light exposure are becoming increasingly apparent [1,2]. Exposure to artificial light at night is linked to increased risk of breast cancer, prostate cancer, metabolic disorders, cardiovascular disease, and psychiatric disorders [3–19]. These clinical effects are remarkably consistent and have been reproduced in animal models of light at night exposure [20–34]. The majority of these studies have focused on adverse health outcomes associated with weeks to years of exposure to light at night. In contrast, this paper will focus on the potential effects of acute exposure to light at night on the health of vulnerable populations, viz., patients in hospital intensive care units.

Before the invention of electric lights, approximately 130 years ago, humans were exposed to minimal light at night. A full moon on a clear night provides about 0.1–0.3 lux (lx) in the temperate zones [35] or up to ~1.0 lx in the tropics [36]. A single candle casts ~11 lx of light on an object a foot away. At the end of the 19th century, with the invention of the electric bulb, exposure to artificial light at night grew rapidly. For the first time, humans could artificially extend the day. As technology advanced, humans were exposed to additional sources of light at night, including television and computer screens, smart phones, and tablet computers. Today, greater than 80% of the World’s human population and >99% of those living in the United States or Europe reside in regions with significant light pollution [37]. Light intensities on an average urban street are estimated to be ~5–15 lx, whereas night-time light intensities in a typical living room are between 100 and 300 lx [35]. Furthermore, as individuals age and their visual systems deteriorate, their needs for indoor lighting may change [38]. Night-time lighting also is typically found in hospitals, out of concern for the safety of patients and the staff working at night. Indeed, assessing the 24-h profiles of light intensity to which patients in a typical ICU are exposed revealed two important outcomes: patients are exposed to relatively low light intensities during the day (~160 lx) and frequent nighttime light intrusions (average light at night exposure of ~10 lx) [39]. These results suggest that ICU patients are likely maintained in an environment without distinct day-night cues, which could lead to dysregulated circadian rhythms, in turn affecting health outcomes [40].

2. Night-time light and circadian dysregulation

There are pronounced circadian rhythms in physiology and behavior that are critical for the optimal functioning of all mammalian species; these rhythms are coordinated by the suprachiasmatic nucleus [41,42]. Because the typical intrinsic circadian period for humans is approximately 24.2 h, the circadian pacemaker needs to be reset daily by zeitgebers (time givers or environmental cues) to achieve circadian entrainment to the solar day [43]. The principal zeitgeber for humans is light, thus humans living without clear daily light-dark cycles risk dysynchronizing their circadian rhythms. However, not all light is an equally effective zeitgeber; for example, critical factors include the irradiance and wavelength of the light, the pattern, duration, and time of day of exposure, as well as the recent photic history of the individual (reviewed in [43]). Even relatively low intensity light (~180 lx) significantly phase-shifts the human circadian pacemaker [44] when paired with dark nights, and under certain circumstances as little as 1.5 lx is sufficient to entrain a 24 h circadian rhythm [45]. Thus, humans are exquisitely sensitive to modulation of circadian rhythms by light.

2.1. Circadian rhythms in melatonin

The nocturnal rise and diurnal decline in endogenous melatonin and its metabolites are commonly used as indicators of circadian phase in humans [46]. Melatonin synthesis and secretion occurs almost exclusively during darkness, and among humans the peak in melatonin typically occurs several hours after the onset of darkness. However, exposure to light during the subjective night causes an immediate suppression of melatonin synthesis [47], which occurs in a dose-dependent manner based on the luminance (lux) of the light source [48]. Other lifestyle factors, such as exercise or caffeine consumption also can shift the night-time melatonin peak [49,50]. Furthermore, the amplitude of the nighttime rise in melatonin tends to become smaller as adults age [51–53]. Importantly, however, beyond its function as a daily and seasonal time-keeper, melatonin is a hormone with a wide array of biological influences, and as such, environmentally-induced alterations in its secretion may have very important health consequences.

2.2. Melatonin and disrupted sleep

Difficulty sleeping is one of the most common complaints about hospital stays [54]. This is not surprising given that light levels up to 1000 lx at night have been reported in ICU settings [55]. Indeed, one recent study reported that exposure to as little as 5–10 lx of light during just two nights significantly affects sleep architecture [56]. Furthermore, night-time light levels between 300 and 500 lx can disrupt the central circadian clock; whereas nocturnal light levels between 100 and 500 lx can affect melatonin secretion in most people [57,58]. Whether light-induced suppression of melatonin is responsible for the observed sleep disruption is not known. The endogenous melatonin rhythm is associated with rhythms in sleep propensity, and among individuals with sleep disorders, providing exogenous melatonin improves sleep–wake cycles [59,60], including among critically ill patients [61]. Likewise, a smaller night-time peak in endogenous melatonin is associated with reports of poorer sleep quality in older women (>60 years of age) [62]. However, the effects of suppressing endogenous melatonin on sleep in humans are not clear [63]. At any rate, both suppressed melatonin and disrupted sleep rhythms have the potential to hamper patient recovery.

Given the numerous disruptive factors commonly experienced in an ICU, including light, noise, pain, administration of medication, and routine patient care activities such as phlebotomy and vital sign monitoring, it is not surprising that abnormal sleep is often reported and observed [64]. Several polysomnography studies have confirmed these observations. Interestingly, a recent study revealed that ICU patients and healthy adults spend comparable amounts of time asleep per day, but the timing is altered in patients; specifically, among ICU patient, sleep occurs in approximately equal amounts during the day and night, as opposed to primarily at night [54]. ICU patients also tend to display prolonged sleep latency (difficulty falling asleep), substantial sleep fragmentation (multiple arousals), decreased sleep efficiency (ratio of time spent asleep to time in bed), elevated stage 2 sleep (“light sleep”), decreased stage 3 (“deep or slow-wave”) sleep, and decreased rapid eye movement (REM) sleep [54,65–67]. Thus, ICU patients exhibit atypical sleep architecture and experience poor sleep quality, which in turn can hamper recovery and precipitate other physiological, cognitive, and behavioral disorders commonly associated with disrupted or insufficient sleep (reviewed in [68]). Indeed, a clinical trial of cardiology ward patients suggests that providing enhanced daytime brightness and restricting nocturnal light exposure improves sleep over the course of the hospital stay as compared to standard hospital room lighting, although that lighting manipulation did not affect length of stay or mood [69].

3. Inflammation and immune function

There are well-established circadian rhythms in immune function that can influence disease course (reviewed in [70,71]). Likewise, many studies have reported that sleep deprivation alters immune function in healthy adults (e.g., [72–74]) and increases susceptibility to infection [75]. Thus, ICU lighting that disrupts circadian rhythms and sleep may imperil patient recovery (reviewed in [76]). Indeed, a study of ICU patients with severe sepsis demonstrated that maintaining the lights low on the unit (varying from ~2 lx/min to ~70 lx/min) prevented the maintenance of a typical circadian rhythm in melatonin; samples were collected every 4 h to assess urinary 6-sulfatoxymelatonin (6-SMT is the primary melatonin metabolite in urine), but yielded no evidence of the typical peak in the early morning or daytime nadir [77]. This study did not examine outcome in these patients, but rodent studies have consistently demonstrated that eliminating the light-dark cycle of animals following the induction of sepsis reduces survival [40,78], and that supplementing with exogenous melatonin improves survival [79]. Similarly, a study using rats demonstrated that there is a circadian effect on pancreatitis-induced inflammation; the protective effect at night is likely due to elevated endogenous melatonin concentrations [80]. The extent to which nighttime light exposure affects inflammation in human patients is not established, but there also is a growing literature in non-human animals indicating that exposure to dim light at night exacerbates inflammation to a variety of proinflammatory stimuli [25,81,82]. Thus, ICU environments that increase exposure to night-time light and are not conducive to maintaining circadian rhythms and typical sleep patterns may not be optimal for patient recovery, particularly for health conditions with an inflammatory component.

4. Post-operative delirium and sundowning syndrome

Delirium occurs in up to 80% of critically ill patients [83] and its etiology is multifactorial [84]. It is particularly common among elderly patients who undergo surgery, and may be linked to disrupted melatonin rhythms during recovery. For example, in a small study of patients who received major abdominal surgery, a typical melatonin secretion pattern was observed among patients who did not develop delirium; in contrast, those who developed delirium without surgical complications had a shifted melatonin pattern, whereas those who developed delirium, as well as surgical complications, had abnormally high melatonin throughout the day [84]. Furthermore, patients with delirium slept primarily during the day, which is consistent with prior studies demonstrating an association between sleep disturbances and post-operative delirium [84]. Similarly, in a study of thoracic esophagectomy patients, ICU psychosis was highly correlated with an irregular melatonin circadian rhythm [85]. These studies do not establish a causal relationship between sleep-disturbances or melatonin rhythm and delirium, but they do demonstrate an intriguing correlation.

Sundowning syndrome, also known as sundown syndrome or “nocturnal delirium”, is a type of delirium that is common among ICU patients and is characterized by a temporally specific pattern of recurring disruptive behaviors [86]. Specifically, individuals with sundowning syndrome become increasingly agitated, aggressive, restless, anxious, vocal, or delirious as the late afternoon or early evening approaches, whereas symptoms improve or resolve during the day. Although the syndrome is most often reported in dementia patients, cognitively-intact elderly individuals also may display sundowning symptoms during ICU hospitalization (e.g., [87]). It has been hypothesized that sundowning is a degradation in mood regulation as circadian rhythms decay in aging [88] or as the result of neurodegenerative disorders [89,90]. Both aging and dementia also are associated with reduced pineal function; indeed, melatonin concentrations in CSF are reduced by approximately 50% in older individuals and 80% in Alzheimer’s patients relative to middle aged adults [91,92]. The absence of clear light-dark phases in the ICU also could contribute to the emergence of sundowning syndrome in the elderly and dementia patients by causing further disruption of melatonin secretion.

Melatonin treatment has been used in several clinical studies to improve delirium and sundowning symptoms, which is consistent with the hypothesis that melatonin may help regulate sleep and serve as a circadian rhythm synchronizer in the absence of salient light-dark cycles. Open label pilot studies demonstrate that providing exogenous melatonin reduces daytime sleepiness and evening agitation among dementia patients residing in nursing homes [93,94]. Likewise, a small study of elderly patients admitted through the emergency department indicates that low-dose melatonin administered nightly significantly reduces the risk of developing delirium during the hospital stay [95]. Meta-analyses support similar conclusions; i.e., that melatonin treatment improves sundowning symptoms and disrupted sleep–wake cycles [59,60]. Likewise, exposure to two hours per day of bright light appears to improve symptoms among Alzheimer’s patients with severe sundowning syndrome [96]. These studies are mainly small pilot studies or case reports, thus large, appropriately powered trials are necessary before a clear understanding about the role of circadian rhythms in sundowning syndrome emerges, and recommendations about lighting to minimize the expression of sundowning within the ICU can be formulated.

5. Other affective consequences of light at night

A recent study reported that greater than 60% of ICU patients exhibited depressive symptoms at discharge [97]. Although the underlying cause of depressive symptoms among this patient population may involve both somatic and psychogenic factors, the sleep disruption and nighttime light exposure commonly experienced in the ICU also may contribute to the manifestation of affective disorders. There are both human and rodent data indicating that light at night increases depressive/depressive-like symptoms. For example, a large cross-sectional study of geriatric individuals in Japan reported that depressed individuals had a significantly higher prevalence of light at night exposure (>5 lx) than non-depressed participants [8]. Likewise, middle-aged people living in San Diego who participated in a similar study had mean bedroom illumination ranging between 13 lx in the early night to 29 lx during the last two hours before waking; among these individuals increased nocturnal illumination was correlated with increased depressive symptomology (CES-D; [98]). Similarly, increases in depressive-like behavior have been reported in rodents after several weeks of exposure to dim light at night [24,25,99]. Importantly, a similar behavioral effect is apparent in diurnal rodents [23]. Thus, light at night increases depressive symptoms/depressive-like behavior in a variety of species, including humans.

6. Cardiovascular and metabolic consequences

The exacerbating effects of long-term night shift work on risk factors for cardiovascular disease are well-described, and include increased carotid atherosclerosis [100], hypertension [101], diabetes [102], metabolic syndrome [103–105], and inflammation [106]. Thus, it is not surprising that sustained shiftwork is associated with an increase in myocardial infarction and other significant coronary events [11,107]; the risk of coronary heart disease declines over time after cessation of shift work, thereby indicating that the negative effects of nightshift are not irreversible [11]. However, they do emerge rapidly; a recent laboratory study convincingly demonstrated that short-term circadian misalignment can have potentially detrimental cardiovascular effects. Inverting behavioral and environmental cycles for three days was sufficient to significantly increase systolic and diastolic blood pressure (particularly during sleep), high sensitivity c-reactive protein (a marker of systemic inflammation), and proinflammatory cytokines concentrations in healthy young adults [108]; similar effects were obtained in shift workers who participated in a laboratory study with a cross-over design that involved simulated day shift work and night shift work [109].

Less is known about potential cardiovascular effects of exposure to light at night among individuals who are not shift workers. However, a home-based study with 528 elderly Japanese people demonstrated a significant increase in night-time systolic and diastolic blood pressure among individuals whose bedroom had nighttime light levels >5 lx; interestingly, this blood pressure effect occurred in the absence of a concomitant changes in urinary melatonin concentrations or actigraphic sleep quality [110]. A home study with a cross-over design comparing the effects of dark versus light (1000 lx) nights on heart function and breathing in healthy young adults reported an increase in the apnea-hypopnea index and the ratio of low frequency power to high frequency power (suggesting altered cardiovascular sympathetic control) on the light compared to dark nights; these changes occurred in the absence of an effect on sleep latency or efficiency [111]. Furthermore, a small sleep lab study that compared blood pressure among individuals exposed to dim light (<10 lx) versus bright light (>3000 lx) at night likewise reported that an elevation of systolic blood pressure among the bright light group emerged during the night, and was reduced by exogenous melatonin treatment [112]. Together, these studies indicate that many of the detrimental cardiovascular effects that have been so well-characterized in nightshift workers, can emerge among non-shift workers who are merely exposed to LAN, sometimes even in the absence of altered sleep. The rapid onset of these cardiovascular effects may have implications for patients experiencing nighttime light exposure in an ICU.

Nightshift work also is associated with glucose intolerance and insulin insensitivity, as indicated by increased postprandial plasma glucose and insulin concentrations [113]. Likewise, increased early evening light exposure at home is associated with an increase in prevalent diabetes among elderly individuals [10]. Similarly, mice housed in either bright or dim light at night for several weeks have significantly increased body mass and reduced glucose tolerance compared with mice housed in standard dark nights, despite comparable caloric intake and locomotor activity among the groups [114]. Thus, there is compelling evidence from both human and animal studies demonstrating a link between chronic light at night exposure and the development of metabolic syndrome, however, indications of disruptions in glucose homeostasis can occur much earlier. For example, a single night of exposure to bright lights (>500 lx) is sufficient to significantly suppress salivary melatonin and increase postprandial plasma glucose and insulin concentrations, thereby suggesting glucose intolerance and insulin insensitivity [115]. Likewise in rats, light at night is sufficient to induce glucose intolerance and insulin insensitivity upon initial exposure, an effect that is influenced by time of night, wavelength, and light intensity [34]. Together, these data suggest that the effects of light at night on blood pressure and glucose homeostasis emerge rapidly, which in turn could be particularly detrimental to recovering patients.

7. Summary

There is a growing and increasingly compelling literature demonstrating the detrimental effects of long-term exposure to light at night on life-threatening health conditions. Although many of the initial studies focused on health disparities identified among night shift workers, recent studies using environmental and in-home light measures suggest that exposure to night-time light also is associated with increased risk for a variety of health conditions, including cancer and metabolic disorders, among individuals who are not shift workers. Indeed, rodent studies have confirmed that chronic light at night alters body mass, glucose regulation, cancer progression, response to chemotherapy, and affective behavior. Furthermore, there are now several laboratory-based clinical studies demonstrating that acute exposure to light at night is sufficient to alter inflammatory measures, glucose regulation, and blood pressure in otherwise healthy individuals. Thus, it is conceivable that the short-term nighttime light exposure experienced in an ICU could alter physiological parameters of immune function, blood pressure, and glucose metabolism, as well as mental health, in a way that would be detrimental to recovering patients. Together, these data provide the rationale for studying whether maintaining a strong light-dark cycle in ICUs could improve patient recovery, as well as psychological outcomes. Although the mechanism underlying night-time light effects on human health remains unspecified, the available data strongly support light-induced suppression of melatonin. These data suggest a possible engineering solution: when nighttime illumination is required, it is possible to restrict the light spectrum to wavelengths that are less likely to suppress melatonin in humans [116]. This could be achieved through the use of lightbulbs with controllable or restricted wavelengths, or through fitting patients with light filtering goggles that block out wavelengths that stimulate non-image forming retinal ganglion photoreceptors and suppress melatonin. Likewise, the efficacy of using appropriately timed exogenous melatonin to counteract the effects of nighttime light on endogenous hormone is worth exploring further. It also would be valuable to examine how exposure to LAN may interact with other common stressors on ICU units, including noise, sleep disruption by staff assessing vital signs, illness-specific physiological changes, and pain. Lastly, studies of nighttime light exposure on physical and psychological health and well-being also have broad implications for integrative medicine and public health.

What is already known about the topic?

• Maintaining appropriate circadian rhythms is crucial to health and well-being.

• Exposure to light at night causes physiological and behavioral changes in a variety of species, including humans.

• Chronic exposure to light at night is associated with increased incidence of life threatening illnesses, including cancer, metabolic disorders, and cardiovascular disease.

What does this paper add?

• This review describes the data associated with acute exposure to light at night, which suggest the potential for rapid effects on health and well-being.

• This review argues that there is a need for studies on the possible effects of nighttime light exposure on critically ill patients.