Use of eye masks and earplugs in intensive care units: systematic review

Abstract

Background

Delirium is a common neuropsychiatric disorder but underdiagnosed in intensive care units due to the different presentations it can manifest. Nurses have an important role to play in preventing its onset and promoting sleep in their patients.

Aim

To describe the efficacy of the use of eye masks and earplugs in adult patients admitted to intensive care units on sleep quality and the development of delirium.

Methods

Systematic literature review of randomized clinical trials. During June and July 2024, the CINHAL, ScienceDirect, Medline, LILACS, ProQuest, Web of Science, and BioMed Central databases were consulted, applying filters based on time period, population, type of study, topic, and scientific journal. Studies were selected that analyzed the use of eye masks and earplugs as nursing care and whose methodological quality achieved a medium-high score using the CASPe tool.

Results

We found (N = 3742) studies. After applying filtering and selection criteria, a total of eight (n = 8) clinical trials were collected. With a total of 611 participants, it was observed that the use of eye masks and earplugs improved subjective sleep quality, while the data obtained on the incidence of delirium reflected an inconclusive result for its prevention due to the heterogeneity of the results.

Conclusions

Studies indicate that the use of sleep masks and earplugs in intensive care unit patients can improve sleep quality and reduce delirium. However, comfort and tolerance vary among patients, and it is crucial to consider duration of use and removals only for short periods of time. Although some studies show significant benefits, others find no relevant differences, suggesting the need for further research to optimize its use. The review has not received any funding.

Resumen

Introducción

El delirium es un trastorno neuropsiquiátrico común pero infradiagnosticado en las unidades de cuidados intensivos debido a las diferentes presentaciones que puede manifestar. El personal de enfermería tiene una importante labor para prevenir su aparición y favorecer el sueño de sus pacientes.

Objetivo

Describir la eficacia del uso de antifaces y tapones para los oídos en el paciente adulto ingresado en unidades de cuidados intensivos sobre la calidad del sueño y el desarrollo de delirium.

Metodología

Revisión sistemática de la literatura de ensayos clínicos aleatorizados. Se consultaron durante los meses de junio y julio de 2024 las bases de datos CINHAL, ScienceDirect, Medline, LILACS, ProQuest, Web of Science y BioMed Central, aplicándose filtros de carácter temporal, poblacional, tipo de estudio, temático y relacionado con la revista científica. Se seleccionaron estudios que analizaran el empleo de antifaces y tapones como cuidados de enfermería y cuya calidad metodológica alcanzase una puntuación media-alta mediante la herramienta CASPe.

Resultados

Se hallaron (n = 3.742) estudios. Tras aplicar filtrado y criterios de selección se recopilaron un total de (n = 8) ensayos clínicos. Con un total de 611 participantes, se observó que el empleo de antifaces y tapones mejoraba la calidad subjetiva del sueño, mientras que los datos obtenidos en incidencia de delirium reflejaron un resultado no concluyente para su prevención debido a la heterogeneidad de los resultados.

Conclusiones

Los estudios indican que el uso de antifaces y tapones para los oídos en pacientes de la unidad de cuidados intensivos puede mejorar la calidad del sueño y reducir el delirium. Sin embargo, la comodidad y la tolerancia varían entre los pacientes, y es crucial considerar la duración de uso y las retiradas solamente durante breves periodos de tiempo. Aunque algunos estudios muestran beneficios significativos, otros no encuentran diferencias relevantes, lo que sugiere la necesidad de más investigación para optimizar su uso. La revisión no ha recibido fuentes de financiación.

Introduction

The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) defines acute confusional syndrome, or delirium, as an alteration of attention, cognition and awareness, in which there is a reduced ability to direct, focus, sustain and shift attention, together with reduced spatial orientation in the surrounding environment. This altered state is not associated with other pre-existing neurocognitive disorders, nor with situations in which patients have reduced arousal levels, such as sedation or coma.¹ It is a common neuropsychiatric disorder in intensive care units (ICU) and its diagnosis can be difficult due to the various forms of presentation it may adopt, such as hyperactive, hypoactive or mixed delirium.² Furthermore, it causes multiple complications for the health of patients, including increased mortality rate, longer hospital stays, and risk of developing cognitive impairment at discharge, costing millions in annual healthcare expenditure.^3,4

It is known that the development of delirium takes place due to the interaction of multiple risk factors, including the administration of sedatives,⁵ antibiotics,⁶ and vasopressors,⁷ advanced age,⁸ the use of mechanical ventilation⁹ (MV) and physical restraints,¹⁰ suffering an infectious process,¹⁰ and admission to an ICU caused by a severe condition.¹¹ Moreover, certain environmental factors coexist at these units, including high levels of noise and light, which do not favour patient rest and may alter their sleep-wake cycle, promoting the onset of delirium. The usual nursing care in these environments includes establishing that nursing staff, as the primary point of reference in managing patient wellbeing, should address these modifiable environmental factors to ensure an environment that is as quiet and dark as possible during night-time, as this significantly improves the quality of sleep of patients at these units.¹² However, when this cannot be achieved, the literature supports the use of eye masks and earplugs as tools to facilitate sleep at these units.^13,14 The purpose of this study is to describe the efficacy of these aids on sleep quality, as well as their impact on the incidence of delirium in the randomised clinical trials selected for analysis.

Objectives

General objective

• •

  To summarise the scientific evidence on the effect of eye masks and earplugs on the quality of sleep and development of delirium in adult patients admitted to intensive care units.

Specific objectives

• •

  To describe the available evidence on the effect of the use of eye masks and earplugs on the quality of sleep of adult patients admitted to intensive care units.

• •

  To describe the results observed in relation to quality of sleep and incidence of delirium among adult patients admitted to intensive care units who used eye masks and earplugs.

• •

  To analyse the assessment tools used to measure sleep and delirium in the studies examined, as well as the heterogeneity derived from their use.

• •

  To describe the relationship between the use of eye masks and earplugs and the length of stay at intensive care units.

• •

  To identify the drawbacks, tolerance problems and adverse effects associated with the use of eye masks and earplugs by patients admitted to intensive care units.

Methodology

Design

This was a systematic review of the literature on randomised clinical trials conducted through searches in the CINAHL, ScienceDirect, MEDLINE, LILACS, ProQuest Central, Web of Science and BioMed Central databases in June and July 2024. These databases were selected for their broad coverage of health-related disciplines and, specifically, nursing.

Research question

The research question was structured in “PIO” format, resulting in the following: What effect does the use of eye masks and earplugs have on the quality of sleep and incidence of delirium in adult patients admitted to intensive care units?

Selection of descriptors

Relevant keywords to respond to the research question were identified through the DeCS and MeSH thesauri. Table 1 presents the descriptors associated with the research question.

Table 1.

Description of the research question.

Descriptors
Natural language		Controlled language
Spanish	English	DeCS	MeSH
Unidad de cuidados intensivos	Intensive care unit	Unidades de Cuidados Intensivos	Intensive care unit
Paciente crítico	Critical patient	−	Critically ill
Delirio	Delirium	Delirio	Delirium
			Disorientation
			Confusion
			Acute confusion
Cuidados de enfermería	Nursing care	Atención de enfermería	Intervention Nursing care
Ensayo clínico aleatorizado	Randomized controlled trial	Ensayo Clínico Controlado Aleatorio	Randomized controlled trial
Antifaces para los ojos	Eye masks	−	−
Tapones para los oídos	Earplugs	−	Earplugs

Source: own elaboration.

Search strategies

To begin the search, the controlled-vocabulary MeSH terms were combined with the “AND” and “OR” Boolean operators. Based on the results, automatic and manual filters were applied to refine the search. These included: time filter (studies published between 2010 and 2024), population filter (studies whose sample included patients aged ≥18 years admitted to an ICU), filter related to the type of study (only randomised clinical trials were selected), filter related to the topic (studies whose main topic was delirium), and filter related to the scientific journal (studies published in journals specialising in intensive care and nursing science). Table 2 presents the search strategies employed in the databases consulted.

Table 2.

Search strategies.

Base of data	Search strategies	Results
CINAHL	AB (intensive care unit OR ICU OR critical care OR critical care unit) AND (delirium OR acute confusion OR confusion OR disorientation) AND randomized controlled trials	229
ScienceDirect	(critically ill OR intensive care unit) AND (delirium OR disorientation) AND (nursing care) AND (randomized clinical trials)	3034
MEDLINE	((((critically ill) OR (intensive care unit)) AND (delirium)) AND (randomised controlled trial)) AND (nursing care)	134
LILACS	(critically ill) OR (intensive care unit) AND (delirium) AND (nursing care) AND (clinical trial)	61
ProQuest Central	delirium AND earplugs AND eye masks	232
Web of Science	delirium AND earplugs AND eye masks	34
BioMed Central	Delirium AND earplugs AND eye masks	18

Source: own elaboration.

Selection criteria

After removing duplicate results in the various databases and excluding 2 publications (as they could not be obtained), a second selection process was conducted, applying the following criteria:

Inclusion criteria:

• •

  Studies with medium-high methodological quality, expressed as a score of ≥7 on the CASPe checklist for clinical trials.¹⁵

• •

  Studies analysing the use of eye masks and earplugs as nursing care among patients admitted to an ICU.

Exclusion criteria:

• •

  Studies which employed family visits, administration of melatonin, early mobilisation, and the use of music therapy as nursing care.

• •

  Studies which employed combined nursing care bundles.

All the phases of this process were conducted by a single researcher, who systematically evaluated all the records identified by the literature searches. No process of independent screening by multiple reviewers was carried out. Nevertheless, so as to minimise the risk of selection bias, the inclusion and exclusion criteria were applied in a strict and consistent manner throughout the entire process.

Results

The final number of articles selected for this review was eight (n = 8). Fig. 1 reflects the methodological process conducted, as a flow diagram. The analysis of results was carried out through a variable coding process, with the goal of finding similarities and differences between the clinical trials selected (Table 3).

Figure 1.

Flow diagram of PRISMA 2020.

Table 3.

Coding of variables.

Title	Coding
Effect of earplug/eye mask on sleep and delirium in intensive care patients^16	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask (from 22:30 h to 06:30 h) + Standard care. Control group → Standard care.
	✓ Sample size: 84 patients (42 in the Intervention group [IG] + 42 in the control group [CG]). IC of the 95%.
	✓ Instruments: Richards-Campbell Sleep Questionnaire (RCSQ), intensive care delirium screening checklist (ICDSC) and Glasgow Coma Scale (GCS).
	• Characteristics of participants:
	✓ Profile of patients: adults without hearing or vision problems, without pain, capable of removing the eye mask and earplugs if necessary and without previous sleep disorders. Glasgow >13 and ICDSC < 4.
	• Characteristics of treatment:
	✓ Should not require MV, or receive any opioid medication or general anaesthesia in the last 24 h.
	• Context characteristics:
	✓ Hospital in Türkiye.
	✓ Polysomnography: no.
	• Extrinsic characteristics:
	✓ Authors: Balci et al., 2021.
	✓ Journal: Nursing in Critical Care.
Effect of night-time earplugs and eye masks on sleep quality in intensive care unit patients^17	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask (from 22:00 h to 07:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 17 patients (8 IG + 9 CG). IC of the 95%.
	✓ Instruments: RCSQ, Pittsburgh sleep quality index (PSQI), Verran/Snyder-Halpern sleep scale (VSHSS), confusion assessment method for the ICU (CAM-ICU), Richmond agitation-sedation scale (RASS) and GCS.
	• Characteristics of participants:
	✓ Profile of patients: adults without deafness or bilateral blindness, severe dementia or encephalopathy, uraemic or hepatic encephalopathy, encephalitis, increased intracranial pressure, metabolic disorders, severe haemodynamic instability, high requirement of vasopressors and severe respiratory failure. ICU stay should be >24 h, Glasgow ≥13 and RASS between −1 and +1.
	• Characteristics of treatment:
	✓ Patients should not require medication to sleep nor interventions which facilitated sleep.
	• Context characteristics:
	✓ Hospital in Thailand.
	✓ Polysomnography and wrist actigraphy: yes.
	• Extrinsic characteristics:
	✓ Authors: Arttawejkul et al., 2020.
	✓ Journal: Indian Journal of Critical Care Medicine.
Effect of using eye masks and earplugs in preventing delirium in intensive care patients: a single- blinded, randomized, controlled trial^18	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask (from 23:00 h to 07:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 72 patients (42 IG + 30 CG). IC of the 95%.
	✓ Instruments: RCSQ, RASS, Nu-DESC scale for delirium screening, Acute physiology and chronic health evaluation (APACHE-II) and GCS.
	• Characteristics of participants:
	✓ Profile of patients: adults who did not suffer delirium upon admission, intoxication by drugs or alcohol, severe neurological deficit defined as coma, known cognitive impairment and patients who received blood transfusion in the first 24 h of hospitalisation. The ICU stay should be >24 h, GCS ≥ 8 and RASS >+3.
	• Characteristics of treatment:
	✓ Undergoing invasive MV (IMV) but conscious. Should not receive sedatives like benzodiazepines or antidepressants.
	• Context characteristics:
	✓ Hospital in Türkiye.
	✓ Polysomnography: no.
	• Extrinsic characteristics:
	✓ Authors: Kiliç and Kav, 2023.
	✓ Journal: Nursing in Critical Care.
Quality sleep using earplugs in the intensive care unit: the QUIET pilot randomised controlled trial^19	• Methodological characteristics:
	✓ Intervention group → Earplugs (from 22:00 h to 06:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 40 patients (20 IG + 20 CG). IC of the 95%.
	✓ Instruments: RCSQ and APACHE-II.
	• Characteristics of participants:
	✓ Profile of patients: adults admitted at ICU following cardiac surgery without pre-existing hearing difficulties or suspected or confirmed tympanic membrane tear.
	• Characteristics of treatment:
	✓ Patients undergoing IMV, placement of earplugs was carried out post-extubation, sedation with propofol and fentanyl.
	• Context characteristics:
	✓ Hospital in Australia.
	✓ Polysomnography: no.
	• Extrinsic characteristics:
	✓ Authors: Litton et al., 2017.
	✓ Journal: Critical Care and Resuscitation.
Impact of earplugs and eye mask on sleep in critically ill patients: a prospective randomized study^20	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask (from 22:00 h to 08:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 61 patients (30 IG + 31 CG). IC of the 95%.
	✓ Instruments: Scale of Sedation of Ramsay, simplified acute physiology score (SAPS-II), Charlson index, visual analogue Scale (VAS) for sleep, CAM-ICU, Hospital Anxiety and Depression Scale (HADS), PSQI and Impact of Event Scale Revised (IES-R).
	• Characteristics of participants:
	✓ Profile of patients: adults with no sedation for over 24 h, levels of sedation <3 on the Ramsay Scale and programmed ICU stay >48 h. In addition, they should not have a history of sleep disorders, psychiatric disease requiring chronic medication, known diagnosis of central neurological impairment, hepatic disease with encephalopathy, uncontrolled sepsis, severe hearing impairment or blindness.
	• Characteristics of treatment:
	✓ Sedation with morphine <0.01 mg/kg/min and vasoactive therapy with noradrenaline <0.3 μg/kg/min.
	• Context characteristics:
	✓ Hospital in France.
	✓ Polysomnography: yes.
	• Extrinsic characteristics:
	✓ Authors: Demoule et al., 2017.
	✓ Journal: Critical Care.
The efficacy of earplugs and eye masks for delirium severity and sleep quality in patients undergoing coronary artery bypass grafting in cardiac intensive care units: A single-blind, randomised controlled trial^13	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask (21:00-22:00h until 06:00-07:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 114 patients (57 IG + 57 CG). 95% CI.
	✓ Instruments: NEECHAM, VSHSS and RASS.
	• Characteristics of participants:
	✓ Profile of patients: adults undergoing cardiac surgery (coronary artery bypass grafting), conscious and oriented to person, place, time, and situation once admitted to the ICU. Patients with hearing, speech or visual impairment which prevented them from participating, pre-existing cognitive impairment and pre-existing delirium were excluded.
	• Characteristics of treatment:
	✓ Analgesia with paracetamol and codeine + morphine if pain was not relieved. No hypnotic medication was received post-surgery.
	✓ Sedatives administered pre-/post-surgery: midazolam, fentanyl, atracurium, thiopentone sodium and morphine.
	• Context characteristics:
	✓ Hospital in Iran
	✓ Polysomnography: no.
	• Extrinsic characteristics:
	✓ Authors: Afshin et al., 2024.
	✓ Journal: Australian Critical Care.
The impact of earplugs and eye masks on sleep quality in surgical ICU patients at risk for frequent awakenings^14	• Methodological characteristics:
	✓ Intervention group → Earplugs + Eye mask + Standard care. Control group → Standard care.
	✓ Sample size: 87 patients (44 IG + 43 CG). IC of the 95%.
	✓ Instruments: RCSQ, Modified Family Satisfaction in the ICU Survey (FS-ICU), CAM-ICU, SAPS-II and ASA-PS.
	• Characteristics of participants:
	✓ Profile of patients: adult women after free-flap breast reconstruction surgery. Pregnant patients, incarcerated or with a diagnosis of sleep apnoea, insomnia or other sleep disorders were excluded.
	• Characteristics of treatment: not indicated.
	• Context characteristics:
	✓ Hospital in the USA.
	✓ Polysomnography: yes.
	• Extrinsic characteristics:
	✓ Authors: Obanor et al., 2021.
	✓ Journal: Critical Care Medicine.
The effect of earplugs during the night on the onset of delirium and sleep perception: a randomized controlled trial in intensive care patients^21	• Methodological characteristics:
	✓ Intervention group → Earplugs (from 22:00h to 06:00 h) + Standard care. Control group → Standard care.
	✓ Sample size: 136 patients (69 IG + 67 CG). 95% CI.
	✓ Instruments: RASS, NEECHAM, GCS, Simplified Acute Physiology Score 3 (SAPS-3), Sepsis-related Organ Failure Assessment (SOFA), Risk, Injury, Failure, Loss and End Stage (RIFLE) and nursing activity through the Simplified Therapeutic Intervention Scoring System (TISS28).
	• Characteristics of participants:
	✓ Profile of patients: adults who spoke Dutch or English with ICU stay >24 h and GCS score ≥10. Patients with known hearing impairment, dementia, confusion or delirium upon admission were excluded.
	• Characteristics of treatment:
	✓ No sedation.
	• Context characteristics:
	✓ Hospital in Belgium.
	✓ Polysomnography: no.
	• Extrinsic characteristics:
	✓ Authors: Van Rompaey et al., 2012.
	✓ Journal: Critical Care.

The number of randomised clinical trials included in the analysis was eight (n = 8). Of these, six (n = 6) examined the use of eye masks and earplugs in combination with standard care in the intervention group, whereas in the control group only standard care was applied. The two (n = 2) remaining studies analysed the use of earplugs together with standard care in the intervention group, while the control group only received standard care. The total number of participants recruited in the eight trials was 611.

Delirium and measurement tools

With regard to delirium, the studies conducted by Arttawejkul et al.,¹⁷ Demoule et al.,²⁰ and Obanor et al.¹⁴ employed the CAM-ICU scale as a diagnostic tool. Arttawejkul et al. reported that the incidence of delirium was similar between both groups, with one case in the control group and one case in the intervention group (P = 1.000). Similarly, Demoule et al. did not find significant differences in the rate of incidence of delirium, with two cases in each group (P = 1.000). Obanor et al. did not observe any statistically significant differences in the CAM-ICU scores between the two groups. By contrast, Van Rompaey et al.²¹ and Afshin et al.¹³ used the NEECHAM scale to assess delirium. Van Rompaey et al. found that the use of earplugs at night significantly reduced the incidence of confusion among ICU patients, with scores of 26 in the intervention group versus 24 in the control group (Mann-Whitney U, P = .04); moreover, patients who slept with earplugs had an improved cognitive status (P = .006). Patients who slept with earplugs developed confusion later compared with those who did not, and this was confirmed by a hazard ratio of 0.47, which suggests 53% less risk of experiencing delirium or confusion in the group with earplugs within the first 24 h post-admission. Afshin et al.¹³ reported significant differences in the severity of delirium between groups in postoperative days two, three and four (P < .001). Kiliç and Kav18 employed the Nu-DESC scale and observed statistically significant differences in the onset of delirium on the second night (P = .019), during the third day (P < .001) and on the third night (P < .001). In addition, they identified that admission at an internal medicine ICU (odds ratio [OR], 11.84; 95% confidence interval [95% CI] 3.00–46.66; P = .017), admission to this service from an operating room, age equal to or greater than 65 years, hearing impairment and lower education levels were factors favouring the development of delirium. Balci et al.¹⁶ used the ICDSC scale and observed that the mean score was significantly lower in the experimental group (U = 614.000, P = .004). Finally, Litton et al.¹⁹ did not employ any specific diagnostic tool for delirium, as they considered this variable as a reason not to use or to remove earplugs.

Quality of sleep and measurement tools

Regarding the evaluation of quality of sleep, a total of five trials were identified which used the Richards-Campbell Sleep Questionnaire (RCSQ) as a measurement instrument. In addition, other scales used in the studies included the Pittsburgh Sleep Quality Index (PSQI), the Verran/Snyder-Halpern Sleep Scale (VSHSS), and the Visual Analogue Scale (VAS). In the study by Balci et al.,¹⁶ patients in the experimental group reported significantly higher scores in the RCSQ at two different times, compared with the control group (U = 198.000, P < .001; U = 70.000, P < .001). In turn, Arttawejkul et al.¹⁷ assessed the quality of sleep during patient enrolment via the PSQI and subsequently measured the same variable using the RCSQ, and found no statistically significant differences between groups (P = .236). Kiliç and Kav18 reported that the mean total score in the quality of sleep scale of the intervention group was significantly higher than in the control group during the three nights they were assessed (P ≤ .001). By contrast, Litton et al.¹⁹ did not detect significant differences between groups using the RCSQ. Similarly, Demoule et al.²⁰ did not observe statistically significant differences in self-evaluated quality of sleep at the time of discharge from the ICU, neither using the VAS scale (60 in the control group versus 70 in the intervention group; P = .63), nor using the PSQI (P = .25). Conversely, Afshin et al.¹³ found significant differences in the dimension of quality of sleep between groups throughout the intervention period (P < .001). Van Rompaey et al.²¹ assessed the quality of sleep via five daily questions with dichotomous answers; after the first night, patients who used earplugs reported significantly better sleep (P = .042). This positive effect on the perception of sleep was maintained during two nights, although its intensity decreased on the third night. Finally, Obanor et al.¹⁴ observed that patients in the intervention group had a mean score of 64.5 in the RCSQ, compared with 47.3 in the control group (P = .0007), indicating a statistically significant improvement in the quality of sleep attributable to the interventions implemented.

Hospital stay and use of devices

Regarding the analysis of the association between the use of eye masks and earplugs and the duration of hospital stay, only 20% of the studies selected (n = 2) incorporated this variable into their design. Arttawejkul et al.¹⁷ reported similar results between groups, without statistically significant differences in duration of ICU stay (P = .572). For their part, when assessing this variable, Demoule et al.²⁰ found no significant differences in the duration of ICU stay or 90-day follow-up.

Tolerance of and adaptation to devices

According to the data reported by the clinical trials selected for this analysis, the main complication associated with the use of eye masks and earplugs among ICU patients was connected to tolerance, comfort and ease of use. Balci et al.¹⁶ managed to have patients in the intervention group maintain these aids for eight consecutive hours on three successive nights, allowing their removal only for brief interventions or communications, without specifying the exact duration of those removals. By contrast, Arttawejkul et al.¹⁷ established that aid removal could not exceed ten minutes, maintaining their use during eight hours each night. Similarly, Kiliç and Kav¹⁸ also set a limit of ten minutes for aid removal, which led to the exclusion of twelve (n = 12) participants due to their poor tolerance of these aids. Similarly, Demoule et al.²⁰ excluded from their analysis nine (n = 9) participants due to a fragmented use of earplugs during the night. By contrast, the study by Litton et al.¹⁹ approached this problem by indicating that viability of the use of the earplugs by these patients had been achieved, expressing that their comfort and ease of use made their use feasible in an ICU. Furthermore, Van Rompaey et al.²¹ did not report withdrawals linked to removal of earplugs among participants. Afshin et al.¹³ established that aid removal was carried out only to administer medication or facilitate communication with patients, without defining a specific period of removal. One positive aspect is that no withdrawals were reported in the intervention group. Finally, Obanor et al.¹⁴ reported that six out of the forty-four patients in the intervention group used only an eye mask.

Discussion

Although there is evidence that acting on environmental factors in an ICU can have a positive effect on quality of sleep and the incidence of delirium in patients, alterations of the sleep-wake cycle remain one of the main problems in this setting. Therefore, it is essential to promote the implementation of non-pharmacological measures focused on improving quality of sleep of critically ill patients, since the disruption of this cycle has a severe impact on the correct functioning of the body. Two systematic literature reviews support these interventions, reporting similar results in terms of delirium and quality of sleep.^22,23 Moreover, several studies conducted outside the scope of this review have concluded that the use of these devices contributes to better quality of sleep, mainly through mitigation of the hostile environment characteristic of an ICU.^24–26

The clinical trial conducted by Khalil et al. identified significant differences in the incidence of delirium following the use of earplugs and eye masks by patients in a neurocritical ICU in Egypt.²⁷ Furthermore, the meta-analysis conducted by Hu et al. also associated the use of these aids with significant reductions in the incidence of delirium.²⁸ By contrast, other studies, including those by Obanor et al.,¹⁴ Demoule et al.,²⁰ and Arttawejkul et al.,¹⁷ have not found statistically significant differences in relation to delirium, highlighting the need to continue investigating this question with larger samples and studies with greater homogeneity. One relevant aspect that could be contributing to the lack of robust results could be the use of different diagnostic tools for delirium. In Spain and in the Spanish-speaking world, the CAM-ICU scale is the most widely accepted and reproducible; however, as observed in the studies analysed in this review, only three of them employed this tool, and found no significant differences in any case.^14,17,20,29 Conversely, the studies that used the NEECHAM, Nu-DESC or ICDSC scales did report significant differences in the reduction of delirium.^13,16,18,21

Moreover, the study by Ely et al.³⁰ showed an increase in the duration of hospital stay and in-hospital mortality associated with delirium. In this cohort study, delirium was identified as an independent predictor of greater mortality at six months and prolonged hospital stay at the ICU, even when taking into account common variables in this setting, including sedation, state of coma and the use of analgesics in patients undergoing MV. However, this association was minimally approached in the clinical trials included in this systematic review, where Demoule et al.²⁰ and Arttawejkul et al.¹⁷ did not find statistically significant differences. It is important to note that the study by Arttawejkul et al. did not include sedated patients in its analysis.^17,20

The main complications observed in the use of eye masks and earplugs were their tolerance and comfort. As previously mentioned, withdrawal of patients during follow-up could occur due to poor tolerance, especially if the subjects were not previously accustomed to using such aids. The level of comfort and noise reduction efficacy varied between the different types of earplugs, and the insertion technique was also a determining factor to be taken into consideration. Previous studies in critically ill patients have employed multiple varieties of earplugs, and many of them did not include a detailed description of the insertion technique, maintenance and levels of noise reduction achieved.^24,31 For example, the pilot study by Litton et al. specifically described how the earplugs were inserted,¹⁹ whereas other studies, such as those by Balci et al., Afshin et al., Van Rompaey et al., Arttawejkul et al., and Demoule et al., did not approach aspects related to this matter.^{13,16,17,20,21} By contrast, Obanor et al., and Kiliç and Kav, reported on the levels of noise reduction achieved with the earplugs selected for their studies.^14,18 Therefore, future studies in this area should seek to specify these aspects, since success or failure clearly depends on the selection and correct use of these aids.

Limitations of the study

The main limitation of this systematic review lies in the difficulty in identifying high-quality randomised clinical trials which evaluated the use of eye masks and earplugs in adults admitted to an ICU. It is possible that some relevant studies were not identified, especially those published in databases not included in the search, like Scopus, and this could have limited the comprehensiveness of the search. The risk of bias was only evaluated through the CASPe tool, whereas the Cochrane Risk of Bias 2.0 (RoB2) tool, currently considered the recommended standard for randomised clinical trials, was not used. Another limitation is that the methodological work was conducted by a single researcher: current methodological standards recommend the participation of at least two independent reviewers to minimise the risk of bias in the selection of articles. The absence of screening by more independent reviewers represents a methodological limitation and could have influenced the final selection of studies. In addition, it is worth highlighting the cultural and contextual variability of the countries where the included studies were conducted, which could have influenced the definition and comparability of standard care. The scarce research work existing in this area, along with the heterogeneity of the available studies, underscores the need for financial investment in this field, in order to update and strengthen the available scientific evidence. Finally, it is worth indicating that the review was not previously registered in any international registry.

Conclusions

The studies analysed show that the use of eye masks and earplugs in ICU patients could have positive effects on quality of sleep and in reducing the incidence of delirium, especially when used in combination with standard care; nevertheless, the available evidence is limited and heterogeneous, so these findings should be interpreted with caution. Although some studies report significant benefits, others do not find statistically relevant differences, which indicates that more research is needed to better define the optimal conditions of use. The most widely used tools for the detection of delirium and the evaluation of quality of sleep were the CAM-ICU and PSQI, respectively. The results showed greater evidence of benefits on improving quality of sleep, whereas no significant effects were observed in relation to delirium employing this scale. No association was observed between the use of eye masks and earplugs and a decrease in the number of days of hospital stay. In addition, the tolerance and comfort of these aids varied between patients, and aspects such as the duration of their use and brief removals were important in order to ensure their acceptance. In general, these aids could constitute a good strategy to improve the wellbeing of patients in the ICU, always taking into consideration individual differences and possible difficulties in their use.

Ethical considerations

The elaboration of this systematic review clearly describes the methodology employed for the search of clinical trials. Selection biases have been avoided insofar as possible, so that the results obtained express the positive or negative effect of the use of eye masks and earplugs at intensive care units. The evaluation of the critical appraisal used the CASPe tool with the objective of reducing selection bias.

Funding

No financial support in the form of grants, scholarships or other sources of funding was received for the elaboration and publication of this work.