Effects of sensory stimulation on level of consciousness in comatose patients after traumatic brain injury: A systematic review

Abstract

BACKGROUND:

A coma is a prolonged unconscious state in which there is no response to various stimuli. In response, sensory stimulation was designed to stimulate brain plasticity and to promote brain regeneration. The effects of sensory stimulation intervention on comatose patients following traumatic brain injury (TBI) remain unclear.

OBJECTIVES:

This study aimed to examine the effects of sensory stimulation on the level of consciousness (LOC) after TBI and to identify the effective treatment dosage.

METHODS:

We searched PubMed, REHABDATA, EMBASE, CINAHL, MEDLINE, PEDro, SCOPUS, and Web of Science from inception to February 2020. Experimental studies investigating the influence of sensory stimulation on the LOC in the comatose patients (Glasgow coma scale < 8) following TBI were selected. The Physiotherapy Evidence Database scale (PEDro) was used to evaluate the methodological quality.

RESULTS:

Eleven studies met the inclusion criteria. Six were randomized controlled trials (RCTs), clinical controlled trials (CCTs) (n = 2), and pilot studies (n = 3). A total of 356 comatose patients (<8 on GCS) post-TBI were included in this study with sample sizes ranging from 5–90 patients. The sample sizes for the selected studies ranged from 5 to 90 patients. The scores on the PEDro scale ranged from three to eight, with a median score of seven. The multimodal sensory stimulation showed beneficial effects on the LOC in the comatose patients following TBI. The evidence for the effects of unimodal stimulation was limited, while the optimal treatment dosage remains unclear.

CONCLUSIONS:

The multimodal sensory stimulation intervention improves the LOC in patients with coma after TBI compared with unimodal stimulation. Further high-quality studies are needed to verify these findings.

Keywords

Brain injury traumatic level of consciousness physical stimulation sensation coma rehabilitation

1 Introduction

Traumatic brain injury (TBI) is a leading cause of disability and hospitalisation in intensive care units (ICU) worldwide [1]. About 17% of people who survive TBI have a period of complete unconsciousness or coma with no awareness of themselves or their surroundings [2]. A coma is a prolonged unconsciousness state in which an individual cannot be awakened, fails to respond regularly to painful stimuli, light, or sound, lacks a normal wake-sleep cycle, and does not initiate voluntary actions [3]. Altered loss of conciseness is associated with adverse effects on the normal functioning, long length of altered consciousness, functional dysfunction severity, slow recovery, and poor prognosis [3].

As people recover from TBI, they normally pass through several phases of recovery in which recovery can stop at any phase [4]. The transition between the phases is usually very gradual and highly distinctive depending on many factors, such as type and location of the injury, past medical history, age, and response to the treatment [4]. The long-term impacts of the TBI can influence the ability to participate in activities of daily living (ADLs) [5]. Acoma occurs due to disturbance in the function either of the brainstem reticular activating system (RAS) above the mid pons or of both cerebral hemispheres [6]. They may have a simple reflex in response to touch or pain, but they show no meaningful response to external stimuli [7].

The coma can last from hours to days, depending on the severity of brain damage [7]. Some people remain in a comatose state for months or even years [7]. A person in a coma may experience some improvement and transition to a vegetative state in which lower brain functions (e.g., sleep cycles, heart rate regulation) and some upper brain functions (e.g., eye-opening, sound production) are present [7, 8]. One of the most serious problems among the unconscious individuals is sensory deprivation because their ability to react to internal and external stimuli is altered [8]. It can cause different mental and perceptual problems and life-threatening conditions for patients hospitalised in the intensive care unit (ICU) [9]. The practical implication of the sensory deprivation is that controlled stimulation may meet the higher threshold of the reticular neurons and increase the cortical activity. The undamaged axons may send out collateral connections called the collateral sprouting that assist in reorganizing the brain activity [10].

The sensory stimulation can stimulate the process of post-TBI plasticity, promote brain regeneration, improve neurologic function [11], shorten the length of the ICU stay, and alleviate anxiety [12]. The multimodal stimulation (e.g., tactile and gustatory, auditory, and tactile) has been investigated concerning the wake cycle and purposeful movements [13]. A growing knowledge implies that the recovery processes are activated immediately after TBI [11]. One such recovery process is plasticity [11]. Enhancement of plasticity occurs through both endogenous factors such as the release of nerve growth factor, and the exogenous factors such as environmental or sensory stimulation [14].

Recently, one systematic review by Padilla and Domina (2016) examined the effectiveness of sensory stimulation on the arousal and alertness of individuals in a coma or vegetative state [14]. It was limited by the number of studies that were published between 2008 to 2013. Their results were not definitive because of the selected studies in their review had a small sample size and short intervention period [14]. Recently, there is an increase in the number of studies that investigate the influence of sensory stimulation in comatose patients following TBI. This review aimed to investigate the effect of sensory stimulation approaches on the LOC in comatose individuals after TB and, if possible, to identify the effective treatment dosage by studying and analyzing the existing evidence.

2 Methods

2.1 Search strategy and study selection

An electronic search was conducted in PubMed, REHABDATA, EMBASE, CINAHL, MEDLINE, PEDro, SCOPUS, and Web of Science from inception to February 2020. The keywords were (sensory stimulation OR stimulation OR acoustic stimulation OR physical stimulation) AND (brain injury OR trauma OR traumatic brain injury OR traumatic) AND (coma OR comatose) AND (level of consciousness OR consciousness OR Glasgow coma scale OR arousal) (Fig. 1).

Fig. 1

Summary of literature review process.

2.2 Eligibility criteria

The present review followed the PRISMA guidelines. And the PICOS approach (P: participants; I: intervention; C: comparison; O: Outcomes; S: study design) [15]. Studies were included if they (a) assessed comatose patients (<8 on Glasgow coma scale [GCS]) following TBI, (b) used sensory stimulation intervention, (c) classified as an experimental study published in English, (d) assessed level of consciousness, and (e) compared with active/passive control group or no control. Articles were excluded if they (a) assessed patients in coma or vegetative caused by non-traumatic disorders, (b) lack of outcome data, (c) descriptive studies, and (d) animal models.

2.3 Methodological quality

Two authors assessed the methodological quality for the included studies using the Physiotherapy Evidence Database scale (PEDro) scale. The PEDro scale provides an overview of a study of the internal and external validity. Furthermore, it assesses the capability of statistical analysis [16]. Four grades of the PEDro scale have been validated, while other grades have face validity [17]. Lastly, inter-rater reliability has been confirmed [16, 18]. The non-randomization, non-blinding of evaluators, no intention to examination, and no measurement of compliance were central threats to the biasing [19]. The methodological quality scores for the selected studies are presented in Table 1.

Table 1
Methodological quality scores

Reference Random allocation Concealed allocation Groups similar at baseline Participant blinding Therapist blinding Assessor blinding < 15% dropouts Intention to treat analysis Between-group differences reported Point estimate and variability reported Total (0 to 10)

Abbasi et al., 2009 [22] Y ? Y N N Y Y Y Y Y 7

Davis and Gimenez, 2003 [11] N N Y N N N Y ? N Y 3

Gorji et al., 2014 [23] Y N Y N N Y Y Y Y Y 7

Mitchell et al., 1990 [28] N N Y N N N Y N Y Y 4

Hall et al., 1992 [29] N N Y N N N Y Y N Y 4

Megha et al., 2013 [24] Y Y Y ? N Y Y Y Y Y 8

Moattari et al., 2016 [25] Y Y Y N N Y Y Y Y Y 8

Oh and Seo, 2003 [30] N N Y N N N Y N N Y 3

Park and Davis, 2016 [31] N N Y N N N Y Y N Y 3

Salmani et al., 2017 [26] Y N Y ? N Y Y Y Y Y 7

Urbenjaphol et al., 2009 [27] Y N Y ? N Y Y Y Y Y 7

Median: 7

Reference	Random allocation	Concealed allocation	Groups similar at baseline	Participant blinding	Therapist blinding	Assessor blinding	< 15% dropouts	Intention to treat analysis	Between-group differences reported	Point estimate and variability reported	Total (0 to 10)
Abbasi et al., 2009 [22]	Y	?	Y	N	N	Y	Y	Y	Y	Y	7
Davis and Gimenez, 2003 [11]	N	N	Y	N	N	N	Y	?	N	Y	3
Gorji et al., 2014 [23]	Y	N	Y	N	N	Y	Y	Y	Y	Y	7
Mitchell et al., 1990 [28]	N	N	Y	N	N	N	Y	N	Y	Y	4
Hall et al., 1992 [29]	N	N	Y	N	N	N	Y	Y	N	Y	4
Megha et al., 2013 [24]	Y	Y	Y	?	N	Y	Y	Y	Y	Y	8
Moattari et al., 2016 [25]	Y	Y	Y	N	N	Y	Y	Y	Y	Y	8
Oh and Seo, 2003 [30]	N	N	Y	N	N	N	Y	N	N	Y	3
Park and Davis, 2016 [31]	N	N	Y	N	N	N	Y	Y	N	Y	3
Salmani et al., 2017 [26]	Y	N	Y	?	N	Y	Y	Y	Y	Y	7
Urbenjaphol et al., 2009 [27]	Y	N	Y	?	N	Y	Y	Y	Y	Y	7
Median: 7

Y: low risk of bias; N: high risk of bias; ?: not clear.

2.4 Data extraction and analysis

The data were extracted by two authors. The data were extracted: (1) study design and participants, (2) treatment protocol, (3) experimental intervention, and (4) control group intervention (Table 2).

Table 2
Study characteristics

Author, year Study design and participants Treatment protocol Experimental intervention Control intervention

Abbasi et al., 2009 [22] Study design: RCTSample (n): 50Sex M/F: 43/7Age (mean): 30.4GCS admission: 6–8Status: comatoseLength of status: several days Session (n): 6Session duration: 15 minsSession frequency: 1 session per day×6 days Multimodal stimulation: tactile &auditory stimulation No intervention

Davis and Gimenez, 2003 [11] Study design: CCTSession duration: 5–15 minsSample (n): 12Sex M/F: 12/0Age (mean): 30GCS admission: 6Status: comatoseLength of status: 6.3 days Session (n): 35–56Session duration: 5–15 minsSession frequency: 5–8 sessions per day×7 days Unimodal stimulation: auditory stimulation Usual care

Gorji et al., 2014 [23] Study design: RCTSample (n): 30Sex M/F: 24/6Age (range): 15–44GCS admission: 5.97Status: comatoseLength of status: 0–6 months Session (n): 28Session duration: 10 minsSession frequency: 2 sessions per day×2 weeks Unimodal stimulation: auditory stimulation Natural voices of environment

Mitchell et al., 1990 [28] Study design: CCTSample (n): 24Sex M/F: 20/4Age (mean): 22.52GCS admission: 4–6Status: ComatoseLength of status: 4–12 days Session (n): 6–12Session duration: 60–120 minsSession frequency: 1-2 sessions×6 days Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation. No intervention

Hall et al., 1992 [29] Study design: PilotSample (n): 6Sex M/F: 5/1Age (mean): 42GCS admission: 3–7Status: ComatoseLength of status: 11–23 days Session (n): 28Session duration: 30 minutesSession frequency: daily×4 weeks Multimodal stimulation: visual, auditory, tactile, and gustatory stimulation. NA

Megha et al., 2013 [24] Study design: RCTSample (n): 30Sex M/F: 30/0Age (mean): 39.7GCS admission:< 8Status: ComaLength of status: 4–12 days Session (n): 28–70Session duration: 20–50 minutesSession frequency: Group A 5 sessions per day, group B &C 2 sessions per day×2 weeks Group A (high frequency group)Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation.5-times a day (20 minutes each session) for 2 weeksGroup B (low frequency group)Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation.Twice a day (50 minutes each session) for 2 weeks Group C (control group) conventional physiotherapy twice daily for 2 weeks.

Moattari et al., 2016 [25] Study design: RCTSample (n): 60Sex M/F: 50/10Age (mean): 36.97GCS admission:< 8Status: ComaLength of status: 3 days Session (n): 14Session duration: 30 minutesSession frequency: 2 sessions per day×7 days Group AMultimodal stimulation: visual, auditory, olfactory, and tactile stimulation delivered by nurse.Group BMultimodal stimulation: visual, auditory, olfactory, and tactile stimulation delivered by family Group C: Usual care

Oh and Seo, 2003 [30] Study design: PilotSample (n): 5Sex M/F: 5/0Age (mean): 50.2GCS admission: 5–10Status: ComaLength of status:< 3 months Session (n): 112 sessionsSession duration: -Session frequency: 2 sessions per day×8 weeks. Multimodal stimulation: auditory, visual, olfactory, gustatory, tactile and physical stimulation NA

Park and Davis, 2016 [31] Study design: PilotSample (n): 9Sex M/F: 9/0Age (mean): 29.7GCS admission:< 8Status: ComaLength of status: - Session (n): 15–40 sessionsSession duration: 30 minsSession frequency: 5–8 sessions per day×3–5 days. Direct auditory stimulationUnimodal stimulation: auditory stimulation by family or nurseNon-direct auditory stimuli Unimodal stimulation: auditory stimulation consisting of general music and TV sounds.

Salmani et al., 2017 [26] Study design: RCTSample (n): 90Sex M/F: 66/24Age (mean): 18–65GCS admission: (6.8) 5–8Status: ComaLength of status: - Session (n): 14Session duration: 30–45Session frequency: 2 sessions per day×7 days Group A (Family-centered affective stimulation group)Multimodal stimulation: auditory, kinetic, and tactile stimulationGroup B (Placebo group)Multimodal stimulation: auditory, kinetic, and tactile stimulation by trained person who was unfamiliar to all patients. Group CUsual care

Urbenjaphol et al., 2009 [27] Study design: RCTSample (n): 40Sex M/F: 28/12Age (mean): 34GCS admission: 3–8Status: ComaLength of status: 3 days Session (n): 70Session duration: 30 minutesSession frequency: 5 sessions per day (one stimulus each session)×2 weeks Multimodal stimulation: tactile, gustatory, olfactory, auditory, and visual stimulation Usual care

Author, year	Study design and participants	Treatment protocol	Experimental intervention	Control intervention
Abbasi et al., 2009 [22]	Study design: RCTSample (n): 50Sex M/F: 43/7Age (mean): 30.4GCS admission: 6–8Status: comatoseLength of status: several days	Session (n): 6Session duration: 15 minsSession frequency: 1 session per day×6 days	Multimodal stimulation: tactile &auditory stimulation	No intervention
Davis and Gimenez, 2003 [11]	Study design: CCTSession duration: 5–15 minsSample (n): 12Sex M/F: 12/0Age (mean): 30GCS admission: 6Status: comatoseLength of status: 6.3 days	Session (n): 35–56Session duration: 5–15 minsSession frequency: 5–8 sessions per day×7 days	Unimodal stimulation: auditory stimulation	Usual care
Gorji et al., 2014 [23]	Study design: RCTSample (n): 30Sex M/F: 24/6Age (range): 15–44GCS admission: 5.97Status: comatoseLength of status: 0–6 months	Session (n): 28Session duration: 10 minsSession frequency: 2 sessions per day×2 weeks	Unimodal stimulation: auditory stimulation	Natural voices of environment
Mitchell et al., 1990 [28]	Study design: CCTSample (n): 24Sex M/F: 20/4Age (mean): 22.52GCS admission: 4–6Status: ComatoseLength of status: 4–12 days	Session (n): 6–12Session duration: 60–120 minsSession frequency: 1-2 sessions×6 days	Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation.	No intervention
Hall et al., 1992 [29]	Study design: PilotSample (n): 6Sex M/F: 5/1Age (mean): 42GCS admission: 3–7Status: ComatoseLength of status: 11–23 days	Session (n): 28Session duration: 30 minutesSession frequency: daily×4 weeks	Multimodal stimulation: visual, auditory, tactile, and gustatory stimulation.	NA
Megha et al., 2013 [24]	Study design: RCTSample (n): 30Sex M/F: 30/0Age (mean): 39.7GCS admission:< 8Status: ComaLength of status: 4–12 days	Session (n): 28–70Session duration: 20–50 minutesSession frequency: Group A 5 sessions per day, group B &C 2 sessions per day×2 weeks	Group A (high frequency group)Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation.5-times a day (20 minutes each session) for 2 weeksGroup B (low frequency group)Multimodal stimulation: visual, auditory, olfactory, tactile and gustatory stimulation.Twice a day (50 minutes each session) for 2 weeks	Group C (control group) conventional physiotherapy twice daily for 2 weeks.
Moattari et al., 2016 [25]	Study design: RCTSample (n): 60Sex M/F: 50/10Age (mean): 36.97GCS admission:< 8Status: ComaLength of status: 3 days	Session (n): 14Session duration: 30 minutesSession frequency: 2 sessions per day×7 days	Group AMultimodal stimulation: visual, auditory, olfactory, and tactile stimulation delivered by nurse.Group BMultimodal stimulation: visual, auditory, olfactory, and tactile stimulation delivered by family	Group C: Usual care
Oh and Seo, 2003 [30]	Study design: PilotSample (n): 5Sex M/F: 5/0Age (mean): 50.2GCS admission: 5–10Status: ComaLength of status:< 3 months	Session (n): 112 sessionsSession duration: -Session frequency: 2 sessions per day×8 weeks.	Multimodal stimulation: auditory, visual, olfactory, gustatory, tactile and physical stimulation	NA
Park and Davis, 2016 [31]	Study design: PilotSample (n): 9Sex M/F: 9/0Age (mean): 29.7GCS admission:< 8Status: ComaLength of status: -	Session (n): 15–40 sessionsSession duration: 30 minsSession frequency: 5–8 sessions per day×3–5 days.	Direct auditory stimulationUnimodal stimulation: auditory stimulation by family or nurseNon-direct auditory stimuli Unimodal stimulation: auditory stimulation consisting of general music and TV sounds.
Salmani et al., 2017 [26]	Study design: RCTSample (n): 90Sex M/F: 66/24Age (mean): 18–65GCS admission: (6.8) 5–8Status: ComaLength of status: -	Session (n): 14Session duration: 30–45Session frequency: 2 sessions per day×7 days	Group A (Family-centered affective stimulation group)Multimodal stimulation: auditory, kinetic, and tactile stimulationGroup B (Placebo group)Multimodal stimulation: auditory, kinetic, and tactile stimulation by trained person who was unfamiliar to all patients.	Group CUsual care
Urbenjaphol et al., 2009 [27]	Study design: RCTSample (n): 40Sex M/F: 28/12Age (mean): 34GCS admission: 3–8Status: ComaLength of status: 3 days	Session (n): 70Session duration: 30 minutesSession frequency: 5 sessions per day (one stimulus each session)×2 weeks	Multimodal stimulation: tactile, gustatory, olfactory, auditory, and visual stimulation	Usual care

Table 3 presents the outcome measures details (GCS). The effect sizes were established by dividing the difference between the means of both groups by the pooled SD [20]. The effect size (ES) was determined using Cohen’s d >0.8 was large, 0.5 moderate, and <0.2 small [21].

Table 3

Outcome measures

Author, Year	Outcome measure	Assessment time	Experimental group (mean±SD)	Control group (mean±SD)	Reported effect	Effect size
Abbasi et al., 2009 [22]	GCS	After each treatment session	1st session: 7.0±0.8	1st session: 6.9±0.9	+	0.12
			2nd session: 7.8±0.9	2nd session: 6.8±0.9		1.11
			3rd session: 8.0±0.8	3rd session: 6.5±1.0		1.66
			4th session: 8.1±0.8	4th session: 6.3±1.2		1.76
			5th session: 8.6±8.0	5th session: 6.6±1.3		1.85
			6th session: 8.8±0.7	6th session: 6.8±1.4		1.81
Davis and Gimenez, 2003 [11]	GCS	Pre and post treatment intervention	Pre: 5.5±–	Pre: 6±–	=	–
			Post: 3.3±–	Post: 1.0±–		–
Gorji et al., 2014 [23]	GCS	Pre and post treatment intervention	Pre: 6.40±1.53	Pre: 5.35±5.53	=
			Post: NR	Post: NR		–
Mitchell et al., 1990 [28]	GCS	Week 1 (baseline), 2, 3, and 4 week	Week 1:5.16±0.58	Week 1:5.08±0.9	+	0.11
			Week 2:7.6±1.6	Week 2:6.4±1.7		0.73
			Week 3:11.4±1.7	Week 3:9±1.3		1.59
			Week 4:14±0.9	Week 4:13.08±1.5		0.74
Hall et al., 1992 [29]	GCS	Pre and post treatment intervention	Pre: 4.5±–	–	+	–
			Post: 13.5±–			–
Megha et al., 2013 [24]	GCS	Pre and post treatment intervention	Group A	Group C	Group A &C:+	1.26
			Pre: 3.40±0.52	Pre: 3.40±1.07	Group B &C:+	1.23
			Post: 6.80±1.48	Post: 4.90±1.60	Group A &B:++
			Group B
			Pre: 3.50±0.97			0.10
			Post: 6.50±1.65			0.98
Moattari et al., 2016 [25]	On admission and after each treatment session	Group A	Group C	Group A &C:=
			On admission: 6.10±1.41	On admission: 6.25±1.44	Group B &C:+	0.11
			1st session: 6.10±1.41	1st session: 6.25±1.44	Group A &B:+	0.11
			2nd session: 6.10±1.41	2nd session: 6.25±1.44		0.11
			3rd session: 6.25±1.44	3rd session: 6.50±1.50		0.17
			4th session: 6.55±1.57	4th session: 6.45±1.53		0.06
			5th session: 6.85±1.53	5th session: 6.55±1.79		0.18
			6th session: 7.15±1.63	6th session: 6.60±1.98		0.30
			7th session: 7.15±1.63	7th session: 6.70±1.97		0.25
			Group B
			On admission:5.75±1.02			0.40
			1st session: 5.75±1.02			0.40
			2nd session: 5.80±1.05			0.36
			3rd session: 6.00±1.29			0.36
			4th session: 6.90±1.51			0.30
			5th session: 7.65±1.92			0.59
			6th session: 8.85±2.20			1.10
			7th session: 9.20±2.16			1.21
Oh and Seo, 2003 [30]	GCS	Pre and post treatment intervention	Pre: 5±–	+	–
			Post: 9.8±–			–
Park and Davis, 2016 [31]	GCS	Pre and post treatment intervention	Direct auditory stimulation	+	–
			Pre: 5.88±1.83
			Post: 7.34±1.53
			Non-direct auditory stimulation
			Pre: 5.88±1.83
			Post: 7.07±1.48
Salmani et al., 2017 [26]	GCS	On admission, and after each treatment session	Group A	Group C	Group A &C:+
			On admission: 5.3±0.6	On admission: 5.3±0.8	Group B &C:=	0.00
			1st session: 5.3±0.6	1st session: 5.3±0.8	Group A &B:+	0.00
			2nd session: 5.4±0.8	2nd session: 5.3±0.8		0.13
			3rd session: 5.8±1.2	3rd session: 5.4±0.8		0.39
			4th session: 7±1.4	4th session: 5.7±1.1		1.03
			5th session: 7.7±1.9	5th session: 5.7±1.2		1.26
			6th session: 8.8±1.9	6th session: 5.9±1.3		1.78
			7th session: 9.1±2.1	7th session: 6.6±1.7		1.31
			Group B
			On admission: 5.1±0.3			0.33
			1st session: 5.1±0.3			0.33
			2nd session: 5.1±0.3			0.33
			3rd session: 5.2±0.5			0.30
			4th session: 5.9±0.6			0.23
			5th session: 6±0.9			0.28
			6th session: 6.9±1.1			0.83
			7th session: 7.2±1.1			0.42
Urbenjaphol et al., 2009 [27]	GCS	Pre and post treatment intervention	Pre: 4.50±1.15	Pre: 4.75±1.33	+	0.20
			Post: 10.45±1.82	Post: 5.9±1.77		2.53

GCS: Glassgow Coma Scale;+: Significant improvement in the experimental group.++: improvement in both groups/conditions without significant difference=: No significant difference between groups.

Strong evidence indicates consistent results from at least two RCTs. Moderate evidence is made based on one RCT, two or more studies with lower levels of evidence. Limited evidence is based on few studies, flaws in available studies, or some inconsistency in the findings across individual studies [21]. After reviewing the results of the included studies, the meta-analysis was not appropriate because of the significant heterogeneity of the treatment protocols.

3 Results

3.1 Study selection

A systematic search of databases produced a total of 879 articles. After eliminating the duplicates, 476 articles were reviewed. Out of those, 427 articles were excluded based on their abstracts Forty-nine full text studies were reviewedThirty-eight studies were removed for the following reasons: case studies/case series and include other neurological disorders. A total of 11 studies were included in the current review (Fig. 1).

3.2 Methodological quality

Six of the selected studies were RCTs [22 –27], CCTs (n = 2) [11, 28], and pilot studies (n = 3) [29 –31]. Five studies [11 , 28–31] did not perform to random allocation, whereas nine studies [11 , 26–31] failed to concealed allocation. Five studies failed to blind the assessors [11 , 28–31]. Three studies failed to initiation to treat analysis [11 , 30]. Finally, four studies failed to report between-group analysis [11 , 29–31]. Six studies were of high methodological quality (>6/10) [22 –27]. Whereas five were of low methodological quality (<6/10) [11 , 28–31]. (Table 1).

3.3 Outcome measures

The selected studies used the Glasgow Coma Scale (GCS) to assess the level of consciousness (LOC). The GCS is a reliable, objective way of recording the individuals’ consciousness level, derived from the eye-opening, verbal response, and motor response [32]. The lowest possible GCS (the sum) is 3 for the deep coma or death, while the highest is 15 for a fully awake person. The inter-rater reliability of the GCS has been reported to be in an acceptable data [33].

3.4 Description of included studies

3.4.1 Participants

A total of 356 comatose patients (<8 on GCS) post-TBI were included in this analysis. Sample sizes ranged considerably from 5 to 90 patients. The majority of patients in the included studies were male (82%), with the mean age (36.45 years old). Regarding the length of coma before the beginning of the intervention, seven studies included patient with <1-month coma status history [11 , 27–29]. The remaining studies did not provide details about the length of coma status [26, 31].

3.4.2 Intervention

Unimodal stimulation: Three studies used unimodal stimulation (i.e., auditory stimulation) to stimulate the LOC in comatose patients post-TBI [11 , 31] compared with usual care [11] and natural voices of environment [23] control groups.

Multimodal stimulation: Eight studies [23 –30] used multimodal stimulation to stimulate the LOC in the comatose patients post-TBI compared with no intervention [22, 28], conventional physiotherapy [24], and usual care [25 –27] control groups.

3.5 Effects of sensory stimulation on the level of consciousness in comatose patients after TBI

3.5.1 Unimodal stimulation

No significant change in the GCS scale scores was reported in the comatose patients following auditory stimulation intervention compared with usual care [11] and the natural voices of the environment [23] control groups. The study by Park and Davis (2016) reported an increase in the GCS scores after the direct and non-direct auditory stimulation conditions with no significant difference [31].

3.5.2 Multimodal stimulation

Eight studies [22 , 24–30] reported a significant increase in the GCS scores following multimodal auditory stimulation intervention compared with no intervention [22, 28], conventional physiotherapy [24], and usual care [25 –27] control groups. Two studies did not include control groups [29, 30]. The results were presented in Table 3.

3.6 Adverse effects or side effects

No adverse effects, side effects, or serious complications were reported after the sensory stimulation intervention in the included studies.

4 Discussion

The present systematic review aimed to examine the influences of sensory stimulation intervention on the LOC in the comatose patients following TBI and to define the optimal treatment dosage. The main finding showed strong evidence for the positive effects of multimodal sensory stimulation on the level of consciousness in comatose individuals post-TBI. We were unable to establish any benefits attributable to unimodal stimulation. Thus, the effects of unimodal stimulation are still limited. Padilla and Domina (2016) showed strong evidence of the impacts of multimodal stimulation on arousal and alertness of individuals in a coma or vegetative state after TBI [14]. As well, Karma and Rawat (2006) found that sensory stimulation is an effective intervention in improving the LOC in the comatose patients following non-traumatic brain injury [34]. Any effective stimulation affects the brain can activate the reticular activating system [34]. This results in increased sympathetic activity in different parts of the body, increased norepinephrine release in nerve terminals, and greater consciousness and arousal [35]. Moreover, it stimulates the nervous system development [36]. The human brain responds to kindly stimulation and behaviors by releasing hormones from the frontal cortex. These hormones increase the overall activity of the brain and improve the LOC [36]. Our results are consistent with the findings of Mitchell et al. (1990) [28], Kater (1989) [37], and Oh and Seo (2003) [30]. According to Kater (1989), the recovery can be explained by alteration such as axonal sprouting, rerouting, and anomalous connections to various regions of the brain and by the compensation mechanisms of the brain [37]. Although the GCS is a reliable outcome measure for recording individuals’ consciousness levels [32]; however, it can lead to misdiagnosing vegetative and minimal consciousness states. This misdiagnosis can lead to significant clinical, therapeutic, and ethical consequences [38].

The efficacy of multimodal sensory stimulation can be explained based on the concept of Perrin et al. (2006). He reported that although the resting metabolism of the comatose individuals is reduced, the auditory cortices (Brodman’s areas numbers 41 and 42) are still responding to the tone [39]. Similarly, Sullivan et al. (1998) showed that the olfactory stimulation raised blood flow to the frontal lobes and the orbitofrontal cortex. Thus, activating the brain regions vital for planning and judgment [40]. Keller et al. (2007) also confirmed that the various kinds of stimulation increase the autonomic nervous system (ANS) activity, which confirms the fact that they have an impact on the Reticular activating system [41].

The selected studies failed to blind the participants and therapists, leading to potential bias. While it is difficult to blind the participants and therapists to the intervention, it is possible to blind the assessor. Therefore, this is important to reduce the bias correlated with measurement based on outcomes [39]. In a total, six studies were of high methodological quality (>6/10) [22 –27]. Whereas five were of low methodological quality (<6/10) [11 , 28–31].

Three studies investigated the effects of unimodal stimulation (i.e., auditory) on the LOC in the comatose patients post-TBI [11 , 31]. Only one study by Park & Davis (2016) demonstrated an improvement in the LOC after both direct and non-direct auditory simulation conditions. This study was of low methodological quality (3/10) and had a small sample size (n = 9) [31]. So the meaningful effects cannot be established. As well, Gorji et al. (2014) and Davis & Gimenez (2003) explained that the lack of support for using the sensory stimulation is related to the small sample size [11, 23].

Eight studies demonstrated improvements in the LOC in the comatose patients post-TBI after multimodal stimulation intervention [22 , 24–30]. Three studies were of low quality on the PEDro scale (High risk of bias) (<6/10) and had a small sample size (<25) [28 –30]. The significant differences are not allowed to calculate with a small sample size [42]. So it is not possible to establish the clinical importance of the reported effects. The remaining five studies were of high methodological quality on the PEDro scale (>6/10) and had a large sample size [22 , 24–27]. The effect sizes were small to large (0.00–2.53). So the meaningful effects can be established.

Three studies administrated multimodal sensory intervention within several days post-coma [22 , 25], while two studies did not provide information about the time of beginning the intervention [26, 27]. As the most neurological recovery occurs early after brain injury [42]; thus, it seems that earlier intervention could provide more efficient improvement in the LOC in the comatose patients after TBI. In agreement with Padilla and Domina (2016), the multimodal sensory stimulation should be administrated early after the coma [14]. The recovery processes in the TBI are activated immediately following the injury [43, 44]. The plasticity recovery permits the brain to modify its organization and function [45]. Improvement of plasticity occurs through exogenous factors, such as environmental or sensory stimulation [14].

Because of the heterogeneity treatment protocols, identifying the optimal treatment dosages remain unclear. Moreover, to date, the long-term effects of the sensory stimulation on the LOC remain unclear. The effects of multimodal sensory stimulation on the LOC in comatose patients with non-traumatic injuries remain ambiguous. Moreover, many studies had small sample sizes and were of poor methodological quality. High-quality studies with large sample sizes and long-term follow-ups are warranted to determine the most effective treatment parameters. Established studies investigated the effects of sensory stimulation on the LOC on the non-traumatic comatose patients strongly needed.

There are several limitations to this systematic review. The search process was limited by studies published in the English language. This leading to bias since the research papers with significant findings are also more likely to be published in the English language [46]. Thus, reviewing only studies published in English could lead to an overestimation of treatment impacts [47]. The meta-analysis was not performed because of the heterogeneity of the treatment interventions between the selected studies. However, the effect sizes were calculated. Lastly, the effect size for some of the selected studies was not calculated because data were presented as a median instead of mean and standard deviation. This limited our ability to perform a meta-analysis.

5 Conclusion

The multimodal sensory stimulation intervention may improve the LOC in the comatose patients following TBI. The unimodal sensory stimulation appeared ineffective relative to multimodal stimulation. The optimal treatment dosage remains unclear. Further high quality and adequately powered randomized controlled trials are needed to verify our findings and hypothesis.

References

Grafman

, Salazar

. Traumatic Brain Injury. 1st ed. Edinburgh: Elsevier; 2015.

Whyte

, Nordenbo

, Kalmar

, Merges

, Bagiella

, Chang

, Yablon

, Cho

, Hammond

, Khademi

, Giacino

. Medical complications during inpatient rehabilitation among patients with traumatic disorders of consciousness. Archives of Physical Medicine and Rehabilitation. 2013;94:1877–83.

Iverson

, Gardner

, Terry

et al. Predictors of clinical recovery from concussion: a systematic review. Br J Sports Med. 2017;51:941–8.

Masel

, DeWitt

. Traumatic Brain Injury Disease: Long-Term Consequences Of Traumatic Brain Injury. New York: Oxford University Press; 2014. pp. 28–56.

Centers for disease control and prevention. Rates of TBI-related emergency department visits, hospitalizations, and deaths—United States, 2001–2010 | concussion | traumatic brain injury | CDC Injury Center. [online] Cdc.gov. Available at: https://www.cdc.gov/traumaticbraininjury/data/rates.html. Accessed January, 2020.

Greenberg

, Aminoff

, Simon

. Clinical Neurology. New York, N.Y.: McGraw Hill Medical; 2012.

White

, Giacino

. Disorders Of Consciousness. Arlington, VA: American Psychiatric Publishing; 2013. pp. 103–130.

Davis

, White

. Innovative sensory input for the comatose brain-injured patient. Critical Care Nursing Clinics of North America. 1995;7:351–61.

Gaugler

, Kane

, Newcomer

. J Am Geriatr Soc. 2005;53:2098–20105.

10.

Sosnowski

, Ustik

. Early intervention: Coma stimulation in the intensive care unit. Journal of Neuroscience Nursing. 1994;26:336–41.

11.

Davis

, Gimenez

. Cognitive-behavioral recovery in comatose patients following auditory sensory stimulation. Journal of Neuroscience Nursing. 2003;35:202–9.

12.

Hetland

, Lindquist

, Chlan

. The influence of music during mechanical ventilation and weaning from mechanical ventilation: A review. Heart & Lung. 2015;44:416–25.

13.

Lombardi

, Taricco

, De Tanti

, Telaro

, Liberati

. Sensory stimulation of brain-injured individuals in coma or vegetative state: results of a Cochrane systematic review. Clinical Rehabilitation. 2002;16:464–72.

14.

Padilla

, Domina

. Effectiveness of sensory stimulation to improve arousal and alertness of people in a coma or persistent vegetative state after traumatic brain injury: A systematic review. American Journal of Occupational Therapy. 2016;70:1–8.

15.

Maher

, Sherrington

, Herbert

, Moseley

, Elkins

. Reliability of the PEDro Scale for Rating Quality of Randomized Controlled Trials. Physical Therapy.. 2003;83:713–21.

16.

Moher

. Assessing The Quality Of Reports Of Randomised Trials. Alton: core research on behalf of NCCHTA; 1990.

17.

Foley

, Bhogal

, Teasell

, Bureau

, Speechley

. Estimates of quality and reliability with the physiotherapy evidence-based database scale to assess the methodology of randomized controlled trials of pharmacological and nonpharmacological interventions. Physical Therapy. 2006;86:817–824.

18.

Ghai

, Driller

, Ghai

. Effects of joint stabilizers on proprioception and stability: A systematic review and meta-analysis. Physical Therapy in Sport. 2017;25:65–75.

19.

Ialongo

. Understanding the effect size and its measures. Biochemia Medica. 2016;26:150–63.

20.

Cohen

. Statistical Power Analysis For The Behavioural Sciences. Hillside, NJ: Earlbaum Associates; 1988. pp. 1.

21.

Abbasi

, Mohammadi

, Sheaykh Rezayi

. Effect of a regular family visiting program as an affective, auditory, and tactile stimulation on the consciousness level of comatose patients with a head injury. Japan Journal of Nursing Science. 2009;6:21–6.

22.

Gorji

, Araghiyansc

, Jafari

, Morad

. Effect of auditory stimulation on traumatic coma duration in intensive care unit of medical sciences university of mazandarn, Iran. Saudi Journal of Anaesthesia. 2014;8:69–72.

23.

Megha

, Harpreet

, Nayeem

. Effect of frequency of multimodal coma stimulation on the consciousness levels of traumatic brain injury comatose patients. Brain Injury. 2013;27:570–7.

24.

Moattari

, Alizadeh Shirazi

, Sharifi

, Zareh

. Effects of a sensory stimulation by nurses and families on level of cognitive function, and basic cognitive sensory recovery of comatose patients with severe traumatic brain injury: A randomized control trial. Trauma Monthly. 2016;21:523–31.

25.

Salmani

, Mohammadi

, Rezvani

, Kazemnezhad

. The effects of family-centered affective stimulation on brain-injured comatose patients’ level of consciousness: A randomized controlled trial. International Journal of Nursing Studies. 2017;74:44–52.

26.

Urbenjaphol

, Jitpanya

, Khaoropthum

. Effects of the sensory stimulation program on recovery in unconscious patients with traumatic brain injury. Journal of Neuroscience Nursing. 2009;41:10–6.

27.

Mitchell

, Bradley

, Welch

, Britton

. Coma arousal procedure: A therapeutic intervention in the treatment of head injury. Brain Injury. 1990;4:273–9.

28.

Hall

, Macdonald

, Young

. The effectiveness of directed multisensory stimulation versus non-directed stimulation in comatose CHI patients: Pilot study of a single subject design. Brain Injury. 1992;6:435–45.

29.

, Seo

. Sensory stimulation programme to improve recovery in comatose patients. Journal of Clinical Nursing. 2003;12:394–404.

30.

Park

, Davis

. Effectiveness of direct and non-direct auditory stimulation on coma arousal after traumatic brain injury. International Journal of Nursing Practice. 2016;22(4):391–6.

31.

Teasdale

, Jennett

. Assessment of coma and impaired consciousness. The Lancet. 1974;304:81–4.

32.

Segatore

, Way

. The Glasgow Coma Scale: time for change. Heart Lung. 1992;21:548–57.

33.

Karma

, Rawat

. Effect of stimulation in coma. Indian Pediatr. 2006;43:856–60.

34.

Khurana

, Khrana

. Textbook of medical physiology. 2nd ed. India: Elsevier india; 2006.

35.

Eysenck

, Flanagan

. Psychology For A2 Level. Hove: Psychology Press; 2001.

36.

Kater

. Response of head-injured patients to sensory stimulation. Western Journal of Nursing Research. 1989;11:20–33.

37.

Schnakers

, Giacino

, Kalmar

, et al.

Does the Glasgow Coma Scale correctly diagnose the vegetative and minimally conscious states?

Critical Care. 2007;11:488.

38.

Perrin

, Schnakers

, Schabus

, Degueldre

, Goldman

, Bre'dart

, Faymonville

, Lamy

, Moonen

, Luxen

, et al. Brain response to one’s own name in vegetative state, minimally conscious state and locked-in syndrome. Archives of Neurology. 2006;63:562–9.

39.

Sullivan

, Warm

, Schefft

, Dember

, O’Dell

, Peterson

. Effects of olfactory stimulation on the vigilance performance of individuals with brain injury. Journal of Clinical and Experimental Neuropsychology. 1998;20:227–36.

40.

Keller

, Hulsdunk

, Muller

. The influence of acoustic and tactile stimulation on vegetative parameters and EEG in persistent vegetative state. Functional Neurology. 2007;22:159–63.

41.

Skilbeck

, Wade

, Hewer

, Wood

. Recovery after stroke. Journal of Neurology, Neurosurgery & Psychiatry. 1983;46:5–8.

42.

Duff

. Review article: Altered states of consciousness, theories of recovery, and assessment following a severe traumatic brain injury. Axone. 2001;23:18–23.

43.

Lannin

, Cusick

, McLachlan

, Allaous

. Observed Recovery Sequence in Neurobehavioral Function After Severe Traumatic Brain Injury. American Journal of Occupational Therapy. 2013;67(5):543–9.

44.

DeFina

, Fellus

, Polito

, Thompson

, Moser

, DeLuca

. The new neuroscience frontier: Promoting neuroplasticity and brain repair in Traumatic Brain Injury. Clin Neuropsychol. 2009;23(8):1391–9.

45.

Portney

, Watkins

. Foundations Of Clinical Research. 3rd ed. Upper Saddle River: Pearson Prentice Hall;2009.

46.

Egger

, Smith

. Meta-analysis bias in location and selection of studies. BMJ. 1998;316:61–6.