The benefits of perioperative music interventions for patients undergoing neurosurgery: a mixed-methods systematic review

Abstract

Background

Several studies have demonstrated music intervention as a safe and inexpensive option for reducing anxiety and pain perioperatively. We performed a systematic review to evaluate its possible benefit in perioperative neurosurgical management.

Methods

The following databases were searched with no restrictions on publication date: PubMed, Embase, CINAHL, PsycINFO, CENTRAL and ClinicalTrials.gov. All studies that quantitatively or qualitatively assessed the effects of any music-based intervention administered within one week of cranial or spinal neurosurgery were eligible. Meta-analysis could not be performed, and quantitative findings were instead summarised narratively. We then synthesised qualitative observations through thematic analysis before conducting an integrative review. PROSPERO registration: CRD42019152626.

Results

Seven studies, of sample sizes ranging from 8 to 60, were included, with the timing, duration, frequency and type of music intervention varying considerably across studies. Quantitative analysis involving five studies showed a significant decrease in anxiety following music intervention in four studies (p < 0.05), and reduced pain perception in three studies (p < 0.05). Mixed evidence was obtained for physiological outcomes including heart rate and blood pressure. Risk of bias was moderate to high. Quantitative findings were generally supported by qualitative analysis which provided additional insight into the factors influencing music intervention’s effectiveness.

Conclusion

Despite the heterogeneity in study characteristics, this review, which is the first mixed-methods systematic review assessing the effects of perioperative music intervention, not only reveals a potential role for music intervention in neurosurgery, but also highlights the possible importance of considering qualitative evidence in future studies to better characterise its effectiveness.

Keywords:

• Anxiety
• music
• neurosurgery
• pain
• systematic review

Introduction

Published evidence for perioperative music intervention arose in 1809, when Jane Todd Crawford sang hymns while undergoing the first known ovariotomy, without the benefit of anaesthesia.¹ Since then, the possible benefits of perioperative music intervention have been widely investigated, demonstrating increasing evidence for its effectiveness in reducing perioperative anxiety, pain and stress in patients.^2–6

When administered under appropriate conditions, music represents a safe and inexpensive option that not only circumvents the side effects associated with pharmacological intervention,^7–9 but also reduces analgesic use and drug interaction considerations.^10–15 The exact underpinnings of music’s beneficial effects remain difficult to establish, but appear to involve psychological as well as physiological influences, creating a peaceful and comfortable experience for the patient, and inducing relaxation by acting on both the autonomic nervous system and limbic system, respectively.⁴ Music may thus be an effective adjunct to anxiolytics such as midazolam, with the added benefit of potentially lowering perioperative indicators of stress including heart rate and blood pressure.^5,6

As with other forms of surgery, neurosurgery can impact patients both physically and emotionally, with certain procedures known to elicit particularly high levels of anxiety across not only patients but also family members and medical staff. Awake craniotomy, for instance, provokes considerable anxiety and fear in patients perioperatively,^16–19 which can interrupt or delay the surgical procedure, consuming time and medical staff resources.²⁰ Craniotomy patients also experience significant pain after surgery, which may increase length of stay and negatively influence postoperative morbidity and mortality.^21,22

Despite growing evidence being demonstrated for the effectiveness of music intervention in relieving anxiety across various surgical specialties, its possible benefit in neurosurgery has not been systematically reviewed to date. Additionally, previous systematic reviews have primarily focussed on quantitative evidence, which fail to adequately capture important aspects of music interventions. The aim of this mixed-methods systematic review is therefore to assess both quantitative and qualitative evidence for the effects of music intervention in cranial and spinal neurosurgery, to explore the benefit of introducing this concept in perioperative neurosurgical management.

Methods

This study was performed according to the PRISMA statement²³ and is registered with The International Prospective Register of Systematic Reviews (PROSPERO) as record number CRD42019152626.

Search strategy

We searched the following databases with no restrictions on publication date: PubMed, Embase, CINAHL, PsycINFO. Search terms included ‘music’, ‘neurosurgery’, ‘brain surgery’, ‘back surgery’, ‘spine surgery’, ‘spinal surgery’, ‘cranial’ and ‘craniotomy’ (Full search strategy available from Appendix A). All titles and abstracts were transferred to Endnote for duplicates to be removed before screening. We also searched for clinical trials in the Cochrane Central Register of Controlled Trials (CENTRAL) and ClinicalTrials.gov using a combination of the previous search terms and screened the results for eligible trials. These were then matched to the corresponding publication on PubMed to identify additional studies. Following this, we reviewed the lists of included studies as well as reference lists of all major systematic reviews on music intervention and surgery for eligible studies. Finally, we screened the reference lists of all eligible studies for potential citations.

Study selection

All studies which involved pre-, per- or post-operative music-based interventions within one week of either cranial or spinal neurosurgery were considered for eligibility. No restrictions were applied to the type of music intervention. Inclusion criteria were the following: Randomised controlled trials (RCTs), quasi-randomised controlled trials (QCTs), non-randomised trials, cohort studies, case–controls for which full-texts are published in a peer-reviewed journal in the English language. Given that this is a mixed-methods systematic review, quantitative, qualitative and mixed-methods studies were all eligible. Exclusion criteria consisted of: Patient population including children less than one year old and patients with hearing impairment. Single case reports, editorials, ideas and opinions were also excluded. Screening of titles and abstracts was carried out independently by two authors (KCNKK and CXK). For studies meeting the inclusion criteria, full articles were retrieved and independently checked for eligibility by KCNKK and CXK. In case of disagreement, the final decision was taken after discussion with the third author (CK).

Data extraction

Data extraction was performed independently by two authors (KCNKK and CXK). Study characteristics were extracted as follows: author, year of publication, study design (quantitative, qualitative or mixed methods, prospective or retrospective), patient population (population size, mean age and sex ratio), type of surgery (cranial or spinal neurosurgery), type of music intervention (recorded or live music therapy), choice of music (patient choice or not), music genre, timing of music intervention (before, during or after neurosurgery), duration and frequency of music intervention, outcomes (anxiety, pain, physiological outcomes including heart rate, systolic and diastolic blood pressures, and respiratory rate, length of hospitalisation and other quality of life measures), outcome scale and results. Findings from qualitative studies including defined themes and subthemes as well as available quotes were also extracted. Inconsistencies in extracted data were discussed with the third author (CK) to reach a consensus.

Quality appraisal

The Mixed-Methods Appraisal Tool (MMAT), which is a scoring system used by mixed-methods systematic reviews for assessing the quality of included studies, was used to evaluate the methodological rigour of qualitative studies, mixed-methods studies, and quantitative studies including RCTs and QCTs.^24–26 Each study is subject to two screening questions and an additional four questions relating to the methodological quality of the study design in question. These questions are then answered by options including ‘Yes’, ‘No’ and ‘Can’t tell’. The risk of bias of RCTs and QCTs was further evaluated using the Cochrane risk of bias tool²⁷ to include the following: random sequence generation (selection bias), allocation concealment (selection bias), selective reporting (reporting bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), and other potential sources of bias. Quality appraisal and risk of bias analysis were carried out by a single author (KCNKK). The results of quality appraisal and risk of bias analysis are available from Appendices B and C, respectively.

Data synthesis

We first described the background characteristics and intervention characteristics for all studies, after which data synthesis was performed according to a framework developed by Thomas and Harden^28,29 in which quantitative synthesis and qualitative synthesis are conducted separately, followed by an integrative review to summarise findings. Narrative synthesis was performed to summarise effectiveness results for the different outcomes including anxiety, pain, heart rate, blood pressure, respiratory rate, length of stay and other quality of life measures. Our qualitative analysis, which aimed to assess the effectiveness and appropriateness of the music-based intervention, involved first identifying all the different themes and subthemes relevant to the extracted quotes and the authors' reports and conclusions, followed by discussions amongst all three authors in order to determine any common themes and ideas. Results from both the quantitative synthesis and qualitative synthesis were then juxtaposed in the form of a matrix to identify any common findings as well as any discrepancies, before drawing a final set of conclusions.

Results

Database searching produced a total of 3392 results which reduced to 2878 results after excluding duplicates. After screening all 2878 results for title and abstract, only eight eligible studies remained (PRISMA Flow Diagram available from Appendix D). One eligible study involved music therapy administered alongside other interventions such as acupuncture and pet therapy and was therefore excluded. Finally, seven studies were included in our data synthesis and these consisted of three RCTs, one QCT, two qualitative studies and one mixed-methods study.

Background characteristics

Population size ranged from 8 to 60 patients, with mean age generally ranging from 40 to 60 years old. Two studies included only adolescent participants.^30,31 The study cohort consisted of patients undergoing cranial neurosurgery in three studies,^20,32,33 two of which involved only awake craniotomy patients.^20,32 The remaining four studies involved spinal surgery patients.^30,31,34,35 None of the control groups across quantitative studies received music intervention with the exception of one study where the intervention group viewed a 12-minute training video prior to receiving live music therapy while the control group received music therapy without watching the video.³¹ Further information regarding the background characteristics of included studies are provided in Table 1.

Table 1. Background characteristics of included studies.

Author/s (Year of publication)	Study design	Population size	Control group	Age (years)	Sex ratio (% male)	Type of surgery
Lin et al. (2011)	QCT	60	30	Mean = 62.2	51.7%	Spinal
Nelson, Adamek & Kleiber (2017)	RCT	41	22	Median (intervention) = 14 Median (control) = 14	9.8%	Spinal (Spinal fusion)
Walworth et al. (2008)	RCT	27	13	Mean = 46.5	44.4%	Cranial
Wu et al. (2017)	RCT	38	19	Mean = 40	65.8%	Cranial (Awake craniotomy)
Mondanaro et al. (2017)	Mixed-methods study (RCT & Prospective qualitative)	60	30	Mean (intervention) = 48.6 Mean (control) = 47.8	41.7%	Spinal (Spinal fusion)
Jadavji-Mithani, Venkatraghavan & Bernstein (2015)	Qualitative (Prospective)	29	–	Mean = 52	51.7%	Cranial (Awake craniotomy)
Kleiber & Adamek (2013)	Qualitative (Retrospective)	8	–	Mean = 14.4	12.5%	Spinal (Spinal fusion)

Intervention characteristics

All included studies involved short-term music intervention administered within three days of neurosurgery. Timing, duration and frequency of music intervention were variable. All studies except one³² involved postoperative music intervention while preoperative music intervention was administered in three studies.^20,33,34 Two studies involved more than one music session.^33,34 Duration of music intervention generally ranged from around 20 to 45 minutes per session, with music being played throughout the duration of awake craniotomy in two studies.^20,32 The type of music intervention also varied, and involved either recorded music being played via earphones or loudspeaker,^20,32,34 or live music being played by a music therapist.^30,31,33,35 These studies allowed the participant to choose their preferred music as part of the intervention and further involved techniques such as relaxation, tension release, breathing control and guided imagery. Studies involving recorded music usually involved a compiled music selection, with music genre varying greatly from pop music to sacred music. Further information regarding the intervention characteristics of included studies are provided in Table 2.

Table 2. Intervention characteristics of included studies.

Author/s (Year of publication)	Type of music intervention	Patient choice	Music genre	Timing of intervention			Duration and frequency of music
				Pre- op	Intra- op	Post- op
Lin et al. (2011)	Recorded music (earphones)	Patient choice from music selection	Pop, classical, nature & sacred	✓		✓	30-minute sessions 7pm the day before surgery, 1 hour before surgery and 3pm on the first and second days after surgery
Nelson, Adamek & Kleiber (2017)	Live music (music therapist)	Patient choice	–			✓	Single 30-minute session first time patient out of bed and seated in a chair on the second day after surgery
Walworth et al. (2008)	Live music (music therapist)	Patient choice	–	✓		✓	20–30-minute session before surgery, and 30-minute sessions each day spent in hospital after surgery
Wu et al. (2017)	Recorded music (loudspeaker)	Patient choice from music selection	Nature, piano, harp, orchestra, jazz & synthesis		✓		Single session for the duration of awake craniotomy
Mondanaro et al. (2017)	Live music (music therapist)	Patient choice	–			✓	Single 30-minute session within 72 hours after surgery
Jadavji-Mithani, Venkatraghavan & Bernstein (2015)	Recorded music (earphones)	Pre-determined music selection	Instrumental	✓	✓	✓	Single session starting 20 minutes immediately before surgery and ending 20 minutes after surgery (Patient had the option to stop the music)
Kleiber & Adamek (2013)	Live music (music therapist)	Patient choice	–			✓	Single 45-minute session first time patient out of bed and seated in a chair, usually on the second day after surgery

Quantitative analysis

Five studies were eligible for quantitative analysis. Anxiety and pain were the two most commonly measured outcomes and were respectively assessed in five and four studies via various rating scales such as the Visual Analogue Scale (VAS) and the State-Trait Anxiety Inventory (STAI). Measured outcomes also included quality of life measures such as relaxation, mood and stress. Two studies assessed physiological outcomes including heart rate, respiratory rate and blood pressure (systolic and diastolic),^32,34 with one study also investigating the length of hospitalisation.³³ Outcomes were generally measured pre- and post-music intervention. Estimation of treatment effect was conducted differently across studies, involving comparison between pre- and post- music intervention scores as well as between control and experimental group values, with variable statistical methods being employed.

Significantly decreased anxiety following music intervention was noted in the experimental group in all studies except in one study.³⁵ Although reduced anxiety was generally observed in the experimental group compared to the control group, the difference reached statistical significance in only two studies, with equivocal findings being reported in one study assessing anxiety via two different rating scales.³⁴ One study did not publish results for the control group³³ while another study administered music therapy to both control and experimental groups.³¹ Despite three studies reporting a decrease in pain following music therapy in the experimental group^31,33,35, the decrease was only significant in one study.³¹ No data on this were available in the other two studies.^32,34 Music intervention produced significantly decreased pain in the experimental group compared to the control group in two studies.^34,35 Differences in pain perception between control and experimental groups were either not reported or could not be assessed in the remaining studies.

Mixed evidence was obtained regarding physiological parameters which were assessed in two studies.^32,34 Music intervention during awake craniotomy significantly decreased heart rate, respiratory rate, systolic and diastolic blood pressures in the experimental group compared to the control group,³² although this was not replicated in another study involving spinal surgery patients.³⁴ No evidence for decreased length of hospitalisation following music therapy was found.³³ Findings regarding other quality of life measures such as depression and stress and relaxation were also equivocal.^33,35 Further information about the outcomes measured by the different studies, the time of measurement and relevant statistics for pre-test and post-test scores for both the control and experimental groups are provided in Table 3.

Table 3. Summary of findings from included studies.

Author/s (Year of publication)	Time of measurement	Outcomes measured		Mean (SD)				Pre- versus Post- test (Experimental group)		Control versus Experimental
				Control group		Experimental group		Statistical test	p value	Statistical test	p value
				Pre-test	Post-test	Pre-test	Post-test
Lin et al. (2011)	Before and after each music intervention session (Except STAI which was measured only on the day before surgery and the second day after surgery)	Anxiety (VAS)	S1		5.1 (2.7)		1.3 (2.1)			Independent t-test	<0.001
			S2		4.9 (2.8)		2.0 (2.1)				<0.001
			S3		2.7 (2.4)		1.4 (1.6)				0.018
			S4		2.1 (1.9)		0.8 (1.3)				0.002
		Anxiety (STAI)	S1		43.4 (11.2)		38.1 (11.5)				0.074
			S4		40.6 (8.2)		38.2 (8.9)				0.286
		Pain (VAS)	S1		6.0 (2.5)		2.3 (2.3)				<0.001
			S2		4.7 (2.1)		1.8 (2.2)				<0.001
			S3		5.2 (1.7)		3.0 (2.3)				<0.001
			S4		4.4 (1.9)		1.7 (1.5)				<0.001
		Diastolic BP (mm Hg)	S1		73.7 (13.7)		76.6 (14.3)				0.426
			S2		76.3 (10)		77.7 (13.2)				0.661
			S3		67.0 (12.3)		65.1 (11.2)				0.548
			S4		65.0 (8.5)		66.9 (9.9)				0.430
		Systolic BP (mm Hg)	S1		135 (22.9)		128 (19.3)				0.210
			S2		136.2 (21)		133.2 (23)				0.596
			S3		122.6 (16.5)		119.4 (21.1)				0.520
			S4		120.8 (15.1)		118.0 (18.5)				0.534
		Mean BP (mm Hg)	S1		93.2 (14.9)		94.2 (16.1)				0.804
			S2		95.2 (12)		95.8 (16.4)				0.886
			S3		85.5 (13)		83.0 (12.8)				0.456
			S4		83.2 (9.9)		83.3 (10.7)				0.970
		Heart rate (beats per minute)	S1		69.8 (11.6)		70.6 (8)				0.767
			S2		70.3 (12.9)		66.7 (10.7)				0.253
			S3		84.7 (14.5)		88 (16.4)				0.412
			S4		82.2 (12)		83.1 (13.1)				0.782
Mondanaro et al. (2017)	Before and after music intervention	Anxiety (HADS)		9.50 (3.64)	10.90 (3.14)	8.53 (3.78)	9.57 (4.30)			RMANOVA*	0.62
		Pain (VAS)		5.20 (2.82)	5.87 (2.58)	6.20 (2.48)	5.09 (2.49)				0.01
		Depression (HADS)		7.83 (3.92)	7.27 (3.53)	7.87 (4.26)	7.47 (4.95)				0.85
		Kinesiophobia (TSK)		31.57 (5.49)	30.53 (5.95)	30.70 (7.34)	31.10 (7.68)				0.93
Nelson, Adamek & Kleiber (2017)	Before and after music intervention	Anxiety (NRS)		6.1 (2.6)	3.7 (2.9)	6.0 (3.0)	3.5 (2.7)	Paired t-test	0.001	Independent t-test	0.521†
		Pain (NRS)		7.6 (1.8)	5.4 (2.5)	7.7 (2.0)	5.1 (2.7)		0.001		0.855†
Walworth et al. (2008)	Before and after each music intervention session	Anxiety (VAS)				33.8‡	24.2‡	Wilcoxon Signed-Rank test	0.03
		Pain (VAS)				30.4‡	22‡		>0.05
		Mood (VAS)				25.9‡	20.9‡		>0.05
		Perception of hospitalisation (VAS)				28.5‡	22.6‡		0.03
		Relaxation (VAS)				39.9‡	21.7‡		0.001
		Stress (VAS)				38.4‡	21.3‡		0.001
		Length of hospitalisation (hours)								Independent t-test	>0.05
Wu et al. (2017)	Before and after music intervention	Anxiety (STAI)		88.58 (6.64)	92.63 (8.76)	92.47 (9.34)	81.05 (10.17)	Wilcoxon Signed-Rank Test	<0.001	Mann-Whitney U test	<0.001
		Diastolic BP (mm Hg)		81.74 (8.10)	81.74 (7.53)	80.00 (6.47)	69.63 (7.07)		<0.001		<0.001
		Systolic BP (mm Hg)		148.95 (14.89)	146.53 (11.77)	131.79 (13.54)	124.37 (12.76)		<0.041		<0.001
		Heart rate (beats per minute)		80.11 (12.15)	81.84 (10.13)	77.79 (10.83)	62.84 (7.33)		<0.001		0.004
		Respiratory rate (breaths per minute)		17.53 (2.12)	17.42 (2.34)	16.58 (2.91)	12.11 (2.77)		<0.035		0.003

VAS: Visual Analogue Scale.

STAI: State-Trait Anxiety Inventory.

NRS: Numeric Rating Scale.

HADS: Hospital Anxiety and Depressions Scale.

TSK: Tampa Scale of Kinesiophobia.

BP: Blood pressure.

SD: Standard deviation.

RMANOVA: Repeated measures analysis of variance.

S₁: First session.

S₂: Second session.

S₃: Third session.

S₄: Fourth session.

*Group (music vs control) × Time (before vs after intervention) interaction.

†Both the control and experimental groups received music therapy in this study (The control group received music therapy only while the experimental group received both music therapy and a training session).

‡In this study, data were scaled such that decreasing numbers represent an improvement in the outcome.

Qualitative analysis

Three studies were included in our qualitative analysis. These consisted of two qualitative studies and one mixed-methods study.^20,30,35 Background characteristics and intervention characteristics are described in Tables 1 and 2, respectively. A semi-structured interview format was used by all three studies, with two studies being of a prospective study design^20,35 and one study being retrospective in nature.³⁰ One study interviewed patients via telephone 2–24 months after the music intervention,³⁰ while the other two studies interviewed patients face-to-face immediately³⁵ or around two weeks after their intervention.²⁰ Thematic analysis resulted in the identification of three key themes: Perception of music intervention, choice of music and preferences with regard to administration of music intervention. The results of our qualitative analysis are described in Appendix E.

Integrating quantitative and qualitative findings

We juxtaposed the findings from quantitative and qualitative analysis to determine their concordance and also to identify possible discrepancies regarding the possible benefits of music intervention in a perioperative setting.

Overall observation from qualitative studies that music intervention helped the participants to relax and feel more at ease was concordant with our quantitative analysis, which supported an anxiety-reducing component to music intervention. Music intervention’s benefit appeared to be more pronounced in awake craniotomy, as suggested by the largely positive response of patients towards intra-operative music intervention.²⁰ This agreed with our quantitative findings, which demonstrated a highly significant reduction in anxiety (p < 0.001), backed by a decrease in various physiological outcomes including heart rate, respiratory rate, systolic and diastolic blood pressures.³² While music intervention could potentially interfere with communication during awake craniotomy,²⁰ it may reduce anxiety by distracting the patient from the procedure and by shielding them from the surgical team’s conversation. Our quantitative synthesis, which showed a general trend towards reduced pain perception, was also consistent with the participants’ description that music diverted their attention from the painful experience.^20,30

Several accounts from our qualitative analysis provide potential insight as to why music intervention did not prove beneficial across studies. Some patients for instance believed music intervention reduced their time spent with loved ones prior to surgery.²⁰ Others simply found a quiet atmosphere more calming than the presence of a music therapist, with post-operative headache also reducing the effectiveness of music following surgery.²⁰ Various aspects of music intervention, such as the volume, choice of music and form of administration, as well as age of participants are likely to influence music’s effectiveness, although we were unable to find robust associations given the small number of studies.^20,30 Finally, other measures described by our qualitative synthesis, such as informing participants beforehand about the benefits of music intervention, may also be relevant.

Discussion

In our review, music intervention was associated with significantly reduced anxiety in most studies, with a trend towards decreased pain also being observed. Quantitative findings were generally consistent with our qualitative analysis which revealed multiple instances where music intervention helped to alleviate anxiety and pain in participants. Music intervention was, however, not always positively viewed by participants, who cited reasons such as reduced time spent with their loved ones and post-operative headaches. Mixed evidence was obtained for other quality of life measures such as depression, mood and stress. A significant decrease in physiological parameters was only apparent in one study involving awake craniotomy patients.

Study results could have been influenced by various factors, including the demographical make-up of studies, which included participants from different countries and with different cultural backgrounds. Other potentially important factors to consider are the gender composition, with male subjects making up only 9.8% of participants in one study,³¹ as well as the age distribution of study cohorts, which consisted of only adolescents in two studies.^30,31 Rating scales, which varied considerably across studies, may also determine study results. Indeed, in one study involving Chinese participants with a mean age of over 60, a decrease in VAS score was noted despite no significant reduction in STAI score.³⁴ Based on the study population, it has been suggested that the VAS measure would have been more appropriate given its accessibility compared to the STAI scoring system.

While overall evidence from our study supports the role of music interventions in reducing anxiety and pain, several limitations require the findings to be interpreted with caution. Firstly, we were only able to identify a small number of eligible studies, which could be due to study results not always being published, or the challenges being associated with patient recruitment. The latter is evidenced by the small sample sizes of included studies, which make our findings more difficult to generalise. Secondly, considerable heterogeneity was apparent across studies, which differed with respect to the type of surgery, the timing, frequency and duration of intervention as well as its administration. Measured outcomes were also highly variable, with certain outcomes not being consistently reported. Interpretation of study results and statistical methodology were additional sources of heterogeneity across studies. We were therefore unable to perform a meta-analysis, and instead summarised our results narratively. It was also not possible to establish how factors such as the timing and nature of music intervention influence its effectiveness.

As with other systematic reviews evaluating the effectiveness music intervention in perioperative management, the risk of bias of studies included in our quantitative analysis was moderate to high.^3,4,36 While one study was a QCT, most included RCTs failed to explicitly describe the randomisation process or the method of allocation concealment. Blinding of outcome assessment was also described by very few studies. Furthermore, blinding of participants was not possible given the intervention in question.

Despite our limited sample size and the considerable heterogeneity observed across studies, our study findings remain consistent with those of other systematic reviews having evaluated the benefits of music intervention in perioperative management.^2,4,36 This suggests a potential role for music intervention in neurosurgery, with patients undergoing procedures such as awake craniotomy appearing to particularly benefit from it. Our study, which to our knowledge, is the first mixed-methods systematic review on music intervention in a perioperative setting, also offers additional insight into the various factors influencing its effectiveness, highlighting the potential importance of considering qualitative evidence in music intervention studies. Given the strong subjective component associated with music intervention, we believe that quantitative assessment fails to adequately capture music intervention’s benefits, which should instead be evaluated via a mixed-methods approach.

Conclusion

Overall, this study suggests a potential role for music intervention in perioperative neurosurgical management, particularly for relieving anxiety and pain. It thus contributes to the growing evidence supporting the benefit of music in a perioperative setting, while additionally demonstrating that qualitative evidence can provide important insight into understanding the factors influencing the effectiveness of music intervention. Although existing evidence appears to support its benefits, future studies should be built on robust study designs and attempt to better characterise its effectiveness in order to facilitate its introduction to routine clinical practice.

Abbreviations
HADS	=	Hospital Anxiety and Depression Scale
MMAT	=	Mixed-Methods Appraisal Tool
NRS	=	Numeric Rating Scale
PRISMA	=	Preferred Reporting Items for Systematic Reviews and Meta-Analyses
PROSPERO	=	International Prospective Register of Systematic Reviews
QCT	=	Quasi-Randomised Controlled Trial
RCT	=	Randomised Controlled Trial
RMANOVA	=	Repeated measures analysis of variance
STAI	=	State-Trait Anxiety Inventory
TSK	=	Tampa Scale of Kinesiophobia
VAS	=	Visual Analogue Scale

Acknowledgements

The authors thank Ruth Jenkins and Marshall Dozier, academic librarians at the University of Edinburgh, for their advice on performing the literature search.

Disclosure statement

The authors report no conflict of interest.

Appendix A: Search strategy

 

PubMed

(music OR musical OR musicotherap*) AND (neurosurgery OR neurosurger* OR neurosurgic* OR “brain surgery” OR “brain operation” OR cranial OR craniotomy OR “brain tumor” OR “brain tumour” OR spine OR spinal OR “back surgery” OR “back operation”)

Embase OvidSP

((music OR musical OR musicotherap*) AND (neurosurgery OR neurosurger* OR neurosurgic* OR brain surgery OR brain operation OR cranial OR craniotomy OR brain tumor OR brain tumour OR spine OR spinal OR back surgery OR back operation)).ti,ab,kw.

Cinahl

(MH music OR MH "music therapy" OR (music OR musical OR musicotherap*)) AND (MH neurosurgery OR (neurosurger* OR neurosurgic* OR “brain surgery” OR “brain operation” OR cranial OR craniotomy OR “brain tumor” OR “brain tumour” OR spine OR spinal OR “back surgery” OR “back operation”))

PsycINFO OvidSP

((music or music therapy).mh. or (music or musical or musicotherap*).ti. or (music or musical or musicotherap*).ab.) AND (neurosurgery.mh. or (neurosurger* or neurosurgic* or brain surgery or brain operation or cranial or craniotomy or brain tumor or brain tumour or spine or spinal or back surgery or back operation).af.)

Cochrane Central

(music OR musical OR musicotherap*) AND (neurosurgery OR neurosurger* OR neurosurgic* OR brain surgery OR brain operation OR cranial OR craniotomy OR brain tumor OR brain tumour OR spine OR spinal OR back surgery OR back operation)

Clinicaltrials.gov

Condition or disease: —

Other terms: music

Study type: All Studies

Study Results: All Studies

Appendix B: Quality appraisal

 

 

Table

Author/s (Year of publication)	Types of mixed methods study components or primary studies	Methodological quality criteria	Responses
			Yes	No	Can’t tell	Comments
Jadavji-Mithani, Venkatraghavan and Bernstein (2015)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Qualitative	Are the sources of qualitative data relevant to address the research question?	✓
		Is the process for analysing qualitative data relevant to address the research question?	✓
		Is appropriate consideration given to how findings relate to the context in which the data were collected?	✓
		Is appropriate consideration given to how findings relate to researchers’ influence?			✓
Kleiber and Adamek (2013)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Qualitative	Are the sources of qualitative data relevant to address the research question?	✓
		Is the process for analysing qualitative data relevant to address the research question?	✓
		Is appropriate consideration given to how findings relate to the context in which the data were collected?	✓
		Is appropriate consideration given to how findings relate to researchers’ influence?			✓
Lin et al. (2011)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Quantitative randomized controlled trial	Is there a clear description of the randomization (or an appropriate sequence generation)?	✓
		Is there a clear description of the allocation concealment (or blinding when applicable)?	✓
		Are there complete outcome data (80% or above)?	✓
		Is there low withdrawal/drop-out (below 20%)?	✓
Mondanaro et al. (2017)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Quantitative randomized controlled trial	Is there a clear description of the randomization (or an appropriate sequence generation)?	✓
		Is there a clear description of the allocation concealment (or blinding when applicable)?		✓
		Are there complete outcome data (80% or above)?	✓
		Is there low withdrawal/drop-out (below 20%)?	✓
	Qualitative	Are the sources of qualitative data relevant to address the research question?	✓
		Is the process for analysing qualitative data relevant to address the research question?	✓
		Is appropriate consideration given to how findings relate to the context in which the data were collected?			✓
		Is appropriate consideration given to how findings relate to researchers’ influence?			✓
Nelson, Adamek and Kleiber (2017)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Quantitative randomized controlled trial	Is there a clear description of the randomization (or an appropriate sequence generation)?		✓
		Is there a clear description of the allocation concealment (or blinding when applicable)?		✓
		Are there complete outcome data (80% or above)?	✓
		Is there low withdrawal/drop-out (below 20%)?	✓
Walworth et al. (2008)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Quantitative randomized controlled trial	Is there a clear description of the randomization (or an appropriate sequence generation)?		✓
		Is there a clear description of the allocation concealment (or blinding when applicable)?		✓
		Are there complete outcome data (80% or above)?	✓
		Is there low withdrawal/drop-out (below 20%)?			✓
Wu et al. (2017)	Screening questions	Are there clear qualitative and quantitative research questions, or a clear mixed methods question?	✓
		Do the collected data allow address the research question?	✓
	Quantitative randomized controlled trial	Is there a clear description of the randomization (or an appropriate sequence generation)?		✓
		Is there a clear description of the allocation concealment (or blinding when applicable)?		✓
		Are there complete outcome data (80% or above)?	✓
		Is there low withdrawal/drop-out (below 20%)?	✓

Appendix C: Risk of bias analysis

 

Appendix D: PRISMA flow diagram

 

Appendix E: Qualitative analysis

 

Perception of music intervention

Exploring patient perception revealed two sub-themes: the participants’ views on music therapy’s effectiveness in relieving anxiety and pain, as well as their thoughts on whether music intervention should be administered.

Several participants in one study reported that live music helped them to relax and calm down emotionally. The adolescents further reported that music therapy enabled them to perceive pain less strongly, adding that music helped to divert their attention from the painful experience. One 17-year-old participant for instance explained: “With the music, it relaxes you with the pain, and it just, like to be very honest, I didn’t feel it.” One adolescent however mentioned that live music could be excessively loud and make the pain worse.

In another study, some of the patients admitted experiencing reduced anxiety as the recorded music helped to distract them from the nervousness surrounding awake craniotomy. The participants described that music made them feel gradually more relaxed as opposed to having a more immediate effect. There were also two patients who mentioned that music helped them to overcome nausea on the operating table. Most patients reported that music did not necessarily trigger happy or sad emotions, although some of them described that the music intervention brought back memories of them playing or learning music. Participants in another study also mentioned playing musical instruments in their childhood or using music to relax in their daily life, which could possibly have contributed to music intervention’s impact on them.

In awake craniotomy, the majority of patients found music to be most helpful during the actual surgical procedure. This was partly because music shielded them from the conversations of the medical team, which patients were afraid of hearing during the awake craniotomy. Some participants however admitted that the music was not as beneficial when they felt impatient during the surgical procedure because the music had lost its calming effect.

Despite the positive views expressed by most patients with regard to music intervention during awake surgery, some patients did not want music playing before their operation. One reason for this was that they preferred to spend this time with their relatives and doctors. A few patients also did not appreciate music intervention following surgery due to post-operative headaches. In such cases, participants did not find music comforting and instead preferred a quiet environment. Nevertheless, some patients found both preoperative and postoperative music beneficial as it helped to reduce anxiety and stress before and after the operation. Overall, patients in this study were highly positive towards the music intervention with some even stating that music should always be coupled to awake surgery rather than being used only for research purposes.

Choice of music

Many of the adolescent participants in one study appreciated the option of choosing, according to their personal preference, what music was to be played by the live music therapist. These participants also enjoyed the modern assortment of music which was available to be performed by their therapist. There were however some adolescents who mentioned that having their own music player would have allowed them to listen to music at their own leisure in the hospital. One adolescent further stated finding live music therapy too loud and described: “I think if I had my iPod with me that might have helped a little because I could have controlled the sound a little more.”

Another study involving adults also reported that about half of the participants would have preferred listening to their own music. These patients stated that having this option would have been more effective at reducing anxiety with regard to their awake craniotomy, with one participant mentioning that he found the music “boring” and that his own music would have been a better source of distraction. There were also some patients in this study who did not want their own music used. Some felt that the music they chose would not have been as effective as the music that was played to them during the study by the researchers, while others were worried of associating their music with the unpleasant experience of undergoing awake surgery.

Choice of key signature was also investigated by one study whereby participants were asked to listen to major and minor key music during the interview and were subsequently asked about their views on anxiety or stress perception. Although most patients had neutral feelings towards the recorded music when it was played to them during the awake surgery, an interview two weeks after the surgery revealed that some patients found the lively nature of major key music more helpful at reducing perioperative anxiety and stress. These patients added that the minor key music on the contrary sounded sad and melancholic and was therefore unsuitable for the purpose of reducing anxiety and stress. Conversely, other patients in the same interview preferred minor key music, finding it more calming due to its slower tempo.

Preferences regarding administration of music intervention

Participants in one study were also given the opportunity, several months after their music therapy, to express their thoughts as to how they preferred music therapy to be administered. In that study, the live music therapist regularly interacted with the participants during the session, which was positively viewed by some adolescents. One 14-year-old participant for instance was quoted as follows: “I remember when she came in she would ask me questions on how it made me feel and how it made me relax and stuff.”

The study also noted that many participants would have preferred to be fully informed about the benefits of music therapy before the surgery so that they were better prepared to take advantage of it. Three adolescents however believed that simply informing participants about music therapy would not be enough, and instead, more effective ways of introducing them to music therapy would be through visual means or by transmitting the information through a peer.