The effect of music on pain perception in women scheduled for elective cesarean section: a systematic review and meta-analysis

Objective: To study the effect of music on pain perception in women scheduled for elective cesarean section (CS) Strategy: We used the following keywords (“music” or “music therapy” and any of the following: cesarean section OR cesarean delivery OR CS OR cesarean OR Caesarean OR "post-op*) Selection Criteria: We included all studies satisfying the following criteria: (1) Population: pregnant women scheduled for cesarean section. (2) Intervention: the addition of any type of music to routine care compared with routine care alone. (3) Study design: randomized controlled


Introduction
Delivery by cesarean section (CS) has become increasingly common and considered a life-saving procedure for both the mother and the infant in certain medical indications. 1 Previous evidence suggests that CS rates of 10-15% could be associated with reduced both maternal and perinatal morbidity and mortality rates. 2,3 Recent studies show that CS rates have dramatically increased throughout the 21 st century particularly in developed countries. 4,5 Some studies have addressed the effect of anxiety on birth outcomes and identified a negative impact on the duration of the operation and even selection of the type of anaesthesia, [6][7][8][9][10] which in turn highlighted the need for psychological interventions to overcome this anxiety. 10 Another major concern of CS is postoperative pain which is more severe when compared to normal vaginal delivery. 11 In a questionnaire study of 220 patients who underwent a CS, 46.8% of the participants complained of immediate postoperative pain. 12 Besides being an irritant, postoperative pain could possibly interrupt with breastfeeding after the operation. 13,14 More recently, studies have investigated the efficacy of non-clinical interventions to deal with the negative physiological and psychological aspects of surgical procedures which may reduce the need for pharmacological interventions. 15,16 One of the widely investigated interventions is music. 17 The impact of music on the negative aspects of surgical procedures including pain and pre-and postoperative anxiety has been thoroughly addressed over the last few decades. 18,19 Recent studies have identified a link between music and cesarean sections and shown a possible impact of a music intervention before, during or after surgery on peri-operative anxiety and postoperative pain. [20][21][22][23][24][25][26][27] The purpose of this systematic review and meta-analysis is to investigate the effect of music with routine care versus routine care alone on postoperative pain and anxiety in women undergoing elective CS and also the impact of music on other physiological measures including: heart rate (HR), systolic blood pressure (SBP) and diastolic blood pressure (DBP).

Materials and Methods
We followed the PRISMA statement guidelines 28 during the preparation of this systematic review and metaanalysis and performed all steps in accordance with the Cochrane handbook of systematic reviews of intervention. 29 No informed consents or ethical committee approval were needed as this is a systematic review of previously published studies

Literature search strategy
We conducted an electronic search of databases including PubMed, Web of Science, SCOPUS, and Cochrane CENTRAL, using the following keywords ("music" or "music therapy" and any of the following: cesarean section OR cesarean delivery OR CS OR cesarean OR Caesarean OR "post-op*). All published articles were considered, and only English articles were selected. We searched the bibliography of included studies for additional relevant records.

Eligibility criteria and study selection
We included all studies satisfying the following criteria

Population:
Pregnant women scheduled for planned cesarean sections.

2.
Intervention: The addition of any type of music to routine care compared with routine care alone before or during CS.

Outcomes:
The main outcome measures were pain intensity, anxiety during and after CS, blood pressure, and heart rate.
We excluded the following 1. non-randomized trials

in vitro and animal studies
3. studies whose data were unreliable for extraction and analysis Two authors independently removed duplicates. Three authors retrieved references and performed the screening in two steps; the first step was to screen titles/abstracts for matching our inclusion criteria and the second step was to screen the full-text articles of eligible abstracts for eligibility to metaanalysis.

Data extraction
Data extraction was independently performed using a standardized form. Data included first author, year of publication, post-operative pain, heart rate, systolic blood pressure, diastolic blood pressure, anxiety score, and morphine usage post-operatively. Two investigators independently scored the studies and collected the information. In case of discrepancies in scoring between the two investigators, a consensus was reached after the involvement of the senior investigator.

Risk of bias assessment
To assess the risk of bias in the retrieved clinical trials, we utilized the Cochrane risk of bias assessment tool, provided in Chapter 8.5 of the Cochrane Handbook of Systematic Reviews of Interventions 5.1.0. 30 Risk of bias assessment included the following domains: sequence generation (selection bias), allocation sequence concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias) and other potential sources of bias, the authors' judgment is categorized as 'Low risk', 'High risk' or 'Unclear risk' of bias. Any discrepancies between the two assessors were resolved through discussion. We utilized the quality assessment table provided in Chapter 8.5 part 2 of the same book. 30

Data synthesis
Data were extracted from assessed articles and entered into RevMan software (Review Manager, version 5.3, The Cochrane Collaboration, 2011; The Nordic Cochrane Centre, Copenhagen, Denmark) for meta-analysis. Afterward, the weighted mean difference was calculated. Statistical heterogeneity between studies was assessed by Isquare (I 2 ) statistics, and values of ≥50% were indicative of high heterogeneity. When heterogeneity was significant, a random-effects model was used for meta-analysis. Fixed effects model of meta-analysis was used when there was no significant heterogeneity.

Results of literature search
Our search strategy resulted in 542 studies. After title and abstract screening, 202 articles were available for full-text screening. We excluded 196 of them, and finally, eight studies matched our inclusion criteria and were included in the final analysis. [20][21][22][23][24][25][26][27] The PRISMA flow diagram for study selection is shown in Figure 1. A total of eight RCTs met our inclusion criteria, which included 818 women. The summary of the included studies is presented in Table 1, and their main results are presented in Table 2.
The following were the sites of the included studies: One study in Israel, 20 one study in China, 22 one study in Turkey, 23 one study in Taiwan, 25 one study in Germany, 27 and three studies in Iran. 21   Women who listened to music before a cesarean section had a significant increase in positive emotions and a significant decline in negative emotions and perceived threat of the situation when compared with women in the control group, who exhibited a decline in positive emotions, an increase in the perceived threat of the situation, and had no change in negative emotions. Women who listened to music also exhibited a significant reduction in systolic blood pressure compared with a significant increase in diastolic blood pressure and respiratory rate in the control group.

Reza 2007 21
100 women scheduled for elective cesarean section There was not statistically significant difference in VAS for pain between two groups up to six hours postoperatively (P>0.05). In addition, morphine requirements were not different between two groups at different time intervals up to six hours postoperatively (P>0.05). There was not a statistically significant difference between two groups regarding postoperative anxiety score and vomiting frequency (P>0.05).

Li 2012 22
60 women undergoing elective cesarean section In the study group the mean HRV, as measured by the low frequency power (LF) value and the LF to high frequency power (LF/HF) ratio during Holter assessment, was significantly less after the music intervention but was not significantly changed in the control group. Moreover, the mean HF value was significantly increased, and the mean anxiety score was significantly decreased after the music intervention but not in the control group. Finally, the mean pain score obtained 6 hours after surgery was significantly lower in the study than in the control group Handan 2017 23 60 women undergoing elective cesarean section.
The Visual Analogue Scale (VAS) scores before and during the procedure showed significantly lower scores for the experimental group, compared to the control group (p<0.05). Music therapy reduces the physiological and cognitive responses of anxiety in patients undergoing multiple cesarean section, and can be used in the clinical practice Ebneshahidi 2008 24 80 Women who underwent cesarean section.
Pain score and postoperative cumulative opioid consumption were significantly lower among patients in the music group (p < 0.05), while there were no group differences in terms of anxiety score, blood pressure, or heart rate (p < 0.05).

Chang 2005 25
64 women who were planning to have a cesarean delivery No significant differences were found between the two groups in any of the physiological indexes. This controlled study provides evidence that music therapy can reduce anxiety and create a more satisfying experience for women undergoing cesarean delivery. Norouzi 2013 26 90 Women who underwent repeat cesarean section.
No significant difference in the overall mean scores of maternal state anxiety score (MSA) between the groups at 6 hours after CS, but the severity of MSA in the experimental groups was less than in the control group (P= 0.02).
At skin suture, significantly lower anxiety levels were reported in the intervention group regarding State anxiety (31.56 vs. 34.41; p = .004) and visual analogue scale for anxiety (1.27 vs. 1.76; p = .018). Two hours after surgery, the measured visual analogue scale for anxiety score in the intervention group was still significantly lower (0.69 vs. 1.04; p = .018).

Risk of bias assessment
We used Cochrane's risk of bias assessment tool and found an overall moderate risk of bias. All included studies reported adequate outcome reporting, proper selective reporting, and no missing data, therefore categorized as low risk of bias. Handan et al. 23 lacked randomization of patients; therefore, it was categorized as high risk. Due to the nature of the intervention (music), blinding of participants could not be achieved in our included studies. Therefore, they were categorized as high risk. Other improperly addressed domains were categorized as an unclear risk ( Figure  2).

Discussion
The present review showed that music significantly reduces postoperative pain after elective CS and helps in the reduction of anxiety in patients. Systolic and diastolic blood pressure (SBP & DBP) together with HR was not affected by music.
Three studies reported postoperative pain scores. Reza et al. 21 found no significant difference while the other two studies 22,24 found a significant decrease in the music group. The net analysis was highly significant in favor of the music group (p=0.005). As for the SBP, DBP, and heart rate outcomes; four studies reported these outcomes. 20,[23][24][25] Two studies found a significant decrease in SBP in the music group, 20,25 and the other two did not find any difference. 23,24 The net analysis showed no significant effect of music on SBP. Only one study found a significant decrease in DBP and HR in the music group, 20 while the other three did not. [23][24][25] As for the anxiety score, seven studies reported a minor non-significant effect of music in reducing anxiety. [21][22][23][24][25][26][27] When combined into a single analysis, the results favored the music group significantly (p=0.04).
These findings are supported by other studies in the literature. Regarding the pain score, Nilsson et al. 31 performed a trial dividing their patients into three groups, two groups listened to music during CS and a control group. Pooled results showed a significant favor of the music groups over the control one. (p=0.001) in terms of pain scores. Good et al. 32 found similar results in their RCT after gynecological surgery, a total of 311 patients were included in the trial which were divided into three groups, the first received normal care, the second was to listen to music, and the third group was given other relaxation techniques. The interventions were delivered once for two days, with each lasting for a quarter an hour. Pooled results showed a significant reduction of pain scores in the music group over the control (p=0.001). Moreover, Laurion et al. 33 carried out a pilot study about the effect of music listening during gynaecological laparoscopy and found a significant decrease in pain scores in the music group (p=0.002).
Regarding other associated outcomes: Nilsson et al. 31 measured post-operative fatigue and nausea scores with selfmeasured 5-degrees, and 4-degrees scores respectively. Results showed that music was not effective in preventing postoperative nausea, however, the music group experienced significantly less fatigue (p=0.001). Hook et al. 35 measured anxiety scores in their study using the Visual Analogue Scale for Anxiety (VASA) and the State-Trait Anxiety Inventory (STAI). Results showed that the music groups had less incidence of anxiety than control groups (p=0.001).

Strengths and Limitations
Including only RCTs gives our study some strength. Our study is limited by the small sample size of the included studies with the total number of participants of 818. Furthermore, heterogeneity seems to be high in the heart rate outcome (I 2 =78%). However, in our defence, this may be normal due to the nature of the study and the intervention. In addition, the risk of bias in the included studies was moderate according to the Cochrane's risk of bias assessment tool. Additionally, cultural differences may be an important confounder in the analysis of the results.
Despite the major differences among the eight studies in evaluation of pain, postoperative painkiller protocols, use of VAS and determination of anxiety, we plotted only data which can be aggregated. They are mixed and the protocol bias is not negligible. The meta-analysis could be improved by separating the expected benefits of music at the different periods (pre-intraand post-operative); this would add some interesting and practical information. However, this was not possible because of the different methods of the included eight studies.

Conclusion
Our study has identified a clinically and statistically significant impact of music on pain scores and decreased anxiety levels in women scheduled for elective CS. Based on our results, we would recommend playing music before, during, and after CS.