Use of music as an adjunct to sedation in patients undergoing transoesophageal echocardiography

Abstract

Aims

Transoesophageal echocardiography (TEE) is widely used for cardiac assessment and requires moderate sedation due to its semi-invasive nature, carrying potential risks. Music is a non-pharmacological strategy shown to reduce anxiety in various clinical settings. We prospectively investigated whether music could reduce anxiety and/or sedation dosage in patients undergoing TEE.

Methods and results

Sixty-three patients (>18 years) referred for TEE were randomized into a music group (n = 31) or control group (n = 32). All completed an anxiety questionnaire and had haemodynamic parameters (blood pressure, heart rate, oxygen saturation) assessed prior to transthoracic echocardiography (TTE). Music was played before TTE and continued throughout TEE in the intervention group. Anxiety and haemodynamic measurements were repeated after TTE, before sedation. Groups were compared for baseline characteristics, anxiety levels, haemodynamic changes, and midazolam dosage. There was no difference between the groups regarding age or sex; control group had more comorbidities. Anxiety levels did not differ between groups at baseline or pre-sedation (P > 0.05). A significant reduction over time was observed only in the music group (P = 0.032 vs. P = 0.069), but group-by-time interaction was not significant (P = 0.347). Regarding anxiety level, it decreased in both groups, more markedly in the music group (P < 0.008 vs. P = 0.04). Diastolic blood pressure also decreased after music exposure (P = 0.024). Midazolam dosage did not differ between groups but was inversely correlated with age.

Conclusion

Although music did not reduce sedation requirements, it significantly decreased anxiety levels and blood pressure; music may be a valuable adjunct to sedation in patients undergoing TEE.

Introduction

Transoesophageal echocardiography (TEE) is widely used as a diagnostic tool in cardiology. Due to the proximity of the probe to the oesophagus, TEE allows us to semi-invasively evaluate various structural cardiac abnormalities, providing greater diagnostic accuracy.^1,2 However, since it requires oesophageal insertion of the probe, it can cause discomfort in the oropharynx and anxiety related to potential breathing difficulties; patients may also experience anxiety due to the unfamiliar nature of the examination and concerns about the potential diagnosis or results.

To enhance patient comfort, mild to moderate (conscious) sedation is used before the procedure, typically using a benzodiazepine (midazolam) in combination with an opioid (Fentanyl).^3,4 Yet, it is noteworthy that the examination carries risks,⁵ mostly related to the use of sedation (hypoventilation, hypoxaemia, and hypotension, among others).^6,7 Therefore, alternatives that could reduce the dose of sedatives would be welcome.

Music is a non-pharmacological, low-cost and easy-to-apply means that has been used for the relief of anxiety or pain in various clinical or surgical situations,^8,9 including both preoperative and post-operative intra-hospital procedures, and ambulatory procedures.¹⁰ Despite being a low-complexity procedure, TEE can cause some degree of anxiety, often due to unfamiliarity with the procedure, fear regarding the diagnostic result or the need for sedation/anaesthesia. The combination of these issues can lead to increased anxiety, which in turn, can result in cardiovascular changes such as an increase in heart rate and blood pressure. Music could possibly help the patient to feel more comfortable and relaxed before TEE.

Objective

The primary objective of this study was to assess whether the presence of music could potentially reduce the degree of anxiety and/or the amount of anaesthetic administered during TEE. The secondary objective was to determine whether variables related to the exam and/or the patient could influence this outcome.

Methodology

Population

We prospectively evaluated all ambulatory or hospitalized patients over 18 years of age, of both sexes, referred for TEE at the Echocardiography Laboratory of Hospital Israelita Albert Einstein in São Paulo, Brazil, from January 2023 to January 2024. All patients provided written informed consent, and the study was approved by the hospital's Ethics Committee under the number CAAE 64738222.7.0000.0071.

Exclusion criteria

Patients were excluded if they met any of the following criteria: significant hearing impairment, contraindication for TEE, reduced cognitive function (such as dementia), or the need for sedation under anaesthesiologist supervision.

Study design

After meeting the inclusion criteria and signing the informed consent form, patients were randomly assigned to either control or intervention (Music) group. Both groups completed a brief questionnaire (Anxiety mini questionnaire) with faces corresponding to a scale of anxiety levels (Figure 1). The anxiety questionnaire was administered at two points: prior to the transthoracic echocardiogram (TTE) and immediately after this examination, before sedation was administered for TEE. Of note, TTE is performed routinely prior to and to complement TEE in our service. The anxiety mini questionnaire (adapted from Cao¹¹), consisted of a simple visual analogue scale for self-assessment to quickly evaluate patient's anxiety level, ranging from 1 (no anxiety), 2 (mild anxiety), 3 (moderate anxiety) to 4 (severe anxiety). It was a fast and simple questionnaire requiring a mark in the facial drawing that showed a distinct anxiety level describing how the patient felt at the time of assessment (Figures 1 and 2). Additionally, clinical data such as blood pressure, heart rate, respiratory rate, and oxygen saturation were recorded concurrently with the administration of the questionnaire at both time points (before the TTE and again before sedation). Demographic data (including comorbidities like diabetes, hypertension, and cardiovascular disease) and the clinical indication for TEE were also routinely documented. The patient’s routine use of anti-anxiety medications (benzodiazepines, selective serotonin reuptake inhibitors) was recorded.

Figure 1.

Anxiety Mini questionnaire used to assess patient's anxiety levels. According to the flow-chart, the mini questionnaire was filled both before TTE, and after it was completed, prior to sedation for the transoesophageal echocardiogram.

Figure 2.

Flowchart for music protocol in patients referred for TEE.

Music

Music began immediately after randomisation, after completing the anxiety questionnaire and obtaining cardiovascular data. Patients in the intervention group could choose from pre-selected categories to be played during the TTE (bossa nova, jazz, classical music) via stereo speakers from a commercial streaming app, continuing till the end of TEE. As previously indicated, haemodynamic data and anxiety levels were checked again before proceeding to TEE.

Transoesophageal echocardiogram

After routine TTE, TEE was undertaken following an established protocol. Briefly, after local anaesthesia of the oropharynx with lidocaine spray and positioning a bite guard, the patient would be laid in the left lateral decubitus to avoid aspiration, and sedation was started. Following sedation, the oesophageal transducer was inserted through the oropharynx into the oesophagus to obtain relevant images for diagnosing structural and/or functional heart abnormalities. The duration of the examination was recorded.

Sedation

Sedation was performed as usual for conscious sedation, with Fentanyl and Midazolam. Initially, Fentanyl was administered at 1 mcg/kg (50 to 100 mcg IV in bolus: for patients over 60 years, the dose was 50 mcg). After 3 min of Fentanyl administration, Midazolam was given with 1 to 2 mg bolus IV. The sedation protocol for TEE at our institution involves administering a dose of 0.1 mg/kg of midazolam, up to a maximum total dose of 10 mg. This dose is given in fractions every two minutes, in 1 mg increments (or 2 mg boluses if the patient is overweight), until conscious sedation is achieved. After each dose, sedation status and clinical parameters are monitored before administering the next dose. Additional Midazolam doses could be administered at the discretion of the attending physician to maintain sedation during the examination.

Statistical analysis

Continuous variables were expressed as means and standard deviations (SD) or medians and quartiles, as appropriate, and categorical variables as absolute frequencies and percentages. The variation in outcomes at baseline (before TTE) and after the end of TTE (before sedation) for Control and Music groups was assessed using linear mixed models, to test for group effects, time effects, and group-by-time interaction effects for the following variables: systolic and diastolic blood pressure, heart and respiratory rates, oxygen saturation, and prevalence of anxiety. For quantitative outcomes, models with Tweedie distribution were adjusted, and for anxiety outcomes, a model with a binomial distribution was used. Comparisons of sample characteristics between groups were performed using the Student's t-test (i) or the Mann–Whitney test (ii), depending on data distribution, and chi-square or Fisher's exact test for qualitative variables. Correlations between two quantitative variables were tested using Pearson or Spearman correlation. Anxiety severity levels between evaluation time points were compared using the marginal homogeneity test. Analyses were conducted using the SPSS statistical programme, version 26.0 (6), with a significance level of 5%. Normality of continuous variables was assessed with the Shapiro-Wilk test.

The necessary sample size for the study was determined to be 64 patients (32 per group) using a 5% significance level and 80% test power. To calculate the sample size for the population, two approaches were considered: one based on the incidence of anxiety and the other on the average dose of midazolam administered for a TEE. However, data on these two variables were not specifically available in the published literature. For midazolam dosage, however, we could use the average dose administered in our institution. To obtain this, we collected data on the midazolam doses used for sedation in all TEEs performed over a three-month period. This yielded an average dose of 2.98 ± 0.39 mg, which was then used for sample size calculation. Eligible participants were randomly assigned to either the control or intervention group in a 1:1 ratio. All data were securely stored in the Research Electronic Data Capture (REDCap) platform hosted at Hospital Israelita Albert Einstein, together with clinical information for subsequent analysis.¹²

Results

Patients were enrolled over a period of 9 months, from February 2023 to November 2023. A total of 68 participants were included in the study protocol; five patients were excluded: two due to hearing impairment, two had a contraindication to TEE and one refused to participate, thus remaining 63 patients: 32 were randomized to the control group and 31 for the music group.

Most participants were outpatients (82.5%, n = 52), aged 52.5 ± 14.8 years, and mostly men (65.1%, n = 41). Average weight was 77.8 ± 15.3, with height 172.2 ± 8.9 cm and body mass index of 26.1 ± 3.9. Most patients were white (87.3%, n = 55) and had a high educational level (95.1%). Comorbidities were not so common, and observed in only 34.9% of the population, the most frequent being hypertension (63.6%), followed by obesity (27.3%). The selected music primarily consisted of bossa nova (36%) and jazz (36%), followed by classical music (28%). Most patients (62%) reported experiencing some degree of anxiety before the examination; among them, anxiety was mild in most patients (40%), moderate in 16% and 6% had severe anxiety.

Regarding the characteristics of the two subgroups, there was no significant difference in demographics (age, weight, height or sex, Table 1); additionally, both groups were similar concerning education, comorbidities and ethnics. The routine use of anti-anxiety medications was not frequent (11%; six patients were using lorazepam and one used buspirone) and was similar in both groups, with 15.6% in the control group and 6.5% in the music group (p = NS). The most common indication for the examination in the control group was valvular heart disease (37.5%), while in the music group, stroke (29%) was the predominant indication. Most patients had been well informed by their assistant physician about the examination (over 90% in both groups), including its potential complications. In both groups, most patients had never undertaken a TEE; those patients that had undertaken the examination before did not refer to having a negative experience related to the examination. There was no difference regarding the prevalence of the type of chosen music between the groups.

Table 1.

Demographic and clinical characteristics of the patients divided into subgroups

	Control	Music	P
Age (years)	55.1 ± 13.7	49.8 ± 15.5	0.2
Weight (kg)	79.2 ± 13.8	78.4 ± 16.9	0.7
Height (cm)	171.9 ± 8.3	172.5 ± 9.6	0.8
BMI (Kg/m2)	26.0 ± 3.7	26.2 ± 4.1	0.8
Male Sex (%)	23	18	0.2

BMI: body mass index; numbers in brackets refer to interquartile range, medians are shown.

Anxiety results: At baseline (prior to TTE) and immediately before sedation, there were no statistically significant differences in the proportion of patients with anxiety between the control and music intervention groups (P = 0.589 and P = 0.239, respectively; Table 2). However, a significant reduction in anxiety over time was observed within the music group (P = 0.032), but not in the control group (P = 0.069), as shown in Table 2 and Figure 3. Despite these within-group differences, the interaction effect between group and time was not statistically significant (P = 0.347).

Table 2.

Haemodynamic, respiratory, and anxiety measures before TTE and after TTE (prior to TEE sedation)

	Control		Music		P
	Baseline	Before sedatiom	Baseline	Before sedation	p*	p**	pa	pb	pc
SBP (mmHg)	125.3 ± 13.7	125.9 ± 16.0	124.1 ± 15.9	123.0 ± 13.8	0.754	0.432	0.754	0.655	0.585
DBP (mmHg)	77.3 ± 8.1	77.4 ± 9.2	77.7 ± 12.0	74.0 ± 13.3	0.876	0.229	0.966	0.024	0.087
HR (bpm)	69 [62; 77]	68 [60; 75]	70 [61; 78]	70 [61; 75]	0.799	0.469	0.273	0.979	0.421
RR (rpm)	15 [12; 16]	15 [13; 17]	14 [12; 16]	14 [12; 16]	0.731	0.155	0.575	0.534	0.406
O2 sat (%)	99 [98; 100]	99 [97; 100]	99 [98; 100]	99 [97; 100]	0.777	0.524	0.546	0.338	0.641
Anxiety (%)	25.0% (13.0%; 42.6%)	15.6% (6.7%; 32.5%)	19.4% (9.0%; 36.9%)	6.5% (1.6%; 22.4%)	0.0589	0.239	0.069	0.032	0.347

DBP, diastolic blood pressure; HR, heart rate; O2 sat, Oxygen saturation; RR, respiratory rate; SBP, systolic blood pressure; TEE, transoesophageal echocardiography; TTE, transthoracic echocardiography.

Numbers in brackets refer to interquartile range, medians prior to brackets.

p* refers to test between groups Control and Music at baseline (before TTE) and p** between groups before sedation for TEE (after the end of TTE).

p^a and p^b: refer to comparison within the same group in different timepoints (baseline compared with before sedation for TEE) for Control group (p^a) and for Music group (p^b).

p^c refers to the interaction between the groups.

Figure 3.

Proportion of anxiety in patients referred to TEE in groups control and music, before TTE and after the end of TTE (post-TEE).

Examining these changes in detail, the Music group demonstrated more improvement in the level of anxiety: two patients who initially informed a high level of anxiety reported only mild anxiety after listening to music, and among the four patients with moderate degrees of anxiety, one reported only mild anxiety and another no anxiety at all (P = 0.046 and P = 0.008 for Control and Music respectively, Figure 4).

Figure 4.

Anxiety levels at baseline (before TTE) and after the end of TTE (prior to sedation for TEE) for music and control groups.

In terms of hemodynamics, there was no difference between the groups Music and Control at Baseline or before TEE (P > 0.05 Table 2); A decrease in blood pressure, however, was seen for the Music group after the intervention compared with Baseline (from 77.7 ± 12.0 mmHg at Baseline to 74 ± 13.3 mmHg just before sedation, P = 0.024), as shown in Figure 5. Table 2 illustrates the haemodynamic and respiratory variables pre- and post-intervention between the groups.

Figure 5.

Diastolic blood pressure comparison between control and music groups at baseline and following TTE (immediately before sedation for TEE).

Comparison of sedation dose between the groups

There was no difference in Fentanyl dose for both music and control group; midazolam dose (both initial and total dose) was also similar for both groups. Because there were more women in the control group, we also adjusted the midazolam dose to weight, but still there was no difference between the groups. These results are shown in Table 3. The proportion of patients that needed an additional midazolam dose for sedation was similar for both groups (59.4% vs. 64.5% for control and music groups respectively, P = 0.6). Duration of TEE and need for an anaesthesiologist due to failure of sedation were similar for both music and control groups (one patient in each group needed general anaesthesia).

Table 3.

Comparison of sedation dose for TEE

	Control	Music	p
Fentanyl dose (mcg)	50.0 [50.0; 50.0]	50.0 [50.0; 50.0]	0.5
Initial Midazolam dose (mg)	4.0 [3.0; 5.5]	5.0 [4.0; 6.0]	0.4
Total midazolam dose (mg)	5.0 [4.5; 7.0]	5.0 [4.0; 8.0]	0.8
Total midazolam dose/weight (mg/kg)	0.08 [0.07; 0.05]	0.08 [0.05; 0.12]	0.9

Numbers in brackets refer to interquartile range, medians are shown prior to brackets.

Correlation of sedation dose with clinical variables

When we tested the entire patient group with the midazolam dose, we found an inverse correlation between the total midazolam dose and age (r = −0.419, P = 0.001), as shown in Figure 6; however, a positive correlation was seen with pre-examination oximetry (r = 0.393, P = 0.001). Fentanyl dose, on the other hand, showed a positive correlation with diastolic blood pressure before the examination (r = 0.320). Total midazolam dose was higher for outpatients (P = 0.013) and those patients without comorbidities (P = 0.004).

Figure 6.

Correlation of patient age with administered total midazolam dose for the whole population.

Discussion

Several studies have previously suggested that music therapy can effectively reduce stress and anxiety in patients undergoing various medical procedures. Music, being a non-pharmacological, low-cost, and easily implementable intervention, has been explored in diverse clinical settings, including preoperative and post-operative periods or during outpatient procedures such as colonoscopies and bronchoscopies^9,10,13,14 Our study, which investigated the potential role of music as an adjunct to sedation during TEE, is consistent with these findings, particularly regarding de degree of anxiety reduction. We observed that anxiety was a very common finding prior TEE, with about 60% of our population feeling anxious before the examination; although both control and intervention groups had mainly low baseline anxiety levels, we observed that music had an effect on those patients with higher degree of anxiety. This reduction highlights the soothing effect of music, which can make the TEE experience less distressing for patients. This is especially important because in terms of healthcare quality, positive experiences have been associated with better health outcomes.

The slight but significant reduction in diastolic blood pressure in the music group aligns with previous reports that have documented the physiological effects of music, including reductions in heart rate and blood pressure,^15,16,17 .This finding suggests that music could contribute to a more stable haemodynamic profile during procedures that typically induce stress or anxiety. However, the absence of significant changes in other haemodynamic variables, such as systolic blood pressure, heart rate, and respiratory rate, indicates that while music can influence some physiological responses, its effects may be limited or somewhat variable across different parameters.

Interestingly, despite the reduction in anxiety and diastolic blood pressure, midazolam dose was not lower in the music group compared with the control group. This finding could be attributed to several factors. First, the standardized sedation protocol used in our study, which aimed to ensure patient comfort and procedural success, may have required regular dosing regardless of the patient's anxiety levels. Additionally, we can expect that the subjective nature of anxiety and its complex interplay with sedation could perhaps explain why reduction in anxiety did not translate into less sedation. Furthermore, the variability in individual responses to sedatives might have influenced the overall impact on sedation requirements. We did not observe any influence regarding the specific type of music used; however, the sample size was likely too small to make statistically significant conclusions on this matter.

It is worth noting that other studies have reported a reduction in sedative doses when music was used during regional anesthesia^18,19 or invasive angiography¹⁵. This discrepancy suggests that the impact of music on sedation requirements may depend on the clinical context, the type of anaesthesia, or the specific patient population involved. The lack of a reduction in sedation dose, despite the decrease in anxiety, is noteworthy and implies that while music can enhance patient comfort, it may not be sufficient on its own to reduce sedative requirements in procedures like TEE. The inherent anxiety linked to the procedure, likely due to its invasiveness and the proximity of the probe to critical structures, might necessitate a certain level of sedation that cannot be adequately addressed by music alone.

Our study suggests potential implications for clinical practice. The ability of music to reduce anxiety and influence haemodynamic stability suggests that it could be a valuable adjunct in the TEE setting, particularly for patients who are more prone to anxiety or those with cardiovascular conditions that could be exacerbated by stress. Although music has been used during other medical procedures, it had not been used or proved to be useful during TEE. However, the lack of an impact on sedation dose indicates that music should be viewed as a complementary therapy rather than a replacement for pharmacological sedation.

This study has some limitations that should be acknowledged. The relatively small sample size may limit the generalizability of our findings, and the lack of a placebo-controlled group (e.g. patients listening to white noise or wearing headphones without music) could mean that some of the effects attributed to music might be due to the mere presence of auditory stimuli. Also, the study design did not have the power to account for different types of music or patient preferences, which could play a role in the efficacy of music therapy. Future studies with larger sample sizes, diverse musical interventions, and placebo controls are warranted to further elucidate the potential of music as a sedation adjunct.

Additionally, the questionnaire used for anxiety assessment was very simple; however, we chose an uncomplicated format to ensure it was quick and easy to administer, as well as easy to evaluate.

Finally, the influence of music on the awakening of the patient was not assessed; we intentionally excluded the evaluation of anxiety and haemodynamic parameters after TEE, due to the potential influence of sedation as an additional variable affecting both anxiety levels and haemodynamic parameters. Once midazolam is administered, most patients experience a decrease in blood pressure, typically managed with saline infusion to stabilize that condition. Additionally, after the examination, anxiety was generally absent for all patients; furthermore, we had no way of confirming whether the patient was listening to the music once they became unconscious from sedation. We believe that might be a topic for future research.

Conclusion

While music therapy did not reduce the dose of sedatives required during TEE, it effectively decreased patient anxiety and diastolic blood pressure; these findings support the use of music as a non-invasive, cost-effective means to enhance patient comfort during TEE, though it should be considered as part of a broader sedation strategy rather than a sole intervention.

Funding

The authors declare that no funds were received during the preparation of this manuscript.

Lead author biography

Amanda de Vasconcelos Eng is a sixth-year medical student at Faculdade Israelita de Ciências da Saúde Albert Einstein, São Paulo, Brazil. She has a strong interest in academic research and knowledge production, alongside internal medicine—particularly cardiology and neurology—as well as medical humanities and patient-centered care. She is currently preparing for residency training.

Data availability

The data underlying this article are available from the corresponding author upon reasonable request.