Effects of music on psychophysiological responses and opioid dosage in patients undergoing total knee replacement surgery
Abstract
Aim
The present authors examined the effects of listening to music on psychophysiological parameters (blood pressure, heart rate, and respiratory rate) during preoperative and postoperative days and determined whether listening to music could lower pain intensity and opioid dosage during postoperative days in patients who underwent total knee replacements.
Methods
This was a two group repeated measures design for 30 subjects aged 53–85 years who were scheduled for total knee replacement. Subjects were randomly assigned to either a music group or a control group. Psychophysiological parameters were obtained from patients' monitors. A visual analog scale was used to assess postoperative pain. Opioid dosage was recorded and converted to standardized units. Mann–Whitney U-test and generalized estimating equation analysis were used to compare groups.
Results
Respiratory rates while in the surgical waiting area were lower for the music group than for the control group (P = 0.02). There was no significant difference between these groups for blood pressure, heart rate, pain intensity, or opioid dosage. However, a within-group comparison showed that systolic blood pressure in the music group was significantly and consistently decreased during postoperative recovery (Wald = 9.21, P = 0.007).
Conclusion
These results suggest that listening to music stabilized systolic blood pressure in patients during postoperative recovery. However, the effects of music on psychophysiological parameters, pain intensity, and opioid dosage in a surgical setting require further research.
Introduction
Taiwan has had the second most rapidly aging population in the world since 2009 (Council for Economic Planning and Development, 2009). This has naturally been accompanied by an increase in the number of elderly people with osteoarthritis (OA). Approximately 9.6–18.0% of men and women aged 60 years or more had OA (Woolf & Pfleger, 2003). OA is expected to be the fourth leading cause of disability by the year 2020 (Woolf & Pfleger, 2003). It is estimated that 94% of patients undergoing total knee replacement (TKR) are diagnosed with OA (Tien et al., 2009). More than 4.5 million Americans have had a TKR (American Academy of Orthopedic Surgeons [AAOS] 2012). The number of TKR has more than doubled during the past decade (AAOS, 2012). In Taiwan, the number of TKR increased 2.4 times from 1996 to 2004 (Tien et al., 2009). In 1995, a national health insurance program was implemented in Taiwan. Based on a Taiwan health insurance bureau announcement in 2000, the medical expense for TKR was $NT 2.63 billion, which was the second highest healthcare expense among total inpatient medical expenses (Corporation Aggregate National Federation of Associations of Physical Therapists, 2012).
Undergoing any surgery is inherently stressful. Spielberger (1972) identified stress as a precursor to anxiety. Patients undergoing surgery may experience anxiety from uncertainty regarding the success of the surgery. Mind/body relaxation has been found to be beneficial for stress management in various populations (Benson, 2009).
Mind and body are reciprocal influences. Anxiety has been consistently characterized by a pattern of reciprocal inhibition between sympathetic activation and vagal deactivation (Kreibig, 2010). Autonomic nervous system arousal increases heart rate and blood pressure, which increase the oxygen demand and the heart workload (Chang, Peng, Wang, & Lai, 2011; Guyton & Hall, 2005). These may have potential adverse effects on overall patients' outcome (Lai, Ortiz-Cardona, & Bendo, 2012).
Providing pain management and improving well-being by delivering optimal nursing care to patients are primary nursing concerns (The American Society for Pain Management Nursing, 2014). Medication is the most commonly used treatment for pain relief (e.g. opioids) in patients undergoing orthopedic surgery. Considering the negative side-effects of opioids, such as respiratory depression, non-pharmacological interventions have long been recognized as adjuvant treatments for pain management (Good, 1996). Music has been used as an adjuvant for pain management (Bernatzky, Presch, Anderson, & Panksepp, 2011; Vaajoki, Pietilä, Kankkunen, & Vehviläinen-Julkunen, 2012) with varying evidence from studies of patients who underwent surgeries (The Joanna Briggs Institute, 2011).
In addition to anxiolytic treatment, the patients prefer to listen to music to relieve anxiety (Chang et al., 2011). Soothing music intervention for pain and anxiety management has been postulated to have mixed results (The Joanna Briggs Institute, 2011). Systematic reviews of randomized controlled trials on the effects of music on anxiety (Nilsson, 2008) and pain relief (Cepeda, Carr, Lau, & Alvarez, 2006; Nilsson, 2008) showed that music reduced preoperative anxiety, postoperative pain (Nilsson, 2008; Vaajoki et al., 2012; Zengin et al., 2013), and the use of analgesic medications (Cepeda et al., 2006; Nilsson, 2008) among various elective surgeries in various hospital settings. However, there is lack of sufficient evidence to recommend the use of music for the physiological parameters of anxiety (The Joanna Briggs Institute, 2011). In addition, no clear effect of music has been observed in reducing postoperative pain intensity levels or opioid requirement (Cepeda et al., 2006). Therefore, it was recommended that additional studies should be conducted to examine anxiety, pain relief, and analgesic medications as outcome measures for the effects of music (Chiang, 2008; The Joanna Briggs Institute, 2011).
Background
Since a decade ago, nursing researchers have had synthesized published works proposing several theories to explain the effectiveness of soothing music (Chiang, 2008; Lai & Good, 2002). Music can induce relaxation and distraction (Nilsson, 2008). Based on a psychophysiological theory, this reduces responses in the neuroendocrine and sympathetic nervous systems (Zatorre & McGill, 2005), which results in decreased stress, heart rate, respiratory rate, and blood pressure (Nilsson, 2008). Based on psychophysiological theory, studies have continuously confirmed part of the mechanism underlying the effects of soothing music (Chang, Lai, Chen, Hsieh, & Lee, 2012; Chang et al., 2011). Reducing stress decreases serum cortisol levels (Lai & Li, 2011), which results in decreased heart and respiratory rates (Guyton & Hall, 2005) along with promotion of psychological well-being (Chang et al., 2011).
- To what extent do the effects of listening to music on the preoperative physiological parameters (systolic blood pressure, diastolic blood pressure, heart rate, and respiratory rate) in the ward and in the surgical waiting area differ between and within groups of patients undergoing TKR?
- To what extent do the effects of listening to music on the postoperative psychophysiological parameters (systolic blood pressure, diastolic blood pressure, heart rate, respiratory rate, and pain intensity), and opioid dosage differ during postoperative recovery (POR) and in the ward between and within groups of patients undergoing TKR?
Methods
A two group repeated measures design was conducted to examine the study aims. A randomized control trial, with sealed envelopes stratified by sex, was used to assign different sex subjects to either a music or a control group. The music group received a soothing music intervention and the control group received the usual care (n = 15/group). To examine the lingering effect of music on psychophysiological parameters, the present authors observed this study population from the time of their admission to the ward to their return to the ward after TKR. Hence, this study could be divided into four phases: phase I, in the ward before surgery; phase II, in the waiting area of the surgical room; phase III, in POR; and phase IV, in the ward after surgery. Physiological parameters were recorded for all four phases. Pain intensity and opioid dosage were recorded for phases III and IV. To examine if there was a linger effect of music (Ho et al., 2011) on the postoperative psychophysiological parameters, no music was provided for both groups during phase IV.
Participants
This study was conducted at a 1000 bed multispecialty hospital in Taiwan. The waiting area for the surgical room was an independent, open space with a total bed capacity (electric transport stretcher) of 11. The total bed (electric transport stretcher) capacity of the POR area where the study was conducted was 14.
Power analysis revealed that a total of 28 subjects was an adequate sample size for an alpha of 0.05 and a beta of 0.2 with a medium effect size, an assumed moderate correlation (r = 0.50) among repeated measures, and for a repeated measures ancova of heart rate (Chang et al., 2011). The study included a total of 30 subjects. Inclusion criteria were: (i) waiting for TKR surgery without premedications; (ii) aged between 45 and 85 years; and (iii) able to communicate in either Mandarin or Taiwanese. Exclusion criteria were: (i) any hearing impairment; (ii) pregnancy; (iii) spinal anesthesia; (iv) alcoholism; and (v) infectious disease. Over the 13 month period, 56 patients were contacted by face-to-face interviews from a list of patients who were scheduled for TKR appointments; all 56 patients were willing to participate. However, one participant was later diagnosed with tuberculosis and 15 patients had changes in their surgery times. Thus, 16 patients were eliminated from this study. The remaining 30 subjects completed the four phases of study assessments.
Experimental intervention
Music intervention comprised five compositions of 30 min soothing piano and Chinese violin music. The music compositions included Spring Rural Field, Woman under the Moon (Chinese music), Going Home (Czech music), Heart Lotus (Taiwanese music), and Memory. These music compositions were collected on a compact disc (CD). Subjects listened to music played on a CD player through broadcast speakers. A recently study revealed that both headphone and broadcast music are effective for reducing the preoperative patient's anxiety (Lee, Chao, Yiin, Chiang, & Chao, 2011). Different musical compositions can achieve the same therapeutic effects if music has similar characteristics (Cepeda et al., 2006). The musical pieces all had similar characteristics, including a smooth rhythm and melody lines with a tempo of 60–80 b.p.m. to achieve relaxation (Chang et al., 2011). Thus, this study used music compositions with similar characteristics, which previous studies revealed to have beneficial effects on relaxation in cancer patient caregivers (Lai, Li, & Lee, 2012), sleep quality of intensive care unit (ICU) patients (Su et al., 2013), and agitation in older adults with dementia (Ho et al., 2011).
Music intervention was provided during phases I, II, and III of the study. The music intervention lasted 30 min (The Joanna Briggs Institute, 2011) in phases I and II. In phase I, the music was played during the evening between 19.30 and 21.00 hours. Hearing is the first sense to return following general anesthesia (Cynthia, 2009). In phase II, the subjects were lying down on a stretcher bed to listen to music on the operation day in the surgical waiting area. In POR, the primary nurse was told that the music would automatically play twice; thus, the music was played for 60 min throughout the period while in POR. The music volume was set at a comfortable level for subjects according to their preferred volumes. While in POR, the music volume was controlled by the investigator. Each subject listened to music for a total of 120 min during the study period. To control for the diffusion of treatment in the open areas, the present authors provided music for only one subject to listen to at the same time period.
Measurements and analysis
Patient characteristics
These included the subjects' demographic data and a brief medical history, including prior TKR and anxiolytic medications. This brief questionnaire was developed by all the investigators involved with this study.
Psychophysiological parameters
The psychophysiological parameters measured included heart rate, blood pressure, respiratory rate, and pain intensity. Heart rate, blood pressure, and respiratory rate were recorded during the four phases using patient monitors (GE Solar 8000; Somatechnology, Bloomfield CT, USA). The monitor was calibrated regularly by the hospital. The accuracy of the monitor was 99.97–98.8%. These physiological responses were measured before and after the intervention. They were recorded by a data collector or primary nurses depending on the phases of the study (Table 1). During phases I and II, the physiological parameters were measured before and after the music intervention. During phase III, the physiological parameters were measured at a 15 min interval. During phase IV, the physiological parameters were measured by the primary nurses at a 4 h interval.
| Variables | Total (n = 30) | Music (n = 15) | Control (n = 15) | χ2 | P | |||
|---|---|---|---|---|---|---|---|---|
| N | % | N | % | N | % | |||
| Sex | 0 | 1 | ||||||
| Male | 10 | 33.3 | 5 | 33.3 | 5 | 33.3 | ||
| Female | 20 | 66.7 | 10 | 66.7 | 10 | 66.7 | ||
| Previous experience of any operation | 0.83 | 0.36 | ||||||
| Yes | 24 | 80 | 13 | 86.7 | 11 | 73.3 | ||
| No | 6 | 20 | 2 | 13.3 | 4 | 26.7 | ||
| Mean | SD | Mean | SD | Mean | SD | t | P | |
|---|---|---|---|---|---|---|---|---|
| Mean age in years | 68 | 9 | 65.93 | 9.3 | 70.07 | 8.50 | −1.27 | 0.21 |
| Operation time (h) | 2.8 | 0.48 | 2.83 | 0.52 | 2.76 | 0.45 | 0.37 | 0.71 |
| Loss of blood during the intraoperative phase | 203.33 | 196.93 | 263.33 | 258.75 | 143.33 | 75.27 | 1.72 | 0.09 |
| Medication during the intraoperative phase | ||||||||
| Fentanyl | 56.83 | 20.94 | 58.33 | 18.09 | 55.33 | 24.01 | 0.38 | 0.70 |
| Xylocaine | 42.66 | 41.93 | 48.00 | 58.08 | 37.33 | 14.37 | 0.69 | 0.49 |
| Propofol | 123.33 | 40.47 | 123.33 | 51.04 | 123.33 | 31.54 | 0.00 | 1.00 |
- SD, standard deviation.
The pain intensity scores were assessed using a visual analog scale (VAS), which consisted of a horizontal line 10 cm in length with the end-points “no pain” and “worst imaginable pain” placed at each end of the line (Huskisson, 1983). The VAS, routinely used by nurses in clinical settings for assessing pain intensity, was used during POR. Its validity and reliability has been tested and the VAS has been found an appropriate measure of postoperative pain (DeLoach, Higgins, Caplan, & Stiff, 1998). The pain intensity scores were recorded based on observations made by the primary nurses during phase III and based on patients' chief complaints during phase IV.
Opioid dosage
Parenteral opioid analgesics included morphine, meperidine (pethidine), and fentanyl. The opioid was provided as patients needed. The opioid dosage amount during phase III and phase IV was recorded by the primary nurses and converted to standardized units (morphine 1 mg = fentanyl 0.01 mg = meperidine 75 mg) by the investigators according to the criteria of the American Pain Society (1996, p. 26).
Procedures
This study protocol was approved by the hospital's research ethics committee (IRB100-03). The investigators assessed participants' eligibility when they arrived in the ward at their appointed time. Detailed explanations of the study were provided to eligible participants and their families. Written informed consent was obtained from all subjects. All were informed that their participation was voluntary and that their confidentiality was assured. Subjects were told that they could withdraw during the intervention at any time. An investigator drew a lot (label) to determine which group the patients would be assigned. To decrease the error rate, all lots were placed in an envelope that was prepared by another individual. Thus, the data collector did not know which group each subject would be assigned. Baseline data were collected before each phase.
Phase I
Subjects were asked to rest for 10 min while the investigator prepared the study equipment. Following baseline assessments, subjects commenced 30 min of music intervention during the night prior to the operative day. Heart rate, blood pressure, and respiratory rate were recorded before and after the intervention. The room temperature of the ward ranged 21–26°C, which was controlled by the subjects.
Phase II
On the second day (op-day), subjects were sent to the surgical room in the morning by a porter using a transport electric stretcher. The music intervention was conducted in the surgical waiting area of the surgical room. An investigator also ensured that no tranquillizer had been consumed for a minimum of 30 min before the baseline data were recorded. Subjects were lying down on a stretcher bed to wait or to listen to music for 30 min. The stretcher bed was adjusted to their desired position between 45° and 60° such that each subject was in a comfortable position. Subjects in the music group were instructed on how to adjust the music volume to meet their personal preferences. This intervention lasted for 30 min and physiological indices were recorded before and after the intervention. During this intervention, subjects were asked to turn off their mobile phones. The room temperature of the surgical waiting area was maintained at a comfortable 22–24°C. The control group went through the same procedures, but without the music intervention.
Phase III
After the surgery was completed, subjects were sent to the POR area. Baseline assessments of the physiological indices and pain scores were recorded immediately after stabilization for 5 min. Hearing is the first sense to return following general anesthesia (Cynthia, 2009). Thus, the music was played for 60 min immediately after the baseline data were collected. Heart rates, blood pressures, respiratory rates, and pain scores were recorded on the POR nursing record sheet at four time points with a 15 min interval between each measurement. No subject was interrupted during the intervention procedure, except when recording psychophysiological parameters. The time interval for the measurements accommodated the POR care routine. Opioid medication usage was also recorded during the stay in POR; this was the sum of the total amount of opioids consumed. The control group went through the same procedures, but without the music intervention. No subject was interrupted during this phase.
Phase IV
One hour later, the subjects were sent back to the ward. The duration between the end of the music in POR and the beginning of phase IV was between 20 and 40 min. Again, baseline assessments of the physiological indices and pain scores were recorded. An adaptation period of approximately 5 min before the baseline measurement recording allowed their heart rates to stabilize. During this phase, there was no music played for both groups. The purpose of this part of the study design was to determine if there was a linger effect (Ho et al., 2011) of the music after receiving a 60 min music intervention in POR. Heart rates, blood pressures, respiratory rates, and pain scores were recorded on the temperature, pulse, respiration (TPR) sheet at four time points with a 4 h interval between each measurement. Opioid medication use within the 12 h in the ward was recorded and was the sum of the total amount of opioid dosage during phase IV.
Statistical analysis
Data were analyzed using PASW version 18.0 for Windows (SPSS, Chicago, IL, USA). Subjects' demographic variables were compared by Student's t-test and χ2-test. A Mann–Whitney U-test was used to compare groups during phases I and II. As data was collected multiple times in phases III and IV, and data across all phases were analyzed, generalized estimating equation (GEE) analysis (Liang & Zeger, 1986) was used to control changes in time and baseline values. The baseline outcome measurements were used as covariates in GEE analysis (Cook & Campbell, 1979). P < 0.05 with Bonferroni correction was considered statistically significant.
Results
Demographic characteristics
The age of subjects ranged 53–85 years (mean = 68 years, standard deviation [SD] = 9). The mean operation time was 2.8 h (SD = 0.48). This study included 10 men (33.3%) and 20 women (66.7%); the sex distribution in each group was equal. Most subjects were married (n = 26, 86.7%) and had history of surgery (n = 24, 80 %). More than two thirds (76.7%) had completed grade school (6 years) (Table 1).
Group comparability
There was no significant baseline difference between the music and control groups in terms of subjects' demographics (Table 1). There was also no difference in these groups for baseline data (Table 2). Moreover, there were no significant differences in medications (e.g. fentanyl, xylocaine, and propofol), and loss of blood (Table 1) and the surgical procedure during the intraoperative phase.
| Physiological parameters | Total (n = 30) | Music (n = 15) | Control (n = 15) | Z | P | |||
|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | |||
| Phase I: preoperation in the ward | ||||||||
| Heart rate | 75.20 | 11.22 | 75.53 | 12.13 | 74.00 | 8.70 | 0.16 | 0.87 |
| Respiratory rate | 17.20 | 1.97 | 17.60 | 1.64 | 16.8 | 2.24 | 1.12 | 0.27 |
| Systolic blood pressure | 145.57 | 20.23 | 146.73 | 22.02 | 144.40 | 18.97 | 0.31 | 0.75 |
| Diastolic blood pressure | 79.97 | 11.96 | 79.93 | 13.59 | 80.0 | 10.54 | −0.02 | 0.98 |
| Phase II: in the surgical waiting area | ||||||||
| Heart rate | 69.50 | 12.76 | 68.8 | 9.59 | 70.20 | 15.64 | −0.30 | 0.77 |
| Respiratory rate | 18.40 | 2.06 | 17.73 | 2.34 | 19.07 | 1.53 | −1.84 | 0.07 |
| Systolic blood pressure | 153.67 | 25.91 | 150.87 | 24.88 | 156.47 | 27.48 | −0.58 | 0.56 |
| Diastolic blood pressure | 83.37 | 14.34 | 81.80 | 13.27 | 84.93 | 15.64 | −0.59 | 0.55 |
| Phase III: in postoperative recovery | ||||||||
| Heart rate | 82.23 | 13.04 | 86.2 | 12.96 | 78.27 | 12.27 | 1.72 | 0.10 |
| Respiratory rate | 17.23 | 2.98 | 17.80 | 3.49 | 16.66 | 2.35 | 1.04 | 0.31 |
| Systolic blood pressure | 145.63 | 26.11 | 152.07 | 19.60 | 139.20 | 30.64 | 1.37 | 0.18 |
| Diastolic blood pressure | 78.70 | 12.79 | 80.33 | 9.94 | 77.06 | 15.30 | 0.69 | 0.49 |
| Pain scorea | 5.00 | 1.20 | 4.90 | 1.04 | 5.09 | 1.38 | −0.35 | 0.73 |
| Phase IV: postoperation in the ward | ||||||||
| Heart rate | 77.27 | 11.96 | 77.87 | 13.99 | 76.67 | 9.99 | 0.27 | 0.78 |
| Respiratory rate | 18.33 | 1.32 | 18.20 | 1.52 | 18.47 | 1.23 | −0.55 | 0.59 |
| Systolic blood pressure | 144.20 | 24.45 | 139.93 | 24.82 | 148.47 | 24.16 | −0.95 | 0.35 |
| Diastolic blood pressure | 77.97 | 11.58 | 76.67 | 13.92 | 79.27 | 8.97 | −0.61 | 0.55 |
| Pain score | 3.77 | 1.19 | 4.00 | 1.31 | 3.53 | 1.06 | 1.07 | 0.29 |
- a Missing = 8 (4/group). SD, standard deviation.
Psychophysiological indices
Blood pressure
Systolic blood pressure and diastolic blood pressure did not significantly differ between the groups during phases I and II (Table 3). Blood pressure was repeatedly measured during phases III and IV. GEE analysis was used to examine group and time effects. After controlling for baseline blood pressure, there was no group difference in blood pressure (Table 4). However, there were significant differences in systolic blood pressure over time (Wald = 9.21, P = 0.007). Interaction effects between time and group were also found (Wald = 24.79, P = 0.001). Systolic blood pressure was consistently decreased in the music group as shown by the slope of 7.9% of the baseline at T4. This indicated that listening to music had beneficial effects on the systolic blood pressure while in POR, as shown by the trend pattern in Figure 1.
| Physiological parameters | Total (n = 30) | Music (n = 15) | Control (n = 15) | Z | P | |||
|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | |||
| Phase I: in the ward before surgery | ||||||||
| Heart rate | 68.7 | 9.67 | 70.53 | 10.69 | 67.07 | 8.78 | 1.04 | 0.30 |
| Respiratory rate | 18.30 | 1.15 | 18.67 | 1.29 | 17.93 | 0.88 | 1.81 | 0.08 |
| Systolic blood pressure | 139.37 | 22.04 | 142.80 | 22.12 | 135.93 | 22.16 | 0.85 | 0.40 |
| Diastolic blood pressure | 76.53 | 11.56 | 79.47 | 12.98 | 73.60 | 9.48 | 1.41 | 0.17 |
| Phase II: in the surgical waiting area | ||||||||
| Heart rate | 71.80 | 14.29 | 71.47 | 8.50 | 72.13 | 18.72 | −0.13 | 0.90 |
| Respiratory rate | 18.43 | 2.29 | 17.53 | 2.20 | 19.33 | 2.05 | −2.13 | 0.02* |
| Systolic blood pressure | 152.8 | 25.24 | 146.07 | 24.73 | 159.53 | 24.72 | −1.49 | 0.14 |
| Diastolic blood pressure | 82.03 | 11.22 | 82.07 | 14.16 | 82.0 | 7.75 | 0.02 | 0.98 |
| Phase III: during postoperative recovery | ||||||||
| Opioid† | 7.14 | 2.46 | 7.39 | 2.66 | 6.86 | 2.29 | 0.57 | 0.57 |
| Phase IV: in the ward after surgery | ||||||||
| Opioid† | 12.41 | 11.71 | 12.04 | 14.43 | 12.90 | 8.05 | −0.14 | 0.89 |
- *P < 0.05. †Parenteral opioid analgesics included morphine, meperidine (pethidine), and fentanyl. The amount of opioid dosage was converted to standardized units (morphine 1 mg = fentanyl 0.01 mg = meperidine 75 mg) according to the criteria of the American Pain Society.
| Outcome measure | Group | Estimate | SE | 95% Wald CI | Wald | P |
|---|---|---|---|---|---|---|
| In postoperative recovery | ||||||
| Heart rate | Musica | 81.20 | 2.90 | 75.51–86.89 | 1.07 | 0.29 |
| Control | 76.77 | 3.13 | 70.63–82.90 | |||
| Respiratory rate | Music | 16.78 | 0.50 | 15.80–17.77 | 0.06 | 0.79 |
| Control | 16.63 | 0.31 | 16.02–17.24 | |||
| Systolic blood pressure | Music | 147.62 | 4.28 | 139.23–156.00 | 0.61 | 0.43 |
| Control | 142.37 | 5.19 | 132.20–152.54 | |||
| Diastolic blood pressure | Music | 78.10 | 2.27 | 73.64–82.56 | 1.23 | 0.27 |
| Control | 74.40 | 2.43 | 69.62–79.18 | |||
| Pain score | Music | 3.22 | 0.22 | 2.79–3.65 | 0.43 | 0.50 |
| Control | 3.00 | 0.25 | 2.52–3.48 | |||
| Postoperation in the ward | ||||||
| Heart rate | Music | 79.41 | 2.62 | 74.27–84.55 | 0.47 | 0.49 |
| Control | 81.72 | 2.12 | 77.57–85.87 | |||
| Respiratory rate | Music | 18.43 | 0.22 | 18.00–18.87 | 0.23 | 0.63 |
| Control | 18.30 | 0.18 | 17.95–18.65 | |||
| Systolic blood pressure | Music | 128.38 | 3.54 | 121.44–135.33 | 0.75 | 0.38 |
| Control | 133.19 | 4.25 | 124.86–141.52 | |||
| Diastolic blood pressure | Music | 71.41 | 2.30 | 66.90–75.93 | 1.52 | 0.22 |
| Control | 95.24 | 19.17 | 57.67–132.82 | |||
| Pain score | Music | 3.07 | 0.26 | 2.56–3.57 | 0.38 | 0.53 |
| Control | 2.87 | 0.18 | 2.51–3.23 |
- a Reference group. CI, confidence interval; SE, standard error.
Changes in systolic blood pressure at postoperative recovery between the groups at different time points. T1, 15th minute; T2, 30th minute; T3, 45th minute; T4, 60th minute. Systolic blood pressure was constantly decreased across time (Wald = 9.21, P = 0.007) in the music group. aStandard error.
Heart rate
Heart rate did not differ significantly between the groups during phases I and II (Table 3). In addition, after controlling for baseline heart rate, there was no difference in group heart rate during phases III and IV (Table 4). This indicated that listening to music did not significantly change the preoperative and postoperative heart rates.
Respiratory rate
The respiratory rates differed significantly between the groups during phase II (t = −2.13, P = 0.02). However, respiratory rates did not differ significantly between the groups during other phases (Tables 3, 4). This indicated that listening to music lowered the respiratory rate while waiting for TKR, but the respiratory rate did not change preoperatively and postoperatively in the ward.
Pain intensity
Pain intensity was assessed after TKR during phases III and IV. Pain intensity did not differ significantly between the groups during these two phases (Tables 3, 4). This indicated that listening to soothing music did not decrease the postoperative pain intensity.
Parenteral opioids
The total amount of opioid dosage was measured after TKR during phases III and IV. Opioid dosage did not significantly differ between the groups during phases III and IV, which indicated that listening to soothing music did not decrease the postoperative opioid dosage.
Discussion
The aims of this study were to examine the effects of music listening on preoperative and postoperative physiological parameters as well as postoperative pain and opioid dosage in patients undergoing TKR. Patients undergoing TKR frequently experience anxiety. Listening to music is one of the preferred activities of patients to relieve their anxiety while waiting for surgery (Chang et al., 2011). The present authors found that the mean changes in scores between the two groups were reflected in their respiratory rates while in the surgical waiting room. Within-group comparisons indicated that systolic blood pressure decreased in the music group significantly and consistently during POR. This suggested that music was effective in reducing the preoperative respiratory rate and postoperative systolic blood pressure. However, no significant difference was found in any other parameter between these groups during any other phases of this study.
Psychophysiological parameters
The findings of this study partially support those of recent studies with different research designs in which music had no significant effect on mean arterial blood pressure and heart and respiratory rates of patients who had undergone heart surgery (Nilsson, 2009, Özer, Özlü, Arslan, & Günes, 2013) or total knee arthroplasty (Allred, Byers, & Sole, 2010). A researcher in nursing reviewed 42 studies that examined the effects of music on patients who underwent different types of surgery and concluded that there was only minimal evidence for the effects of music on preoperative, postoperative, or intraoperative heart rate, blood pressure, and respiratory rate (Nilsson, 2008).
In the present study, there was a statistically significant difference in the preoperative respiratory rate between groups while they were in the surgical waiting room. During the four phases of this study, the preoperative respiratory rate was the only significantly different parameter between the two groups. Regarding blood pressure, the present authors' study demonstrated a consistent reduction in systolic blood pressure in the music group during POR.
Although there is still no consistent evidence for the effects of music on reduction in the respiratory rates of patients who underwent surgery or medical procedures, the present authors' study found a positive effect of music on decreased respiratory rates of patients while waiting for TKR and decreased postoperative systolic blood pressure during POR.
There was no significant difference in postoperative pain between the music and control groups. The present authors found that there was no significant difference in pain intensity between the groups during phase III. Also, there was no linger effect during the last phase of this study. This finding of a lack of effect of music on postoperative pain was consistent with recent randomized clinical trial studies that also found that music had no effect in patients who underwent TKR (Allred et al., 2010) and those who underwent open coronary artery bypass grafting or aortic valve replacement surgery (Nilsson, 2009).
However, other studies have found that music was effective for the postoperative pain of patients who underwent gynecologic surgeries (Good & Ahn, 2008; Li & Dong, 2012). These inconsistencies may be because of differences in study populations and the types of surgery. Although there is no agreement among studies on the effectiveness of music on postoperative pain relief, music still can be used as an adjuvant in pain management (Engwall & Duppils, 2009; The Joanna Briggs Institute, 2011).
Opioid dosage
There are very few previous studies that examined the effects of listening to music on opioid dosage in TKR. The present study found that there was no difference in postoperative opioid dosage between the two groups. This was similar to the results of a recent study by Chi et al. (2011) in which music did not reduce opioid dosage in women who were receiving intracavitary brachytherapy for gynecological cancer. The present authors' study findings also confirmed the results of a study that showed that after 6 h of initiating a music intervention, there was no difference in the amount of opioids used between groups who underwent TKR (Allred, Byers, & Sole, 2010). Fewer than half of previous studies found that music could reduce the use of analgesic medications (Nilsson, 2008). The present study findings supported those from a systematic review of randomized controlled trials by Cepeda et al. (2006) in which the clinical significance of listening to music in decreasing opioid requirements was uncertain. The lack of favorable group effects on pain intensity and opioid dosage may be partly because of the limited number of subjects per group. Because the present authors' sample size calculation was based on physiological parameters reported in previous studies, the lack of differences in the other parameters may be because of the lack of power to determine the effects on pain and opioid usage.
Efficacy of music
The efficacy of music is influenced by how well the listeners like the music (Good, 1996). Although the subjects' music preferences were not monitored in the present authors' study, all subjects expressed that they liked the music provided. Although the same music pieces used in the present study have been tested for their effectiveness in relaxing older adults with dementia (Ho et al., 2011), ICU patients (Su et al., 2013), and cancer patient caregivers (Lai, Li, & Lee, 2012), the effectiveness of these music selections on the physiological parameters of patients undergoing TKR is not entirely clear. Thus, personal music preferences may need to be considered when selecting music intervention for these subjects. Although the independent effects of individualized music preferences were not investigated in this study, this was the first study to provide evidence for the cumulative effect of music on systolic blood pressure during POR. Perhaps the subjects' systolic blood pressure was more responsive to increased and decreased stress than the other physiological measures. The present study results provided evidence for nurses that the use of soothing music reduces patients' anxiety during POR.
Based on a psychophysiological theory derived from the published work (Lai & Good, 2002), certain types of music induce relaxation and the present authors' study only partially supported the psychophysiological theory of music. Therefore, more reliable evidence of effects of music on physiological parameters, pain intensity, and opioid dosage needs to be established.
Limitations and implications
All the present authors' subjects were undergoing TKR; the homogeneity of the subjects should have improved the study power. However, the events and care process in the operating room may affect the physiological status of the patients, thus this perspective needs to be further studied. The present study added to evidence-based approaches for nurses to improve the respiratory rate and systolic blood pressure of patients, although its findings should be interpreted with caution. Because of concerns regarding nosocomial infection control by the present authors' surgical team, intraoperative measurements were not performed to provide a more comprehensive understanding of the effects of music during the entirety of perioperative phases. Future studies should include the intraoperative phase and the timing of the music intervention to provide more comprehensive results. Music preference plays a large role because people generally like familiar music and dislike unfamiliar music. Further research will be necessary to investigate whether the likeability of music is related to psychophysiological parameters, pain intensity, and opioid dosage in this patient population. Another caution of the interpretation of the study is the numbers of the participants. However, this study applied a multiphase influence using repeated assessments of physiological responses, which helped limit the sample size needed. However, in practice, healthcare professionals can encourage surgical patients to listen to soothing music as a self-care intervention to release stress.
Conclusion
The present study examined the effects of listening to music on preoperative and postoperative physiological parameters as well as on postoperative pain and opioid dosage in patients undergoing TKR. According to a psychophysiological theory, music improves relaxation. This study demonstrated that listening to soothing music that was broadcast for 30 min while awaiting TKR surgery decreased the respiratory rates of these subjects as compared with the control group that did not listen to music. A beneficial effect of listening to music was also manifested by a consistent decrease in systolic blood pressure in the music group during POR. However, listening to soothing music did not reduce postoperative pain or the amount of analgesic medications required to manage pain for patients undergoing TKR. Although the mechanisms of music need to be further studied, the results of this study provide evidence to support the use of music in the surgical waiting and POR areas before and after TKR.
Conflict of interest
The authors have no financial conflicts of interest to disclose.
Author contribution
H. J., T. Y., Y. M., C. Y., and H. L. contributed to the conception and design of this study; H. L., C. Y., and H. J. performed the statistical analysis and drafted the manuscript; and H. L. and C. Y. critically reviewed the manuscript and supervised the whole study process. All authors read and approved the final manuscript.
