The World Health Organization defines preterm birth as a birth before 37 completed weeks of gestation or fewer than 259 days after the first day of the woman's last menstrual period. In 2020, an estimated 13.4 million babies were born preterm, accounting for 10% of all live births worldwide. In 2020, the preterm birth rate varied widely, from 4%–16% among different countries (1). In 2019, 5.30 million children younger than 5 years died—17% of whom died due to preterm birth complications (2)—and most survivors were hospitalized in the neonatal intensive care unit (NICU), with a length of hospitalization ranging from a few days to months. Hospitalized neonates experience acute episodic pain every day, including pain caused by heel lancing, naso- and endotracheal suctioning, venipuncture, arterial puncture, tracheal intubation, tracheal extubation, peripherally inserted central venous catheter (PICC) placement, retinopathy of prematurity (ROP) examination and other special invasive operations according to their individual clinical requirements (3). Most preterm infants require hospitalization in the NICU, where pain exposure is even more prevalent. According to a European multicenter study, many painful and stressful procedures were performed in the NICU, and the majority of them were performed without analgesic measures (4). According to the traditional definition, pain is an unpleasant sensory and emotional experience that leads to systemic physiological changes. However, this definition strongly relies on patients to describe their pain, establishing the primacy of self-reports as the “gold standard” for newborns (who are nonverbal) and proposing that pain in newborns be classified as acute episodic, acute recurrent, prolonged, persistent or chronic pain (5). Based on a global, four-stage consensus and validation, acute episodic pain is described as a painful response to a procedure or an event. An episode is defined as a single event or a sequence of events related to a procedure (6). Although newborns are unable to describe their pain, physiological and behavioral responses are difficult to ignore. The physiological changes include increased heart rate and blood pressure, decreased blood oxygen saturation, fluctuating cortisol levels, and tissue damage (7). The physiological changes include increased heart rate and blood pressure, decreased blood oxygen saturation, fluctuating cortisol levels, and tissue damage (5). In addition to erratic fluctuations in vital signs caused by painful stimuli, the long-term adverse effects of painful experiences are of concern. Recent studies have shown that exposure to persistent or recurrent pain or stress in the infancy period has long-term consequences that not only affect brain and behavioral development but also affect all organs, leading to serious morbidity and potential mortality (6, 7). Moreover, preterm infants who survive are more susceptible to prolonged negative effects throughout their lives, such as permanent changes in the endocrine and immune systems and impaired reactivity toward stressful events (8–10). According to the evidence-based clinical practice guidelines for the management of newborn pain, pain management in newborns requires a multidimensional approach that includes environmental, pharmacological and non-pharmacological measures (3). Although pharmacological treatments for neonates are established early and have relatively certain effects, the immature drug metabolism and related negative side effects, such as respiratory depression, apnea, bradycardia and hypotension, are difficult to overlook, especially for premature infants (11). Therefore, there has been growing interest in nonpharmacological analgesic options such as oral sucrose administration, nonnutritive sucking (NNS), kangaroo care, breastfeeding, and massage (11, 12).

Among all nonpharmacological analgesic options, music has long been used to enhance well-being and reduce pain and suffering. According to the definition put forward by the American Music Therapy Association, music therapy refers to the clinical and evidence-based use of music as a therapeutic intervention to accomplish individualized goals facilitated by a credentialed professional who has completed an approved music therapy program (13). Music therapy has been utilized as a strategy to ameliorate pain in various healthcare settings due to its noninvasiveness, cost effectiveness, reduction in anesthetic dosage and promising analgesic effects. However, the effects of music on pain management in preterm infants during painful procedures from the aspect of pain assessment have not been determined. For newborn patients, especially preterm infants, self-expression is not possible. Therefore, valid and reliable tools for assessing the degree of pain in newborns are needed (14). There are approximately 40 pain assessment instruments available to date, with various evaluation dimensions (15). The different assessments include different items and have been validated for patients with various kinds of pain and ages (16). The Premature Infant Pain Profile (PIPP) was developed by Stevens, Johnston, Petryshen, and Taddio to evaluate preterm infants' pain responses. This scoring system includes seven items: two items that describe baseline characteristics (gestational age and behavioral state); two items that are physiological indicators (heart rate and oxygen saturation); and three items that describe the neonate's facial expression (brow bulge, eye squeeze, and nasolabial furrow); the score for each item ranges from 0 to 3. A PIPP score of 0–6 points indicates mild pain, 7–12 points indicates moderate pain, and 13–21 points indicates severe pain (17). The PIPP is used to evaluate acute procedural pain in preterm and term infants (18). As the instrument takes contextual factors such as gestational age into account, the PIPP is recommended for its ability to judge pain in preterm and even extremely preterm neonates (16).

In addition, new studies with more detailed data and higher evidence levels have been published (19–22). Thus, we performed the present systematic review and meta-analysis of randomized controlled trials (RCTs) to investigate the effects of music on pain management in premature newborns. The results of this investigation may guide future decision-making regarding the use of music for managing pain in premature newborns.

This systematic review and meta-analysis followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement and the Cochrane Handbook for Systematic Reviews of Interventions. Ethical approval and patient consent were not required because all analyses were based on previously published studies.

Increasing evidence has shown that nonpharmacological analgesic methods are essential components of all effective interventions designed for managing procedural pain in preterm infants. However, there is no comprehensive understanding of the effect of music interventions on pain control and its physiological parameters. Considering that the PIPP has been validated to be a reliable assessment of pain in premature infants (28, 29), we performed this PIPP-centered meta-analysis to evaluate the analgesic effects of music and obtain higher-level evidence.

Our results showed that music had a nonsignificant effect on reducing the absolute change in the PIPP score during invasive procedures. However, due to the high heterogeneity (I²= 89%) and the lack of clarity regarding the specific duration of the intervention may have led to unreliable conclusions. Furthermore, music can significantly reduce pain during painful procedures, and the heterogeneity decreased precipitously from 86% to 0% after removing the study by Tang et al. (24) due to its use of a different methodology. In their study, compared to other studies, music was played for a longer duration (10 min) before the start of the invasive procedure. Moreover, music had a significant effect on pain reduction at the end of the invasive procedure according to the PIPP score after the painful procedure. By comparing the results obtained during and after the operation, the importance of the duration of music before the start of the procedure was reemphasized. Thus, future studies are needed to verify the analgesic effect of playing music for a long period before painful procedures and explore a related standard scheme. According to our results and the findings of previous studies, the analgesic effect of music is understood, and music interventions have great potential for application in clinical practice to ease the pain of preterm infants, reduce the duration of invasive procedures, decrease the use of narcotic and analgesic drugs, and avoid the long-term effects of repetitive, acute episodic pain. Questions for future research remain. The optimal duration and amplitude of music therapy for pain relief still have yet to be determined. There are no studies that aimed to define the best choice of music for music therapy. To our knowledge, the aspect of overstimulation has not been studied. If possible, we strongly recommend that more experts in the field of music therapy focus on and participate in preterm infant acute episodic pain alleviation projects, assisting clinicians and nurses in clinical decision making. Admittedly, we should emphasize that as a supplementary measure, the pain-relieving effect of music is limited because the pain category was unchanged.

Our results also indicated that music could significantly reduce pain levels during daily painful procedures. First, as an auditory stimulus, music not only provides a distraction but also has a potential effect on modulating perception by reducing delta-band activity in the cingulate gyrus and increasing gamma-band activity in somatosensory brain structures at different pain processing stages (30), thereby helping relieve painful responses. Moreover, studies have suggested that the analgesic effects of music can be partly explained by the endorphin release-induced reduction in the sympathetic nervous system response (31). Second, hearing music can physically reduce disturbing noises from the armamentarium and protect neonates against experiencing negative feelings (32, 33). Third, music can block pain pathways by facilitating sensorial saturation (34).

Improvement in physiological parameters is another indication of stress levels that reflect pain management outcomes. Several studies have demonstrated that music leads to an increase in oxygen saturation levels, as it reduces stress and has positive effects on both the heart rate and rhythm as well as on the respiratory rate (20, 35). Our results further validated this result, as music can increase the blood oxygen saturation of neonates during and after painful procedures. According to a recent systematic review, music/vocal interventions could reduce the heart rate in preterm infants during procedures, and the effect was greater after the procedures (36). Nevertheless, music did not have an obvious advantage in decreasing the peak heart rate in the present study. Blood oxygen saturation is affected by multiple factors, such as the whole respiratory and circulatory system. Therefore, the improvement in blood oxygen saturation may be more meaningful than the decrease in the peak heart rate.

As a recognized stress hormone, cortisol has always been a widely used indicator of stress levels (37). Studies have shown that music exposure can decrease stress hormone levels in preterm neonates in the NICU environment (38). Previous studies have confirmed that music decreases the levels of stress hormones before, during, and after surgery (39–41). However, only one of the four studies measured cortisol levels, and the cortisol concentration in the music intervention group was significantly lower than that in the control group, which further confirmed the stress-relieving effect of music. Therefore, additional studies are needed to further analyze cortisol levels to obtain a more visual representation of the influence of music on the degree of stress alleviation.

In this systematic review and meta-analysis, we first investigated the impact of music on pain management in preterm infants during acute episodic pain. This study has several limitations. First, the sample size in each study was relatively small. Second, the included studies involved diverse invasive operations and different evaluation time points, which contributed to a major part of the heterogeneity. Third, the duration and genre of the music and the approach of the music intervention largely varied in this study, and the related standard is worth investigating in future studies. Additionally, the selected studies involved various durations. Finally, missing and unpublished data also led to bias in the true impact of the music intervention. Thus, robust RCTs with large sample sizes and a standard protocol should be conducted in the future to obtain more accurate data and to verify our results.

In conclusion, this systematic review demonstrated that music is an effective intervention for relieving procedural pain in preterm infants. Our results indicated that music can reduce stress levels and improve blood oxygen saturation. Figure 7 shows the ideal outcome of music for procedural pain management in preterm infants. As a noninvasive, nonpharmaceutical, relatively low-cost intervention that can be applied at the bedside, music has an extremely impressive application in the NICU daily nursing routine. Due to the limitations of the present study, large-scale, prospective RCTs should be performed to validate these results.