The study recruited 51 children and adolescents who underwent surgery and were followed by the Acute Pain Service at Cincinnati Children’s Hospital Medical Center between July 2019 and March 2020. The study was approved by the Institutional Review Board (IRB #2018-2892) and conducted per the rules and regulations for ethical research. It was registered with ClinicalTrials.gov on September 21, 2020 (NCT04556747). A written consent (and assent for patients >11 years) was obtained from all participants and the study adhered to the CONSORT (Consolidated Standards of Reporting Trials) guidelines. The patients were recruited and enrolled in the study after surgery following hospital admission, and they were not compensated for participation.

Patients were suitable for recruitment if the Acute Pain Service was managing their pain. The patients managed by the Acute Pain Service experience high postoperative pain unmanageable solely by the surgery team. These patients necessitate multimodal analgesia and management with patient-controlled analgesia (PCA) and regional or neuraxial analgesia.

The primary outcome was the change in pain intensity after VR-GR. Secondary outcome measures included changes in pain unpleasantness and anxiety.

All patients used the Starlight Xperience VR device, a commercially available headset supplied by the Starlight Children’s Foundation (Figure 2). It is a customized version of the Lenovo Mirage Solo with Daydream VR headset. The integrated headphones deliver audio content, and the patients interact and navigate within the VR environment using head movements and a handheld controller. The patients used the “Mindful Aurora” guided relaxation–based application to learn slow breathing and relaxation, an application developed by the Stanford CHARIOT (Childhood Anxiety Reduction through Innovation and Technology) program (Figure 3). The users were “transported” to an alpine meadow in the virtual world, where a 10-minute relaxation narrative teaches focused, slow, and paced breathing.

The patient visits occurred on postoperative day (POD) 1 or 2. After determining eligibility and obtaining consent, the patients completed the PCS-C and CASI questionnaires and rated their baseline pain intensity, pain unpleasantness, and anxiety levels on the NRS.

Following these assessments, the patients were oriented to the VR headset and given a tutorial on the device application. They were instructed to remove the headset for discomfort, nausea, or dizziness. After completing the 10-minute session (based on the standard duration of relaxation sessions), the VR device was removed. Pain intensity, pain unpleasantness, and anxiety were recorded immediately, at 15 minutes, and at 30 minutes after the experience. The patients and parents then completed the experience questionnaires.

All statistical analyses were performed using SAS 9.4 (SAS Institute). A P value of .05 was the cutoff value for statistical significance. Additionally, statistical significance with the Bonferroni adjustment for multiple comparisons for the primary outcome (change from the baseline pain intensity at three time-points after VR-GR) was assessed. The first-order autoregressive, AR(1), covariance structure was used in all the mixed-effects models. The missing data were examined, and all available data were used in statistical analyses.

We enrolled 51 patients over 8 months. All patients completed pain and anxiety assessments at baseline and immediately following VR-GR; 100% (n=51) completed the pain assessment at 15 minutes, and 98% (n=50) completed the anxiety assessment at 15 minutes; and 88% (n=45) completed the assessments at 30 minutes following VR-GR. The patients did not receive any analgesic medications during participation in the study. Missing data resulted from limitations in the clinical environment, including patients undergoing imaging studies, receiving care from the care team, or falling asleep.

The patients were primarily adolescent, male, and Caucasian (Table 1). Of the 51 patients recruited, 19 (37.3%) underwent abdominal surgery, 21 (41.2%) underwent Nuss repair of pectus excavatum or chest surgery, and 11 (21.6%) underwent orthopedic procedures (such as posterior spinal fusion or major hip surgery). Half of the recruited patients were ASA status I/II or III/IV. The patients reported moderate pain intensity, unpleasantness, and mild anxiety levels before the VR-GR (Table 1).

Patients had moderate pain catastrophizing and average anxiety sensitivity [21]. Furthermore, higher PCS-C scores were associated with higher baseline anxiety (Spearman ρ=0.41, P<.001) (Table 2).

Pain catastrophizing and anxiety sensitivity were not associated with changes in pain intensity or pain unpleasantness.

VR-GR was associated with a small reduction in pain intensity (Figure 4A, Table 3). The Wilcoxon signed-rank test showed that pain intensity decreased immediately (median –1.0, IQR –2.0 to 0, P<.001) following the VR-GR session and remained significant at 15 minutes (median 0, IQR –1.0 to 0.50, P=.03), and at 30 minutes (median 0, IQR –1.5 to 0, P=.02) (Table 3).

We used a series of adjusted, mixed-effects models to observe the correlation between psychological factors and changes in the baseline outcomes. It was observed that anxiety sensitivity levels were associated with greater reductions in pain intensity (β=–.06, SE 0.03, P=.04) and pain unpleasantness (β=–.09, SE 0.04, P=.01). Pain catastrophizing was not associated with changes in pain and anxiety following VR-GR (Table 4).

When assessing the association between changes from the baseline and covariates, we found that the Caucasian patients had a smaller decrease in pain unpleasantness from the baseline than the non-Caucasian participants—difference in least squares mean (LSM=1.20, 95% CI 0.11-2.28, P=.03). Additionally, the older participants had a smaller decrease in pain intensity (β=.12, SE 0.05, P=.03) than the younger patients. No other variables (ASA, POD, or race) were associated with changes in outcomes.

The patients reported a very positive VR experience. Overall, 96% of children would recommend VR to friends and family, and most (n=45, 88%) believed they felt “calmer and less anxious after having used VR” and that VR “made it easier for (them) to tolerate (their) pain.” Parents reported a similar positive overall experience when asked the same questions. The parents (n=44, 100%) who completed the questionnaire would recommend VR; 93% (n=41) believed that VR made their child calm, and 84% (n=37) believed that VR helped their child tolerate pain. No patient experienced any self-reported side effects during the VR session.

VR-GR may be beneficial in pediatric postoperative pain management. Our pilot study assessed the association between a single VR-GR session delivered after surgery and changes in pain intensity, pain unpleasantness, and anxiety in children and adolescents with acute pain following surgery. The most significant changes were observed for pain unpleasantness, with lesser reductions in pain intensity and anxiety. More considerable reductions in pain intensity and unpleasantness after VR-GR were associated with higher anxiety sensitivity. Pain catastrophizing did not appear to be associated with changes in these outcomes. The qualitative assessments suggested that the patients had positive experiences with VR-GR with no reported side effects.

Although pain management strategies increasingly incorporate multimodal analgesia, the percentage of patients experiencing severe postoperative pain did not change in the last 20 years [2]. Besides prolonged pain, children are also at risk of persistent opioid use after surgery; opioid exposure after surgery can confer nearly a 50-fold increase in opioid use [3]. It was reported that more than 25% of children with chronic pain transitioned into chronic opioid use after using opioids for surgery-associated pain [6].

Few studies have used VR for acute postoperative pain in adults [23], and none have used it to help manage pain in children after surgery. Most VR studies use distraction–based VR (VR-D) to reduce pain temporarily by redirecting attention [7,9,10,13]. Without VR, distraction alone provides little pain management benefit [24], with no significant lasting impact on pain [15]. The improved efficacy of VR-D to temporarily reduce pain versus distraction alone may be due to the immersion with VR [23]. Although VR-D is more effective than distraction alone, its use is limited. Incorporating VR with other pain-reducing strategies, like guided relaxation, may promote sustained pain relief beyond the temporary impact of distraction [16].

Mind-body therapies, like relaxation and guided imagery, decrease anxiety and help manage pain in children undergoing surgery [17]. Unfortunately, their integration into postoperative clinical care is fraught with challenges, including access to care, high cost, and provider availability. As a result, the use of these therapies is minimal even at a tertiary care institution like ours, besides their inability to engage children. Using VR to deliver these therapies can increase accessibility and enhance acceptability in children versus methods without VR, making this therapy more engaging and relevant. Importantly, VR and mind-body therapies have very little side effects, a significant advantage over traditional pharmacologic interventions.

This study was the first to integrate a mind-body therapy with VR to help pediatric postoperative pain. The study’s goals were to preliminarily assess the use of VR-GR to decrease acute postoperative pain and anxiety and determine whether pain catastrophizing and anxiety sensitivity influence the VR-GR effect in a broad pediatric population. The “Mindful Aurora” application taught patients to slow down their breathing and relax. A single VR-GR session was associated with immediate, acute changes in pain intensity, pain unpleasantness, and anxiety. Reductions in pain unpleasantness were observed at all intervals (immediately, 15 minutes, and 30 minutes following VR-GR). Anxiety reduced immediately and at 15 minutes following VR-GR. An association was observed between higher CASI scores and greater reductions in pain intensity and unpleasantness following VR-GR; however, such an association was absent with pain catastrophizing.

This study had several limitations. The study design did not include a control group, which restricted us from determining a causal relationship between the use of VR-GR and pain and anxiety reduction. It is possible that interaction with the study team could influence changes in pain and anxiety. Additionally, we did not collect data on analgesic use in the study population or standardize the timing of the postoperative visit. This study was designed to test the feasibility of the VR-GR technology, obtain pilot data in children hospitalized after surgery, and assess a possible association between VR-GR and pain and anxiety reduction. The results of this study support the need to conduct a randomized controlled efficacy trial comparing the use of VR-GR to active control, data collection on the timing and use of all analgesic medications, and standardizing the timing of the VR-GR sessions. The study used a single, short VR session in the postoperative period. Although we saw small changes in pain intensity, pain unpleasantness, and anxiety, these changes did not reach any clinical significance. Previous literature report that a reduction in pain intensity of 2 or more points on the NRS or a 30% reduction in pain is considered clinically significant [25]. We recognize that a single session is unlikely to produce a large and sustained effect, and we will pursue further clinical studies using repeated sessions to determine the impact on these outcomes. Finally, the potential bias when self-reporting pain scores could have produced lower self-reported pain and anxiety scores following the VR-GR session. This will be addressed in future studies by including a control group.

To summarize, our study demonstrates the successful use of VR-GR in children after surgery. A single VR-GR session was associated with transient decreases in pain and anxiety. Future research is needed to investigate the effect of VR-GR in reducing pain and anxiety in the postoperative setting compared to a control.