Effect of Virtual Reality on Pain Management and Opioid Use Among Hospitalized Patients After Head and Neck Surgery

This randomized clinical trial evaluates the use of virtual reality on postoperative pain management in hospitalized patients after head and neck surgery.

Key Points

Question

Can use of a virtual reality (VR) experience facilitate pain control among hospitalized patients after head and neck surgery?

Findings

This randomized clinical trial of 30 patients hospitalized after head and neck surgery showed that use of VR reduced postintervention pain scores more than a control intervention. Use of VR was also associated with meaningful 4- and 8-hour reductions in postintervention opioid use compared with preintervention opioid use.

Meaning

Virtual reality may be a useful nonpharmacologic adjunct for postoperative pain control among patients after head and neck surgery.

Abstract

Importance

Optimal postoperative pain management is challenging. Virtual reality (VR) provides immersive, 3-dimensional experiences that may improve pain control and reduce reliance on pharmacologic pain management.

Objective

To evaluate use of VR on postoperative pain management after head and neck surgery.

Design, Setting, and Participants

This prospective, pilot randomized clinical trial was conducted at Oregon Health & Science University from July 2020 to October 2021 and included patients hospitalized after major head and neck surgery.

Interventions

Similar 15-minute interactive gaming experiences (Angry Birds) using an Oculus Quest VR headset (VR intervention) or a handheld smartphone device (control).

Main Outcomes and Measures

The primary outcome was postintervention pain reduction. Pain scores were obtained preintervention, immediately after intervention, and then hourly for 4 hours. Secondary outcomes included changes in opioid use, measured as milligram morphine equivalents (MMEs), and patient experiences with their intervention using 5-point Likert scales.

Results

Of the 30 patients randomized for inclusion, the final population included 14 patients in the VR cohort and 15 patients in the control cohort; the majority of patients were male (26 of 29 [90%]), and the mean (SD) age was 58.3 (13.8) years. After outlier removal, there were clinically meaningful reductions in postintervention pain among patients in the VR group immediately after intervention (mean difference, −1.42; 95% CI, −2.15 to −0.70; d = 1.50), at 1 hour (mean difference, −0.86; 95% CI, −1.90 to 0.14; d = 0.67), 2 hours (mean difference, −1.07; 95% CI, −2.30 to 0.14; d = 0.69), and 3 hours (mean difference, −1.36; 95% CI, −2.80 to 0.13; d = 0.71) compared with patients in the control group. Patients in the VR group also demonstrated reductions in 4-hour postintervention opioid use compared with 4-hour preintervention opioid use (mean difference, −9.10 MME; 95% CI, −15.00 to −1.27 MME; d = 0.90) and 8-hour postintervention opioid use compared with 8-hour preintervention opioid use (mean difference, −14.00 MME; 95% CI, −25.60 to −2.40 MME; d = 0.94). There were no meaningful differences in subjective patient experiences with their respective interventions.

Conclusions and Relevance

In this randomized clinical trial, VR reduced pain scores and opioid use compared with a control intervention. Virtual reality may be a useful adjunct for postoperative pain management after head and neck surgery.

Trial Registration

ClinicalTrials.gov Identifier: NCT04464304

Introduction

Postoperative pain in patients undergoing head and neck surgery is a major factor when aiming to improve functional recovery, reduce hospital stays, and increase quality of life.¹ Major interventions, including bedside procedures, can also affect the experience of pain. Narcotic analgesics are an important part of postoperative pain management, but these have risks, including nausea, sedation, respiratory depression, and possible dependence.² Current pain-control strategies are often inconsistent and may be frequently inadequate.³ Multimodal analgesia is an area of exploration for optimal pain control, but patients may have coexisting medical conditions such as kidney or liver disease that can limit use of certain medications.^3,4,5,6 Nonpharmacologic measures for postoperative pain control may provide novel and cost-effective strategies to confront this complex issue.

Virtual reality (VR) provides experiences in a 3-dimensional, immersive environment and has been used in a variety of health care scenarios, including treatment of anxiety, physical rehabilitation, distraction during procedures, education, and pain control.^7,8,9,10,11 Virtual reality offers a distraction from pain by providing a multisensory environment of auditory, visual, and proprioceptive stimuli. In the general inpatient setting, VR use has shown to be feasible and associated with pain reduction compared with a 2-dimensional (2D) control.^12,13,14 However, there has been limited study of VR in the setting of postoperative pain management.^{15,16,17,18,19} There are different types of VR content that may influence the sensory experience.²⁰ Passive VR experiences may include observing a virtual 3-dimensional environment such as nature scenes or popular destinations. Interactive VR experiences may provide an increased element of engagement compared with passive VR experiences through participant-driven exploration of the VR environment or participation in virtual games.²¹ The element of patient game participation in health care has received increased attention, particularly in pain distraction.^22,23

The purpose of this randomized clinical trial was to evaluate the effect of a VR active game compared with a similar 2D game on postoperative pain among hospitalized patients after head and neck surgery. The primary aim was to evaluate changes in subjective pain scores after intervention use. The secondary aims were to evaluate changes in opioid use and subjective patient experiences with their intervention. We hypothesized that patients using VR would have larger postintervention pain score reductions compared with the control group.

Methods

Study Design

This was a randomized clinical pilot trial at Oregon Health & Science University, a tertiary academic hospital, for hospitalized patients undergoing head and neck surgery from July 2020 to October 2021 (Supplement 1). The study was approved by the Oregon Health & Science University Institutional Review Board and followed Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines.

Participant Recruitment

Patients who required hospitalization after head and neck surgery were asked to participate if they had moderate to severe pain, defined by an average pain score on their electronic medical record of 3 or greater out of 10 during the 24 hours prior to screening for study eligibility.^11,12 Written informed consent was obtained from all patients prior to their participation in the study. Patients were included if they were able to understand the goals of the study and provide informed consent, were 18 years or older, and were English speaking. Patients were excluded if they were in the intensive care unit; had symptoms concerning for an active respiratory tract infection; had active eye discharge or active nausea or emesis; had a history of nausea, vertigo, or motion sickness; had a history of seizure, epilepsy, or hypersensitivity to flashing light; or had postoperative reconstruction or wound care conditions that prevented safe or comfortable use of the VR headset.¹² Participant inclusion is shown in Figure 1.

Figure 1. CONSORT Diagram.

VR indicates virtual reality.

Demographics and Clinical Characteristics

Demographics included age, self-reported sex, comorbidities measured using the Charlson Comorbidity Index, and preoperative opioid use. Surgical characteristics included diagnosis and surgical procedures. Time to intervention after most recent opioid use was documented. Postoperative analgesic use was recorded, and opioid use was measured as milligram morphine equivalents (MME) for standardization.

Interventions

Participants were randomly assigned using software-generated block randomization to the VR intervention or the control group.²⁴ Interventions were provided to patients by a coauthor (V.C.P.). The VR headset used was the Oculus Quest (Facebook, Inc), which provides the VR video and audio and comes with hand controllers. A silicon pad was added on each headset before use, disposable sanitary covers were used between patients, and disposable head caps, eye masks, and face masks (personal protective equipment) were also placed on patients before use to minimize direct device contact. Patients in the VR group participated in a 15-minute interactive VR gaming experience (Angry Birds VR: Isle of Pigs [Resolution Games]). Patients participated in this game using controllers while sitting in bed. This content is interruption free and includes music, sound, and animation. Patients in the control group played Angry Birds 2 (Rovio Entertainment) for 15 minutes on a handheld smartphone device (iPhone XR [Apple Inc]) while sitting in bed, with audio played through the device speaker. Although these 2 games had differences based on platform used, they were similar in terms of themes, animations, audio, and general game play. Interventions were not provided on the day of surgery.

Outcome Measures

The primary outcome was change in pain score after intervention use. Prior to the intervention, patients self-reported pain on an 11-point numeric rating scale (NRS; from 0-10), an easy to administer and validated pain-assessment modality.²⁵ Patients repeated the NRS assessment immediately after intervention use and then hourly for 4 hours after the intervention. The value for the change in NRS assessment that is clinically meaningful has not been determined. Secondary outcomes included changes in opioid use and patient satisfaction with their intervention. Pharmaceutical analgesic use, including opioid and nonopioid pain medication, was documented from the electronic medical record in the 8-hour preintervention and 8-hour postintervention periods. After intervention use, participants were asked to complete responses for 2 questions regarding their experience and desire to use their audiovisual intervention in the future. These were rated on 5-point Likert scales with the following answer choices: strongly disagree, disagree, neither agree nor disagree, agree, and strongly agree. Adverse effects were also evaluated.

Statistical Analysis

A formal sample-size calculation was not performed because this was a pilot study and sufficient comparative published data were not available. Statistical data analyses were performed using SPSS Statistics for Windows, version 26 (IBM). Mean values were reported with SDs, and median values were reported with IQRs (reported as a single value [75th percentile value minus 25th percentile value]). Additionally, 95% CIs were reported, and Cohen d effect sizes were reported as appropriate and interpreted as small (d = 0.2), medium (d = 0.5), and large (d = 0.8).²⁶ P values were the result of 2-sided tests, and P < .05 was determined to be statistically significant.

Results

Study Population

There were 30 patients randomized for inclusion in the study. One patient was unable to complete the VR intervention owing to discomfort in wearing the associated personal protective equipment. The final population included 14 patients in the VR cohort and 15 patients in the control cohort (Table). The majority of patients were male (26 of 29 [90%]), and mean (SD) postoperative day of intervention was 3.9 (3.4) days after surgery. The patients’ primary surgical sites were the oral cavity (12 of 29 [42%]) or oropharynx (11 of 29 [38%]), and the majority of secondary surgical sites involved the neck (20 of 29 [69%]). Of the 29 patients, 13 (45%) underwent tracheostomy, and 15 (52%) underwent free-tissue reconstruction. One patient underwent 2 free-tissue reconstruction procedures during the same surgery. Eleven of the 29 (38%) patients received opioids preoperatively, and all patients had opioids available to them for postoperative analgesia. The mean (SD) age was 58.3 (13.8) years, Charlson Comorbidity Index score was 4.1 (2.3), and time to intervention after most recent opioid use was 3.1 (3.7) hours. The differences in the VR and control groups were obtained for the time to intervention after most recent opioid use (median [IQR], 1.5 [2.4] hours vs 2.0 [4.2] hours; median difference, −0.04 hours; 95% CI, −2.00 to 1.00 hours), preintervention pain score (median [IQR], 5.0 [3.0] vs 4.0 [2.0]; median difference, 1.00; 95% CI, −1.00 to 2.00), total 4-hour preintervention opioid use (median [IQR], 21.8 [24.5] MME vs 7.6 [22.5] MME; median difference, 7.50 MME; 95% CI, 0.00-24.00 MME), and total 8-hour preintervention opioid use (median [IQR], 43.5 [37.8] MME vs 22.5 [37.5] MME; median difference, 15.00 MME; 95% CI, −1.50 to 30.00). The differences in the median values for the 2 groups among these variables revealed no clinically meaningful differences. Finally, there were no clinically meaningful differences in preintervention use of acetaminophen, ibuprofen, celecoxib, or gabapentin between the 2 groups.

Table. Patient Demographics and Clinical Characteristics.

Variable	Intervention, No. (%)
	VR (n = 14)	Control (n = 15)
Age, mean (SD), y	57.4 (13.2)	59.2 (14.8)
Sex
Female	1 (7)	2 (13)
Male	13 (93)	13 (87)
CCI score, mean (SD)	4.6 (3.0)	3.6 (1.3)
Preoperative opioid use	6 (43)	5 (33)
Primary surgical sites
Face	0	0
Oral cavity	4 (29)	8 (53)
Oropharynx	8 (57)	3 (20)
Larynx	1 (7)	2 (13)
Neck	2 (14)	2 (13)
Secondary surgical sites
None	3 (21)	2 (13)
Face	0	0
Oral cavity	1 (7)	1 (7)
Oropharynx	1 (7)	1 (7)
Larynx	0	0
Neck	8 (57)	12 (80)
Chest	1 (7)	0
Free-tissue reconstruction	5 (36)	10 (67)
Free-tissue donor site
Radial forearm	2 (14)	3 (20)
Anterolateral thigh	3 (21)	5 (33)
Fibula	0	2 (13)
Latissimus dorsi	0	1 (7)
Tracheostomy	6 (43)	7 (47)
Preintervention pain score, mean (SD)	5.1 (2.0)	4.4 (1.6)
Preintervention opioid use, median (range), MME
4 h	21.8 (0.0-78.5)	7.6 (0.0-22.5)
8 h	43.5 (0.0-96.1)	22.5 (0.0-45.0)

Abbreviations: CCI, Charlson Comorbidity Index; MME, milligram morphine equivalents; VR, virtual reality.

Change in Pain Score After Intervention

There were no meaningful differences in preintervention pain scores between the 2 groups (mean difference, 0.67; 95% CI, −0.67 to 2.00; d = 0.38). Graphical analysis identified an extreme control group outlier with pain reduction from 7 to 0 immediately after intervention use, identified to be more than 3 times the IQR below the first quartile. After outlier exclusion, among these 28 patients there was a meaningful difference in pain reduction among the VR group vs the control group immediately after intervention use (mean difference, −1.42; 95% CI, −2.15 to −0.70; d = 1.50; Figure 2). Patients in the VR group also had medium and clinically meaningful pain score reductions at 1 hour (mean difference, −0.86; 95% CI, −1.90 to 0.14; d = 0.67), 2 hours (mean difference, −1.07; 95% CI, −2.30 to 0.14; d = 0.69), and 3 hours (mean difference, −1.36; 95% CI, −2.80 to 0.13; d = 0.71). The difference in pain score at 4 hours was considerably smaller than the differences at the other time points (mean difference, −0.68; 95% CI, −2.10 to 0.71; d = 0.38).

Figure 2. Change in Pain Score Across Different Postintervention Time Periods Compared With the Baseline Preintervention Pain Score Among Patients in the Virtual Reality (VR) Group or Control Group.

Opioid Use

Figure 3 shows trends in reduction of 4- and 8-hour postintervention opioid use compared with preintervention opioid use. There was a meaningful decrease in 4-hour postintervention opioid use compared with 4-hour preintervention opioid use among patients in the VR group (mean difference, −9.10 MME; 95% CI, −15.00 to −1.27 MME; d = 0.90). One patient did not have complete data available for 8-hour preintervention use owing to use of patient-controlled analgesia, which did not allow opioid use to be time stamped for this period. Of the remaining patients, there was a similar meaningful decrease in 8-hour postintervention opioid use compared with 8-hour preintervention opioid use among patients in the VR group (mean difference, −14.00 MME; 95% CI, −25.60 to −2.40 MME; d = 0.94).

Figure 3. Reduction in Opioid Use Postintervention vs Preintervention Among Patients in the Virtual Reality (VR) Group or Control Group.

A, Four hours postintervention compared with 4 hours preintervention (mean difference, −9.10 milligram morphine equivalents [MME]; 95% CI, −15.00 to −1.27 MME; d = 0.90). B, Eight hours postintervention compared with 8 hours preintervention (mean difference, −14.00 MME; 95% CI, −25.60 to −2.40 MME; d = 0.94).

In both panels, the top line of the box indicates the third quartile; the bottom line, the first quartile; the line within the box, the median value; the whiskers, the range of values from minimum to maximum; dots, outliers; and diamonds, the mean value.

Patient Experience

There were no clinically meaningful differences in the median score for responses to the question “I enjoyed my audiovisual experience” between the VR and control groups (median [IQR], 4.5 [1.0] vs 4.0 [1.0]; median difference, 0.00; 95% CI, 0.00-1.00). All 14 patients in the VR group responded “agree” (7 of 14 patients) or “strongly agree” (7 of 14 patients). Among the 15 patients in the control group, 1 patient responded “strongly disagree,” 2 patients responded “neither agree nor disagree,” and 12 patients were evenly divided into responses of “agree” or “strongly agree.” Additionally, there were no clinically meaningful differences in the median value for responses to the question “I would like to see my audiovisual experience used more often in my healthcare” between the VR and control groups (median [IQR], 5.0 [1.0] vs 4.0 [2.0]; median difference, 0.00; 95% CI, 0.00-1.00): 2 patients in the VR group reported “neither agree nor disagree,” 4 patients reported “agree,” and 8 patients reported “strongly agree,” and among patients in the control group, 1 patient responded “strongly disagree,” 4 patients responded “neither agree nor disagree,” and 10 patients were evenly divided into responses of “agree” or “strongly agree.”

Adverse Events

There were no adverse events related to intervention use noted in either group. One patient in the VR group could not complete intervention use owing to discomfort associated with wearing a mask, one of the pieces of personal protective equipment worn during device use.

Discussion

In this study, use of a VR intervention was associated with meaningful reductions in postintervention pain and opioid use compared with a similar 2D intervention. Optimal postoperative pain management is challenging, and multimodal approaches are increasingly used to act on different areas of the pain pathway and reduce reliance on opioid analgesics.²⁷ However, alternatives to opioid medications such as nonsteroidal anti-inflammatory drugs, acetaminophen, and gabapentin still have risks and limitations owing to various comorbid conditions. Thus, there is a role for nonpharmacologic adjuncts to pain control. The application of VR in health care is steadily expanding, including procedural distraction, anxiety reduction, and pain control.^{12,14,17,18,28} One theory is that diversion of attention to pain through distraction may provide a slower response to pain signals.²⁹ The immersive environment may explain the increased effectiveness of VR in pain attenuation compared with traditional forms of distraction such as 2D videos.¹² Furthermore, while analgesics interrupt pathways that sense pain, VR instead acts both indirectly and directly on pain signals and perception through multisensory awareness, emotion, and concentration.^29,30,31

Tashjian et al¹² and Spiegel et al¹¹ previously reported on the use of VR in pain control among hospitalized patients compared with a 2D control. However, the VR intervention included an active gaming experience, while the control consisted of viewing a 2D video.^11,12 While these results are encouraging, the inclusion of an active game in the VR group may be a contributing factor to pain reduction by heightening the distraction experience compared with a passive viewing experience.¹² The current study used comparable experiences in the VR and control groups because both groups participated in a similar game (Angry Birds), using similar strategies to progress through the game and experiencing similar animations and audio. The present findings provide support for use of VR as a more effective means of pain reduction than a 2D intervention. Future studies may consider evaluating postintervention outcomes at more frequent intervals to further assess the duration of postintervention benefits. Additionally, patients in the control group reported a small reduction in immediate postintervention pain scores and overall high enjoyment with their intervention. This suggests that for patients who may not be able to tolerate use of a VR headset, there may be a role for 2D gaming interventions as an adjunct to pain control and to improve patient quality of life.

While there is a rise in studies evaluating the use of VR for procedural anxiety and pain control, only a few studies have aimed to evaluate the use of VR after surgery.^28,32,33 Specht et al¹⁷ performed a randomized clinical trial and found that VR was associated with reduced postoperative pain scores compared with tablet use among pediatric patients. Olbrecht et al^19,34 also found VR to reduce pain and anxiety among pediatric patients after surgery. A small, noncontrolled study by Mosso-Vázquez et al¹⁶ reported reduced postoperative pain as well as psychological stress among patients using VR after cardiac surgery. Ding et al³⁵ published a systematic review and meta-analysis of 8 randomized clinical trials using VR for postoperative pain after surgeries, including hemorrhoidectomy, dental surgery, craniotomy or spine surgery, episiotomy repair, and knee surgery. They found that perioperative use of VR resulted in a greater reduction in postoperative pain, but none of these studies evaluated outcomes on opioid use.³⁵ To our knowledge, the present study is the first to evaluate use of VR for postoperative pain control among hospitalized patients after head and neck surgery, and it is the first study to demonstrate a reduction in opioid use associated with patients using VR compared with a control intervention. Use of VR was associated with a reduction in opioid use up to 8 hours. This is a meaningful finding that may encourage larger adoption and application of VR experiences for management of postoperative pain.²⁷

This study identified a long-term reduction in post-VR opioid use, as well as medium and clinically meaningful pain score reductions up to 3 hours post-VR use by evaluation of the Cohen d effect size. However, owing to the small sample size, the 95% CIs around the mean differences in pain scores were imprecise for the time periods of 1 to 3 hours after intervention use and included the null value of zero. Thus, no definitive conclusions can be made about the effect of VR compared with the control group for these prolonged pain score changes. Nevertheless, VR use appears to deliver clinically meaningful reductions in pain scores up to 3 hours after use, as measured by Cohen d. There did not appear to be a meaningful pain score difference 4 hours postintervention use, but post-VR opioid use appeared reduced for up to 8 hours. There may be other factors apart from subjective pain intensity that may influence opioid use, such as psychiatric comorbidities.³⁶ Use of VR has shown benefit in supporting treatment of anxiety and depression, so it is possible that there may have been therapeutic effect on causes aside from subjective pain associated with the reduction in opioid use.^37,38 These findings suggest that there may be a role for inpatient VR use during nursing or clinician bedside procedures, such as local wound care, or as an adjunct while patients await pharmacologic pain control.^30,38,39,40 These findings also suggest the possible benefit of incorporation of VR experiences in enhanced recovery after surgery protocols to facilitate pain control, reduce opioid use, and improve patient quality of life in the hospital. Furthermore, the direct costs of using VR in this study were overall low because the headset could be reused, and we used a headset associated with a 1-time cost of approximately $400. Larger-scale studies are needed to provide a more comprehensive evaluation of the clinical use of VR in postoperative recovery.

Patients who undergo head and neck surgery pose a unique challenge to the use of VR owing to incisions or reconstruction that may preclude wearing of the VR headset and prevent the comfortable and safe use of VR in the postoperative period. Thus, some patients were excluded from the current study owing to surgical-site considerations, including patients undergoing parotidectomy and excision of cutaneous facial or scalp malignant tumors with reconstruction. While direct pressure from the headset may adversely affect local wounds or cause discomfort, given the limited duration of periodic VR use at a time it is unclear what the true hazards are and how long after surgery these risks last. Future studies are necessary to better understand risks of using VR headsets in the postoperative period and to better define inclusion criteria for patients with head and neck cancer undergoing surgery.

Limitations

While there are many strengths to this study, there are several limitations. Because this study represented a pilot study for the use of postoperative VR among patients hospitalized after head and neck surgery, the sample size was small. This small sample size is likely associated with imprecision in the estimates for differences in pain score changes between the VR and control groups, as noted by the wide range of 95% CIs that included the value zero, which limits the ability to make definitive conclusions. A larger number of patients in the control group underwent free-tissue reconstruction, which is associated with additional surgical sites, and this possibly contributed to differences in pain scores and may have been avoided with a larger sample size. However, there were no clinically meaningful differences in baseline pain scores between the 2 groups, suggesting higher preintervention pain was not observed among patients in the control group. Owing to the type of interventions used, blinding could not be performed, and there are limitations to allocation concealment. There were approximately 23% of patients approached for recruitment who declined participation, and enrollment may have selected for patients more likely to enjoy gaming experiences and receive benefit from use of these interventions. Novelty of VR use among patients without a previous exposure to this technology may also potentially influence the findings from this study, and future studies would benefit from evaluating outcomes after repeated use of a VR platform. This study attempted to provide similar experiences in VR and using a 2D intervention, but there are differences in these experiences that may be associated with the findings, such as use of hand controllers with the VR platform. Additionally, timing of intervention use was not standardized.

Conclusions

In this randomized clinical trial, use of VR after head and neck surgery reduced pain scores up to 3 hours after intervention and opioid use up to 8 hours after intervention when compared with a control intervention. Use of VR also received high ratings of enjoyment and desire to be used by patients in their future health care. Virtual reality may be a useful adjunct in pain control after head and neck surgery.

Supplement 1.

Trial Protocol

Supplement 2.

Data Sharing Statement