A Systematic Review of Virtual Reality Therapeutics for Acute Pain Management

Abstract

Background:

Pain management is a fundamental human right. Numerous side effects, rising rates of chronic opioid use, and increased incidence of opioid-related deaths have led to greater use of non-pharmacologic alternatives for acute pain management. One such tool is virtual reality (VR), a non-pharmacologic, virtual platform for pain management delivery.

Aims:

The purpose of this systematic review is to examine the delivery and clinical efficacy of VR therapeutics for acute pain management in adults and identify practical considerations of VR deployment, as well as current gaps in the literature.

Methods:

A systematic review of all pertinent articles published between January 1^st, 2000 and August 1^st, 2020 was conducted according the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines.

Data Sources:

A search of PubMed, CINAHL, PsychINFO, Embase, Compendex, and Inspec was completed using MESH and keyword search terms related to acute pain and VR.

Results:

Twenty-three articles met final inclusion criteria and were included in this review. Studies utilized VR in a variety of settings for wound care, procedure-induced pain, physical or occupational therapy, dental treatment or generalized acute pain. The primary means by which included studies promoted analgesia was via distraction. Of the reviewed studies, 19 (83%) reported decreases in pain intensity while using VR compared to no VR use or to a non-VR group.

Conclusions:

This systematic review found VR to be an effective tool for acute pain management. Findings from this review also underscore the importance of addressing patient’s sense of presence and levels of immersion, interaction and interest when deploying VR. Future VR studies should consider incorporation of anxiety, presence, and VR side effect measures in addition to acute pain metrics.

Introduction

Pain management is a fundamental human right (Brennan et al., 2007; Mitra et al., 2018). Utilizing best practices for pain management helps to minimize its negative effects on healing and quality of life (Chou et al., 2016) and potentially prevents its transition to chronic pain (Meissner et al., 2015). Other factors like anxiety can exacerbate acute pain (AP), which usually occurs in the moment and for a short duration (Melzack & Katz, 2006). While pharmacologic management of AP – especially using opioid medications – is highly effective, (Angell, 1982; Chou et al., 2016) caution is also required. Numerous side effects (Ricardo Buenaventura et al., 2008), rising rates of chronic opioid use (Brummett et al., 2017) and increased incidence of opioid-related deaths (Alho et al., 2020; Gomes et al., 2018) have led to greater use of non-pharmacologic alternatives for AP management, such as virtual reality (VR).

Background

As an immersive digital platform, VR utilizes visual and auditory stimuli to give users a sense of physical presence in a virtual space (Howard, 2019; Slater & Wilbur, 1997; Yumurtacı, 2016). Delivering VR interventions necessitates considering four components: presence, immersion, interaction, and interest (Stark, 1995). For VR interventions that use distraction as a mechanism of action a higher degree of presence (the more users “feel” they have entered a virtual world) is linked to greater analgesic efficacy (Gutierrez-Martinez et al., 2010, 2011; Hoffman et al., 2004; Slater & Wilbur, 1997). Immersion dictates the degree of presence, via the quality of visual and auditory equipment (hardware) used (Slater & Wilbur, 1997; Yumurtacı, 2016). This is most effectively done using a wired (with linked equipment) or standalone headmounted device and head-tracking that shifts the digital field of view (FOV) – i.e. what they see – as the user looks, and “moves”, about the virtual environment (Howard, 2019; Stark, 1995). Both VR equipment and content (software) determine the level of interaction users have with the virtual environment. Active VR denotes the ability to interact with, move through, or change the virtual environment, whereas passive VR limits interaction and makes the user a reflexive viewer (Stark, 1995). Interest in the virtual environment holds these elements together and is positively correlated to presence (Hoffman et al., 2004). Active VR content increases presence and is also more interesting than passive content (Gutierrez-Maldonado et al., 2011; Gutierrez-Martinez et al., 2011; Hoffman et al., 2004, 2006; Wender et al., 2009). Laboratory studies have found that interventions with a high degree of presence, and that use active (versus passive) VR modalities, to significantly reduce AP (Gutierrez-Martinez et al., 2010, 2011; Hoffman et al., 2004, 2006; Wender et al., 2009). Though these components dictate how well VR (as a platform) is utilized, immersive interventions become therapeutic when clinicians deploy them in therapeutically appropriate ways (Levac & Galvin, 2013). Thus, it is also necessary to examine the content of therapeutics to determine their potential for therapeutic efficacy.

Systematic reviews have been performed in order to explore VR’s efficacy for burn and procedural pain (Chan et al., 2018; Scapin et al., 2018), among inpatient randomized controlled trials (Dascal et al., 2017), and generally for acute and chronic pain management (Mallari et al., 2019). Given the evolving nature of VR an updated and comprehensive review of clinical studies that use VR for AP management in adults is needed. Additionally, few address practical considerations for deployment and none have specifically addressed the components of VR interventions that are vital to ensuring an efficacious therapeutic environment (presence, immersion, interaction and interest). The primary aim of this systematic review is to examine the delivery and efficacy of VR therapeutics for clinical acute pain management in adults for the treatment of burn patients and for procedural pain. As secondary aims, we examine VR’s effect on pain-associated metrics (anxiety and physiologic measures), explore how included studies utilize VR components, determine the duration and kinds of VR content utilized, and recognize how studies address infection control and training for VR use.

Methods

Protocol and Eligibility Criteria

This systematic review was completed in accordance with Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (Moher, 2009; Moher et al., 2015). Articles for review included original research articles in published journal and conference proceedings. Inclusion criteria included articles addressing adults (18 years and older), acute pain, procedural pain, surgical pain, postoperative pain, physical pain, measures for pain, pain perception, and virtual reality (VR). For purposes of this review, VR was defined as “an immersive 3D display that [excludes] the external (real-world) environment” (Chan et al., 2018, p. 2). If their title or abstract matched the inclusion criteria, articles were reviewed. Articles in a language other than English were translated, using Google Translate, and reviewed against criteria. Exclusion criteria included: 1) articles addressing children and adolescents or containing either in their study in combination with adults, 2) studies focused on chronic pain and chronic pain syndromes (including chronic diseases with chronic pain elements – such as rheumatoid arthritis, neuropathic pain), 3) those that used augmented reality, and 4) studies that did not include AP metrics or measures. Articles were also excluded that lacked author names or access to full text. A complete list of inclusion criteria based on possible journal filters is included in Supplementary Materials.

Search and Study Selection

A search of PubMed, CINAHL, PsychINFO, Embase, Compendex, and Inspec was completed using MESH and keyword search terms related to AP and VR. Each database required database-specific terminology and terms, a complete list of the individual MESH and keyword terms is listed in Supplemental Materials. As the majority of clinical VR research has occurred since 2000, the search included all pertinent articles between January 1, 2000 and August 1, 2020. A database search was performed. Title and abstracts were uploaded to Rayyan, a free web- and app-based tool for title and abstract screening that has been utilized previously (Fedorowicz et al., 2012) and shown to reduce screening time (Cohen et al., 2006; Ouzzani et al., 2016). Studies were excluded if they failed to meet eligibility criteria. If studies contained titles that were relevant, but lacked abstracts for review, they were included in the final screening. Final screening of studies involved review of complete manuscripts whose title and abstract met the inclusion criteria. Studies were omitted if content failed to meet all stated criteria. Disagreements between reviewers were resolved based on consensus of all authors.

Risk of Bias Assessment

Eligible articles included in the review were assessed for bias using the Cochrane Review’s tool for assessing risk in randomized trials (Higgins et al., 2011). Each paper’s bias was rated as “high”, “low”, or “unclear” based on these respective categories. Non-randomized trials were not assessed and instead marked as not applicable (“n/a”). Disagreements at all stages of the selection process were resolved by consensus or in consultation with the senior author (H.T.).

Results

Study Selection and Assessment

Details of the study assessment process are shown in Figure 1. The initial search resulted in 3,847 articles, 856 of which were duplicates. The list of 2,991 non-duplicate articles 2,860 articles were excluded based on stated criteria (vide supra), leaving 131 articles. Full texts were obtained and 106 of them were subsequently excluded for not meeting inclusion criteria. Two articles were excluded due to duplicate publication. Thus, 23 articles met final inclusion criteria and were included in the review.

Figure 1: Flow Diagram for Study Selection and Inclusion.

Study Characteristics

Among eligible studies, sample sizes ranged from 8 to 182 participants, and 11 studies used a randomized controlled trial design. Studies utilized VR in Inpatient, Outpatient and Procedural Clinics as well as Emergency Room settings for wound care (8), procedure-induced pain (6), physical or occupational therapy (4), dental treatment (3) or generalized acute pain (2). Complete information on study populations, interventions, comparators, primary outcome measures and study designs (PICOS) can be found in Table 1.

Table 1:

Study Populations, Interventions, Comparators, Outcome Measures and Study Designs (PICOS)

Author (year)	Population	Content	Comparators	Origin of Pain	Pain Measure(s)	Study Design
Alshatrat et al (2019)	Patients at Dental Clinic	video [comedy or documentary]	VR + standard of care (SOC) vs SOC	Dental procedure	Visual analog scale (VAS) 0–10cm [5 dimensions : time thinking about pain, pain unpleasantness, how much teeth/gums bothered them, pain intensity, average pain]	Within-Subjects
Basak, Duman and Demirtas (2020)	Emergency room patients	underwater video	Distractio n (VR or Cards) + SOC vs SOC	Peripher al intravenous catheter insertion	VAS 0–10cm [pain intensity], 1–10cm [“satisfaction with procedure”]	Single-Blind, Randomized, Controlled Clinical Trial
Carrougher et al (2009)	Inpatient burn patients	n/a	VR + SOC vs SOC	Physical or Occupati onal Therapy	Graphic rating scale (GRS) – 0–100 [3 dimensions : pain intensity, time spent thinking about pain and pain unpleasant ness]	Within-Subjects
Ding et al (2019)	Hemorrhoide ctomy patients	“Snow World”	VR + SOC vs SOC	First hemorrh oid dressing change	VAS 0–10cm [pain intensity]	Open- Label, Randomi zed Clinical Trial
Ford et al (2018)	Outpatient burn patients	VR landscape videos [8 different to choose from]	n/a	Burn wound care	4-point Likert-like scale [provider’s impression of patient’s pain relief]	Feasibility and Acceptability Clinical Trial
Furman et al (2009)	Patients at Dental Clinic	“Second Life” [botanical garden]	VR + SOC vs movie + SOC vs SOC	Dental procedure	VAS 0–10cm [5 dimensions - time thinking about pain, pain unpleasant ness, how much teeth/gums bothered them, worst pain, and average pain]	Within- Subjects
Glennon et al (2018)	Patients with hematologica l diagnosis	Nature scene video [3 different to choose from], and “relaxing music”	VR + SOC vs SOC	Bone marrow aspiration and biopsy procedure	Numeric Rating Scale (NRS) 0–10 [pain intensity]	Quasi-experimental
Guo et al (2015)	Patients with hand injuries necessitating dressing change	“Afanda”/A vatar [3D film]	VR + SOC vs SOC	Hand wound dressing change	VAS [pain intensity, unidentified number/type]	Randomized Clinical Trial
Hoffman et al (2000)	Inpatient burn patients	“SpiderWo rld”	VR + SOC vs SOC	Physical or occupational therapy	VAS 0–10cm [5 dimensions - time thinking about pain, worst pain, average pain, how much wound bothered them, unpleasant ness of occupation al therapy]	Within- Subjects
Jahanisho orab et al (2015)	Postpartum primaparous women	IMAX Dolphin and Whales 3D	VR + SOC vs SOC	Episioto my repair	NRS 0–100	Randomized Clinical Trial
Jin et al (2018)	Patients with osteoarthritis after total knee arthroplasty	VR “Rowing”	VR + SOC vs SOC	Physical therapy	VAS [pain intensity, unidentified number/typ e]	Randomized Clinical Trial
Konstantatos et al (2009)	Inpatient burn wound patients	Virtual Medicine’s “Relaxatio n DVD” [hypnotherapy with relaxing visual scenery and audio; further details omitted]	VR + Patient-controlled analgesia (PCA) vs PCA	Burn wound care	VAS 0–10cm [pain intensity]	Randomized Clinical Trial
Maani et al (2011)	Inpatient burn wound patients	“SnowWorld”	VR + SOC vs SOC	Burn wound care	GRS 0–10 [3 dimensions : pain intensity, time spent thinking about pain, unpleasant ness of pain]	Within-Subjects
Morris et al (2010)	Inpatient burn wound patients	“Chicken Little” [PC game]	VR + SOC vs SOC	Physical therapy	NRS [pain intensity, no metrics listed, referred to as numeric pain rating scale in study]	Within-Subjects
Mosso Vasquez et al (2019)	Patients during outpatient lipoma removal	“Space coast”, “inMind VR”, and “Dyno VR games” [smartphon e], “Enchante d Forest” and “Magic Cliff” [computer-linked VR]	Smartpho ne VR vs Computer -linked VR	Lipoma removal	VAS 0–10cm [pain intensity]	Within-Subjects
Pandya et al (2017)	Patients receiving adductor canal catheter placement prior to unilateral primary total knee arthroplasty	“Titans of Space” [interactive with gazedirection], “Lanterns for Google Cardboard “ and “SeaWorld VR2” [both passive] [all had backgroun d music]	VR + SOC vs SOC	Ultrasou nd-guided adductor canal catheter placeme nt [nerve block procedure]	NRS 0–10 [pain intensity, only measured posttreatment]	Quasi-experimental
Patterson et al (2006)	Inpatient burn wound patients	“SnowWorl d” [no interaction] , audio hypnosis	VR + SOC vs SOC	Burn wound care	GRS 0–10 [3 dimensions : pain intensity, time spent thinking about pain, unpleasant ness of pain]	Within-Subjects
Sikka et al (2019)	Emergency room patients	24 different environments	VR + SOC vs SOC	Pain score > 3/10; Defined sources of pain	Verbal NRS 0–10 [pain intensity]	Within-Subjects
Spiegel et al (2019)	Inpatients in hospital	20 different environments	VR + SOC vs SOC	Pain score > 3/10; Undefine d sources of pain	Verbal NRS 0–10, [pain intensity]	Randomized Clinical Trial
Tanja-Dijkstra et al (2018)	Study 2: Patients undergoing dental treatment	Virtual coastal path and virtual urban environment	VR Content #1 + SOC vs VR Content #2 + SOC vs SOC	Dental treatment	Immediatel y post-intervention : NRS 0–10 [pain intensity], McGill Pain Questionna ire (SF-MPQ) [15-item short form]; At 1-week follow-up: NRS 0–10 [recall pain intensity, intrusive thoughts of experience, and vividness Of memories of experience]	Randomized Clinical Trial
Tse et al (2003)	Patients with leg ulcers necessitating wound debridement and dressing changes	Opera, cartoons television show, or natural environment	VR + SOC vs SOC	Leg ulcer wound debridement and dressing change	NRS 0–10 [pain intensity]	Within-Subjects
Walker et al (2014)	Men undergoing flexible cystoscopy	“SnowWorl d” [game]	VR + SOC vs SOC	Flexible cystoscopy	VAS 0–10cm [4 dimensions - time thinking about pain, pain intensity, average pain, pain unpleasant ness]	Randomized Clinical Trial
Zschaler (2010)	Inpatient burn wound care	Cold landscape with snowmen and canyons	VR + PCA vs PCA	Burn wound care	VAS [pain intensity] [no metrics defined]	Randomized Clinical Trial

Components of VR

Just under one third of studies (n=7) used patient-reported measures of presence in the virtual world (Table 2). In three of these studies, participants reported a moderate, or greater, sense of presence during VR (Alshatrat et al., 2019; Hoffman et al., 2000; Maani et al., 2011). While individual types of hardware for immersion varied drastically among studies (Table 2), the categories of immersive hardware could be divided into several groups. Computer-based VR content was used in 7 studies, one study used both computer- and smartphone-based content (Mosso Vázquez et al., 2019), five used smartphone-based content, 7 used 3D glasses (or equivalent), and three studies omitted details on their hardware (Jin et al., 2018; Walker et al., 2014; Zschaler, 2010). Over one third of studies (n=9) used active – as opposed to passive – VR, and one study compared active to passive VR (Mosso Vázquez et al., 2019). Interest (“fun” or another correlate to interest) was measured in four studies (Table 2).

Table 2:

Virtual Reality Components and Considerations

Presence Measures
Graphic Rating Scale (GRS) 0–10	(Maani CV et al., 2011)
Numeric Rating Scale (NRS) 0–10	(Tanja-Dijkstra et al., 2018)
Visual Analog Scale (VAS) 0–10/0–100	(Alshatrat et al., 2019; Hoffman et al., 2000; Walker et al., 2014)
Other	(Furman E et al., 2009; Guo et al., 2015)
Immersion Method
3D Glasses	(Alshatrat et al., 2019; Glennon et al., 2018; JahaniShoorab et al., 2015; Konstantatos et al., 2009; Tanja-Dijkstra et al., 2018; Tse et al., 2003)
Computer	(Ding et al., 2019; Furman E et al., 2009; Hoffman et al., 2000; Maani CV et al., 2011; Morris et al., 2010; Patterson DR et al., 2006)
Computer vs Smartphone	(Mosso Vázquez et al., 2019)
Smartphone	(Basak et al., 2020; Ford et al., 2018; Pandya et al., 2017; Sikka et al., 2019; Spiegel B. et al., 2019)
Unknown	(Carrougher et al., 2009; Jin C. et al., 2018; Walker et al., 2014; Zschaler, 2010)
Interaction Type
Active	(Carrougher et al., 2009; Ding et al., 2019; Furman E et al., 2009; Hoffman et al., 2000; Jin C. et al., 2018; Maani CV et al., 2011; Morris et al., 2010; Tanja-Dijkstra et al., 2018, p.; Walker et al., 2014)
Active + Passive (varying content)	(Pandya et al., 2017; Spiegel B. et al., 2019)
Passive	(Alshatrat et al., 2019; Basak et al., 2020; Ford et al., 2018; Glennon et al., 2018; Guo et al., 2015; JahaniShoorab et al., 2015; Konstantatos et al., 2009; Patterson DR et al., 2006; Sikka et al., 2019; Tanja-Dijkstra et al., 2018; Tse et al., 2003)
Unknown	(Mosso Vázquez et al., 2019; Zschaler, 2010)
Interest Measures
5-Point Bipolar Adjective Scale (“attractiveness” of VR environment)	(Tanja-Dijkstra et al., 2018)
Numeric Rating Scale (NRS) 0–10 (“enjoyment” in watching video)	(Tse et al., 2003)
Visual Analog Scale (VAS) 0–10 (“fun” in VR)	(Maani CV et al., 2011)
Visual Analog Scale (VAS) 0–10 (how “entertaining” was the virtual world)	(Walker et al., 2014)
Average Time in VR
0–5 min	(Basak et al., 2020; Guo et al., 2015; Hoffman et al., 2000)
6–10 min	(Carrougher et al., 2009; Ford et al., 2018; Maani CV et al.,2011; Morris et al., 2010; Sikka et al., 2019; Spiegel B. et al., 2019)
11–15 min	(Glennon et al., 2018)
16–20 min	(Furman E et al., 2009; Konstantatos et al., 2009; Patterson DR et al., 2006)
21–25 min	(Ding et al., 2019)
26–30 min	(Jin C. et al., 2018)
>30 min	(Mosso Vázquez et al., 2019)
Unknown	(Alshatrat et al., 2019; Pandya et al., 2017; Tanja-Dijkstra et al., 2018; Tse et al., 2003; Walker et al., 2014; Zschaler, 2010)

Of the included studies, 12 used digitally rendered VR environments, 8 studies used a video or 3D video, one used a computer game (Morris et al., 2010), one study described the environment but not its name (Zschaler, 2010), and one study omitted details on content (Carrougher et al., 2009); see Table 1. Of note, one digitally-rendered VR environment (Patterson et al., 2006) and one video (Konstantatos et al., 2009) utilized a specified therapeutic (hypnosis); no other studies used content based on evidence-based therapeutics. Though four studies used the same VR environment (“SnowWorld”), all other studies with smartphone or computer-based VR used different environments. Across studies, participants reportedly spent between two and 30+ minutes in virtual environments (Table 2).

Effect of VR on Pain Intensity

In all but one study, pain intensity was assessed using a visual analog scale (VAS), graphic rating scale (GRS) or numeric rating scale (NRS; Table 1). Of the 23 studies reviewed, 19 reported decreases in pain intensity while using VR. Of these 19 studies, 15 studies reported statistically significant decreases in pain intensity (p<0.05). Seven were in comparison to a no-VR condition (Alshatrat et al., 2019; Furman E et al., 2009; Hoffman et al., 2000; Maani et al., 2011; Sikka et al., 2019; Spiegel B. et al., 2019; Tse et al., 2003), and seven to a no-VR group (Ding et al., 2019; Guo et al., 2015; JahaniShoorab et al., 2015; Jin et al., 2018; Konstantatos et al., 2009; Pandya et al., 2017; Tanja-Dijkstra et al., 2018). Yet not all studies have shown these positive results. Two studies noted no difference between groups receiving VR for AP and standard of care (Glennon et al., 2018; Walker et al., 2014) and Morris et al (2010) found that there was no difference in pain intensity when individuals were, or were not, using VR. Lastly, Zschaler et al (2010) found VR to increase pain intensity when paired with patientcontrolled analgesia. Further details about study pain measures can be found in Table 1.

Secondary Outcomes and Practical Considerations

Anxiety is a correlate of pain, and physiologic measures (heart rate and blood pressure) are pain indicators, yet these were rarely assessed in studies. Less than half of studies (10 of 23) measured anxiety (Table 3). Virtual reality decreased anxiety, “stress”, or “nervousness” in three of these studies (Tanja-Dijkstra et al., 2018; Hoffman et al., 2000; Patterson et al., 2006), but had no effect on anxiety in others (Glennon et al., 2018; Konstantatos et al., 2009; Morris et al., 2010). In the seven studies that reported physiologic measures, VR had mixed effects. During some painful events VR was found to decrease systolic blood pressure (Alshatrat et al., 2019; Mosso Vázquez et al., 2019), yet in the same study by Alsharat and colleagues, no change was seen in diastolic blood pressure or heart rate. Other studies have also reported VR to have little effect on blood pressure (Furman E et al., 2009; Pandya et al., 2017).

Table 3:

Secondary Outcomes and Practical Considerations

Anxiety Measures
Corah Dental Anxiety Scale	(Furman et al., 2009)
Likert-like scale 0–4	(Glennon et al., 2018)
Visual Analog Scale (VAS) 0–10	(Hoffman et al., 2000; Walker et al., 2014)
(“anxiety”)
Modified Dental Anxiety Scale 0–4	(Tanja-Dijkstra et al., 2018)
Patient-Reported Outcome Management Information System (PROMIS) Anxiety Short-Form 8a	(Sikka et al., 2019)
Burn-Specific Anxiety Rating Scale (BSARS)	(Konstantatos et al., 2009; Morris et al., 2010; Patterson et al., 2006)
Physiologic Measures
Blood Pressure	(Alshatrat et al., 2019; Furman E et al., 2009; Glennon et al., 2018; Mosso Vázquez et al., 2019; Pandya et al., 2017; Walker et al., 2014)
Heart Rate	(Carrougher et al., 2009; Ding et al., 2019; Furman et al.,2009; Glennon et al., 2018; Walker et al., 2014)
Respiratory Rate	(Glennon et al., 2018; Walker et al., 2014)
Oxygen Saturation	(Ding et al., 2019; Glennon et al., 2018)
Temperature	(Glennon et al., 2018; Walker et al., 2014)
Galvanic Skin Response	(Walker et al., 2014)
VR Side Effects Measures
Visual Analog Scale (VAS) 0–10 (“nausea”)	(Alshatrat et al., 2019; Hoffman et al., 2000; Walker et al., 2014)
Graphic Rating Scale (GRS) 0–10 (“nausea”)	(Maani et al., 2011)
Unknown	(Carrougher et al., 2009; Furman E et al., 2009)
Practical Considerations
Mention Infection Control	(Sikka et al., 2019; Spiegel B. et al., 2019)
Train Participants for VR Use	(Ding et al., 2019; Ford et al., 2018; Pandya et al., 2017; Patterson DR et al., 2006; Sikka et al., 2019; Spiegel et al., 2019)

Less than half of included studies measured, or commented on, VR-related side effects. Of the potential side effects of VR (Cobb et al., 1999), nausea was the only adverse VR symptom reported by articles that were included in this review. Four studies reported no subjects who experienced nausea (Maani et al., 2011; Pandya et al., 2017; Walker et al., 2014; Wright et al., 2005), three studies reported subjects who experienced mild nausea (Carrougher et al., 2009; Furman et al., 2009; Hoffman et al., 2000), and one study reported a small number of subjects with greater than mild nausea after VR use (Alshatrat et al., 2019). Another finding of included studies was that practical considerations – such as infection control and participant training – were seldom mentioned. Only two studies mentioned infection control measures taken during the study (Sikka et al., 2019; Spiegel B. et al., 2019), and 6 studies mentioned training participants in VR use prior to beginning the study. Details on anxiety, physiologic and VR side effect measures used and inclusion of infection control or VR training are found in Table 3.

Risk of Bias

Of the studies included in the bias assessment (12), one-half clearly utilized random sequence generation, while others omitted how randomization occurred. Only one study reported clear allocation concealment (Tanja-Dijkstra et al., 2018), while three had high potential for bias (Jin et al., 2018; Spiegel et al., 2019; Walker et al., 2014). The remainder (n=8) failed to address it altogether. The majority of studies suffered from a high degree of bias in blinding of participants and personnel, as well as the outcome assessment (n=9); the remainder failed to mention these details. Only one study noted incomplete data that could have biased the study, and no studies selectively reported on their data. Just under one-half (n=12) of studies were randomized controlled trials and were evaluated using the bias rating tool; other studies were not evaluated but are assumed to contain more bias due to study design. Further details of studies included in the bias tool can be found in Supplementary Materials.

Discussion

Pain and Related Outcomes

The results of this systematic review show VR to be an effective tool for AP management. A likely mechanism by which VR promoted analgesia in these studies is distraction. Distraction is a simple but effective approach that has been successfully utilized to assist with AP management (Hudson et al., 2015; Miller et al., 1992; Primack et al., 2012). Distraction leverages elements of the Gate Theory of Pain by directing cognitive attention away from noxious stimuli towards visual and auditory input, or an activity (Melzack & Katz, 2006; Melzack & Wall, 1965). As the brain only has so much ability to process incoming stimuli (Kahneman, 1973), cognitive-related brain areas provide analgesia by inhibiting passage of noxious stimulation to the brain (Melzack & Katz, 2006; Terkelsen et al., 2004). VR for AP utilizes these principles to shift focus away from noxious stimuli and into a computer-generated environment (Gold et al., 2007). This was not the only successful modality for managing AP with VR, as studies by Konstantatos et al (2009) and Patterson et al (2006) delivered hypnotic suggestions via a virtual environment to reduce AP. However, each study delivered VR hypnosis prior to the pain-inciting event, as opposed to during the painful event as described in other VR distraction studies. This review’s findings extend results from laboratory studies showing VR’s efficacy for AP analgesia (Boylan et al., 2017; Czub & Piskorz, 2018; Karaman et al., 2019; Sharar et al., 2016) by deploying VR interventions into real-world clinical settings. The implications of this translational research supports the potential use of VR content for acute-pain analgesia, though further study is needed to confirm distraction as the analgesic mechanism of the action in these VR-based therapeutics. Similarly, hypnosis-based VR content also shows promise as a modality for AP, but further research is needed for clinical validation.

Few studies included measures of anxiety, and those that did varied greatly in the measures used. This made comparison across studies problematic. Even fewer studies utilized physiologic measures, and those that did showed mixed results. This is likely due to the fact that while a majority of physiologic measures have been shown to be valid for AP assessment, it is often necessary to combine them with other AP measures or use interpretive algorithms to ensure accuracy and minimize confounding elements (Cowen et al., 2015; Korving et al., 2020).

Virtual Presence and Negative Outcomes

Four studies showed VR to have little effect, or a negative effect, on AP management. Of these four studies with negative outcomes, only Walker et al (2014) utilized VR content and measured presence; participants were found to have a low sense of presence in the virtual environment. Glennon et al (2018) and Morris et al (2010) both utilized non-VR content (a nature video and a computer game, respectively) in their studies, while Zschaler et al (2010) described their content, but omitted further details. While all three used headsets to deliver content and block external stimuli, it is difficult to determine intervention fidelity as none of these studies measured presence.

One reason for neutral or negative outcomes could have also been participant positioning. In the study by Glennon et al (2018), participants were placed in a prone position and then watched a nature video, and Walker et al (2014) placed participants in a supine position prior to immersion in VR. The point of view for most VR environments is from an upright position; these participants were not in upright positions for their procedures, yet these articles make no mention of the VR environment’s orientation (or changes in VR content to match these non-upright positions). A mismatched visual-physical experience or poorly matched content (ex: staring at the ground or up at the sky – unable to see any horizontal content) could detract from the efficacy of VR. This could have been responsible for the decreased presence found by Walker et al (2014). In addition, when participants used the computer game in the study by Morris et al (2010), it is unknown if participants were immersed in the environment and could look around the virtual world or if their view was fixed, regardless of head position (virtually: playing video games on a television screen in an otherwise dark room). The latter would be an example of using non-VR content with VR hardware – essentially watching a movie at a drive-in. You may feel present in that three-dimensional environment, but content (software) limitations do not allow you to actually be virtually present in that environment. The capability to be virtually present in a digital environment is what sets VR content apart from other console-, computer- or smartphone-based interventions (ex: personal computer (PC), Playstation^®, XBOX^®, iOS, Android and other platforms). In the study by Zschaler et al (2010), participants were asked to use a patient-controlled analgesia (PCA) device while in VR. It’s unknown if this device was visible in VR or how its use may have impacted users as this study lacked a measure of presence. Konstantatos et al (2009) used patient controlled analgesia during the same procedure and in the same population, but as their intervention (VR hypnosis) was delivered prior to the procedure, participants did not have the same issue. These studies underscore the need to measure presence in clinical VR studies and consider how the presentation of content and a participant’s positioning might impact the analgesic efficacy of VR interventions. Studies have shown how increasing presence also increases the analgesic efficacy of the virtual environment (Gutierrez-Martinez et al., 2010; Hoffman et al., 2004), but further research is needed to inform deployment of VR in the clinical context.

Considering Elements of Interaction

Active VR and passive VR content varied greatly across included studies. Interaction is another VR component that influences presence and analgesic efficacy. Only Mosso Vasquez et al (2019) compared active VR and passive VR in the clinical context; they found that both kinds of interaction were effective for analgesia, though active (computer-based) VR had greater analgesic efficacy. These findings align with several laboratory studies by Hoffman et al (2004, 2006) and Wender et al (2009) that found active VR to significantly decrease pain intensity, compared to passive VR. This potential for variation in analgesic efficacy based on interaction with the virtual environment could prove problematic in studies with varied kinds of interactive content. To avoid confounding, it is important for future clinical VR studies to consider the kind of interaction being deployed when designing the study, assigning participants, and performing analyses. Further research comparing the efficacy of active and passive VR content for AP analgesia is needed.

Varying Content and Hardware

While each of the studies included in this review used what they termed “virtual reality”, there was a large variation in the content that was used. Alsharat et al (2019), Guo et al (2015) and Tse et al (2003) utilized non-VR video content that was delivered via a headset (the equivalent of a seeing a large television screen). Other studies utilized VR hardware to deliver nature videos (Basak et al., 2020; Ford et al., 2018; Glennon et al., 2018; JahaniShoorab et al., 2015) and digitally-rendered nature content (Furman et al., 2009; Tanja-Dijkstra et al., 2018). VR games were used in several studies (Ding et al., 2019; Hoffman et al., 2000; Jin et al., 2018; Maani et al., 2011; Walker et al., 2014), while another deployed a non-VR computer game f(Morris et al., 2010). Studies by Mosso-Vasquez et al (2019), Pandya et al (2017), Sikka et al (2019) and Spiegel et al (2019) deployed a variety of differing VR content that users could pick from that included 360-degree videos, VR games, and other VR content. Because content differed across the majority of studies and were deployed on a diverse array of devices (with varying levels of screen quality, field of view and some with and others without head-tracking), it is difficult to compare efficacy across studies. That said, laboratory studies have shown that playing non-VR video games and using VR are both effective in decreasing pain intensity compared to a no-VR condition - though VR showed the greatest analgesic efficacy (Boylan et al., 2017). Future clinical studies comparing each of these interventions – video or television, video games, and VR content – are needed to ascertain which modalities are best deployed to properly address AP analgesic needs and pragmatic constraints across clinical settings. Comparison between types of VR content (360-degree video versus digitally-rendered content) and differing VR modalities (games versus open-world) is also needed to better inform clinical deployment. Lastly, further development of evidence-based therapeutic VR content is needed.

Practical Considerations

While VR is highly effective for AP management, the most immersive and interactive VR content has historically been more expensive and required computer equipment to process the virtual environment (Hoffman et al., 2004, 2006; Wender et al., 2009). In spite of being less interactive, Ford et al (2018) and Mosso Vasquez et al (2019) both note that reduced cost and ease-of-deployment of passive VR content via smartphone-based headsets could allow for more widespread use and should be considered. Additionally, newer VR headsets with greater processing power that do not require a computer are also available on the consumer market at significantly reduced costs. Usability testing of this type of hardware in clinical settings is needed.

Two other barriers to widespread deployment of VR therapeutics are infection control and training in VR use. Few of the included studies mentioned infection control measures, yet this is a vital consideration for clinical deployment. Additionally, infection control procedures – like other technology used in healthcare spaces (Brady et al., 2012; Pratt et al., 2001) – may degrade VR devices’ hardware and potentially reduce their performance. Using VR also necessitates training of both clinical staff and participants. Few studies noted taking time to orient and instruct participants in how to use the headsets and navigate the VR content; failure to do so may have impacted participant’s experience. Future studies must consider implementation of infection control measures and assess the time needed to adequately train both staff and participants in VR use. Further research is needed to explore means maintaining infection control without harming VR equipment, enhance current means of cleaning VR hardware between uses and develop more streamlined approaches to training participants in VR therapeutic use for AP management.

Areas for Future Research

After reviewing the current literature on VR’s use for AP management, there are several areas for future research that become clear. While VR has been used extensively in wound care populations, its use in virtually all other healthcare populations for AP management is limited, and in most settings, nonexistent. Use of anxiety, presence, and VR side effects measures in VR studies for AP is limited; future studies should incorporate these measures to provide better insight into VR’s efficacy for AP analgesia. The majority of the reviewed VR content lacks firm grounding in evidence-based interventions. Researchers, designers and developers must cultivate collaborations to develop theory-driven and pain-specific content that maximizes VR principles of presence, immersion, interaction, and interest in adult populations. Further clarity regarding the mechanism of action of VR content for AP management is needed, as well as a greater understanding of how the various elements of VR itself might impact the intended therapeutic effect. Further research is needed to test evidence-based VR content over time and across various clinical populations to ensure its safety and maximize its potential efficacy.

Implications for Pain Management Practice

There is a dearth of information currently available for practitioners related to VR content, hardware and deployment. Results from this review are intended to guide translation of VR therapeutics for clinical use, and inform future deployment of VR interventions for AP management. Based on the studies included in this review, there is strong evidence for utilizing VR-based distraction content during pain-inciting events, and promising evidence for utilizing hypnosis-based content prior to pain-inciting events. Providers should be trained on VR deployment prior to use, and utilize infection-control standards in accordance with the population using the intervention and location of deployment.

Limitations

While the body of evidence for using VR for AP management with burn patients and procedural pain is compelling, less than half of the included studies employed randomization, and the majority of studies show clear biases in their designs, methods, and analyses. Studies without immersive 3D displays were excluded, which may have omitted some high-quality VR studies (i.e. using a projector instead of a VR headset). Overall, cross-study comparison is difficult due to the extensive heterogeneity in hardware and software use.