Secondary Logo

Journal Logo

Obstetric Anesthesiology

Virtual Reality Analgesia in Labor: The VRAIL Pilot Study—A Preliminary Randomized Controlled Trial Suggesting Benefit of Immersive Virtual Reality Analgesia in Unmedicated Laboring Women

Frey, David P. DO*; Bauer, Melissa E. DO*; Bell, Carrie L. MD; Low, Lisa Kane PhD, CNM; Hassett, Afton L. PsyD*; Cassidy, Ruth B. MA*; Boyer, Katherine D. BS*; Sharar, Sam R. MD§

Author Information
doi: 10.1213/ANE.0000000000003649

Abstract

Childbirth is considered one of the most memorable and painful experiences in a woman’s life. There is great variability in the duration and intensity of pain during this experience and in the chosen analgesic modalities.1–3 In the United States, approximately 60% of women receive neuraxial analgesia for their labor.4 While only 17% of women report an unmedicated birth in the United States,2 75% use some form of nonpharmacologic method either alone or in combination with medical therapies when surveyed.2 Although epidural analgesia is considered the most effective treatment for labor pain,1,2,4 it is not without side effects. Using nonpharmacologic approaches for labor pain has the potential to reduce side effects, increase rates of breast feeding, increase shared decision making and sense of control over the birth process, and ultimately increase satisfaction.3,5

One possible nonpharmacologic intervention that has been effective in a wide variety of clinical settings to manage acute pain is immersive virtual reality (VR) distraction.6 VR is a phrase coined by Jaron Lanier in the mid-1980s to refer to a user-computer interface that allows for real-time simulation of an environment that users can interact with via multiple sensory channels in an intuitive manner.7,8 Unlike other forms of media, VR can create an illusion of presence or an uncanny feeling of actually being in a computer-generated environment.8 VR is thought to offer potential therapeutic advantages in acute pain over other forms of distraction because of its ability to create a multisensory distraction while isolating a patient from the immediate clinical setting and replacing it with a more attractive virtual environment.9 Indeed, research has shown greater analgesic effects achieved with increasing degrees of VR immersion when compared to less immersive VR systems,10,11 traditional videogames,12 and music alone.13 Although studied for more than a decade in acute pain settings, the high cost of VR systems was inhibitory to widespread application and required proprietary software. However, with the proliferation of high-definition screens used on mobile phones, the cost of high-quality VR headsets has fallen dramatically, making it a potentially scalable pain management modality.

This current pilot study is one of the first to investigate using immersive VR analgesia to treat labor pain. Our primary hypothesis was that VR distraction would show statistically significant decreases in reported sensory pain (“worst pain” scores) compared with no VR distraction in women experiencing unmedicated labor. Secondarily, we hypothesized beneficial effects on cognitive and affective components of pain, anxiety, nausea, and patient satisfaction scores.

METHODS

This study was conducted at Michigan Medicine’s Von Voigtlander Women’s Hospital between November 2016 and April 2017 and was approved by the Institutional Review Board (HUM00116129). Information was made available during prenatal appointments and through direct mailings to all women scheduled to deliver at the study site. Written informed consent was obtained from qualifying patients on presentation for delivery. It adheres to applicable Enhancing the QUAlity and Transparency Of health Research (EQUATOR) guidelines. It was also registered on clinicaltrials.gov #NCT02926469 by the principal investigator David P. Frey, DO, on October 6, 2016.

Eligible patients were otherwise healthy women at ≥32 weeks’ gestation giving birth for the first time and in the first stage of labor with an anticipated vaginal delivery. Exclusion criteria included current use of pharmacologic analgesia (including neuraxial analgesia), age <18 or >45, presence of fetal or placental anomaly, high-risk pregnancy, or fetal concerns for which a delay in labor epidural placement would have been undesirable due to risk of urgent delivery (eg, body mass index >40, difficult airway, hemorrhage, nonreassuring fetal tones, malpresentation), inability to indicate pain intensity or complete study measures, requiring an interpreter, hearing or vision deficits, psychiatric disorders, seizure history, or reported predisposition for motion sickness.

This was a prospective, randomized controlled, counterbalanced, crossover (within-subjects) study that recruited 28 patients during the first stage of labor who met inclusion criteria. Subjects were randomly assigned to receive unmedicated labor (without analgesics, alternative therapies, or systematic distraction) either alone or with VR, followed by the opposite condition (order randomized and counterbalanced with a random number generator). This crossover design allowed each patient to serve as her own control to decrease the variability between separate patients’ labor experiences. Participants were observed for equivalent times during unmedicated contractions (also without alternative therapies or systematic distraction) in both the VR and non-VR conditions. After each condition, subjects were provided with questionnaires to assess pain, anxiety, nausea, and perception of the VR experience.

Similar to other clinical VR studies,9,14 subjects were asked to provide ratings of 3 separate pain outcomes using numeric rating scale tools. Specifically, patients rated the amount of time spent thinking about pain (cognitive pain dimension), pain unpleasantness (affective pain dimension), and worst pain intensity (sensory pain dimension) they experienced (Supplemental Digital Content 1, Appendix, http://links.lww.com/AA/C490). These pain dimensions are separately measurable15 and have reliably assessed pain outcomes to detect treatment effects.16

MATERIALS AND PROCEDURE

Using only consumer-ready components, an immersive and interactive VR system was developed using a Samsung GearVR (Samsung, San Jose, CA) head-mounted display powered by a Galaxy S7 phone, a hand control, and noise-reducing headphones powered by a parallel S5 phone. The study began after regular contraction pain scores of ≥4/10 were reported. Each condition lasted 10 minutes (or 3 contractions) to replicate previously studied exposure times of VR for acute pain.9 Orientation to the device and instructions took <60 seconds. Each patient experienced the same scene of curious manatees from the Ocean Rift (www.ocean-rift.com) scuba diving simulation with sounds of manatee calls and breathing underwater. Additional relaxing music was supplied from nighttime sleep by Brain.fm (www.brain.fm). User input consisted of head tracking and a hand control that simulated taking underwater photos. After each condition, patients completed questionnaires as described above.

Statistical Analysis

By utilizing a within-subjects randomized crossover design, each subject served as her own control. Separate regression models for each outcome (pain, anxiety, and nausea) were used to assess mean differences and their corresponding 95% CIs, using generalized estimating equations. Differences were calculated as control (no VR) condition minus the treatment (VR) condition. Additionally, means and standard deviations were used to summarize rating responses. The fitted models accounted for repeated measurements on each subject. Significance was determined by 95% CIs that do not include zero. Average differences, standard deviations of differences, and standardized mean differences (calculated as mean difference/standard deviation of difference) were also used to assess effect sizes.

Differential carryover effects were evaluated by examining mean differences (VR − no VR) in types of pain, nausea, and anxiety and performing Student t test across the randomization groups. This test compared the average differences between the VR first and VR second randomization groups. To assess evidence of imbalance between randomization groups, we also compared patients’ baseline characteristics by the VR first and VR second groups (Supplemental Digital Content 2, Table S1, http://links.lww.com/AA/C491). Means and standard deviations, frequency (n [%]), and medians and interquartile ranges were calculated as appropriate, in addition to appropriate measures of standardized mean differences.17 For skewed continuous variables, medians and interquartile ranges were calculated, and modified standardized differences were computed using rank statistics. Imbalance was defined as the absolute standard difference exceeding 1.96*sqrt[(nVR1 + nVR2)/nVR1nVR2] = 0.755.18

With a 2×2 crossover design, assuming that the standard deviation of the paired differences is 1.8492 and α is .05, a target sample size of ≥30 participants would detect a difference in the means between the 2 treatment groups of 1 point on the worst pain score with 80% power in a 2-sided t test.

Statistical analyses were completed using SAS 9.4 (SAS Institute Inc, Cary, NC).

RESULTS

Twenty-eight subjects were enrolled, with 27 completing the study. One withdrawal occurred due to the patient’s desire to enter the bathtub during the study to avoid risk of water damage to the VR equipment and potential confounding of results with complementary water bath therapy. The subjects ranged in age from 19 to 38 years old (mean ± SD: 27.9 ± 5.6), with 78% identifying as Caucasian and 67% having a bachelor’s degree or higher.

As shown in the Figure, the numeric rating scale scores of the primary outcome for worst pain intensity (sensory pain) were significantly lower in the VR condition (slope estimate −1.5 [95% CI, −0.8 to −2.2] and standardized mean difference −0.8). The NRS scores for secondary outcomes of affective pain (slope estimate −2.5 [95% CI, −1.6 to −3.3] and standardized mean difference −1.0), cognitive pain (slope estimate −3.1 [95% CI, −2.4 to −3.8], and standardized mean difference −1.7) were also significantly lower in the VR condition. The NRS scores for anxiety were also lower in the VR than non-VR condition (slope estimate −1.5 [95% CI, −0.8 to −2.3] and standardized mean difference −0.7).

Figure.
Figure.:
Study outcomes between conditions. Mean pain ratings on a numeric rating scale (NRS) with virtual reality (VR; white) and without VR (black) during contractions. Standard deviations are indicated by error bars.

Eighty-two percent reported very much/completely enjoying VR use during labor. Seventy percent of participants would be very/completely interested in new VR development specifically for childbirth. There were no adverse events and no significant differences in nausea (slope estimate mean −0.4 [95% CI, 0.1 to −0.8] and standardized mean difference −0.3). Ultimately, 15% of subjects who did not eventually require cesarean delivery went on to complete an unmedicated vaginal delivery.

To assess for differential carryover effect, mean differences (VR − no VR) were compared by order of treatment group (VR first or VR second), and there were no statistically significant differences found in the primary or secondary outcomes as assessed via t test. All standardized differences across order of treatment group were below the threshold of 0.755 (Supplemental Digital Content 2, Table S1, http://links.lww.com/AA/C491), so the 2 groups appeared similar in terms of demographics, and therefore the 2 groups were treated as similar in the differential carryover effect and main statistical analyses.

To determine whether rare outlying observations might have skewed results to significant findings, boxplots and histograms of the differences for pain ratings were completed (Supplemental Digital Content 3, Figure S1, http://links.lww.com/AA/C492). The only measure showing outliers was nausea; however, the generalized estimating equation model results for nausea were not statistically significant. Therefore, it does not appear that outlying observations are driving the results toward significance.

DISCUSSION

In this pilot study, we observed that it is feasible to create a VR experience to decrease pain and anxiety during the first stage of labor using low-cost, consumer-ready products without increasing nausea. The primary outcome for worst pain intensity (sensory pain) scores was significantly lower in the VR condition, with similar decreases seen in the secondary outcomes of reported affective and cognitive pain scores. These reductions in labor pain scores observed in this study are similar in trend and magnitude to those of previous studies of immersive VR in different clinical scenarios of acute pain outside of the labor process.9,12

There were several limitations to this study. Participants self-selected for this study, and results may be affected by this selection bias. However, real-life users of VR would also self-select. Additionally, the treatment condition was not blinded, and participants may have underreported their pain scores after the VR experience. This limitation is inherent to a within-subject crossover design, although our design allowed for treatment order randomization while also minimizing the large variability seen in labor experiences. A between-groups design utilizing placebo VR would potentially avoid this limitation.

While cervical dilation measurements would ideally have been obtained immediately before the study condition, this was a low-intervention clinical population with infrequent cervical checks, and we did not want to require such an intervention for participation. However, by using individually reported pain scores, the within-subjects design better controls for subject variability by ensuring factors like variability in contractions, previous pain experiences, previous VR experience, catastrophizing, coping ability, and pain expectations were identical in both study conditions regardless of cervical dilation variability between patients. In addition, we fell just short of our recruiting goal in the targeted recruitment period due to the strict exclusion criteria and the uniqueness of the type of patient interested in participating.

One final limitation is that the optimum VR experience for use in labor has yet to be defined, including its intended duration and frequency of use during the labor process. Further, at the time of this study, there were no readily available VR experiences thought to be appealing for use in laboring women, and a VR experience was created by recontextualizing existing consumer content and devices. The current VR experience was designed to last ≤10 minutes (similar to previously studied uses of VR in acute pain),9 but labor averages many hours.2 Creating VR experiences for prolonged labors will be limited by the current absence of labor-specific VR content, lack of female developers, headset comfort, and lens fogging.

In summary, the results of this pilot study support the concept that consumer VR may be an accessible and useful nonpharmacologic modality of pain and anxiety management during labor without major side effects and supports the need for continued research. Future development for VR applications in laboring women—either alone or as an adjunct therapy—should focus on ease of use and intuitive design, prehospital education, custom-tailored virtual environments, adaptability to variable patient positions (bed, chair, birth balls, tubs, and operating tables if requiring operative delivery), positive motivational biofeedback, goal-oriented tasks including position changes and mobility, and emotionally engaging content appealing to laboring women.

ACKNOWLEDGMENTS

The authors thank Michelle Housey, MPH, for assisting with the design of this study.

DISCLOSURES

Name: David P. Frey, DO.

Contribution: This author helped design the study, collect the data, and analyze and prepare the manuscript.

Name: Melissa E. Bauer, DO.

Contribution: This author helped design and interpret the study and prepare the manuscript.

Name: Carrie L. Bell, MD.

Contribution: This author helped design the study and prepare the manuscript.

Name: Lisa Kane Low, PhD, CNM.

Contribution: This author helped design and interpret the study and prepare the manuscript.

Name: Afton L. Hassett, PsyD.

Contribution: This author helped design and interpret the study and prepare the manuscript.

Name: Ruth B. Cassidy, MA.

Contribution: This author helped analyze, interpret, and prepare the manuscript.

Name: Katherine D. Boyer, BS.

Contribution: This author helped collect the data and prepare the manuscript.

Name: Sam R. Sharar, MD.

Contribution: This author helped design the study, interpret the data, and prepare the manuscript.

This manuscript was handled by: Jill M. Mhyre, MD.

REFERENCES

1. Koyyalamudi V, Sidhu G, Cornett EM, et al. New labor pain treatment options. Curr Pain Headache Rep. 2016;20:11.
2. Declercq ER, Sakala C, Corry MP, Applebaum S, Herrlich A. Listening to MothersSM III: Pregnancy and Birth. New York: Childbirth Connection; 2013. Available at: http://transform.childbirthconnection.org/wp-content/uploads/2013/06/LTM-III_Pregnancy-and-Birth.pdf. Accessed May 25, 2018.
3. Simkin P, Bolding A. Update on nonpharmacologic approaches to relieve labor pain and prevent suffering. J Midwifery Womens Health. 2004;49:489–504.
4. Osterman MJ, Martin JA. Epidural and spinal anesthesia use during labor: 27-state reporting area, 2008. Natl Vital Stat Rep. 2011;59:1–13, 16.
5. Bonapace J, Gagné GP, Chaillet N, Gagnon R, Hébert E, Buckley S. No. 355-physiologic basis of pain in labour and delivery: an evidence-based approach to its management. J Obstet Gynaecol Can. 2018;40:227–245.
6. Li A, Montaño Z, Chen VJ, Gold JI. Virtual reality and pain management: current trends and future directions. Pain Manag. 2011;1:147–157.
7. Lanier J. Virtually there. Sci Am. 2001;284:66–75.
8. Gonzalez-Franco M, Lanier J. Model of illusions and virtual reality. Front Psychol. 2017;8:1125.
9. Hoffman HG, Chambers GT, Meyer WJ 3rd, et al. Virtual reality as an adjunctive non-pharmacologic analgesic for acute burn pain during medical procedures. Ann Behav Med. 2011;41:183–191.
10. Hoffman HG, Seibel EJ, Richards TL, Furness TA, Patterson DR, Sharar SR. Virtual reality helmet display quality influences the magnitude of virtual reality analgesia. J Pain. 2006;7:843–850.
11. Wender R, Hoffman HG, Hunner HH, Seibel EJ, Patterson DR, Sharar SR. Interactivity influences the magnitude of virtual reality analgesia. J Cyber Ther Rehabil. 2009;2:27–33.
12. Hoffman HG, Patterson DR, Carrougher GJ. Use of virtual reality for adjunctive treatment of adult burn pain during physical therapy: a controlled study. Clin J Pain. 2000;16:244–250.
13. Johnson S, Coxon M. Sound can enhance the analgesic effect of virtual reality. R Soc Open Sci. 2016;3:150567.
14. Schmitt YS, Hoffman HG, Blough DK, et al. A randomized, controlled trial of immersive virtual reality analgesia, during physical therapy for pediatric burns. Burns. 2011;37:61–68.
15. Gracely RH, McGrath F, Dubner R. Ratio scales of sensory and affective verbal pain descriptors. Pain. 1978;5:5–18.
16. Jensen MP. The validity and reliability of pain measures in adults with cancer. J Pain. 2003;4:2–21.
17. Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS. Paper 335–2012. 2012. Proceedings of SAS Global Forum Conference. Cary, NC, SAS Institute Inc. Available at: http://support.sas.com/resources/papers/proceedings12/335–2012.pdf. Accessed May 25, 2018.
18. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28:3083–3107.

Supplemental Digital Content

Copyright © 2018 International Anesthesia Research Society