Virtual reality and its use in post-operative pain following laparoscopy: a feasibility study

Study design

The study was an open-label single-centre randomized crossover pilot trial in 35 women in a tertiary university teaching hospital, between April and August 2019. This trial had approval from the Monash Health Human Research and Ethics Committee (HREC/45131/MonJ- 2018-150802) and was performed in accordance with the Good Clinical Practice principles. This trial was also registered on the Australia-New Zealand Clinical Trials Registry (ACTRN12618001398291p; registration date: 20/08/2018).

Participants

Women were eligible for inclusion in this study if they were above 18 years of age and undergoing a laparoscopic procedure for a gynecological condition.

Clinical exclusion criteria were: conversion to laparotomy; chronic narcotic use or narcotic dependence; pregnancy; patients with an intellectual impairment; patients with co-morbidities such as pre-existing heart disease; and people in existing dependent or unequal relationships with researchers.

Technology-related exclusion criteria were: prior sensitivity to VR technology; motion sickness; vertigo; seizures; epilepsy; and active nausea and/or vomiting. Women were recruited at the time of their attendance at the day surgery center via convenience sampling.

Randomization and blinding

Randomization for the sequence of active and passive content was carried out via randomized permuted blocks using Microsoft Excel 2016, with block sizes of 10 and a 1:1 ratio at the time of recruitment. Due to the nature of VRT and the fact that each type of content required explanation before administration, the allocation was not concealed from the researcher nor participant. No blinding was implemented in the analysis nor the interpretation of the results considering this was a proof of concept study.

Outcomes

The main primary outcome of this study was patient-reported pain scores. POP was evaluated using an 11-point numerical rating scale (NRS), which was verbally delivered to participants. This scale was chosen as it demonstrates validity and reliability in a clinical setting20,21. The NRS, a 10 cm horizontal line, was defined by the end points: 0 = “no pain at all”; oath, 10 = “worst pain I can imagine”. Women were asked to rate their pain by assigning it a number between 0 and 10, based on the above definitions. Patients were evaluated for the pain felt at the specific time points of 0, 10, 20 and 30 min.

As an additional primary outcome, we assessed the feasibility and acceptability of VRT in the immediate post-laparoscopy setting. Feasibility was assessed through recruitment and dropout rates, where it was determined a priori that a recruitment rate ≥ 50% and a dropout rate of ≤ 20% following randomisation would be necessary for the trial to be deemed a success. The outcome of acceptability was assessed through a post-intervention questionnaire aimed at determining patients’ opinions of the device and VRT as an analgesic. Women were asked to rate the experience of using the device, whether they would use it again, recommend it to others as well as other questions (Supplementary Materials).

Secondary outcomes included:

  1. 1.

    Opioid use, defined as the amount of PO opioids administered to patients were recorded at the time of administration by the trial investigator. Later, for analysis, they were converted to morphine sulphate equivalents via the Australian and New Zealand College of Anaesthetists (ANZCA) validated dose equivalence analgesic table22. PO opioids were analyzed based on the time period in which they were administered (time periods: prior to protocol; between 0–10; 10–20; and, 20–30 min).

  2. 2.

    Adverse outcomes, defined as nausea, active vomiting, dizziness, vertigo, seizures and were evaluated through a questionnaire administered to women at 0, 10, 20 and 30 min. For the trial to be deemed safe, it was determined a priori that the number of participants experiencing side effects must be ≤ 30% under a per protocol analysis. This threshold was determined with consideration to the level of post-operative side effects that may be anticipated within this population which can be a confounding factor due to the similar symptoms with cybersickness. Participants who did not complete the protocol are reported separately.

In addition, a pre-operative questionnaire was administered to patients to gather their demographic data as well as their perception towards pain and anxiety associated with the procedure (Supplementary Materials).

Study procedures

Eligible patients were approached prior to their surgery by the trial investigator (unrelated to their care), when written informed consent was obtained, and the trial procedure was explained.

Participants then underwent their laparoscopic procedure, which was carried out by an experienced surgeon (> 100 procedures per surgeon). The surgical technique was determined at the discretion of the operating surgeon and as per clinical guidelines. Similarly, the anesthetic care was provided by the anesthetic team at their discretion. This was chosen to allow the findings to be generalizable to any PO patient, in line with real world evidence.

Following surgery, patients were immediately taken to the post-acute care unit (PACU). The PACU clinician caring for the patient as well as the patient used an 11-point visual analogue scale (VAS) to define the patient’s cognitive state. The end points of the VAS were defined as: 0 = “non-responsive” and 10 = “fully awake”. A score of ≥ 5 was required from both the clinician and the patient for the participant to be randomized into the study. If a patient or clinician reported a VAS < 5 approximately 2 h after surgery, they were excluded from the study. A score of > 5 was chosen based on consensus opinion in this setting as it was intended that the patients would act as their own control for continued participation in the trial based on their own perception of their wakefulness. The clinician’s input would be to moderate this self-assessment based on their clinical experience in managing numerous post-operative women.

Once randomized, the participant’s bed was positioned between 10° and 45°. Following which, patients began the 30-min protocol that consisted of two 10-min segments of VRT interspersed with one 10-min washout period. For patients assigned to intervention group 1, this entailed active VRT (10 min), followed by a washout period (10 min) and then passive VRT (10 min). Intervention group 2 underwent passive VRT, followed by a washout period and then active VRT. The washout period consisted of routine clinical care and was intended to allow for the analgesic effects of VRT to dissipate. A 10-min washout has been deemed adequate in experimental studies for the effect of VRT to dissipate19.

During the study, participants were welcome to stop VR at any stage of the trial, in which case, they were recorded as a withdrawal and their reason for cessation was also documented.

VR content

Two pieces of VR content, delivered via the Oculus Go headset, were used during this trial, both of which were developed by ALO VR (Singapore) specifically for this trial. The pieces of content were “Sky Lights 2” and “Cosmic You”, the former being the active distraction content and the latter, the passive meditation content.

In Sky Lights 2 (Fig. 1), the user is placed lying down on their back in a quiet field, staring at a starry night sky with several unlit Chinese lanterns floating gently above. By focusing their gaze on a lantern, the user can set it alight, causing the lantern to rise upwards and away. Occasionally, as a reward for continued participation; a lit lantern will either set off a series of fireworks or form Lantern Festival shapes such as a dragon or a giant fish. Relaxing background music is also played to provide auditory stimulation. For this trial, user control was achieved by head tracking and lanterns were lit through triggers on either a Bluetooth hand controller or touchpad on the head-mounted display, based on user preference. Orientation to the device and instructions required approximately 60 s and the procedure itself only began once the headset was secured onto the patient and verbal confirmation was received that the game had started.

Figure 1
Figure 1

Picture of a user experiencing the Sky Lights 2 immersive pain distraction content.

In Cosmic You, participants are listening to a guided meditation with relaxing background music while looking at colorful shooting stars in the night’s sky. Users are not required to interact with this content but are encouraged to participate in the meditation.

The rationale for comparing the analgesic effects of active distraction VR versus passive meditation VR was to determine if there is a relationship between the therapeutic effect and the degree of interaction required by the participant. In a systematic review by Smith et al., it was identified that the analgesic effect is significantly greater for active VR than for passive VR, although no clinical studies had explicitly investigated this difference14. As such, this study is perhaps one of the first clinical studies to do so.

Statistical methods

Sample size

In order to assess the primary outcome of comparing the differences between the active and passive VR content and their respective effects on pain scores across the various time points, power calculations were prepared using a within subject repeated measures ANOVA model for 2 groups and 4 repeated measurements . The assumptions were an effect size of 0.23, a correlation between repeated measures of 0.5 and a non-sphericity correction of E = 1, β = 95% and α = 0.0523. This approximated a sample size of 30. Considering a 10% loss to follow up, 34 women were to be recruited into the study with n = 17 in each treatment arm. It was also established a priori that the trial will continue recruiting until the adequate numbers were available in each arm to also establish the level of drop out with VRT use.

Data analysis

The raw data for numerical variables in the study were assessed for their distribution using the Shapiro–Wilk test. Normally distributed data were expressed as mean (± standard deviation), while skewed data were expressed as median (inter-quartile range). Categorical variables were expressed as frequency counts (percentages).

For the primary outcome, comparing the effects of the active and passive VR contents on pain scores, two-way mixed model ANOVA analysis was employed to investigate differences between the groups. For each parameter, all data points: 0, 10, 20, and 30 min were assessed. The data were primarily assessed for the two-way interaction of treatment and time. If significant, post-hoc analysis with a Bonferroni correction for multiple comparisons was employed. In the event of a non-significant two-way interaction term, subsequent analysis for the effect of time and the effect of treatment were carried out.

For the secondary outcome, comparing the effects of active and passive VR content on opioid requirements, Mann–Whitney U test was used to analyze between group differences. To compare the effects of active distraction and passive meditation VR content on side effects, Fisher’s exact test or the corrected Fisher–Freeman–Halton exact test was employed.

For the remaining variables, the statistical test used to compare the intervention groups was dependent on the data. Continuous variables were analyzed for differences either via the independent sample t-test or the Mann–Whitney U test, depending on their distribution. Categorical variables were compared using either chi-squared test of homogeneity, Fisher’s exact test or the corrected Fisher–Freeman–Halton exact test, depending on the expected values ​​in the frequency table. In each of the aforementioned statistical tests, the null hypothesis was defined as no difference between the intervention groups’ mean, median or frequency.

Unless otherwise specified, the assumptions of all statistical tests were met. For all tests, statistical significance was set at an alpha level of 0.05.

Statistical analysis was completed on SPSS v24.0 and figures were created using GraphPad v7.0b.

Leave a Reply

Your email address will not be published. Required fields are marked *