Reducing Periprocedural Pain and Anxiety of Child Patients with Guided Relaxation Exercises in a Virtual Natural Environment: A Clinical Research Study

Abstract

Fear of needles is common among child patients. It causes stress and can lead to difficulty in procedures and future treatment avoidance. Virtual reality (VR) has emerged as a promising tool to reduce pain and anxiety non-pharmacologically. However, a research gap exists regarding what VR content is most effective in decreasing periprocedural stress. This article reports a VR feasibility study conducted with 83 child patients aged 8–12 years during a cannulation procedure. It has a between-subjects design with four groups, comparing deep breathing and mindfulness-based relaxation in a virtual nature environment (VNE) to passive VNE and standard care. The results from both relaxation exercise groups have been previously reported. This follow-up article adds findings from passive VNE and control groups, comparing all four for effectiveness and patient experience. The key findings highlight that deep breathing was highly effective according to heart rate variability (HRV) data, but less enjoyable than the mindfulness-based relaxation, which achieved higher patient satisfaction but was less effective according to HRV. Passive VNEs were pleasant but did not cause measurable stress reduction. All VR interventions improved patient experience over standard care. Relaxation exercises in a VNE reduce periprocedural stress more efficiently than passive VNEs or standard care in pediatrics.

1. Introduction

Almost all children are afraid of needles during the early years of their childhood. This fear tends to decrease as we grow older, but by the age of 10, the majority of children still suffer from it [1]. Bad treatment experiences tend to sustain or even increase this fear. Negative effects of it include patient and caregiver stress and anxiety before and during the treatment, increased difficulty in performing the related medical procedure for the patient, and increased likelihood of treatment avoidance in later life. This results in a need for methods to prevent or overcome needle-related distress during the procedure in pediatrics [1,2,3].

Traditional methods to address the common fear of needles—or its more severe form, needle phobia—include educating the child about the medical procedure and using a distraction to alleviate the stress during the procedure [4]. Sedative medication can also be used to address severe distress, but this requires additional resources from the healthcare provider and includes the risk of adverse effects for the patient [5]. In recent years, Virtual Reality (VR) interventions have shown strong promise to address pain and anxiety in clinical care, but while the use of VR distraction as a non-pharmacological intervention to reduce pain and stress in pediatrics is generally recommended, information regarding effective content for these medical VR interventions remains limited [6].

This article contributes to narrowing this research gap by reporting a pediatric clinical feasibility study with a medical VR intervention based on the use of guided relaxation exercises in a virtual natural environment (VNE) to address pain and anxiety in pediatrics. The study involved 83 eight to twelve-year-old child patients and their legal guardians and had a between-subjects design with four groups. While all child patients received a peripheral intravenous (IV) cannulation procedure [7] related to their treatment during the study, participants in three groups used different versions of the VR intervention developed for this study during the IV cannulation procedure. The fourth group was a Control group that received only standard care.

Previously, the authors have reported findings regarding the two guided relaxation exercise groups of this study. A Deep Breathing group of 21 participants used a deep breathing exercise in a VNE, with results including reduced stress levels during the VR intervention according to heart-rate variability (HRV) analysis [8] and good user experience among child patients, their caregivers, and pediatricians, with the perceived usability of the method being higher among patients with increased levels of anxiety or needle phobia [9]. A Mindfulness group of 20 participants used a mindfulness-based relaxation exercise with the HRV analysis indicating a slightly weaker but significant stress-reducing effect [10]. Children in this group reported a very positive user experience and experienced less pain than they expected, with results suggesting the best feasibility for patients with mild to moderate treatment-related anxiety. These previous articles studied the topic from within-subjects perspectives.

In this article, we report the results from the last two participant groups of this study—Passive VNE and Control groups, both consisting of 21 participants—and compare this data to the complete dataset. This report includes a detailed analysis of the effects and user experience of VNEs and guided relaxation exercises in reducing perioperative pain and anxiety in pediatrics. With this, we aimed to answer the following research questions: Do guided relaxation exercises in a VNE cause increased HRV in child patients during an IV cannulation procedure, indicating a stress-reducing effect? Does the type of relaxation guidance—deep breathing or mindfulness-based—affect these results, and do the results differ from using a passive VNE in this context? The key findings confirm the objective efficacy of guided relaxation exercises in a VNE, but also reveal a conflict between this efficacy and subjective enjoyment regarding the deep breathing exercise in a VNE.

2. Background

VR (Virtual Reality) refers to immersive virtual environments that are commonly experienced with a head-mounted display (HMD). It is the most traditional form of extended reality (XR) technologies [11]. VR experienced with an HMD replaces the visual information from the real world with a virtual one. This virtual environment reacts to the user’s head movements similarly to how the real world does, creating a convincing illusion of being in another place, especially if accompanied by matching auditory input. This fundamental feature of VR experienced with an HMD makes it a very powerful distraction method.

2.1. Virtual Reality (VR) in Acute Pain Management

As distraction is already a common method for addressing acute short-term pain in clinical care [4], distraction by VR can be considered as a technologically advanced tool for distraction, rather than a completely new treatment method. The use of distraction in pain management is based on our limited cognitive capacity as human beings and the importance of cognition in mediating experiences of pain—the more the distraction method used manages to capture the available attentional capacity, the less is available for the experience of pain [12]. Distraction is especially suitable for addressing acute pain in infants and children, but current understanding remains limited regarding which distraction method is the most suitable for clinical use [13,14,15].

Compelling evidence on the benefits of VR in this context has emerged during the last few years. One recent narrative review forecasts VR as a fundamental component of future pain management protocols due to its unique capabilities regarding distraction, education, and therapeutic engagement with the patients, which effectively complement traditional clinical methods [6]. While the statement of the review is bold, it is backed by recent systematic reviews reporting strong evidence for the pain-reducing effects of VR interventions [16,17,18,19,20,21,22].

Addressing periprocedural (near and during the medical procedure) pain and anxiety of child patients with VR is an especially promising use context, as VR pain management methods are more effective when addressing a clear, specific pain source, such as an IV cannulation procedure [6]. This analgesic effect is especially strong with patients who experience moderate to severe pain, and children seem to benefit more than older patients [17].

Only limited information exists on how different VR contents affect the efficacy of these interventions. The contents used in the studies vary from 360-degree videos to VR games, with some studies suggesting the superiority of interactive VR [23,24,25]. Some meta-analyses contradict this, reporting a larger pain-reducing effect for passive VR methods [17,21]. Other reviews favor interactive content [16,20] or found no differences between the effects of different VR types [18]. The mixed findings can be explained by the large variance of different VR content used in the studies and the rather simplified division between interactive and passive VR.

The current literature highlights the need for clinical studies evaluating the efficacy and feasibility of different VR contents for pain and anxiety management, with child patients being one of the most under-researched user groups. Study designs targeting specific use cases and incorporating objective research metrics and a control group are recommended [6,16,17,19,21,22,26].

2.2. Challenges of Using Medical VR Applications in Pediatrics

While strong evidence for using medical VR applications to effectively address perioperative stress in pediatrics exists [27,28]—with the benefits including good acceptance among both the child patients [29] and healthcare professionals [30], and providing both economical [31,32] and non-pharmaceutical alternative to sedative medication [33]—several challenges also exist that need to be addressed.

The possibility of cybersickness is the most well-known side effect of VR experiences, which is similar to motion sickness both regarding the symptoms and the affective factors [34,35]. Use of VR devices in a healthcare setting also causes a need for training and technical support for the pediatric staff, who are traditionally not accustomed to using these methods in their work [30,36]. Additional factors to consider are the need for true multidisciplinary collaboration when developing medical VR applications [37] and the potential reduction in novelty value in repeated use of these VR methods [36]. The need for infection control can cause limitations for clinical use, especially regarding surgical procedures [38].

2.3. Virtual Natural Environment (VNE) in Pain and Anxiety Management

A recurring theme in the relevant studies is the use of nature-based VR interventions. In a systematic review of VR for acute pain management from 2022, 19 out of 23 included studies were based on the use of VNEs in some form. Out of these, 11 studies used a passive VNE, while 8 also included some other affective factor in the VR content, such as music, guided relaxation, or interactive games and activities within the nature experience [20]. With passive VNEs, we refer to VR nature experiences which consist only of experiencing virtual nature without other affective factors within the application.

The use of VNEs is based on the virtual replication of experiencing real nature environments for similar effects, especially regarding anxiety reduction [39]. For acute pain and anxiety management, it offers a new stress-reducing factor in addition to the benefits of distraction by VR. VNEs can provide similar but reduced positive effects as real nature environments, being especially suitable for reducing negative affect such as stress and anxiety, while only real nature experiences seem to evoke positive emotions [40,41,42,43]. The stress-reducing effect seems to be equally strong between VNEs and real nature experiences [44]. The presence of natural sounds, such as birdsong, wind, and water sounds [45] and natural light seems to be among the key factors in achieving stress reduction [42,46].

Most of the studies regarding VNEs in pain and anxiety reduction are based on green environments [41], with forest environments being especially effective for cognitive recovery and open environments, such as plains and fields, being better for emotional recovery [43]. Blue nature environments, as environments that include a water element, seem to have similar yet less studied stress-reducing qualities [47,48]. A recent systematic review goes as far as suggesting water elements to be among the key factors that contribute to stress reduction in VNEs [42], but very few studies exist that compare the effects of blue nature environments to other nature environments. In addition, it has been suggested that cultural familiarity and personal experiences regarding the nature type affect the stress recovery in VNEs [49], but evidence contradicting this has also emerged [50].

One recent narrative review [51] proposes a theoretical framework for nature-based analgesia with five key features the VNE should meet: (i) enable the sense of being away (immersive presence away from stressors), (ii) address individual’s needs and desires (fitting the activity or goal), (iii) include potential for exploration (optional movement and/or audiovisual complexity and biodiversity), (iv) be engaging and fascinating without demanding too much attention, and (v) be associated with safety (no signs of threats or danger). Smith et al. conclude that meeting these features can result in activation of brain areas involved in mediating stress, cognition, mood, and autonomic function, thus reducing both anxiety and experiences of pain.

2.4. Guided Relaxation Exercises in a Virtual Natural Environment (VNE)

Contemplative activities, such as deep breathing and mindfulness exercises, have the potential to improve the pain and anxiety-reducing effects of VNEs while also being unlikely to cause cybersickness due to the minimal amount of movement required [52]. They also fit well within the proposed theoretical framework for nature-based analgesia [51].

Deep breathing, which refers to slow and deep conscious breathing, can improve autonomic function by reducing sympathetic activity and diastolic blood pressure, effectively decreasing stress and anxiety [53,54]. Positive results have been found with combining deep breathing exercises with VNEs, with findings including positive user experience and sense of presence, reduced stress and improved relaxation, and a close connection between respiration and attention [55,56]. Furthermore, no differences seem to exist between VR and non-VR deep breathing exercises regarding the effects on stress, anxiety, and mood [57]. Biofeedback methods have been studied in this context for improved focus on the desired breathing rhythm with positive findings, but the studies have rarely compared biofeedback to other methods of breathing guidance [58,59,60,61].

Mindfulness exercises can relax the mind by guiding the attention to the details of the present moment without judgment [62,63]. In clinical healthcare, it can be used to decrease stress by teaching the patient to calmly accept negative sensations [64,65]. However, as VR experienced with an HDM is fundamentally distracting, mindfulness-based meditation exercises in VNEs often guide the attention towards elements in the VR, to bodily sensations that are not related to the medical procedure, or even to imagination and memories. This approach has been used to effectively reduce experienced anxiety [66] and pain in a pediatric setting [67,68].

More detailed background information regarding deep breathing [8,9] and mindfulness-based [10] methods in this use context can be found in the previous articles.

3. Materials and Methods

The clinical study this research article is based on was conducted between 2022 and 2025 in two Finnish hospitals with a total of 87 child patients of 8–12 years of age. Participating children were due to receive an IV cannulation as part of their normal treatment. The child patients and their caregivers were contacted on the day of their treatment and offered the possibility to participate in this research. Four participants were excluded from this article due to missing data or receiving a blood draw instead of IV cannulation.

The study used a VR application designed for this purpose (VirNE, Virtual Natural Environments) to conduct research with a between-subjects design and four participant groups: Deep Breathing, Mindfulness, Passive VNE, and Control. Participants in the first three groups used a variation in the VirNE application during the IV cannulation procedure, while the Control group received only standard care. The aim was to compare the effects and user experience of two different relaxation exercises in a VNE to a passive VNE and standard care to evaluate the feasibility of these methods for addressing acute pain and periprocedural anxiety in pediatric use.

Participants were divided into groups in the order of participation. The Deep Breathing group (n = 21) used a deep breathing exercise in a VNE, and the within-subjects results were published in 2023 [8,9]. Mindfulness group (n = 20) used a mindfulness-based exercise in a VNE [10]. This article extends these prior articles by reporting the last two groups of this study with a between-subjects design: the Passive VNE group (n = 21), who used VirNE without a relaxation exercise, and the Control group (n = 21), who received only standard care with no VR intervention.

The rationale for the non-randomized participant allocation was the study’s preliminary nature, the positive findings of the previous articles, and the need to verify those findings with control data. The Control group was the last participant group of this study. Participants in this group were informed of their group allocation during initial contact for ethical reasons, although this increased the risk of allocation bias.

While the information regarding Deep Breathing and Mindfulness groups has already been published, this article uses this data in the between-subjects analysis. To study possible negative effects between the groups, the earlier decision to exclude two outliers from the Deep Breathing group [8,9] was overruled.

Although this section provides a complete summary of the protocol, a more detailed explanation regarding the guided relaxation exercises is available in the previous articles.

3.1. Virtual Natural Environments (VirNE) Application

Virtual Natural Environments (VirNE) is a VR relaxation application designed for this research with Unity [69] by a multidisciplinary team of experts from the fields of child psychiatry, human-technology interaction, and media production. VirNE was designed to address acute pain and anxiety in pediatrics. It is based on combining the relaxing effects of VNEs with the relaxing effects of guided relaxation exercises. In this study, the child patients used a Meta Quest 2 HMD (Meta, Menlo Park, CA, USA) to experience VirNE.

VNEs in VirNE are based on stationary 360-degree video recordings from Finnish nature environments, recorded with an Insta360 Pro in 6K 3D 30 fps format during summer 2021. These videos are presented with a stereo sound file of local nature sounds, which mainly consists of various summer birds singing. This same audio file was used for all VNEs used in this study. When VirNE is used as a passive VNE, no other elements are included in the VR experience. Figure 1 presents the five VNEs used in this study.

VirNE includes two relaxation exercises: a deep breathing exercise and a mindfulness-based guided relaxation exercise. These exercises are studied and reported in previous articles, which also feature more in-depth coverage of the exercise designs and the development of the application [9,10]. Both exercises are based on dialog guidance and are 6–7 min long, including a pre-designated point for conducting a medical operation at about 4 min into the exercise. Figure 2 illustrates the differences between the exercises and the passive VNE experience.

3.2. Research Methods

The research protocol assessed both subjective and objective data to analyze the impact of the studied method. It combined insights from human–computer interaction and medical perspectives. Subjective data were collected through questionnaires, interviews, and observations made by the nurses conducting the research, while the objective data involved heart rate (HR) recordings to identify potential changes in participants’ stress and relaxation levels. HR data was collected with a Polar H10 Heart Rate Sensor and recorded with a Heart Rate Monitor application designed for this research [8], which enabled local data recording without third party involvement and marking specific points of interest in the recording. Two nurses were present in the research situation, one attending to the child patient and their legal guardian, while the other managed the technical equipment used in the research. Figure 3 visually presents the research scenario.

All participant groups in this study followed a three-phase research protocol: pre-procedure, VR intervention and IV cannulation, and post-procedure. The pre-procedure phase began with clinical measurements and connecting the equipment. The HR recording was started as soon as the child patient reached a half-sitting posture in a hospital bed, and the equipment was ready. A structured background interview of the child patient and the caregiver followed, along with filling out digital questionnaires with a tablet. Excluding the Control group, the child patients then selected their preferred VNE for the VR intervention, and patients in guided relaxation groups also selected the avatar character for the experience.

At the beginning of the next phase, VR intervention and IV cannulation, the patient had worn the HMD, and the VR intervention began. For participant groups with a relaxation exercise (Deep Breathing and Mindfulness), this phase had a fixed length of about 6 min and a predesignated timeframe for conducting the IV cannulation about 4 min into the exercise. For the Passive VNE group, the nurses let the child patient experience the VNE for a few minutes before proceeding to the IV cannulation. After the medical procedure, the child could continue experiencing the VNE for a minute or two. For the Control group, this phase did not include VR intervention, and they received standard care during the IV cannulation. This included an option to use their own smartphone for distraction, and it was also suggested if the patient was in clear distress. For all groups, this phase included a verbal description of the IV cannulation process by the nurses while they conducted it, as performed in standard care in Finland.

The post-procedure phase began when the HMD was removed, or in the Control group when one minute had passed from the IV cannulation. This phase consisted of digital questionnaires and further clinical measurements for the child patient, followed by a brief unstructured interview. When everything was ready, the nurses ended the HR data recording and concluded the final data collection phase, and participants proceeded with their normal treatment.

Further data for this article were acquired with two formal structured interviews with the head of the pediatric team conducting the research, in 2023 and in early 2025. The one in 2023 consisted of 23 questions regarding how well the research protocol was followed, the nurses’ experience regarding the VirNE application, and the evaluation between the two relaxation exercises studied in the first two groups (Deep Breathing and Mindfulness). The interview in 2025 consisted of 10 questions and had similar themes, but the focus was on Passive VNE and Control groups.

3.3. Measures

3.3.1. Measures from Interviews

Two measures are derived from the participant interview data: the estimated level of needle phobia [3] and general treatment experience, with the evaluations for these measures being conducted by the primary authors of this article.

The needle phobia was assessed from the pre-procedure interviews with a 3-point Likert scale (0–2): 0 indicated no signs of needle phobia; 1 suggested mild indications of needle phobia; and 2 implied clear signs of needle phobia. General treatment experience was measured from post-procedure interviews with three possible values: positive, neutral, and negative.

3.3.2. Screen for Child Anxiety Related Emotional Disorders (SCARED) Questionnaire

SCARED questionnaires [70] were used in the pre-procedure phase of this research for both the child patient and their legal guardian to evaluate stress and anxiety symptoms of the child patient in all participant groups. These questionnaires consist of 41 statements regarding negative emotions in various everyday situations. The statements are evaluated with valued answers—whether the emotion is experienced constantly (2), occasionally (1), or never (0). A total score of 25 or more can indicate the presence of an anxiety disorder. This article uses a SCARED Average value per participant, which was calculated by dividing the sum of the answers of child patients and their caretakers by two.

3.3.3. Adapted Visual Analog Scale for Anxiety (VAS-A) Questionnaire

Child patients answered adapted VAS-A questionnaires [71,72] in the pre- and post-procedure phases of this research, regarding their expectations and experiences of pain and anxiety during the IV cannulation. The adaptation was based on an earlier study with a VR relaxation application [73] and was first presented in the article about the deep breathing exercise [8].

Both the pre- and post-procedure versions included two questions, formatted as “How much [pain/anxiety] do you expect to experience/you experienced during the procedure?”. A 7-point Likert scale (0–6) was used to measure the answers, with 0 being labeled as “Not at all”, and 6 as “Extreme [pain/anxiety]”.

3.3.4. Customized User Experience Questionnaire

A customized User Experience questionnaire was used in the post-procedure phase of the VR intervention groups (Deep Breathing, Mindfulness, and Passive VNE). The questionnaire was a simplified version of the VR user experience methods used in prior research [73] and was first presented in the article regarding deep breathing exercises [9]. It consisted of four questions targeted for 8–12-year-old children: “It was easy for me to adjust to being in VR” (Q1); “It was easy for me to focus on the exercise” (Q2); “The application was helpful/useful for me” (Q3); “The application was boring for me” (Q4). A 7-point Likert scale (0–6) was used to measure the answers, with 0 being labeled as “Completely disagree” and 6 as “Completely agree”.

3.3.5. Heart Rate Variability (HRV) Analyses

Heart rate (HR) data were collected during all three phases of the research from all participants. This data was used for HRV analyses to objectively measure the changes in the stress levels of the child patients [74,75]. The HRV data variables used in these analyses included the standard deviation of normal-to-normal intervals (SDNN), root mean square of successive differences between normal heartbeats (RMSS), square root of Baevsky’s Stress Index (SI), and the low frequency to high frequency ratio (LF/HF ratio).

SDNN and RMSSD values depict the state of the parasympathetic nervous system, which is related to the subject’s level of relaxation, with higher values suggesting a more relaxed state of mind [76]. SI reflects cardiovascular system stress and the activity of the sympathetic nervous system, which is related to stress and responsible for the “fight or flight” response, with higher values suggesting increased stress [77]. LF/HF ratio depicts changes between sympathetic and parasympathetic dominance, with higher values suggesting higher stress and lower values suggesting a more relaxed state of mind [78].

Two HRV analyses were conducted for this article: the VR intervention HRV analysis and the Periprocedural HRV analysis. The VR intervention analysis studied the effect on HRV during the VR intervention but before the IV cannulation, while aiming to avoid the stress-inducing effects of the IV cannulation. In contrast, the Periprocedural analysis studied the effect near and during the IV cannulation, aiming to compare the effects on HRV between the groups during the immediate negative affect of the medical procedure.

The VR intervention analysis was based on the earlier articles [8,10] and compared the data from the Passive VNE and Control groups to the earlier results. This analysis focused on two specific points of interest—S1 (120 s) was collected in the pre-procedure phase between 5:00–7:00 into the HR recording and served as a baseline, while S2 (120 s) was collected in the VR intervention and IV cannulation phase between 2:45 and 0:45 (m:ss) before the IV cannulation in Passive VNE and Control groups. In Deep Breathing and Mindfulness groups, the location of the S2 (120 s) was calculated from the start of the exercise, being located approximately between 3:15 and 1:15 (m:ss) before the IV cannulation. The 30 s difference in the location of the S2 (120 s) was due to a shorter timeframe between the VR intervention and the IV cannulation in the Passive VNE group.

The Periprocedural HRV analysis was conducted to compare the groups in moments close to the IV cannulation. This analysis used data from all four groups to compare the stress and relaxation levels in the moments close to the IV cannulation. It included 10 samples: S1–S10 (30 s). S1 (30 s) was considered as the baseline and was located between 5:00 and 5:30 (m:ss) into the HR recording. S2–S10 (30 s) consisted of a timeframe between 75 s before to 75 s after the cannulation, with each sample overlapping with the previous one by 15 s, and S6 (30 s) being centered on the moment of the IV cannulation. Figure 4 illustrates these samples.

3.4. Statistical Analysis

All analyses were conducted with IBM SPSS Statistics (Version 29) with the alpha threshold (p-value) for statistical significance set to p < 0.05. No outliers were excluded from the analyses. Missing data are summarized at the beginning of the Section 4, and details of partially missing data are reported within related subsections.

Data characteristics are described using mean and standard deviation (SD) for normally distributed data, and median with interquartile range (IQR) for non-normal distributions. For consistency, all values within a single dataset are reported using the same format. Normality was assessed using the Shapiro–Wilk test.

As no outliers were excluded, non-parametric statistical methods were used to avoid false positives. Wilcoxon signed-rank tests were used to compare two related conditions. When comparing groups, Mann–Whitney U tests were used to compare two groups, and Kruskal–Wallis H tests to compare more than two groups. Effect size calculations and covariance analyses (ANCOVA) were not performed due to small and uneven group sizes, the presence of outliers, and the use of non-parametric methods, as these factors could lead to unstable and potentially misleading estimates.

3.5. Research Ethics Approval and Clinical Trials Registration

This study was approved by the Authority Ethics Committee of Pirkanmaa Welfare Area (R21068L) and Finnish Medicines Agency (2021/007366).

The first participants of this study were enrolled in May 2022. Clinical trial registrations were not a common requirement for clinical VR feasibility studies in Finland at this time, being reserved mainly for pharmaceutical interventions. During the relatively long data-gathering phase of this study, the implementation of these policies has changed.

Because of this, the study was retrospectively registered with the ISRCTN registry (ISRCTN12712970; https://www.isrctn.com/ISRCTN12712970) accessed on 25 May 2025. This registry entry is related to a broader feasibility study, which also includes other sub-studies that are excluded from the scope of this article.

4. Results

A total of 87 child patients and their legal guardians participated in this research. These results include 83 participants in four groups: Deep Breathing (n = 21), Mindfulness (n = 20), Passive VNE (n = 21), and Control (n = 21). Four participants were excluded from the results due to missing data or receiving a venipuncture for blood draw instead of IV cannulation. The number of participants included in the results varies between the measures.

Table 1. Participants included and excluded from the results.

Data Group	Included (n) 1	Excluded (n) 1	Primary Reason for Exclusion
Enrolled participants	83	4	No IV cannulation
SCARED Average	82	1	Missing data
VNE selection	58	4 (+21)	Missing data (+Control)
Adapted VAS-A questionnaire: Pain	81	2	Missing data
Adapted VAS-A questionnaire: Anxiety	80	3	Missing data
Customized User Experience questionnaire	59	3 (+21)	Missing data (+Control)
VR intervention HRV analysis	69	14	Insufficient data (in Passive VNE)
Periprocedural HRV analysis (S1–S5)	79	4	Insufficient data quality
Periprocedural HRV analysis (S6–S10)	67	16	First IV cannulation attempt failed

¹ Including participants from the previously reported Deep Breathing and Mindfulness groups.

summarizes the included and excluded participants along with the primary reasons for exclusion.

The background information of the patients in all groups is presented in

Table 2. Participant background information in all groups.

Data Group	Deep Breathing 1	Mindfulness 1	Passive VNE	Control
Participants (n)	21	20	21	21
Age, mean (SD)	9.95 (±1.24) years	10.65 (±1.27) years	9.95 (±1.32) years	9.86 (±1.39) years
Sex division	13 females, 8 males	12 females, 8 males	12 females, 9 males	7 females, 14 males
SCARED Average, median (IQR)	15.00 (10.75)	11.00 (10.25)	11.00 (9.75) 2	15.50 (14.75)
Prior IV cannulations (0–3/4+)	11/10	14/6	16/5	15/6
Fear of needles (no/mild/clear)	7/7/7	8/6/6	3/9/9	7/5/9
Prior virtual reality experience (yes/no)	12/9	17/3	11/10	-
Motion sickness sensitivity (low/medium/high)	11/6/4 (participants)	15/4/0 (participants)	16/5/0 (participants)	-

¹ Data from Deep Breathing and Mindfulness groups has been previously reported, but now includes previously excluded outliers. ² Data missing from one participant.

. Passive VNE and Control groups were fairly similar to the previously published groups. Most notable differences were the male-dominant sex division in Control and higher sensitivity for motion sickness in the Deep Breathing group. Other differences include a higher prevalence of needle phobia in the Passive VNE and Control groups, and a low amount of VR experience and slightly higher mean age in the Mindfulness group. SCARED Average was also somewhat higher for participants in the Deep Breathing and Control groups, but the median values of all groups were well below the threshold of anxiety disorder indication (25).

4.1. Virtual Natural Environment (VNE) Selection

VNE selection results include pooled data from all VR intervention groups. This data was missing from 4 patients, resulting in a total of 58 participants. Out of these, 35 were female and 23 were male children. As illustrated in Figure 5, blue environments (A and B) were a notably common choice, being selected by 70% of the participants, while only 30% of participants selected a green environment (C, D, or E). No significant sex-based differences were found in the VNE selections.

4.2. Interviews and Observations

4.2.1. Passive VNE Group

Out of 21 participants in the Passive VNE group, 17 expressed a positive experience in the post-procedure interviews, while the experience of three participants was evaluated as neutral, and one had a negative experience. 20 participants gave positive comments regarding the VR experience. These were mostly short general praises regarding the nature experience or the VR headset. Common statements used to describe the experience include: “Good” (10), “I liked it” (7), “Nice” (5), and “Helpful” (3). Regarding more specific comments, 12 participants gave positive remarks regarding the VNE, with three specifically mentioning nice visuals and three mentioning good sounds. Five participants stated wanting to use the VirNE application again in the future, and two mentioned liking how the HMD blocked the view from the cannulation.

“It was pretty nice. I did not pay attention to the cannulation. The landscape was pretty nice, the sunshine and the rocks were nice.” [Participant, 11 years old]

Three participants reported or were observed to have difficulties concentrating on the VR content, with two trying to observe the cannulation process instead. The participant with the negative experience started to relax at the beginning of the VR experience, but quickly pulled the hand away and started crying just before the cannulation after seeing the needle from the narrow space between the face and the HMD. The HMD was then removed, and the participant needed the comfort of the caregiver and 15 min of conversation before accepting the cannulation. Another participant experienced brief nausea in the post-procedure phase, but the caregiver still evaluated that the VR intervention helped with the cannulation.

“I liked the VR goggles, the picture, and the sounds, but since I’m not afraid of needles, it was not so useful for me.” [Participant, 12 years old]

Only very few negative comments were obtained from the post-procedure interviews. However, the participant with a negative experience did not participate in the post-procedure phase. One participant found the VR experience to be a bit boring and longed for additional animal content to the VNE, such as birds flying or fishes jumping from the water. Another one longed for better graphics in VR, and one child expressed a desire for foreign landscapes such as the Alps.

4.2.2. Control Group

Out of 21 participants in the Control group, three expressed a positive experience, while 14 were evaluated as having a neutral experience, and four had a clearly negative experience. The large number of neutral experiences is partially explained by the lack of detailed comments from the participants—while all participants were asked about the experience, the responses of this group were brief and often insignificant.

Six participants were observed to be clearly nervous, with three resorting to distraction by their own smartphones, from which they watched various material of their preference. The procedure was delayed for these six patients due to the need to calm their nervousness, with patient symptoms including being visibly afraid (4), excessive talking and moving (2), crying (1), and not answering verbal questions (1). These patients needed a considerable amount of convincing before allowing the cannulation to be conducted.

4.2.3. General Treatment Experience Between the Groups

Over 75% of the participants in all three VR intervention groups had a positive treatment experience, while standard care resulted in a positive experience for less than 15% of the patients. Most patients receiving standard care had a neutral experience. While all groups had at least one participant with a negative experience, Control had the highest number of negative cases (4). Negative cases in VR intervention groups were all related to low adherence to the intervention.

Table 3. General treatment experience of all four participant groups.

Data Group	Deep Breathing 1	Mindfulness 1	Passive VNE	Control
Participants (n)	21	20	21	21
General treatment experience (positive/neutral/negative)	16/2/3	16/3/1	17/3/1	3/14/4

¹ Data from Deep Breathing and Mindfulness groups has been previously reported, but now includes previously excluded outliers.

presents these differences in general experience between the groups.

4.3. Questionnaires

4.3.1. Adapted Visual Analog Scale for Anxiety (VAS-A) Questionnaires

Four groups with a total of 81 participants are included in the Adapted VAS-A results. The Passive VNE group (n = 20) is missing data from one participant. The Control group includes 21 participants, but data regarding anxiety is excluded from one participant due to highly conflicting data between VAS-A results and research observations. From the previously reported groups, the Deep Breathing group includes all 21 participants, while the Mindfulness group includes 19 participants, with data from one participant missing.

Table 4. Adapted Visual Analog Scale for Anxiety (VAS-A) results in all groups.

Data Group	Deep Breathing 1	Mindfulness 2	Passive VNE	Control
Participants (n)	21	19	20	21
VAS-A Pain expectation, mean (SD)	2.52 (±1.75)	2.63 (±1.80)	2.30 (±1.45)	2.90 (±1.37)
VAS-A Pain experience, mean (SD)	2.29 (±1.68)	1.42 (±1.39)	1.50 (±1.57)	2.00 (±1.70)
VAS-A Anxiety expectation, mean (SD)	1.52 (±1.66)	1.68 (±1.95)	1.65 (±1.42)	2.85 (±1.73)
VAS-A Anxiety experience, mean (SD)	1.57 (±1.86)	1.11 (±1.63)	1.00 (±1.41)	1.65 (±1.84)

¹ Data from the Deep Breathing group has been previously published, but now includes two participants previously excluded as outliers. ² Data from the Mindfulness group has been previously reported.

presents mean values from Adapted VAS-A results for all participant groups. Value 0 represents “not at all” and 6 “extreme pain/anxiety” on a 7-point Likert scale.

The mean pain expectation was within 0.60 points between all groups, with Control having the highest mean value. VR intervention groups expected the procedure to cause relatively low anxiety and moderate pain, while the Control group expected moderate anxiety and pain. The mean pain experience was lower than expectations in all groups, with the highest difference found in the Mindfulness group and the smallest in the Deep Breathing group. Mean anxiety decreased in all groups except Deep Breathing, where anxiety experiences were similar to expectations.

Statistical analysis between the four groups revealed a significant difference χ²(3) = 7.998, p = 0.046 from expected anxiety, with pairwise comparison revealing higher anxiety expectations for Control in comparison to the Deep Breathing (z = 2.492, p = 0.013), Mindfulness (z = −2.303, p = 0.021) and Passive VNE groups (z = 2.045, p = 0.041). No significant differences were found in expected pain or experiences between the groups. Patients receiving only standard care expected a more anxious experience than patients who knew they would use the VR intervention during the procedure.

A statistically significant difference between the two conditions regarding anxiety was found from Passive VNE (z = −2.565, p = 0.010) and Control (z = −2.572, p = 0.010) groups, and the conditions regarding pain were close to a relevant difference (p = 0.051) in both groups. Previously published results found no significant differences between the conditions regarding pain or anxiety in the Deep Breathing group [8], while in the Mindfulness group, the decrease was statistically significant regarding pain (p = 0.037) but not regarding anxiety [10]. No significant differences were found from the differences between the expectations and experiences of the four groups, but pairwise comparison revealed a significant difference in anxiety between the Deep Breathing and Control groups (z = −2.242, p = 0.025).

From the patient perspective, mindfulness-based relaxation seems to feel better for pain and anxiety reduction than deep breathing exercises in a VNE, which performed poorly in these subjective evaluations. Passive VNE reduced anxiety from expectations and fared well in pain ratings, while the positive performance of standard care can be explained by higher stress expectations.

4.3.2. Customized User Experience Questionnaire

Three groups with a total of 59 participants are included in the user experience results. The Passive VNE group includes 20 participants, with data from one participant missing. From the previously reported groups, Deep Breathing includes 20 participants, with data from one participant excluded due to a user input error [9], and Mindfulness includes 19 participants, with data from one participant missing [10]. The mean values for the four questions in this questionnaire for all three VR intervention groups are presented in

Table 5. Customized User Experience Questionnaire results in all VR intervention groups.

Data Group	Deep Breathing 1	Mindfulness 2	Passive VNE
Participants (n)	20	19	20
Q1: It was easy for me to adjust to being in VR, mean (SD)	4.70 (±1.89)	5.21 (±1.13)	5.00 (±1.56)
Q2: It was easy for me to focus on the exercise, mean (SD)	4.15 (±1.84)	4.68 (±1.16)	5.00 (±1.21)
Q3: The application was helpful/useful to me, mean (SD)	3.50 (±2.14)	4.68 (±1.45)	5.40 (±1.05)
Q4: The application was boring for me, mean (SD)	1.25 (±1.80)	1.00 (±1.70)	0.30 (±.73)

¹ Data from the Deep Breathing group has been previously published, but now includes two participants previously excluded as outliers. ² Data from the Mindfulness group has been previously reported.

. A value of six indicates complete agreement and zero indicates complete disagreement with the statement.

Passive VNEs were very well received according to Q1, Q3, and Q4 results, and while Q2 results included six participants who reported some problems regarding focus, the mean value was still higher than in the guided relaxation groups. Statistical analysis was conducted to compare the results between the groups. A statistically very significant difference χ²(2) = 10.978, p = 0.004 was found in the answers to Q3, where pairwise comparison revealed that passive nature experience felt more helpful than guided deep breathing (z = 3.313, p < 0.001). While no significant differences were found between the groups from the answers to Q1, Q2, or Q4, the difference in Q4 between the passive VNE and the guided relaxation exercises is notable, suggesting that passive VNEs felt more interesting to the children.

4.4. Heart Rate Variability (HRV) Analyses

4.4.1. The VR Intervention HRV Analysis of All Groups

The VR intervention analysis included 69 participants in four groups. The Passive VNE group had only 11 participants eligible for this analysis, due to the data of the excluded participants having bad data quality or an insufficient amount of time between the start of the VR intervention and the IV cannulation. The Control group had 20 eligible participants, with one participant excluded due to insufficient data quality. From the previously reported groups, Deep Breathing included 20 participants, including the previously excluded two outliers [8], and the Mindfulness group included 18 participants [10]. Data from three participants were missing or excluded due to poor quality in these groups. The median values of the HRV variables of S1 and S2 (120 s) of all four groups are presented in

Table 6. The median values and interquartile ranges (IQR) of HRV variables in S1 and S2 (120 s) for all four participant groups.

Data Group	Deep Breathing 1	Mindfulness 2	Passive VNE	Control
Participants (n)	20	18	11	20
HR S1 (120 s) median (IQR) (bpm)	87.50 (18.75)	92.50 (19.50)	87.00 (16.00)	82.50 (18.75)
HR S2 (120 s) median (IQR) (bpm)	85.00 (22.25)	91.50 (24.50)	91.00 (26.00)	87.50 (17.00)
SDNN S1 (120 s) median (IQR) (ms)	36.25 (22.10)	42.30 (34.28)	37.00 (21.90)	58.60 (25.85)
SDNN S2 (120 s) median (IQR) (ms)	62.70 (31.78)	59.10 (53.35)	34.80 (28.50)	51.75 (41.38)
RMSSD S1 (120 s) median (IQR) (ms)	32.15 (29.63)	32.00 (38.40)	28.80 (32.30)	53.55 (45.95)
RMSSD S2 (120 s) median (IQR) (ms)	53.50 (44.33)	49.35 (56.65)	27.90 (26.50)	49.55 (49.65)
LF/HF ratio S1 (120 s) median (IQR)	1.554 (1.962)	1.021 (1.487)	1.250 (2.125)	0.955 (1.290)
LF/HF ratio S2 (120 s) median (IQR)	0.456 (0.532)	1.055 (1.377)	1.467 (1.910)	1.506 (1.661)
SI S1 (120 s) median (IQR)	14.00 (6.08)	12.80 (10.07)	13.70 (5.70)	9.15 (4.27)
SI S2 (120 s) median (IQR)	8.40 (5.30)	9.05 (11.80)	14.00 (9.80)	9.70 (6.40)

¹ Data from the Deep Breathing group has been previously published, but now includes two participants previously excluded as outliers. ² Data from the Mindfulness group has been previously reported.

.

Statistical analysis was conducted to evaluate differences in the baseline (S1, 120 s) condition between the groups. Significant differences were found from SDNN χ²(3) = 8.190, p = 0.042, while HR (p = 0.053), RMSSD (p = 0.064), and SI (p = 0.051) were close to statistical significance. Pairwise comparisons revealed significant differences between Control and other groups; Deep Breathing (SDNN z = −2.522, p = 0.012; RMSSD z = −2.085, p = 0.037; SI z = 2.404, p = 0.016); Mindfulness (HR z = 2.682, p = 0.007; SDNN z = −2.323, p = 0.020; RMSSD z = −2.346, p = 0.019; SI z = 2.250, p = 0.024); and the differences with Passive VNE were close to significance (SDNN p = 0.072; RMSSD p = 0.055; SI p = 0.058). The differences in baseline values are likely related—at least partially—to the male-dominant sex division of Control [79].

Comparing the S1 and S2 (120 s) conditions, a statistically significant difference was found in HR for both Passive VNE (z = 2.405, p = 0.016) and Control (z = 2.447, p = 0.014) groups. No significant differences were found between the conditions from SDNN, RMSSD, LF/HF ratio, or SI in either of these groups. The elevated HR close to the procedure indicates increased stress, and all HRV variables indicate a minor shift towards stress in S2 (120 s), both with passive VNE and standard care. Regarding the earlier groups, median HR decreased with both guided relaxation exercises, and the previously reported findings indicate reduced stress and increased relaxation during the VR intervention for both Deep Breathing (SDNN, p < 0.001; RMSSD, p = 0.002; LF/HF ratio, p = 0.010; SI p < 0.001) [8] and Mindfulness (SDNN p = 0.006; SI p = 0.037) [10] groups.

4.4.2. The Periprocedural HRV Analysis of All Groups

The Periprocedural HRV analysis included measurements from 79 participants in four groups: Deep Breathing (n = 20), Mindfulness (n = 18), Passive VNE (n = 21), and Control (n = 20). Four participants were excluded due to missing or poor-quality data. For samples S1–S5 (30 s) all 79 participants were included, while for S6–S10 (30 s), the patients whose first IV cannulation attempt failed during the research were excluded, resulting in 67 participants: Deep Breathing (n = 15), Mindfulness (n = 15), Passive VNE (n = 18), and Control (n = 19).

The baseline condition (S1, 30 s) was compared between all groups. Statistically significant differences were found from SDNN χ²(3) = 9.061, p = 0.028, RMSSD χ²(3) = 8.125, p = 0.044 and SI χ²(3) = 9.254, p = 0.026. As in the VR intervention HRV analysis, the baseline measures of the VR intervention groups were similar, and the Control group differed from the other groups, with the values closely resembling the S1 (120 s) measurements.

Differences to S1 (30 s) for S2–S10 (30 s) were compared between the groups (S2d–S10d). Figure 6 illustrates these differences with line charts of mean differences (S2d–S10d). Statistically significant differences were found between the groups from S2d: SDNN χ²(3) = 14.305, p = 0.003 and SI χ²(3) = 17.658, p < 0.001; S3d: SDNN χ²(3) = 9.131, p = 0.028 and SI χ²(3) = 15.428, p = 0.001; S8d: SI χ²(3) = 8.428, p = 0.038; S9d: SDNN χ²(3) = 9.600, p = 0.022 and SI χ²(3) = 9.693, p = 0.021; S10d: SDNN χ²(3) = 9.600, p = 0.022 and SI χ²(3) = 11.683, p = 0.009. Pairwise comparisons found no significant differences between Passive VNE and Control groups before (S2d–S5d), during (S6d), or after the IV cannulation (S7d–S10d). However, both guided relaxation exercises reduced stress before and after the procedure, with a significant difference from passive VNE or standard care.

The deep breathing exercise was the most effective condition, with significantly different changes to stress levels found until S4d (45 to 15 s before the IV cannulation), and again at S8d (15 to 45 s after the IV cannulation). Comparison to passive VNE found differences from S2d (SDNN z = −3.207, p = 0.001; RMSSD z = −2.082, p = 0.037; SI z = −3.292, p < 0.001), S3d (SDNN z = −2.056, p = 0.040; SI z = −2.706, p = 0.007), S8d (SI z = −2.504, p = 0.012), S9d (SDNN z = −2.663, p = 0.008; SI z = −2.734, p = 0.006), and S10d (SDNN z = −2.653, p = 0.008; RMSSD z = −2.224, p = 0.026; SI z = −3.016, p = 0.003). Comparison to control found differences from S2d (SDNN z = 3.207, p = 0.001; RMSSD z = 2.219, p = 0.027; SI z = 3.445, p < 0.001), S3d (SDNN z = 2.925, p = 0.003; RMSSD z = 1.981, p = 0.048; SI z = 3.590, p < 0.001), S4d (SDNN z = 1.957, p = 0.050), S8d (SDNN z = 2.489, p = 0.013; SI z = 2.415, p = 0.016), S9d (SDNN z = 2.788, p = 0.005; RMSSD z = 2.003, p = 0.045; SI z = 2.483, p = 0.013), and S10d (SDNN z = 2.278, p = 0.023; SI z = 2.068, p = 0.039).

The mindfulness-based exercise caused significant differences in stress reduction until S3d (60 to 30 s before the IV cannulation), at S7d (0 to 30 s after the IV cannulation), and at S10d (45 to 75 s after the IV cannulation). Comparison to passive VNE found a difference from S2d (SI z = −2.325, p = 0.020) and S10d (SI z = −2.512, p = 0.012). Comparison to control found differences from S2d (HR z = 2.281, p = 0.023; SI z = 2.486, p = 0.013) and S3d (HR z = 2.196, p = 0.028; SI z = 2.448, p = 0.014). This relaxation exercise performed poorly near the IV cannulation, with the deep breathing exercise being more effective at S7d (SDNN z = −2.019, p = 0.043). At S6d, the mean SDNN values were similarly different between the relaxation exercises, but due to high variance in both groups, the difference was not statistically significant (p = 0.099).

5. Discussion

5.1. Main Findings

This study compared the effects and user experience of two different guided relaxation techniques in a VNE to a passive VNE and standard care in reducing the stress of child patients during IV cannulation. The aim was to address the research gap regarding VR content in this context [6].

Only relaxation exercises in a VNE provided objective evidence of stress reduction. Deep breathing guidance reduced stress according to all HRV measures used (SDNN, RMSSD, LF/HF ratio, SI), with a significant effect found until 30 s before the cannulation, and the effect was observed again 30 s after the cannulation. Mindfulness-based relaxation guidance also reduced stress similarly, but with a weaker effect and poorer performance close to the cannulation. Poor performance was also previously found among patients with a fear of needles [10]. No objective evidence of stress-reducing effects was found from passive VNE or standard care.

In contrast, while user experience was positive across all VR intervention groups, patient self-evaluations indicated that passive VNE feels more useful and more pleasant than a deep breathing exercise in the same VNE. Passive VNE also reduced the experience of anxiety from expectations, while no similar evidence was found with the deep breathing exercise. Mindfulness-based relaxation reduced pain from expectations, and the user experience was slightly better than with the deep breathing exercise across all relevant measures, including general treatment experience. The Control group reported reduced anxiety and pain, but also expected a more stressful experience than the rest. While all groups reported low pain and anxiety experience, deep breathing exercise performed worse than standard care, although post-procedure interviews and observations—and HRV data—indicate superiority of deep breathing in a VNE over standard care.

In pediatric research, differences between patient self-evaluations and objective results are not uncommon. Children can both exaggerate and understate their experiences regarding pain and anxiety, with potential underlying factors including desire to do well, past experiences, anxiety issues, and sex [80,81,82].

Even so, differences in self-evaluations between the VR intervention groups suggest a problem in the patient experience of the deep breathing exercise. This might be due to the monotonous nature of the implementation. The exercise included a four-minute period of steady deep breathing—starting two minutes and 30 s before the IV cannulation. As the stress-reducing effect was almost immediate, this was an unnecessarily long time to focus on breathing only. It might have felt boring, uncomfortable, or even anxiety-inducing to some patients. This user experience issue is likely related to half of the group reporting problems with focus [9].

Baseline HRV of the Control group differed from other groups. The sex division was similarly different between the groups, offering at least a partial explanation—the differences were in line with known effects of age and sex to HRV [79]. VAS-A data support this assumption, with higher stress expectations found from Control. If we ignore the effect of sex, baseline HRV samples would imply a more relaxed state of mind for Control, rather than a more stressed one. Data regarding SCARED questionnaires, the level of needle phobia, and previous cannulation experiences do not provide evidence that could explain the difference. The exact causality remains undefined, as prior knowledge of receiving only standard care could also contribute to this. However, the baseline HRV was similar in all VR intervention groups, and stress-reducing effects were found from both relaxation exercises, also when comparing to passive VNE. In practice, the data from the Passive VNE group can be considered as an alternative control, which verifies the stress-reducing effects of guided relaxation exercises.

In general, the VirNE application was positively received among both the child patients and their caretakers, and the nurses—who had not previously used VR to ease the IV cannulation process—found the technology both beneficial and easy to use. A case report related to this study indicated good suitability for repeated use [83]. Finnish children seem to prefer blue VNEs, which refer to virtual nature with a water element, but both individual and cultural differences regarding this preference are possible. Visiting blue nature environments is a common recreational summer activity among Finnish children [84], which can possibly explain the popularity of blue VNEs in our study. Studies comparing the effects of green and blue VNEs remain limited and have a moderate to high risk of bias [85]. Nevertheless, our findings support the idea that blue VNEs can cause similar stress-reducing effects as green VNEs [42,47].

5.2. Related Work

The results strongly support the general idea of using nature-based VR relaxation to reduce acute stress in pediatrics, being well in line with previous studies [6,20,42]. The application used (VirNE) met all five key features for nature-based analgesia, as recently proposed [51]: (i) enables the sense of being away, (ii) addresses patient’s need to endure treatment and individual preferences, (iii) potential for exploration (in the form of audiovisual biodiversity), (iv) engaging and fascinating without demanding too much attention (can be untrue regarding the deep breathing exercise), and (v) associated with safety (no signs of danger). Thus, our results offer a concrete and effective example of this theoretical framework in practice.

Very few pediatric studies compare guided relaxation exercises in a VNE to a passive VNE or standard care in addressing acute stress. In general, most of the related studies have been conducted with adults, and pediatric studies incorporating objective measures and a control group have been rare [6,17].

A relatively similar VR intervention was studied with 55 children (mean age (SD) 10.88 (±3.17) years) for inpatient relaxation (n = 36) or addressing periprocedural stress (n = 19). The VR contents, which included guided relaxation exercises and minigames in a VNE, were evaluated to be feasible and well-accepted for both use cases [68]. Another study compared mindfulness-based relaxation in a VNE to distraction with tablet games in a pediatric emergency department with adolescents (n = 110, mean age (SD) 15.0 (±1.3) years). The VR intervention was more effective in reducing short-term anxiety, while the experiences of pain were similar between the groups [66]. Both studies align well with our results.

Nature-based VR games are among the more studied VR contents in this context. Most notably, gamified deep breathing exercises have been found to reduce periprocedural stress and increase treatment satisfaction of child patients [86,87]. Good feasibility and superiority to standard care have also been found with various minigames in VNEs [88,89,90]. While lacking objective measures of stress, these studies imply a strong potential for improved patient experience with a gamification approach to VR relaxation—especially regarding the deep breathing method.

Other VR contents studied in pediatric use include various 360-degree videos [91,92,93], cartoons [94], age-appropriate procedural information [95], and hypnotherapy [96]. These studies suggest good general feasibility for VR contents that are designed or curated for pediatric use. Most notably, Badke et al. [93] found reduced peak stress levels from HRV with 360-degree videos in pediatric intensive care unit (n = 115, median age (IQR) 10.0 (7.0) years). Patients could select the theme of the videos from three options: adventure, animals, and nature landscapes (passive VNEs). Only 13% selected the landscapes, with animals being popular among younger patients and adventure among older patients. As we found no evidence of stress reduction from HRV with passive VNEs, visible animal content seems likely to improve the stress-reducing qualities of VNEs for younger children. However, Badke et al. found no reduction in average stress levels, which is in line with our HRV analyses regarding passive VNE. As all HRV variables used in our study are more influenced by average than peak stress levels, it remains plausible that passive VNE could reduce peak stress.

Studies with adult patients offer additional insights. While VNEs have been found to reduce acute stress both objectively and subjectively [42], visual nature elements seem to be more related to the anxiety-reducing effect, with nature sounds reducing pain even without the visual element [97]. In general, VR distraction seems to be more effective in reducing periprocedural anxiety than pain [98], further highlighting the importance of sound in VNEs [99] targeted for acute pain reduction [42]. With this in mind, the professional sound design of VirNE potentially contributed to the positive results found in our study, especially regarding the positive performance of passive VNEs in subjective evaluations. In general, the audiovisual quality of the experience could be a factor in why some studies have found passive VNEs to perform poorly in pain reduction [25,100].

Our results show that the analgesic effects of VNEs can be improved with guided relaxation exercises. Outside of VR studies, both deep breathing and mindfulness-based relaxation guidance have been found to objectively reduce stress among adults [101]. While deep breathing seems universally effective, a low tendency for spirituality can reduce the effect of mindfulness-based methods [102]. Spirituality—as an individual dynamic characteristic related to introspection and meaning [103]—affects the subjective well-being of children [104], thus potentially affecting the effectiveness of mindfulness-based methods in pediatrics as well.

Deep breathing in a VNE has been found to objectively reduce stress in adult populations [58,59,105]. The evident benefits of deep breathing methods in stress management are likely related to attention, as steady breathing helps maintain it, and loss of respiration stability contributes to fluctuating attention [55].

Rather than addressing acute pain, mindfulness-based relaxation in a VNE has been more commonly studied in a mental health context, where it has been found to reduce anxiety and improve emotional regulation [106,107]. Regarding pain, the focus has been on more vague forms, such as chronic pain [108] and post-operative pain [109]. Including non-VR studies, a systematic review and meta-analysis for acute pain management [110] found mindfulness-based interventions to increase tolerance of pain, but found no effect on pain severity.

We found no other clinical studies that would compare these two relaxation methods in a VNE. A study comparing VR deep breathing to non-VR guided mindful breathing found that while both practices increase pain thresholds, the analgesic brain mechanisms differ notably. VR deep breathing activated the visual-auditory cortex and limited connection to pain processing (primary somatosensory cortex), while mindful breathing increased this connection, inhibiting its role in pain processing with sensory-interoceptive processing of breathing [111]. The analgesic effects of mindfulness-based relaxation in a VNE seem to differ from other forms of VR pain management [112], potentially explaining the weaker effect close to IV cannulation.

Biofeedback is an emerging theme in relevant studies, with findings suggesting strong potential to improve patient adherence to deep breathing. Studied methods to achieve this include visually presenting the difference to the desired breathing rhythm and gamification elements, such as using breathing as a game controller and rewarding the user for correct breathing rhythm [58,59,86,87,105].

5.3. Limitations and Recommendations

This was a preliminary study with a sequential non-randomized participant allocation to groups. Participants in the Control group were recruited to participate in the control group due to study design and ethical reasons. This causes a difference in the research scenario between the Control and VR intervention groups. It possibly affected the patient demographics during recruitment and the mindset of these participants during the research.

The differences in sex division and baseline measures indicate a high likelihood of allocation bias in the Control condition. The Passive VNE condition was used as an alternative control in this study. Randomized clinical trials remain highly recommended for further research. Detailed sex- or gender-based analyses are excluded from the scope of this article and remain topics for future research.

The study population was diverse regarding the evaluated level of needle phobia. While the results show good general applicability for addressing periprocedural pain and anxiety, further research with a higher number of participants suffering from fear of treatment is needed to evaluate the feasibility for severe cases.

The use of stationary 360-degree videos of local nature is a feasible method to implement VNEs for this use context. Gamification of the exercise, avoiding prolonged use of the deep breathing method, and incorporating biofeedback into the exercise are identified as promising methods to improve the user experience and patient adherence to the VR intervention.

6. Conclusions

Our results show that guided relaxation exercises improve the efficacy of VNEs as a non-pharmacological method to address periprocedural stress in child patients.

Both relaxation methods were superior to passive VNEs and standard care in stress reduction according to HRV data. The deep breathing exercise was more effective but less enjoyable than mindfulness-based relaxation in a VNE, which had a weaker objective effect—especially in moments close to the IV cannulation—but resulted in higher patient satisfaction. No objective evidence of stress reduction was found with passive VNEs or standard care.

All three VR conditions outperformed standard care in general patient experience, with Finnish children preferring blue VNEs (lake/ocean shore) over forest environments. Nurses evaluated both relaxation exercises as superior to passive VNEs in this use context, while children found passive VNEs to be highly pleasant and more useful than deep breathing in a VNE. Passive VNEs reduced anxiety from expectations, while mindfulness-based relaxation in a VNE reduced pain.

To conclude, VNEs with guided relaxation exercises are pleasant, feasible, and effective methods to reduce the acute stress of child patients, thus presenting highly promising alternatives to sedatives in addressing treatment-based anxiety. Stationary 360-degree nature videos can be used to create realistic and highly pleasant VNEs that cause no adverse effects, being ideal for use cases where patient movement is undesired. Additional content, such as guided relaxation exercises, is needed to ensure the stress-reducing effect. User experience and patient adherence to these exercises can plausibly be improved with interactive features, such as gamification of the exercise and the use of biofeedback.