The study was designed as a single-blind, randomized-controlled trial. The Tokat Gaziosmanpasa University, Clinical Research Ethics Committee (Approval number: 22- KAEF-270) approved the protocol for this study. The trial was registered on https://clinicaltrials.gov/ web site (Trial number: NCT05829564). All participants were informed about the study's purpose and the interventions before the study. All participants signed consent forms.

Individuals with CNP who were referred to the physical therapy program at Hacettepe University Faculty of Physical Therapy and Rehabilitation, were offered to be enrolled in the study. The inclusion criteria were as follows: neck pain persisting for at least 3 months, between the ages of 18–65 years, sedentary life style (not receiving physical therapy in the last 6 months), a score of 10 or more out of 50 on the Neck Disability Index, individuals who can read and write. Exclusion criteria were; history of previous spinal surgery, neurological deficit (motor deficits), vestibular pathology, neurological, cardiopulmonary, musculoskeletal system conditions affecting physical performance, pathology in the shoulder joint, history of spinal trauma, pregnancy.

The sociodemographic characteristics included age, body mass index, gender, education, and occupation. Additionally, pain characteristics were recorded based on location (upper neck, lower neck, right, left), duration (3–6 months, 6 months-1 year, 1–2 years, 2–5 years, and > 5 years), and frequency (<1 and 1 time, 2–3 times, >3 times per week) for all participants. Outcome measures were recorded pre and post interventions.

All participants received 10 sessions at a frequency of 2–3 times a week for 4 weeks. The study included three groups; 1) VR group (VR+traditional exercise), 2) CM group (cervical mobilization+traditional exercise), 3) traditional exercise group (Fig. 1).

Fig. 1.

Flow diagram of the study.

(0.54MB).

The sample size was calculated with G*Power software (version 3.1) based on a repeated measures design with two time points (baseline and 4 weeks later) and three groups. An effect size of 0.25, which was derived from the pain intensity data of a reference study,38 was estimated, with the significance level set at 5 % and the statistical power set at 80 %. The analysis indicated a total sample size of 36 participants. Considering a 20 % drop-out rate, we planned to enroll 15 participants per group, resulting in a total of 45 participants for the study.

A researcher who was separate from the intervention team selected 45 numbers from a random number table and randomly divided them into the three groups. Then, the paper with numbers was placed into an opaque envelope. The envelopes were scrambled and numbered in turn. After inclusion, patients received the numbered envelopes consecutively and were divided into the corresponding groups according to the number on the paper within the envelope. Assessors who were responsible for the measurements were unaware of the group assignments and remained distant from the intervention.

Statistical analyses were performed using SPSS Version 26.0 (IBM Corp., Armonk, NY, USA). Normality and homogeneity of variances were tested using the Shapiro-Wilk and Levene tests, respectively. Descriptive data were presented as mean (standard deviation); mean (95 % CI), and number (frequency). GPE scores, which were not normally distributed, were compared among the three groups using the Kruskal-Wallis test. For analysis of primary and secondary outcome results, Generalized Estimating Equations (GEE) were used to analyze changes over time and interactions between groups. Time, group, and time*group interaction were included as independent variables in the model. Bonferroni correction was used for multiple comparisons. Changes in the outcome variables were calculated by subtracting the pre-intervention value from the post-intervention value. Standardized effect sizes for the differences between groups were calculated using Cohen's d formula. Cohen's d was classified as small (0.2–0.5), moderate (0.5–0.8), and large (≥0.8).39 The significance level for all statistical tests was set at p < 0.05.

The baseline and post-treatment values for JPSE and pain are shown in Table 2. The group*time interactions for JPSE values in six cervical directions were not statistically significant between groups (p > 0.05). The comparison between VR and CM shows a negligible effect size of 0.01 for JPSE-flexion; 0.02 for JPSE-extension; 0.12 for JPSE-right rotation; and 0.16 for JPSE-left rotation. For JPSE-RLF and JPSE-LLF, the effect size is respectively 0.54 and 0.53, indicating a moderate difference between VR and CM.

Similarly, for the VAS, the group*time interaction was not statistically significant (p > 0.05). The comparison between VR and CM shows a small to moderate effect size of 0.44 for VAS.

The baseline and post-treatment values for the baropodometry measurements, 10MWT, FSST, and SLST assessments are shown in Table 3. There were no statistically significant differences between the groups regarding the baropodometry measurements, 10MWT, SLST, and FFST (p > 0.05).

The baseline and post-treatment values for the NDI and SF-36 assessments for the groups are shown in Table 4. The group*time interaction was not statistically significant except SF-36social functioning (p > 0.05). The mean difference in social functioning is 21.15 with a 95 % CI of 9.46, 32.85, indicating a significant improvement in CM group (Cohen’s d = 1.52). Additionally, no statistically significant difference was observed in GPE scores among the three groups at post-treatment (p > 0.05) Table 4.

One participant had mild dizziness in the first session in the VR group, however dizziness disappeared within 5 min. In the CM group, two participants had mild pain in the first session. No observed pain in the other sessions.

Based on this study, which started from the premise that VR and CM applications have similar physiological mechanisms, the addition of VR and CM to traditional exercises did not provide additional benefits for JPS. Similar to the current study, Bahat et al.20 investigated the effects of cervical kinematic training with or without VR, and found no difference in effects.20 Guo et al.37 also stated that there was no difference of improvement in proprioception between VR therapy combined with conventional rehabilitation and conventional rehabilitation alone.37 In another study about comparison of the effect of VR and conventional proprioceptive training, no significant group-by-time interaction were seen in all directions of JPSE.40 In contrast, Cetin et al. compared the effects of VR and motor control exercises on JPS, and they indicated that JPS showed more improvement in the VR group.24

Similarly, adding VR and CM to traditional exercises did not lead to additional benefits. Those results are consistent with those from the studies by Bahat et al.20 and Guo et al. .24 In contrast, Guo et al. ,24 in a different study, indicated that VR therapy provided more pain reduction compared to conventional rehabilitation.37 In a 2024 review study on the management of neck pain using VR, it was noted that combining VR therapy with conventional rehabilitation is effective in alleviating pain.41 A meta-analysis study indicated that moderate evidence supports VR as a beneficial nonpharmacological approach to reducing neck pain intensity, but limited and heterogeneous studies limit the strength of evidence.42 In another study, CM has been shown to be effective in reducing pain in patients with CNP compared to routine physical therapy alone.43 We can say that VR is a promising tool for chronic pain management; however, there is a lack of VR intervention design consistency, follow-up reporting, and large sample sizes.44 These results show that there are studies in the literature presenting different outcomes on this topic, and further research is needed. According to the results of our study, there was no benefit to adding VR and CM to traditional exercises in improving JPS and reducing pain.

Changes in baropodometric measurements, 10MWT, SLST, and FSST were similar for all 3 groups. In a previous study comparing CerviGame®-VR and conventional proprioceptive training, neither intervention was superior in improving dynamic balance.16 Also, Bahat et al.20 indicated that no difference was observed between the kinematic training with or without VR groups in terms of eyes closed balance and single leg stance values.20 Our study also include parameters such as foot pressure distribution and postural oscillations that are not widely researched in the literature. In a study conducted to compare the peak pressure distribution in individuals with CNP compared to healthy participants, the authors stated that there was no difference in peak plantar pressure or center of pressure oscillation values between groups.11 The only notable finding was for the greater difference in quality of life for those receiving a combination of traditional exercises and CM or VR compared to traditional exercises alone.

The study has several limitations. The VR application in the study is not an application specifically developed for patients with CNP. There is a need for applications to be developed specifically for this population. Additionally, the absence of long-term follow-up data further restricts the generalizability of the findings.