Effects of audio and visual distraction on patients’ vital signs and tolerance during esophagogastroduodenoscopy: a randomized controlled trial

This study protocol is included in additional Figure 1. This study was designed as a prospective, single-blinded randomized controlled trial, and it was performed at Shikoku Central Hospital of the Mutual Aid Association of Public School teachers. The Ethics Committee in Shikoku Central Hospital of the Mutual Aid Association of Public School teachers approved the study protocol, which was registered in the University Hospital Medical Information Network (UMIN Clinical Trials Registry, number UMIN000029637).

Figure 1 shows a flow diagram of the enrollment and procedures of this study. A total of 360 subjects were scheduled to receive EGD at a regular health check-up at our hospital participated in this study between October 2017 and March 2018. The study design was explained, and all subjects provided written informed consent. Seventy-one subjects were excluded from the study if they met any of the following criteria: (1) current medication use; (2) a history of severe heart failure, renal failure, hepatic failure, or chronic obstructive pulmonary disease; (3) previous abdominal surgery, including endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD); (4) audio or visual disability; (5) previous experience of bad feelings from audio or visual stimuli; (6) a history of anxiety or psychiatric disorders; (7) pregnant or a possibility of pregnancy; and (8) receiving a diagnosis of gastrointestinal cancer or required biopsy. Subjects presented to the endoscopy floor in the morning after a longer than 12-h fasting period. Subjects who was divided into two categories those underwent EGD for the first time and those experienced EGD previously and each subject was randomly divided into 4 groups using a sealed opaque numbered envelope method by the endoscopy nurse who assisted at EGD. All subjects sat on a sofa and rested quietly for 5 min in a private room near the endoscopy room. Subjects in control group continued to sit on the sofa and rest quietly for 10 min prior to EGD. Subjects in audio group sat on the sofa and listened to music for 10 min. Subjects in visual group sat on the sofa and watched a silent natural image for 10 min. Subjects in combination group sat on the sofa and watched a natural image while listening to music for 10 min. The study used healing music, such as country and classical music, based on the tone of a music box, which was chosen as good by 20 volunteers in a pre-meeting prior to the start of this study. The moving images used in this study were various natural images, including a mountain, forest, river, waterfall, lake, and sunset. Music and natural images were delivered using a wall-type Hi-vision liquid crystal television (TH-42AS650; Panasonic Corporation, Osaka, Japan). Pharyngeal anesthesia with lidocaine pump spray (Xylocaine Pump Spray 8%; AstraZeneca, Osaka, Japan) without any sedative agents was applied, and 5 endoscopy specialists with greater than 5 years of experience in endoscopy performed a standard EGD, including observations of the esophagus, stomach, and duodenum, using a conventional single channel endoscope (GIF-H260; Olympus, Tokyo, Japan) without knowledge of the group of the subject. The profile of mood states (POMS) and the visual analog scale (VAS) of impressions for EGD were performed at pre- and post-distraction. VAS of the acceptance of distraction was performed after EGD.

Pulse rate (PR) and blood pressure (BP) were measured in the right upper arm, and peripheral blood oxygen saturation (SpO₂) was measured at the left finger using a monitor unit (BSM-7100 Life Scope; NIHON KOHDEN CORPORATION, Tokyo, Japan). These parameters were measured 5 and 15 min after sitting on the sofa, during EGD, and 5 min after EGD procedure. Parameters during EGD were measured just after insertion of the endoscope through the esophagogastric junction (approximately 2 min from the start of EGD) and just after moving the endoscope from the stomach to the esophagogastric junction (approximately 5–7 min from the start of EGD).

We assessed autonomic nervous function from pre-EGD to post-EGD using power spectral analysis (PSA). HRV was measured using a Heart Rhythm Scanner (HRV analysis system from Biocom Technologies, Ark Trading Pacific, Inc.) equipped with software that performed algorithms for short-term HRV analysis. A Biocom HRS-08 Bluetooth Wireless Pulse Wave Sensor photoplethysmography monitor was used in this study, and it was clipped to the right earlobe. Data of the average R-R intervals for 5 min were subjected to PSA using the software of the HRV analysis system. The amplitudes of the low-frequency (LF) range (LF, 0.04–0.15 Hz) and high-frequency (HF) range (HF, 0.15–0.40 Hz) were analyzed using complex demodulation. These LF and HF values were designated as the LF power and HF power, respectively. HF power is fluctuation in the heart rate caused by respiration, which is mediated by cardiac parasympathetic nervous activity [17, 18]. The ratio of LF power to HF power is an index of sympathetic nervous activity [19‐22]. The HF power data were converted to a logarithmic scale to analyze using linear regression in the present study.

We used the POMS2 questionnaire for psychological assessments between pre- and post-distraction because POMS is a self-report measure that quickly assesses transient, fluctuating feelings and enduring affective states [23, 24]. The POMS2 is composed of 35 items and 8 subscale scores. We also used the VAS, which consists of a 100-mm horizontal line scored from 0 to 100 to rate the degrees of strain, anxiety, and fear of EGD. All subjects answered the POMS and provided their impressions of EGD immediately after sitting on the sofa and 15 min after sitting on the sofa, but prior to EGD. After EGD, subject used the VAS to rate their degrees of satisfaction, usefulness, and willingness to assess the subject’s acceptance of distraction.

The primary outcome measures were psychological factors including POMS and impression for EGD at post-distraction and acceptance of distraction including degrees of satisfaction, usefulness, and willingness for the next use at post EGD. The secondary outcome measures were vital signs, HRV at post-distraction, during EGD, and at post-EGD. Vital signs and HRV were measured at pre-EGD (5 min after sitting on the sofa), pre-EGD (15 min after sitting on the sofa), during the early and the latter half of EGD, and 5 min after the end of EGD.

We assumed that the appropriate sample size for the randomized subjects was over 180 subjects based on the requirement of a significant difference between 4 groups with a significance level of 0.05, power of 80%, and, effect size of 0.25. Additionally, the rate of subjects who fill exclusion criteria or who received a diagnosis of gastrointestinal cancer or a biopsy was 30–40% by referring to our previous prospective randomized trial on endoscopy. Therefore, the planned required number of subject who receive EGD was over 300. Quantitative data, including subject characteristics, vital signs, POMS score, and VAS scores of impressions for EGD and acceptance of distraction, are expressed as the means ± standard deviation (SD). Parameters of autonomic nervous function are expressed as the means ± standard error of the mean (SEM). All significant differences at a P value less than 0.05 were considered significant. The χ²-test or Mann-Whitney U-test was used for comparisons between 2 groups or pre- and post-distraction in same group. The m × n χ²-test or Kruskal Wallis test was used to analyze differences among 3 or 4 groups. If the Kruskal Wallis test revealed differences between the groups, then post-hoc pairwise comparisons were performed using the Mann-Whitney U test with Bonferroni correction. All analyses were performed using Med Calc Software (Broekstraat, Mariakerke, Belgium).

Table 1 shows the baseline characteristics in the 4 groups. There was no significant difference in age, gender, smoking, drinking, experience or duration of EGD, POMS, and the impression for EGD among the 4 groups.

Figure 2 shows the changes in PR from pre-EGD to post-EGD in the 4 groups. PRs during the early and the latter half of EGD, and 5 min after the end of EGD were significantly higher than pre-EGD (5 min after sitting on the sofa) in control group and audio group (p <  0.001, p <  0.001, and p <  0.05). PRs during the early and the latter half of EGD were significantly higher than pre-EGD (5 min after sitting on the sofa) in visual group (p <  0.001 and p <  0.005) and combination group (p <  0.001 and p <  0.05), but there was no significant difference in PR between pre-EGD (5 min after sitting on the sofa) and 5 min after the end of EGD, and PR at pre-EGD (15 min after sitting on the sofa) was significantly lower than pre-EGD (5 min after sitting on the sofa) in visual group (p <  0.005) and combination group (p <  0.001).

Figure 3 shows the changes in BP from pre-EGD to post-EGD in the 4 groups. BPs during the early and the latter half of EGD, and 5 min after the end of EGD were significantly higher than pre-EGD (5 min after sitting on the sofa) in control group (p <  0.001, p <  0.05, and p <  0.05). BP during the early half of EGD was significantly higher than pre-EGD (5 min after sitting on the sofa) in the 3 distraction groups (p <  0.001), but no significant BP elevation was observed during the latter half of EGD and 5 min after the end of EGD. BP at pre-EGD (15 min after sitting on the sofa) was significantly lower than pre-EGD (5 min after sitting on the sofa) in the 3 distraction groups (p <  0.05).

Fig. 4 shows the changes in SpO₂ from pre-EGD to post-EGD in the 4 groups. SpO₂ during the latter half of EGD was significantly higher than pre-EGD (5 min after sitting on the sofa) in control group, audio group, and visual group (p <  0.01, p <  0.005, and p <  0.005).

Table 2 shows a comparison of PR at each point among the 4 groups. There was no significant difference in PR at pre-EGD (5 min after sitting on the sofa), during the early half of EGD, and the latter half of EGD among the 4 groups. However, there was a significant difference in PR at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD among the 4 groups on the Kruskal Wallis test (p <  0.001 and p <  0.01). Post-hoc pairwise comparisons revealed that PRs at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD in the 3 distraction groups were significantly lower than control group.

Table 3 shows a comparison of BP at each point among the 4 groups. There was no significant difference in BP at pre-EGD (5 min after sitting on the sofa), during the early half of EGD, and the latter half of EGD among the 4 groups. However, there was a significant difference in BP at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD among the 4 groups on the Kruskal Wallis test (p <  0.005 and p <  0.05). Post-hoc pairwise comparisons revealed that BP at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD in visual group was significantly lower than control group.

Table 4 shows a comparison of SpO₂ at each point among the 4 groups. There was no significant difference in SpO₂ at pre-EGD (5 min after sitting on the sofa), pre-EGD (15 min after sitting on the sofa), during the early half of EGD, and the latter half of EGD among the 4 groups. However, there was a significant difference in SpO₂ at 5 min after the end of EGD among the 4 groups on the Kruskal Wallis test (p <  0.01). Post-hoc pairwise comparisons revealed that SpO₂ at 5 min after the end of EGD in combination group were significantly higher than control group.

Figure 5-A shows a comparison of Log HF power from pre-EGD to post-EGD among the 4 groups. There was a significant difference in Log HF power during the early and the latter half of EGD, and 5 min after the end of EGD among the 4 groups on the Kruskal Wallis test (p <  0.001, p <  0.01, and p <  0.05). Post-hoc pairwise comparisons revealed that Log HF powers during the early half of EGD in the 3 distraction groups was significantly higher than control group.

Figure 5-B shows a comparison of LF power/ HF power from pre-EGD to post-EGD between the 4 groups. There was a significant difference in LF power/ HF power at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD among the 4 groups on the Kruskal Wallis test (p <  0.001 and p <  0.001). Post-hoc pairwise comparisons revealed that LF power/ HF power at pre-EGD (15 min after sitting on the sofa) and 5 min after the end of EGD in the 3 distraction groups was significantly lower than control group.

Table 5 shows a comparison of POMS and the impression for EGD between pre- and post-distraction in the 3 distraction groups. The score for negative mood at post-distraction was significantly lower than at pre-distraction in the 3 distraction groups (p <  0.05). The VAS scores of strain, anxiety, and fear for EGD at post-distraction were significantly lower than pre-distraction in audio group (p <  0.05, p <  0.05, and NS), visual group (p <  0.005, NS, and NS), and combination group (p <  0.001, p <  0.05, and p <  0.05).

Table 6 shows a comparison of the acceptance of the distraction after EGD among the 3 distraction groups.