To determine the effect of inhalation aromatherapy on sedation level, analgesic dosage, and bispectral index (BIS) values during donor site dressing in patients with burns.
This trial was conducted on 62 patients with burns requiring donor site dressing who were admitted to the Burn Center of Imam Reza Hospital, Mashhad, Iran. In the intervention group, the patients inhaled damask rose 40% and lavender 10% essential oils during donor site dressing change, whereas in the control group, the site was dressed using routine protocol. Sedatives and analgesics were prescribed until the levels of brain activity achieved light sedation. The brain activity and sedation levels were measured before and after the donor site dressings using the BIS. Data were analyzed using the analysis of covariance and the two-way analysis of variance with repeated measures.
All 62 patients completed the study. The required doses of ketamine (P < .001), fentanyl (P = .003), morphine (P < .001), and propofol (P < .001) were significantly lower in the intervention group. The BIS was also significantly lower in the intervention group (P < .001). Heart rate decreased significantly during the aromatherapy, as well as after analgesic and sedative consumption (P < .001).
The inhalation of damask rose and lavender essential oils is an effective intervention to reduce the doses of sedative and analgesic drugs administered as well as BIS during donor site dressing change in patients with burns.
Burns are among the major public healthcare problems worldwide and the most common cause of mortality, disability, and adverse outcomes for patients.1 According to World Health Organization reports, more than 300,000 deaths are recorded annually from burn injuries, and 95% of these occur in low- and moderate-income countries.2 In Iran, there are 150,000 cases of burns each year, with 3,000 mortalities.3
Skin grafting is one of the essential treatments for patients with burns; as a result, donor site pain is one of the most important complications after grafting that can affect pain management strategies adopted, opioid use, and hospitalization duration.4 Further, this pain is one of the most distressing symptoms for patients during the postoperative period.5 Specifically, the first donor site dressing change is associated with severe pain and anxiety6,7 and may even require general anesthesia.8 But because most research focuses on patients’ perception of pain from the primary wound, reports on donor site-related pain are insufficient.9 Minimizing pain and distress during the procedure is critical and has positive psychosocial effects for patients and their families while promoting re-epithelialization and wound healing.10–12
Numerous studies have been conducted on the effect of analgesic and sedative agents such as ketamine, fentanyl, propofol, and midazolam for pain reduction during dressing change in patients with burns.13,14 Despite the administration of high dosages of analgesics, patients still experience intense pain during these treatments.15 In addition, the harmful complications of these medications in high doses can aggravate patients’ condition.16 Higher levels of anxiety further increase the patients’ need for sedative agents.17
Given that pain control is one of the most important nursing responsibilities and is considered an index of quality of care,18,19 it is essential to find practical and safe pain management strategies, including both pharmacologic and nonpharmacologic interventions. Nonpharmacologic approaches for pain management include a wide range of techniques that not only reduce the physical sensation of pain, but also reduce the mental suffering associated with it.20 One such approach is aromatherapy.21 The American Nurses Association has introduced the use of this therapeutic approach as a part of holistic nursing services.22
Lavender (Lavandula angustifolia Mill), known as “ostokhoddous” in Iran, is a plant native to Mediterranean regions. The linalool and linalyl acetate contained in this plant can stimulate the parasympathetic system. Linalyl acetate has narcotic properties, and linalool acts as a sedative.23 The positive effects of its aroma on pain/anxiety reduction and relief, sleep quality, and rheumatic diseases have been studied.24–29 Inhalation of lavender essential oil can reduce respiratory rate and BP levels during panic attacks. The anxiolytic and antidepressant effects of lavender essential oil in both human/animal studies have been reported, too.30–32
Damask rose (Rosa damascena mill L.), called “gole mohammadi” in Iran, is one of the most important flowers of the Rosaceae family. Besides its pleasant scent, this plant has many pharmacologic effects including antioxidant, anti-HIV, antibacterial, hypnotic, antidiabetic, and relaxant effects.33 Damask rose essential oil also contains ester, ketone, aldehyde, and terpene compounds that produce numerous mental effects by stimulating the olfactory center in the brain. The scent of this flower is beneficial for reducing heartbeat frequency in patients with burns during dressing changes, sleep quality, improving the severity of pain and anxiety in the first stage of labor, and for postpartum depression.34–37
Aromatherapy stimulates the parasympathetic nervous system via the hypothalamus and thus alters heart rate (HR), BP, respiration rate, oxygen consumption, stress/anxiety levels, and pain perception.38 In this regard, several studies suggest positive effects of aromatherapy on knee and neck pain, menstrual pain, pain during hemodialysis, and anxiety/stress surrounding bypass surgery.39–41 Nevertheless, there is a lack of adequate evidence on the effectiveness of aromatherapy in relieving burn complications, including the pain experienced during donor site dressing change.42 Given that the use of essential oil combinations is more effective than using them separately,43 the present study sought to determine whether inhalation aromatherapy using damask rose and lavender can reduce the bispectral index (BIS) and sedative/analgesic dosage during donor site dressing change in patients admitted to the authors’ burn center.
This randomized controlled clinical trial (ID: IRCT20171123037599N2) was conducted on 62 patients with burns admitted to the burn center of Imam Reza Hospital who required a donor site dressing. The study was approved by the research ethics committee (ID: IR.MUMS.NURSE.REC.1397.027). Confidentiality of the data was fully maintained. All participants signed an informed consent form and were assured of their anonymity and right to withdraw from the study at will.
The inclusion criteria were informed consent to participate, no olfactory disturbance or history of allergy to plant scents, having 5% to 25% burns according to Lund and Browder criteria, second- and third-degree burns, age 18 to 55 years, donor site located on the anterior of the thigh, wound grafting in one session, absence of inhalation burns, and burns not caused by self-immolation.
Exclusion criteria were any unwillingness to receive or intolerance to rose and lavender essential oils, delay in donor site dressing change, emergency situations, clinical suspicion of donor site infection, and serosanguineous drainage.
The sample size was calculated based on the results of a pilot study on 10 participants per group that compared the mean sedation level and analgesic dosage and BIS with 95% confidence interval and 80% test power. The mean and SD of BIS from the pilot study were 75.2 ± 5.3 in the control group and 71.6 ± 4.2 in the intervention group. The desired sample size was estimated to be at least 56 overall and 28 per group based on G*Power software (version 3.1; Heinrich Heine University, Düsseldorf, Germany). Considering a dropout rate of 20%, the target sample size was increased to 68 (34 per group).
A statistician randomly assigned patients who were admitted to the burn center and met the inclusion criteria into two groups using SPSS software (version 25; IBM Corp, Armonk, New York, US) by producing a random chain of ones (control) and twos (intervention). These random numbers were kept in sealed envelopes and opened only when an eligible patient was recruited.
Data Collection Tool
The data collection tool was a questionnaire containing the demographic characteristics of each patient, the type and dose of sedative/analgesic agents, BIS, pain score using a visual analog scale, and changes in physiologic parameters (systolic/diastolic BP, HR/respiration means, and oxygen saturation [SPo2]). The patients’ physiologic parameters were recorded in both groups. A BIS monitor (Model A2000 v 3.21; Aspect Medical Systems Inc, Newton, Massachusetts) was used to evaluate the depth of anesthesia in the patients by demonstrating the electrical characteristics of their brain cortex.44 The BIS values are inversely related to the depth of anesthesia;45 therefore, applying painful stimuli produces a significant increase in BIS.46 A BIS monitor offers a reliable and appropriate method for assessing the level of consciousness at all stages of anesthesia.47 The electroencephalogram converts patient data to scaled values from 0 (electroencephalogram off) to 100 (completely awake),48 showing the degree of consciousness quantitatively. Typically, BIS levels of 65 to 75 indicate light sedation or sedation, values of 50 to 60 implicate deep sedation, and lower degrees indicate a progressive decrease in brain electrical activities.49
First, sedatives/analgesics (midazolam, fentanyl, ketamine, morphine, and propofol) were injected into 10 patients with burns (total body surface area, 5%–25%) during donor site dressing to determine the required dose of sedative/analgesic agents to achieve light sedation (65 < BIS <75). The following values were obtained: midazolam (0.01–0.1 mg/kg), ketamine (0.5–1.5 mg/kg), fentanyl (0.3–1 μg/kg), morphine (0.1 mg/kg), and propofol (0.1–1.5 mg/kg).
In both groups, the donor site dressing was changed by nurses with a similar level of expertise and work experience (more than 10 years). This process is usually performed 72 hours after grafting and at 8 o’clock in the morning on the same day. An anesthesiologist and his/her assistant were present during donor site dressing.
In the intervention group, at 7:30 am on the day of donor site dressing (half an hour before changing the dressing), the BIS sensor was placed on the patient’s forehead, and after monitoring their physiologic indices (systolic/diastolic BP, HR/respiration means, and SPo2), damask rose (40%) and lavender (10%) essential oils (Barij Essence Pharmaceutical Company, Kashan, Iran; registration no. 1028) were administered by an ultrasonic nebulizer (402B model; Yuwell, Kolkata, India), placed 20 cm from the patient’s face for 20 minutes. After the aromatherapy session, sedatives/analgesics (midazolam, fentanyl, ketamine, morphine, and propofol) were injected until brain activities approached BIS levels of 65 to 75. Then, the donor site was dressed. If the BIS increased to levels higher than 75, a bolus dose of sedative/analgesic was used.
In the control group, the patients were monitored by BIS and physiologic indicators before changing the donor site dressing. The sedatives and analgesics were then injected until the level of brain activities approached BIS levels of 65 to 75. To help maintain light sedation, the same measures of injecting a bolus dose of sedatives and analgesics were used.
A baseline BIS reading and continuous monitor readings were recorded every 5 minutes until 15 minutes after the procedure was terminated. Then, the following parameters were recorded and compared in both groups: physiologic parameters, the amount of sedative/analgesic agents required to achieve the desired BIS, BIS fluctuations during dressing change, pain score, and any need for bolus doses to achieve light sedation (65 < BIS <75). If SPo2 was reduced to less than 94%, supplemental oxygen was administered.
Statistical analysis was conducted using SPSS software. The normality of the numeric variables was taken as a skewness of ±1.5 and kurtosis of ±2 distribution measures. Data were presented using mean (SD) for the normal numeric variables and frequency (percent) for the categorical variables. The between-group comparisons of the baseline measures and demographic variables were carried out via independent t test and χ2 tests wherever appropriate. For within-group comparisons among three measurements, the paired-sample t test and repeated-measures analysis of variance (ANOVA) were used. The analyses of covariance (ANCOVAs) were conducted in two models to assess the effect of the intervention. Model 1 controlled for baseline measures, and model 2 controlled for baseline measures and confounders including age, sex, body mass index, education, occupation, burn percent, burn depth, and cause of burns. Two-way ANOVA with repeated measures was performed to assess the interaction effect of measurements by the study groups. For the repeated-measures ANOVA, the assumption of sphericity was evaluated by Mauchly test, and to correct the deviation from the assumption, the Greenhouse-Geisser correction was used. All the analyses were carried out using the per-protocol approach, and P < .05 was considered significant. The reporting of this study adheres to the CONSORT (Consolidated Standards of Reporting Trials) statement.30 In addition, a per-protocol strategy was pursued for the analyses.
Investigators recruited 107 patients for this study. In the initial eligibility assessment step, 37 patients were excluded (20 patients did not meet the inclusion criteria and 17 declined to participate). Seventy patients were allocated to the intervention (n = 35) and control (n = 35) groups. Three patients in the intervention group and five in the control group discontinued the intervention, and ultimately, the data from 62 patients were analyzed (n = 32 in the intervention group and n = 30 in the control group; Figure 1).
Table 1 presents patient characteristics. The results showed no significant differences between the intervention and control groups in terms of age, sex, body mass index, education, occupation, marital status, reason for burns, location of burns, and burn depth (P > .05 for all variables).
|Characteristic||Intervention (n = 32)||Control (n = 30)||P a|
|Body mass index, kg/m2||24.1||(3.2)||24.7||(3.9)||.548|
|Education (high school diploma and higher)||24||75.0%||22||73.3%||.248|
|Marital status (married)||23||71.9%||23||76.7%||.775|
Comparing Medication Dosage Administered
The results showed significant differences between the intervention and control groups in terms of the consumed dosage of fentanyl (P < .001), morphine (P < .001), and propofol (P < .001) at baseline; however, the differences were not significant for ketamine (P > .05; Table 2). The differences were adjusted in the subsequent analyses. In addition, the consumed dosage of midazolam was assessed only at baseline, and the results showed a significant difference between the intervention (mean [SD] = 1.31 [0.52]) and control (mean [SD] = 2.65 [0.58]) groups (P < .001).
|Drug||Timepoint||Intervention (n = 32)||Control (n = 30)||MD (95% CI)||P a||MD (95% CI)||P a||Interaction
|Ketamine||Initial||47.81||5.52||49.33||2.54||−1.52 (−3.73 to 0.69)||.174b||—|
|Final||0.00||0.00||42.50||10.40||−42.43 (−46.20 to 38.67)||<.001c||−42.42 (−46.32 to 38.52)||<.001|
|MD (95% CI), P d||−47.81 (−49.81 to 45.81), <.001||−6.83 (−10.77 to 2.90), <.001||<.001|
|Fentanyl||Initial||27.03||7.61||50.0||0.01||−22.96 (−25.75 to 20.19)||<.001b||—|
|Final||0.00||0.00||29.83||22.76||−29.83 (−48.96 to 10.71)||.003c||−31.39 (−49.87 to 12.92)||.001|
|MD (95% CI), P d||−27.03 (−29.77 to 24.29), <.001||−20.17 (−28.67 to 11.67), <.001||0.112|
|Morphine||Initial||4.03||1.38||4.93||.37||−0.90 (−1.42 to −0.38)||.001b||—|
|Final||0.32||1.22||4.13||1.89||−3.56 (−4.44 to 2.69)||<.001c||−3.61 (−4.40 to 2.83)||<.001|
|MD (95% CI), P d||−3.71 (−4.32 to 3.17), <.001||−0.80 (−1.46 to 0.14), .013||<.001|
|Propofol||Initial||18.59||8.16||32.00||8.87||−13.40 (−17.73 to 9.08)||<.001b||—|
|Final||1.25||4.92||53.00||18.41||−49.66 (−58.34 to 40.98)||<.001c||−52.36 (−60.62 to 44.10)||<.001|
|MD (95% CI), P d||−17.34 (−20.17 to −14.52), <.001||21.00 (13.51 to 28.49), <.001||<.001|
Table 2 presents the paired t test results for the within-group comparison of the consumed dosage of drugs in each group. A significant decline was observed in the consumed dosage of ketamine, fentanyl, morphine, and propofol (P < .05 for all variables).
The intervention effect
Table 2 presents the results of the ANCOVA based on the two models. In both models, the results showed that the intervention had a significant effect on the consumed dosage of the drugs (P < .05 for all). The amount of decrease was significantly higher in the intervention group than the control group.
The group-measurement interaction
Table 2 presents the group-measurement interactions. Significant interactions were observed between measurements and group for ketamine, morphine, and propofol (P < .001 for all), indicating the significantly different time trends between the intervention and control groups. Nonetheless, fentanyl showed no significant interaction (P = .112), which suggests a similar trend in both groups.
Comparing BIS, BP, and Vital Signs
The results showed no significant differences between the intervention and control groups in terms of BIS, systolic BP, diastolic BP, SPo2, and respiratory rate (P > .05). Nevertheless, the difference was significant in terms of HR (P < .05 for all; Tables 3 and 4). The differences were adjusted in subsequent analyses.
|Variable||Time||Intervention (n = 32)||Control (n = 30)||MD (95% CI), P||MD (95% CI), P a||Interaction
|BIS||Before inhalation||96.91||1.73||97.20||1.16||−0.29 (−1.05 to −0.46), .438b||—|
|With drug||71.43||2.47||74.03||1.13||−2.61 (−3.61 to −1.60), <.001c||−2.53 (−3.46 to −1.59), <.001|
|Variables||Time||Intervention (n = 32)||Control (n = 30)||MD (95% CI), P||MD (95% CI), P a||Interaction
|With inhalation||106.31||11.06||97.97||9.75||8.35 (3.04 to 13.66), .003b||—|
|With drug||112.69||10.28||105.47||13.80||0.48 (−4.33 to 5.28), .843c||0.36 (−4.20 to 4.92), .875|
|After drug||104.53||13.32||97.73||10.60||−1.37 (−4.93 to 2.18), .442c||−1.58 (−5.30 to 2.13), .398|
|With inhalation||127.25||17.21||121.23||9.76||6.02 (−1.16 to 13.19) .099b||—|
|With drug||141.09||17.66||136.70||15.02||−1.42 (−6.23 to 3.39), .557c||0.30 (−4.18 to 4.78), .894|
|After drug||126.19||17.25||121.87||13.59||−1.21 (−5.75 to 3.33), .597c||−0.51 (−5.22 to 4.20), .830|
|With inhalation||72.59||10.82||72.87||12.31||−0.27 (−6.15 to 5.60), .926b||—|
|With drug||78.09||9.55||79.57||14.29||−1.27 (−5.66 to 3.12), .564c||−0.53 (−4.57 to 3.51), .794|
|After drug||70.72||8.89||72.50||17.80||−1.66 (−8.31 to 4.99), .619c||−0.92 (−7.78 to 5.95), .790|
|With inhalation||96.19||1.99||95.63||2.19||0.55 (−0.51 to 1.62), .301b||—|
|With drug||92.44||3.03||94.20||5.81||−2.01 (−4.33 to 0.31), .089c||−1.95 (−4.35 to 0.44), .108|
|After drug||91.78||3.55||92.17||2.28||−0.71 (−2.13 to 0.72) 0.326c||−0.61 (−2.04 to 0.83), .401|
|With inhalation||17.09||2.53||17.00||5.07||0.09 (−1.92 to 2.11), .926b||—|
|After drug||14.94||2.09||15.43||1.74||−0.52 (−1.39 to 0.36), .243c||−0.51 (−1.22 to 0.20), .156|
Tables 3 and 4 also present the repeated-measures ANOVA for the within-group comparisons of the measurements for BIS, BP, and vital signs in each group. Significant decreases were observed in terms of BIS, SPo2, and respiratory rate in both the intervention and control groups (P < .001 for all). Nonetheless, a different trend was observed for HR, systolic BP, and diastolic BP. HR decreased significantly “with inhalation” (mean difference, 9.82; 95% confidence interval, 7.01–12.62; P < .001), followed by a nonsignificant increase with drug consumption and again a significant decline after drug consumption (mean difference, 9.32; 95% confidence interval, 3.88-14.76; P < .001). Systolic BP and diastolic BP showed no significant differences at baseline and “with inhalation” measures (P > .05 for all), but they increased significantly with drug consumption and then decreased significantly after drug consumption (P < .05).
As summarized in Table 2, the ANCOVA test indicated a significant intervention effect on the consumed drugs (P < .05). This indicates that the consumption of analgesics and sedative agents in the intervention group was significantly lower than in the control group.
The mean pain score assessed by visual analog scale was statistically lower in the intervention group compared with the control group during donor site dressing (mean ± SD, 4.5 ± 0.6 vs 8.3 ± 1.4; P < .001).
The results of the ANCOVA with the two models are presented in Tables 3 and 4. In both models, the results suggest a significant effect of the intervention on BIS (P < .05 for both). The amount of decrease was significantly higher in the intervention group than the control group; however, no significant intervention effects were observed for HR, systolic BP, diastolic BP, SPo2, or respiratory rate based on the baseline-adjusted and fully adjusted ANCOVA models.
The measurement-group interaction
Tables 3 and 4 present the interaction between measurements and groups. Significant measurement-group interactions were observed for BIS (P < .001 for all), indicating the significantly different time trends of the measures between the intervention and control groups. Nonetheless, HR, systolic BP, diastolic BP, SPo2, and respiratory rate showed no significant interaction effects (P = .112).
The present study investigated the combined effect of inhalation aromatherapy using damask rose (40%) and lavender (10%) essential oils on sedative/analgesic usage during donor site dressing change in patients with burns. Damask rose 40% has been used successfully by Bikmoradi et al36 to assess the effect of aromatherapy on the dressing-related pain intensity in patients with burns. The effect of lavender 10% on the sleep quality of patients with burns also has been evaluated, although without any positive outcome on sleep quality.50
These results indicated a decrease in the average BIS score (index of anesthesia depth) and reduction in sedative and analgesic administration after the intervention. In other words, aromatherapy results in a progressive reduction in brain electrical activities and thus reduces the amount of sedatives/analgesics prescribed during donor site dressing.
The most common form of burn pain is inflammatory nociceptive pain, which is exacerbated by procedural pain related to the treatment of burn wounds—in particular, donor site dressing change. When dressing burn wounds, the stimulation of local nociceptors transmits an impulse via A-delta and C fibers to the dorsal horn of the spinal cord, and at the end, the perception of pain is regulated by cortical areas and the thalamus. These mechanisms necessitate the application of multimodal approaches for pain relief.51 In contrast, the antinociceptive effect of essential oils is mediated by hypothalamic neurons.52 The effect of aromatherapy has previously been examined using lavender essential oil to reduce the severity of labor pain,53 surgery pain,54 pain intensity during vascular needle insertions in hemodialysis,55 and enhanced anxiety relief in patients with burns.56
Patients with burns usually experience high levels of severe pain and anxiety57 and require large amounts of narcotics and sedatives.58 Therefore, finding an effective treatment for pain reduction and decreasing the dosage of analgesics and sedatives are essential for these patients. In the present study, the authors used a combination of essential oils for reducing pain and anxiety among the patients. The BIS showed lower pain score and better sedation with lower analgesic/sedative administration.
It has been demonstrated that the BIS value increases in response to a pain stimulus;46 therefore, measuring brain activity based on BIS can help with the recognition of pain. The patients with burns in the control group (with an increase in their BIS) experienced intense pain during dressing change. The lower BIS index in the intervention group is probably indicative of a milder pain experienced by these patients. The findings are consistent with the findings of other studies that indicate the positive effects of aromatherapy on pain in patients with burns. Inhalation aromatherapy with lavender essential oil could reduce short-term pain in patients with burns.28 The short-term effect in that study was possibly attributable to the application of only 10 drops of the fragrance, whereas in the present study, two fragrances were used simultaneously for aromatherapy for 20 minutes.
The positive effects of massage and inhalation aromatherapy were also reported in another study on pain alleviation in patients with burns.59 These patients had second-degree burns and did not require skin grafts, so massage and inhalation aromatherapy were used to reduce general pain. In the present study, patients required more than aromatherapy, because donor site dressing changes are painful and one of the most distressing complications in the postoperative period.5,60
Similar results have been observed in other studies.61 Some studies have shown that aromatherapy for reducing episiotomy pain decreases the analgesic consumption.62,63 Using this method after laparoscopic gastric surgery resulted in lower morphine usage for patients in the intervention group.64 The comparison of the effect of aromatherapy with oral acetaminophen on pain in children undergoing tonsillectomy also showed a reduction in analgesic prescription in the aromatherapy group.25 However, an examination of inhalation aromatherapy with damask rose essential oil revealed that inhalation aromatherapy does not affect the amount of analgesic use.35 This difference can be attributed to the amount and method of aromatherapy inhalation and the timing of pain measurement.
The findings regarding the effect of aromatherapy on patients’ vital signs during aromatherapy inhalation showed that this approach reduces HR, which seems to result from a decrease in the severity of anxiety after aromatherapy inhalation, but there was no statistically significant difference in the mean respiratory rate, systolic/diastolic BP, and SPo2; that is, inhalation aromatherapy had no effect on these outcomes. The results obtained in a study by Najafi et al65 were consistent with the present findings, concluding that inhalation aromatherapy impacts the average number of pulses and reduces the mean pulse rate but had no effects on the mean respiration rate or systolic/diastolic BP.
Patients with higher-degree burns were not included in this study. The intervention was performed only in one center, which limits the generalizability of the results. Further, this was a small sample size in the context of the literature. Future larger investigations could be done to mitigate this issue.
The simultaneous use of two fragrances in inhalation aromatherapy may reduce BIS and the intensity of patients’ pain during donor site dressing change, reducing the required dose of sedatives and analgesics. Therefore, aromatherapy is a potentially safe and effective approach for reducing pain in these patients. Given the increasing use of complementary therapies, inhalation aromatherapy with damask rose and lavender could be considered a noninvasive method to alleviate the pain of donor site dressing changes.
[ Tratto da: www.journals.lww.com ]