Keywords breast reconstruction - aromatherapy - reconstruction - breast cancer - lavender
Up to 50% of women may experience psychosocial distress, depression, and anxiety following
a cancer diagnoses and mastectomy.[1 ]
[2 ]
[3 ]
[4 ]
[5 ] Despite the potential for breast reconstruction to mitigate cancer-related distress,
a sizable portion of patients experience anxiety and depression in the acute perioperative
period after reconstructive surgery.[1 ]
[4 ]
[5 ] Recently, there has been increased emphasis on implementing enhanced recovery after
surgery (ERAS) protocols to reduce the number of opioid analgesics and anxiolytics
prescribed to patients who undergo microvascular breast reconstruction.[6 ]
[7 ] This is necessary to not only reduce risk factors, such as overdose and respiratory
distress, but to also reduce the incidence of comorbid psychological conditions (i.e.,
anxiety and depression) that often present in the patients with breast cancer diagnoses.[8 ]
[9 ]
[10 ] Thus, further investigations into low-risk therapeutics that relieve perioperative
pain, anxiety, and depression are warranted.
Previous studies have demonstrated the benefits of aromatherapy with essential oils
to benefit patients' mood and health.[11 ]
[12 ]
[13 ] Specifically, lavender oil is composed of chemical constituents, linalool, and linalyl
acetate that have been suggested to have anxiolytic effects through the inhibition
of the GABA(A) binding receptor in the central nervous system.[14 ]
[15 ] Studies examining these effects in the surgical patient have presented conflicting
evidence and are largely hindered by inadequate study design and poor methodology.[12 ]
[13 ]
[16 ]
[17 ]
[18 ] Furthermore, no studies have examined the benefits of aromatherapy in breast cancer
patients who undergo microvascular breast reconstruction. The aim of this study was
to assess the potential benefit of lavender oil as a perioperative adjunct to mitigate
anxiety, depression, pain, and improve sleep in women undergoing microvascular breast
reconstruction.
Methods
This was a prospective, single-blinded, randomized controlled trial designed to evaluate
the use of lavender oil to reduce perioperative anxiety, depression, pain, and improve
sleep in patients who undergoing microvascular breast reconstruction. Procedures were
performed by three surgeons at a single surgery center between December 2017 and January
2020. The trial was approved by the Institutional Review Board at Duke University
Medical Center. The study sponsor, dōTERRA International, provided the essential oils
that were used in this study.
Patients
Adult female patients' ages 18 to 85 years who were previously diagnosed with breast
cancer and undergoing microvascular breast reconstruction were eligible for this study.
Patients were excluded if they were pregnant, did not carry a breast cancer diagnosis,
or had a history of sensitivity to lavender oil or any of its ingredients.
Study Design
After meeting all eligibility criteria, a total of 58 patients (29 in the experimental
group and 29 in the control group) were enrolled in this study ([Fig. 1 ]). This was a prospective, single-blinded, randomized controlled trial with a 1:1
treatment allocation of patients placed in two parallel groups: (1) patients who received
lavender oil and (2) the control group who received coconut oil. Fractionated coconut
oil was selected as the control substance due to its colorless, odorless, and inert
nature. A randomized block design with study arm allocation known only to the study
coordinator until the day before surgery was used for randomization.
Fig. 1 Patient enrollment.
Beginning in the preoperative holding area, patients had four drops of lavender or
coconut oil placed on their left and right wrists. The patients were then instructed
to rub their wrists and hands together and then inhale and exhale slowly for 1 minute.
Intraoperatively, the anesthesia team applied four drops of lavender or coconut oil
on the patient's temple at every 2 hours until completion of the surgery. Postoperatively,
nurses were instructed to administer the lavender or coconut oil to patients at every
4 hours in the same manner as previously described between the hours of 6:00 a.m.
and 6:00 p.m. Between 6:00 p.m. and 6:00 a.m., the essential oil was applied to a
cotton ball that was placed within 20 cm of the patient. This method of administration
was adapted from previous studies that examined the benefits of essential oils on
perioperative anxiety and nausea and vomiting.[19 ]
[20 ] Blood pressure, pulse rate, and respiratory rate were monitored in the intraoperative
and postoperative period approximately 30 minutes after the administration of the
essential oil. The scales used to measure patient pain, sleep quality, anxiety, and
depression were administered to patients in the preoperative holding area and then
every postoperative day (POD) throughout the patient's hospitalization.
Postoperative Care
All patients were managed postoperatively according to previously published ERAS pathways
for patients undergoing microvascular breast reconstruction.[21 ]
[22 ] Briefly, patients receive multimodal pain management beginning in preoperative holding
that consists of acetaminophen, gabapentin, celecoxib, naproxen, oxycontin, and a
scopolamine patch. Intraoperatively, patients receive ketamine boluses in addition
to fentanyl and transversus abdominis plane blocks with liposomal bupivacaine. Postoperatively,
pain management regimens consist of nonopioid medications (e.g., acetaminophen, celecoxib,
and gabapentin) with oral opioids available for breakthrough pain. On POD1, patients
are encouraged to get out of bed to the hospital chair and ambulate with assistance
beginning on POD2. Patients are typically discharged from the hospital on POD3 or
POD4.
Data Collection and Outcome Measures
Baseline patient demographic and clinical variables included age, body mass index
(BMI), race, smoking status, medical comorbidities, psychiatric history, oncologic
breast history (e.g., breast cancer type and hormone receptor status), breast surgery
history, receipt of adjuvant or neoadjuvant chemo/radiation therapy, and receipt of
hormonal therapy. Reconstructive variables included timing of breast reconstruction,
laterality, and type of microvascular breast reconstruction. Surgical complications,
including delayed wound healing, hematoma, seroma, and infection were captured. Additionally,
adverse events, defined as a documented allergic reaction that occurred in either
experimental arm were collected.
The main outcomes of interest included pain scores, sleep scores, and measures of
anxiety and depression. Secondary outcomes of interest included an analysis of how
blood pressure, heart rate, and respiratory rate were influenced by lavender oil in
the perioperative period. Pain scores were captured using the visual analogue scale
and recorded on a scale of 1 to 10.[23 ] Sleep scores were measured using the Richards–Campbell Sleep Questionnaire.[24 ] This is a brief 6-item questionnaire where a total score is calculated to represent
the overall quality of a person's sleep, with higher scores representing better sleep.
The Hospital Anxiety and Depression Scale (HADS) was used to quantify levels of anxiety
and depression.[25 ] This 14-component scoring system is used to tabulate a total sum score (0–42) or
separate anxiety and depression scores (0–21). A higher score represents a more severe
degree of anxiety and/or depression. The severity of a patient's anxiety or depression
may then be grouped into three categories based on the score (normal, 0–7; borderline
abnormal, 8–10; and abnormal 11–21).
Statistical Analysis
Categorical variables were summarized using frequency and percentage and compared
using Fisher's exact tests. Continuous variables were summarized with median, interquartile
range (IQR), and compared using Wilcoxon's rank sum tests. Linear mixed-effect models
were used to investigate HADS (total, anxiety, and depression), sleep, and pain scores.
Variables for time point (preoperative/POD1/final POD) and arm were analyzed while
adjusted for the patient's reconstruction laterality. First the interaction of time
point by study arm was tested. If the interaction was nonsignificant, it was dropped
from the model before testing the main effects for time and arm. The underlying covariance
structure used for all mixed models was unstructured and estimation was based on the
restricted maximum likelihood method. Adjustment for multiple pairwise comparisons
used the Scheffe method. Analyses were conducted based on the intent to treat principle.
p -Values of <0.05 were considered statistically significant. Line plots illustrate
changes in vital signs over time. At each time point, the mean and corresponding 95%
confidence interval has been plotted. All analyses were conducted using SAS software
(Version 9.4; SAS Institute Inc., Cary, NC), and plots were created in the R language
and environment for statistical computing (R Foundation for Statistical Computing,
Vienna, Austria).
Results
Baseline Characteristics of Study Population
From December 2017 to January 2020, 58 patients were enrolled in this study. Nine
patients (two in the experimental group and seven in the control group) were excluded
from final analysis after withdrawing from the study or due to incomplete data and/or
medical records. The baseline characteristics of the study cohort are in [Table 1 ]. The median age was 47.6 years (range, 32–68 years). The two study groups were similar
with respect to BMI, race, smoking status, medical comorbidities, and presence of
past psychiatric history (including anxiety and depression). Most patients underwent
a delayed (46.9%, n = 23) form of microvascular breast reconstruction, received bilateral reconstruction
(57.1%, n = 28), and received a deep inferior epigastric artery perforator flap (55.1%, n = 27). Reconstruction laterality was the only variable that varied significantly
between the experimental and control groups with a larger number of patients in the
experimental group undergoing unilateral breast reconstruction (59.3%, n = 16 vs. 22.7%, n = 5; p = 0.02).
Table 1
Patient characteristics
Placebo/control (n = 22)
n (%)
Experimental (n = 27)
n (%)
Total (n = 49)
n (%)
p -Value
Age at surgery (y)
n
22
27
50
0.7938[a ]
Median (IQR)
46.7 (40.6–54.9)
48.2 (41.9–56.0)
47.6 (41.1–54.9)
Range
31.6–67.5
31.8–66.9
31.6–67.5
BMI
n
22
27
49
0.6152[a ]
Median (IQR)
27.8 (25.5–31.8)
27.9 (24.4–30.4)
27.9 (24.7–30.7)
Range
20.9–40.1
19.6–36.0
19.6–40.1
Time elapsed between date of diagnosis and date of reconstruction? (mo)
n
21
26
47
0.7891[a ]
Median (IQR)
21.6 (14.1–112.5)
20.0 (14.8–33.1)
20.0 (14.4–47.2)
Range
3.3–274.6
2.1–218.1
2.1–274.6
Race
Asian
2 (9.1)
0 (0.0)
2 (4.0)
Black or African American
2 (9.1)
5 (17.9)
7 (14.0)
White
12 (54.5)
20 (71.4)
32 (64.0)
Multiracial
3 (13.6)
0 (0.0)
3 (6.0)
Other
1 (4.5)
0 (0.0)
1 (2.0)
Unavailable
2 (9.1
3 (10.7)
5 (10.0)
Smoker, current or former
6 (27.3)
10 (38.5)
16 (33.3)
0.5421[b ]
Medical comorbidities
Hypertension
4 (18.2)
6 (21.4)
10 (20.0)
1.0000[b ]
Diabetes
1 (4.5)
1 (3.6)
2 (4.0)
1.0000[b ]
Hyperlipidemia
1 (4.5)
1 (3.6)
2 (4.0)
1.0000[b ]
Chronic pain disorder
2 (9.1)
3 (11.5)
5 (10.4)
1.0000[b ]
Documented psychiatric history
9 (40.9)
13 (50.0)
22 (45.8)
0.5729[b ]
Psychiatric illnesses
Anxiety
7 (31.8)
9 (32.1)
16 (32.0)
1.0000[b ]
Depression
4 (18.2)
8 (28.6)
12 (24.0)
0.5116[b ]
Other
1 (4.5)
0 (0.0)
1 (2.0)
0.4400[b ]
Breast surgery history
Simple mastectomy
17 (77.3)
23 (82.1)
40 (80.0)
0.7317[b ]
Nipple sparing mastectomy
2 (9.1)
3 (10.7)
5 (10.0)
1.0000[b ]
Sentinel lymph node biopsy
11 (50.0)
11 (39.3)
22 (44.0)
0.5685[b ]
Axillary lymph node dissection
7 (31.8)
12 (42.9)
19 (38.0)
0.5595[b ]
Lumpectomy
3 (13.6)
1 (3.6)
4 (8.0)
0.3136[b ]
Breast cancer type
DCIS
5 (22.7)
3 (10.7)
8 (16.0)
0.2770[b ]
Invasive
17 (77.3)
22 (78.6)
39 (78.0)
1.0000[b ]
Not documented
1 (4.5)
0 (0.0)
1 (2.0)
0.4400[b ]
Hormone receptor status
ER+
15 (68.2)
17 (60.7)
32 (64.0
0.7676[b ]
PR+
10 (45.5)
17 (60.7)
27 (54.0)
0.3926[b ]
Her2+
7 (31.8)
4 (14.3)
11 (22.0)
0.1781[b ]
Radiation therapy
13 (59.1)
20 (74.1)
33 (67.3)
0.3612[b ]
Chemotherapy
16 (72.7)
20 (74.1)
36 (73.5)
1.0000[b ]
Chemotherapy timing
Neoadjuvant
6 (27.3)
10 (35.7)
16 (32.0)
0.5589[b ]
Adjuvant
10 (45.5)
9 (32.1)
19 (38.0)
0.3888[b ]
Other
0 (0.0)
1 (3.6)
1 (2.0)
1.0000[b ]
Hormonal therapy
14 (66.7)
19 (70.4)
33 (68.8)
1.0000[b ]
Documented psychiatric history
9 (40.9)
13 (50.0)
22 (45.8)
0.5729[b ]
Psychiatric illnesses
Anxiety
7 (31.8)
9 (32.1)
16 (32.0)
1.0000[b ]
Depression
4 (18.2)
8 (28.6)
12 (24.0)
0.5116[b ]
Other
1 (4.5)
0 (0.0)
1 (2.0)
0.4400[b ]
Reconstruction type
Unilateral
5 (22.7)
16 (59.3)
21 (42.9)
0.0193[b ]
Bilateral
17 (77.3)
11 (40.7)
28 (57.1)
Contralateral prophylactic reconstruction
16 (72.7)
12 (44.4)
28 (57.1)
0.0808[b ]
Type of free flap
Unilateral DIEP
3 (13.6)
2 (7.4)
5 (10.2)
Bilateral DIEP
16 (72.7)
11 (40.7)
27 (55.1)
Stacked DIEP
2 (9.1)
13 (48.1)
15 (30.6)
Other
1 (4.5)
1 (3.7)
2 (4.1)
Abbreviations: BMI, body mass index; DCIS, ductal carcinoma in situ; DIEP, deep inferior
epigastric artery perforator; ER, estrogen receptor; HER2, human epidermal growth
factor 2; IQR, interquartile range; PR, progesterone receptor.
a Wilcoxon's rank sum p -value.
b Fisher's exact p -value.
Hospital Anxiety and Depression Scores
Summary results regarding the effects of lavender oil on HADS anxiety and depression
scores are shown in [Table 2 ]. HADS anxiety scores revealed that the mean preoperative, POD1, and final POD scores
for all patients were 9.4 (standard deviation [SD] ± 3.62), 7 (SD ± 3.49), and 6.9
(SD ± 3.16), respectively. Preoperatively, 36.6% (n = 15), 24.4% (n = 10), and 39% (n = 16) of patients were classified as “normal,” “borderline abnormal,” and “abnormal”
based on the anxiety scores, respectively. At the time of the final postoperative
survey, 59% (n = 23), 28.2% (n = 11), and 12.8% (n = 5) of patients were classified as “normal,” “borderline abnormal,” and “abnormal”
based on the anxiety scores, respectively. HADS depression scores revealed that the
mean preoperative, POD1, and final POD scores for all patients was 14.6 (SD ± 1.73),
14.9 (SD ± 2.55), and 15.7 (SD ± 1.94), respectively. Preoperatively, 4.9% (n = 2), and 97.5% (n = 39) of patients were classified as “borderline abnormal” or “abnormal,” respectively,
based on the depression scores. At the time of the final postoperative survey, 2.6%
(n = 1) and 97.4% (n = 38) of patients were classified as “borderline abnormal” and “abnormal,” respectively,
based on the depression scores.
Table 2
Summary results for outcomes
Placebo/control (n = 22)
Experimental (n = 27)
Total HADS, preoperative
n
17
24
Mean (SD)
10.9 (7.07)
11.9 (7.32)
Median (IQR)
10.0 (7.0–14.0)
10.0 (7.0–16.0)
Range
0.0–28.0
3.0–30.0
Total HADS, POD1
n
16
24
Mean (SD)
9.9 (5.65)
10.3 (7.79)
Median (IQR)
9.5 (6.5–13.5)
9.0 (4.0–14.0)
Range
0.0–21.0
2.0–39.0
Total HADS, final POD
n
16
23
Mean (SD)
9.4 (6.21)
8.0 (5.40)
Median (IQR)
9.5 (4.5–12.5)
6.0 (4.0–12.0)
Range
1.0–24.0
0.0–19.0
HADS anxiety, preoperative
n
17
24
Mean (SD)
9.6 (4.40)
9.2 (3.03)
Median (IQR)
10.0 (6.0–11.0)
9.0 (7.0–12.0)
Range
3.0–21.0
4.0–15.0
HADS anxiety, preoperative
n (%)
0–7: normal
5 (29.4)
10 (41.7)
8–10: borderline abnormal
5 (29.4)
5 (20.8)
11–21: abnormal
7 (41.2
9 (37.5)
HADS anxiety, POD1
n
16
24
Mean (SD)
7.6 (3.65)
6.5 (3.40)
Median (IQR)
7.0 (4.0–11.0)
6.0 (4.0–8.0)
Range
3.0–13.0
2.0–16.0
HADS anxiety, final POD
n
16
23
Mean (SD)
7.7 (3.52)
6.3 (2.82)
Median (IQR)
7.0 (5.0–9.5)
6.0 (4.0–8.0)
Range
3.0–16.0
2.0–12.0
HADS anxiety, final POD
n (%)
0–7: normal
9 (56.3)
14 (60.9)
8–10: borderline abnormal
4 (25.0
7 (30.4)
11–21: abnormal
3 (18.8)
2 (8.7)
HADS depression, preoperative
n
17
24
Mean (SD)
14.9 (1.68)
14.4 (1.77)
Median (IQR)
15.0 (14.0–16.0)
15.0 (14.0–16.0)
Range
12.0–18.0
10.0–16.0
HADS depression, preoperative
n (%)
8–10: borderline abnormal
0 (0.0)
2 (8.3)
11–21: abnormal
17 (100.0)
22 (91.7)
HADS depression, POD1
n
16
24
Mean (SD)
14.8 (2.14)
15.0 (2.84)
Median (IQR)
14.0 (13.5–16.5)
16.0 (13.0–16.5)
Range
11.0–18.0
7.0–19.0
HADS depression, final POD
n
16
23
Mean (SD)
15.8 (2.77)
15.7 (1.11)
Median (IQR)
15.5 (14.0–17.5)
16.0 (15.0–16.0)
Range
10.0–21.0
14.0–18.0
HADS depression, final POD
n (%)
8–10: borderline abnormal
1 (6.3)
0 (0.0)
11–21: abnormal
15 (93.8)
23 (100.0)
Sleep scores, preoperative
n
17
24
Mean (SD)
6.5 (2.61)
5.6 (2.39)
Median (IQR)
6.5 (4.2–8.3)
5.3 (3.8–7.7)
Range
1.3–10.0
1.7–10.0
Sleep scores, POD1
n
16
24
Mean (SD)
5.5 (1.97)
5.1 (2.08)
Median (IQR)
5.7 (4.3–7.2)
5.7 (3.2–6.3)
Range
1.5–8.3
1.8–9.2
Sleep scores, final POD
n
16
23
Mean (SD)
6.6 (2.43)
7.4 (1.71)
Median (IQR)
6.8 (4.5–8.9)
7.8 (5.8–8.7)
Range
2.7–10.0
3.5–10.0
Pain score, preoperative
n
18
24
Mean (SD)
0.5 (1.36)
0.9 (1.80)
Median (IQR)
0.0 (0.0–0.0)
0.0 (0.0–0.5)
Range
0.0–5.5
0.0–5.0
Pain score, POD1
n
14
22
Mean (SD)
3.5 (1.91)
3.1 (2.74)
Median (IQR)
4.0 (2.0–4.7)
2.6 (1.2–4.5)
Range
0.0–6.0
0.0–9.0
Pain score, final POD
n
15
23
Mean (SD)
2.5 (1.92)
2.9 (2.81)
Median (IQR)
2.2 (0.6–4.0)
2.3 (0.0–5.0)
Range
0.0––6.0
0.0–9.0
Abbreviations: HADS, hospital anxiety and depression scale; IQR, interquartile range;
POD, postoperative day; SD, standard deviation.
Mixed linear-effect models tested the interaction of the randomization arm and time
point on all scores. No significant interaction effects were observed. Next, the main
effects for arm and time were tested. Overall, no significant differences were found
based on randomization into the experimental or control groups when comparing total
HADS (p = 0.36), HADS anxiety (p = 0.82), or HADS depression scores (p = 0.21; [Table 3 ]). However, when considering the total patient cohort, significant differences in
HADS scores were seen based on time (p < 0.01). Notably, HADS anxiety scores were found to be significantly lower in the
postoperative period as compared with preoperative scores (p < 0.001), while HADS depression scores were found to be significantly higher in the
postoperative period as compared with the preoperative period (p = 0.005; [Table 4 ]; [Fig. 2 ]). This suggests that patients may experience lessened anxiety but increased symptoms
of depression in the immediate postoperative period following microvascular breast
reconstruction.
Fig. 2 Least squares mean estimates and 95% confidence intervals obtained for the total
patient cohort from the linear mixed models: p -values are from mixed model test of main effect for time. HADS, hospital anxiety
and depression scale; POD, postoperative day.
Table 3
Mixed-model results
Outcome
Effect
p -Value
HADS total scores
Arm × time
ns
Arm
0.3589
Time
0.0004
HADS anxiety scores
Arm × time
ns
Arm
0.8163
Time
<0.0001
HADS depression scores
Arm × time
ns
Arm
0.2069
Time
0.0131
Sleep scores
Arm × time
ns
Arm
0.8585
Time
<0.0001
Pain scores
Arm × time
ns
Arm
0.3002
Time
<0.0001
Abbreviations: HADS, hospital anxiety and depression scale; ns, nonsignificant then
dropped from model before testing main effects.
Table 4
Pairwise comparisons
Score
Pairwise comparison
Scheffe's p -value
HADS total
Pre-op and POD1
Pre-op and final POD[a ]
POD1 and final POD[a ]
0.13
<0.001
0.036
HADS anxiety
Pre-op and POD1[a ]
Pre-op and final POD[a ]
POD1 and final POD
<0.001
<0.001
0.94
HADS depression
Pre-op and POD1
Pre-op and final POD[a ]
POD1 and final POD
0.48
0.005
0.052
Sleep score
Pre-op and POD1
Pre-op and final POD[a ]
POD1 and final POD[a ]
0.082
0.021
<0.001
Pain score
Pre-op and POD1[a ]
Pre-op and final POD[a ]
POD1 and final POD
<0.001
<0.001
0.14
Abbreviations: HADS, hospital anxiety and depression scale; POD, postoperative day;
Pre-op, preoperative.
a Indicates a pairwise comparison with adjusted p < 0.05.
Sleep Scores, Pain Scores, and Vital Signs
Outcomes regarding sleep and pain scores are summarized in [Table 2 ]. Mean preoperative, POD1, and final POD sleep scores were 6.0 (SD ± 2.49), 5.3 (SD ± 2.02),
and 7.1 (SD ± 2.04), respectively. Mean preoperative, POD1, and final POD pain scores
were 0.7 (SD ± 1.62), 3.3 (SD ± 2.43), and 2.7 (SD ± 2.48), respectively. Overall,
no significant differences were seen between the arms for sleep (p = 0.86) or pain (p = 0.30) scores ([Table 3 ]). However, when considering the total patient cohort, significant differences in
sleep and pain scores were seen based on time. In general, patients demonstrated improved
sleep scores (p = 0.021) and improved pain scores throughout the length of their hospitalization
(p < 0.001; [Table 4 ]; [Fig. 2 ]).
The physiologic effect of lavender oil on respiration rate, heart rate, systolic blood
pressure, and diastolic blood pressure was measured in the preoperative and postoperative
period and compared between groups. No significant differences were seen at any of
the measured time points ([Fig. 3 ]). In addition, no significant differences in the rates delayed wound healing (22.7
vs. 21.4%; p = 1.00), hematoma (4.5 vs. 17.9%; p = 0.21), seroma (4.5 vs. 14.3%; p = 0.37), infection (22.7 vs. 32.1%; p = 0.54), or flap congestion (4.5 vs. 0%; p = 0.44) were seen, and no adverse events related to the essential oil were documented
throughout the trial ([Table 5 ]).
Table 5
Complications by arm
Placebo/control (n = 22)
Experimental (n = 27)
Total (n = 49)
p -Value
Surgical complication
n (%)
No
9 (45.0)
11 (44.0)
20 (44.4)
1.0000[a ]
Yes
11 (55.0)
14 (56.0)
25 (55.6)
Surgical complications
n (%)
Delayed wound healing
5 (22.7)
6 (21.4)
11 (22.0)
1.0000[a ]
Hematoma
1 (4.5)
5 (17.9)
6 (12.0)
0.2109[a ]
Seroma
1 (4.5)
4 (14.3)
5 (10.0)
0.3681[a ]
Infection
5 (22.7)
9 (32.1)
14 (28.0)
0.5374[a ]
Flap congestion
1 (4.5)
0 (0.0)
1 (2.0)
0.4400[a ]
a Fisher's exact p -value.
Fig. 3 Trends of vital signs over time before and after essential oil application in the
treatment arms. Time points represented are (1) clinic visit, (2) preoperative before
lavender administration, (3) preoperative after lavender administration, (4) intraoperative,
(5) PACU, (6) postoperative day (POD) 1, (7) POD2, and (8) POD3. PACU, post anesthesia
care unit.
Discussion
To our knowledge, this is the first randomized controlled trial to directly assess
the therapeutic benefits of lavender oil to alleviate perioperative pain, anxiety,
depression, and improve sleep in patients undergoing microvascular breast reconstruction.
The results of this study suggest that there are no significant differences in HADS
anxiety or depression scores, pain scores, sleep scores, or vital signs when comparing
patients who received lavender oil and the control group who received coconut oil.
In addition, the application of lavender oil did not result in an increased incidence
of adverse events or surgical-related complications when compared with the control
group. Overall, the results of this study suggest that lavender oil may be safely
used as an adjunct in patients undergoing microvascular breast reconstruction as part
of multimodal analgesic treatment; however, it should not be used as a sole therapy
to treat anxiety, depression, sleeplessness, or perioperative pain.
Linalool and linalyl acetate are chemical constituents of lavender that are believed
to contribute to its therapeutic effect through inhibition of GABA(A) receptors in
the central nervous system to induce a state of relaxation and mitigate pain perception.[14 ]
[15 ] This proposed mechanism has also been suggested to reduce physiological stress responses
and trigger reductions in blood pressure and heart rate.[26 ]
[27 ] The results of our study suggest that there are no meaningful differences in blood
pressure or heart rate when comparing the vital signs of patients who receive lavender
oil and those in the control group. However, further research is needed to elucidate
the potential physiological effects of inhaled lavender oil on the surgical patient.
Prior studies examining the therapeutic benefit of aromatherapy and lavender oil have
presented conflicting results. Essential oils have been suggested to reduce perioperative
anxiety and pain in patients admitted to intensive care units and those awaiting ambulatory
surgery, in addition to patients undergoing cesarean sections and coronary artery
bypass grafting.[11 ]
[12 ]
[16 ]
[19 ]
[20 ]
[21 ]
[28 ]
[29 ]
[30 ]
[31 ]
[32 ] Specifically, Olapour et al reported that among patients undergoing a cesarean section,
those who received lavender oil as part of their multimodal pain regimen demonstrated
less postoperative pain and improved satisfaction with pain control.[32 ] However, the authors emphasize in this study that while lavender oil appears to
be safe and demonstrates a therapeutic effect, it should not be used as a sole measure
for pain management. In contrary, Kim et al reported on the analgesic effects of lavender
oil in patients undergoing a breast biopsy, finding that there was no objective difference
in patient-reported pain scores. However, patients' who received lavender oil reported
improved satisfaction with their pain control.[33 ] Salamati et al demonstrated similar findings when assessing the benefits of lavender
oil in patients undergoing open-heart surgery, found that lavender oil did not confer
an objective, measurable benefit with pain control.[21 ] Our results support these findings and suggest that while lavender aromatherapy
may not display a direct analgesic or mood stabilizing effect, it does provide a low-cost
and safe adjunct that has the potential to positively impact a patient's subjective
perception of their treatment-related pain.[34 ] It should be noted, however, that the utilization of an ERAS pathway may have influenced
the results of this study. The use of ERAS pathways in microvascular breast reconstruction
have been shown to reduce opioid use and improve pain control as compared with traditional
postsurgical pathways.[35 ]
[36 ] It is unclear if the effects of lavender aromatherapy may be more pronounced in
the absence of the perioperative multimodal pain regimens seen in conjunction with
ERAS pathways.
While no meaningful difference existed with respect to the severity of HADS depression
and anxiety scores, interesting trends were observed when examining the total patient
cohort. In general, breast reconstruction patients were found to display a reduction
in their HADS anxiety scores prior to discharge. However, HADS depression scores persisted
in the “borderline abnormal” to “abnormal” range signifying the pervasiveness of mental
health issues, like depression, throughout the perioperative period of breast reconstruction.
These results are like previous studies examining the long-term psychosocial outcomes
associated with breast reconstruction.[2 ]
[3 ]
[4 ] Metcalfe et al reported on the psychosocial functioning of women diagnosed with
breast cancer, found that breast reconstruction patients demonstrated improved psychosocial
scores over time with relatively low levels of distress and depression over a 6-year
period.[2 ] However, the authors note that a significant portion of women continue to experience
moderate-to-severe cancer-related distress which is especially notable among patients
who undergo a form of delayed breast reconstruction.[2 ] Women who undergo mastectomy, with or without breast reconstruction, tend to experience
heightened psychosocial distress in the acute perioperative period following a cancer
diagnoses and surgery. This may be attributed to concerns regarding a patient's postmastectomy
appearance and factors such as a lower household income, lack of social support, higher
tumor stage, and a history of depression and/or anxiety.[37 ]
[38 ]
[39 ]
[40 ] Patients may be at the highest risk for anxiety and depression during first year
after a cancer diagnoses. This may be related to a fear of recurrence, loss of social
support, and fatigue and pain after surgery which prevents a return to normal daily
activities.[41 ] Previous studies note, however, that psychosocial distress related to a breast cancer
diagnoses or surgery tends to resolve overtime and does not appear to have a long-standing
effect on patient wellbeing.[42 ]
[43 ] Our study population was largely composed of women who underwent a form of delayed
breast reconstruction which may account for the high level of depression seen in this
cohort. It is important for reconstructive surgeons to recognize that patients seeking
breast reconstruction may have higher baseline levels of cancer-related distress and
should be ready to offer the appropriate avenues for support and counseling when needed.
Limitations
We acknowledge several limitations. The study sample size was not calculated a priori
but was chosen arbitrarily. This is the first study designed to evaluate the use of
lavender oil in breast reconstruction patients and thus a power analysis could not
be performed with a degree of confidence. Therefore, a type-II error cannot be ruled
out as the statistical power of this study may limit our conclusions. Most of the
hospital discharges took place on POD3 or POD4, and this length of follow-up time
may not have been adequate to observe an effect between the patient groups. Additionally,
complications experienced by patients may have affected perceptions of pain, anxiety,
depression, and sleep. However, due to the low incidence of complications in this
cohort, the independent effect of postoperative complications on individual scores
could not be assessed. In addition, we did not capture a baseline incidence of sleep
disturbance in our patient cohort, and variations in a patient's baseline emotional
state, pain tolerance, and the effect of standard postoperative pharmacologic agents
on lavender essential oil were not controlled in this study. Lastly, while this trial
was blinded to the patient, the distinct scent of the lavender oil makes it difficult
to utilize a placebo in this study.
Conclusion
We hypothesized that in patients undergoing microvascular breast reconstruction, topical
and aromatherapy with lavender oil would result in objective improvements in perioperative
pain, anxiety, depression, and sleep. However, the results of this study suggest that
lavender oil does not confer measurable advantages when used as an adjunct to multimodal
pain regimens. Despite this, lavender oil is a low-cost therapeutic that may be safely
used in the postoperative management of the breast reconstruction patient. Further
investigation into the potential role of aromatherapy in patients undergoing breast
reconstruction is warranted.
