Introduction

Nausea and vomiting in pregnancy (NVP) is a common condition. It is stimated that the  prevalence of nausea  during pregnancy  is  between  50-80%,  whilst  the  prevalence  of  vomiting  and  retching  during  pregnancy  is  approximately  50%  (1).  NVP  can  significantly  reduce  one’s quality  of  life and have negative effects upon daily activities,  marital  life,  parenting  roles,  work life, and social  functioning whilst also accounting for  33%  of  all  cases  of  work-related  sickness  absence  during  pregnancy (2-3). NVP can also impose a  great  economic  burden  on both individuals and society  due  to the loss  of  productivity  and  rising healthcare  costs it creates  (4). Such findings point to  the  need  for effective  evidence-based interventions in this context.

Despite  its  high  prevalence  and  associated  social  and  economic  burden,  a wholly effective  treatment  for NVP has  yet  to  be  established (5). Moreover, the  teratogenic  effects  of  medications  previously  used  for  the  management  of  NVP  have  led  to  caution  in  the  prescription  of  medications, particularly  during  the  first  trimester  of  pregnancy  (4).  Indeed, pregnant people generally prefer non-pharmacological interventions due  to  concerns  about  their  side  effects  on  the  fetus  (1). As such, alternate interventions and a thorough examination of their effectiveness are required.

Complementary  and  alternative  medicine  (CAM)  is often preferred as  a  natural  and  safe alternative to  alleviate pregnancy  symptoms in general (6).  The  use of  CAM  during  pregnancy  ranges  from  20%  to  60%,  with  aromatherapy  being  one  of  the  most  popular (7). Indeed, aromatherapy  is  one  of  the  fastest  growing  therapies  in  the  world  today (8). Yet there is a paucity of high-quality evidence for the use of aromatherapy in pregnancy, particularly about NVP, and findings remain inconclusive in many cases.

Lavender  oil  is  the  most  commonly  used  oil  in  pregnancy  (9),  and  is  considered  safe  for use  during  pregnancy  (10–14).  Indeed,  an  in-vitro  safety  assessment demonstrated  no  detrimental  effects using  both  low  and  high  lavender  concentrations  in  a  widely  used  placental  cell  model  (14). The  key  constituents  of  Lavandula  angustifolia,  the  most  commonly  used  species  of  lavender,  are  linalyl  acetate  and  linalool  (7).  Linalool   and  linalyl  acetate  in  lavender  works by stimulating  the  parasympathetic  system,  linalyl  acetate  has  narcotic  properties  and  linalool can act  as  a  sedative  (11).  Such effects show promise for the alleviation of NVP, and one  pilot  randomized  controlled  trial (RCT) has already demonstrated  that  a  combination  of both peppermint  and  lavender  oil can  reduce  the  mean  score  of  nausea  and  vomiting, along with fatigue, and promote an  increase in  energy  levels during  pregnancy  (11).  Nevertheless, there remains a paucity  of  available studies in this area and thus there is a need for further  research  to  reach  a  definitive  conclusion  regarding  the  effect  of  lavender  as  an  aromatherapy  agent  on  NVP.

Alongside NVP,  it  is  well-established  that  anxiety  is  one  of  the  most  prevalent  conditions  affecting  almost  one-third  of  pregnancies  (15).  This is concerning as evidence  suggests  that  anxiety  may  also  increase  the  risk  of  preterm  birth  and  low  birth  weight infants (15).  Some  research  has  thus  far  focused  on  the  effects  of  aromatherapy  using  lavender  oil  on  anxiety  levels  during  pregnancy  (12,13). Yet, none have examined the effect of lavender oil on both anxiety and NVP together. Some  studies  have  suggested  that  anxiety  plays  a  role  in  the  incidence  of  nausea  and  vomiting  (16).  Moreover,  some  researchers  have  reported  an  association  between  anxiety  and  NVP  (17–19).  Whilst  the  use  of  mint  oil  has  been  found  to  positively  reduce  NVP,  without  affecting  state  anxiety  (20),  a  Cochrane  review  has  demonstrated  little  conclusive  evidence  regarding  the  overall  efficacy  of  aromatherapy  in  alleviating  NVP  (4).  Considering the above, further  research is required to  provide  conclusive  recommendations  on  the  use  of  aromatherapy,  particularly  with the use of  lavender  oil  and its effect on both  nausea,  vomiting,  and  anxiety  in  pregnancy.  The aim of this study was therefore to determine the effect of lavender aromatherapy on nausea, vomiting, and anxiety in pregnancy. 

Materials and Methods

A  non-randomized,  single-blind,  quasi-experimental  study  was  conducted  with  two  parallel  groups  of  lavender  oil  (intervention)   and  placebo.  It  was  recorded  in  the  Iranian  Registry  of  Clinical  Trials  Center  with  the  code  IRCT201306082324N12  on  09/23/2014  (URL of  registry:   https://en.irct.ir/trial/1967).  This  study  was  conducted  between  June,  2015  and January,  2016  at  the  Baharloo  hospital,  a  training  and  treatment  center  affiliated  with  Tehran  University  of  Medical  Sciences  (TUMS),  Iran.  Participants  included  those  pregnant   and  with  a  gestational  age  of  between  6  and  16  weeks  who  were  experiencing  mild  to   moderate  symptoms  of   nausea  and  vomiting. 

Based  on  a previous study where the  Rhodes Index of Nausea, Vomiting, and Retching (RINVR) was used on day three of the intervention period (11), and the  mean  (SD)  of  the  placebo   (n=23)  (6.37)  and  the  intervention  group  (n=18)  (6.08) as used  to  calculate  differences  among  the  two  groups,  a  sample  of  33  participants in  each  group  was  estimated. The study with this sample size had a power of 90% at a 5% significance level. However, considering 10% drop out, a sample of  36  participants  per  group  was  estimated.

In a convenience sampling, of the total number of participants screened for eligibility (n=83), 72  met  the  inclusion  criteria  and  were  assigned to  either the  lavender (n=36) or placebo group  (n=36). To prevent participants contacting each other, they were  allocated to either the intervention or placebo group on alternate days. Aroma bottles were dispensed to participants by a  research  assistant  who  was unaware of the research objectives. The research assistant was blinded to the two groups.

Recruitment commenced after ethical approval was granted by the Research Ethics Committee of TUMS.  Subsequently, the researcher explained the objectives and procedure of the study to potential participants who met the inclusion criteria, who were then encouraged to ask any questions about the research before providing their informed consent.  Written informed consent, was obtained from those who were willing to participate in the study. 

The socio-demographic characteristics questionnaire prepared through a literature review, consisted: age, gestational age, BMI (body mass index), gravidity, economic and occupational status, and level of education of participants. 

The RINVR consists of eight items rated on a five-point Likert scale to measure nausea and vomiting.  The total score of this index ranges from zero to 32.  A score of zero indicates an absence of symptoms, other scores range from 1-8, 9-16, and 17-32, each indicating mild, moderate, and severe symptoms respectively. It has been validated  in  a separate  study  assessing  complications,  such  as  NVP  (21).  The  reliability  of  the  Persian  version  of  the RINVR was  confirmed  based  on  Cronbach’s  alpha  coefficient  test  (α=0.80) (22).  However, in the present study, the Cronbach’s alpha coefficient of the RINVR has been calculated as 0.89. 

The state anxiety scale used and  prepared  by  Spielberger  consists  of  20  self-report  statements, with  each  item  ranging  from  1-4  (23).  The  total  score  ranges  from  20  (the  lowest  level  of  anxiety)  to  80  (the  highest  level  of  anxiety)  (24).  Scores are categorized as follows; mild anxiety, 20-39; moderate anxiety, 40-59; and severe anxiety, 60-80.  The  reliability  of  the  Persian  version  of  this  tool  was  confirmed  based  on  a Cronbach’s  alpha  coefficient  of  0.90  (25).  However, the Cronbach’s alpha coefficient for this tool was found to be 0.89 in the present study. The reliability of both the Spielberger and RINVR scales was assessed with a two-week test-retest design with 15 pregnant people who were not included in the main study.

Participants met the inclusion criteria if they were aged between 20-34 years old and currently pregnant with a singleton planned pregnancy. Participants were also eligible if they reported experiencing mild to moderate NVP and had a state anxiety score  of  <60.  Participants were excluded if they had  following: A diagnosis  of  a  physical  or  psychological  disorder;  exposure  to  stressful  events  <3  months;  allergies  to  herbs;  olfactory  disorders;  antiemetic  or  anti-anxiety  drug  (herbal  or  chemical  drugs)  usage  <24-hours;  and  a  lack  of  self-reporting  on  the  use  of  sedative  medications.  Participants were also excluded  during  the  seven-day  intervention  period  if  they  experienced  a crisis  or  stressful  event;  were  unwilling  to  continue  participating  in  the  study;  did  not  comply  with  inhalation  instructions  for  two  consecutive  or  non-consecutive  days; were using other antiemetic or anti-anxiety  drugs  or  any other  complementary  medications  that  could  potentially  reduce  NVP  or  anxiety  during  the  study.

The instructed frequency  and  duration  of  inhalation during this aromatherapy intervention  were  guided  by  Ghani’s  previous  study  (11).  As it was considered inappropriate for participants to use an  oil  burner  at  home,  dripping  a  higher  percentage  of  lavender  (10%)  onto  cotton for inhalation  was  suggested as an appropriate and comparable alternative by  an aromatherapy specialist.

Participants were advised to stop aromatherapy if they experienced any  of  the  following  allergic  reactions:  runny  nose,  skin  rash,  itching,  headache,  burning  eyes,  and/or  abdominal  pain.  They also received seven copies of the  RINVR to complete after  oil  inhalation  every  night  for a period of seven  days.  To ensure regular inhalation, a daily record checklist was also provided.  Participants were asked to submit their completed Rhodes nausea-vomiting index to the research  team  after  the  seven-day  intervention  period.  They were also asked to complete the state anxiety scale both before and after the intervention period. The RINVR and the state anxiety scale were completed through self-reporting. The RINVR typically required around 3-5 minutes, while the state anxiety scale took about 5-10 minutes to complete.

On the first day of the intervention period, participants completed  a  socio-demographic  data  questionnaire,  the  RINVR,  and  a  state  anxiety  scale.  All  participants  also  received  routine  training  on  how  to  relieve  NVP,  such  as  avoiding  certain  foods,  odors,  activities,  or  situations  that  may  exacerbate  NVP  (11).  NVP was considered to be a  primary  outcome, and state anxiety was considered to be  a  secondary outcome.

Participants in the lavender group inhaled lavender oil whilst participants in the placebo group inhaled  sesame oil. Inhalation occurred  twice a day (once before afternoon rest and again just before nighttime sleep) for seven days.  For each inhalation participants extracted four drops of the bottle’s contents onto a piece  of  cotton  and  then  inhaled  this  for  20  minutes  at  a  distance  of  20  cm  from  their  nose.

Identical glass containers (aroma bottles) were pre-coded by the pharmacist using the letters A and B. Assignment remained blind to the research team.  A research assistant, who  was also  unaware  of  the  research  objectives,  assigned  bottles  to  participants.  Nevertheless, we considered that participants would be able to ascertain the bottle’s content once inhaled. Thus, only  the  assessor  was truly blinded  to  each  group’s  assignment.

The lavender oil (lavender group) and the sesame oil used as a placebo (placebo group) were both provided by Kashan Barij Essence Pharmaceutical  Company  (Kashan,  Iran).  This lavender oil was extracted from the flowering aerial parts of the lavender plant (Lavandula angustifolia) and then diluted with odorless sesame oil at a ratio of 10%.  The sesame  oil  used as a placebo  was  procured  as  a  harmless  oil  used  during  pregnancy  (26,27).

Descriptive statistics were used to interpret the sample characteristics.  The normal distribution of quantitative data was evaluated using the Kolmogorov-Smirnov test.  The Chi-square and Fisher’s exact test were also used to compare qualitative variables.  In addition, both paired and independent sample t-tests were used to determine differences in quantitative data both within and between the groups.  Furthermore, a repeated measures analysis of variance (ANOVA) was employed to compare  the  mean  RINVR scores  both  within  and  between  the  groups  during  the seven-day  intervention  period.  The mean RINVR score was compared two-by-two over time using the Bonferroni test.  To examine comparisons between the two groups, the Greenhouse-Geisser test was  also used  where  there  was  no  assumption  of  sphericity.  SPSS version 16 was used for all data analyses.  A P-value of <0.05 was considered statistically significant.

Results

In total, 66 of the 72 recruited participants completed this study (33 participants in each group).  Three participants were excluded from the study because they were inhaling  the  oils irregularly and outside of the timeframes prescribed (one  in  the  lavender group and  two in the placebo group).  A further three participants were removed due to their  use  of  antiemetic  and/or  anti-anxiety  drugs (one from the placebo group and  two  participants  from  the  lavender  group)  (Figure  1).

There was no statistically significant difference between the two groups regarding socio-demographic data.  The 47% of the participants were primigravida (Table 1). 

Prior to the intervention, there  were  no statistically significant  differences  between  the  two  groups  regarding  the  mean  RINVR  scores (P= 0.914) (Tables  2  and  4).   The independent  sample  t-test conducted  on scores taken on the  seventh  day of the intervention period revealed that  the  overall  mean RINVR score  of  the  lavender  group  was  lower  than  the overall mean  RINVR score of the placebo group  (6.91±3.58 and 9.58±4.31, respectively)  (P=0.008).  This result is similar to those reported when comparing the severity of  NVP between the two groups  (P=0.032)  (Table  3).

Results from the Bonferroni test demonstrated  no statistically significant difference  in  the  mean  RINVR scores  of  the  lavender (intervention) and  placebo groups  over the time  (P=0.117). Yet  a statistically significant decreasing trend was observed in both groups over the time (P=0.001 and P=0.004,  respectively)  (Table  2  and  Figure  2).

On  intervention day 7, the mean score  of  state  anxiety in  the  lavender  group  was  markedly lower  than  the mean score of  state  anxiety in the placebo group (36.21±5.68 and 40.27±7.02,  respectively,  P=0.012) (Table 4). This result was also similar when  comparing  the severity of state anxiety between the two groups (P=0.027) (Table 5).  The paired t-test equally revealed that the mean score of state anxiety had also distinctly decreased in both groups (Table  4).

Discussion

This study investigated the effect of aromatherapy using lavender oil  and  sesame  oil  as  a  placebo  on  nausea,  vomiting,  and  anxiety between the  6^th and 16^th  weeks  of  pregnancy  over  a  period  of  seven days.  The mean score of nausea and vomiting were compared between the two groups (Intervention and placebo). A statistically significant difference was only identified  between  the  two  groups  in  terms  of  the  total  RINVR score at the end of the intervention period whereby on the seventh day, the mean RINVR score  was statistically, significantly  lower  in  the  lavender  group  than  in  the  placebo group.  State anxiety was also compared between the two groups on the seventh day of the intervention period, and our statistical analysis  demonstrated that both severity and the mean score of state anxiety were markedly lower  in  the  lavender  group  than they were in the placebo group.      

This study contributes to the limited number of valid studies conducted thus far on the effects of aromatherapy using lavender  oil  on  NVP.  Ghani et al. (2013)  previously investigated the impact of a lavender and peppermint  oil  combination  on  NVP  at  8-16  weeks  gestation. In their pilot study, between one and four drops of peppermint and lavender  oil  2%,  respectively were  added  to  a  large  spoon  of  water  in  an  oil  burner  in  a  closed  room.  Participants were subsequently asked to breathe deeply in the  room  saturated with this  aroma  for  20 minutes, twice a  day.  After three days, the total RINVR score was significantly reduced compared  to scores in  the  pre-intervention  phase.  Nevertheless,  despite  this  reduction,  the  severity  of  NVP  remained  high  (11). Conversely in the present study, the severity of NVP on the  seventh  day  of the intervention period was either mild  (69.7%)  or  moderate  (27.3%), and  none  of our  participants  had  severe  NVP.  A  key strength of our study is that the lavender aroma was studied in isolation whereas when aromas are mixed such as in this earlier pilot study (11), the impact  of  the  intervention  cannot  be  attributed  to  any  one  particular  aroma.  This also has implications for any future research.

In  our sample, the mean total RINVR score decreased  in  both  the  lavender  and  placebo groups  during  seven  days  of  the  intervention  period.  This decline in the placebo group may be attributed to placebo effects. For example, in the reporting of clinical trials it has been speculated that placebo  treatments  exert  their  effects  through  psychological  and  mental  factors (28).  In this sense, one’s belief in the efficacy of treatment can positively affect the sensory experience of treatment. Considering the above it will be important not to overclaim the effects of placebos in this context, particularly where they are not statistically significant.

Lavender  oil  is  the  most  studied  aromatic  oil  in the amelioration of  anxiety  and  depression  in  obstetrics (7). Yet  few  studies have examined the effect of lavender oil on anxiety in pregnancy.  In two  randomized  controlled  trials it was  reported that lavender  oil  was  one  of  the  three  oils  (lavender,  petitgrain,  and  bergamot)  used  in  the  third  trimester  of  pregnancy  (12,13). Yet Igarashi  et al.  (2010)  reported that aromatherapy was evidenced to have no statistically  significant  effects  on  state  anxiety  between  the  32^nd  and  36^th  weeks  of  pregnancy   (13).  In  another  trial  conducted  by  Igrashi  et al. (2012),  five  minutes  of  inhaling  an  essential  oil  with  high  linalool  and  linalyl  acetate  content was also evidenced to have little to no effect in improving scores related to tension and anxiety (12). Our results differ in that  the positive effect  of  lavender  oil on  state  anxiety we have found is  promising.  The reason for this contradictory finding may be related to differences  in  gestational  age  and  the frequency with which aromatherapy had been administered.   Yet often, earlier studies in this field have included small sample sizes and the type of oil used by  participants  is  not  reported.  Thus, the conclusions of these earlier studies remain unreliable, whilst ours may be more reliable due to our larger sample size and detailed reporting. Overall, few studies  have  investigated  the  use  of  essential  oils  for  the  treatment  of  NVP  and  anxiety  in  pregnancy, and the use of different oils can affect results, for example where the  positive  effects  of  aromatherapy are observed in relation to  NVP,  but  not state  anxiety  (20). As such, further studies are also required  to  investigate  the  effect  of  different  aromatic  essential  oils  on  the  alleviation  of  NVP. 

Lavender  oil  is  an  aromatic  herbal  extract  with  carminative  (a  smooth  muscle  relaxant),  sedative,  and  antidepressant  properties (29).  Its  anticholinergic,  antihistaminic,  and  anti-inflammatory  activities  resemble  those  of  antiemetic  drugs  (30).  Linalool  and  linalyl  acetate,  the  main  chemical  constituents  of  lavender  (31)  are  known  to  depress  the  central  nervous  system  and  induce  sedative  effects,  and  marked  narcotic-like  activities  (29).  Lavender inhibits the hypothalamic-pituitary-adrenal axis and reduces the secretion of corticotropin-releasing hormone.  It  also  reduces  the  secretion  of  adrenocorticotropic  hormones  from  the  pituitary  gland,  decreases  the  release  of  cortisol,  increases  the  secretion  of  serotonin  from  the  adrenal  gland,  and thus leads  to  anxiety  alleviation  (32).  Accordingly, we hypothesize that lavender oil used in aromatherapy would have a statistically significant positive effect on NVP and state anxiety. Whilst we only identified a statistically significant difference between the two  groups  in  terms  of  the  total  RINVR score at the end of the intervention period (7 days), future studies could usefully measure these effects over longer time periods. 

The  primary  effect  of  aromatherapy  is  the  stimulation  of  olfactory  receptor  cells  in  the  nasal  epithelium  and  the transmission of  neural  signals  to  the  brain,  limbic  system,  and  thalamus  to achieve both mental and physical  balance through  the  release  of  endorphins  and  serotonins  (33).  Aromatherapy  can  directly  affect  the  limbic  system by  bypassing oral  intake  and  protecting  against  adverse  reactions  during  pregnancy  (7).  There  are  no  known  adverse  effects  or  complications  associated  with  this type of aromatherapy  during  pregnancy,  whether  by  inhaling  or  applying  essential  oils  topically  (11,34–36). Indeed, we observed no adverse effects in our sample. This suggests that lavender oil may be used effectively and safely  as  a  low-cost  intervention  for  the  management  of  nausea,  vomiting,  and  anxiety  during  pregnancy. 

Whilst this study has been conducted rigorously, it has not been possible to blind  the  participants as to which group they were in due to the nature of aromatherapy (e.g., participants having prior recognition of lavender scent). Moreover, the research team had no control over participants’ psychological state during the intervention period. Whilst these are limitations, they could not be mitigated. A key strength of this study was that we were able to test the effectiveness of a single oil (lavender oil) on both NVP and state anxiety on a large sample size.  Future studies may usefully compare the effects of different essential oils delivered via aromatherapy over longer time periods in this context.

Conclusion

Drawing from the results presented here, aromatherapy with lavender oil resulted in a statistically significant positive effect on day seven of the intervention period. So, it can be considered safe and effective in reducing nausea,  vomiting,  and  anxiety  during  pregnancy. Increasing the duration of this intervention may enable further positive effects to emerge. Due to the simplicity, safety and cost-effectiveness of using aromas in this context, care providers could promote the use of lavender aromatherapy during pregnancy for the amelioration of NVP and anxiety. 

Declarations

Acknowledgements

This study was submitted as a partial fulfillment of the requirements for a master’s degree in Midwifery, which was supported by TUMS. The authors express their appreciation to the participants and personnel of Baharloo teaching hospital affiliated with TUMS.  We thank the Research Deputy, School of Nursing and Midwifery, Research Ethics Committee of TUMS as well as Barij Essence Company.

Conflicts of interest

The authors declared no conflicts of interest.

Funding

The study was supported by the TUMS, Tehran, Iran.  Kashan Barij Essence Company provided the essential oils used during the intervention period of this study. This work was financially supported by the TUMS and Kashan Barij Essence Company (No. PC9402).  The company had no role in either the design or the undertaking of this research (e.g., collection, analysis, and interpretation of data; and/or in the writing of this manuscript).

Ethical approval

The  study  protocol was written in accordance with declaration of Helsinki and was approved by the Research Ethics Committee of TUMS, Tehran, Iran. Informed consent was obtained from all subjects.  Participants were ensured of the confidentiality of their information, and they were informed of their right to leave the study  at  any  time  without  consequence  and  without  giving  a  reason.  All data were kept in a secured and locked personal file which was only accessed by the research team.

Code of Ethics

IR.TUMS.REC.  2478/130/92.

Use of Artificial Intelligence (AI)

No AI was used in all part of this research.

Authors’ contribution

AA conceptualization, methodology, investigation, resources, validation, formal analysis, writing, original draft.  MK conceptualization, methodology, investigation, resources, supervision, formal analysis, writing – review, and editing.  SP writing, editing and interpretation.  RB conceptualization, methodology, formulating and supplying the aromatherapy oils used in the study. AM formal analysis and SHJ conceptualization and critical appraisal. All authors read and approved the final manuscript and agreed to be accountable for all aspects of the work.