Plant Fragrances Are Like Music for Our Senses: A Scoping Review of Aromatherapy in Gynecologic Cancers and Breast Cancer Care

Selection process

The databases PubMed and CINAHL were searched from January to April 2021, using Boolean operators for the terms listed in Table 1 in a multi-stage, iterative process.

Inclusion criteria included empirical primary studies on EO applications in women with gynecologic and breast cancers, via nasal application and/or percutaneous application, in English, and without limitations in date of publication. Excluded were research not focused on the targeted group, basic biochemistry research, publications in other languages and/or without access to abstracts and full text. The first author (J.C.) undertook a three-step review process to rank titles, review abstracts, and full text. Then, this was repeated by a second researcher. Findings were regularly discussed within the larger research team.

Data analysis

Data analysis began with a full-text reading of all included papers to generate a first broad overview. Parallel to this, the literature was tracked in a table matrix including categories of interest to the research, including: author, country of research and journal, purpose, objective of the study, EO used, targeted population, research design, methods of data collection and analysis, and most relevant results.

In a second round, a qualitative, structuring content analysis was conducted, based on Mayring ³⁴ : First, a deductive category system was developed according to topics stemming from the research questions. Corresponding passages from the publications were reduced to the essential statements. Then, in a second and third pass the passages were summarized and sorted to a higher abstraction level. Unique or interesting material raised in the literature was developed into inductive categories to iteratively refine the category system.

Aromatherapy research in women with gynecologic cancers and breast cancer: description of key features

The 10 included studies were published between 1995 and 2021. Five of the 10 studies were published in nursing journals, and the other 5 were published in medical journals. Nine studies are quantitative studies, of which four are randomized controlled trials (RCTs), ^{13,36 –38} two have cross-over designs, ^39,40 one is a semi-comparative trial, ²⁴ and two are pilot studies. ^27,41 It is striking that only one of these studies calculated effect sizes for the outcomes of interest. ⁴¹ One RCT included a qualitative sub-survey, ³⁷ and one study had a qualitative design with conducted semi-structured interviews. ⁴²

All studies were conducted with women with breast cancer, however, in different stages of treatment. Only two studies explicitly targeted women with breast cancer in aftercare. ^24,42 Study outcomes of interest included: symptom management, ³⁶ QOL, ^36,37,39,41 nausea and emesis, ^27,39 sleep, ^37,38 skin reactions, ⁴¹ anxiety, ^13,24,38 pain, ^38,41 depression, ^24,38 effects on lymphodema, ⁴² dietary intake, ⁴⁰ adverse events, ³⁷ vital signs, ³⁷ and immunologic parameter. ²⁴ Five studies used single EO: ginger ^39,40 and lavender. ^13,38,42

The other studies used blended EO in different mixtures. ^{24,27,36,37,41} Precise description about the composition of the intervention product was given in some studies with the botanical name detailed, ^{24,27,39 –41} the mixing ratios of blended EO mixtures, ^27,36,41 the origin of the oils, ^{24,39 –41} and the storage of the EO during the study. ²⁷ In three studies, the reasons for the selection of the EO remain unknown. ^24,36,37 In one study, the patients had the possibility to choose between different EO. ³⁷ Only Zorba and Ozdemir ²⁷ referred to an underlying theoretical framework, namely, the theory of unpleasant symptoms.

All studies reported on a heterogenic array of outcomes. Statistical significance were determined by six studies on the following topics: reduced anxiety, ^13,24 lowered aggression hostility, ²⁴ alleviated acute nausea, ³⁹ increased QOL, ^36,39 symptoms measured by the Rotterdam symptom checklist, ³⁶ reduced incidence of nausea and pharyngeal reflex, ²⁷ and increased energy. ⁴⁰ Four studies, on the other hand, showed no significant effects concerning skin reactions, QOL, or patient satisfaction, ⁴¹ overall treatment effect in reducing chemotherapy-induced nausea and vomiting, ³⁹ QOL, sleep quality, and vital sign, ³⁷ anxiety, depression, and sleep. ³⁸

Qualitative results showed overall positive subjective effects of applications with EO, such as experiencing relaxation, comfort, and general enjoyment of the intervention. ^37,42 Kirshbaum ⁴² reported the additional subjective perception of pain alleviation, reduction in swellings, improvement in self-esteem, and a greater feeling of support.

Four studies reported that no adverse events occurred, ^27,37,38,41 or only minor incidents were reported such as mild dizziness. ³⁹ In the remaining five studies, adverse events were not mentioned. ^{13,24,36,40,42} Thus, no study explicitly reported major adverse events.

Specifications and context: what does aromatherapy mean?

Descriptions of the term aromatherapy varied, not only concerning the degree of complexity but also regarding the content itself. These included: listing ingredients, describing the application form, reporting effects, and classifying context. Two publications explained aromatherapy based on all the earlier mentioned descriptions. They specified the use of EO via inhalation or absorption of the oil into the skin, contextualized them as part of CIM, and distinguished the effects aiming at alleviating physical and/or emotional symptoms. ^40,43

Tamaki et al. ³⁷ add the improvement of spiritual well-being and “various diseases” to the spectrum of potential effects and differentiate between EO extractions from herbs, flowers, and other plants, however, without referring to potential forms of applications. These authors and one other refer to aromatherapy's historical roots as a healing art in ancient Egypt and India. ^37,42 General statements describe aromatherapy as widely used practices, but they do not define meanings of the concept explicitly. ^{13,24,27,36,42}

One reference points to the spiritual effects of aromatherapy. ⁴¹ The terminology used to distinguish the separate effects of aromatherapy is inconsistent and includes: physical and psychological effects of aromatherapy, ⁴¹ physical and emotional effects, ^39,40 physical, emotional effects and effects on spiritual well-being, ³⁷ mental, physical, and spiritual effects, ¹³ and effects on body and mind. ⁴² This implicitly links to different fields of discourse.

Five studies embed aromatherapy as a practice in the field of CIM therapies. ^{13,24,37,39,40} Only one of these studies specifically embeds aromatherapy as a practice in the field of nursing. ¹³

The study by Halm et al. ⁴¹ occupies a special position within the included studies: The pilot study investigates the effect of a blend of EO on skin reactions in women undergoing radiotherapy for breast cancer. Although the use of EO in form of a topical application is an aromatherapeutic approach, the term aromatherapy is not used throughout the publication. Similarly, the study by Shammas et al. ³⁸ carries out an intervention with EO, without explicitly referring to the practice of aromatherapy and only listing the term aromatherapy as a keyword. Thus, in both studies, the term aromatherapy is neither conceptualized nor described, as is the case in the other included studies.

Principle of action: how does aromatherapy work?

Most publications cite the possible effects of aromatherapy based on findings from previous research. References to the specific effects of aromatherapy cited include: reducing nausea (with ginger, ³⁹ mint, ³⁶ or “scented oils” in general ⁴⁰ ); anxiolytic properties (lavender ³⁸ ); pain control (rosemary and lavender ³⁶ ); or depression (lavender, bergamot, and frankincense ³⁶ ).

Other studies, however, remain quite unspecific in descriptions of effects: Beyliklioğlu and Arslan ¹³ describe the probable effects as increased attention, decreased anxiety, and relaxation, also as effects on vital signs such as pulse, respiratory rate, and blood pressure but do not specify the substances linked to these effects. Similarly, in the oldest included study, Kirshbaum ⁴² remarks only broadly on the effects associated individual EO. Only a few studies refer to the underlying principle of action or question reasoning behind potential effects. A distinction can be made between two explanatory models, which can be described as (1) biochemical explanation and (2) olfactory explanation ^† :

Reported challenges: what to consider in aromatherapy research?

Although the included studies differ from each other in terms of research design, objectives, and target population, similarity can be found in the specific methodological challenges associated with aromatherapy, which are of interest for planning future aromatherapy research. The most frequently mentioned challenges included the issues of (1) blinding the intervention and (2) influence of standard treatment on the effects of aromatherapy:

Further challenges are mentioned by single studies concerning the following: investigator bias because of researchers' subjective attitudes toward aromatherapy ⁴⁰ ; no or minimal reaction to the aromatherapy intervention due to preconditioning of the control-group in a cross-over design, for instance in combination with chemotherapy and nausea ³⁹ ; difficulties to show significant effects of aromatherapy due to a ceiling effect ³⁷ ; irritation or dropout because of the smell of the intervention product ⁴¹ ; and difficulties in asking questions about the subjective experiences with aromatherapy without provoking a thematic bias. ⁴²

Implications for future research: what improvement are recommended?

Analysis of the included studies gave specific points that should be considered when investigating aromatherapy. The first point underscored the relevance of qualitative approaches in aromatherapy research. Some studies highlighted the seemingly strong effects of aromatherapy expressed through subjective perceptions, despite little demonstrated objective effects. ³⁹ Moreover, Ovayolu et al. ³⁶ state their concern that patient feedback may not be considered when data do not demonstrate objectifiable measurement.

However, even minimal positive effects may have the potential to make a difference in the QOL for those affected, a sentiment echoed by Shammas et al. ³⁸ For these reasons, Tamaki et al. ³⁷ recommend the use of aromatherapy in clinical practice to offer the patients options, better tailored to their expectations. This accentuates the need to take subjective odor preferences into consideration. ¹³ Recommendations to contend with an intervention that cannot be blinded is only given by Halm et al. ⁴¹ They suggest the management of patients' expectations of aromatherapy and long-term adherence by transparently explaining these in the informed consent.

What the studies tell us …

The results described reflect the obvious challenges in aromatherapy research. First, questions arise about heterogeneous conceptualization and definition of the key term “aromatherapy,” for example: What does the term aromatherapy mean precisely within a given context and what context should be used? Answers should be addressed more directly in the use of aromatherapy in cancer patients. Our results found that aromatherapy is often contextualized as a part of CIM. ^{14,44 –46} Despite the frequent association with CIM, aromatherapy is not explicitly mentioned in relevant clinical practice guidelines. ³

One reason may be that the guidelines only include recommendations using integrative therapies with a sufficient evidence base and aromatherapy may fall in the category of integrative treatment for which there is insufficient evidence to enable a clear recommendation. This demonstrates the need for high-quality research on aromatherapy in clinical practice to make such recommendations.

The second challenge is the often in-transparent selection and justification of the EO used, their composition and mixing ratios. This problem has already been addressed through guidelines for reporting on studies with herbal interventions in general. ⁴⁷ Such ambiguities make it difficult to translate applications precisely into practice, and to replicate or conduct further research on presented evidence. The elaborated CONSORT statement for controlled trials of herbal interventions can and should be used as a guide in this regard.^(ibid.)

A third challenge is the blinding of aromatherapy applications due to the characteristic smells of EO. ^38,41 In an attempt to blind an intervention in a cross-over design, two research teams used fragrance-matched artificial placebos differing in composition but not in smell. ^39,40 In both cases, there was no explicit mention that the blinding procedure controls the biochemical principle of action only, yet the olfactory effect may occur in both the control and intervention as the same fragrance was used. This point should be considered in future research for an explicit understanding of what is exactly controlled and measured—and what is not.

A real blinding of aromatherapy interventions with consideration of both, the olfactory and the biochemical principle of action, seems only possible when the intervention itself is not known to the participants. However, an approach to obfuscate the intervention would then lack distinction between the two principles of action. Further, such an approach raises ethical concerns.

Moreover, the studies implicitly refer to the relevance of subjective perceptions within research about aromatherapy: The analysis of the included studies illustrates neither homogenous results about the effectiveness of aromatherapy in general nor comparable statistical results with quantitative outcomes. One reason for this is that the included publications are extremely diverse in terms of study design, research question, outcomes of interest, EO used, application, data collection, and indication of effect sizes.

In addition, the complexity of EO compositions and the two different principles of action make it extremely challenging to capture the differentiated effects with quantitative measurements. The perceptions of the participants, when collected, are homogenous regarding the positive evaluation. Despite this, research activities within this field rely disproportionately on quantitative approaches: Nine of our 10 included studies are quantitative studies. Only one of these ³⁷ includes a qualitative survey with open questions.

Another (quantitative) study implicitly points to the relevance of qualitative methodological approaches in aromatherapy research, by stating how “[p]ositive feedback by patients themselves could be of particular concern […]” (ibid.: 403), aiming at explaining, expanding, and completing what quantitative results cannot capture. Hence, the findings of this scoping review demonstrate that methodological diversity is necessary to capture the subtle and complex effects of aromatherapy interventions, requiring creativity and customization in study design. This in line with research about aromatherapy in patients with cancer in general.

What the studies neglect to tell us …

The included studies exclusively target women with breast cancer, and no primary study that focused on women with gynecologic cancers was found. This research gap needs to be addressed in further research to focus on the specific needs of women in and also after gynecologic cancer treatment. The need exists: A recent study emphasized how stress reduction may be a means to support care needs of gynecologic cancers survivors by helping manage side effects of disease and/or treatment and cope with anxiety. ⁴⁸

Surprisingly, challenges in capturing the effects of EO are hardly discussed in connection with their complex compositions. The EO are multi-substance mixtures made up of plant components that differ, depending on the nature of the substrate, weather conditions, and harvesting time, and therefore will always vary in composition, which may lead to different research results. To minimalize this challenge, it is appropriate to classify EO consistently, as other publications have. ^10,49 Important information would be: What is the composition? What company are they from? Which plants are the EO derived from? And, in the case of blended EO, what are the mixing ratios?

Consistency in reporting is likewise important concerning the two principles of action of EO applications, which are only rarely described and not explicitly considered in methodological rationale in the included studies, despite the call to do so for several decades. ⁵⁰ Due to the different principles of action of aromatherapy, the use of EO per inhalation and through the skin goes beyond a mere fragrance therapy. Fragrance is only absorbed through the sense of smell and subjectively evaluated, meaning it has an olfactory effect ⁵¹ that can vary widely from person to person, even when the EO used is the same.

Beyond subjective olfactory, the biochemical effect through skin and mucous membranes should also be taken into account in aromatherapy research. Yet disagreement still exists about which principle of action has the stronger effect or is the most important to research. According to Herz, ¹² for example, the strongest effect of smell comes from the meaning that the smell has for a person, thus from the olfactory principle of action, suggesting that the biochemical effect via the skin plays a secondary role.

This research suggests that expectations play an important role in changing moods and emotions through smell, ¹¹ and raises further questions about the role of odor-induced mood changes. The indications for linking odor to emotion introduce an important perspective into aromatherapy research within the context of integrative medicine. This could concern the effects of emotion on stress, fatigue, coping with pain, and sleep disturbance. Although the two principles of action always interact and should be considered in research designs, research on aromatherapy rarely takes both principles into account.

Finally, it should be noted that even though the studies included are limited to clinical research and to a specific target group, the potential, key issues, and challenges identified are certainly relevant to other settings, sectors, and targeted people. Nevertheless, consideration of specific conditions and characteristics should be taken into account for further research.

Limitations

This study was conducted under a number of restraints that contributed to some limitations, despite our best efforts. This research was limited to only two searched databases considered most relevant for our topic. To minimize this limitation, the reference lists of our included publications were also manually searched to identify additional relevant studies. Despite this, some relevant studies may have been missed. Further, the study language was limited to English. Following the character of a scoping review and our research's underlying objectives, the included studies were not evaluated according to the quality of the evidence but rather assessed according to the characteristics of the findings instead.