Safety and stress-reducing effects of rosemary extract: an open-label trial and randomized double-blind crossover study

Abstract

This study comprised two trials that evaluated the safety and anti-stress effects of rosemary extracts containing rosmarinic acid and diterpenes. Study I was a 4-week, single-arm, open-label safety trial involving 22 healthy adults who consumed high doses of the rosemary extract daily. No clinically relevant adverse events or abnormal laboratory findings were observed, thus confirming its safety. Study II was a randomized, double-blind, placebo-controlled, crossover trial with 40 participants that assessed psychological and physiological stress markers. After four weeks of rosemary extract intake, significant improvements were found in trait anxiety scores measured using the State-Trait Anxiety Inventory and heart rate variability, as indicated by a decreased low-frequency/high-frequency ratio. Additionally, a single dose administered before the mental workload test significantly reduced state anxiety scores. No significant differences were observed in salivary cortisol levels or visual analog scale scores between the groups. These results suggest that rosemary extract is safe for consumption and may effectively reduce chronic and acute stress in healthy individuals. These findings support the use of rosemary extract as a natural dietary supplement for stress management.

Introduction

In modern society, stress is a prevalent issue that many individuals encounter daily and significantly affects their physical and mental health. Stress is recognized as a key contributor to various health issues, including depression, sleep disturbances, and lifestyle diseases. These health problems not only diminish the quality of life but also impose substantial economic costs on society. The coronavirus disease 2019 (COVID-19) pandemic has further exacerbated this situation, leading to increased stress levels across all age groups, from children to adults, and a corresponding rise in depression, anxiety, and other mental health disorders, making mental health issues more pronounced as a societal concern.^(1–4) Addressing these societal issues and preventing stress-related health conditions is essential.

As the importance of mental health has become widely recognized and stress management is emphasized as a key component of overall well-being, dietary supplements and plant-derived compounds have garnered significant attention as convenient approaches for reducing stress. Rosemary (Salvia rosmarinus) has been used in cooking and traditional medicine for centuries, and its components have been shown to possess antioxidant and anti-inflammatory properties, suggesting their potential stress-reducing effects.^(5,6) Rosmarinic acid, a polyphenol found in rosemary, has potent antioxidant effects.^(7,8) Furthermore, rosmarinic acid was reported to exert an anti-stress effect via hippocampal neurogenesis in a forced swimming test in mice.⁽⁹⁾ This may protect the nervous system by scavenging free radicals and reducing cellular oxidative stress.

Diterpenes found in rosemary mainly consist of carnosic acid and carnosol, which have been reported to possess anti-inflammatory and neuroprotective properties. Diterpenes are believed to play a role in reducing oxidative stress in nerve cells and maintaining nervous system health.⁽⁸⁾ Research involving animals implied that diterpenes exhibit anti-stress activities.^(10,11) Specifically, they may enhance the activity of intracellular antioxidant enzymes and suppress the production of inflammatory cytokines, thereby modulating stress responses and protecting nerve cells. This makes them promising candidates for the prevention of stress-related disorders.⁽¹²⁾

Rosemary is a valuable source of essential oils and extensively used in cosmetics, perfumes, and aromatherapy.^(13,14) Despite their benefits, the direct oral intake of essential oils poses risks of gastrointestinal and renal inflammation, leading to safety concerns. Considering the potential risks associated with the direct oral intake of essential oils, we previously focused on the residue remaining after essential oil extraction. Our investigation revealed that this residue contains anti-stress compounds, such as rosmarinic acid and diterpenes.⁽¹⁵⁾ Based on these findings, extraction from the residue after essential oil removal offers a promising approach for creating safe and easily usable anti-stress food ingredients for both culinary and medicinal applications. Previous studies have suggested that the intake of rosemary extract (RME) may improve physiological responses to stress. However, research specifically examining the effects of RME containing rosmarinic acid and diterpenes (carnosic acid and carnosol) as the main components is limited. Therefore, the primary objective of this study was to assess the safety of the excessive consumption of RME derived from the residue of essential oil extraction with rosmarinic acid and diterpenes (Study I), followed by an investigation of its effects on chronic and acute stress to determine its efficacy (Study II).

Materials and Methods

Study I

Study design

This study was designed as a single-arm, nonrandomized, open-label trial conducted at Clinical Creative Co., Ltd. (Sapporo, Japan) from December 23, 2021, to February 25, 2022. Each participant received RME for four weeks. The study was performed in accordance with the Declaration of Helsinki (2013) and the Ethical Guidelines for Medical and Health Research Involving Human Subjects (2015) [Ministry of Education, Culture, Sports, Science and Technology (MEXT), and the Ministry of Health, Labour and Welfare (MHLW), Japan]. The study protocol was approved by the Ethics Committee of NIPPN Corporation (Tokyo, Japan; permission number: 21-03) and the Sapporo Yurinokai Hospital Ethics Committee (Sapporo, Japan; permission number: 021). All experiments were conducted by Clinical Creative Co., Ltd. and Sapporo Yurinokai Hospital. This trial was registered at www.umin.ac.jp/ctr/index.htm (registration number: UMIN000046475).

Study participants

The study participants were selected from 31 candidates through a volunteer bank. Informed consent was obtained from all participants. The inclusion criteria were as follows: (1) Japanese men and women aged 20–65 years at the time of providing consent and capable of attending outpatient visits; (2) healthy individuals without severe organ dysfunction or specific diseases who were not undergoing related treatments or any pharmacotherapy; (3) those who could consume the test food for four weeks; (4) individuals who could maintain their daily lifestyle habits during the study period; and (5) those who could provide written informed consent.

The exclusion criteria were as follows: (1) individuals with a history of treatment for heart failure, myocardial infarction, or other cardiovascular diseases; (2) individuals currently undergoing treatment for atrial fibrillation, arrhythmias, liver dysfunction, renal impairment, cerebrovascular disorders, rheumatism, diabetes, dyslipidemia, hypertension, or other chronic diseases; (3) individuals classified as Category D (requiring medical intervention) according to the Japan Health Checkup Association criteria; (4) those who regularly use medications (including herbal medicines) or supplements; (5) individuals with allergies to medications or rosemary; (6) pregnant or breastfeeding individuals, or those who may become pregnant during the study period; (7) individuals who have participated in another clinical trial within the past month or are currently participating in one; and (8) any other individuals deemed unsuitable for participation in this study by the principal investigator. A sample size of 22 participants was used to test statistical significance.

Intervention

NIPPN Corporation provided the RME, and the test diets were obtained from Sunsho Pharmaceutical Co., Ltd. (Fuji, Japan). The test diet contained microcrystalline cellulose, caramel color, calcium stearate, silicon dioxide, and RME [rosmarinic acid 2.5 ‍mg, rosemary-derived diterpenes (carnosic acid and carnosol) 4 ‍mg/1 capsule] (Table 1). The participants consumed 10 capsules of the test diet per day for four weeks. On the day of testing, participants were required to fast for at least 10 ‍h before undergoing the examinations.

Table 1.

Composition of the capsules

	RME capsules	Placebo capsules
Ingredients	Rosemary extract powder containing rosmarinic acid and diterpene (carnosic acid and carnosol), microcrystalline cellulose, caramel color, calcium stearate, silicon dioxide	Microcrystalline cellulose, caramel color calcium stearate, silicon dioxide
Amount of active ingredents per capsule	Rosemary extract powder containing rosmarinic acid and diterpene (carnosic acid and carnosol) 100 ‍mg (contains 2.5 ‍mg of rosmarinic acid, 4 ‍mg of diterpene)	0 ‍mg

Safety evaluation

Safety was assessed through physical measurements, blood biochemical analyses, hematological assessments, and urinalysis, which were performed at screening and again at four weeks after the intervention. Blood biochemical analyses included total protein, albumin, albumin-to-globulin ratio, total cholesterol, low-density and high-density lipoprotein cholesterol, triglycerides, fasting blood glucose, free fatty acids, phospholipids, blood urea nitrogen, creatinine, uric acid, total bilirubin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, lactate dehydrogenase, gamma-glutamyl transferase, creatine phosphokinase, sodium, potassium, chloride, and iron. Hematological assessments included white and red blood cell counts, hemoglobin, hematocrit, platelet count, and glycated hemoglobin. Urinalysis included specific gravity, pH, protein, glucose, urobilinogen, and occult blood. Blood biochemical analyses, hematological assessments, and urinalysis were conducted by Daiichi Kishimoto Clinical Laboratories Co., Ltd. (Sapporo, Japan). Adverse events were monitored throughout the study period, and their relationship with the study product assessed by the principal investigator.

Statistical analysis

Statistical analyses were conducted for each measurement parameter, with results expressed as mean ± SD. A paired t test was performed to assess the differences between baseline (week 0) and week 4 using a significance level of 5% for two-tailed tests. All statistical analyses were performed using R software ver. 4.0.3 (Institute for Statistics and Mathematics, Vienna, Austria; www.r-project.org).

Study II

Study design

This study was designed as a randomized, double-blind, placebo-controlled, crossover trial conducted at HUMA R&D Co., Ltd. (Tokyo, Japan) from June 24, 2024, to September 27, 2024. Each participant received both the active treatment and placebo in two separate 8-week periods, with a 4-week washout phase between these periods. The study was performed in accordance with the Declaration of Helsinki (2013) and the Ethical Guidelines for Medical and Health Research Involving Human Subjects (2015) [Ministry of Education, Culture, Sports, Science and Technology (MEXT), and the Ministry of Health, Labour and Welfare (MHLW), Japan]. The study protocol was approved by the Ethics Committee of NIPPN Corporation (permission number: 24-01) and the Tokyo Shinjuku Clinic Ethics Committee (Tokyo, Japan; permission number: RD01001AS04). All experiments were conducted by HUMA R&D Co., Ltd. and Tokyo Shinjuku Clinic. This trial was registered at www.umin.ac.jp/ctr/index.htm (registration number: UMIN000054368).

Study participants

The study participants were selected from 106 candidates through a volunteer bank. Informed consent was obtained from all participants. The inclusion criteria were as follows: (1) healthy Japanese adults aged 20–60 years at the time of providing written informed consent; (2) individuals who were stressed on a daily basis; (3) those whose Beck Depression Inventory-II scores were under 20; (4) individuals who exhibited a systolic blood pressure between 101 and 139 mmHg and diastolic blood pressure between 61 and 89 mmHg; (5) those who had a regular or part-time job; and (6) individuals who had the capacity to consent, fully understood the study, voluntarily applied, and provided written consent to participate after receiving a full explanation of the study’s purpose and content.

The exclusion criteria were as follows: (1) individuals who were currently under medication or receiving outpatient treatment for a serious disease; (2) those who were previously engaged in exercise or nutritional therapy under the guidance of a physician; (3) individuals with allergies to medications (aspirin) or foods (rosemary) related to the study; (4) those with current or previous history of drug or alcohol dependence; (5) individuals who were in the hospital for mental (depression, etc.) or sleep disorders (insomnia, sleep apnea syndrome, etc.), or those who had previous history of a psychiatric disorder; (6) those who had extremely irregular life rhythms, such as shift or night workers; (7) individuals whose eating, sleeping, and other habits were extremely irregular; (8) those who were having an unbalanced diet; (9) heavy alcohol drinkers (average net alcohol intake of approximately 60 ‍g/day or more) or excessive smokers (21 cigarettes/day or more); (10) individuals with current or previous history of brain disease, malignancy, immunological disease, diabetes mellitus, hepatic disease (hepatitis), renal disease, cardiac disease, thyroid disease, adrenal disease, metabolic disease, or other serious diseases; (11) individuals using health foods, supplements (such as gamma-aminobutyric acid, rosemary, caffeine, and theanine), or medications that may affect stress or those unable to refrain from their use; (12) those who regularly drank energy drinks; (13) individuals who anticipated a major change or stressful situation in their home or working environment (e.g., job change, move, or marriage) from one month before screening to study completion; (14) those who participated in other clinical studies/research within three months prior to providing consent or planned to participate in other clinical studies/research during the study period; (15) individuals who donated more than 200 or 400 ‍ml whole blood, plasma, or platelets within one or three months, respectively, before providing consent; (16) pregnant or lactating women, or those who may become or are intending to become pregnant during the study period; (17) individuals who have difficulty complying with recording each survey form; (18) those whose laboratory test values or measurements at screening indicate their ineligibility to participate in the study, and (19) individuals deemed inappropriate for the study by the principal doctor. Based on previous studies, the required sample size was determined to be approximately 36 cases. Considering potential dropouts and exclusions due to protocol deviations during the trial, the target sample size was set to 40 cases.

Intervention

Participants were randomly assigned to receive either two RME or placebo capsules daily for a 4-week period. After this initial phase, a 4-week washout period was observed, during which the participants did not consume any study capsules. Subsequently, the participants were administered the other set of capsules for an additional four weeks. Test diets were obtained from Sunsho Pharmaceutical Co., Ltd. The RME capsules contained microcrystalline cellulose, caramel color, calcium stearate, silicon dioxide, and RME provided by NIPPN Corporation [rosmarinic acid 5 ‍mg, rosemary-derived diterpenes (carnosic acid and carnosol) 8 ‍mg/2 capsules]. In the placebo capsule, RME was replaced with microcrystalline cellulose (Table 1). Both diets were provided as capsules and could not be distinguished based on appearance or smell. The allocator strictly managed allocation details until the study director provided instructions for unblinding. To ensure blinding, this study was conducted by investigators, participants, and assessors who were blinded to the group allocation by keeping the test capsules indistinguishable from the placebo or RME.

Randomization and blinding

The participants were randomly allocated to the placebo or rosemary groups in a 1:1 ratio using computer-generated random numbers. Stratified randomization was employed to ensure an equal distribution of participants based on sex, state anxiety scores, and trait anxiety on the State-Trait Anxiety Inventory-Form JYZ (STAI). The allocation process was conducted and closely monitored by an individual who was not directly involved in the research until the principal investigator instructed the unblinding of the study.

Primary outcome

The primary outcome was assessed using the STAI, which was administered at screening, at the start of the intervention, and four weeks thereafter. Additionally, to evaluate the effect that single consumption has on work-related stress, the Uchida–Kraepelin test (UKT) was administered as a stressor on the first day of the intervention after consuming the test food, followed by an evaluation using the STAI.

Secondary outcome

The primary outcomes were assessed using the low-frequency/high-frequency (LF/HF) ratio, visual analog scale (VAS), and salivary cortisol secretion level measurements, all of which were performed at screening, at the start of the intervention, and four weeks thereafter. Furthermore, on the initial day of the intervention, measurements were taken before test food intake and again after conducting the UKT 30 ‍min post-consumption.

The LF/HF ratio was measured using the Fatigue Stress Measurement System VM302 (Fatigue Science Laboratory Inc., Osaka, Japan). Both fingers were placed in the device to simultaneously record photoplethysmography and electrocardiography, from which the LF/HF ratio was calculated.

The VAS was used to measure “Fatigue”, “Stress Levels”, “Relaxation”, “Mental Fatigue”, and “Concentration”. The level for each item was indicated by marking a vertical line on a 100-mm line segment, and the score determined by measuring the distance from the left end to the marked line. For “Fatigue”, the left end was set as “No fatigue at all”, and the right end as “Extreme fatigue”. For the items “Stress levels” and “Mental fatigue”, the scale was set with “Least ever felt” at the left end and “Most ever felt” at the right end. Conversely, for “Relaxation” and “Concentration”, the left end was labeled as “Most relaxed/concentrated”, and the right end as “Least relaxed/concentrated”.

Salivary cortisol levels were analyzed by outsourcing them to Yanaihara Institute Inc. (Shizuoka, Japan).

Safety evaluation

Safety was assessed through physical measurements, blood biochemical analyses, hematological assessments, and urinalysis performed at screening and again at four weeks after the second phase of the intervention. Physical measurements included height, weight, body mass index, systolic and diastolic blood pressure, and heart rate. Blood biochemical analyses included total protein, albumin, total cholesterol, low-density and high-density lipoprotein cholesterol, triglycerides, blood glucose, free fatty acids, phospholipids, blood urea nitrogen, creatinine, uric acid, total bilirubin, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, lactate dehydrogenase, gamma-glutamyl transferase, creatine phosphokinase, and glycated hemoglobin. Hematological assessments included white and red blood cell counts, hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin concentration, and platelet count. Urinalysis included specific gravity, pH, protein, glucose, urobilinogen, bilirubin, ketone bodies, and occult blood. Blood biochemical analyses, hematological assessments, and urinalysis were conducted by BML, Inc. (Tokyo, Japan). Adverse events were monitored throughout the study period, and their relationship with the study product assessed by the principal investigator.

Questionnaire survey on daily life and medication

Participants were instructed to complete a daily questionnaire from the day of the week 0 examination in Phase I to the day of the week 4 examination in Phase II. The questionnaire included items on test food intake, health status, medication use (including the names of any drugs taken), exercise, alcohol consumption and quantity, dietary supplements, food intake that may affect the research area, and lifestyle changes.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics 29 (IBM Corp., Armonk, NY). The significance level for statistical hypothesis testing was set at 5% (two-tailed test), and the confidence interval set at 95% (two-tailed). The primary analysis was conducted on the full analysis set, which included all participants allocated through stratified randomization, excluding those with no post-allocation data and those who did not consume the test food at least once. The primary and secondary outcomes at each time point were analyzed using a linear mixed-effects model, with order, timing, and food as fixed effects and subjects as variable effects. The baseline measurements were included as covariates. The main intervention comparisons were evaluated based on main effects of the groups using the least-squares mean at the 4-week post-intervention examination time point for each phase. Safety data were summarized descriptively, and the incidence of adverse events compared between groups using Fisher’s exact test.

Results

Study I: Overdose safety study

Participants

A total of 31 candidates were recruited and screened (SCR), and 22 eligible individuals (5 men and 17 women, aged 29–62 years) enrolled (Fig. 1). Among the 22 trial participants, one dropped out because of COVID-19 during the intake period. The mean age of the participants was 42.4 ± 9.9 years, with a relatively even distribution (Table 2). In terms of test diet adherence, all participants achieved a 100% consumption rate.

Fig. 1.

Study participant flow in the overdose safety study (Study I).

Table 2.

Participant background in the overdose safety evaluation

Parameter	Week 0	Week 4
Sex (number)	Male (5)/Women (16)
Age	42.9 ± 9.9
Height (cm)	161.1 ± 9.5	161.1 ± 9.5
Weight (kg)	57.4 ± 11.8	57.1 ± 11.8
BMI (kg/m2)	21.9 ± 3	21.8 ± 3
Systolic blood pressure (mmHg)	117.3 ± 12.6	118.9 ± 13.2
Diastolic blood pressure (mmHg)	77.0 ± 7.4	74.6 ± 6.8
Heart rate	74.2 ± 9.4	68.9 ± 7.2

Values are presented as mean ± SD. BMI, body mass index.

Safety assessment

Safety assessment parameters, including blood biochemical, hematological, and urinalysis tests, were evaluated at weeks 0 and 4 from the start of consumption. The results indicated that although some parameters showed fluctuations within normal laboratory values, no abnormal changes in the clinical laboratory values were observed, and no fluctuations suspected to be related to the test food (Supplemental Table 1 and 2*). Adverse events were observed in 15 cases and 20 events during the trial period. The principal investigator determined that none of the adverse events were related to the test diet.

Study II: Efficacy evaluation study

Participants

In total, 106 candidate participants were selected for the SCR, and 40 eligible individuals randomized (Fig. 2). However, during the blinded review after the trial, two participants were unable to attend visits during the trial period, leading to missing data for the efficacy assessment and a final analysis population of 38 individuals.

Fig. 2.

Study participant flow in the efficacy evaluation study (Study II).

Baseline characteristics of the study participants are shown in Table 3. The mean age of the participants was 44.3 ± 9.6 years, with a relatively even distribution of men (n = 20) and women (n = 18). In terms of test diet adherence, all participants achieved a 100% consumption rate, except for two who had a consumption rate of 93.1% and seven who had a rate of 96.6%.

Table 3.

Participant background in the efficacy evaluation

Parameter
Sex (number)	Men (20)/Women (18)
Age	44.3 ± 9.6
BMI (kg/m2)	21.78 ± 1.82
Systolic blood pressure (mmHg)	118.1 ± 10.4
Diastolic blood pressure (mmHg)	76.7 ± 6.8
BDI-II score	12.6 ± 4.9
Score of state anxiety on STAI	50.3 ± 5.6
Score of trait anxiety on STAI	52.9 ± 6.3

Values are presented as mean ± SD. BDI-II, Beck Depression Inventory-II; BMI, body mass index; STAI, State-Trait Anxiety Inventory.

State-Trait Anxiety Inventory

Analysis of the crossover study revealed no significant order or period effects (Table 4). After four weeks of continuous intake, the Trait Anxiety Scale of the STAI showed a significant reduction in the RME group compared with that in the placebo group. However, no significant differences were observed between the groups regarding state anxiety after four weeks of continuous intake (Table 5). On the first day of the intervention in each phase, following the consumption of the test food and imposition of work-related stress through the UKT, the STAI evaluation indicated that state anxiety scores in the RME group were significantly lower than those in the placebo group (Table 5). Conversely, no significant differences in trait anxiety were found between the groups after work-related stress.

Table 4.

Carryover effects in the efficacy evaluation study

Parameter	Order effects	Period effects
	p value	p value
State anxiety	0.396	0.308
Trait anxiety	0.927	0.855

Table 5.

Scores in the STAI after treatment

Parameter	Group	Continuous intake				Single intake
		Week 0 score	Week 4 score	p value		Before UKT score	After UKT score	p value
State anxiety	Placebo	44.0 ± 7.8	44.5 ± 6.5			44.0 ± 7.8	47.9 ± 6.5
	RME	44.5 ± 7.9	42.9 ± 7.3	0.153		44.5 ± 7.9	46.0 ± 7.8	0.041
Trate anxiety	Placebo	47.3 ± 7.4	48.2 ± 7.6			47.3 ± 7.4	47.6 ± 8.2
	RME	48.3 ± 8.9	46.0 ± 8.6	0.014		48.3 ± 8.9	47.5 ± 8.6	0.948

Scores at week 0 are presented as mean ± SD, whereas scores at week 4 are presented as least-squares mean ± SD. Bold values indicate statistically significant differences at p<0.05. STAI, State-Trait Anxiety Inventory; UKT, Uchida–Kraepelin test.

Secondary outcome

Four weeks of test food consumption resulted in a significant reduction in the LF/HF ratio in the REM group compared with that in the placebo group (Table 6). Conversely, no differences in the LF/HF ratio were noted between groups or in its change after work stress on the first day of the study.

Table 6.

Scores of the LF/HF ratio after treatment

Parameter	Group	Continuous Intake				Single Intake
		Week 0 score	Week 4 score	p value		Before UKT score	After UKT score	p value
LF/HF ratio	Placebo	4.05 ± 7.44	4.57 ± 5.81			4.05 ± 7.44	2.83 ± 3.61
	RME	4.35 ± 7.94	2.53 ± 1.83	0.019		4.35 ± 7.94	2.30 ± 2.36	0.392

Scores at week 0 are presented as mean ± SD, whereas scores at week 4 are presented as least-squares mean ± SD. Bold values indicate statistically significant differences at p<0.05. LF/HF rate, low-frequency/high-frequency rate.

No significant differences were observed in the VAS items between groups at both the post-work stress load assessment on the first day of test diet intake or the evaluation at the end of the 4-week test diet period (Table 7).

Table 7.

VAS scores of fatigue and stress after treatment

Parameter	Group	Continuous intake				Single intake
		Week 0 score	Week 4 score	p value		Before UKT score	After UKT score	p value
Fatigue	Placebo	52.8 ± 16.1	54.5 ± 17.3			52.8 ± 16.1	61.8 ± 17.1
	RME	53.5 ± 17.3	52.6 ± 17.3	0.516		53.5 ± 17.3	60.7 ± 16.8	0.681
Stress levels	Placebo	52.9 ± 17.0	53.3 ± 16.3			52.9 ± 17.0	57.6 ± 19.1
	RME	55.4 ± 16.7	50.0 ± 15.2	0.275		55.4 ± 16.7	57.2 ± 17.2	0.889
Relaxation	Placebo	49.1 ± 15.5	48.6 ± 16.8			49.1 ± 15.5	50.1 ± 18.5
	RME	51.7 ± 14.7	49.5 ± 13.8	0.721		51.7 ± 14.7	51.2 ± 17.9	0.708
Mental fatigue	Placebo	53.1 ± 19.0	55.2 ± 18.6			53.1 ± 19.0	63.5 ± 17.5
	RME	54.6 ± 19.1	50.4 ± 17.4	0.171		54.6 ± 19.1	62.8 ± 15.1	0.783
Concentration	Placebo	50.4 ± 15.8	52.4 ± 14.4			50.4 ± 15.8	56.2 ± 15.6
	RME	51.0 ± 16.4	48.9 ± 14.8	0.221		51.0 ± 16.4	55.4 ± 18.9	0.814

Scores at week 0 are presented as mean ± SD, whereas scores at week 4 are presented as least-squares mean ± SD. UKT, Uchida–Kraepelin test; VAS, visual analog scale.

The secretion levels of salivary cortisol showed no significant differences between the RME and placebo groups following work stress load on the first day of test diet intake and after four weeks of test diet consumption (Table 8).

Table 8.

Scores of salivary cortisol levels after treatment

Parameter	Group	Continuous intake				Single Intake
		Week 0 score	Week 4 score	p value		Before UKT score	After UKT score	p value
Salivary cortisol	Placebo	0.186 ± 0.075	0.192 ± 0.097			0.186 ± 0.075	0.139 ± 0.084
	RME	0.183 ± 0.129	0.180 ± 0.091	0.520		0.183 ± 0.129	0.137 ± 0.078	0.841

Scores at week 0 are presented as mean ± SD, whereas scores at week 4 are presented as least-squares mean ± SD. UKT, Uchida–Kraepelin test.

Safety assessment

Safety assessment items, including vital signs, physical measurements, blood biochemical tests, hematological tests, and urinalysis, were evaluated at SCR and at weeks 0 and 4 of each phase. In this study, fluctuations in the clinical laboratory values were within the normal range, and no abnormal fluctuations detected (Supplemental Table 3 and 4*). During the trial period, adverse events were observed in 13 cases and 23 events in the RME group and in 14 cases and 31 events in the placebo group (Table 9). The principal investigator determined that these events were unrelated to the test diet and that no significant differences in the incidence of adverse events were evident.

Table 9.

Observed adverse events in the efficacy evaluation study

	Cases	Incidence rate	p value
Placebo	14	0.18
RME	13	0.16	0.999

Discussion

The safety trial was conducted at a dose five times higher than that used in the efficacy evaluation. While some parameters showed significant changes at week 4 compared with those at baseline, these were minor and remained within physiological ranges, suggesting that they were not clinically relevant. We recorded 15 cases and 20 adverse events related to menstrual factors or temporary reactions to the COVID-19 vaccination that resolved quickly. Therefore, the trial was conducted according to the principal physician’s judgment. No adverse events were attributed to the test compounds. These findings confirm the safety of excessive RME intake over a 4-week period.

This study investigated the effects of RME containing rosmarinic acid and diterpenes (carnosic acid and carnosol) on daily and work-related stress in healthy Japanese adults and evaluated its safety in cases of excessive intake. The results demonstrated that RME intake significantly improved subjective evaluation measures, such as the STAI, as well as physiological indicators, such as the LF/HF ratio, compared with those of the control group. Safety evaluations indicated no issues with the extract, even in cases of excessive intake. These findings suggest that RME intake is safe and exhibits anti-stress effects through both single and continuous intake methods.

Among the components known for their stress-reducing effects in rosemary are rosmarinic acid, carnosic acid, and carnosol, along with 1,8-cineole and luteolin.^(8,16–19) The RME used in the present study did not contain 1,8-cineole or luteolin. Reports indicate that approximately 90% of the diterpenes in rosemary are composed of carnosic acid and carnosol.⁽²⁰⁾ Thus, the anti-stress effects observed in the present study were likely caused primarily by rosmarinic acid and diterpenes (carnosic acid and carnosol). No carryover effects were observed. Given that the half-lives of rosmarinic acid and diterpenes are less than one day, the 4-week washout period implemented in this study was considered sufficient.

The efficacy evaluation study used the STAI as a primary endpoint. Use of the STAI as a measure of mental stress is supported by its established reliability and validity in various populations. This inventory is widely used in clinical and research settings to assess anxiety levels in response to stress-induced situations. Its sensitivity to changes in anxiety over time makes it a suitable tool for evaluating the effectiveness of stress-reducing interventions. Consequently, the STAI can serve as a robust indicator of mental stress, allowing for a better understanding of the psychological impact of stressors in clinical studies. The findings of this study support those of previous research indicating that RME, which is rich in rosmarinic acid and diterpenes, possesses antidepressant and anxiolytic properties, as demonstrated in animal models.⁽²¹⁾ Our results suggest that these benefits can also be extended to humans. Specifically, a subjective assessment using the STAI revealed that trait anxiety scores, which are indicative of chronic stress, improved with ongoing RME consumption. Additionally, a single dose of the extract reduced state anxiety scores, reflecting acute stress. However, no significant differences were observed in state anxiety after continuous intake or in trait anxiety following workload during a single intake. Considering that state anxiety after continuous intake showed numerical trends toward stress reduction and assessed “what is currently being felt”, conducting a work stress test at the end of continuous intake could likely have demonstrated an effect. In contrast, the lack of efficacy observed in trait anxiety after workload stress during single intake may be attributed to the nature of trait anxiety, which evaluates “feelings typically experienced in daily life”.⁽²²⁾ Based on these results, the present study revealed that RME comprising rosmarinic acid and diterpenes has beneficial effects on both daily stress through continuous consumption and work-related stress through single consumption.

The results of this trial demonstrated that RME consumption decreased the LF/HF ratio, suggesting a reduction in stress. This alteration in the LF/HF ratio may be linked to changes in the secretion of monoamine-derived hormones. RME acts on the oxytocin pathway and enhances the secretion of monoamine hormones, such as dopamine, adrenaline, noradrenaline, and serotonin, which are associated with anxiolytic and antidepressant effects.⁽²²⁾ Diterpenes and rosmarinic acid in rosemary have also been reported to exert antidepressant-like effects by modulating sympathetic and parasympathetic nervous system activity.^(16,23) Additionally, monoamine-derived hormones play a role in regulating autonomic nervous system balance.^(24–26) Therefore, anti-stress effects observed with RME consumption can be regarded as one of the underlying mechanisms of action, likely resulting from the stimulation of monoamine hormone secretion, which affects both the sympathetic and parasympathetic nervous systems to promote autonomic balance.

The stress-relieving effects of RME may also be associated with its anti-inflammatory properties. Stress increases inflammatory cytokine levels in the brain, leading to decreased levels of brain-derived neurotrophic factor (BDNF). This reduction results in the inhibition of neurogenesis and dendritic atrophy, which can trigger depression-like symptoms.^(27–29) Conversely, both rosmarinic acid and diterpenes have been confirmed to possess strong anti-inflammatory effects, and intake of RME containing these compounds has been reported to increase BDNF levels in the brain. Additionally, both rosmarinic and carnosic acids have been shown to promote dendritic growth in neurons.^(7,8,30) Therefore, the antioxidant properties of rosmarinic acid and diterpenes in RME are hypothesized to contribute to an increase in BDNF levels, facilitating dendritic growth and thereby exerting stress-relieving effects. However, as inflammatory markers were not measured in the present study, future research should consider evaluating the levels of inflammatory markers and BDNF in the blood.

This study assessed the stress-relieving effects of both continuous and single intake of RME. The UKT employed in this study is recognized as a measure of stress from occupational fatigue and continuous task performance, such as work or study.^(31–33) The significant decrease in STAI state anxiety scores observed following a single intake of RME indicates its potential to reduce mental stress related to work and study. Although the long-term anti-stress effects of rosemary essential oil and its components have been previously evaluated, reports on the anti-stress effects of a single intake are limited.^(34,35) This study revealed that a single dose of RME containing rosmarinic acid and diterpenes can effectively reduce temporary stress experienced during learning and work.

While significant stress-relieving effects of RME were observed at the primary endpoints of the STAI and in heart rate variability, no significant differences were found in the VAS or salivary cortisol levels. While the VAS did not show significant differences, the observed numerical trends in both chronic and acute stress suggest the positive impact that RME had on stress reduction. Similarly, the salivary cortisol secretion levels showed numerical changes in the direction of stress improvement. Salivary cortisol levels exhibit diurnal variations that peak upon awakening.⁽³⁶⁾ In the evaluation of a single intake, the time elapsed between the pre- and post-assessments may have been influenced by diurnal variations, making it challenging to confirm the effects of RME. Additionally, for long-term intake, cortisol secretion is influenced by sex hormone secretion, suggesting that factors such as hormonal balance may obscure the stress-relieving effects of RME.⁽³⁷⁾ Future research should focus on selecting biomarkers with minimal diurnal variations that influence hormonal balance, such as inflammatory markers, when measuring stress-related biomarkers.

Previous studies have highlighted the anti-stress properties of rosmarinic acid. The present study explored the anti-stress effects of RME containing both rosmarinic acid and diterpenes and considered the effects of a single administration. The study results indicate that the actions of these beneficial components through the autonomic nervous system and their influence on brain neurons can lead to significant improvements in subjective stress evaluations using the STAI, effectively addressing both everyday stress and stress from work or study. These findings suggest that the intake of RME may contribute to the reduction of daily stress and stress associated with work and study, potentially serving as a preventive measure against stress-related diseases.

Conclusion

In this study, we investigated the effect that RME containing rosmarinic acid and diterpenes (carnosic acid and carnosol) has on stress through a randomized, placebo-controlled, double-blind crossover trial involving healthy adult men and women, and also conducted an open-label trial to evaluate the safety of excessive RME intake. Overall, the study findings verify the safety of RME. Moreover, continuous consumption of the extract improved STAI scores and heart rate variability, and single consumption showed benefits for the STAI after work-related stress, suggesting that RME effectively alleviated stress.

Author Contributions

HK, KO, KS, SF, and KA designed the study and interpreted the data. HK and KO were involved in data analysis. HK and KO wrote the manuscript. All authors critically revised the report, commented on drafts of the manuscript, and approved the final version.

Ethics Approval and Consent to Participate

Study I and Study II were performed in accordance with the Declaration of Helsinki (2013) and the Ethical Guidelines for Medical and Health Research Involving Human Subjects (2015) [Ministry of Education, Culture, Sports, Science and Technology (MEXT), and the Ministry of Health, Labour and Welfare (MHLW), Japan]. The protocol of Study I was approved by the Ethics Committee of NIPPN Corporation (Tokyo, Japan; permission number: 21-03) and the Sapporo Yurinokai Hospital Ethics Committee (Hokkaido, Japan; permission number: 021). The protocol of Study II was approved by the Ethics Committee of NIPPN Corporation (permission number: 24-01) and the Tokyo Shinjuku Clinic Ethics Committee (Tokyo, Japan; permission number: RD01001AS04). Informed consent was obtained from all participants in Study I and Study II.

Abbreviations

BDNF

brain-derived neurotrophic factor

COVID-19

coronavirus disease 2019

LF/HF

low-frequency/high-frequency

RME

rosemary extract

SCR

recruitment and screening

STAI

State-Trait Anxiety Inventory-Form JYZ

UKT

Uchida–Kraepelin test

VAS

visual analog scale

Conflict of Interest

All authors are employed by NIPPN Corporation. Clinical experiments were conducted by Clinical Creative Co., Ltd., Sapporo Yurinokai Hospital, HUMA R&D Co., Ltd., and Tokyo Shinjuku Clinic, with funds provided by NIPPN Corporation. The authors have no conflicts of interest to declare.