Assessment of the efficacy of plant-derived essential oils against Mycobacterium tuberculosis: A systematic review and meta-analysis

Abstract

Mycobacterium tuberculosis (MTB), the causative agent of tuberculosis (TB), remains a major global health challenge, exacerbated by the rise of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains. Limitations in current antibiotic therapies have intensified interest in alternative antimicrobials, including plant-derived phytochemicals. Essential oils (EOs), with their complex chemical composition and long-standing traditional use, represent promising candidates for novel anti-TB agents. This study conducted a systematic review and meta-analysis to evaluate the in vitro antimycobacterial activity of plant-derived EOs. Following PRISMA guidelines, four major databases, Scopus, MEDLINE Central/PubMed, Embase, and Web of Science, were searched through May 31, 2025, and 31 eligible studies were included after screening and quality assessment. Descriptive and comparative analyses of minimum inhibitory concentrations (MICs) were performed using R software. The results revealed substantial variability in EO efficacy across plant species and geographic origins. The most potent activity was observed in Euclea sp. and Croton sp., which showed exceptionally low MICs of 1 μg/mL and 4.88 μg/mL, respectively. In contrast, EOs from Dichrostachys cinerea, Dorstenia elliptica, Imperata cylindrica, Mondia whitei, Pentadiplandra brazzeana, and Tetrapleura tetraptera exhibited weaker effects, with MICs up to 2048 μg/mL. Plant anatomical sources also influenced activity, with leaves and stems showing higher efficacy than roots and fruits. Overall, these findings highlight the therapeutic potential of specific EOs as adjunct or alternative treatments for DR-TB. Further studies involving compound standardization, and in vivo validation are necessary to support their development into clinically applicable anti-TB agents.

Graphical abstract

Highlights

• •A global systematic review and meta-analysis of 31 studies on EOs against MTB, including MDR/XDR strains.
• •Euclea sp. and Croton sp. showed the strongest anti-MTB activity with very low MIC values.
• •Piper was the most studied genus, with consistent anti-MTB activity.
• •Asian plant EOs had the lowest average MICs, indicating high potency.
• •Roots and fruits exhibited more potent antimicrobial activity than leaves.

1. Introduction

Mycobacterium tuberculosis (MTB) is the primary cause of tuberculosis (TB), a dangerous respiratory infection that has afflicted millions of people throughout history; also, we can call it one of human society's murderers for the high rate of death caused by established infection (single-cause death) [[1], [2], [3]]. After many years of struggle with MTB, scientists have discovered various chemical compounds to address it. They were applicable and valuable, but before the emergence of drug resistance (DR) [4,5]. In this phenomenon, bacteria can adapt to environmental conditions when exposed to antibiotics. Molecular investigations reveal changes in the MTB genome [6] that enable it to evade drugs by remodeling its cellular structures and molecular elements, explicitly targeting the sites of action of prescribed antibiotics, such as the RNA polymerase subunit beta (rpoB), thereby evading rifampicin (RIF). The rate of drug resistance in MTB has risen in recent decades. Resistance to the first-line treatment [multidrug-resistant (MDR)] and the second-line treatment [extensively drug-resistant (XDR)] has complicated treatment. Statistics show about 10.8 million affliction, 1.25 million deaths, and 400,000 cases with MDR worldwide for 2023 [7,8]. Current therapeutic options are increasingly limited due to the rise of DR-MTB strains, underscoring the urgent need for novel antimicrobial compounds. Emerging fields such as nanotechnology have contributed innovative approaches to antibiotic design; however, MTB possesses sophisticated adaptive mechanisms that enable it to circumvent the cytotoxic effects of various nanoparticles [9]. An Iranian proverb states, “The ancients knew something when they did such and such a thing,” which serves as a testament to the ingenuity and practical knowledge of past civilizations in solving problems. In the meantime, we can look back and ask: how did our ancestors fight infections with their basic equipment?

One of our ancestral legacies is nature. In past and primary civilizations, humans utilized plants in their daily lives. They used plants in different ways to treat their pain. Additionally, they found that plants are an effective material for alleviating respiratory disorders [10]. These effects are attributed to secreted substances by various parts of a plant (e.g., leaf, root, flower, etc.). Different compounds are synthesized and secreted by plants in nature. Aromatic plants constitute a significant portion of the flora and can produce essential oils (EOs) with diverse practical and industrial applications (e.g., the pharmaceutical and cosmetic industries) [11,12]. These compounds, with various chemical groups [20–60 different components], such as esters, alcohols, terpenes, phenolic compounds, and others, exhibit biological activity against bacteria and can kill them by disrupting their cellular structures and components [13,14]. EOs, as secondary metabolites of plants, are part of the plant's defense against pests and infectious plant diseases. The biosynthesis of essential oil constituents is primarily mediated through specialized (secondary) metabolic pathways, which function independently of the core essential metabolic processes required for cell growth and survival. The biodiversity of plants and their intricate taxonomy present a wide variety of products [15]. Experiments showed that EOs extracted by different distillation methods (e.g., steam and hydro) from plants inhibited various bacterial strains [16]. These plant-derived compounds also demonstrated measurable inhibitory activity against MTB, indicating their potential as candidates for developing novel therapeutic strategies targeting DR-TB [17].

Investigations have shown that EOs have beneficial effects on MTB, including inhibiting its growth and activity. They were considered to have the potential to cause infectious diseases, particularly respiratory infections, and were treated in traditional medicine with anti-decongestant effects [18]. On the other hand, natural compounds often have a greater appeal to customers than synthetic and chemical products. Therefore, there is a better market for pharmaceutical companies [19]. Accordingly, this systematic review and meta-analysis were undertaken to critically evaluate the antimycobacterial efficacy of plant-derived EOs based on published global research. Emphasis was placed on standardized biological metrics, particularly minimum inhibitory concentration (MIC), to enable meaningful cross-study comparisons. By normalizing and synthesizing the available data, this study aimed to reduce heterogeneity and provide a robust evidence base regarding the therapeutic potential of EOs against MTB. The results not only highlight the promising inhibitory activity of several EO-yielding plant species but also emphasize the importance of integrating natural product research into contemporary anti-TB drug discovery pipelines. In an era marked by the growing threat of MDR and XDR-TB, these findings support the reconsideration of phytotherapeutic agents as viable candidates for adjunctive or standalone therapy. Ultimately, this work contributes to a more nuanced understanding of the potential role of botanical compounds in addressing persistent global health challenges posed by TB.

2. Methods

The authors performed a systematic review and meta-analysis following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [20]. Our research protocol was registered in the Prospective Register of Systematic Reviews (PROSPERO, CRD420251243199; available at https://www.crd.york.ac.uk/PROSPERO/view/CRD420251243199). Ethical approval was not sought for this study, which includes an analysis of secondary data.

2.1. Study design and search strategies

A systematic literature search was conducted by two investigators (SMAMS and MM) to identify studies published by May 31, 2025. Relevant studies were retrieved from MEDLINE Central/PubMed, Scopus, Web of Science, and Embase using keywords including “Mycobacterium tuberculosis,” “essential oil∗," “treatment,” “antimycobacterial,” and their combinations using the (AND) and (OR) operators without language or geographical restrictions. Reference lists of included studies were manually reviewed to ensure comprehensive coverage. Appendix File No. 1. manifested our search strategies.

2.2. Study selection

The information of all searched articles was transferred into the EndNote reference management software (Version X7, Thomson Reuters, New York, NY, United States), and duplicate entries were removed. Next, all identified articles were screened by title and abstract using the inclusion/exclusion criteria, and the screening was conducted independently by four investigators (HH, MGH, MF, and AHH) to minimize bias. Disagreements were resolved through discussion and consensus, or by consulting a fourth reviewer (SMAMS and MM), who resolved them based on the study's judgments. The reference lists within the identified studies were manually searched (using backward and forward citation searches) to ensure a comprehensive understanding of the collected articles (SFS and MRR). Studies that satisfied the inclusion criteria were selected for final analysis.

2.3. Eligibility criteria

The following criteria were used to keep an identified study in the systematic review and for meta-analysis [1]: In vitro studies that reported MIC values of EOs against MTB [2], Studies with their full-text available [3], Studies that provided sufficient data (EO name, MIC value, MTB), and [4] Studies written and published in English. On the contrary, Exclusion criteria included [1]: Studies reporting insufficient quantitative data [2], All types of studies except original (reviews [narrative, systematic, and meta-analyses], case reports, case series, conference abstracts, editorials, book chapters, or letters, and [3] duplicate datasets.

2.4. Data extraction and quality assessment

Data were extracted, independently and cross-checked for accuracy, into a Microsoft Excel 2016 spreadsheet (Microsoft, Redmond, Washington, USA) from eligible studies by four investigators (HH, MGH, MF, and AHH) using a predesigned form, which had been pretested and adjusted before use by two investigators (SMAMS and MM). The data were extracted, including the paper title, the first author's name, year of publication, plant name, plant part (leaf, root, etc.), MIC values, etc.

2.5. Data analysis

All statistical analyses were performed using R (version 4.5.1, “Great Square Root,” released June 3, 2025) in conjunction with RStudio (latest version as of January 2025). Data subgrouping was initially performed in Microsoft Excel, followed by a comprehensive analysis using R packages developed explicitly for meta-analysis, including the meta and metafor packages [21,22]. Visualization of results, including forest plots, bar charts, and pie charts, was performed using the ggplot2 package and its supporting libraries in R.

For the assessment of the amount of effectiveness, the means of all exported results (MIC (μg/mL)) were considered for each species. Additional statistical values [homogeneity and heterogeneity], such as median, min, max, and standard deviation (SD), were calculated for better insight into the results. Due to differences in the types and values of the controls across the included articles, we are unable to calculate the ratios and means or to illustrate the forest plot. On the other hand, each study investigated different types of plants' EOs, so we cannot launch risk of bias assessment tests (e.g., Egger's or Begg's test) for included articles bias assessment; further, for assessing bias risk based on each EO, heterogeneity values are not meaningful due to: 1. Present insufficient data by articles, 2. Lack of sufficient studies for assessment and comparison, 3. The existence of the same values across different samples, or launching a single sample that has led to heterogeneity factors (e.g., standard error and standard deviation) becoming incalculable.

3. Results

3.1. Search results and selection process

Of 2055 papers extracted from databases, 634 were duplicates; after removing them, 1421 papers were screened. Non-full-text articles are excluded in the next step. Review articles, book chapters, letters, editorials, conference papers, and non-English papers were excluded from the systematic process. Finally, 31 articles were included in the present study; a list of them is available in Appendix File No. 2. More details are observable in Fig. 1.

Fig. 1.

Flowchart of systematic review. Demonstrates process from identification to inclusion.

3.2. Efficacy of different taxa

After calculating the mean for each plant genus, Euclea and Croton are the two most effective genera, with MIC values of 1 μg/mL and 4.88 μg/mL, respectively. In contrast, Dichrostachys, Dorstenia, Imperata, Mondia, Pentadiplandra, and Tetrapleura show the highest MIC at 2048 μg/mL. Then, Solanum and Cymbopogon ranked as the second- and third-highest MIC genera, with MICs of 1536 μg/mL and 1250 μg/mL, respectively. More details about all examined taxa are visible in Fig. 2. Different efficacy rates are observed across genera, but significant differences were first observed in Trifolium at 150 μg/mL. The majority of the presented data concern MDR and XDR MTB strains that exhibit resistance to RIF and isoniazid (INH), with MICs of <10 μg/mL.

Fig. 2.

Efficacy of different plants' EOs on MTB inhibition (μg/mL). Euclea is known as the most effective genus, and six genera (Dichrostachys, Dorstenia, Imperata, Mondia, Pentadiplandra, and Tetrapleura) with more than 2048 μg/mL MIC have less potency in inhibition of MTB.

Based on the meta-analysis, the efficacy of different genera was assessed using MIC values; in Table 1, we describe the most effective genera with the lowest mean MIC values, as reported by different studies. Table 2 presents statistical values for the efficacy of various herbal genera in inhibiting MTB. Most of the genus has one species for its efficacy. However, some others have more data about their different species. For example, Cinnamomum is one of them, with three examined species, and its mean MIC is 208.29 μg/mL, indicating efficacy. Lantana and Lippia are two other genera with more than one studied species. Lantana at 90 μg/mL is more effective than Lippia at 657.94 μg/mL. Piper is another genus with many studied species (17 spp.) that have shown an efficacy value of 183.70 μg/mL. Xylopia is another effective genus with two species that has a MIC of 768 μg/mL.

Table 1.

Seven most potent genera in MTB inhibition based on MIC value. Euclea as most effective genus ranked first.

Genus	MIC (μg/mL)	Rank
Euclea	1	1st
Croton	4.88	2nd
Cuminum	12.5	3rd
Spondias	13.18	4th
Salvia	16.5	5th
Brownea	17.57	6th
Trachyspermum	19.5	7th

Table 2.

Statistical details of efficacy of EOs on MTB inhibition (μg/mL). Mean, max, min, median and SD have been calculated to give insight about efficacy rates, no. of studies also has been mentioned to check accuracy and fault possibility.

Genus	Species	Mean	Max	Min	Median	SDa	No. of studies
Afrostyrax	lepidophyllus mildbr	512	–	–	–	–	1
Allium	sativum (L.)	31	–	–	–	–	1
Arracacia	tolucensis var. multifida	73.6	–	–	–	–	1
Brownea	b	17.57	–	–	–	–	1
Cinnamomum	zeylanicum	208.29	768	12.5	81.62	293.43	6
	glanduliferum
	verum
Croton	cajucara Benth	4.88	–	–	–	–	1
Cuminum	cymi num	12.5	–	–	–	–	1
Cymbopogon	citratus	1250	–	–	–	–	1
Daucus	littoralis Sibth.	198	–	–	–	–	1
Dichrostachys	cinerea (L.)	2048	–	–	–	–	1
Dorstenia	psilirus wellis	2048	–	–	–	–	1
Echinops	giganteus	24	–	–	–	–	1
Elionurus	tristis	32	–	–	–	–	1
Eucalyptus	citriodora	53.2	–	–	–	–	1
Eugenia	caryophyllata	25	–	–	–	–	1
Fagara	macrophylla Engl	1024	1024	1024	1024	0	2
	xanthoxyloides
	lam
Imperata	cylindrica Beauv.	2048	–	–	–	–	1
Juniperus	communis L.	60.33	–	–	–	–	1
Lantana	fucata	90	100	80	90	14.14	2
	trifolia
Laurus	nobilis	100	–	–	–	–	1
Lippia	sidoides Cham alba (Mill) NE Brown	657.94	1000	82.26	649.75	556	3
	americana
	scaberrima Sond.
Mondia	whitei (Hook F) Skeelz	2048	–	–	–	–	1
Monodora	myristica	768	–	–	–	–	1
Morus	alba	10.85					1
Pelargonium	graveolens	78	–	–	–	–	1
Pentadiplandra	brazzeana Baillon	2048	–	–	–	–	1
Pimpinella	anisum	100	–	–	–	–	1
Piper	aduncum	183.70	512	14.06	179.15	112.74	7
	guineense
	regnellii
	capense L.F
	Ihotzkyanum
	mikanianum
	mosenii
	multinodum
	sarmentosum
	xylosteoides
	cernuum
	diospyrifolium
	arboretum
	aduncum
	gaudichaudianum
	mosenii
	rivinoides
Plectranthus	amboinicus (Lour.) Spreng	351.6	–	–	–	–	1
Salvia	fruticose	16.50	79.03	0.1	1.6	34.95	2
	officinalis
	tomentosa
	cilicica
	aratocensis
Satureja	khuzestanica	117	156	78	117	55.15	1
	rechingeri
Scorodophleus	zenkeri Harms	768	–	–	–	–	1
Solanum	melongera Hierm	1536	–	–	–	–	1
Spondias	pinnata	13.18	–	–	–	–	1
Tetrapleura	tetraptera Tamb	2048	–	–	–	–	1
Thymus	vulgaris L	1024	–	–	–	–	1
Trachyspermum	copticum	19.5	–	–	–	–	1
Tridax	procumbens	600	–	–	–	–	1
Turnera	diffusa	88.73	–	–	–	–	1
Xylopia	parviflora (A. Rich.) Benth	768	1024	512	768	362.08	1
	aethiopica Dunal
Zataria	multiflora	78	–	–	–	–	1
Euclea	natalensis	1	–	–	–	–	1
Entada	elephantina	75	–	–	–	–	1
Plantago	lanceolate	125	–	–	–	–	1
Trifolium	burchellianum	150	–	–	–	–	1
Tridax	procumbens	600	–	–	–	–	1
Zanthoxylum	leprieurii	32	–	–	–	–	1
Sphaeranthus	indicus	125	–	–	–	–	1

^aSD has been calculated for values between species and is not meaningful based on studies.

^bThe reference mentioned Brownea sp. and does not refer to a special species.

3.3. Parts of plants

Different parts of plants have been studied in various experiments. Accessibility, ease of extraction, and the amount of EOs in various parts differ. In total, leaves and other aerial parts are the most suitable and popular parts for EOs extraction, followed by barks and stems. Nevertheless, leaves, stems, and bark are the most effective parts based on the calculated mean MIC. In this section, we explore the most and least effective herbal species for MTB inhibition. Additionally, the usage rates of different parts were calculated and presented in Fig. 3-A.

Fig. 3.

A: Number of species used based on part of them. (Seed 6, Fruit 10, Root 10, Stem and bark 10, Leaves and other aerial parts 38). B: Efficacy of different parts of plants based on inverse MIC.

3.3.1. Efficacy

Fruit and root of plants demonstrate high MICs for their secreted EOs of 873.68 μg/mL and 835.25 μg/mL, respectively. Meanwhile, the MICs of Tetrapleura tetraptera and Dichrostachys cinerea are 2048 μg/mL. Fruits of Spondias pinnata show high efficacy with the lowest MIC value. On the other hand, leaves, stems, and bark show high potency against MTB, with the lowest calculated mean MIC. Anti-TB activity of the Trachyspermum copticum, related to their seeds, with an MIC of 48.75 μg/mL. In seed, the average MIC is 605.45 μg/mL, with a median of 640 μg/mL. Leaves and aerial parts with the highest population in the studied species (38 species) have an MIC of 166.51 μg/mL. Furthermore, stem and bark have been reported to contain 177.53 μg/mL in the second situation, based on published studies. Table 3 and Fig. 3-B provide details on the efficacy of different parts of plants.

Table 3.

Details of efficacy based on part of plants (μg/mL). Efficacy was calculated based on mean of included data analysis, and effectiveness of parts mentioned for major and minor observed MIC based on each specie.

Part	Mean	Median	Major MIC (Less effective)	SD	Minor MIC (Most effective)	No. of species
Fruit	873.68	512	Tetrapleura tetraptera Tamb (2048) Dichrostachys cinerea (2048)	761.03	Spondias pinnata (7.81)	10
Leaves and other aerial parts	166.51	80	Lippia alba (Mill) NE Brown (1250)	266.35	Salvia fruticosa (0.1)	38
Root	835.25	55	Dorstenia psilirus Wellis	1043.92	Salvia cilicica (0.2)	10
			Imperata cylindrica Beauv.
			Mondia whitei (Hook F) Skeelz
			Pentadiplandra brazzeana Baillon
			(2048)
Seed	605.45	640	Fagara macrophylla Engl	404.36	Trachyspermum copticum (48.75)	6
			Fagara xanthoxyloides Lam
			(1024)
Stem and bark	177.53	78.7	Scorodophleus zenkeri Harms (1024)	281.51	Cinnamon [8]	10

3.4. Geographical distribution

Published studies on our topic are predominantly from Africa and America, followed by Asia. The ancient use of plants as herbal medicine for treating infections was commonplace in these regions. The distribution of plants near seas and oceans is more diverse than in other regions, especially in the southern hemisphere.

3.4.1. Continent-based efficacy

The most efficient plants are grown in Asia, with a MIC of 90.41 μg/mL. However, most species diversity is observed in America, with 33 species. America is the third continent with an effective fauna for fighting TB, exhibiting an MIC of 159.13 μg/mL against MTB. Plants grown in Africa show an MIC of 696.16 μg/mL against anti-MTB in vitro. Studies in Europe are so limited. Just Polish and British scientists reported anti-TB activity of EOs obtained from plants. They examined three plant species to determine an average MIC of 130.72 μg/mL for their research and ranked second among continents for plant efficacy. Table 4 and Fig. 4 provide detailed information on continents.

Table 4.

Details of statistical values for MIC based on the continents where the studies occurred (μg/mL). Asia has most effective plants genera, then Europe and America, and finally Africa.

Continent	Mean (μg/mL)	Median	SD	No of sp.
Africa	696.16	512	750.47	30
America	159.13	100	237.82	33
Europe	130.72	128.16	118.93	3
Asia	90.41	50	149.44	14

Fig. 4.

Average amount (μg/mL) of anti-TB activity of plants based on the place of studies in a bar chart. Africa has the highest MIC, followed by America and Europe. Asia has the most efficient EOs against MTB.

3.4.2. Country-based efficacy

Cameroon, Egypt, South Africa, Uganda, Brazil, Colombia, Mexico, the United States, India, Iran, Turkey, Poland, and the United Kingdom are countries with related studies. The obtained results demonstrated that the average MICs of EOs calculated across countries are 1140.4, 17.65, 198.27, 32, 201.29, 87.99, 54, 96, 173, 95.55, 2.7, 132, and 60.33 μg/mL. Fig. 5 shows a world map of MIC values for EOs in each country. Most published studies have been conducted in Africa and South America, and some in West Asia (India and Iran). Mexico and Brazil in South America have the most-studied plant faunas.

Fig. 5.

Efficacy of plants based on country. Cameroon has the least efficient plants with 1140 μg/mL, and Turkey has the most efficient plants with 2.7 μg/mL.

3.4.3. Plant distribution per continent

Plants which have been assested for anti-TB activity, as reported in published studies, exhibit high diversity in America and Africa, with 33 and 30 distinct species, respectively. In Asia, 14 species have been reported as anti-TB agents based on in vitro studies on their EOs. Fig. 6 shows all taxa present across different continents. The green nature of South Asia, alongside Africa and South America, which are located near the equator, and the high prevalence of TB in the Middle East and North Africa have led to increased research on related subjects in these regions.

Fig. 6.

Taxonomic diversity in continents. America [especially South America] has the most diverse nature, followed by Africa and Asia [especially West and South], which have diversity in efficient plants against MTB.

4. Discussion

The global burden of MTB, especially in its MDR and XDR species, presents an escalating challenge to public health. Conventional treatment regimens based on RIF and INH are increasingly compromised by chromosomal mutations in genes that confer antimicrobial resistance, undermining treatment outcomes [1,4,5]. According to WHO estimates, over 400,000 cases of MDR-TB occurred worldwide in 2023, highlighting the pressing need for alternative or adjunctive therapeutics. In this context, plant-derived EOs offer a promising route for drug discovery due to their wide chemical diversity and historical application in respiratory and infectious diseases. This systematic review and meta-analysis compiled 31 studies retrieved from MEDLINE Central/PubMed, Scopus, Embase, and Web of Science, and filtered according to eligibility criteria. The dataset spans several continents, multiple plant taxa, and both drug-sensitive and drug-resistant MTB strains, providing a globally representative overview of EO activity against MTB.

Dichrostachys, Dorstenia, Imperata, Mondia, Pentadiplandra, and Tetrapleura have been reported to have MIC values as high as 2048 μg/mL. These African-origin species are well-established in regional ethnobotanical systems and have demonstrated weak inhibitory effects in vitro [[23], [24], [25], [26], [27], [28]]. For instance, Dichrostachys cinerea and Tetrapleura tetraptera, both of which have been shown in studies from sub-Saharan Africa, are among the EOs that have shown weak activity, especially against MDR-TB strains, which in the present study were also reported among the weakest EOs.

The genus Piper, although showing a more moderate average MIC (183.7 μg/mL), was the most taxonomically represented, with 17 species studied. This includes Piper aduncum [29], Piper regnellii [30], and Piper guineense [31], all of which are rich in monoterpenes and phenylpropanoids such as safrole, 1,8-cineole, and eugenol, compounds known to disrupt microbial cell walls and impair cellular respiration. The consistent efficacy observed across diverse Piper species emphasizes its pharmacological value as a lead genus. Similarly, Cinnamomum spp., including Cinnamomum zeylanicum, Cinnamomum glanduliferum, and Cinnamomum verum [[32], [33], [34]], demonstrated moderate antimycobacterial activity, with a pooled average MIC of 208.29 μg/mL. Cinnamaldehyde, a primary bioactive component, has been demonstrated to interfere with ATP synthesis and mycolic acid metabolism in MTB [35]. Moreover, the EO of C. zeylanicum has demonstrated synergy with conventional antibiotics, such as INH, supporting its potential for use in combination therapy [33]. Other taxa of interest include Lippia sp., Salvia sp., and Zanthoxylum leprieurii, Lippia sidoides, and Lippia scaberrima, commonly found in South America and Africa, were associated with weak MICs (∼658 μg/mL) but are noted for their safety and availability. Salvia fruticosa, Salvia tomentosa, and Salvia officinalis, reported from Turkey, showed MIC values as low as 16.5 μg/mL, which show strong antibacterial effects. Meanwhile, Z. leprieurii, a lesser-known species from Uganda, demonstrated high efficacy (MIC: 32 μg/mL) and is rich in alkaloids and flavonoids with known anti-TB potential.

Geographically, Africa and South America contributed the most species, 30 and 33, respectively, while Asia yielded 14, and Europe yielded just 3. Africa had the highest mean MIC (696.16 μg/mL), reflecting both the low EO potency in selected species and their ethnomedical usage. Asia, although less represented, exhibited the lowest average MIC (90.41 μg/mL), suggesting the high pharmacological potential of species such as Sphaeranthus indicus (India) [23], Trachyspermum copticum (Iran) [32], and Salvia spp. (Turkey [36]).

Plant part analysis revealed that fruits and roots were associated with the lowest antimycobacterial activity (mean MICs of 873.68 μg/mL and 835.25 μg/mL, respectively), while leaves and aerial parts, though the most commonly used, yielded the highest average efficacy (166.51 μg/mL). For instance, the root EOs of Dorstenia psilirus, as well as the fruit EOs of Tetrapleura tetraptera and Dichrostachys cinerea, ranked among the lowest potent samples analyzed.

Mechanistically, EOs act through multiple avenues. Their lipophilic constituents, such as terpenes, phenols, and aldehydes, interact with the lipid-rich mycobacterial envelope, increasing membrane permeability, disrupting ion gradients, and leading to energy depletion [34]. Some components also inhibit efflux pumps and enzymes critical to cell wall biosynthesis. Compounds such as 1,8-cineole, thymol, and cinnamaldehyde have also demonstrated immunomodulatory effects, potentially enhancing the host's macrophage response to infection [35].

Notably, over 80 % of the studies included in this analysis specifically targeted MDR and XDR MTB strains, increasing the clinical relevance of the findings. The fact that several EOs demonstrated MICs comparable to standard anti-TB drugs (e.g., RIF or INH) supports their consideration in adjunctive therapy. For instance, the EO of Zanthoxylum leprieurii was effective against resistant strains and showed synergy with RIF [23], while Cinnamomum zeylanicum EO enhanced the activity of INH in vitro [33].

Nonetheless, translating in vitro efficacy into clinical application requires caution. Variability in EO composition due to plant species, geographical origin, extraction method, and environmental factors introduces challenges in standardization [37]. In addition, pharmacokinetics, toxicity, and therapeutic index remain insufficiently characterized for most EOs in this study. While some studies employed analytical fingerprinting for compound identification [29,32], comprehensive toxicological evaluations and bioavailability studies are lacking. Although the present review provides a comprehensive overview of the antimycobacterial potential of plant-derived EOs, a major limitation encountered during data extraction was the scarcity of cytotoxicity information across the included studies. Most investigations focused exclusively on in vitro antimicrobial assays and did not evaluate toxicity in mammalian cell lines or animal models, thereby preventing a systematic assessment of the therapeutic window or selectivity index. The absence of standardized cytotoxicity reporting underscores a critical gap in the current literature, as safety profiling is indispensable for advancing EOs toward preclinical development. Future studies should therefore integrate parallel cytotoxicity testing using harmonized methodologies to enable robust comparisons and support the translation of promising compounds into viable anti-TB drug candidates.

Future studies should prioritize integrated investigations that bridge phytochemistry, pharmacology, microbiology, formulation science, and translational medicine. A foremost priority is the isolation, purification, and structural characterization of active phytoconstituents within the most potent EOs identified. While crude EOs from genera such as Cinnamomum, Piper, Salvia, and Zanthoxylum exhibited potent antimycobacterial activity, their complex and variable compositions hinder precise attribution of efficacy to specific compounds. Employing analytical techniques such as gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), and nuclear magnetic resonance (NMR) spectroscopy will enable the identification and quantification of key bioactive. Isolated compounds should then undergo structure–activity relationship (SAR) analyses to refine molecular features responsible for antimycobacterial activity. Simultaneously, there is an urgent need to assess the synergistic or antagonistic interactions between EOs and existing anti-TB drugs. Preliminary evidence of synergism between compounds such as cinnamaldehyde and RIF or INH warrants systematic evaluation using standardized in vitro combination assays, including checkerboard dilution methods and time-kill kinetics. Determining fractional inhibitory concentration indices (FICIs) will elucidate the additive or synergistic potential, which may facilitate dose reduction of conventional drugs, thereby lowering toxicity and mitigating the development of resistance. Moreover, preclinical in vivo studies must be prioritized to address the pharmacokinetics, pharmacodynamics, and safety profiles of both crude oils and purified components. Most EOs studies remain limited to in vitro MIC assays, which do not reflect systemic absorption, bioavailability, metabolic fate, or toxicity. Rodent models of pulmonary TB should be employed to assess therapeutic efficacy, tissue distribution (especially lung targeting), dosing regimens, and potential adverse effects. Repeated-dose toxicity, genotoxicity, and immunomodulatory profiling will be essential to determine therapeutic windows.

To overcome challenges of volatility and poor aqueous solubility, novel drug delivery systems should be explored to enhance EOs' bioavailability and stability. Nanoparticle-based encapsulation (e.g., liposomes, polymeric nanoparticles, and solid lipid nanoparticles) can enhance pharmacokinetics, protect bioactive compounds from degradation, and provide sustained or targeted release at infection sites. Formulations should be validated through in vitro release kinetics and in vivo biodistribution studies. Further, mechanistic studies at the molecular and cellular levels are needed to elucidate the specific pathways disrupted by EO constituents in MTB. Transcriptomic, proteomic, and metabolomic profiling can provide insight into how EOs affect MTB gene regulation, protein synthesis, membrane integrity, and metabolic pathways. Such mechanistic clarity will strengthen the case for EO-derived compounds as viable drug candidates. Finally, future discovery efforts should integrate ethnopharmacological knowledge and computational modeling. Plants historically used for respiratory ailments, particularly those documented in Appendix 2, can be prioritized for bio-chemometric and in silico screening to identify high-affinity binders to validated MTB drug targets. Molecular docking, QSAR modeling, and predictive ADMET analyses can complement empirical research and streamline lead optimization.

5. Conclusion

The anti-TB activity of different materials, whether industrial or natural, is a crucial issue when MTB emerges as a DR bacterium. Herbal treatment of various disorders and diseases, such as infections, holds a bright future for herbal medication in combating infectious agents. MTB, one of the most dangerous bacterial species, is also susceptible to plant-derived compounds, such as EOs. EOs produced and secreted by plants exhibit suitable activity against MTB (especially about Euclea sp. and Croton sp.), as determined by MIC tests in various laboratories worldwide. In this meta-analysis, we attempted to describe these activities using statistical measures to gain a deeper understanding with detailed information about the possibility of utilization of EOs in future clinical studies. Although global efforts toward TB eradication remain feasible and scientifically justified, MTB exhibits remarkable adaptive capacity, enabling it to modulate metabolic pathways and develop mechanisms that diminish or evade the antimycobacterial effects of plant-derived compounds.

CRediT authorship contribution statement

Seyyed Mohammad Amin Mousavi-Sagharchi: Writing – original draft, Validation, Software, Methodology, Formal analysis, Data curation. Mina Rezghi Rami: Writing – review & editing, Writing – original draft, Data curation. Hanieh Hasani: Writing – original draft, Data curation. Maede Mohammad Ghasemi: Writing – original draft, Data curation. Maryam Farahhal: Writing – original draft, Data curation. Amir Hossein Hasani: Writing – original draft, Data curation. Seyyedeh Fatemeh Seyyedian-Nikjeh: Writing – original draft, Data curation. Maryam Meskini: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Methodology, Investigation, Conceptualization. Seyed Davar Siadat: Writing – review & editing, Validation, Supervision, Project administration, Methodology.

Ethics approval and consent to participate

Not applicable.

Clinical trial number

Not applicable.

Funding declaration

Not applicable.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: