Medicinal plants used in managing diseases of the respiratory system among the Luo community: an appraisal of Kisumu East Sub-County, Kenya

Background Poor access to healthcare in rural communities causes many people to seek herbalists who use medicinal plants for the treatment of various disease conditions. Most knowledge of traditional herbal medicine makes use of indigenous remedies which are often undocumented and are at risk of being lost. The preservation of this knowledge may facilitate scientific inquiry into promising new therapeutic molecules. Methods Semi-structured questionnaires were used to collect the sociodemographic information of 30 herbalists in Kisumu East Sub County. The local names of medicinal plants used in managing illnesses of the respiratory system, their habit, active parts, indications, methods of preparation, routes of administration, scientific identity, and conservation status were also recorded. Other reported traditional uses, pharmacological activities, and toxicological data were identified via a literature search. Results Most herbalists were female (86.7%), aged between 61 and 70 years (43.3%) with no formal education (56.7%), and had 21–30 years of practice (30%). 44 plant species, belonging to 43 genera and 28 families were identified. Leguminosae and Rutaceae plant families were predominant, leaves were frequently used (33%), and trees were the most common habit (44.4%). Most plants were collected in the wild (79.2%), preparation was mainly by decoction (68.8%), and the administration was mainly orally. The main indication was cough and 79.5% of all documented plant species had previously been reported to have a pharmacological activity relevant to the mitigation of respiratory illnesses. Toxicological data was available for 84.1% of the plant species identified. Conclusions The predominant use of roots, root barks, and root tubers by herbalists in Kisumu East Sub County threatens to negatively impact the ecological survival of some plant species. The preservation of herbalists’ knowledge of medicinal plants in the study area is a pressing concern considering their advanced age and little formal education. There is a need to conserve some of the medicinal plants documented in this study. The medicinal claims made by herbalists also warrant scientific scrutiny.


Background
The global burden of respiratory diseases makes for daunting reading. Lower respiratory tract infections (LRTI) and chronic obstructive pulmonary disease (COPD) reportedly claimed 6 million human lives in 2016 [1]. The prevalence of COPD in Sub Saharan Africa has been reported to be between 4 and 25% and > 100,000 deaths have been linked to non-communicable diseases including those of the respiratory system [2,3]. Diseases of the respiratory system hurt individual productivity and are responsible for more than 10% of all disabilityadjusted life years [4].
According to a 2013 Kenya National Bureau of Statistics (KNBS) economic survey, pneumonia, and tuberculosis were responsible for 13.7% of all total deaths in the Nyanza region [5]. It is important to note that illnesses of the upper respiratory tract are the second leading cause of death in Kisumu County [6]. Poor access to healthcare and scarcity of health resources in rural areas such as many parts of Kisumu East Sub County causes many inhabitants of such areas to rely on indigenous plant resources to manage common diseases including those that affect the respiratory system. Plant-based indigenous remedies may be key in the future management of respiratory system diseases [7]. However, the potential of this resource is largely untapped due to inadequate documentation by the herbalists who prepare the remedies.
The rapid development of infrastructure in Sub Saharan Africa including Kenya threatens to destroy cultural lands where medicinal plants are cultivated. This is problematic given that the knowledge of these plant resources is mostly an extension of people's culture [8,9]. Herbalists are usually the custodians of medicinal plants in these communities. By documenting the knowledge held by herbalists, vital information on the medicinal plants may be preserved. The current study aimed to collect ethnobotanical data on medicinal plants used by herbalists in the management of respiratory diseases in Kisumu East Sub County.

Ethical approval and consent to participate in the study
Ethical approval for the study was obtained from the Biosafety, Animal Use and Ethics committee of the University of Nairobi (Ref: FVM BAUEC/2019/210). Approval was additionally sought from regional administrators (the area chief and assistant chief ) who were duly notified of the study's objectives. The scope, possible benefits, and risks of the study were explained to willing participants (herbalists) and consent forms were made available to them for signing.

Study area
The study was conducted in Kisumu East Sub County in Western Kenya (Fig. 1). The study area is approximately 365 km from Nairobi (the administrative capital of Kenya) and covers an area of approximately 135 km 2 . It lies within latitudes 0° 20′ South and 0° 50′ South and longitudes 33° 20′ E and 35° 20′ E and comprises of several administrative wards including Kolwa Central, Kolwa East, Manyatta B, Nyalenda A, and Kajulu East and West [10]. Moreover, the population in this area is about 220,977 according to the 2019 Kenya Population and Housing Census [5]. It receives an annual relief rainfall of between 1200 and 1300 mm and annual temperatures range between 20 and 35 °C. The major economic activities of residents include fish farming, and agriculture (sugar, livestock, and poultry farming) [10].

Data collection
The study was conducted between March and September 2019. Ethnobotanical data were obtained by using semistructured questionnaires. The target respondents were local herbalists with good ethnobotanical knowledge of the plants used in managing respiratory diseases and related symptoms. Thirty local herbalists were selected for interviews which were conducted both in Kiswahili and Luo dialect with the aid of a botanist familiar with the languages. Each of the respondents was interviewed individually to ensure confidentiality. The interviews sought to answer the following questions; • Which plant parts are most commonly used in preparing the indigenous remedies indicated for respiratory illnesses? • Which methods are adopted in preparing the indigenous remedies? • Which respiratory illnesses are most commonly treated with medicinal plants in the study area? • Which plant species are used in the preparation of the remedies? • How are the indigenous remedies administered? See Additional file 1.

Collection and identification of plant specimens
Several trips were made to the homesteads of the herbalists where voucher specimens were collected and pressed and later identified by a botanist before being deposited at the University of Nairobi Herbarium. Information on the vernacular name, plant part used, plant habit (i.e. the general appearance, growth form, or architecture), plant status, method of preparation, and route of administration were collected.

Literature search strategy
A literature search was conducted on MEDLINE, Pub-Med, PubMed Central (PMC), Google Scholar, the Directory of Open Access Journals (DOAJ), The Journal Author Name Estimator (JANE), University repositories, and from grey literature to identify relevant articles/ theses/ reference material containing information on previously reported traditional uses, pharmacological/ chemical activities, and toxicological data on the medicinal plants indicated for the management of respiratory illnesses in Kisumu East Sub County. Studies were excluded if they were not in English.

Data analysis
Frequencies and percentages were used to analyze the sociodemographic data of the herbalists. The relative frequency of citation (RFC) was used to evaluate the ethnobotanical data.

Relative frequency of citation (RFC)
This was done to determine the number of herbalists who considered particular plant species were worth mentioning in the management of diseases of the respiratory system. The value was calculated using the formula described by Tardio and Santayana [11]; where Fc is the number of herbalists who cited a particular species and N is the total number of herbalists (Table 1).   Fig. 3. Leaves were the most frequently used parts (33%), followed by roots (28%) and stem bark (24%). Root bark, fruits, corms, bulbs, and root tubers accounted for 15%. Roots, root bark, root tuber, and stem bark accounted for 60% of plant parts used in the management of diseases of the respiratory system (Fig. 3).

Dosage, mode of preparation, and route of administration
Various methods were used to prepare herbal medicine used for managing diseases of the respiratory system in the study area ( Table 2). The most common method was decoction (68.8%), concoction (20.8%), and chewing (4.2%) ( Table 2). Other methods of preparation included cold maceration, powdering, and crushing before instillation in the nostrils which accounted for 2.1% respectively ( Table 2). The main route of administration of the indigenous remedies prepared by the traditional medicine practitioners was oral ( Table 2).

Pharmacological and toxicological reports on the medicinal plants documented in this study
Of the 44 plant species documented in this study, 95.5% had studies that had reported their pharmacological/ chemical activity (Table 3). Moreover, 79.5% (35/44) of the documented medicinal plants had previously been reported to be effective against microorganisms that are associated with respiratory illnesses and 84.1% of the plant species had toxicological data ( Table 3).

Socio-demographic information of herbalists in the study area
Many of the herbalists interviewed in this study were older members of the society. It has previously been reported that traditional herbal practice is usually a preserve of the older members of the society [240,241]. It is also important to note that it is often harder for the younger generation of herbalists to be accepted by their communities as they are considered to be inexperienced in key tenets of traditional herbal medicine [240,241]. The observation that many of the interviewed herbalists had not received any formal education seems to agree with what has been observed by other authors [241].

Diversity of medicinal plants identified in the study area and their use
The Leguminosae plant family was the most dominant family indicated for respiratory illnesses in the study area. According to Christenhusz and colleagues, Leguminosae has a large global distribution and is the 3rd largest plant family in the world (after Orchidae and Asteraceae) [242]. The worldwide distribution of this plant family may have some influence on the decision of herbalists to use the plants from this family [243].
The predominance of trees as a source of herbal therapies may have something to do with their abundance, easy availability throughout the year, and resistance to   [243][244][245]. Leaves are considered by herbalists to be important photosynthetic organs [241,243]. Thus, it is not surprising that they were the most frequently used plant parts in the study area. It was disturbing to note that many of the herbalists in the area were uprooting the plants that they used for making some of the indigenous remedies. Furthermore, in the course of the interview, some of the herbalists had reported that Warburgia salutaris and Zanthoxylum gilletii were no longer available in some parts of Kisumu East Sub County owing to poor conservation practices. According to Maroyi, it is not advisable to over use the roots and stem barks of plants for medicinal value as this may sabotage plant conservation efforts [246]. Notwithstanding, some herbalists reported that they only collected plant parts in quantities that were enough for   Antimalarial activity [14] In vivo (Swiss albino mice) [14] The methanol leaf extract was reported to be non-toxic in mice with a median lethal dose of > 2000 mg/ kg [14] Aloe kedongensis Reynolds Malaria [15] Antiplasmodial activity (aqueous leaf extract), leishmanicidal activity (aqueous and methanol extracts) [16] In vitro (semi-automated microdilution assay, anti-leishmanial assay, anti-promastigote assay, anti-amastigote assay, MTT assay) [16] The aqueous and methanol leaf extracts were reported to have low cytotoxicity against human embryonic lung fibroblast (HELF) cell lines (CC 50 > 500 µg/mL) [16] Allium sativum L.
Malaria, wound disinfectant, intestinal infections [17], cold [18], aphrodisiac [19] Chemoprophylaxis against lead nitrate induced toxicity in mice [20], increase in the weight of seminal vesicles and epididymis of male animals and elevation of sperm count [21], antibacterial and antifungal activity (essential oil extracts) [22] In vivo (Swiss albino mice) [20], in vivo (Swiss albino mice) [21], in vitro (disc diffusion and yeast glucose Chloramphenicol Agar method) [22] The LD 50 in rabbits was reported to be 3034 mg/kg with a maximum tolerated dose of 2200 mg/kg [23]. Mortality in rabbits was recorded at 3200 and 4200 mg/kg. Anorexia and paralysis were observed in rabbits at high doses [23] The aqueous extract at a 300 mg/ kg dose was reported to have mild toxicity symptoms in Wistar rats, but doses of 600 mg/kg and 1200 mg/kg were reported to elevate biochemical parameters. No toxicity was reported up to a dose of 2500 mg/kg and LD 50 was reported to be > 5000 mg/ kg [24] Rhus natalensis Bernh.
Malaria, fever [75], swollen testicles, and abdominal pains [39], pneumonia [25] Antifungal and antibacterial activity (Ethanolic root extract) [76] in vitro (Agar diffusion technique) [76] The 95% ethanol extract was reported to be nontoxic in mice and no mortality was observed even at concentrations of up to 5000 mg/ kg. However, drowsiness in doses between 1200 and 5000 mg/kg was reported [77] The compounds isolated from the methanol stem bark fraction (resveratrol derivatives) were reported to have low cytotoxicity on prostate cancer cell lines [78]. The ethanol root extract was reported to be nontoxic in brine shrimp larvae [74] [84] in vitro (Hole plate diffusion method, microdilution method) [83], in vitro (Bioautography assay) [84] The acetone leaf extract was reported to be cytotoxic against cancer cell lines [85] Carica papaya L.
Malaria, liver disease [12], tuberculosis [30], malaria, [86,87], fever [18] Antibacterial activity (Methanol root extract) [88], antitumour activity and immunomodulatory effects (Aqueous leaf extract) [89] in vitro (Cup plate agar diffusion method) [88], in vitro (Cell viability assay, caspase assay, microarray analysis) [89] The aqueous and ethanol leaf extracts were reported to be cytotoxic on human oral squamous cell carcinoma SCC25 cell lines [90] The aqueous leaf extract was reported to disrupt cell division and to induce mitotic spindle disturbance in Allium cepa [91] The methanol leaf extract was reported to be cytotoxic against LLC-MK2 cell lines [92] The aqueous leaf extract was reported to be non-toxic in Sprague Dawley rats at a 2000 mg/kg dose [93] No morphological alterations were reported in Sprague Dawley rats treated with a 28-day repeated oral dose of 2000 mg/kg [94] Aqueous and ethanol leaf extracts were reported to be nontoxic at doses of up to 5000 mg/kg [95] The methanol leaf, root, and stem bark extracts were reported to be nontoxic against MRC-5 cell lines  Croton megalocarpus Del.
A 400 mg/kg dose of the iso saline leaf extract administered intraperitoneally in Sprague-Dawley rats significantly elevated serum levels of alanine and aspartate aminotransferase, and significantly lowered the blood glucose levels [119].

No reports
Tamarindus indica L.

Cissus rotundifolia (Forssk.) Vahl
Threatened abortion/contraception [113] Pain [128] Malaria, liver disease and otitis [235] Malaria [159] Antibacterial activity (Buffered methanol (80% methanol and 20% PBS) and acetone) [236], hypoglycemic activity(Aqueous leaf extracts) [237] in vitro (Agar well disc diffusion assay) [236], in vivo (Wistar rats) [237] The methanol (70%) extract of aerial parts was reported to be more cytotoxic on MCF-7 (breast cancer) cell lines than doxorubicin (IC50 = 0.77 µg/mL and 3.45 µg/mL respectively) [238] Rhoicissus revoilii Planch Pneumonia, tonsillitis [239] Antifungal activity against C. albicans (Ethanol extract) [239] in vitro (Agar well disc diffusion assay) [239] No reports Mailu et al. Chin Med (2020) 15:95 their work and which would not hamper conservation efforts. It is also worth mentioning that a local name for Acanthus polystachyus was not available. Instead, there was a consensus among the interviewed herbalists that 'Nyanandi' was the closest semblance to a name that this plant could be given on account of the assertion that it may have originally have been brought in from Nandi County which happens to be an immediate neighbor of Kisumu County.

Dosage, mode of preparation, and route of administration
Teaspoons and tablespoons were used for measuring the dosages of powdered plant materials such as barks, stems, or roots while glasses or cups were used for measuring doses of concoctions or decoctions. While the use of 300/500 mL cups was commonly recommended by the herbalists as a means of measuring the dosages of concoctions/decoctions to be used, there was ambiguity in how this was applied. This trend was also observed in a previous report where medicinal plants used for maternal healthcare in Katsina state, Nigeria were surveyed [18].
Decoctions and concoctions were the most common method of preparing indigenous remedies and was done by the herbalist or by the patient who was given instructions on how to make the preparation. The process often involved harvesting the plants, drying them in the sun or in the house for a period of several days, and crushing them into powder with the aid of a homemade mortar and pestle.
The preparations would then be stored in plastic soda bottles that varied between 500 mL and 2 L and sold to the patients directly or in the market. Powdered plant parts could be included in tea and administered orally.
The route of administration was majorly orally. In the case of Eucalyptus camaldulensis, decoctions were prepared by boiling the leaves in an earthen pot and the patient was advised to cover themselves with a blanket such that the emanating steam completely engulfed them. This was done over a period of time and the patient would later be advised to take 2 teaspoons of the decoction in the event that they had a common cold. Patients were asked to revert back to the herbalist for further directions in case they did not feel better. It is worth noting that many of the interviewed herbalists were of the opinion that their remedies rarely failed. In the minds of the herbalists, the failure of the remedies to work was largely due to the incapacity of the patients to follow the instructions issued by the herbalists.
The interviewed herbalists were of the opinion that their remedies had minimal side effects. However, it is not clear whether these herbalists had the capacity to identify any adverse events or whether they had any mechanisms to report such cases whenever they occurred.

Pharmacological reports and toxicology of the medicinal plants documented in this study
To the best of our knowledge, this is the first study to document the medicinal plants used in the management of respiratory illnesses by herbalists in Kisumu East Sub County. It is interesting to note that up to 84.1% of the medicinal plants documented in this study have previously been reported to be effective against Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, Aspergillus spp, and Candida albicans. These microorganisms have been associated with pneumonia and tonsillopharyngitis [247].
The most cited plants in this study were Warburgia salutaris, Zanthoxylum gilletii, Carissa edulis, Tylosema fassoglensis, and Harrisonia abyssinica. Carissa edulis and Clerodendrum myricoides have been reported to be useful in the management of asthma, cough, and cold [37,105]. The similarity of our observations to those made by previous authors seems to suggest that there may in fact be a consensus among herbalists from different communities with regard to the usefulness of some of the medicinal plants in their environment.
Toxicological data was not available for 4 species of plants including Croton dichogamous, Rhynchosia elegans, Mollugo nudicaulis, and Rhoicissus revoilii. Moreover, there was no pharmacological data on Croton dichogamus, and Rhynchosia elegans. This may be a potential gap that may need filling in the future.

Conclusions
The predominant use of roots, root barks, and root tubers in preparing decoctions by herbalists in the study area threatens the ecological survival of some of the plant species used. The preservation of ethno medicinal knowledge in the study area is a pressing concern considering the advanced age and little formal education of the herbalists interviewed. Plans to conserve some of the medicinal plants documented in this study should be initiated. There is a need to scientifically scrutinize the medicinal claims made by the herbalists interviewed in this study.

Limitations
The dosage frequency, duration of treatment, and storage condition of the powdered plant material, decoctions, or concoctions were not captured during the interviews. Information on the duration of treatment was also not captured.