Medicinal plants used by herbalists in the treatment of snakebites envenomation in the Acholi, Teso and Karamoja sub-regions of Uganda

Background There is high mortality and morbidity due to poisonous snakebites globally, with Sub Saharan African having one of the highest rates. However, Traditional Medicine Practitioners (TMP) have been treating snakebites in Uganda for long. However, few studies have been conducted to document such vital traditional indigenous knowledge before its lost. The aim of this study was to document the medicinal plant species used by experienced TMP in treating snake envenomation in selected post-conict parts of northern Uganda. Methods An ethnopharmacological survey was conducted in Serere, Kaberamaido and Kaabong districts in Uganda. Twenty-ve TMP with expertise in treating snakebites were purposively identied using the snowball technique, and interviewed using semi structured questionnaires. Data were analysed using simple descriptive statistics.

with neurotoxic venom include cobras, mambas, sea snakes, kraits and coral snakes [5]. Snakes with haemotoxic venom include rattlesnakes, copper head and cottonmouths [7]. Snake venoms can be neutralized by antibodies obtained after immunizing domestic animals with them. This led to production of anti-venom called antisera. A major drawback of serum therapy is its prohibitive cost and chance that victims are often some distance away from medical care when bitten [8]. The search for novel venom inhibitors from natural products is therefore relevant because of their potential to complement serum therapy in neutralizing mainly the local damages of envenomation. Plants extracts constitute an excellent alternative with a range of antivenom activities [7].
Africans have traditionally been treating poisonous snakebites using herbs [9][10][11][12]. For instance, 147 patients bitten by snakes were seen between November 1995 and October 1996 and 90% of them used herbs in KwaZulu-Natal, South Africa [13]. In Kenya, 32 medicinal plants have been documented for treatment of snakebites [14,15]. In central Uganda, 36 plant species were documented for treating snakebites [10]. A total of 25 plants were documented for treatment of snakebites during survey of traditional herbal drugs of Bulamogi county, in Uganda [11]. Five other medicinal plant species were documented for treatment of snakebites in Northern sector of Kibale National Park in Western Uganda [16].
The current population of Uganda is over 45.3 million [17]. More than 80% of Ugandans are involved in agriculture and live in rural areas [18] where making the populace highly vulnerable to snake envenomation without access to antisera in health facilities.
There is widespread use of medicinal plants for treatment of snakebites in Uganda although there are no statistics available snakebite treatment generally and the use of herbs to manage them, let alone their e cacy. Additionally, ethnopharmacological surveys of plants used for treatment of snakebites have not been done in many parts of Uganda. The aim of this study was to document the plant species used in the treatment of snakebites envenomation in the Acholi, Teso and Karamoja sub-regions of Uganda. These are post con ict regions were affected during the war led by Joseph Kony's Lord's resistance army rebel (LRA) out t. The LRA war begun in 1986 and lasted over 18 years [19]. Anecdotal evidence seems to suggest a high prevalence snakebites envenomation experienced by returnees during the post-con ict resettlement in northern Uganda. This is because as many as 2 million people who had ed the ghting were forced into internally displaced people's camps in northern Uganda [20].

Study design and setting
An ethnopharmacological study was conducted in the districts of Soroti in Mukura/Asuret sub-counties (1.7229° N, 33.5280° E), Serere (4994° N, 33.5490° E), Kaberamaido, Anyara sub-county (1.6963° N, 33.2139° E) in the Teso sub region, Kitgum, Namukora & Orom sub-counties (3.3397° N, 33.1689° E, Acholi sub region) and Kaabong, Timu sub-county (3.5126° N, 33.9750° E, Karamoja Sub region) (Fig. 1). The data was collected between August and October 2017 using interviews with semi-structured questionnaires These areas have tropical and savanna type vegetation [21]. The study areas were selected because are recovering from protracted LRA war, they are remote. Additionally, these areas have limited access to modern health facilities with antisera and have been reported to have frequent snakebites [22,23].

Characteristics of participants
Traditional medicine practitioners or herbalists with expertise in treating patients bitten by snakes were purposively selected and identi ed using the snowball technique [24].

Plant collection & identi cation
Voucher specimens of the plant species mentioned in the study were collected using standard procedures [25] and taken to Makerere University herbarium for identi cation. The scienti c names of the plant species were identi ed based on the Kew database at http://www.theplantlist.org accessed on 4th January, 2018 at 18:09 local time. Plant families were veri ed using the angiosperm phylogeny group IV at http://www.mobot.org/MOBOT/research/APweb/

Data analysis
The data were analysed using simple descriptive statistics in Microsoft Excel 2019.

Results
Twenty-ve TMP were purposively selected and interviewed. Only ve were women, the rest were men. The average age of the respondents was 54.7 years and ranged from 36 to 95 years. The majority of the respondents (80%) were illiterate, with only 20% having attained primary education and were all peasant farmers.
Sixty plant species from 28 families and fty-one genera were documented. The plant families with most species were Asteraceae (8), Fabaceae (7), Asparagaceae and Amaranthaceae with 4 species each and Euphorbiaceae, Meliaceae and Solanaceae with 3 species each ( Table 1). The genus with the most plant species was Echinops (3). This was followed by: Annona, Chlorophytum spp, Eucalyptus and Solanum with two species each (Table 1). The most commonly mentioned plant species were: Steganotaenia araliaceae (16), Microglossa pyrifolia and Gladiolus dalenii both at 13, Aframomum mildbraedii (11), Jasminum schimperi, Cyathula uncinulata, Crinum macowanii and Cyphostemma cyphopetalum (10), Annona senegalensis (9) Echinops longifolius (9), Gloriosa superba and Indigofera spicata (8) Tamarindus indica (7) S. araliaceae was mentioned by all the TMP in the Ik community. It was used as rst aid and is said to cause immediate vomiting only when used by someone bitten by a venomous snake.

Medicinal plant preparation & administration
The methods of preparation and administration were grouped into thirteen categories. Most of the herbal medicines were prepared for oral administration (62.5%). The rest were administered topically (32.5%) with the exception of inhalation of smoke from burnt plant material (1.2%). The commonest methods of oral application were cold water infusions (31.8%), decoctions (21.2%) and chewing or squeezing juice from the plant material and drinking it (5.9%). The commonest topical methods of application were poultices (9.4%) and direct application of powders to the bitten site or juice (8.2%), then application of powder to bite area after making small cuts with a razor blade (4.7%). Some of the plant species were used as snake repellents (3.6) and one specie was used for making an eye wash (1.2%) for cases of ocular envenomation by spiting cobras. One herbalist reported burning the plant material and making the patient to inhale the smoke in cases where they were unconscious (1.2%).
In the Ik community in Kaabong district, herdsmen, farmers and hunters usually moved with small quantities of G. dalenii powder as a quick remedy in case of being bitten by a poisonous snake. In case of a snakebite, small cuts are made at the site and the powder applied. Generally, the consensus among the TMP was high in the relatively closed and isolated Ik community. Additionally, the medicinal plant species they used were not used by the other communities interviewed elsewhere but only mentioned by the Ik community. These included G. dalenii, E. longifolius, Cyathula uncinulata and Steganotaenia araliaceae.

Knowledge acquisition and transfer
Most herbalists acquired their knowledge on the use of medicinal plants for snakebite management from their parents and grandparents (80%) other relatives (12%). Two unique cases (8%) involved eld observation of self-medication in snakes. In the rst case, one herbalist reported that he acquired knowledge on the use of Microglossa pyrifolia for treatment of snakebites by observing a snake wounded in a ght with another snake using it and reportedly recovering from its injuries. In the latter case, another herbalist who mainly uses the root of Opilia amentacea for all snakebite cases because of its perceived e cacy. The choice of this species was based on its whitish scaly stems which look line the scales of a snake.

Type of snakebites treated
Cyathula uncinulata, Astripomoe amalvacea, Kalanchoe sp. and Hoslundia opposita were speci cally mentioned as being used for treating puff adder bites. Euphorbia hypericifolia was also used for treating scorpion and spider stings. Microglossa pyrifolia was used for treating all types of snakebites except puff adder. Bryophyllum delagoense and Steganotaenia araliaceae were used for treating cobra bites. In addition, S. araliaceae was used as rst aid for all snakebites. Most of the plant species used are harvested from the wild (68.75%), 24.24% domesticated and 6.25% from both domesticated and wild. Thirty-seven (61.6%) of the documented plant species did not have any previous references about use in snakebite treatment in the literature.

Unidenti ed medicinal plant species used
An additional nine plant species were mentioned by the herbalists for treating snakebites envenomation in Kitgum district (Table 2). However, we were not able to collect voucher specimen for these species for identi cation for several reasons including wild res that had destroyed some of their habitats, drought, insecurity near the Uganda Sudan boarder and di culty in locating some of the species because they were naturally rare.

Discussion
Even though most herbs were reportedly used singly, most of the herbalists prepared a polyherbal formulations for use. One of the TMP used powder consisting of Pseudocedrela kotschyi, Gardenia ternifolia, Zanthoxylum chalybeum, Indigofera arrecta and Capsicum frutescens.
Although we reported most of the plant species recorded in this study not having any previous documentation for use in snakebite treatment in the literature, some unique cases are presented as follows. We recorded for the rst the use of species Opilia amentacea in the treatment of snakebite envenomation in Uganda. Interestingly, the same species is used in India for snakebite envenomation [26]. However, in Uganda, the single mention was one renowned traditional healer specialising in treating snakebites commonly known as "Dr. Snake" was associated with doctrine of signatures (DoS) or similarities. The selection of plant species for treatment of particular conditions because of their resemblance to particular organs is not a new concept. This DoS or similarities attributes the therapeutic properties of some plants to particular morphological characters or features they possess, i.e. "like treats like" [41]. This particular herbalist begun using this plant because of the scaly and dotted appearance of its bark and its creeping habit. This is the rst report on the doctrine of signatures with O. amentacea with reference to snakebite. According to Bennett, [42], the Doctrine of Signatures is found throughout the world and has had a long history of use. He further argues that considering the DoS from the classical morphological perspective has rarely led to the discovery of medicinal plants and the approach is therefore unproductive and largely untestable. The DoS cannot therefore be considered scienti c [41,42], although parts of its utility lies in facilitating the process of understanding the subjective, psychological, and spiritual dimensions of nature [43].
Another interesting observation was the routine use of prayers during healing. One particular healer was observed to always began his plant collection routine in the eld with prayers. He prayed to God beseeching him to give the plant species their healing power before he begun harvesting. He professed the catholic faith and attributed his success to his God. This particular healer had a medicinal plant garden and a special treatment room/hut in which he treated his patients. He got o cial recognition with a certi cate from the ministry of culture in Uganda as early as 1986. The citation of prayers by herbalists who profess Christianity during healing rituals has previously been reported in western Uganda [44]. Prayers form an integral part of the belief system and are believed to make the treatment successful.
The transfer of traditional knowledge is by word of mouth. The TMP identify and train particular children on the identi cation, preparation and administration of the herbs. We report a unique case of self-medication in snakes. This proving an insight into the antivenom potential of Microglossa pyrifolia. Although previous studies have reported cases of self-medication especially in primates such as chimpanzees [45,46] we have not come across previous reports of self-medication in snakes. However, according to Shurkin, [47] some lizards are believed to survive venomous snake bites by eating roots of particular plants. It is therefore not farfetched to consider self-medication in snakes.
According to the in-charge of Timu health centre II in Kaabong district, there were relatively many reports of snake bites in the Ik community, but there were few cases reporting to the health facility health centre. Even those who reported to the health centre came several days after being bitten by snakes for supportive treatment after initially managing the snakebites with herbs. The health centre also did not have any antisera for treatment of snake bites.

Conclusion
TMP widely use several medicinal plant species for treating snakebites envenomation in the post-con ict sub-regions of Acholi,