Abstract
Background. People with diabetes, herbalists, and traditional medicine practitioners (TMPs) from Kinshasa use plants to treat diabetes, but no study has inventoried the plants used by these populations. The present study was conducted to identify the plants used in Kinshasa to treat diabetes mellitus. Methods. The survey conducted in the form of a semistructured interview between March 2005 and August 2006 made it possible to collect ethnobotanical information from people with diabetes (n = 126), herbalists (n = 80), and TMPs (n = 120). Results. The 326 subjects consulted (sex ratio M/F = 0.6, age 51 ± 7 years, and experience: 17 ± 5 years) provided information on 71 plants, most of which are trees (35%), belonging to 38 families dominated by Fabaceae (19.7%) and indicated in 51 other cases of consultation dominated by malaria (12%). From these 71 plants derived, 86 antidiabetic recipes were administered orally, where the leaf is the most used part (>50%) and the decoction (>46%) is the most common mode of preparation. This study reports for the first time the antidiabetic use of 11 species, among which Tephrosia vogeliiX (0.08), Chromolaena corymbosaX (0.06), and Baphia capparidifoliaX (0.06) present the highest consensus indexes (CI) and Marsdenia latifoliaW (UVp = 0.08) and Rauvolfia manniiX (UVp = 0.06) present the highest UVs. Conclusion. The results show that Kinshasa people treat diabetes using several plants. Some are specific to the ecological environment; others are used in other regions. Pharmacological studies are underway to assess the therapeutic efficacy of these plants.
1. Introduction
Diabetes mellitus (DM) is a metabolic and chronic disease involving inappropriately elevated blood glucose levels (hyperglycemia). Hyperglycemia alone can impair pancreatic beta-cell function and contributes to impaired insulin secretion. The body cannot produce and secrete sufficient insulin hormone or use it effectively. This insulin deficiency leads to elevated blood glucose levels and decreases carbohydrate and protein metabolism [1, 2].
Worldwide, 537 million adults (20–79 years old) live with diabetes. This number is expected to rise to 643 million by 2030 and 783 million by 2045. More than 3 out of 4 adults with diabetes live in low-income countries like DR Congo, and diabetes has been responsible for 6.7 million deaths in 2021, i.e., one end every 5 seconds [1]. Cardiac, vascular, neurological, and renal damage and neuropathy may occur without appropriate treatment. Treatment includes diet, exercise, and medication [3]. Given that most cases occur in low-income countries, where a significant fraction of the population resorts to traditional medicine, which essentially uses plant resources, medicinal plants constitute a credible alternative in the fight against diabetes mellitus.
Natural products have been shown to play an essential role in regulating pathophysiological signaling pathways, particularly in diabetes [4]. Over 800 plant species showing hypoglycemic activities can be vital sources in discovering and developing new types of antidiabetic molecules [5].
Medicinal plants have a high potential to treat various ailments due to the presence of their significant bioactive phytoconstituents. Certain plants are rich sources of compounds reputed to be antidiabetic such as flavonoids, alkaloids, phenolic compounds, and tannins, improving pancreatic tissues’ efficiency by increasing insulin secretion or decreasing intestinal glucose absorption [6].
The World Health Organization insists that scientists conduct ethnomedicinal, ethnobotanical, and ethnopharmacological investigations to record and preserve traditional knowledge, create databases, and validate scientifically traditional claims from the perspective of developing improved medication [7].
In the Democratic Republic of the Congo, a country with similarities to other developing countries, data on the prevalence of diabetes mellitus and the use rate of medicinal plants in their therapeutic load is not sufficiently available. Although less systematic, some studies have reported presumed antidiabetic plants in Kinshasa [8, 9].
This study completes the data reported in these studies while highlighting plants used by TMPs, diabetics, and herbalists in Kinshasa.
2. Methods
2.1. Experimental Framework
The city-province of Kinshasa is located between 4°18′ and 4°25′ south latitude and between 15°18′ and 15°22′ east longitude. It is bounded to the north and east by the province of Kwilu, to the south by the province of Congo Central, and to the west by the Republic of the Congo, with an average altitude of 300 m above the sea. Kinshasa city’s climate is tropical. It is characterized by a long rainy season lasting for 8 months (October–July), discontinuous between January and February, followed by August and September in favor of the short dry season. The vegetation of Kinshasa consists of degraded primitive forests, savannahs, and aquatic and semiaquatic formations in the valleys and Pool Malebo. It belongs to the Guinean-Congolese region, the Congolese basin domain, and the Congolese-Zambezian transition sector [10].
2.2. Ethnobotanical Data Collection
This cross-sectional descriptive ethnobotanical study was conducted between March 2005 and August 2006 through semistructured interviews based on a questionnaire. The discussion focused on knowledge about plants used to manage diabetes mellitus. Three groups of subjects were consulted: people with diabetes, herbalists, and TMPs from Kinshasa.
People with diabetes were met in 4 health zones of the city of Kinshasa: Bumbu, Kalamu, Limete, and Makala, covering the biomedical centers for the care of people with diabetes in Kinshasa supervised by BDOM (diocesan office of biomedical works). The reference health centers (RHCs) Libundi, Bondeko, Saint Clément, 2e Rue, the Bondeko clinic, and the Saint Joseph General Reference Hospital were concerned.
The herbalists consulted during this study were met in 5 popular markets in Kinshasa, the central market, the Selembao market, the Matadi Kibala market, the Mariano market, and the UPN market.
The TMPs were met in 4 communes of Kinshasa (Bumbu, Kalamu, Limete, and Makala). These municipalities are covered by the health zones selected during the survey of people with diabetes. TMPs were reached by snowballing from the TMPs provided by the population in each area concerned, and the investigations were carried out in Lingala. The listed plant species were collected and placed in herbariums, then compared to the reference herbariums from Kisantu (National Agricultural Study and Research Institute), to identify the scientific names. These were then formatted according to the Plants of the World online database: Plants of the World Online/Kew Science (https://powo.science.kew.org/) or African Plant Database (https://africanplantdatabase.ch/) or the World Flora Online (https://www.worldfloraonline.org/).
2.3. Data Analysis
Three ethnobotanical indexes were calculated to assess the significant species: the usual value (UV), the fidelity index (FI), and the consensus index (CI). The following formula determined the usual value (VU): VU = (∑Ui)/Ni, where Ui is the number of uses mentioned by an informant for a plant (organ) and Ni = the number of informants who cited the plant. The following formula calculated the consensus index on the plant (organ) (CI): CI= Np/N, where Np is the number of people who cited the plant (organ) and N is the number of people who were consulted in the study. The fidelity index of the recipe was calculated by the formula FI = nr/Np, where nr = the number of people who cited the recipe and Np is the number of people who cited the plant. Apart from the 3 ethnobotanical indexes mentioned above, the relative citation frequency (RCF) was also determined. It was calculated by the formula FCR = nx100/N with n being the number of occurrences of the factor examined and N = total number of the population concerned. This factor was used to quantify various factors analyzed in this study, except those involved by the ethnobotanical indexes.
In this study, UV makes it possible to evaluate the medicinal importance of a plant in the environment; CI makes it possible to identify the level of consensus of the population on the use of a species in the management of diabetes mellitus; and FI makes it possible to establish the level of loyalty that emerges among informants on an antidiabetic recipe.
3. Results
3.1. Ethnobotanical Profile of Inventoried Plants
Seventy-one plants were inventoried during this study. These plants were informed by 3 types of informants, herbalists (class Y), diabetics (class X), and TMPs (class W). The coexistence of these 3 types of informants gave rise to two other classes, the class of plants common to the 3 sources (class Z) and plants familiar to herbalists and TMPs (class V). As such, the results show that 16 plants are from diabetics, 7 plants are from herbalists, 22 are from TMPs, 6 are from both herbalists and TMP, and 20 are from TMPs, herbalists, and diabetics (Table 1). It should be noted, however, that the TMP class provided the largest most significant number of species, i.e., 31% (Figures 1(c) and 1(d)).

(a)

(b)

(c)

(d)
Each of these plants is named in one of the 15 Congolese ethnic groups reported in this study, of which Kikongo (50.7%) and Lingala (11.3%) occupy the first two places (Figure 2). These plants belong to 38 families dominated by the Fabaceae (19.7%), followed, respectively, by Rubiaceae Juss (7.0%), Apocynaceae Juss (5.6%), Asteraceae Bercht and J. Presl (5.6%), and Phyllanthaceae Martinov with 5.6% (Figure 3). These plant taxa are mostly trees (35%) or perennial herbs (31%) endemic to tropical Africa (Figures 1(a) and 1(b)). Among the 71 inventoried species, only Persea americana Mill (30.67%), Senna alata (L.) Roxb (30.67%), Garcinia kola Heckel (18.40%), Piliostigma reticulatum (D.C.) Hochst (17.79%), and Vachellia karroo (Hayne) Banfi and Galasso (17.79%) showed a relative citation frequency >15% (Table 1).


85% of identified medicinal plants in the present study have been previously used in local medicines as antidiabetic plants (13%). However, 72% have demonstrated an antihyperglycemic effect. Nevertheless, this study highlights 11 plant species never reported before as antidiabetic plants, among which Tephrosia vogeliiX (FCR = 7.98), Chromolaena corymbosaX (FCR = 6.44), and Baphia capparidifoliaX (FCR = 5.52) are the most cited (Table 1).
3.2. Ethnomedical Profile of Inventoried Plants
The 71 plants inventoried during this study are used in 86 antidiabetic recipes, of which 80 recipes use one plant, and 6 combine two plants (Table 2). The administration of the two types of recipes is essentially done orally, the leaf is the most used organ (50 and 58.3%), and decoction is the most predominant mode of preparation with 46.3 and 66.7% (Figure 4).

Figure 4 indicates that Baphia capparidifolia, Chromolaena corymbosa, Marsdenia latifolia, Rauvolfia mannii, and Tephrosia vogelii are among the critical plant species, according to their consensus index and use value.
In these recipes (R81–R86), we can observe the associations of leaves-leaves, roots-roots, and leaves-roots, representing half of these associations. Only 12 of the 71 plants are involved in the recipes combining two plants. For these recipes of plants in association, the treatment duration varies between 30 and 45 days, and it is the combination of Albizia adianthifolia (leaves) and Annona senegalensis (roots) with an RCF of 4.6 which is the most cited: R85 (Table 3).
The consensus indexes of the identified antidiabetic recipes (ICR) varied between 0.01 and 0.31, with the highest value observed for the recipes R61, based on Persea americanaX leaves, and R72, based on leaves of Senna alataX.
For the 11 plants reported for the first time by this study as antidiabetics, the ICRs vary between 0.08 and 0.01, with Tephrosia vogeliiX (0.08), Chromolaena corymbosaX (0.06), and Baphia capparidifoliaX (0.06) showing the highest values. Usual plant UVp values vary between 0.02 and 0.12; the highest values were observed in Annona senegalensisW, Nauclea latifoliaY, and Erythrina abyssinicaW with an UVp = 0.12 each. Note also that in the group of 11 taxa reported for prayer times by this study, UVp varies between 0.02 and 0.08; these are Marsdenia latifoliaW (UVp = 0.08) and Rauvolfia manniiX (UVp = 0.06) which presented the highest values in this class (Table 2).
The 71 plants inventoried are involved in 51 other causes of consultation, including malaria (12%), cough (9%), diarrhea (7%), wounds (6%), and sexual weakness, which occupy the first 5 places. The other pathologies are below the 5% mark (Figure 5).

3.3. Sociodemographic Profile of Subjects Consulted
The surveyed and consulted during this study were either diabetics (38.7%) or herbalists (24.5%) or practitioners of traditional medicine (36.8%), primarily women (62.3%) whose majority age is between 48 and 58 years old. Still, the extremes are 18 and 70 years old. They mainly live either in the municipality of Kalamu (29.8%) or in Makala (29.4%), and their level of education is, on average, secondary (47.2%). They mainly exercise 4 types of profession, the most representative of which is either commerce (28.5%) or liberal work (24.2%). In most cases, they have experienced more than 11 years of use of medicinal plants in managing diabetes (Table 4).
We also sought to determine the correlation between a few variables that characterize the subjects consulted within the framework of this study. The results show a positive linear correlation (R = 0.95, y = 0.4134x-6.5581) between the age of diabetic subjects and the number of years they have lived with diabetes. Similarly, a positive linear correlation is also observed between the age of herbalists and their years of experience (y = 0.5148x − 13.198, R2 = 0.8372). In the same way, we noted a positive between the TMP age of the TMPs and the profession’s expertise (y = 0.4076x − 8.3606, R2 = 0.807) and their age with the number of patients they receive per quarter, y = 0.4525x − 10.394, R2 = 0.9482 (Figure 6).

(a)

(b)

(c)

(d)
4. Discussion
The Congolese, particularly the flora of Kinshasa, have plants that traditional medicine practitioners and the general population use to manage diabetes. This study was interested in inventorying these plants and presenting their ethnobotanical and ethnomedicinal profiles. The study reported 71 taxa used in 86 recipes to treat diabetes mellitus.
These 71 plants are primarily trees from the Fabaceae family, endemic to tropical Africa, and are named mainly in Kikongo (Table 1 and Figures 2 and 3). These results are in agreement with the literature. Indeed, Fabaceae is reported to be the most prominent family of trees in Africa’s tropical and dry forests [113]. This importance of Fabaceae is observed both within the plant kingdom [114] and within African medicinal plants in particular [115]. The numerical predominance of Fabaceae in sub-Saharan Africa has been attributed to their ability to scavenge atmospheric nitrogen, allowing them to grow in nutrient-poor and rich soils [116]. In our experimental framework, no accessible study has addressed the question of the preponderance of a botanical family over all the taxa used in traditional medicine in Kinshasa. Such a study would be desirable. However, the analysis of some ethnobotanical studies in Kinshasa shows that the Fabaceae constitutes one of the three most evoked families. Some previous studies in the Kinshasa reported the predominance of certain botanical families over others. As an illustration, a study on ethnopharmacological surveys of plants used in female intimate baths [116] mentioned the predominance of Rubiaceae (37%) and Fabaceae (11%). In contrast, another study on plants used to treat the symptoms of tuberculosis [117] reported the predominance of Fabaceae, Apocynaceae, and Lamiaceae, each with an FCR of 8.3%. In the same way, FCR = 47% and FCR = 12.7% were found respectively in the families of Rubiaceae and Fabaceae for a study on plants sold on the Kinshasa market [118], and an FCR of 7.7% in Fabaceae, Rubiaceae and Zingiberaceae for medicinal plants of the Lukunga district [119]. There are also studies done on the management of diabetes in the region. This is the case of a survey carried out in the Kimbanseke and Selembao communes: n = 21, Rubiaceae FCR = 33.3%, Fabaceae FCR = 19.1% [9], or the study in Kwango and Kinshasa: n = 68, Fabaceae FCR = 19.1% [8]. It would nevertheless be interesting to carry out an inventory of plants known to be medicinal in the DRC to highlight their specific ethnobotanical characteristics. This concern has not been the subject of this study, which nevertheless contributes to highlighting the need for such data.
The literature review carried out on 71 taxa shows that it is possible to group the plants into 3 classes: species for which no ethnobotanical or pharmacological information is available (class α), those with ethnopharmacological use without any scientific evidence (class δ), and species for which scientific evidence is available (class β) (Figure 1(d)). The fact that 70% of species found are reported to be antihyperglycemic in vivo models (using rats) reinforces the credibility of the information obtained from our study. It suggests a high probability of finding those with antihyperglycemic activity among the 30% of the remaining species.
The 6 plant species whose antidiabetic ethnomedicinal knowledge is reported for the first time during this study and for which no pharmacological antidiabetic activity survey is registered in the accessible literature to date are Acalypha paniculata, Baphia capparidifolia, Chromolaena corymbosa, Crotalaria medicaginea, Platymitra arborea, and Rauvolfia mannii. On the other hand, 11 medicinal plants, including Antidesma venosum, Costus phyllocephalus, Crinum ornatum, Diospyros heudelotii, Gladiolus gregarius, Lippia multiflora, Millettia drastica, Mitragyna stipulosa, Palisota schweinfurthii, Tephrosia vogelii, and Terminalia mollis, have been quoted in traditional medicine as treating diabetes mellitus in various countries. However, they are not yet scientifically validated (Table 1). In both cases, these plants constitute a great richness in the database for future preclinical investigations. However, the antidiabetic evaluation of Costus lucanusianus has been validated after several research attempts; no study has confirmed the local antidiabetic use of this plant.
Previous studies on traditional healers and medicinal plant vendors have identified some species used in Kinshasa city (Figure 7) to manage diabetes mellitus. Thus, Abelmoschus esculentus, Albizia adianthifolia, Alchornea cordifolia, Azadirachta indica, Bridelia ferruginea, Catharanthus roseus, Costus phyllocephalus, Cymbopogon citratus, Dysphania ambrosioides, Erythrina abyssinica, Ficus benghalensis, Gymnanthemum amygdalinum, Lippia multiflora, Morinda lucida, Nauclea latifolia, Phaseolus vulgaris, Phyllanthus niruri, Psidium guajava, Schwenckia americana, Senna alata [8], Annona senegalensis [8, 73, 119], Garcinia kola [73, 119] Gladiolus gregarius [119], Monodora myristica [120], and Persea americana [8, 118] have been cited.

Concerning ethnobotanical indexes, it should be noted that for an ethnobotanical study targeted towards a specific pathology, the relative frequency of citations corresponds to the consensual citation index of the plant, which translates the consensus reached around a particular plant species on targeted use. There is a higher consensus for the 11 taxa not studied on using Tephrosia vogelii leaves (CIp = 0.08) as an antidiabetic than any other species of the same category (Table 2). This precedence which may result from this ethnobotanical index (CIp) is, however, to be put into perspective by the fact that a single category only told the plant of the subjects consulted, namely, people with diabetes, unlike, in particular, the leaves of the Crinum ornatumZ species which present the weakness of ethnobotanical index whose value is twice lower than that of Tephrosia vogelii (CIp = 0.04), but whose strength lies in the fact that it was informed simultaneously by the three sources: diabetics, TMPs, and herbalists, which reinforces the consensus around its use as an antidiabetic in the study environment.
The study population uses three plant species most because of their highest UV of the series (0.12). These are Annona senegalensis, Erythrina abyssinica, and Nauclea latifolia (Table 2). This justifies their lower usual value of 0.02 (Table 2). All of its plants belong to the taxa category, whose ethnobotanical and pharmacological knowledge of diabetes has previously been reported. Therefore, the most common plants in the study would not be the most interesting in the context of this study, which aims to enhance the ethnopharmacological knowledge of Kinshasa. If, in most cases, each plant inventoried during this study is also known and used by our resource persons for at least one other pathology, it should be emphasized that this is not the case with Tephrosia vogeliiX and Platymitra arboreaY for which our informants only use it in the management of diabetes mellitus.
Malaria (FCR = 12%), cough (FCR = 9%), and diarrhea (FCR = 7%) constituted the 3 first pathologies, apart from diabetes, for which the plants listed during this study are used. These pathologies are among the 10 most deadly pathologies in the DRC [121], suggesting that traditional medicine from Kinshasa will likely contribute to managing specific pathologies of concern in DR Congo. It would be interesting to carry out studies to validate this ethnopharmacological knowledge to rightly understand the assistance of conventional medicine from Kinshasa in managing pathologies such as diabetes.
The 71 inventoried plant taxa made it possible to list 86 antidiabetic recipes, 80 of which use a single plant (Table 2), and 6 recipes combine 2 simultaneous medicinal plants (Table 3). Nonetheless, considering all previous ethnopharmacological studies conducted in DR Congo [8, 15, 38, 48], this one reports antidiabetic recipes of the plants used in association for the first time.
Of all the antidiabetic recipes using a single plant and reported during this study, the leaf and the decoction constitute the most used organ and mode of preparation. These results (Figure 8) are in agreement with the studies carried out on the plants of Kisangani: leaf (57.6%), decoction (78.8%), and n = 33 [38], and plants of DR Congo gathered in a review article, leaves 39.2%, decoction: 60.5%, and n = 213 [52]. Some disparities are nevertheless observed in a study in South Katanga, where the decoction (62.5%) was found to be the primary way of preparation and the root was found to be the most used organ: 41.3%, n = 95 [15]. The same is true in the study carried out simultaneously in Kwango and Kinshasa (n = 68), where the leaves at 65% proved to be the most used organ and maceration with 63% was found to be the most common mode of preparation solicited [8]. The herbal antidiabetic recipes used in Kinshasa align with the national trend, unlike those used in Lubumbashi, the country’s second-largest city.

(a)

(b)
According to the subjects consulted, using a decoction would aim to extract and activate the active ingredient. This is a mixed idea, especially since this extraction procedure is more beneficial than harmful. Indeed, as much as it could facilitate the release of certain active principles often present in the plant in the form of glucosides, it could not only release specific toxic secondary metabolites such as cyanogenic glycosides [122] but also deteriorate the active ingredient when the latter is thermolabile [123–125]. Therefore, this practice remains to be assessed case-by-case, and only experimental work could determine its fair value.
About the various correlations established on specific characteristics of the subjects consulted (Figure 8), this study shows that the older the diabetic issues, the longer they have lived with diabetes, which suggests that most of the subjects consulted would be diabetics from type 1. Similarly, the study reveals that the oldest TMPs, the most experienced and skilled, see more patients. This supposes that traditional medicine requires time to retain patients who would be more confident towards the most experienced. This attitude is also observed in conventional medicine, where rationality would like the patient to be reassured by the doctor’s experience before being consulted, especially for specific pathologies.
5. Conclusion
Several plants are used in Kinshasa to manage diabetes mellitus by herbalists as well as by people with diabetes and traditional medicine practitioners. This study highlights not only antidiabetic recipes but also plants reported only in Kinshasa for the management of diabetes. It also evokes their usual values in this environment. It opens the way for a subsequent study to validate the antidiabetic use, particularly of particular species of the environment and those of specific special interest there.
Data Availability
The data used in this study are included within the article.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Authors’ Contributions
BCV designed, formatted, and analyzed the data, developed methods, and drafted the first manuscript. FMK and MMH performed study checks. BCV, ONP, and FMK were responsible for manuscript corrections.
Acknowledgments
The authors thank the Herbarium of Kisantu, the National Agricultural Study and Research Institute of Kinshasa, for identifying plant species and the TMPs, herbalists, and people with diabetes in Kinshasa for the acquisition of ethnobotanical information on plants. The late Professor Joseph Moswa Lokonda is greatly thanked for initiating this project.