1. Introduction
Diabetes increasingly represents a major global concern, especially in developing countries and particularly in Sub-Saharan Africa. According to WHO, worldwide prevalence of diabetes has almost doubled since 1980, from 4.7% to 8.5% in 2014 in the adult population [1] . According to the International Diabetes Federation (IDF), diabetes is one of the six leading causes of death worldwide [2] . According to Djrolo F et al., the prevalence of diabetes in Cotonou was estimated at 3.3% in 2002 and 4.6% in 2007 [3] . Diabetes is a chronic disease with serious complications, which requires expensive long-term treatment. Its treatment considers dietary measures and industrial treatment. The use of plants medicinal use by diabetics is experiencing exponential growth. The percentage of diabetic patients using medicinal plants increases and we note a significant diversification of plants with the discovery new plants.
In Benin, traditional medicine based on plant extracts and natural active ingredients is used in the treatment of diabetes. Thus, some diabetic use medicinal plants either exclusively or in a combine way with medical treatment. Studies have revealed that many diabetic patients use medicinal plants for controlling their blood sugar levels. In 2003 in Conakry, out of 397 diabetic patients, 33% used plants to control their blood sugar [4] . In Morocco in 2015, out of 509 diabetic patients, 54% used with medicinal plants [5] . In 2008 in Algeria, out of 634 diabetic patients 62%regularly used medicinal plants [6] .
2. Method
This was a descriptive and analytical cross-sectional study which took place over a 3 months period from June 20 to September 16, 2022. Population was diabetic followed or hospitalized in the university clinic of Endocrinology, Metabolism and Nutrition of CNHU-HKM during the period and who gave informed consent. Excluded are patients who have difficulty or unable to indicate the medicinal plants used. The dependent variables are plant use and the independent variables are sociodemographic characteristics: age; gender; housing; work ; level of education; socio-economic level; duration of diabetes; chronic complications; glycated hemoglobin level; family history of diabetes; treatment used; high blood pressure and dyslipidemia; plants parts used (stems, roots, leaves, seeds, fruits); method of administration (decoction, maceration, infusion, powder); dosage; the exclusive or association nature; reasons for use; side effects ; duration of phytotherapy. Data collection was undertaken using a standardized inquiry form, through a direct question-answer interview on the use of medicinal plants.
Data analysis was done using the software: Epi info 7.1, Epi-data 3.1 and Excel 2016. Data entry was done using Epi info 7.1 software. Verification and clearance were carried out with STATA MP14. The comparison of frequencies and percentages (represented by the qualitative variables) was carried out with the Pearson’s Chi2 test or the Fisher exact test (if the Chi2 test condition was not respected); the comparison of the means (represented by the qualitative variables) was made with the Student’s test. A p < 0.05 was considered statistically significant. In multivariate analysis, the logistic regression model was used. Logistic regression made possible to measure the association between the use of hypoglycaemic medicinal plants and its determinants. The choice of variables integrated into the logistic regression model is based on the statistical association between the variable and the event, measured by the odds ratio.
3. Results
We identified 173 diabetic patients. Sixty-five patients consume medicinal plants. The prevalence is 37.6% among diabetic using hypoglycaemic medicinal plants in the University Clinic of Endocrinology Metabolism Nutrition of the CNHU-HKM.
Sociodemographic characteristics of the population.
The average age of the patients is 59 ± 1.43 years. The median age is 61 years and the extremes are 31 and 75 years. Most patients were female with a sex ratio (male/female) of 0.69.
Diabetic patients living in urban areas were the majority (64.7%).
Civil servants were the most represented (59.5%) followed by traders (23.1%) and housewives (7.5%). The majority of patients were educated (97.7%). Majority of patients (83.8%) had an average socioeconomic level. The bride and groom were the most represented (80.4%). Factors related to medicinal plants.
Medicinal plants used:
Sixty-five patients (37.57%) used medicinal plants. Twenty-nine (29) species have been recorded, grouped into 20 botanical families. The most represented botanical families are Annonaceae and Fabaceae. The most frequently species used are “Hlenwé”, “Mangatin”, “Gninsikin”.
The scientific names, families belonging, vernacular names and the frequencies of use are gathered in the Table 1.
Most frequently used species (Figures 1-3).
Parts of the plant: leaves are the part most used by patients in their treatment 89.2%.
Reasons for use: sometimes-complicated reasons for use are numerous. The experience of other patients convinced most patients (84.6%) to use anti-hyperglycaemic plants.
Association with oral treatment: 61/65 (96.8%) users of medicinal plants add it with oral treatment. For almost one in two patients, treatment lasts less than a year.
Method of use: plants are much more used in the form of herbal tea and prepared by infusion.
Table 1. Scientific names of the different plants used, families they belong to.
Figure 1. Phyllanthus amarus (hlenwé): Euphorbiaceae.
Figure 2. Mangifera indica (mangatin: Anacardiaceae).
Figure 3. Momordia charantia (gninsikin): Annonaceae.
Dosage: plants used in the form of herbal tea were estimated using two forms of glasses. The glass locally called “talokpévi” with a content of 40 ml and the bamboo glass 300 ml. No adverse effects related to medicinal plants have been reported by users.
Satisfaction: less more than half of users are satisfied with the treatment of their diabetes with medicinal plants (58.5%).
Complications in patients using medicinal plants: chronic complications of diabetes are observed mainly in patients consuming hypoglycemic medicinal plants: diabetic nephropathy 4.6% versus 1.7%, diabetic retinopathy 4.6% versus 1.7%, stroke 1.2% versus 0.6, diabetic neuropathy 41.2% versus 27%.
Analytic study (Table 2).
The use of medicinal plants is significantly associated with family history of diabetes, personal history of hypertension, glycemic control and the occurrence of chronic complications of diabetes.
4. Discussion
One hundred and seventy-three (173) diabetic were collected. Population were larger in studies carried out by Selihi Z et al. (509) in 2015 in Morocco [5] and by Allali H et al. (634) in 2008 in Algeria [6] . This can be explained by the fact that our study was carried out in a hospital environment and that the prevalence of diabetes in Morocco and Algeria is higher than that of Benin because of a larger population. Most of patients (94.8%) were at least 40 years. These results
Table 2. Factors associated with the use of medicinal plants in the treatment of diabetes by CNHU patients.
agree with those of Allali H et al. [6] and those of Selihi Z et al. [5] , who found 97% and 83.3% respectively. The predominance is female in our study. It is similar to the study carried out by Errajrari A et al. 62% in Morocco [7] . This is explained by the susceptibility of women to developing diabetes, linked to several factors such as a sedentary lifestyle, stress and obesity. The majority of patients had poorly controlled diabetes (60.1%), as in the study by Selihi Z et al. [5] . This could be explained by the less observance in diet means and therapy.
Sixty-five patients or 37.7% used medicinal plants. These results are different from those obtained in the study of Allali H et al. (62%) [6] and that of Errajrari A et al. (52%) [7] . This study allowed us to identify twenty-nine plants (29) belonging to 20 botanical families. Other ethnobotanical studies have revealed similar diversity of antidiabetic plants; Ziyyat A et al. identified 41 plants belonging to 36 families [8] ; Allali H et al. identified 58 plants belonging to 38 botanical families [6] . It emerged from our study that the species most used by diabetic patients at CNHU-HKM are: Annonaceae and Fabaceae. These results have some similarities with some previous work. Thus in the study by Gbekley E et al. the most represented families were Fabaceae, Euphorbiaceae and Compositae [9] . In the study by Karou S et al., Fabaceae were the most represented [8] .
The leaves of the plants are the most used parts (89.2%) in the different preparations. Baldé N et al. had the same results [4] , as in the study by Gbekley E et al. [9] . This could be explained by the fact that the leaves are easier to handle.
The positive experience of a third person convinced most patients (84.6%). Many studies have had similar results such as those of Errajari A et al. 80% (7) and Baldé N et al. 78% [4] .
Plants are mainly used in combination with drug treatment (96.8%) as in the study by Errajraji A et al. (7) (69%). This could be linked to the fact that patients use plants to accelerate the action of medications fearing the occurrence of hyperglycemia.
The recipes are mainly consumed in the form of herbal teas. The preparation of recipes is mainly done by infusion (64.6%). These results are like the studies of Selihi et al. [5] . On the other hand, Gbekley E et al. found that the preparations are mainly made by decoction [9] . No adverse effects were reported by patients. Which is not the case with Baldé N et al., 18% of adverse effects [4] Pharmaco-toxicological studies must be carried out to determine the potential adverse effects induced by these different plants.
Patients who consume plant-based recipes in combination with drug treatment mostly have poorly controlled diabetes compared to patients only taking OADs, as in the study by Selihi Z et al. [5] .
This could be linked to the priority gave to plants over medical treatment, through patients who take plants adding with OADs, which is often find expensive. Chronic complications of diabetes were present in 95.4% of medicinal plant consumers and in 31.5% of non-consumers. These results are like those of Selihi Z et al. [5] . In their study, complications were found in 75.1% of plant consumers and in 50.8% of non-consumers.
5. Conclusion
13