Prevalence and Risk Factors for the Metabolic Syndrome among Patients with Epilepsy Attending a Neuropsychiatric Hospital in Kigali, Rwanda

International Journal of Medical Research & Health Sciences (IJMRHS)
ISSN: 2319-5886 Indexed in: ESCI (Thomson Reuters)

Research - International Journal of Medical Research & Health Sciences ( 2021) Volume 10, Issue 3

Prevalence and Risk Factors for the Metabolic Syndrome among Patients with Epilepsy Attending a Neuropsychiatric Hospital in Kigali, Rwanda

Francois Xavier Ndayambaje1*, Jean Bosco Gahutu2, Simon Peter Rugera1 and Bernard Natukunda1
1Department of Medical Laboratory Sciences, Mbarara University of Science and Technology (MUST), Mbarara, Uganda
2Department of Physiology, Faculty of Medicine, University of Rwanda (UR), Kigali, Rwanda
*Corresponding Author:
Francois Xavier Ndayambaje, Department of Medical Laboratory Sciences, Mbarara University of Science and Technology (MUST), Mbarara, Uganda, Tel: 256-4854-21373, Email:

Received: 23-Feb-2021 Accepted Date: Mar 23, 2021 ; Published: 30-Mar-2021


Background: Cardiovascular mortality and morbidity are more frequent in people with epilepsy than in the general population. The explanation for this may be the change in biochemical components due to the use of anti-epileptic drugs. We conducted a study to determine the prevalence and risk factors for Metabolic Syndrome (MetS) among Rwandan adults with epilepsy emphasizing the respective anti-epileptic drugs that the patients were receiving. Methods: This was a cross-sectional study conducted from December 2018 to December 2019 at Ndera Neuropsychiatric Hospital in Kigali, Rwanda. Consenting adult Rwandan patients with epilepsy, who had been on anti-epileptic drugs for at least two years, were recruited into the study. Participants had their anthropometric measurements taken and their fasting blood glucose plus lipids assayed. Using a data collection form, patients had their demographic and clinical characteristics recorded. Results: There were 1076 participants (male-to-female ratio, 1.4; age range, 42). Using the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) criteria, there were 329 individuals with MetS giving an overall prevalence rate of 30.6%. Significant risk factors for MetS included use of valproic acid (p=0.007), a sedentary lifestyle (p=0.025), waist circumference >102cm (p=0.001), high triglycerides (p=0.001), high blood pressure (p=0.001), and fasting blood glucose >6.1mmol (p=0.001). Conclusion: MetS is highly prevalent among Rwandan patients with epilepsy. Therefore, local physicians are advised to carefully select the type of anti-epileptic medication administered and to regularly request anthropometric as well as laboratory checkups for such patients to predict a diagnosis of MetS and the complications thereof.


Metabolic syndrome, Epilepsy, Prevalence, Anti-epileptic drugs, Rwanda


Metabolic Syndrome (MetS), a group of metabolic risk factors including hyperglycemia, dyslipidemia (elevated triglycerides, reduced high-density lipoproteins), elevated blood pressure, and obesity is increasingly becoming a global public health concern [1]. In the general population, westernization characterized consumption of foods with fat and sugar content combined with less physical exercises are the leading causes of this syndrome, which is a great predictor of cardiovascular disease [2,3].

Compared with the general population, people with epilepsy have elevated rates of cardiovascular mortality and morbidity, which may be driven by metabolic changes [4,5]. Studies have found that this patient group also has an increased risk of developing metabolic syndrome and obesity, both of which are risk factors for vascular disorders [6-9].

The connection between metabolic syndrome and epilepsy has emerged as a public health question of importance to both specialists and primary care practitioners [10]. Studies that investigated the presence of MetS in patients with epilepsy yielded contradicting results. Some of these studies targeted only specific groups of populations, such as females, young adults, children [6,11,12]. Other researchers investigated different metabolic side effects of Anti- Epileptic Drugs (AEDs) on different organs, tissues, and biochemical components of the human body such as bone and calcium metabolism, glucose metabolism, mental health, blood components, and lipid metabolism [13-17].

In the present study, we estimated the prevalence of MetS in Rwandan Patients with Epilepsy in a cross-sectional survey conducted from December 2018 and December 2019 at Ndera Neuropsychiatric Hospital in Kigali, Rwanda. Data on lifestyle and metabolic risk factors were collected on adult patients with epilepsy. We used the National Cholesterol Education Program (NCEP) Adult Panel III (ATP III) criteria to confirm MetS [18].



The study was carried out in the Out-patients’ Department of Ndera Neuro-Psychiatric Hospital, in Kigali, Rwanda. The study duration was 12 months from December 2018 to December 2019 during which the prevalence of MetS and its components among patients with epilepsy was determined. Study participants, who were adult volunteers having epilepsy and aged between 18 and 60 years, were consecutively enrolled. Participating individuals were on treatment for epilepsy using a single type of anti-epileptic drug.

We excluded patients with malignancies, pregnancy, mental or legal incapacity that prevented them from giving valid consent or hindered them from full participation in the study process or strict adherence to study protocol.

Data Collection

Data were collected from patients who agreed to participate in the study. They signed a written consent form following a detailed explanation of the study. Participants were then interviewed and examined in a special room for privacy and confidentiality. Demographic details (names, sex, age, and marital status), risk factors (diabetes, hypertension, physical activity, and smoking), and anthropometric measurements (height, weight, and waist circumference) were collected using data collection forms. The interview was carried out by the Principal Investigator (PI) and one Research Assistant (RA) who had been trained before commencing the study.

Anthropometric Measurements

Waist circumference was taken without outer clothing, using a non-stretchable tape measure at the level of the uppermost edge of the hip bone on the abdomen with the tape parallel to the ground and recorded to the nearest 0.5 centimeters. Blood Pressure (BP) was taken from the arm (brachial artery) for all respondents on the first encounter by using a digital sphygmomanometer. BP measurements were performed in the sitting position with the arm supported at heart level and repeated after 5 minutes; the average of the two measurements was taken as the correct BP for the individual. To calculate the Body Mass Index (BMI), subjects were weighed (in kilograms) on a weighing scale while fully dressed but without shoes. Height (in meters) was measured without shoes by using a stadiometer. The BMI was calculated as the weight divided by the height squared (kg/m2).

Biochemical Tests

Blood samples were collected with BD Vacutainer® venous blood collection tubes. Five milliliters of venous blood were taken from the antecubital fossa and placed in empty sterile vacutainer tubes. The blood samples were separated by centrifugation at 1000 × g for 15 min and the supernatant serum was collected. Serum levels of lipids Triglycerides (TG), Total Cholesterol (TC), High-Density Lipoprotein Cholesterol (HDL-C), blood glucose were measured using commercially available kits (Roche System Reagents) by a Cobas C 311 autoanalyzer using appropriate quality control measures for the reagents and autoanalyzer.

Statistical Analysis

Data were entered and analyzed using the Statistical Package for Social Sciences (SPSS) version 24.0 software (SPSS Inc. Chicago, IL). Means of numeric variables were compared between groups by the student’s t-test. Proportions were compared using the Chi-square test. Multivariate logistic regression was used to determine the risk factors associated with MetS in patients with epilepsy. For tests and models, a p-value of less than 0.05 was taken for statistical significance.

Ethical Clearance

Before the commencement of the study, ethical clearance was sought from Mbarara University of Science and Technology-Research Ethics Committee (MUST-REC) in Mbarara, Uganda. Ethical approval was then secured from both Rwanda National Health Research Committee (NHRC) and Rwanda National Ethics Committee (RNEC). We conducted this study following the guidelines of the Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects. Information concerning the study was given to all study participants before the interviews and sample collection. A written and duly signed consent was obtained from all participants. Confidentiality was ensured for all individuals who participated in the study by removing all identifiers at the time of data entry. The participation was fully voluntary devoid of any kind of coercion and no participant was paid for their involvement in the study.


During the study period, 1076 adult patients with epilepsy meeting the inclusion criteria were carefully selected to participate in this study. They were 629 males, 58.5%, and 447 females, 41.5%. The mean age of participants was 40.22 ± 10.37 that of males was 40.20.04 ± 10.34 and of females was 40.24.04 ± 10.42. The distribution of age was 33.3% (n=358) between 18-35 years, 34.8% (n=374) between 36-45 years, 32.0% (n=344) between 46-60 years. Valproate treated patients were 328 (30.5%), Carbamazepine; 207 (19.2%), Leveracetam; 145 (13.5%), Phenytoin; 130 (12.1%), Phenobarbital; 135(12.5%), Clonazepam; 95(8.8%) and Topiramate treated patients were 36 (3.3%). Other clinical characteristics of participants are presented in Table 1 and Table 2.

Table 1 Clinical characteristic of subjects with epilepsy with and without metabolic syndrome

Total number of subjects (n=1076) Participants without metabolic syndrome [n=747 (69.4%)] Participants with metabolic syndrome
[n=329 (30.6%)]
Males 629 (58.5%) 440 (40.9%) 189 (17.6%) 0.655
Females 447 (41.5%) 307 (28.5%) 140 (13.0%)
Age groups N (%)
18-35 358 (33.3%) 261 (24.3%) 97 (9.0%) 0.214
36-45 374 (34.8%) 254 (23.6%) 120 (11.2%)
46-60 344 (32.0%) 252 (21.6%) 329 (30.6%)
Antiepileptic drugs N (%)
Valproic acid 428 (30.5 %) 209 (19.4%) 119 (11.1%) 0.007
Carbamazepine 207 (19.2%) 140 (13.4%) 67 (6.2%) 0.534
Leveracetam 145 (13.5%) 104 (9.7%) 41 (3.8%) 0.518
Phenytoin 130 (12.1%) 97 (9.0%) 33 (3.1%) 0.799
Phenobarbital 135 (12.5%) 95 (8.8%) 40(3.7%) 0.171
Clonazepam 95 (8.8%) 74(6.9%) 21 (2.0%) 0.061
Topiramate 36 (3.3%) 28 (2.6%) 8 (0.7%) 0.268
Metabolic parameters (mean ± SD )
FBG (mmol/L) 5.75 ± 0.70 5.63 ± 0.65 6.01 ± 0.74 0.001
Total cholesterol (mmol/L) 3.78 ± 0.88 3.75 ± 0.82 3.84 ± 0.98 0.141
LDLc (mmol/L) 3.20 ± 1.18 3.09 ± 1.11 3.44 ± 1.28 0.001
HDLc (mmol/L) 1.15 ± 0.20 1.18 ± 0.19 1.09 ± 0.20 0.001
Triglycerides (mmol/L) 1.34 ± 0.58 1.25 ± 0.53 1.54 ± 0.62 0.001
Component of metabolic syndrome
High abdominal obesity 378 (35.2%) 182 (16.9%) 196 (18.3%) 0.001
High blood pressure 316 (29.4%) 90 (8.4%) 101 (9.5%) 0.001
Diabetes mellitus 155 (14.4%) 84 (7.8%) 71 (6.6%) 0.001
Low HDLc 283 (26.3%) 144 (13.4) 139 (12.9%) 0.001
High triglycerides 303 (28.2%) 141 (13.1%) 162 (15.1%) 0.001

Table 2 Comparison of clinical and metabolic parameters between males and females

Males [N=629 (58.5%) Females [N=447 (41.5%)] p value
Age groups N (%)
18-35 214 (19.9%) 144 (13.4%) 0.214
36-45 215 (20.0%) 159 (14.8%)
46-60 200 (18.6%) 144 (13.4%)
Ages in years, (mean ± SD ) 40.20 ± 10.34 40.24 ± 10.42
Antiepileptic drugs N (%)
Valproic acid 199 (18.5%) 129 (12.0%)
Carbazepine 126 (11.7%) 81 (7.5%)
Leveracetam 81 (7.5%) 64 (5.9%)
Phenytoin 77 (7.2%) 53 (4.9)
Phenobarbital 72 (6.7%) 63 (5.9%)
Clonazepam 58 (5.4%) 37 (3.4%)
Topiramate 16 (1.5%) 20 (1.9%)
Metabolic parameters (mean ± SD )
FBG (mmol/L) 5.74 ± 0.69 5.75 ± 0.72 0.805
Total cholesterol (mmol/L) 3.72 ± 0.85 3.85 + 0.90 0.017
LDLc (mmol/L) 3.19 ± 1.15 3.20 + 1.21 0.89
HDLc (mmol/L) 1.15 ± 0.20 1.16 ± 0.20 0.666
Triglycerides (mmol/L) 1.34 ± 0.56 1.33 ± 0.60 0.767
Component of metabolic syndrome
High abdominal obesity 215 (20%) 163 (15.1%) 0.885
High blood pressure 112 (10.4%) 79 (7.3%) 0.763
Diabetes mellitus 97 (0.9%) 58 (5.4%) 0.26
Low HDLc 166 (15.4%) 117 (10.9% 0.937
High triglycerides 183 (17.0%) 120 (11.2%) 0.419

The prevalence of metabolic syndrome in people with epilepsy was 30.6% (329 subjects) and out of these, the number of males was 189 (57.4%) whereas the number of females was 140 (42.6%). The prevalence of metabolic syndrome weighted by age group was 9.0% (97) among the age group of 18-35 years old, 11.2% (120) among the age group of 36-45 years old, and 30.6% (329) among the age group of 46-60 years old. Regarding individual components of MetS, 378 subjects (35.2%) had high abdominal obesity, 316 subjects (29.4%) had high blood pressure, 155 subjects (14.4%) had diabetes mellitus, 283 subjects (26.3%) had low HDLc, and 303 subjects (28.2%) had high level of serum triglycerides.

Regarding the occurrence of MetS in patients under different anti-epileptic drugs, we found that valproic acid was significantly associated with MetS (p=0.007). Indeed, among 328 valproic acid-treated patients, 119 (36.3%) of them had MetS. We did not find any statistically significant differences between the other AEDs and MetS. The distribution of MetS in other AEDs treated patients is found in Table 3.

Table 3 Frequency of metabolic syndrome in patients under different anti-epileptic drugs

Metabolic syndrome present
No Yes Total
Valproic acid 209 (63.7%) 119 (36.3%) 328
Carbamazepine 140 (67.6%) 67 (32.4) 207
Levetiracetam 104 (71.1%) 41 (28.3%) 145
Phenytoin 95 (70.4%) 40 (29.6%) 135
Phenobarbital 97 (74.6%) 33 (25.4%) 130
Clonozepam 74 (77.9%) 21 (22.1%) 95
Topiramate 28 (77.8%) 8 (22.2%) 36

In this study, multivariate logistic regression analysis did not reveal any significant correlation between the occurrence of MetS in patients with epilepsy and some clinical characteristic such as sex, age gender, family history of diabetes, and family history of hypertension (Table 4).

Table 4 Multivariate analysis of risk factors associated with metabolic syndrome

Component Odd Ratio (OR) 95% confidence interval
Lower Upper p-value
Gender 1.272 0.933 1.734 0.128
Age 1.001 0.986 1.016 0.903
Waist circumference >102 cm 1.067 1.052 1.016 0.001
Fasting blood glucose >6.1 mmol 1.81 0.866 1.809 0.001
HDLc <1.03 mol 0.066 0.029 0.148 0.001
Triglyceride >130 mol 2.797 2.122 3.685 0.001
Systolic blood pressure >130 mmHg 1.032 1.018 1.046 0.001
Diastolic blood pressure >85 mmHg 1.102 1.071 1.232 0.001
Family history of diabetes 0.978 0.029 0.148 0.906
Family history of hypertension 1.252 0.866 1.809 0.232

On the other hand, a significant correlation was found between the occurrence of MetS and quite number of components such as waist circumference (OR=1.067; p=0.001; 95% CI=1.052-1.016), fasting blood glucose (OR=1.810; p=0.001; 95% CI=0.866-1.809), serum triglyceride (OR=2.797; p=0.001; 95% CI=2.122-3.685), systolic blood pressure (OR=1.32; p=0.001; 95% CI=1.018-1.046) and diastolic blood pressure (OR=1.102; p=0.001; 95% CI=1.071-1.232).


To the best of our knowledge, this was the first published study that sought to determine the prevalence of MetS and cardiovascular risk factors among Rwandan patients with epilepsy. In this study, MetS were defined as a constellation of interconnected physiological, biochemical, clinical, and metabolic factors that directly increase the risk of atherosclerotic cardiovascular disease, type 2 diabetes mellitus, and all-cause mortality [19]. In Rwanda, a country with approximately 0.7% patients with epilepsy, the prevalence of MetS among these individuals was found to be 30.6% [20]. This was significantly higher than for other prevalence rates reported from surveys conducted among the general population in Africa. For instance, using the ATP III criteria, the prevalence of MetS in Ethiopia has been reported as 10.0% in men and 16.2% in women [21]. Regarding other countries in the African region, the prevalence of MetS was found to be as follows: South Africa, 18.5%; Ghana, 15.0%; Nigeria, 25.2% and Cameroon, 1.8%-1.9% [22-25]. However, some research findings from overseas among the general population indicated higher prevalence rates of MetS than that of Rwanda. For instance, a community study from Eastern India reported an overall prevalence of 33.5% [26]. A study conducted in Croatia among obese children and adolescents presenting with some components of MetS reported a prevalence rate of 30.3% which is comparable to our present study. Other studies which reported higher prevalence rates of MetS than ours include a study from Brazil which reported a prevalence ranging from 55.6%-74% [27]. Some studies that were looking at the prevalence of metabolic syndrome in certain categories of patients with epilepsy found closely related prevalence. In young adults Indian with epilepsy the MetS prevalence has been reported as 29.5%, while the MetS prevalence of 25.8% has been reported from Estonian in adults patients with epilepsy who were under VA treatment [6,28]. Other studies reported a higher prevalence of MetS: 47% among Chinese obese patients with epilepsy on VA and 43.5% in Italian overweight epileptic patients treated with VA [12,29].

In our study, all age groups had different frequencies of MetS. However, this difference did not reach statistical significance (9.0% in the 18-35 years old group, 11.2% in the 36-45 years old group, and 30.6% in the 46-60 years old group, p=0.214). In this regard, if we compare our results with results from other studies conducted among patients with epilepsy elsewhere, we agree with Jiaji Fang who also did not find the significant difference of MetS across age groups [29]. However, our results disagree with the results of Nair who found a significantly higher prevalence of MetS in the oldest age group [6].

Men and women were all affected though with different proportions [189 subjects (17.6%) versus 140 subjects (13.0%), p=0.655)]. The interpretation is that AEDs and other risk factors can trigger MetS in people with epilepsy regardless of their gender and age group. Our results are therefore concordant with those published by Verroti, et al. Nair, et al. and Jiaji Fang, et al who also found no difference in the occurrence of MetS between males and females [6,12,29]. However, Prasad reported significant gender differences with a higher prevalence of MetS in females compared to males (p=0.00) [26].

We observed that across all age groups, Valproic Acid (VA) treated patients are more prone to develop MetS than other EADs treated patients (11.1%, p=0.007). The least prevalence of MetS was found in topiramate-treated patients (0.7%, p=0.268).

This confirms earlier reports from several studies which emphasize that the use of VA for treating epilepsy is linked with a high probability of experiencing metabolic and hormonal disturbances [22,23,30-36].

Regarding individual components of MetS, we found that the most prevalent one was high abdominal obesity with diabetes mellitus being the least prevalent. However, in some general population studies, the most frequent MetS component was low serum HDL cholesterol in women and elevated High Blood Pressure (HBP) in men [22,25]. Similar results have been found by other investigators and abdominal obesity was significantly correlated with MetS (p=0.001) [37].

Increased abdominal obesity has constantly been reported by several researchers working on studies of patients with epilepsy who are under AEDs medication especially VA and current evidence supports that abdominal obesity is predictive of metabolic risk factors [10-12,38]. In our study, HBP was the second most MetS individual component we found. According to Jiaji Fang, et al. HBP was also found to be very common in Chinese obese patients with epilepsy on VA. VA causes MetS in some patients but not all. Genetic factors and several molecular pathways of energy homeostasis might influence the occurrence of MetS [12]. Our study showed that the prevalence of hyperglycemia was 14.4% which suggests that hyperglycemia is higher among patients with epilepsy than in the general population. The Rwanda Non-communicable Diseases Risk Factors Report released in 2015 pointed out that in the general population of Rwanda, impaired fasting glucose levels and raised blood glucose affect 3.1% of the population in Rwanda [40].

Hyperglycemia may be due to drugs that are known to cause hyperglycemia like VA or due to a combination of MetS risk factors commonly found among patients with epilepsy like obesity and lower exercise capacity that make body cells less sensitive or resistant to insulin.


Results of our study form an informative tool for clinicians and other policymakers in the management of epilepsy in Rwanda. Understanding the prevalence of metabolic syndrome and other metabolic abnormalities among patients with epilepsy is immensely useful to practitioners to help affected patients undergo risk-reducing interventions. This also lays a good foundation for further research in related areas such as the interactions of AEDs and other medications, markers of genetic susceptibility of MetS in patients with epilepsy.



All authors played a role in the study design. FXN and JBG conducted the study and collected data. FXN, BN, and SPR analyzed and interpreted the data. FXN, BN, and SPR wrote the manuscript and approved its final version before submission for publication.


The authors would like to thank interviewers and nurses who participated in recruiting study participants and data collection. The principal investigator acknowledges the financial support he got from the German Academic Exchange Service (Deutscher Akademischer Austausch Dienst, DAAD).

Conflicts of Interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.


Select your language of interest to view the total content in your interested language

Scope Categories
  • Clinical Research
  • Epidemiology
  • Oncology
  • Biomedicine
  • Dentistry
  • Medical Education
  • Physiotherapy
  • Pulmonology
  • Nephrology
  • Gynaecology
  • Dermatology
  • Dermatoepidemiology
  • Otorhinolaryngology
  • Ophthalmology
  • Sexology
  • Osteology
  • Kinesiology
  • Neuroscience
  • Haematology
  • Psychology
  • Paediatrics
  • Angiology/Vascular Medicine
  • Critical care Medicine
  • Cardiology
  • Endocrinology
  • Gastroenterology
  • Infectious Diseases and Vaccinology
  • Hepatology
  • Geriatric Medicine
  • Bariatrics
  • Pharmacy and Nursing
  • Pharmacognosy and Phytochemistry
  • Radiobiology
  • Pharmacology
  • Toxicology
  • Clinical immunology
  • Clinical and Hospital Pharmacy
  • Cell Biology
  • Genomics and Proteomics
  • Pharmacogenomics
  • Bioinformatics and Biotechnology